The Tumor Microenvironment (Cancer Drug Discovery and Development)

Cancer Drug Discovery and Development Series Editor: Beverly A. Teicher

For further volumes: http://www.springer.com/series/7625

Rebecca G. Bagley Editor

The Tumor Microenvironment

Editor Rebecca G. Bagley Genzyme Corporation Framingham MA, USA [email protected]

ISBN 978-1-4419-6614-8 e-ISBN 978-1-4419-6615-5 DOI 10.1007/978-1-4419-6615-5 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2010934382 © Springer Science+Business Media, LLC 2010 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer soft-ware, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)

Preface

The fact that tumors are composed of both tumor cells and host cells has long been known. These tumor-associated cells include vascular endothelial cells and pericytes, as well as inflammatory cells such as neutrophils, monocytes, macrophages, mast cells and eosinophils, and lymphocytes. The tumor cells also interact with stromal cells and with elements of the tissue extracellular matrix. What has been less appreciated is the role that these cells could have in modulating the growth, invasion, and metastasis of the tumor. Early on, the elements of what we now call the tumor microenvironment were considered to be more or less innocent bystanders to the role of the tumor cells as they grew and invaded local sites. Today, there is an increased understanding of the critical role of the tumor microenvironment as dramatically influencing the course of tumor development and dissemination. This volume represents a superb compilation of the latest thoughts and data regarding the role of each essential component of the tumor microenvironment in cancer development and progression. Perhaps, the earliest recognition of the role of nonmalignant cells as cancer regulators was the recognition that lymphocytes can participate in what was termed “immune surveillance” in the 1960s. Our understanding of tumor immunity has improved markedly since then, and there are now successful clinical studies showing the potential use of immune-based therapies in cancer treatment. The role of tumor infiltrating lymphocytes is nicely discussed here in the chapter by Fu and Yu, with a special emphasis on the potential prognostic implications of the presence of these cells at a tumor site. Natural killer cells, first shown to be involved in tumor surveillance by Herlyn in the 1980s, also have multiple effects on tumors as discussed in the chapter by Arai. Additional immune and inflammatory cells are also discussed. These include macrophages, which have effects on both tumor cells and blood vessels, and mast cells, which have a potent and somewhat underappreciated effect on tumor growth that is clarified in the chapter by Angelidou and Zhang. A picture emerges of the tumor as a city filled with diverse cellular inhabitants who operate in the role of a “zoning board” in telling the tumor how fast it can expand and into what spaces it can grow. An extremely important regulator of tumor growth and metastasis is the tumor vasculature. Again, tumors were known to be filled with functional blood vessels for many decades before Judah Folkman explained how they might influence tumor v

vi

Preface

growth. His theory of tumor angiogenesis predicted that anti-angiogenic agents might be developed to combat cancer and this prediction has been borne out. In many cases, the predominant signal for angiogenesis is tissue hypoxia and a very strong section of this book deals with the effects of hypoxia on the tumor and on the vascular components of the tumor. The potential translational aspects of this field is just emerging and a compelling discussion of the novel therapies that target the hypoxia induced transcription factor, HIF-1-alpha can be found in the book’s opening chapter by Rapisarda and Melillo. Along with the immune cells, inflammatory cells, and vascular cells, the tumor surrounds stromal cells including fibroblasts and other mesenchymal cells. The interplay between stromal cells and tumor cells is an extremely vibrant area of study and is nicely discussed in the chapter by Orimo. Of course, a principal function of these cells is to produce and remodel the elements of the extracellular matrix as described in the contribution by DeClerck. These proteins can present a barrier to tumor invasion and metastasis. Excellent and comprehensive discussions of the role of collagens and of fibronectin as well as their binding partners the integrins and tetraspanins can be found in the section on the extracellular matrix. The modulation of the matrix by cancer cells is discussed in detail in the chapter on matrix metalloproteases and cancer cell invasion by Zucker and Cao. The interplay of tumor cells and the microenvironment is mirrored in the complex interactions of tumor cells, fibroblasts, matrix elements, and the various cells that produce either matrix metalloproteases or their natural inhibitors. No contemporary discussion of the role of the tumor microenvironment could be complete without an in-depth examination of the existence and potential role of stem cells in the establishment of human tumors and in their metastatic dissemination. This is a relatively new field and comes with the attendant controversies that often characterize new areas of investigation. This can make for very exciting reading, as it does in the chapters on cancer stem cells by Giordano, on endothelial cell precursors by Shaked and on mesenchymal stem cells by Banerjee. It is safe to say that the exact contribution of all of these progenitor, precursor and stem cell populations to cancer is an emerging and truly important field of inquiry. These chapters beautifully capture the scientific foment that characterizes the study of these cells and their role in cancer. Perhaps, the most exciting aspect of the increased attention paid to the tumor microenvironment is the promise that new approaches to cancer therapy can be derived from our paying attention to these non-tumor elements and to their interactions with the tumor cells. This is beautifully captured throughout this book. One senses an emerging optimism that the tumor microenvironment is not only important, but that it represents a treasure trove of potential therapeutic opportunities such as the ones described in the chapter on secreted growth factors as therapeutic targets by Teicher. Whether one starts out as a skeptic, or as a devotee, this excellent and comprehensive compilation will educate the reader and stimulate new ideas about tumors as an integrated combination of cellular, secreted, stromal, and matrix elements that each represents a potential therapeutic target. Boston, MA

Bruce Zetter

Contents

Part I: Physiological Parameters   1 Combination Strategies Targeting Hypoxia Inducible Factor 1 (HIF-1) for Cancer Therapy.................................................... Annamaria Rapisarda and Giovanni Melillo

3

  2 The Tumor Microenvironment: New Insights into Regulation of Tumor pH by Carbonic Anhydrases.............................. Pawel Swietach, Adrian L. Harris, and Richard D. Vaughan-Jones

23

  3 Hypoxia, Gene Expression, and Metastasis........................................... Olga V. Razorenova and Amato J. Giaccia   4 Molecular Mechanisms Regulating Expression and Function of Cancer-Associated Carbonic Anhydrase IX..................... Jaromir Pastorek and Silvia Pastorekova   5 Glycolytic Pathway as a Target for Tumor Inhibition.......................... Weiqin Lu and Peng Huang

43

59 91

Part II: Malignant Cells   6 Aberrant DNA Methylation in Cancer Cells......................................... Toshikazu Ushijima

121

  7 DNA Repair and Redox Signaling.......................................................... Mark R. Kelley, Millie M. Georgiadis, and Melissa L. Fishel

133

  8 Cancer Stem Cells and Microenvironment............................................ Mario Federico and Antonio Giordano

169

  9 Epithelial–Mesenchymal Transition in Development and Diseases...................................................................... Yadi Wu and Binhua P. Zhou

187 vii

viii

Contents

10 Invasion and Metastasis......................................................................... Douglas M. Noonan, Giuseppina Pennesi, and Adriana Albini 11 Dormancy of Disseminated Tumor Cells: Reciprocal Crosstalk with the Microenvironment.................................................. Paloma Bragado, Aparna C. Ranganathan, and Julio A. Aguirre-Ghiso

213

229

Part III: Vasculature and Stroma 12 Impact of Endothelial Progenitor Cells on Tumor Angiogenesis and Outcome of Antiangiogenic Therapy: New Perspectives on an Ongoing Controversy.................................... Robert S. Kerbel, Francesco Bertolini, and Yuval Shaked 13 Bone Marrow Derived Mesenchymal Stem/Stromal Cells and Tumor Growth....................................................................... Pravin J. Mishra and Debabrata Banerjee 14 Integrin Signaling in Lymphangiogenesis............................................ Barbara Garmy-Susini 15 Role of Pericytes in Resistance to Antiangiogenic Therapy................................................................................................... Koji Matsuo, Chunhua Lu, Mian M.K. Shazad, Robert L. Coleman, and Anil K. Sood 16 Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas........................................................................... Urszula M. Polanska, Kieran T. Mellody, and Akira Orimo

257

275 289

311

325

Part IV: Immune-Mediated Cells 17 Mast Cells and Tumor Microenvironment.......................................... Theoharis C. Theoharides, Konstantinos-Dionysios Alysandratos, Asimenia Angelidou, and Bodi Zhang

353

18 Macrophages in the Tumor Microenvironment.................................. Monica Escorcio-Correia and Thorsten Hagemann

371

19 The Prognostic Significance of Tumor-Infiltrating Lymphocytes........................................................................................... Ping Yu and Yang-Xin Fu

385

Contents

20 The Pro-inflammatory Milieu and Its Role in Malignant Epithelial Initiation................................................................................ Adam Yagui-Beltrán, Qizhi Tang, and David M. Jablons 21 Natural Killer Cells for Adoptive Immunotherapy............................ Jonathan E. Benjamin and Sally Arai

ix

409 431

Part V: Extracellular Matrix 22 Fibronectin.............................................................................................. Andreas Menrad

457

23 Collagen in Cancer................................................................................. Janelle L. Lauer and Gregg B. Fields

477

24 Integrins and Cancer............................................................................. Laurie G. Hudson and M. Sharon Stack

509

25 Matrix Metalloproteinases and Cancer Cell Invasion/Metastasis................................................................................ Stanley Zucker and Jian Cao 26 Tetraspanins and Cancer Metastasis.................................................... Margot Zöller

531 555

Part VI: Secreted Proteins 27 Chemokines and Metastasis.................................................................. Kalyan C. Nannuru, Seema Singh, and Rakesh K. Singh

601

28 Transforming Growth Factor-b in Lung Cancer, Carcinogenesis, and Metastasis............................................................ Sonia B. Jakowlew

633

29 Cooperative Interactions Between Integrins and Growth Factor Signaling in Pathological Angiogenesis.................................... Jennifer Roth, Eric Tweedie, and Peter C. Brooks

673

30 The Extracellular Matrix and the Growth and Survival of Tumors......................................................................... Yves A. DeClerck

695

x

Contents

31 Secreted Growth Factors as Therapeutic Targets............................... Beverly A. Teicher

711

32 Adrenomedullin...................................................................................... Rebecca G. Bagley

733

Index................................................................................................................

749

Contributors

Julio A. Aguirre-Ghiso Division of Hematology and Oncology, Departments of Medicine and Department of Otolaryngology, Tisch Cancer Institute, Mount Sinai School of Medicine, One Gustave L Levy Place, Box 1079, New York, NY 10029, USA Adriana Albini IRCCS Multimedica, Via Fantoli 16/15, 20138 Milano, Italy Konstantinos-Dionysios Alysandratos, MD Tufts University School of Medicine, 136 Harrison Ave., Boston, MA 02111, USA Asimenia Angelidou, MD Tufts University School of Medicine, 136 Harrison Ave., Boston, MA 02111, USA Sally Arai, MD Division of Blood and Marrow Transplantation, Stanford School of Medicine, 300 Pasteur Drive, H3249 MC 5623, Stanford, CA 94305, USA, Rebecca G. Bagley Genzyme Corporation, 49 New York Ave, Framingham, MA 01701, USA Debabrata Banerjee Department of Medicine and Pharmacology, Robert Wood Johnson Medical School, The Cancer Institute of New Jersey, University of Medicine and Dentistry of New Jersey, 195 Little Albany Street, New Brunswick, NJ 08901, USA Jonathan E. Benjamin Division of Blood and Marrow Transplantation, Stanford University Medical Center, Stanford, CA, USA

xi

xii

Contributors

Francesco Bertolini, MD/PhD Department of Hematology-Oncology, European Institute of Oncology, Milan, Italy Paloma Bragado, PhD Division of Hematology and Oncology, Department of Medicine and Department of Otolaryngology, Mount Sinai School of Medicine – NYU, New York, NY 10029, USA Peter C. Brooks Maine Medical Center Research Institute, Center for Molecular Medicine, 81 Research Drive, Scarborough, ME, USA Jian Cao, PhD Stony Brook University, Room 004, Life Sciences Building, Stony Brook, NY 11794-5200, USA Robert L. Coleman, MD Department of Gynecologic Oncology, M. D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030, USA; Center for RNA Interference and Non-Coding RNA, University of Texas, Houston, TX, USA Yves A. DeClerck Departments of Pediatrics and Biochemistry and Molecular Biology, University of Southern California and The Saban Research Institute of Childrens Hospital Los Angeles, Los Angeles, CA 20027, USA Monica Escorcio-Correia, PhD Barts and The London School of Medicine and Dentistry, Institute of Cancer, Charterhouse Square, London, UK EC1M 6BQ Mario Federico, MD Sbarro Health Research Organization, Temple University, BioLife Science Building, Suite 333, 1900 N. 12th Street, Philadelphia, PA 19122, USA; Section of Clinical Oncology, University of Palermo, Palermo, Italy Gregg B. Fields Department of Biochemistry, University of Texas Health Science Center, 7703 Floyd Curl Drive, San Antonio, TX 78229, USA Melissa L. Fishel, PhD Herman B. Wells Center for Pediatric Research, Indiana University, 980 W. Walnut Street, Indianapolis, IN 46202, USA

Contributors

Yang-Xin Fu, MD/PhD Department of Pathology and Section of Dermatology/ Department of Medicine, University of Chicago, 5841 S. Maryland Ave, MC 3083, Chicago, IL 60637, USA Barbara Garmy-Susini Unité mixte Inserm U858, Institut de Médecine Moléculaire de Rangueil, IFR 150, 1, Avenue Jean Poulhès, BP 84225 31432, Toulouse Cedex 4, France Millie M. Georgiadis, PhD Herman B. Wells Center for Pediatric Research, Indiana University, 980 W. Walnut Street, Indianapolis, IN 46202, USA Amato J. Giaccia Division of Radiation and Cancer Biology, Stanford University, Stanford, CA 94305, USA Antonio Giordano Biology Department, Sbarro Health Research Organization, Temple University, Philadelphia, PA, USA Thorsten Hagemann Barts and The London School of Medicine and Dentistry, Institute of Cancer, Charterhouse Square, London, EC1M 6BQ, UK Adrian L. Harris Weatherall Institute of Molecular Medicine, Oxford University, Oxford OX3 9DS, UK Peng Huang Department of Molecular Pathology, The University of Texas MD, Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030, USA Laurie G. Hudson, PhD University of New Mexico College of Pharmacy, Albuquerque, NM, USA David Jablons, MD Department of Surgery, University of California San Francisco, San Francisco, CA 94143, USA Sonia B. Jakowlew National Cancer Institute, Center for Cancer Training, Cancer Training Branch, Bethesda, MD 20892, USA

xiii

xiv

Contributors

Mark R. Kelley Department of Pediatrics, Section of Hematology/Oncology, Herman B Wells Center for Pediatric Research, Indiana University, Indianapolis, IN 46202, USA Robert S. Kerbel, PhD Department of Cellular and Molecular Biology, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, ON, Canada Janelle L. Lauer, PhD The University of Texas Health Science Center, 7703 Floyd Curl Drive, San Antonio, TX 78229-3900, USA Chunhua Lu, MD/PhD Department of Gynecologic Oncology, M. D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030, USA Weiqin Lu, PhD MD Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030, USA Koji Matsuo, MD Department of Gynecologic Oncology, M. D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030, USA Giovanni Melillo SAIC-Frederick, Inc., DTP-Tumor Hypoxia Laboratory, Developmental Therapeutics Program, National Cancer Institute – Frederick, Building 432, Room 218, Frederick, MD 21702, USA Kieran T. Mellody, BSc./MPhil Paterson Institute for Cancer Research, The University of Manchester, Wilmslow Road, Manchester, England M20 4BX Andreas Menrad Antibody Therapeutics Genzyme Europe Research, 310 Cambridge Science Park, Milton Road, Cambridge, CB4 OWG, UK Pravin J. Mishra, MS Department of Medicine, Robert Wood Johnson Medical School, The Cancer Institute of New Jersey, University of Medicine and Dentistry of New Jersey, 195 Little Albany St, New Brunswick, NJ 08901, USA Kalyan C. Nannuru, PhD Department of Pathology and Microbiology, University of Nebraska Medical Center, 985900 Nebraska Medical Center, Omaha, NE 68198-5900, USA

Contributors

xv

Douglas M. Noonan, PhD Universita degli Studi dell’Insubria, Varese, Italy Akira Orimo CR-UK Stromal-Tumour Interaction Group, Paterson Institute for Cancer Research, The University of Manchester, Wilmslow Road, Manchester, M20 4BX, UK Jaromir Pastorek Centre of Molecular Medicine, Slovak Academy of Sciences, Dubravska cesta 9, 845 05 Bratislava, Slovak Republic Silvia Pastorekova Centre of Molecular Medicine, Slovak Academy of Sciences, Dubravask cesta 9, Bratislava 845 05, Slovak Republic Giuseppina Pennesi, MD IRCCS Multimedica, Via Fantoli 16/15, 20138 Milano, Italy Urszula M. Polanska, Msc/PhD Paterson Institute for Cancer Research, The University of Manchester, Wilmslow Road, Manchester, England M20 4BX Aparna C. Ranganathan, PhD Division of Hematology and Oncology, Department of Medicine and Department of Otolaryngology, Mount Sinai School of Medicine – NYU, New York, NY 10029, USA Annamaria Rapisarda, PhD SAIC-Frederick, Inc., DTP-Tumor Hypoxia Laboratory, Developmental Therapeutics Program, National Cancer Institute – Frederick, Building 432, Room 218, Frederick, MD 21702, USA Olga V. Razorenova, PhD Division of Radiation and Cancer Biology, Stanford University School of Medicine, 269 Campus Drive, Stanford, CA 94305, USA Jennifer M. Roth, MS Center for Molecular Medicine, Maine Medical Center Research Institute, 81 Research Drive, Scarborough, ME 04074, USA Yuval Shaked Department of Molecular Pharmacology, Rappaport Faculty of Medicine, Technion – Israel Institute of Technology, Haifa, Israel

xvi

Contributors

M. Sharon Stack University of Missouri School of Medicine, M214C Medical Sciences Building, 1 Hospital Drive, Columbia, MO 65212, USA Mian M. K. Shazad, MS/MD Department of Gynecologic Oncology, M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030, USA; Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, TX, USA Rakesh K. Singh Department of Pathology and Microbiology, University of Nebraska Medical Center, 985900 Nebraska Medical Center, Omaha, NE 68198-5900, USA Seema Singh, PhD Department of Pathology and Microbiology, University of Nebraska Medical Center, 985900 Nebraska Medical Center, Omaha, NE 68198-5900, USA Anil K. Sood Department of Gynecologic Oncology, Cancer Biology, M. D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030, USA Center for RNA Interference and Non-Coding RNA, University of Texas, Houston, TX, USA Pawel Swietach, DPhil Deparment of Physiology, Anatomy & Genetics, Oxford University, Oxford OX1 3PT, UK Qizi Tang, PhD Department of Surgery, University of California San Francisco, San Francisco, CA 94143, USA Beverly A. Teicher Genzyme Corporation, 49 New York Avenue, Framingham, MA 01701-9322, USA [email protected] Theoharis C. Theoharides Tufts University School of Medicine, 136 Harrison Ave., Boston, MA 02111, USA Eric Tweedie, BS Center for Molecular Medicine, Maine Medical Center Research Institute, 81 Research Drive, Scarborough, ME 04074, USA

Contributors

xvii

Toshikazu Ushijima National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo, Japan Richard D. Vaughan-Jones, PhD Deparment of Physiology, Anatomy & Genetics, Oxford University, Oxford OX1 3PT, UK Yadi Wu, PhD Department of Molecular and Biomedical Pharmacology, Markey Cancer Center and University of Kentucky School of Medicine, BBSRB room B336, 741 South Limestone, Lexington, KY 40506-0509, USA Adam Yagui-Beltrán Department of Surgery, Division of Adult Thoracic Surgery, The Helen Diller Family Comprehensive Cancer Center Thoracic Surgery, University of California San Francisco, 1600 Divisadero, Room A-743, San Francisco, CA 94143-1724, USA Ping Yu The Committee on Immunology and Department of Pathology and Section of Dermatology/Department of Medicine, University of Chicago, 5841 S. Maryland Ave, Chicago, IL 60637, USA Bruce Zettler, PhD Childrens’s Hospital, Boston Harvarel Medical School, 300 Lougwood Ave., Boston, MA 02115, USA Bodi Zhang, MD/MPH Tufts University School of Medicine, 136 Harrison Ave., Boston, MA 02111, USA Binhua P. Zhou Markey Cancer Center, University of Kentucky School of Medicine, Lexington, KY, USA and Department of Molecular and Cellular Biochemistry, University of Kentucky School of Medicine, Lexington, KY, USA

xviii

Margot Zöller Department of Tumor Cell Biology, University Hospital of Surgery, Im Neuenheimer Feld 365, D-69120, Heidelberg, Germany and Department of Tumor Progression and Immune Defense, German Cancer Research Center, Heidelberg, Germany and Department of Applied Genetics, University of Karlsruhe, Karlsruhe, Germany Stanley Zucker Veterans Affairs Medical Center, Northport, NY 11768, USA and Stony Brook University, Stony Brook, NY 11794, USA

Contributors

Part I

Physiological Parameters

Chapter 1

Combination Strategies Targeting Hypoxia Inducible Factor 1 (HIF-1) for Cancer Therapy Annamaria Rapisarda and Giovanni Melillo

Abstract  Solid tumors often present regions of decreased oxygen levels (hypoxia) due to an imbalance between increased oxygen consumption and insufficient oxygen delivery from the aberrant tumor vasculature. Intratumor hypoxia is associated with altered cellular metabolism, an invasive and metastatic phenotype, as well as resistance to radiation and chemotherapy. The discovery of Hypoxia Inducible Factor-1 (HIF-1), a transcription factor critically involved in cellular responses to hypoxia and tumor progression, has provided evidence of a potential molecular target of intratumor hypoxia that could be exploited for the development of novel cancer therapeutics. A growing number of small molecule inhibitors of HIF-1, which act by distinct molecular mechanisms, have been described so far. However, HIF-1 expression in human cancers is focal and heterogeneous, consistent with the possibility that single agent HIF-1 inhibitors may have limited clinical activity. It is then plausible that combination strategies aimed at maximizing the clinical potential of HIF-1 inhibition may be more effective. We will discuss current approaches used for targeting HIF-1, emphasizing in particular opportunities for rationally designed combination strategies aimed at exploiting vulnerable features of the tumor microenvironment.

Introduction Whereas each tumor type reflects a heterogeneous collection of genetic and epigenetic alterations, hypoxia represents a unifying characteristic of human tumors, being a widely represented feature of the tumor microenvironment. Intratumor hypoxia not only affects cells that are located far from a functional blood vessel but also originates from an imbalance between increased oxygen consumption of rapidly G. Melillo (*) DTP-Tumor Hypoxia Laboratory, Bldg. 432, Room 218, SAIC – Frederick, Inc., NCI at Frederick, Frederick, MD 21702, USA e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_1, © Springer Science+Business Media, LLC 2010

3

4

A. Rapisarda and G. Melillo

growing tumor cells and inadequate oxygen supply from a structurally and functionally abnormal tumor vasculature. The fluctuation of oxygen levels in the tumor microenvironment results in the induction of an array of responses in tumor cells, such as inactivation of apoptotic pathways and activation of pro-survival pathways, induction of a more invasive and metastatic phenotype, switch to a glycolytic metabolism and induction of angiogenesis, responses aimed at providing adaptation to, or escape from, the hypoxic microenvironment. Clinically, hypoxia is associated with resistance to standard treatment, including chemotherapy and radiation therapy, and is predictive of metastasis and poor outcome in a variety of tumor types (Hockel and Vaupel 2001; Brown and Giaccia 1998; Brizel et  al. 1996; Semenza 2007). Hypoxia controls the expression of hundreds of genes, many of which are regulated by Hypoxia Inducible Factor-1 (HIF-1), a master regulator of the transcriptional response to oxygen deprivation (Semenza 2007; Manalo et al. 2005). HIF-1 is a heterodimeric protein comprised of a constitutively expressed HIF-1b (also known as aryl hydrocarbon receptor nuclear translocator, ARNT) subunit and a HIF-a subunit, which is regulated by oxygen concentrations (Wang et al. 1995; Wang and Semenza 1995). Under normoxic ­onditions, HIF-a is rapidly hydroxylated on two proline residues by oxygen-dependent prolyl hydroxylases (PHDs) and targeted for ubiquitylation and proteasomal degradation in a VHL-dependent fashion. In addition, HIF-1a can be hydroxylated on an asparagine residue (Asn803) by Factor Inhibiting HIF-1 (FIH), hence inhibiting the recruitment of the co-activator p300/CBP and transcriptional activity. In contrast, under hypoxic conditions the HIF-a subunit is stabilized and translocates to the nucleus where it dimerizes with HIF-1b and, by binding to hypoxia responsive elements (HREs), activates transcription of hundreds of target genes involved in key steps of tumorigenesis, including angiogenesis, metabolism, proliferation, metastasis, and differentiation (Semenza 2008) (Fig. 1.1). Overwhelming evidence indicates that HIF-a (HIF-1a and/or HIF2a) is indeed over-expressed in the majority of human cancers (Zhong et al. 1999; Talks et al. 2000; Bos et al. 2001), where it is associated with patient mortality and poor response to treatment (Table 1.1) (Aebersold et al. 2001; Birner et al. 2000, 2001b; Bos et al. 2003; Koukourakis et al. 2006). Notably, increased HIF-1a expression is associated with a variety of genetic alterations frequently detected in human cancers, including loss of function of tumor suppressor genes, such as VHL, PTEN, and CDKN2A (which encodes ARF) or oncogenic gain of function, such as RAS, SRC, and BCR-ABL (Semenza 2003). In addition, the involvement of receptor tyrosine kinase-dependent signaling pathways, including EGFR, HER2/Neu, and the PI3K/AKT/mTOR and MAPK pathways, in the induction of HIF-1a expression also suggests that many of these pathways converge on or implicate HIF-1 in mediating cell survival and growth (Pouyssegur et al. 2006; Majumder et al. 2004; Pore et al. 2006; Hudson et al. 2002). Therefore, HIF-1 has become an attractive target for the development of novel cancer therapeutics and despite the challenges associated with the discovery and development of small molecule inhibitors of transcription factors, many HIF-1 inhibitors which may potentially be useful for cancer therapy have been described so far (Onnis et al. 2009; Melillo 2007; Welsh et al. 2006; Melillo 2006).

1  Combination Strategies Targeting Hypoxia Inducible Factor 1

5

Angiogenic factors

(VEGF, FGF2, PDGF, PIGF, Tie2)

VHL binding Proteosomal degradation

Inhibition of transcriptional activity

P402 OH

Drug resistance

(Telomerase, Oct4)

(MDR1/ABCB-1, ABCG-2)

Glycolitic metabolism, pH regulation

HIF- α P564 OH HIF- α N803OH PHDs

RTKs/mTOR

(GLUT1, PDK1, PGK1, HK2, LDH-A, CA9, MCT4)

FIH

Survival, Proliferation

HIF- α

HIF-1 translation

mRNA transcription

O2 concentration

Immortalization and stem cells

β

α

Dimerization Cofactors recruitment DNA binding

(Cyclin D1, VEGF, IGF2, TGF-α)

Migration (CXCR4, SDF-1 α)

Invasion Metastasis (UPAR, Met, LOX)

Fig. 1.1  Regulation of Hypoxia Inducible factor-a (HIF-a) and HIF-1 dependent gene expression. Under normal oxygen conditions, HIF-a is continuously translated and rapidly degraded through the VHL-proteasome pathway following hydroxylation of proline residues by PHDs. Factor inhibiting HIF-1 mediates the hydroxylation of an asparagine residue, thus inhibiting the recruitment of the co-activator p300/CBP and HIF-1 transcriptional activity. All these processes are inhibited when oxygen concentration decreases, hence the HIF-a subunit is stabilized, translocates to the nucleus where it dimerizes with HIF-1b and induces the transcription of a large number of genes involved in crucial aspects of tumorigenesis. In addition, receptor tyrosine kinase-dependent ­signaling pathways can increase the rate of HIF-1a translation, at least in part, through the activation of mTOR

Small Molecule Inhibitors of HIF-1 The approach most widely used so far to identify HIF-1 inhibitors has been cell-based high throughput screens (HTS) using reporter gene assays (Rapisarda et al. 2002). The main advantage of cell-based HTS is the potential for the identification of unrecognized pathways associated with the activation of the target being investigated. However, determining the mechanism of action of active “hits” may be time consuming and chances to identify highly specific inhibitors is very low. Indeed, the majority of HIF-1 inhibitors described so far lack specificity, which complicates the pharmacological validation of HIF-1 inhibitors for cancer therapy, and either target signaling pathways that are involved in HIF-1 regulation or inhibit HIF-1a protein accumulation and/or transcriptional activity. Examples of agents that target signaling pathways implicated in the HIF-1a expression include

6

A. Rapisarda and G. Melillo

Table 1.1  Human cancers in which HIF-1a over-expression has been associated with patient mortality Cancer type

Reference

Astrocytoma (diffuse) Bladder (superficial urothelial)a Bladder (transitional cells) Breast Breast (c-Erb2 positive) Breast (positive linfonode) Breast (negative linfonode) Cervix (early stage) Cervix (radiation therapy) Cervix (grade IB-IIIB, radiation therapy) Endometrial Gastric Gastrointestinal stromal tumor (stomach) Lung (nonsmall cell lung carcinoma) Melanoma (malignant)b Oligodendrioglioma Oropharynx-squamous cell carcinoma Ovariana Pancreas

Korkolopoulou et al. (2004) Theodoropoulos et al. (2004) Theodoropoulos et al. (2005) Vleugel et al. (2005) and Dales et al. (2005) Giatromanolaki et al. (2007) Schindl et al. (2002) Bos et al. (2003) Birner et al. (2000) Burri et al. (2003) Bachtiary et al. (2003) Sivridis et al. (2002) Griffiths et al. (2007) Takahashi et al. (2003) Swinson et al. (2004) Giatromanolaki et al. (2003) Birner et al. (2001a) Aebersold et al. (2001) Birner et al. (2001b) Sun et al. (2007)

 Combination of HIF-1a over-expression and mutant p53  HIF-2a over-expression

a

b

inhibitors of mTOR (Majumder et al. 2004; Hudson et al. 2002), AKT (Tan et al. 2005), Her2/Neu (Laughner et  al. 2001), EGFR (Pore et  al. 2006; Koukourakis et al. 2003; Zhong et al. 2000), and Bcr-Abl (Mayerhofer et al. 2002) (Fig. 1.2). Since HIF-1a accumulation is controlled primarily at the level of protein translation or protein degradation, many of the HIF-1 inhibitors identified thus far do indeed decrease HIF-1a protein levels by a number of distinct molecular mechanisms. HIF-1 inhibitors that affect protein translation, although by mechanisms not always clearly elucidated, include topoisomerase I (topotecan and EZN-2208, a pegylated form of SN38) (Rapisarda et al. 2004a, b; Sapra et al. 2008) and topoisomerase II inhibitors (NSC644221) (Creighton-Gutteridge et al. 2007), microtubule stabilizing (taxanes and epothilones) and destabilizing agents (Mabjeesh et al. 2003; Escuin et al. 2005), cardiac glycosides (Zhang et al. 2008), PX-478, which also increases HIF-1a degradation (Koh et al. 2008) and farnesyltransferase inhibitors (Han et al. 2005). Agents that appear to affect primarily HIF-1a degradation include inhibitors of the chaperon protein Hsp90 (17-AAG, 17-DMAG) (Isaacs et  al. 2002; Mabjeesh et  al. 2002), and HDAC inhibitors (vorinostat, LAQ824, FK228) (Ellis et al. 2009), which may also block HIF-1a transcriptional activity (Kong et al. 2006), and YC-1, which activates soluble guanylate cyclase (sGC) (Kim et al. 2006a). Another mechanism by which HIF-1 may be inhibited is at the level of transactivation. In addition to chetomin (Kung et al. 2004), which

1  Combination Strategies Targeting Hypoxia Inducible Factor 1

HIF-1 β

α

Co-factors Recruitment Transcriptional activity DNA binding

NSC50352

7 Chetomin Bortezomib Vorinostat Amphotericin

HIF-1-dependent gene expression

Echinomycin Polyamides Doxorubicin

Dimerization

HIF-1 β 17AAG,17DMAG YC-1, SCH66336 NS398, ibuprofen celecoxib

HIF-1 α

RTKs/mTOR

Translation Degradation

Temsirolimus Everolimus Gefitinib Erlotinib Cetuximab Herceptin

TPT,SN38 NSC644221, Digoxin Candidaspongiolide PX478, 2ME2

Proteasome HIF-1 α mRNA

HIF-1 α antisense RNA Aminoflavone

Fig. 1.2  Signaling pathways targeted by small molecule inhibitors of HIF-1

was originally identified as a small molecule that interferes with the interaction between HIF-1a C-TAD and p300/CBP, inhibition of the proteasome (bortezomib, Velcade®) has also been associated with blockade of HIF-1 transcriptional activity, despite the paradoxical increase of HIF-1a protein, although the exact mechanism of action remains to be fully elucidated (Shin et  al. 2008; Kaluz et  al. 2008). Finally, modulation of HIF-1a mRNA levels using antisense oligonucleotides (EZN-2968) (Greenberger et al. 2008) or the AhR ligand aminoflavone (Terzuoli et al. submitted) also leads to inhibition of HIF-1 expression. However, in order to identify HIF-1 small molecule inhibitors with a higher degree of selectivity, efforts are ongoing to target protein–protein interaction, a challenging yet potentially highly rewarding strategy, in an attempt to interfere with HIF-a/HIF-1b dimerization (Yang et al. 2005; Park et al. 2006; Scheuermann et al. 2009). In addition, examples of HIF-1 inhibitors that affect HIF-1 DNA binding to the HRE have been provided, including synthetic polyamides (Olenyuk et al. 2004), echinomycin (Kong et al. 2005), and anthracyclines (Lee et al. 2009), which represent a conceptually attractive strategy to inhibit HIF-1. Many of the HIF-1 inhibitors described above are either FDA approved or are in early clinical development (Table 1.2). Results of ongoing clinical trials of HIF-1 inhibitors in patients with cancer will provide information regarding both inhibition of HIF-1 and potential clinical activity of this emerging therapeutic approach.

8

A. Rapisarda and G. Melillo

Table 1.2  HIF-1 inhibitors FDA approved or in clinical development with potential mechanism of action Agent

Target

Mechanism

Reference

EZN-2968 Aminoflavone

HIF-1a mRNA AhR

Greenberger et al. (2008) Terzuoli, submitted

Topotecan EZN-2208 PX-478 Erlotinib, gefitinib, cetuximab Trastuzumab Temsirolimus, everolimus 17AAG, 17DMAG

Topoisomerase 1 Topoisomerase 1 unknown EGFR

Degradation HIF-1a mRNA processing Translation Translation Translation Signaling

Her2/Neu mTOR

Translation Translation

HSP90

Degradation

Doxorubicin Bortezomib

DNA Proteasome

Vorinostat, LAQ824, FK228

HDAC

DNA binding Transcriptional activity Degradation, transcriptional activity

Rapisarda et al. (2004b) Sapra et al. (2008) Welsh et al. (2004) Pore et al. (2006) and Luwor et al. (2005) Koukourakis et al. (2003) Del et al. (2006) and Wan et al. (2006) Isaacs et al. (2002) and Mabjeesh et al. (2002) Lee et al. (2009) Kaluz et al. (2006) Kong et al. (2006), Mie et al. (2003), and Qian et al. (2006)

Targeting HIF-1: Single Agent or Combination? The potential therapeutic benefit of HIF-1 inhibition has been extensively documented in preclinical models using both genetic and pharmacological tools, and consistent inhibition of tumor growth and angiogenesis has been demonstrated in different tumor types (Semenza 2003; Melillo 2006). However, the lack of selective HIF-1 inhibitors and reliable biomarkers associated with HIF-1 inhibition in tumor tissue, associated with the hardly predictive nature of results obtained in xenograft models, have significantly hampered the validation of HIF-1 as a target in human cancers. For instance, it is still largely unclear whether tumor regression should be an expected, although desirable, outcome of single agent HIF-1 inhibition. Evidence has been provided that HIF-1 inhibition might be most effective when used in early, rather than late, stages of tumor progression (Li et  al. 2005), a situation rarely encountered in clinical oncology practice, where the majority of patients present with metastatic disease. In addition, even conceptually, inhibition of HIF-1 expressing cells, as a single agent strategy, may not be particularly effective for a number of reasons, including the focal and heterogeneous expression of HIF-1a in solid tumors, the reliance of cancer cells on HIF-1-independent pathways in oxygenated areas of the tumor, and the overall well-established redundancy of oncogenic signaling pathways that may be hardly affected by single agent strategies. In addition, the broad involvement of HIF-1 in a number of biological pathways that are

1  Combination Strategies Targeting Hypoxia Inducible Factor 1

9

relevant for tumorigenesis provides unique opportunities for combination therapies that may take advantage of the different approaches available in the current therapeutic armamentarium. A better understanding of the role that HIF-1 plays in the biology of human cancers and in the context of available therapeutic strategies will provide valuable information for the design of rational combinations that may target vulnerable aspects of the tumor microenvironment.

Molecularly Targeted Agents and HIF-1 Inhibition Combination strategies aimed at targeting multiple and redundant signaling pathways that contribute to oncogenesis are a mainstay of current therapeutic efforts in oncology for the potential to both prevent the emergence of resistant clones and translate in a better therapeutic outcome. Indeed, combinations of molecularly targeted agents, with or without chemotherapy, are currently being tested in virtually any tumor type with the goal of assessing safety of administration and potential therapeutic efficacy. As mentioned above, several novel targeted agents currently in clinical development or FDA approved also inhibit HIF-1, raising the possibility that combination strategies incorporating these agents may already provide examples in which inhibition of HIF-1 may potentially contribute to the therapeutic efficacy. Experimental evidence in support of this conclusion has been provided by combination of the mTOR inhibitor rapamycin with the HDAC inhibitor LBH589, which showed increased antitumor activity in prostate and renal cell cancer xenografts models (Verheul et  al. 2008). Interestingly, the concomitant inhibition of HDACs and mTOR resulted in a greater inhibition of HIF-1a protein levels relative to what was achieved by each agent alone, possibly due to the activity of these inhibitors at different levels of HIF-1a regulation including translation, protein stability, and transcriptional activity. In addition, experiments conducted in colon cancer xenograft models using low doses of rapamycin in combination with the camptothecin analog irinotecan (a topoisomerase I inhibitor) showed a marked increase in antitumor activity associated with a profound inhibition of HIF-1a protein accumulation that was not observed with either agent alone (Pencreach et  al. 2009). Interestingly, previous studies in a human glioma xenograft model had suggested that daily, but not intermittent, administration of single agent topotecan, a topoisomerase I HIF-1 inhibitor, downregulated HIF-1a expression in tumor tissue and caused inhibition of tumor growth and angiogenesis (Rapisarda et al. 2004c). These examples emphasize how HIF-1 inhibition may be an underappreciated therapeutic consequence of molecularly targeted agents currently in clinical development. A better understanding of the potential contribution of HIF-1 inhibition to therapeutic activity of these combinations may help in selecting patients more likely to respond and in designing optimal schedules of administration to maximize inhibition of HIF-1-dependent responses.

10

A. Rapisarda and G. Melillo

Hypoxic Cells Are More Resistant to Chemotherapy and Radiation Therapy It has long been recognized that hypoxic cells are more resistant to conventional therapeutic approaches, including chemotherapy and radiation therapy. Likewise, it is well appreciated that resistant hypoxic cells may contribute to treatment failure and poor prognosis of patients whose cancers are more hypoxic. Resistance to radiotherapy and chemotherapy in hypoxic tumors has been attributed to a multitude of mechanisms, including direct effects (decreased generation of oxygendependent radical formation or decreased drug accessibility to the more distant hypoxic cells) and indirect effects (clonal selection of cells adapted to hypoxia) (Brown and Giaccia 1998; Vaupel et al. 2001). Hypoxia exerts selective pressure on cells for loss of p53, regulates the transcription of several genes pivotal for the selection of a resistant phenotype, such as multidrug resistance gene 1 (MDR-1) and ABCG-2 and ABCB-1 (ABC transporters), induces expression of VEGF and other pro-angiogenic factors (important for the survival of endothelial cells following radiation), downregulates genes involved in DNA mismatch repair (MSH2, MSH6, NSB1, and BRCA1) leading to increased genomic instability (Huang 2008), and finally induces genes important for survival of cancer cells (Bertout et al. 2008; Huang 2008). Many of the effects described above are mediated by HIF-1; therefore, it is conceivable that inhibition of HIF-1 may contribute in a number of ways to combination therapies by counteracting mechanisms that are activated in the tumor microenvironment to generate resistance and by sensitizing cancer cells to available therapeutic strategies.

Combination of HIF-1 Inhibitors with Chemotherapy The potential contribution of HIF-1a to cellular resistance to chemotherapy has been suggested by evidence in both experimental models and the clinical setting. Initial clinical observations indicated that high levels of HIF-1a expression were associated with incomplete responses to radiation and chemotherapy in patients with head and neck squamous cancers (Aebersold et al. 2001; Koukourakis et al. 2003). Formal evidence that HIF-1a may be implicated in mediating resistance to conventional chemotherapeutic agents was provided by studies showing that MDR-1, one of the most widely expressed efflux pumps that mediate resistance to chemotherapy, is a direct HIF-1 target gene (Comerford et al. 2002). In addition, it was shown that HIF-1a deficient mouse embryonic fibroblasts were more sensitive to carboplatin, etoposide (a topoisomerase II inhibitor), and irradiation (agents that cause double-strand DNA breaks) but not to the topoisomerase 1 inhibitor SN38 (that generates single-strand DNA breaks) (Unruh et al. 2003), suggesting that HIF-1-mediated resistance to chemotherapy cannot simply be explained by upregulation of MDR-1. Resistance to etoposide, a

1  Combination Strategies Targeting Hypoxia Inducible Factor 1

11

topoisomerase II inhibitor, also appears to correlate with HIF-1-dependent induction of anti-apoptotic genes and downregulation of topoisomerase IIa (Sermeus et al. 2008; Sullivan and Graham 2009). The potential synergistic activity of HIF-1 inhibition in combination with chemotherapy has been formally tested in a glioma xenograft model in which HIF-1a expression was downregulated by an inducible shRNA system (Li et  al. 2006). Earlier evidence had indicated that HIF-1a downregulation by shRNA effectively inhibited tumor growth, although the effect was more pronounced in early rather than late stages of tumor progression (Li et al. 2005). The combination of the cytotoxic drug temozolomide with HIF-1a knockdown exhibited a super-additive therapeutic effect compared with either approach alone, demonstrating that HIF-1 inhibition impacted on sensitivity to chemotherapy (Li et al. 2006). However, the same authors showed no benefit of a combination between HIF-1 inhibition and the angiogenesis inhibitor ABT-869 (a multitargeted receptor tyrosine kinase inhibitor). It is interesting to note that following treatment with ABT-869, large tumors that are less likely to be responsive to HIF-1a inhibition became sensitive to HIF-1a knockdown, suggesting that anti-angiogenic treatment might partially alleviate the “resistance” of large tumors to HIF-1a knockdown and sensitize tumors to HIF-1 inhibition. Further studies in preclinical models, ultimately with validation in clinical trials, are required to fully explore the potential of HIF-1 inhibition to modulate cellular sensitivity to chemotherapeutic agents and to block stress pathways induced by chemotherapy in the tumor microenvironment. Combination of HIF-1 Inhibitors with Radiation Therapy Hypoxia is an important factor contributing to tumor radioresistance, a phenomenon initially explained by the reduced generation of oxygen radicals in hypoxic tissues and consequent decrease in radiation-induced DNA damage (Moeller et al. 2007). In reality, more complex biological pathways are implicated in the effects of radiation therapy on hypoxic tumors. Clinical studies have suggested that high levels of HIF-1a expression correlate with poor local control in patients with oropharyngeal cancers (Aebersold et al. 2001) and evidence from preclinical studies has indicated that HIF-1 deficient tumors are more radiosensitive than their wild-type counterpart (Zhang et al. 2004; Moeller et al. 2005; Williams et al. 2005). Moreover, it has been shown that radiation increases HIF-1a levels and induces the expression of HIF-1-dependent genes in tumors (Moeller et  al. 2004), consistent with the possibility that HIF-1 activation may be part of compensatory pathways that might mediate resistance to radiation therapy. However, further studies have shown that HIF-1 may have divergent effects on radiosensitivity. On the one hand, HIF-1 might be involved in the induction of apoptosis by increasing p53 phosphorylation following radiation treatment, thus sensitizing cancer cells to radiation (Moeller et al. 2005). On the other side, it is well appreciated that the tumor vasculature is an important target of radiation therapy and a major determinant of response

12

A. Rapisarda and G. Melillo

(Garcia-Barros et  al. 2003), thus the induction of angiogenic factors, including VEGF, by HIF-1 may actually protect endothelial cells from radiation-induced apoptosis (Gupta et  al. 2002). Indeed, combination of the HIF-1 inhibitor YC-1 with radiation blocked radiation induced HIF-1 upregulation and resulted in significant vascular destruction, suggesting that combining HIF-1 blockade with radiation may increase antitumor efficacy (Moeller et al. 2004). In addition, the HIF-1 inhibitor PX-478 increased therapeutic efficacy of radiation by blocking HIF-1-dependent effects on the tumor microenvironment in addition to a direct effect on tumor cell survival (Schwartz et al. 2009). Finally, recent evidence also suggests that HIF-2a inhibition sensitizes renal cancer cells to radiation therapy by increasing p53 activity and inducing apoptosis (Bertout et al. 2009). Given the complexity of the role that HIF signaling may play in the response of tumor tissue to radiation therapy, careful consideration should be given to the identification of tumor types and features of the tumor microenvironment that may predict for or be associated with a positive interaction between radiation therapy and HIF-1 inhibition.

Intratumor Hypoxia as a Potential Mechanism of Resistance to Anti-angiogenic Therapies The excitement associated with the introduction of anti-angiogenic agents for ­cancer therapy has been partially blunted by the limited single agent activity of this therapeutic strategy and by the development of resistance, a phenomenon originally unanticipated for an approach targeted to “normal,” genetically stable endothelial cells. While anti-angiogenic therapies have shown clinical benefit in combination with chemotherapy in metastatic colorectal cancer, advanced nonsmall cell lung cancer, and metastatic breast cancer (Ferrara 2005; Shojaei and Ferrara 2007; Ellis and Hicklin 2008), the single agent activity has been well below expectations and has been demonstrated only in few tumor types, including in renal cell carcinoma and more recently gliomas. More importantly, clinical benefit has only been transient and most tumors eventually become resistant and relapse. In addition to “intrinsic” resistance in patients who do not respond to anti-angiogenic agents, several mechanisms of “acquired” resistance have been proposed, including upregulation of alternative pro-angiogenic signals, production of pro-angiogenic factors by stromal cells, recruitment of bone marrow derived pro-angiogenic cells, increased pericyte coverage of the vasculature, and activation of invasive phenotypes (Rapisarda and Mellilo 2009). Interestingly, many of the responses implicated in resistance to anti-angiogenic therapy may be mediated by the products of genes induced by HIF-1. The consequences of effective anti-angiogenic therapy on the tumor microenvironment are still poorly understood and partly controversial, which limits our ability to develop rational combination therapies. On the one hand, “normalization” of tumor vasculature has been proposed, with consequent decrease in interstitial pressure and

1  Combination Strategies Targeting Hypoxia Inducible Factor 1

13

better delivery of chemotherapy (Jain 2005); on the other hand, vascular “regression” has been demonstrated in several experimental models and appears to be a logical consequence of effective inhibition of tumor vasculature (Kerbel and Folkman 2002). The latter scenario is associated with an increase in intratumor hypoxia, which not only might be mechanistically involved in mediating resistance to anti-angiogenic therapies but also might provide unique opportunities for combination strategies (Paez-Ribes et al. 2009; Loges et al. 2009). Indeed, while expression of HIF-1a in solid tumors is focal and heterogeneous, detected predominantly in perinecrotic areas, “therapy-induced” hypoxia, which may be associated with the administration of antiangiogenic therapies, will lead to a more persistent and protracted expression of HIF-1a. Higher levels of HIF-1 in the tumor microenvironment may not only contribute to drug resistance in general, but also provide a mechanism of “escape” for tumor cells overcoming the potential therapeutic effects of anti-angiogenic agents. Thus, it is conceivable that combination of anti-angiogenic therapy with HIF-1 inhibition may have synergistic therapeutic activity by targeting HIF-1 in the context of a “hypoxic stressed” tumor microenvironment. Combination of Anti-angiogenic Therapies and HIF-1 Inhibitors To test the hypothesis that anti-angiogenic therapies may be more efficacious in combination with HIF-1 inhibition, we combined topotecan, a topoisomerase I inhibitor that downregulates HIF-1a protein in vitro and in vivo (Rapisarda et al. 2004b, c), with the anti-VEGF antibody bevacizumab in a glioma xenograft model. Indeed, we found that inhibition of HIF-1a by topotecan in a hypoxic stressed tumor microenvironment resulted in a more pronounced antitumor effect, relative to either agent alone. The effects on tumor growth were associated with significantly decreased HIF-1 transcriptional activity and reduced tumor cell proliferation (Rapisarda et al. 2009), consistent with the hypothesis that targeting HIF-1a activity may abrogate compensatory pathways required for cancer cell survival. Moreover, recent studies showed that genetic disruption of both HIF-1a and HIF-2a expression in colon cancer xenografts improves tumor response to sunitinib (a multitargeted receptor tyrosine kinase inhibitor) (Burkitt et al. 2009). This effect was mediated by a marked decrease in tumor angiogenesis and perfusion, through inhibition of multiple pro-angiogenic factors, as well as a marked decrease in tumor cells proliferation. To further support the role that HIF-1 inhibition might play in combination with anti-angiogenic therapies, it is interesting to point out that the combination of bevacizumab and irinotecan (another topoisomerase I inhibitor that also inhibits HIF-1) has shown clinical benefit in glioblastoma patients with a six-month overall survival of 62–77% (Vredenburgh et  al. 2007; Chen et al. 2007). Further studies are required to better define the patient population that may benefit from these combination approaches, in particular by better understanding when and to what extent anti-angiogenic therapy is indeed associated with increased intratumor hypoxia and HIF-1a expression.

14

A. Rapisarda and G. Melillo

Cancer Cell Metabolism and the Hypoxic Tumor Microenvironment More than 80 years ago, Otto Warburg observed that cancer cells avidly take up glucose and produce lactic acid under aerobic conditions, a process subsequently referred to as the Warburg effect or aerobic glycolysis. The exact molecular mechanism(s) regulating cancer cells dependence on glycolysis remain unclear, but it is likely that a combination of effects due to oncogenic alterations and the hypoxic tumor microenvironment play a role. Many genetic alterations associated with the cancer phenotype promote glycolysis and, at the same time, suppress oxidative phosphorylation. For instance, p53 plays an important role in regulating both “aerobic” and “anaerobic” respiration, while c-Myc expression activates the transcription of several glycolytic enzymes (Kroemer and Pouyssegur 2008). The largest functional group of genes consistently upregulated by HIF-1 in a number of cell types is associated with glucose metabolism. HIF-1 increases the expression of the transporters necessary for the entry of glucose into the cell, the genes involved in the enzymatic breakdown of glucose to pyruvate and the enzymes involved in the metabolism of pyruvate. In hypoxic cells, pyruvate is converted by lactate dehydrogenase (LDH) to lactate, which is then released into the extracellular space. HIF-1 also controls cellular oxygen consumption by (a) inducing the expression of PDH kinase 1 (PDK1), which inhibits pyruvate dehydrogenase (PDH) and the conversion of pyruvate to acetyl-CoA, ultimately decreasing mitochondrial respiration (Papandreou et al. 2006; Kim et al. 2006b) and (b) reducing mitochondrial mass, by counteracting the stimulatory action of c-Myc on mitochondrial biogenesis (Zhang et al. 2007). It is conceivable then that these metabolic changes utilized by cancer cells to survive in a hypoxic, nutrient deprived tumor microenvironment might also offer therapeutic opportunities to selectively target hypoxic cells and spare normal tissues.

HIF-1 Inhibitors in Combination Strategies Targeting Tumor Metabolism Cancer cells, but not normal cells, are thought to be “addicted” to glycolysis and attempts are being pursued to explore the possibility to target glycolysis for cancer therapy. However, the activity of 2-deoxyglucose (2-DG, a competitive inhibitor of glucose), as single agent, has been limited by evidence of toxicities in clinical trials (Denko 2008). Interestingly, genetic knockdown of HIF-1a, as well as treatment with CCI-779 (an inhibitor of mTOR that decreases HIF-1 activity), sensitizes cancer cells in vitro to 2-DG (Maher et al. 2007; Wangpaichitr et al. 2008), suggesting that HIF-1 inhibitors might be used to increase the therapeutic efficacy of 2-DG. An original approach has been proposed to increase oxygen consumption in cancer cells by reversing the inhibition of mitochondrial function using either the HIF-1 inhibitor echinomycin or the PDK-1 inhibitor dichloroacetate (DCA) (Cairns et al. 2007). The pharmacological induced increase in oxygen consumption, associated

1  Combination Strategies Targeting Hypoxia Inducible Factor 1

15

with the limited oxygen supply due to the dysfunctional tumor vasculature, made the  tumors more hypoxic and sensitized cancer cells to the hypoxic cytotoxin tirapazamine. These data suggest that HIF-1 inhibitors might have a potential role in combination strategies aimed at targeting tumor metabolism or oxygen consumption.

Conclusion In the complex biology of human cancers, HIF-1 is yet another attractive target for the development of cancer therapeutics. The role of HIF-1 in human cancers has been extensively investigated and accumulating evidence implicates HIF-1 in a number of biological processes associated with resistance to therapy and tumor progression. Hence, HIF-1 inhibition appears as a logical therapeutic strategy and many small molecule inhibitors of HIF-1 have been described, several of which are validated in preclinical models. However, we have still limited understanding of when and to what extent inhibition of HIF-1 in cancer patients may be effective. Single agent studies are required to have a better appreciation of the biological consequence associated with HIF-1 inhibition in human cancers, yet the lack of selective pharmacological inhibitors of HIF-1 significantly hinders this task. On the other hand, we have learned over the last few years that targeting multiple signaling pathways deregulated in cancer cells may be a more effective therapeutic strategy. The broad involvement of HIF-1 in many biological processes associated with and required for tumor progression provides unique opportunities for the development of combination therapies. A better understanding of the role played by HIF-1 in pathways associated with the administration of chemotherapy, radiation and anti-angiogenic therapies may unveil vulnerable aspects of  the tumor microenvironment that could be exploited for the rational design of combination strategies. Acknowledgments  The authors would like to thank members of the Tumor Hypoxia Laboratory and Dr. R. H. Shoemaker for helpful discussion. This project has been funded in whole or in part with Federal funds from the National Cancer Institute, National Institutes of Health, under Contract No. N01-CO-12400. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. government. This research was supported in part by the Developmental Therapeutics Program, DCTD, of the National Cancer Institute, NIH.

References Aebersold DM, Burri P, Beer KT et  al (2001) Expression of hypoxia-inducible factor-1alpha: a novel predictive and prognostic parameter in the radiotherapy of oropharyngeal cancer. Cancer Res 61:2911–2916.

16

A. Rapisarda and G. Melillo

Bachtiary B, Schindl M, Potter R et al (2003) Overexpression of hypoxia-inducible factor 1alpha indicates diminished response to radiotherapy and unfavorable prognosis in patients receiving radical radiotherapy for cervical cancer. Clin Cancer Res 9:2234–2240. Bertout JA, Patel SA, Simon MC (2008) The impact of O2 availability on human cancer. Nat Rev Cancer 8:967–975. Bertout JA, Majmundar AJ, Gordan JD et al (2009) HIF2{alpha} inhibition promotes p53 pathway activity, tumor cell death, and radiation responses. Proc Natl Acad Sci USA 106:14391–14396. Birner P, Schindl M, Obermair A et al (2000) Overexpression of hypoxia-inducible factor 1alpha is a marker for an unfavorable prognosis in early-stage invasive cervical cancer. Cancer Res 60:4693–4696. Birner P, Gatterbauer B, Oberhuber G et al (2001a) Expression of hypoxia-inducible factor-1 alpha in oligodendrogliomas: its impact on prognosis and on neoangiogenesis. Cancer 92:165–171. Birner P, Schindl M, Obermair A et al (2001b) Expression of hypoxia-inducible factor 1alpha in epithelial ovarian tumors: its impact on prognosis and on response to chemotherapy. Clin Cancer Res 7:1661–1668. Bos R, Zhong H, Hanrahan CF et  al (2001) Levels of hypoxia-inducible factor-1 alpha during breast carcinogenesis. J Natl Cancer Inst 93:309–314. Bos R, van der GP, Greijer AE et al (2003) Levels of hypoxia-inducible factor-1alpha independently predict prognosis in patients with lymph node negative breast carcinoma. Cancer 97:1573–1581. Brizel DM, Scully SP, Harrelson JM et al (1996) Tumor oxygenation predicts for the likelihood of distant metastases in human soft tissue sarcoma. Cancer Res 56:941–943. Brown JM, Giaccia AJ (1998) The unique physiology of solid tumors: opportunities (and problems) for cancer therapy. Cancer Res 58:1408–1416. Burkitt K, Chun SY, Dang DT et al (2009) Targeting both HIF-1 and HIF-2 in human colon cancer cells improves tumor response to sunitinib treatment. Mol Cancer Ther 8(5):1148–1156. Burri P, Djonov V, Aebersold DM et al (2003) Significant correlation of hypoxia-inducible factor1alpha with treatment outcome in cervical cancer treated with radical radiotherapy. Int J Radiat Oncol Biol Phys 56:494–501. Cairns RA, Papandreou I, Sutphin PD et al (2007) Metabolic targeting of hypoxia and HIF1 in solid tumors can enhance cytotoxic chemotherapy. Proc Natl Acad Sci USA 104:9445–9450. Chen W, Delaloye S, Silverman DH et  al (2007) Predicting treatment response of malignant gliomas to bevacizumab and irinotecan by imaging proliferation with [18F] fluorothymidine positron emission tomography: a pilot study. J Clin Oncol 25:4714–4721. Comerford KM, Wallace TJ, Karhausen J et al (2002) Hypoxia-inducible factor-1-dependent regulation of the multidrug resistance (MDR1) gene. Cancer Res 62:3387–3394. Creighton-Gutteridge M, Cardellina JH, Stephen AG et  al (2007) Cell type-specific, topoisomerase II-dependent inhibition of hypoxia-inducible factor-1alpha protein accumulation by NSC 644221. Clin Cancer Res 13:1010–1018. Dales JP, Garcia S, Meunier-Carpentier S et al (2005) Overexpression of hypoxia-inducible factor HIF-1alpha predicts early relapse in breast cancer: retrospective study in a series of 745 patients. Int J Cancer 116:734–739. Del BD, Ciuffreda L, Trisciuoglio D et  al (2006) Antiangiogenic potential of the mammalian target of rapamycin inhibitor temsirolimus. Cancer Res 66:5549–5554. Denko NC (2008) Hypoxia, HIF1 and glucose metabolism in the solid tumour. Nat Rev Cancer 8:705–713. Ellis LM, Hicklin DJ (2008) Pathways mediating resistance to vascular endothelial growth factortargeted therapy. Clin Cancer Res 14:6371–6375. Ellis L, Hammers H, Pili R (2009) Targeting tumor angiogenesis with histone deacetylase inhibitors. Cancer Lett 280(2):145–153. Escuin D, Kline ER, Giannakakou P (2005) Both microtubule-stabilizing and microtubuledestabilizing drugs inhibit hypoxia-inducible factor-1alpha accumulation and activity by disrupting microtubule function. Cancer Res 65:9021–9028.

1  Combination Strategies Targeting Hypoxia Inducible Factor 1

17

Ferrara N (2005) VEGF as a therapeutic target in cancer. Oncology 69(Suppl 3):11–16. Garcia-Barros M, Paris F, Cordon-Cardo C et al (2003) Tumor response to radiotherapy regulated by endothelial cell apoptosis. Science 300:1155–1159. Giatromanolaki A, Sivridis E, Kouskoukis C et al (2003) Hypoxia-inducible factors 1alpha and 2alpha are related to vascular endothelial growth factor expression and a poorer prognosis in nodular malignant melanomas of the skin. Melanoma Res 13:493–501. Giatromanolaki A, Koukourakis MI, Simopoulos C et  al (2007) Metastatic cancer cells from c-erbB-2 negative primary breast cancer maintain the original c-erbB-2/HIF1alpha phenotype. Cancer Biol Ther 6:153–155. Greenberger LM, Horak ID, Filpula D et al (2008) A RNA antagonist of hypoxia-inducible factor1alpha, EZN-2968, inhibits tumor cell growth. Mol Cancer Ther 7:3598–3608. Griffiths EA, Pritchard SA, Valentine HR et al (2007) Hypoxia-inducible factor-1alpha expression in the gastric carcinogenesis sequence and its prognostic role in gastric and gastro-oesophageal adenocarcinomas. Br J Cancer 96:95–103. Gupta VK, Jaskowiak NT, Beckett MA et al (2002) Vascular endothelial growth factor enhances endothelial cell survival and tumor radioresistance. Cancer J 8:47–54. Han JY, Oh SH, Morgillo F et al (2005) Hypoxia-inducible factor 1alpha and antiangiogenic activity of farnesyltransferase inhibitor SCH66336 in human aerodigestive tract cancer. J Natl Cancer Inst 97:1272–1286. Hockel M, Vaupel P (2001) Tumor hypoxia: definitions and current clinical, biologic, and molecular aspects. J Natl Cancer Inst 93:266–276. Huang LE (2008) Carrot and stick: HIF-alpha engages c-Myc in hypoxic adaptation. Cell Death Differ 15:672–677. Hudson CC, Liu M, Chiang GG et  al (2002) Regulation of hypoxia-inducible factor 1alpha expression and function by the mammalian target of rapamycin. Mol Cell Biol 22:7004–7014. Isaacs JS, Jung YJ, Mimnaugh EG et al (2002) Hsp90 regulates a von Hippel Lindau-independent hypoxia-inducible factor-1 alpha-degradative pathway. J Biol Chem 277:29936–29944. Jain RK (2005) Antiangiogenic therapy for cancer: current and emerging concepts. Oncology (Williston Park) 19:7–16. Kaluz S, Kaluzova M, Stanbridge EJ (2006) Proteasomal inhibition attenuates transcriptional activity of hypoxia-inducible factor 1 (HIF-1) via specific effect on the HIF-1alpha C-terminal activation domain. Mol Cell Biol 26:5895–5907. Kaluz S, Kaluzova M, Stanbridge EJ (2008) Comment on the role of FIH in the inhibitory effect of bortezomib on hypoxia-inducible factor-1. Blood 111:5258–5259. Kerbel R and Folkman J (2002) Clinical translation of angiogenesis inhibitors. Nat Rev Cancer 2:727–739. Kim HL, Yeo EJ, Chun YS et al (2006a) A domain responsible for HIF-1alpha degradation by YC-1, a novel anticancer agent. Int J Oncol 29:255–260. Kim JW, Tchernyshyov I, Semenza GL et  al (2006b) HIF-1-mediated expression of pyruvate dehydrogenase kinase: a metabolic switch required for cellular adaptation to hypoxia. Cell Metab 3:177–185. Koh MY, Spivak-Kroizman T, Venturini S et al (2008) Molecular mechanisms for the activity of PX-478, an antitumor inhibitor of the hypoxia-inducible factor-1alpha. Mol Cancer Ther 7:90–100. Kong D, Park EJ, Stephen AG et al (2005) Echinomycin, a small-molecule inhibitor of hypoxiainducible factor-1 DNA-binding activity. Cancer Res 65:9047–9055. Kong X, Lin Z, Liang D et al (2006) Histone deacetylase inhibitors induce VHL and ubiquitinindependent proteasomal degradation of hypoxia-inducible factor 1alpha. Mol Cell Biol 26:2019–2028. Koukourakis MI, Simopoulos C, Polychronidis A et al (2003) The effect of trastuzumab/docatexel combination on breast cancer angiogenesis: dichotomus effect predictable by the HIFI alpha/ VEGF pre-treatment status? Anticancer Res 23:1673–1680.

18

A. Rapisarda and G. Melillo

Korkolopoulou P, Patsouris E, Konstantinidou AE et al (2004) Hypoxia-inducible factor 1alpha/ vascular endothelial growth factor axis in astrocytomas. Associations with microvessel morphometry, proliferation and prognosis. Neuropathol Appl Neurobiol 30:267–278. Koukourakis MI, Bentzen SM, Giatromanolaki A et al (2006) Endogenous markers of two separate hypoxia response pathways (hypoxia inducible factor 2 alpha and carbonic anhydrase 9) are associated with radiotherapy failure in head and neck cancer patients recruited in the CHART randomized trial. J Clin Oncol 24:727–735. Kroemer G and Pouyssegur J (2008) Tumor cell metabolism: cancer’s Achilles’ heel. Cancer Cell 13:472–482. Kung AL, Zabludoff SD, France DS et  al (2004) Small molecule blockade of transcriptional coactivation of the hypoxia-inducible factor pathway. Cancer Cell 6:33–43. Laughner E, Taghavi P, Chiles K et al (2001) HER2 (neu) signaling increases the rate of hypoxiainducible factor 1alpha (HIF-1alpha) synthesis: novel mechanism for HIF-1-mediated vascular endothelial growth factor expression. Mol Cell Biol 21:3995–4004. Lee K, Qian DZ, Rey S et al (2009) Anthracycline chemotherapy inhibits HIF-1 transcriptional activity and tumor-induced mobilization of circulating angiogenic cells. Proc Natl Acad Sci USA 106:2353–2358. Li L, Lin X, Staver M et al (2005) Evaluating hypoxia-inducible factor-1alpha as a cancer therapeutic target via inducible RNA interference in vivo. Cancer Res 65:7249–7258. Li L, Lin X, Shoemaker AR et  al (2006) Hypoxia-inducible factor-1 inhibition in combination with temozolomide treatment exhibits robust antitumor efficacy in  vivo. Clin Cancer Res 12:4747–4754. Loges S, Mazzone M, Hohensinner P et  al (2009) Silencing or fueling metastasis with VEGF inhibitors: antiangiogenesis revisited. Cancer Cell 15:167–170. Luwor RB, Lu Y, Li X et al (2005) The antiepidermal growth factor receptor monoclonal antibody cetuximab/C225 reduces hypoxia-inducible factor-1 alpha, leading to transcriptional inhibition of vascular endothelial growth factor expression. Oncogene 24:4433–4441. Mabjeesh NJ, Post DE, Willard MT et al (2002) Geldanamycin induces degradation of hypoxiainducible factor 1alpha protein via the proteosome pathway in prostate cancer cells. Cancer Res 62:2478–2482. Mabjeesh NJ, Escuin D, LaVallee TM et al (2003) 2ME2 inhibits tumor growth and angiogenesis by disrupting microtubules and dysregulating HIF. Cancer Cell 3:363–375. Maher JC, Wangpaichitr M, Savaraj N et al (2007) Hypoxia-inducible factor-1 confers resistance to the glycolytic inhibitor 2-deoxy-D-glucose. Mol Cancer Ther 6:732–741. Majumder PK, Febbo PG, Bikoff R et al (2004) mTOR inhibition reverses Akt-dependent prostate intraepithelial neoplasia through regulation of apoptotic and HIF-1-dependent pathways. Nat Med 10:594–601. Manalo DJ, Rowan A, Lavoie T et al (2005) Transcriptional regulation of vascular endothelial cell responses to hypoxia by HIF-1. Blood 105:659–669. Mayerhofer M, Valent P, Sperr WR et  al (2002) BCR/ABL induces expression of vascular endothelial growth factor and its transcriptional activator, hypoxia inducible factor-1alpha, through a pathway involving phosphoinositide 3-kinase and the mammalian target of rapamycin. Blood 100:3767–3775. Melillo G (2006) Inhibiting hypoxia-inducible factor 1 for cancer therapy. Mol Cancer Res 4:601–605. Melillo G (2007) Targeting hypoxia cell signaling for cancer therapy. Cancer Metastasis Rev 26:341–352. Mie LY, Kim SH, Kim HS et al (2003) Inhibition of hypoxia-induced angiogenesis by FK228, a specific histone deacetylase inhibitor, via suppression of HIF-1alpha activity. Biochem Biophys Res Commun 300:241–246. Moeller BJ, Cao Y, Li CY et al (2004) Radiation activates HIF-1 to regulate vascular radiosensitivity in tumors: role of reoxygenation, free radicals, and stress granules. Cancer Cell 5:429–441. Moeller BJ, Dreher MR, Rabbani ZN et al (2005) Pleiotropic effects of HIF-1 blockade on tumor radiosensitivity. Cancer Cell 8:99–110.

1  Combination Strategies Targeting Hypoxia Inducible Factor 1

19

Moeller BJ, Richardson RA, Dewhirst MW (2007) Hypoxia and radiotherapy: opportunities for improved outcomes in cancer treatment. Cancer Metastasis Rev 26:241–248. Olenyuk BZ, Zhang GJ, Klco JM et al (2004) Inhibition of vascular endothelial growth factor with a sequence-specific hypoxia response element antagonist. Proc Natl Acad Sci USA 101: 16768–16773. Onnis B, Rapisarda A, Melillo G (2009) Development of HIF-1 inhibitors for cancer therapy. J Cell Mol Med in press.. Paez-Ribes M, Allen E, Hudock J et al (2009) Antiangiogenic therapy elicits malignant progression of tumors to increased local invasion and distant metastasis. Cancer Cell 15:220–231. Papandreou I, Cairns RA, Fontana L et al (2006) HIF-1 mediates adaptation to hypoxia by actively downregulating mitochondrial oxygen consumption. Cell Metab 3:187–197. Park EJ, Kong D, Fisher R et al (2006) Targeting the PAS-A domain of HIF-1alpha for development of small molecule inhibitors of HIF-1. Cell Cycle 5:1847–1853. Pencreach E, Guerin E, Nicolet C et al (2009) Marked activity of irinotecan and rapamycin combination toward colon cancer cells in vivo and in vitro is mediated through cooperative modulation of the mammalian target of rapamycin/hypoxia-inducible factor-1alpha axis. Clin Cancer Res 15:1297–1307. Pore N, Jiang Z, Gupta A et al (2006) EGFR tyrosine kinase inhibitors decrease VEGF expression by both hypoxia-inducible factor (HIF)-1-independent and HIF-1-dependent mechanisms. Cancer Res 66:3197–3204. Pouyssegur J, Dayan F, Mazure NM (2006) Hypoxia signalling in cancer and approaches to enforce tumour regression. Nature 441:437–443. Qian DZ, Kachhap SK, Collis SJ et al (2006) Class II histone deacetylases are associated with VHL-independent regulation of hypoxia-inducible factor 1 alpha. Cancer Res 66:8814–8821. Rapisarda A, Melillo G (2009) Role of the hypoxic tumor microenvironment in the resistance to anti-angiogenic therapies. Drug Resist Updat 12:74–80. Rapisarda A, Uranchimeg B, Scudiero DA et al (2002) Identification of small molecule inhibitors of hypoxia-inducible factor 1 transcriptional activation pathway. Cancer Res 62:4316–4324. Rapisarda A, Shoemaker RH, Melillo G (2004a) Targeting topoisomerase I to inhibit hypoxia inducible factor 1. Cell Cycle 3:172–175. Rapisarda A, Uranchimeg B, Sordet O et al (2004b) Topoisomerase I-mediated inhibition of hypoxia-inducible factor 1: mechanism and therapeutic implications. Cancer Res 64:1475–1482. Rapisarda A, Zalek J, Hollingshead M et al (2004c) Schedule-dependent inhibition of hypoxiainducible factor-1alpha protein accumulation, angiogenesis, and tumor growth by topotecan in U251-HRE glioblastoma xenografts. Cancer Res 64:6845–6848. Rapisarda A, Hollingshead M, Uranchimeg B et al (2009) Increased antitumor activity of bevacizumab in combination with hypoxia inducible factor-1 inhibition. Mol Cancer Ther 8:1867–1877. Sapra P, Zhao H, Mehlig M et al (2008) Novel delivery of SN38 markedly inhibits tumor growth in xenografts, including a camptothecin-11-refractory model. Clin Cancer Res 14:1888–1896. Scheuermann TH, Tomchick DR, Machius M et  al (2009) Artificial ligand binding within the HIF2alpha PAS-B domain of the HIF2 transcription factor. Proc Natl Acad Sci USA 106:450–455. Schindl M, Schoppmann SF, Samonigg H et al (2002) Overexpression of hypoxia-inducible factor 1alpha is associated with an unfavorable prognosis in lymph node-positive breast cancer. Clin Cancer Res 8:1831–1837. Schwartz DL, Powis G, Thitai-Kumar A et  al (2009) The selective hypoxia inducible factor-1 inhibitor PX-478 provides in  vivo radiosensitization through tumor stromal effects. Mol Cancer Ther 8:947–958. Semenza GL (2003) Targeting HIF-1 for cancer therapy. Nat Rev Cancer 3:721–732. Semenza GL (2007) Hypoxia and cancer. Cancer Metastasis Rev 26:223–224. Semenza GL (2008) Hypoxia-inducible factor 1 and cancer pathogenesis. IUBMB Life 60:591–597.

20

A. Rapisarda and G. Melillo

Sermeus A, Cosse JP, Crespin M et al (2008) Hypoxia induces protection against etoposide-induced apoptosis: molecular profiling of changes in gene expression and transcription factor activity. Mol Cancer 7:27. Shin DH, Chun YS, Lee DS et al (2008) Bortezomib inhibits tumor adaptation to hypoxia by stimulating the FIH-mediated repression of hypoxia-inducible factor-1. Blood 111:3131–3136. Shojaei F, Ferrara N (2007) Antiangiogenic therapy for cancer: an update. Cancer J 13:345–348. Sivridis E, Giatromanolaki A, Gatter KC et  al (2002) Association of hypoxia-inducible factors 1alpha and 2alpha with activated angiogenic pathways and prognosis in patients with endometrial carcinoma. Cancer 95:1055–1063. Sullivan R, Graham CH (2009) Hypoxia prevents etoposide-induced DNA damage in cancer cells through a mechanism involving hypoxia-inducible factor 1. Mol Cancer Ther 8:1702–1713. Sun HC, Qiu ZJ, Liu J et al (2007) Expression of hypoxia-inducible factor-1 alpha and associated proteins in pancreatic ductal adenocarcinoma and their impact on prognosis. Int J Oncol 30: 1359–1367. Swinson DE, Jones JL, Cox G et al (2004) Hypoxia-inducible factor-1 alpha in non small cell lung cancer: relation to growth factor, protease and apoptosis pathways. Int J Cancer 111:43–50. Takahashi R, Tanaka S, Hiyama T et al (2003) Hypoxia-inducible factor-1alpha expression and angiogenesis in gastrointestinal stromal tumor of the stomach. Oncol Rep 10:797–802. Talks KL, Turley H, Gatter KC et  al (2000) The expression and distribution of the hypoxiainducible factors HIF-1alpha and HIF-2alpha in normal human tissues, cancers, and tumorassociated macrophages. Am J Pathol 157:411–421. Tan C, de Noronha RG, Roecker AJ et al (2005) Identification of a novel small-molecule inhibitor of the hypoxia-inducible factor 1 pathway. Cancer Res 65:605–612. Theodoropoulos VE, Lazaris AC, Sofras F et al (2004) Hypoxia-inducible factor 1 alpha expression correlates with angiogenesis and unfavorable prognosis in bladder cancer. Eur Urol 46:200–208. Theodoropoulos VE, Lazaris AC, Kastriotis I et al (2005) Evaluation of hypoxia-inducible factor 1alpha overexpression as a predictor of tumour recurrence and progression in superficial urothelial bladder carcinoma. BJU Int 95:425–431. Unruh A, Ressel A, Mohamed HG et al (2003) The hypoxia-inducible factor-1 alpha is a negative factor for tumor therapy. Oncogene 22:3213–3220. Vaupel P, Kelleher DK, Hockel M (2001) Oxygen status of malignant tumors: pathogenesis of hypoxia and significance for tumor therapy. Semin Oncol 28:29–35. Verheul HM, Salumbides B, van EK et  al (2008) Combination strategy targeting the hypoxia inducible factor-1 alpha with mammalian target of rapamycin and histone deacetylase inhibitors. Clin Cancer Res 14:3589–3597. Vleugel MM, Greijer AE, Shvarts A et  al (2005) Differential prognostic impact of hypoxia induced and diffuse HIF-1alpha expression in invasive breast cancer. J Clin Pathol 58:172–177. Vredenburgh JJ, Desjardins A, Herndon JE et al (2007) Bevacizumab plus irinotecan in recurrent glioblastoma multiforme. J Clin Oncol 25:4722–4729. Wan X, Shen N, Mendoza A et al (2006) CCI-779 inhibits rhabdomyosarcoma xenograft growth by an antiangiogenic mechanism linked to the targeting of mTOR/Hif-1alpha/VEGF signaling. Neoplasia 8:394–401. Wang GL and Semenza GL (1995) Purification and characterization of hypoxia-inducible factor 1. J Biol Chem 270:1230–1237. Wang GL, Jiang BH, Rue EA et al (1995) Hypoxia-inducible factor 1 is a basic-helix-loop-helixPAS heterodimer regulated by cellular O2 tension. Proc Natl Acad Sci USA 92:5510–5514. Wangpaichitr M, Savaraj N, Maher J et al (2008) Intrinsically lower AKT, mammalian target of rapamycin, and hypoxia-inducible factor activity correlates with increased sensitivity to 2-deoxy-D-glucose under hypoxia in lung cancer cell lines. Mol Cancer Ther 7:1506–1513. Welsh S, Williams R, Kirkpatrick L et  al (2004) Antitumor activity and pharmacodynamic properties of PX-478, an inhibitor of hypoxia-inducible factor-1alpha. Mol Cancer Ther 3:233–244.

1  Combination Strategies Targeting Hypoxia Inducible Factor 1

21

Welsh SJ, Koh MY, Powis G (2006) The hypoxic inducible stress response as a target for cancer drug discovery. Semin Oncol 33:486–497. Williams KJ, Telfer BA, Xenaki D et al (2005) Enhanced response to radiotherapy in tumors deficient in the function of hypoxia-inducible factor-1. Radiother Oncol 75:89–98. Yang J, Zhang L, Erbel PJ et al (2005) Functions of the Per/ARNT/Sim domains of the hypoxiainducible factor. J Biol Chem 280:36047–36054. Zhang X, Kon T, Wang H et al (2004) Enhancement of hypoxia-induced tumor cell death in vitro and radiation therapy in vivo by use of small interfering RNA targeted to hypoxia-inducible factor-1alpha. Cancer Res 64:8139–8142. Zhang H, Gao P, Fukuda R et  al (2007) HIF-1 inhibits mitochondrial biogenesis and cellular ­respiration in VHL-deficient renal cell carcinoma by repression of C-MYC activity. Cancer Cell 11:407–420. Zhang H, Qian DZ, Tan YS et al (2008) Digoxin and other cardiac glycosides inhibit HIF-1alpha synthesis and block tumor growth. Proc Natl Acad Sci USA 105:19579–19586. Zhong H, De Marzo AM, Laughner E et  al (1999) Overexpression of hypoxia-inducible factor 1alpha in common human cancers and their metastases. Cancer Res 59:5830–5835. Zhong H, Chiles K, Feldser D et  al (2000) Modulation of hypoxia-inducible factor 1alpha expression by the epidermal growth factor/phosphatidylinositol 3-kinase/PTEN/AKT/FRAP pathway in human prostate cancer cells: implications for tumor angiogenesis and therapeutics. Cancer Res 60:1541–1545.

Chapter 2

The Tumor Microenvironment: New Insights into Regulation of Tumor pH by Carbonic Anhydrases Pawel Swietach, Adrian L. Harris, and Richard D. Vaughan-Jones

Abstract  Intracellular pH (pHi) is a powerful modulator of cell function. There is a narrow range of pHi, usually between 7.0 and 7.5, over which cell growth and development is optimal. Metabolism, particularly cellular respiration, tends to disturb this favorable steady state. The survival of cells undergoing such an acid challenge relies on efficient mechanisms for regulating pHi. In solid tumors, pHi regulation is challenged by their heavy demand for energy and poor blood supply. Inadequately perfused cancer cells deprive the milieu of oxygen, forcing a switchover to the less energy efficient, anaerobic mode of respiration. To meet energy demands, cells upregulate respiration and excrete more acid which lowers extracellular pH (pHe). The tumor extracellular milieu is believed to exert Darwinian selection in favor of cancer cells and against normal cells, based on the cell’s capacity to protect pHi under hypoxia and low pHe. In this chapter, we describe the pathways operating in tumors for removing acid and regulating pHi, with particular emphasis on membrane transporters and carbonic anhydrase (CA) enzymes. We discuss these pathways in terms of survival adaptations and possible targets for anticancer therapy.

Biological Importance of pH The hydrogen atom is very common in the human body, and despite being the lightest of elements, it comprises about a tenth of total body mass. Virtually all hydrogen atoms are bonded by means of electron transfer or sharing with other elements. In many such compounds, the hydrogen nucleus can dissociate reversibly from the remainder of the molecule. In the case of the most common isotope 1 H, the hydrogen nucleus is solely a proton, and therefore the H+ dissociation and

A.L. Harris (*) Weatherall Institute of Molecular Medicine, Oxford University, Oxford OX3 9DS, UK e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_2, © Springer Science+Business Media, LLC 2010

23

24

P. Swietach et al.

association reactions are dubbed deprotonation and protonation, respectively. Water, the biological solvent that contributes towards 60% of the body’s mass, also undergoes reversible deprotonation, and in pure form equilibrates at a “neutral” H+ ion concentration ([H+]) of 1027 M. Acids and bases are defined as molecules that increase and decrease [H+], respectively. To avoid the use of exponential notation, the acidity/basicity of a solution is commonly expressed as the negative logarithm of molar [H+], and given the symbol pH (Boron 2004). Many biologically relevant molecules are weak acids and weak bases. Proteins, for instance, are chains of weak acids/bases. Weak acids/bases exist in dynamic equilibrium between their protonated and unprotonated forms, the concentration ratio of which is very sensitive to pH. Since the H+ ion has very high charge density, pH will have a major impact on the physico-chemical properties of weak acids/ bases. This is particularly relevant to proteins, where charge distribution is linked to structure and function (Boron 2004; Swietach et al. 2007). Indeed, intracellular pH (pHi) is a potent and universal regulator of cellular physiology (Boron 2004; Swietach et al. 2007; Roos and Boron 1981; Vaughan-Jones et al. 2008; Isfort et al. 1993; McConkey and Orrenius 1996). Displacements of pHi by only a fraction of a unit can lead to patho-physiological responses and apoptotic death (Park et  al. 1999). Cells must defend their optimal pHi by corrective (homeostatic) mechanisms. A major challenge to these housekeeping mechanisms is metabolic acid production, a process that takes place in all respiring cells throughout their life. To counteract metabolic acid loading, excess intracellular H+ ions must be removed across the cell membrane into the extracellular space and out of the body at the lungs and kidneys. Despite its tiny radius, the highly-polarizing H+ ion is poorly permeant across the lipid bilayer of cell membranes. To allow transmembrane H+ - traffic, cells have evolved transport proteins for H+ ions (e.g. Na+/H+ exchange), or other means of facilitated H+ ion diffusion (e.g. its reaction with HCO3– to form membrane-permeant CO2). The performance of these mechanisms can be compromised by the accumulation of extracellular acid, either as a result of allosteric H+ - inhibition of transport proteins (Vaughan-Jones and Wu 1990; Aronson 1985) or by the collapse diffusion gradients that drive acid-efflux (Swietach et al. 2008, 2009). Through these mechanisms, extracellular pH (pHe) can feedback onto pHi homeostasis. In turn, pHe depends on the diffusive coupling across the extracellular space and with blood. Efficient pHi regulation is central to the survival of metabolically-active tissues, particularly those with relatively poor blood perfusion. Solid tumors are characterized by a high acid production rate (Gatenby and Gillies 2004; Gillies et al. 2008) and inadequate blood supply (Swietach et al. 2007; Vaupel et al. 1989; Kallinowski et al. 1989; Stubbs et al. 2000). Although cancer cells, by definition, harbor a number of mutated gene products, the bulk of cellular protein is no different from ­normal cells and bears a similar pHi sensitivity, carving a narrow pHi range for survival. The growth of tumors may depend critically on the ability of cancer cells to remove intracellular acid more efficiently than normal cells. The conditions that are characteristic of tumors, notably hypoxia and low pHe (Griffiths et  al. 2001; Gatenby et  al. 2006), may be tolerated by cancer cells, but not by normal cells, thereby selecting against the latter (Fang et al. 2008). In this chapter, we describe

2  The Tumor Microenvironment

25

the mechanisms for controlling pHi in solid tumors, with particular emphasis on the role of CO2/HCO3– buffer and its enzyme catalyst, carbonic anhydrase (CA).

Sources of Cellular Acid Cellular Respiration A large number of biochemical reactions produce or consume H+ ions, but production of lactic acid by anaerobic respiration and carbon dioxide (CO2) by aerobic respiration are quantitatively the most important acid fluxes that can alter pHi. Glucose, the major respiratory substrate in tumors, produces 2 × H+ plus 2 × lactate (i.e. ionized lactic acid) by the cytosolic O2-independent glycolytic pathway, or 6 × CO2 plus 6 × H2O through O2-dependent mitochondrial pathways. In addition, during periods of nucleic acid synthesis, glucose can yield 1 × CO2 through an additional cytosolic pathway called the pentose phosphate shunt (Helmlinger et al. 2002). Upon hydration, CO2 yields H+ and HCO3−. Assuming equilibrium at pH 7, the molar H+ – yield per aerobically and anaerobically respired glucose is 5.3 and 2.0, respectively (aciddissociation constants for CO2 and lactic acid are 1026.2 and 1024 M, respectively). Tumors demand a particularly high respiratory rate to energize their substantial growth and development. Glucose yields almost 20 times more energy (ATP molecules) when it is respired to CO2 rather than lactic acid. The carbon atoms of lactic acid are less oxidized than those in CO2, and still have considerable energy trapped in their bonds. Indeed, lactic acid can be respired further to CO2. Based on energy efficiency alone, it would be sensible for tumors to prefer aerobic respiration. But to benefit from this higher energy yield, tumors need functional mitochondria and adequate O2 supply. As solid tumors develop, they grow away from their basement membrane and vasculature. This increases the diffusion distance for O2 delivery. Furthermore, the elevated O2 demand in cancer cells that are closer to blood vessels greatly reduces O2 availability in more distal cells, and shortens the distance over which cells are adequately oxygenated. This distance, known as the Krogh radius (Krogh 1919), can be as little as a few cell layers thick in tumors (Dewhirst et al. 1994). Beyond the Krogh radius, tissue becomes hypoxic. In addition to inadequate and highly disorganized vasculature (Thomlinson and Gray 1955), the extent of hypoxia within tumors is increased by variable and discontinuous blood flow (Vaupel et al. 1989; Helmlinger et al. 1997). The size of the hypoxic region can be visualized by pimonidazole staining (Fig. 2.1a). Low O2 tension (<1%) blocks the degradation of hypoxia-inducible factor (HIF), a transcription factor that otherwise has a very short lifetime under normoxia. HIF was discovered as an inducer of erythropoietin expression (Semenza and Wang 1992), and is now an established regulator of genes involved in glucose metabolism (e.g. Glut-1 for glucose uptake, Fig. 2.1b) and pH regulation (e.g. CA9, a carbonic anhydrase isoform, Fig. 2.1c) (Semenza 2003). Hypoxia forces glucose to be respired anaerobically by blocking O2-dependent mitochondrial reactions and

26

P. Swietach et al.

Fig. 2.1  Tumor hypoxia and its correlation with hypoxia-induced proteins, modified from Airley et al. (2003). Serial sections of cervix carcinoma biopsies, immuno-stained for (a) pimonidazole, (b) Glut-1, and (c) CAIX, showing similarities in distribution and pattern. Magnification 200×. Shaded arrows point in the direction of decreasing O2, as assessed by increasing pimonidazole staining

by HIF-induced expression of pyruvate dehydrogenase kinase (PDK) and lactate dehydrogenase (LDH), which divert pyruvate away from mitochondrial reactions (PDK) and towards lactic acid production (LDH) (Semenza et al. 1994).

The Warburg Effect Under uncompromised O2 supply, it is in the interest of cellular energetics to respire aerobically. This phenomenon, called the Pasteur effect, is a coordinated effort by adenosine nucleotides (Ramaiah et al. 1964), redox potentials (Cerdan et al. 2006), and transcription factors (Hardie 2003) to favor the production of CO2 over lactic acid. In an apparent violation of the Pasteur effect, cancer cells typically show constitutive upregulation of lactic acid production even under normoxic conditions. This observation, called the Warburg effect (Gatenby and Gillies 2004; Warburg 1930), is central to cancer biology and is proposed to arise as a result of the stabilization of HIF by lactic acid (Lu et al. 2002), aberrant HIF degradation by mutated von Hippel–Lindau protein (Ivanov et al. 1998), or dysfunctional transcription factors, such as c-myc, which regulate carbohydrate turnover (Liao and Dickson 2000). The persistence of the Warburg effect in cancer cells must provide a benefit for survival that outweighs the lower ATP yield of anaerobic respiration. One advantage of the Warburg effect is its lack of dependence on O2 supply, which is highly unreliable in tumors (Gatenby and Gillies 2004). Fluctuating blood flow has been proposed to promote cancer development by selecting for cells that survive repeated hypoxia-reoxygenation cycles (Gatenby and Gillies 2004; Kimura et  al. 1996), i.e. those supporting the Warburg phenomenon. A second plausible benefit of the Warburg effect is the greater acid-yield per ATP molecule generated, seven-fold higher than aerobic respiration. It has been suggested that lactic acid overproduction sets up an acidic microenvironment that is critical for the transition from pre-invasive to malignant carcinoma (Gatenby and Gillies 2004). A positive correlation has been established between glycolytic flux (and hence acid production) and tumor aggressiveness and invasiveness (Gatenby

2  The Tumor Microenvironment

27

and Gillies 2004). Since glucose supply is usually plentiful, it is feasible to upregulate lactic acid production (viz., the Warburg effect) and deposit a greater amount of acid than possible with aerobic respiration, while still catering for cellular energy demands. The near-universal enhancement of glucose utilization by tumors, shown by 18fluorodeoxyglucose positron emission tomography (Gambhir 2002), supports this model of enhanced acid production. According to modern hypotheses of cancer progression, an acidic extracellular milieu (a) encourages cancer cell development and (b) is lethal to normal cells. There is growing evidence for an important role for H+-sensitive G-protein coupled receptors (OGR1 (Ludwig et al. 2003) and G2A (Murakami et al. 2004)) in transducing extracellular acidity into a proliferation signal (Huang et al. 2008). In a neuroblastoma cell line, low pHe activates the inositol-1,4,5-trisphosphate (IP3) signaling pathway, which triggers Ca2+ release from intracellular stores and the activation of the MEK/ERK pathway. In addition, acidic pHe has been shown to favor tumor invasiveness by increasing cell migration, decreasing cell adhesion, and disrupting the extracellular matrix (Gatenby and Gillies 2004; Gatenby and Gawlinski 1996; Kato et  al. 1992; Martinez-Zaguilan et  al. 1996; Schlappack et  al. 1991; Stock and Schwab 2009). Cancer cells are believed to survive better in an acidic extracellular microenvironment by acquiring a more potent acid-extruding phenotype that protects the intracellular compartment from acid buildup (Swietach et al. 2007; Gatenby et al. 2007). Inadequate acid extrusion in normal cells may explain the lethality of low pHe therein (Fig. 2.2).

Fig. 2.2  Cartoon of acid production from a normal cell (smooth outline) and a cancer cell (rigged outline) in an hypoxic region of a tumor, beyond the Krogh radius from a blood vessel (wavy cylinder on right). The cancer cell (bottom cell with jagged outline) produces more acid (gray-filled H+ ions) than the normal cell (top cell with smooth outline, emitting white-filled H+ ions). Extracellular acid (produced mainly by the overactive cancer cell) favors cancer cell growth and development, but has detrimental effects on normal cells due to their lack of sufficient acid-extruding mechanisms to protect the intracellular space from acidification

28

P. Swietach et al.

Transport of Acid Across the Surface Membrane Intracellular and Extracellular pH in Tumors For many decades after Otto Warburg’s seminal finding that tumors overproduce lactic acid, it was assumed that both pHi and pHe are acidic in cancers. The low pH detected using microelectrodes (Kallinowski et al. 1989; Wike-Hooley et al. 1984) was deemed to support this hypothesis, until 31P nuclear magnetic resonance (NMR) revealed alkaline pH in solid tumors (Griffiths et  al. 1981). These apparently conflicting reports are easily reconciled by considering that 31P NMR reports mainly pHi, whereas electrodes are sensitive mainly to pHe. Subsequent NMR measurements with extracellular probes such as 3-aminopropylphosphonate confirmed low pHe (Gillies et  al. 1994). Therefore, the major difference between cancer and normal tissue is that the former has lower pHe, typically around 6.9 (Vaupel et al. 1990), but as low as 6.0 in some reports (Gillies et al. 2002). Tumors, like normal tissue, require modestly alkaline pHi to support growth, as shown by in vitro (Park et al. 1999; Chiche et  al. 2009) and in vivo (Chiche et  al. 2009) experiments. The maintenance of high pHi at low pHe requires efficient acid extrusion, a phenotype that is now considered characteristic of cancer cells.

Regulation of Tumor pH by Membrane Transport There are four challenges to efficient pH–i regulation in tumors. Firstly, the demand placed on pHi - regulating mechanisms is considerable. Tumors respire glucose at 0.1–1.0 mM/min (Vaupel et al. 1989) and generate acid equivalents at a rate that could promptly disturb resting free [H+] (~0.06 mM at pHi 7.2). Secondly, the transmembrane efflux of CO2 and lactic acid is limited by their membrane permeability and intracellular dissociation into membrane-impermeant ions (H+, HCO3–, lactate). Of the two acid products, lactic acid is less readily excreted than CO2 because (a) lactic acid is a larger and more polar molecule than CO2, and therefore is less able to cross the membrane’s lipid bilayer, (b) H+ and lactate produced anaerobically stay 99.9% dissociated in cytoplasm because of the high acid dissociation constant of lactic acid (10–4 M), whereas CO2 dissociates less strongly, with a ~90% yield at equilibrium (at pH 7; CO2 dissociation constant of 10–6.2 M). For every molecule of weak acid produced, anaerobic respiration leads to more intracellular H+ ion trapping than aerobic respiration, and therefore places a greater demand for H+ extruding mechanisms. However, intracellular conversion of CO2 into H+ and HCO3– can be catalyzed by intracellular isoforms of CA, a ubiquitously expressed enzyme (Swietach et al. 2007; Maren 1967; Supuran 2008). In a diffusion-reaction steady-state, intracellular hydration of CO2 reduces the capacity of its diffusive flux out of cells.

2  The Tumor Microenvironment

29

Thirdly, the share of cellular ATP for active acid extrusion is limited and cannot impinge on ATP reserves for growth and development. This may be particularly rate limiting in tumors that rely on energy inefficient anaerobic respiration. Finally, acid extruding mechanisms must be adequately active at low pHe. Extracellular H+ ions have been shown to have an inhibitory effect on transporters such as Na+/H+ exchange (Vaughan-Jones and Wu 1990; Aronson 1985). This would significantly reduce the capacity for acid extrusion in tumors with low pHe. To compensate for this, cancer cells may express a higher density of acid/base transporters, or express isoforms that are less susceptible to inhibition by low pHe.

Efflux of Metabolic Acid Figure 2.3 illustrates pathways for pHi regulation in tumor cells. Without specific mechanisms for transporting acid equivalents, cells would have to rely on the passive efflux of their acid products. CO2 and, to a lesser extent, lactic acid can cross the lipid bilayer, but efficient efflux requires a facilitated permeation pathway. Such

Fig. 2.3  The role of membrane transporters in facilitating acid efflux. Aerobic (left) and anaerobic (right) respiration of glucose yield the acid products CO2 and lactic acid, respectively. CO2 and, to a lesser degree, lactic acid are membrane-permeant and can exit across the surface membrane. At near-neutral pH, however, a fraction of CO2 and lactic acid will ionize. To control intracellular pH, cells express several transporters to facilitate transmembrane acid removal: AQP, aquaporins; MCT, monocarboxylic acid transport; HE, H+-ion extrusion; BT, HCO3– transport

30

P. Swietach et al.

facilitated transport can occur through aquaporin gas channels for CO2 (Nakhoul et  al. 1998) (Fig. 2.3a) and by monocarboxylic acid transporters (MCT) for H++lactate (equivalent to lactic acid) co-transport (Halestrap and Price 1999) (Fig. 2.3b). Isoforms MCT1 and MCT4 have been identified in cancer. Furthermore, hypoxia induces MCT4 expression to match the HIF-induced rise in lactic acid production (Ullah et al. 2006).

Buffering of H+ Ions Like all weak acids, lactic acid and hydrated CO2 undergo deprotonation in aqueous medium. The magnitude of the intracellular [H+] rise can be reduced by means of pH buffering, for example: Buffer + H + ↔ HBuffer + Cells have high intracellular buffering capacity because of the proteins, peptides, amino acids, and other weak acids/bases that they harbor. On their own, buffers offer limited protection from acid loading as they can minimize, but cannot remove, pHi disturbances. Typically, it takes many mmol/L of acid to reduce pHi by one unit (Swietach et al. 2008, 2009). However, chronic acid loading over the lifetime of a cell would eventually saturate buffer sites and drive pHi into the lethal range. Cells can counteract buffer saturation by keeping the concentration of protonated and unprotonated buffer constant. In a so-called open buffer system, the concentration of either form of buffer can be restored by transport to or from the extracellular space. The major “open” buffer system in cells is CO2/HCO3–. Intracellular H+ ions react with HCO3– forming membrane permeant CO2 that can exit cells freely. Intracellular buffering is then restored by HCO32 uptake, e.g. by means of Na+-dependent Cl–/HCO3– exchange (Lee and Tannock 1998) (Fig. 2.3c). The effectiveness of this mechanism at protecting pHi depends critically on the availability of extracellular HCO3– for cellular uptake and on efficient CO2 venting. The extracellular space of most tissues is often assumed to be an infinite compartment with solutes at well-controlled concentrations. However, in tumors with poor blood flow, the extracellular milieu may not provide adequate HCO3– (particularly at low pHe) to perpetuate pH buffering by CO2/HCO3–. Under those circumstances, excess H+ ions may have to be extruded by HCO3–-independent mechanisms.

Extrusion of H+ Ions Excess H+ ions can be removed from cells by means of membrane-bound transporter proteins. High levels of Na+/H+ exchangers (NHE) (Swietach et al. 2008;

2  The Tumor Microenvironment

31

Lee and Tannock 1998; McLean et  al. 2000) and vacuolar-type H+ ATPase (V-ATPase) pumps (Martinez-Zaguilan et  al. 1993) have been detected in many types of cancer cell (Fig. 2.3d). Fibroblasts transfected with V-ATPase (Gillies et  al. 1990) or NHE (Reshkin et  al. 2000) become tumorigenic, suggesting an important role for H+ extruders in developing the cancer phenotype. In breast cancer, expression levels of V-ATPase are high and in excess of NHE, and correlate with metastatic potential (Sennoune et al. 2004). Experimental evidence supports the notion that H+ extruders contribute to extracellular acidification and help to keep pHi in a range that is favorable for growth (Sennoune et al. 2004; Fais et al. 2007). The enhanced capacity for secreting acid into the extracellular space may be a means of autocrine/paracrine signaling, in which cell proliferation is triggered via extracellularly activated G-protein coupled receptors sensitive to H+ (Huang et al. 2008). It is noteworthy that H+ extrusion and buffering do not remove the anionic conjugate base of the metabolic acid (lactate and HCO3–); therefore, additional mechanisms are required to remove or metabolize these. HCO3– efflux in exchange for extracellular Cl– can occur by means of transporters belonging to the SLC4 (Alper 2006) and SLC26 (Mount and Romero 2004) families. Recently described pathways for organic anion transport on SLC26 (Mount and Romero 2004) and SLC21 (Hagenbuch and Meier 2004) carriers may offer non-H+-coupled means of transporting lactate or its catabolic derivatives.

Regulation of pH Using Nonrespiratory Sources of H+ Ions In the discussion so far, the substrate for transmembrane acid flux was the acid generated by cellular respiration. In metabolically active cancer cells, this is the most substantial source of acid. However, the finding that solid tumors with genetically attenuated glycolysis are still able to acidify their extracellular space suggests that cancer cells can excrete acid from intracellular H+ donors other than metabolically generated acid (Newell et al. 1993; Yamagata et al. 1998). Intracellular buffers provide a substantial pool of H+ ions that can be used to fine-tune pHi independently of the cell’s metabolic activity. Without this reserve of H+ ions, the only means of fine-tuning pHi would be to adjust the balance between metabolic acid production and its transmembrane removal. If, for instance, the production and excretion of metabolic acid is perfectly matched, but a rise in pHi is desired, then the cell must extrude its reserve H+ ions. NHE is a major candidate for finetuning pHi, because of the plethora of signaling molecules that target its activity (Vaughan-Jones and Swietach 2008). Experiments have shown that NHE-driven intracellular alkalinization triggers early stages of neoplastic transformation (Reshkin et  al. 2000), and is concomitant with cell cycle progression (Cardone et al. 2005).

32

P. Swietach et al.

Role of Carbonic Anhydrase in Acid-Equivalent Transport Intracellular and Extracellular Carbonic Anhydrase Isoforms Several steps in the pathways illustrated in Fig. 2.3 depend on the forward and backward rate constants in the equilibrium: CO2 + H 2 O ↔ HCO3− + H + The kinetics of pH buffering by CO2/HCO3– and CO2 hydration (and the reverse reaction) depend critically on CA expression levels. The first description of CA activity was based on the intracellular compartment of red blood cells (Meldrum and Roughton 1933). The high intracellular CA (CAi) activity in these cells is now considered to be an exception rather than rule, as typical CAi levels in most cells accelerate CO2 hydration by several, rather than several thousand, fold (Swietach et al. 2007, 2008, 2009; Leem and Vaughan-Jones 1998). The moderate CA catalysis inside cells is a compromise that (a) supports adequate intracellular CO2/HCO3– buffering kinetics (e.g. for the reaction between intracellular HCO3– and lactic acid, Fig. 2.3c) and (b) prevents excessive hydration of cell-generated CO2 (e.g. from aerobic production, Fig. 2.3a) that would trap H+ and HCO3– inside cells. Known catalytically-active intracellular CA isoforms include CA1, CA2, CA3, CA7, and CA13 (Maren 1967; Supuran 2008). Recently, a number of membrane-bound CA isoforms with extracellular-facing catalytic sites (CAe) have been described, including CA4, CA9, CA12, and CA14 (Maren 1967; Supuran 2008; Geers and Gros 2000; Pastorek et al. 1994). Two of these, CA9 and CA12, are hypoxically induced (Wykoff et  al. 2000; Loncaster et al. 2001; Airley et al. 2003) and correlate positively with cancer aggressiveness and poor prognosis (Haapasalo et  al. 2006; Driessen et  al. 2006; Beasley et  al. 2001; Chia et al. 2001; Koukourakis et al. 2001), although CA12 is also induced by estrogen in breast cancer and is associated then with a good outcome. It has been postulated that CAe activity in tumors accelerates acid extrusion (Swietach et  al. 2007, 2008, 2009; Chiche et al. 2009; Svastova et al. 2004). Figure 2.4 illustrates the possible mechanisms by which CAe activity achieves such facilitation.

Facilitated CO2 Diffusion Efficient CO2 efflux from aerobically respiring cells (Fig. 2.3a) can be curtailed by CO2 buildup in poorly perfused extracellular spaces. The outward transmembrane [CO2] gradient can be increased by hydrating extracellular CO2 into HCO3– and H+ (Swietach et al. 2009; Svastova et al. 2004) (Fig. 2.4a). Theoretical modeling predicts that optimal CO2 venting out of cells can be achieved by favoring extracellular over intracellular CO2 hydration (Swietach et al. 2009). This can be implemented by expressing CAe at a higher level than CAi. In multicellular spheroid models,

2  The Tumor Microenvironment

33

expression of CA9 activity has been shown to reduce intracellular acidity (Swietach et al. 2008, 2009) and increase extracellular acidity (Swietach et al. 2009) in a manner that is sensitive to CAe inhibitors. There is evidence, at least from some cell lines, that overexpression of CA9 decreases overall CAi activity (Swietach et  al. 2008, 2009), thereby shifting the site of CO2 hydration from the intracellular to the extracellular compartment. This CAe catalyzed pathway is an example of facilitated CO2 diffusion, and results in a fall of pHe. An apparent contradiction may be inferred from the role of an hypoxia-inducible enzyme (CA9, CA12) in facilitating the efflux of aerobically generated CO2 (Fig. 2.4a). However, the CA9 gene is among the most hypoxia sensitive (Wykoff et al. 2000). Along a radial gradient of decreasing oxygen, CA9 levels are likely to rise ahead of the downregulation of aerobic respiration. The mechanism illustrated in Fig. 2.4a may operate in the intermediate zone, where hypoxia is sufficient to induce CAe but inadequate to block aerobic respiration. In more proximal, oxygenated regions, CAe activity becomes redundant due to the small cell-to-capillary distance for CO2 diffusion (Swietach et al. 2009). CAe-facilitated CO2 diffusion can also take in anaerobically respiring cells. The substrate for CAe catalysis could be CO2 generated by the pentose phosphate

Fig. 2.4  The role of extracellular carbonic anhydrase (CAe) in facilitating efflux of aerobic (left) and anaerobic (right) acid-products. (a) Aerobically generated CO2 crosses the membrane passively down a diffusion gradient, maintained in the outward direction by CAe-catalyzed hydration of excreted CO2. (b) Anaerobically generated lactic acid (HLac), buffered by intracellular HCO3–, produces CO2. The CO2 thus generated exits the cells down a diffusion gradient maintained in the outward direction by CAe-catalyzed hydration. HCO3– generated extracellularly is transported into the cell by BT (e.g. Na+–HCO3– cotransport) to complete the cycle. (c) The products of intracellular CO2 hydration, H+ and HCO3–, are transported out of cells by H+ extruders (HE, e.g. Na+/H+ exchange) and bicarbonate transporters (BT, e.g. Cl–/HCO3– exchange). CAe-catalyzed buffering of H+ ions with HCO3– minimizes possible inhibition of HE by low pHe. (d) HLac flux on monocarboxylic acid transporters (MCT) is maintained in the outward direction by CAe-catalyzed titration of HLac and HCO3–. Water omitted for clarity. Pathways in the upper half tend to decrease pHe, and those in the lower half tend to increase pHe. All pathways stimulate acid efflux and raise pHi

34

P. Swietach et al.

shunt. A more important source of CO2 is intracellular buffering of lactic acid by HCO3–, a reaction catalyzed by CAi, if present (Figs. 2.3c and 2.4b). In addition to supporting an outward [CO2] gradient, CAe activity also generates extracellular HCO3– that is then taken up by cells to replenish intracellular CO2/HCO3– buffering capacity.

Facilitated H+ Diffusion A fraction of aerobically generated CO2 will undergo intracellular hydration, particularly in the presence of CAi activity. Extrusion of the resulting H+ ions will then reduce pHe, which may feedback negatively on transport activity, e.g. by means of allosteric auto-inhibition, as shown for NHE (Vaughan-Jones and Wu 1990; Aronson 1985). CAe activity can support H+ - transport by catalyzing the buffering of cell-extruded H+ ions with extracellular HCO3– (Fig. 2.4c). As a consequence of such facilitated H+ diffusion, pHe increases. In anaerobically respiring cells, the yield of intracellular lactic acid titration can be compromised, particularly under conditions of limited HCO3– uptake, e.g. inadequate transport activity or low extracellular substrate. Intracellular lactic acid that does not react with HCO3– can be extruded by MCT as H++lactate (Fig. 2.3b). CAe activity can stimulate this pathway as well, again by titrating the extruded acid with extracellular HCO3–. This converts lactic acid into CO2, a more mobile and less ionizing acid. Consequently, pHe increases and maintains a steeper outward gradient for H+-lactate co-transport (Fig. 2.4d). This pathway is another example of facilitated H+ diffusion.

The Transport Metabolon In the aforementioned pathways, CA activity works in functional association with membrane transporters. This functional coupling has been dubbed the “transport metabolon” and some reports have also proposed a structural basis for the CA-transporter interaction (Sterling et al. 2001; Becker et al. 2005). The transport metabolon was originally proposed for the collaboration between CAi isoforms and Cl−/HCO3− exchange (Sterling et  al. 2001), but there is now evidence for a similar interaction between CAe isoforms and HCO3− transporters (Alvarez et al. 2003; Morgan et al. 2007). As membrane transport is a diffusion-reaction process, the transport metabolon ensures that, through the provision and consumption of transport substrate, the membrane-translocation step is not rate-limited by the cis supply and trans removal of solute. The functional significance of the transport metabolon has, however, been questioned (Lu et al. 2006), particularly under conditions of freely available and equilibrated extracellular CO2/HCO3− buffer (Swietach et al. 2008).

2  The Tumor Microenvironment

35

The Dominant Role for Carbonic Anhydrase in Tumors The balance between the four CAe-assisted pHi-regulatory pathways illustrated in Fig. 2.4 will depend on (a) the dominant mode of cellular respiration, (b) the expression and activity of transporter proteins, and (c) the availability of substrate, such as extracellular HCO3– (Swietach et al. 2009). In hypoxic regions, and in cancer cells showing a pronounced Warburg effect, lactic acid production dominates. Elsewhere, in regions of tumors with functional mitochondria and adequate oxygenation, aerobic respiration to CO2 will form the bulk of acid production. However, the transmembrane flux of CO2 versus H+ (lactic acid) may not follow the pattern of aerobic versus anaerobic respiration. In several cancer cell lines, HCO3–dependent transport has been shown to dominate pHi regulation over HCO3–independent fluxes, such as NHE (Swietach et  al. 2008, 2009; Lee and Tannock 1998). This provides the transport-protein machinery to support intracellular lactic acid titration with HCO3–, and hence transmembrane CO2 efflux in glycolytic cells. In the corresponding tumors, the source of CO2 for transmembrane efflux would vary with distance from blood vessels, from aerobic CO2 production to titration of lactic acid by HCO3–. A high overall CO2 secretion rate is in agreement with in vivo data showing high extracellular CO2 levels in rodent solid tumors (Gullino et  al. 1965) and that 70% of acid output in colon cancers is in the form of CO2 (Griffiths et al. 2001; Holm et al. 1995). CAe-assisted CO2 versus H+ diffusion can be distinguished on the basis of the effect of CAe inhibitors on pHe. In spheroids grown from CA9-transfected cells, CAe inhibitors increase pHe (Swietach et  al. 2009). This suggests that the major effect of CAe catalysis in this preparation is to facilitate CO2 diffusion (Fig. 2.4a, b). The balance between CAe-assisted CO2 versus H+ diffusion in cancers in vivo awaits such pharmacological investigation. A dominance of the former would offer an elegant mechanistic link between two well-established correlations, that between tumor aggressiveness and low pHe (Vaupel et al. 1989; Gatenby et al. 2006; Gillies et al. 2002) and between tumor aggressiveness and the expression of CA9 and/or CA12 (Chia et  al. 2001; Generali et  al. 2006; Giatromanolaki et  al. 2001; Koukourakis et al. 2006).

Future Directions and Outlook for Therapy Acidification of the extracellular milieu, in combination with powerful mechanisms to protect pHi, are believed to be a central hallmark of the aggressive cancer phenotype. Restoration of alkaline pHe is predicted to remove the survival advantage that cancer cells have over normal cells and to discourage invasive behavior and progression towards malignant forms of the cancer (Fang et al. 2008). In addition, raising pHe reduces ionization of weak basic drugs used in chemotherapy, such as epirubicin (Generali et al. 2006), and thereby increases drug penetrability to intracellular targets (Raghunand et al. 2001).

36

P. Swietach et al.

The use of bicarbonate salts to alkalinize blood, with the anticipated rise in tumor pHe, has been proposed as a means of tackling cancer (Raghunand et  al. 2001). A more selective and specific therapy proposes pH-regulating transporters and enzymes as targets. Inhibition of these proteins has been proposed to be cancer’s Achilles heel (Fang et  al. 2008). Recently, three targets for inhibition have been postulated as plausible therapeutic approaches to treating cancer: (a) H+ pumps, (b) monocarboxylic acid (H+-lactate) transporters, and (c) extracellular CA isoforms. Inhibition of V-ATPase activity using interference RNA reduces metastasis (Fais et  al. 2007), and its pharmacological inhibition (e.g. with omeprazole) has been shown to improve responses to chemotherapy in solid tumors (Luciani et al. 2004) and promote death of B-cell tumors (De Milito et al. 2007). Fibroblasts transfected with the E7 oncogene failed to undergo intracellular alkalinization and neoplastic transformation when incubated with the NHE inhibitor, 5-(N,N-dimethyl) amiloride (Reshkin et al. 2000). Inhibition of MCT by cinnamate derivatives and, putatively, by lonidamine leads to cellular lactic acid accumulation and intracellular acidification, which are shown to be lethal in neuroblastoma cells (Fang et  al. 2006). Although the efficacy of inhibiting V-ATPase, NHE, and MCT has been demonstrated in some in vitro cell lines and in vivo animal tumor models, the use of these drugs in anticancer therapy is not without drawbacks. Cancer cells tend to overexpress isoforms of V-ATPase, NHE, and MCT, but many noncancer cells also rely on these transporters for survival. The lack of a basis for tumor specificity would trigger drug toxicity and uncalibrated effects on whole-body physiology. Before experimental evidence for tumor-associated CAe isoforms was produced, the broad spectrum CA inhibitor acetazolamide has been shown to reduce in vivo tumor growth (Teicher et al. 1993). Following the discovery of CA9, the sulfonamide CA9-inhibitor indisulam was demonstrated to have powerful anticancer effects (Owa et  al. 1999). The tumor-associated CAe isoforms, CA9 and CA12, offer an inhibitory target that is almost exclusive to cancer (with the exception of normal gastric mucosa). Membrane-impermeant, CA9- and CA12-selective inhibitors would target tumors, with minimal predicted side effects on other tissues. An impressive range of membrane-impermeant CAe inhibitors has been synthesized, and the effectiveness of these as anticancer drugs is currently being researched (Supuran 2008). Prodrugs activated by the extracellular acid environment would be another selective approach. Other therapeutic strategies based on tumor-specific CAe isoforms may include inhibitory antibodies, vaccines to develop T-cell antiCA9 or anti-CA12 responses, and targeted antibody therapy. Regulation of tumor pHi remains an intensively researched and hotly debated topic. Furthering our knowledge about the special adaptations found in tumors should improve our overall understanding of pH homeostasis, and may reveal more efficacious and safer anticancer therapy. Acknowledgments  This work has been supported by the Medical Research Council and Royal Society (PS), Cancer Research UK (ALH) and British Heart Foundation (RDV-J).

2  The Tumor Microenvironment

37

References Airley RE, Loncaster J, Raleigh JA, Harris AL, Davidson SE, Hunter RD, West CM, Stratford IJ (2003) GLUT-1 and CAIX as intrinsic markers of hypoxia in carcinoma of the cervix: relationship to pimonidazole binding. Int J Cancer 104:85–91. Alper SL (2006) Molecular physiology of SLC4 anion exchangers. Exp Physiol 91:153–161. Alvarez BV, Loiselle FB, Supuran CT, Schwartz GJ, Casey JR (2003) Direct extracellular interaction between carbonic anhydrase IV and the human NBC1 sodium/bicarbonate co-transporter. Biochemistry 42:12321–12329. Aronson PS (1985) Kinetic properties of the plasma membrane Na+-H+ exchanger. Annu Rev Physiol 47:545–560. Beasley NJ, Wykoff CC, Watson PH, Leek R, Turley H, Gatter K, Pastorek J, Cox GJ, Ratcliffe P, Harris AL (2001) Carbonic anhydrase IX, an endogenous hypoxia marker, expression in head and neck squamous cell carcinoma and its relationship to hypoxia, necrosis, and microvessel density. Cancer Res 61:5262–5267. Becker HM, Hirnet DC, Fecher-Trost, Sultemeyer D, Deitmer JW (2005) Transport activity of MCT1 expressed in Xenopus oocytes is increased by interaction with carbonic anhydrase. J Biol Chem 280:39882–39889. Boron WF (2004) Regulation of intracellular pH. Adv Physiol Educ 28:160–179. Cardone RA, Casavola V, Reshkin SJ (2005) The role of disturbed pH dynamics and the Na+/H+ exchanger in metastasis. Nat Rev Cancer 5:786–795. Cerdan S, Rodrigues TB, Sierra A, Benito M, Fonseca LL, Fonseca CP, Garcia-Martin ML (2006) The redox switch/redox coupling hypothesis. Neurochem Int 48:523–530. Chia SK, Wykoff CC, Watson PH, Han C, Leek RD, Pastorek J, Gatter KC, Ratcliffe P, Harris AL (2001) Prognostic significance of a novel hypoxia-regulated marker, carbonic anhydrase ix, in invasive breast carcinoma. J Clin Oncol 19:3660–3668. Chiche J, Ilc K, Laferriere J, Trottier E, Dayan F, Mazure NM, Brahimi-Horn MC, Pouyssegur J (2009) Hypoxia-inducible carbonic anhydrase IX and XII promote tumor cell growth by counteracting acidosis through the regulation of the intracellular pH. Cancer Res 69: 358–368. De Milito A, Iessi E, Logozzi M, Lozupone F, Spada M, Marino ML, Federici C, Perdicchio M, Matarrese P, Lugini L, Nilsson A, Fais S (2007) Proton pump inhibitors induce apoptosis of human B-cell tumors through a caspase-independent mechanism involving reactive oxygen species. Cancer Res 67:5408–5417. Dewhirst MW, Secomb TW, Ong ET, Hsu R, Gross JF (1994) Determination of local oxygen consumption rates in tumors. Cancer Res 54:3333–3336. Driessen, A., Landuyt W, Pastorekova S, Moons J, Goethals L, Haustermans K, Nafteux P, Penninckx F, Geboes K, Lerut T, Ectors N (2006) Expression of carbonic anhydrase IX (CA IX), a hypoxia-related protein, rather than vascular-endothelial growth factor (VEGF), a pro-angiogenic factor, correlates with an extremely poor prognosis in esophageal and gastric adenocarcinomas. Ann Surg 243:334–340. Fais S, De Milito A, You H, Qin W (2007) Targeting vacuolar H+-ATPases as a new strategy against cancer. Cancer Res 67:10627–10630. Fang J, Quinones QJ, Holman TL, Morowitz MJ, Wang Q, Zhao H, Sivo F, Maris JM, Wahl ML (2006) The H+-linked monocarboxylate transporter (MCT1/SLC16A1): a potential therapeutic target for high-risk neuroblastoma. Mol Pharmacol 70:2108–2115. Fang JS, Gillies RD, Gatenby RA (2008) Adaptation to hypoxia and acidosis in carcinogenesis and tumor progression. Semin Cancer Biol 18:330–337. Gambhir SS (2002) Molecular imaging of cancer with positron emission tomography. Nat Rev Cancer 2:683–693. Gatenby RA, Gawlinski ET (1996) A reaction-diffusion model of cancer invasion. Cancer Res 56:5745–5753.

38

P. Swietach et al.

Gatenby RA, Gillies RJ (2004) Why do cancers have high aerobic glycolysis? Nat Rev Cancer 4:891–899. Gatenby RA, Gawlinski ET, Gmitro AF, Kaylor B, Gillies RJ (2006) Acid-mediated tumor invasion: a multidisciplinary study. Cancer Res 66:5216–5223. Gatenby RA, Smallbone K, Maini PK, Rose F, Averill J, Nagle RB, Worrall L, Gillies RJ (2007) Cellular adaptations to hypoxia and acidosis during somatic evolution of breast cancer. Br J Cancer 97:646–653. Geers C, Gros G (2000) Carbon dioxide transport and carbonic anhydrase in blood and muscle. Physiol Rev 80:681–715. Generali D, Fox SB, Berruti A, Brizzi MP, Campo L, Bonardi S, Wigfield SM, Bruzzi P, Bersiga A, Allevi G, Milani M, Aguggini S, Dogliotti L, Bottini A, Harris AL (2006) Role of carbonic anhydrase IX expression in prediction of the efficacy and outcome of primary epirubicin/ tamoxifen therapy for breast cancer. Endocr Relat Cancer 13:921–930. Giatromanolaki A, Koukourakis MI, Sivridis E, Pastorek J, Wykoff CC, Gatter KC, Harris AL (2001) Expression of hypoxia-inducible carbonic anhydrase-9 relates to angiogenic pathways and independently to poor outcome in non-small cell lung cancer. Cancer Res 61:7992–7998. Gillies RJ, Liu Z, Bhujwalla Z (1994) 31P-MRS measurements of extracellular pH of tumors using 3-aminopropylphosphonate. Am J Physiol 267:C195–203. Gillies RJ, Martinez-Zaguilan R, Martinez GM, Serrano R, Perona R (1990) Tumorigenic 3T3 cells maintain an alkaline intracellular pH under physiological conditions. Proc Natl Acad Sci USA 87:7414–7418. Gillies RJ, Raghunand N, Karczmar GS, Bhujwalla ZM (2002) MRI of the tumor microenvironment. J Magn Reson Imaging 16:430–450. Gillies RJ, Robey I, Gatenby RA (2008) Causes and consequences of increased glucose metabolism of cancers. J Nucl Med 49(Suppl 2):24S–42S. Griffiths JR, Stevens AN, Iles RA, Gordon RE, Shaw D (1981) 31P-NMR investigation of solid tumours in the living rat. Biosci Rep 1:319–325. Griffiths JR, McIntyre DJ, Howe FA, Stubbs M (2001) Why are cancers acidic? A carrier-mediated diffusion model for H+ transport in the interstitial fluid. Novartis Found Symp 240:46–62. Gullino PM, Grantham FH, Smith SH, Haggerty AC (1965) Modifications of the acid-base status of the internal milieu of tumors. J Natl Cancer Inst 34:857–869. Hagenbuch, B., Meier PJ (2004) Organic anion transporting polypeptides of the OATP/SLC21 family: phylogenetic classification as OATP/SLCO superfamily, new nomenclature and molecular/functional properties. Pflugers Arch 447:653–665. Halestrap AP, Price NT (1999) The proton-linked monocarboxylate transporter (MCT) family: structure, function and regulation. Biochem J 343(Pt 2):281–299. Hardie DG (2003) Minireview: the AMP-activated protein kinase cascade: the key sensor of cellular energy status. Endocrinology 144:5179–5183. Helmlinger G, Yuan F, Dellian M, Jain RK (1997) Interstitial pH and pO2 gradients in solid tumors in vivo: high-resolution measurements reveal a lack of correlation. Nat Med 3:177–182. Helmlinger G, Sckell A, Dellian M, Forbes NS, Jain RK (2002) Acid production in glycolysisimpaired tumors provides new insights into tumor metabolism. Clin Cancer Res 8:1284–1291. Haapasalo JA, Nordfors KM, Hilvo M, Rantala IJ, Soini Y, Parkkila AK, Pastorekova S, Pastorek J, Parkkila SM, Haapasalo HK (2006) Expression of carbonic anhydrase IX in astrocytic tumors predicts poor prognosis. Clin Cancer Res 12:473–477. Holm E, Hagmuller E, Staedt U, Schlickeiser G, Gunther HJ, Leweling H, Tokus M, Kollmar HB (1995) Substrate balances across colonic carcinomas in humans. Cancer Res 55:1373–1378. Huang WC, Swietach P, Vaughan-Jones RD, Ansorge O, Glitsch MD (2008) Extracellular acidification elicits spatially and temporally distinct Ca2+ signals. Curr Biol 18:781–785. Isfort RJ, Cody DB, Asquith TN, Ridder GM, Stuard SB, LeBoeuf RA (1993) Induction of protein phosphorylation, protein synthesis, immediate-early-gene expression and cellular proliferation by intracellular pH modulation. Implications for the role of hydrogen ions in signal transduction. Eur J Biochem 213:349–357.

2  The Tumor Microenvironment

39

Ivanov SV, Kuzmin I, Wei MH, Pack S, Geil L, Johnson BE, Stanbridge EJ, Lerman MI (1998) Down-regulation of transmembrane carbonic anhydrases in renal cell carcinoma cell lines by wild-type von Hippel-Lindau transgenes. Proc Natl Acad Sci USA 95:12596–12601. Kallinowski F, Schlenger KH, Runkel S, Kloes M, Stohrer M, Okunieff P, Vaupel P (1989) Blood flow, metabolism, cellular microenvironment, and growth rate of human tumor xenografts. Cancer Res 49:3759–3764. Kimura H, Braun RD, Ong ET, Hsu R, Secomb TW, Papahadjopoulos D, Hong K, Dewhirst MW (1996) Fluctuations in red cell flux in tumor microvessels can lead to transient hypoxia and reoxygenation in tumor parenchyma. Cancer Res 56:5522–5528. Koukourakis MI, Giatromanolaki A, Sivridis E, Simopoulos K, Pastorek J, Wykoff CC, Gatter KC, Harris AL (2001) Hypoxia-regulated carbonic anhydrase-9 (CA9) relates to poor vascularization and resistance of squamous cell head and neck cancer to chemoradiotherapy. Clin Cancer Res 7:3399–3403. Koukourakis MI, Pitiakoudis M, Giatromanolaki A, Tsarouha A, Polychronidis A, Sivridis E, Simopoulos C (2006) Oxygen and glucose consumption in gastrointestinal adenocarcinomas: correlation with markers of hypoxia, acidity and anaerobic glycolysis. Cancer Sci 97: 1056–1060. Krogh A (1919) The number and distribution of capillaries in muscles with calculations of the oxygen pressure head necessary for supplying the tissue. J Physiol 52:409–415. Lee AH, Tannock IF (1998) Heterogeneity of intracellular pH and of mechanisms that regulate intracellular pH in populations of cultured cells. Cancer Res 58:1901–1908. Leem CH, Vaughan-Jones RD (1998) Out-of-equilibrium pH transients in the guinea-pig ventricular myocyte. J Physiol 509 (Pt 2):471–485. Liao DJ, Dickson RB (2000) c-Myc in breast cancer. Endocr Relat Cancer 7:143–164. Loncaster JA, Harris AL, Davidson SE, Logue JP, Hunter RD, Wycoff CC, Pastorek J, Ratcliffe PJ, Stratford IJ, West CM (2001) Carbonic anhydrase (CA IX) expression, a potential new intrinsic marker of hypoxia: correlations with tumor oxygen measurements and prognosis in locally advanced carcinoma of the cervix. Cancer Res 61:6394–6399. Lu H, Forbes RA, Verma A (2002) Hypoxia-inducible factor 1 activation by aerobic glycolysis implicates the Warburg effect in carcinogenesis. J Biol Chem 277:23111–23115. Lu J, Daly CM, Parker MD, Gill HS, Piermarini PM, Pelletier MF, Boron WF (2006) Effect of human carbonic anhydrase II on the activity of the human electrogenic Na/HCO3 cotransporter NBCe1-A in Xenopus oocytes. J Biol Chem 281:19241–19250. Luciani F, Spada M, De Milito A, Molinari A, Rivoltini L, Montinaro A, Marra M, Lugini L, Logozzi M, Lozupone F, Federici C, Iessi E, Parmiani G, Arancia G, Belardelli F, Fais S (2004) Effect of proton pump inhibitor pretreatment on resistance of solid tumors to cytotoxic drugs. J Natl Cancer Inst 96:1702–1713. Ludwig MG, Vanek M, Guerini D, Gasser JA, Jones CE, Junker U, Hofstetter H, Wolf RM, Seuwen K (2003) Proton-sensing G-protein-coupled receptors. Nature 425:93–98. Maren TH (1967) Carbonic anhydrase: chemistry, physiology, and inhibition. Physiol Rev 47:595–781. Martinez-Zaguilan R, Lynch RM, Martinez GM, Gillies RJ (1993) Vacuolar-type H(+)-ATPases are functionally expressed in plasma membranes of human tumor cells. Am J Physiol 265:C1015–1029. Martinez-Zaguilan R, Seftor EA, Seftor RE, Chu YW, Gillies RJ, Hendrix MJ (1996) Acidic pH enhances the invasive behavior of human melanoma cells. Clin Exp Metastasis 14:176–186. McConkey DJ, Orrenius S (1996) Signal transduction pathways in apoptosis. Stem Cells 14:619–631. McLean LA, Roscoe J, Jorgensen NK, Gorin FA, Cala PM (2000) Malignant gliomas display altered pH regulation by NHE1 compared with nontransformed astrocytes. Am J Physiol Cell Physiol 278:C676–C688. Meldrum NU, Roughton FJ (1933) Carbonic anhydrase. Its preparation and properties. J Physiol 80:113–142. Morgan PE, Pastorekova S, Stuart-Tilley AK, Alper SL, Casey JR (2007) Interactions of transmembrane carbonic anhydrase, CAIX, with bicarbonate transporters.. Am J Physiol Cell Physiol. 293:C738–C748.

40

P. Swietach et al.

Mount DB, Romero MF (2004) The SLC26 gene family of multifunctional anion exchangers. Pflugers Arch 447:710–721. Murakami N, Yokomizo T, Okuno T, Shimizu T (2004) G2A is a proton-sensing G-protein-coupled receptor antagonized by lysophosphatidylcholine. J Biol Chem 279:42484–42491. Kato Y, Nakayama Y, Umeda M, Miyazaki K (1992) Induction of 103-kDa gelatinase/type IV collagenase by acidic culture conditions in mouse metastatic melanoma cell lines. J Biol Chem 267:11424–11430. Nakhoul NL, Davis BA, Romero MF, Boron WF (1998) Effect of expressing the water channel aquaporin-1 on the CO2 permeability of Xenopus oocytes. Am J Physiol 274:C543–C548. Newell K, Franchi A, Pouyssegur J, Tannock I (1993) Studies with glycolysis-deficient cells suggest that production of lactic acid is not the only cause of tumor acidity. Proc Natl Acad Sci USA 90:1127–1131. Owa T, Yoshino H, Okauchi T, Yoshimatsu K, Ozawa Y, Sugi NH, Nagasu T, Koyanagi N, Kitoh K (1999) Discovery of novel antitumor sulfonamides targeting G1 phase of the cell cycle. J Med Chem 42:3789–3799. Park HJ, Lyons JC, Ohtsubo T, Song CW (1999) Acidic environment causes apoptosis by increasing caspase activity. Br J Cancer 80:1892–1897. Pastorek J, Pastorekova S, Callebaut I, Mornon JP, Zelnik V, Opavsky R, Zat’ovicova M, Liao S, Portetelle D, Stanbridge EJ et al (1994) Cloning and characterization of MN, a human tumorassociated protein with a domain homologous to carbonic anhydrase and a putative helixloop-helix DNA binding segment. Oncogene 9:2877–2888. Ramaiah A, Hathaway JA, Atkinson DE (1964). Adenylate as a Metabolic Regulator. Effect on Yeast Phosphofructokinase Kinetics. J Biol Chem 239:3619–3622. Raghunand N, Mahoney B, van Sluis R, Baggett B, Gillies RJ (2001) Acute metabolic alkalosis enhances response of C3H mouse mammary tumors to the weak base mitoxantrone. Neoplasia 3:227–235. Reshkin SJ, Bellizzi A, Caldeira S, Albarani V, Malanchi I, Poignee M, Alunni-Fabbroni M, Casavola V, Tommasino M (2000) Na+/H+ exchanger-dependent intracellular alkalinization is an early event in malignant transformation and plays an essential role in the development of subsequent transformation-associated phenotypes. FASEB J 14:2185–2197. Roos A, Boron WF (1981) Intracellular pH. Physiol Rev 61:296–434. Schlappack OK, Zimmermann A, Hill RP (1991) Glucose starvation and acidosis: effect on experimental metastatic potential, DNA content and MTX resistance of murine tumour cells. Br J Cancer 64:663–670. Semenza GL (2003) Targeting HIF-1 for cancer therapy. Nat Rev Cancer 3:721–732. Semenza GL, Wang GL (1992) A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation. Mol Cell Biol 12:5447–5454. Semenza GL, Roth PH, Fang HM, Wang GL (1994) Transcriptional regulation of genes encoding glycolytic enzymes by hypoxia-inducible factor 1. J Biol Chem 269:23757–23763. Sennoune SR, Bakunts K, Martinez GM, Chua-Tuan JL, Kebir Y, Attaya MN, Martinez-Zaguilan R (2004) Vacuolar H+-ATPase in human breast cancer cells with distinct metastatic potential: distribution and functional activity. Am J Physiol Cell Physiol 286:C1443–C1452. Sterling D, Reithmeier RA, Casey JR (2001) A transport metabolon. Functional interaction of carbonic anhydrase II and chloride/bicarbonate exchangers. J Biol Chem 276:47886–47894. Stock C, Schwab A (2009) Protons make tumor cells move like clockwork. Pflugers Arch 458(2):981–992. Stubbs M, McSheehy PM, Griffiths JR, Bashford CL (2000) Causes and consequences of tumour acidity and implications for treatment. Mol Med Today 6:15–19. Supuran CT (2008) Carbonic anhydrases: novel therapeutic applications for inhibitors and activators. Nature Rev (Drug Discoveries) 7:1–14. Svastova E, Hulikova A, Rafajova M, Zat’ovicova M, Gibadulinova A, Casini A, Cecchi A, Scozzafava A, Supuran CT, Pastorek J, Pastorekova S (2004) Hypoxia activates the capacity of tumor-associated carbonic anhydrase IX to acidify extracellular pH. FEBS Lett 577:439–445.

2  The Tumor Microenvironment

41

Swietach P, Vaughan-Jones RD, Harris AL (2007) Regulation of tumor pH and the role of carbonic anhydrase 9. Cancer Metastasis Rev 26:299–310. Swietach P, Wigfield S, Cobden P, Supuran CT, Harris AL, Vaughan-Jones RD (2008) Tumorassociated carbonic anhydrase 9 spatially coordinates intracellular pH in three-dimensional multicellular growths. J Biol Chem 283:20473–20483. Swietach P, Patiar S, Supuran CT, Harris AL, Vaughan-Jones RD (2009) The role of carbonic anhydrase 9 in regulating extracellular and intracellular pH in three-dimensional tumor cell growths. J Biol Chem 284:20299–20310. Teicher BA, Liu SD, Liu JT, Holden SA, Herman TS (1993) A carbonic anhydrase inhibitor as a potential modulator of cancer therapies. Anticancer Res 13:1549–1556. Thomlinson RH, Gray LH (1955) The histological structure of some human lung cancers and the possible implications for radiotherapy. Br J Cancer 9:539–549. Ullah MS, Davies AJ, Halestrap AP (2006) The plasma membrane lactate transporter MCT4, but not MCT1, is up-regulated by hypoxia through a HIF-1alpha-dependent mechanism. J Biol Chem 281:9030–9037. Vaughan-Jones RD, Swietach P (2008) Pushing and pulling the cardiac sodium/hydrogen exchanger. Circ Res 103:773–775. Vaughan-Jones RD, Wu ML (1990) Extracellular H+ inactivation of Na(+)-H+ exchange in the sheep cardiac Purkinje fibre. J Physiol 428:441–466. Vaughan-Jones RD, Spitzer KW, Swietach P (2008) Intracellular pH regulation in heart. J Mol Cell Cardiol 46(3):318–313. Vaupel P, Kallinowski F, Okunieff P (1989) Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: a review. Cancer Res 49:6449–6465. Vaupel P, Kallinowski F, Okunieff P (1990) Blood flow, oxygen consumption and tissue oxygenation of human tumors. Adv Exp Med Biol 277:895–905. Warburg O (1930) The metabolism of tumours. Arnold Constable, London. Wike-Hooley JL, Haveman J, Reinhold HS (1984) The relevance of tumour pH to the treatment of malignant disease. Radiother Oncol 2:343–366. Wykoff CC, Beasley NJ, Watson PH, Turner KJ, Pastorek J, Sibtain A, Wilson GD, Turley H, Talks KL, Maxwell PH, Pugh CW, Ratcliffe PJ, Harris AL (2000) Hypoxia-inducible expression of tumor-associated carbonic anhydrases. Cancer Res 60:7075–7083. Yamagata M, Hasuda K, Stamato T, Tannock IF (1998) The contribution of lactic acid to acidification of tumours: studies of variant cells lacking lactate dehydrogenase. Br J Cancer 77:1726–1731.

Chapter 3

Hypoxia, Gene Expression, and Metastasis Olga V. Razorenova and Amato J. Giaccia

Abstract  Metastasis is the primary cause of death from cancer due to the spread of disease throughout the body. Increasing evidence suggests that the hypoxic microenvironment serves as a driving force for the metastatic process. Fifty to sixty percent of solid tumors contain hypoxic areas, where the gene expression is reprogrammed by low oxygen microenvironment leading to aggressive invasive cancer cell behavior. Hypoxia upregulates multiple genes involved in different steps of metastatic process, including angiogenesis, proliferation, migration, invasion, motility, adhesion, ECM remodeling, and survival. Moreover, hypoxia confers tumor cells with chemo- and radio-resistance. At the end of this chapter, we discuss the facts linking hypoxia and cancer stem cells (CSC) mainly through the ability of hypoxic microenvironment to shift cells toward the undifferentiated phenotype. Abbreviations AKT ARNT CC-RCC CNS CSC EMT FAK FIH GSC HIF HRE

protein kinase B arylhydrocarbon receptor nuclear t­ranslocator clear cell renal cell carcinoma central nervous system cancer stem cell epithelial–mesenchymal transition focal adhesion kinase factor inhibiting HIF glioma stem cell hypoxia-inducible factor hypoxia-responsive element

A.J. Giaccia (*) Division of Radiation and Cancer Biology, Stanford University, Stanford, CA 94305, USA e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_3, © Springer Science+Business Media, LLC 2010

43

44

IDH1 iPS a-KG LOH LOX MAPK NSC PHDs 1–3 PI-3K pVHL VHL WT

O.V. Razorenova and A.J. Giaccia

isocitrate dehydrogenase 1 induced pluripotent stem cell a-ketoglutarate loss of heterozygosity lysyl oxidase mitogen-activated protein kinase neural stem cell prolyl-4-hydroxylases 1–3 phosphoinositide 3 kinase von Hippel–Lindau protein von Hippel–Lindau gene wild-type

The Link Between Hypoxia and Metastasis Metastatic disease is the primary cause of death from cancer due to its spread beyond the primary affected organ to multiple organs in the body (secondary sites). The metastatic process is complex and can be divided into several major steps: (1) invasion and migration of tumor cells, leading to intravasation; (2) survival in the circulation (resistance to anoikis); (3) adhesion and migration, leading to extravasation; (4) survival in the new microenvironment; and (5) proliferation in the secondary site (Fig. 3.1) (Chan and Giaccia 2007; Nguyen et al. 2009). These ­processes require the involvement of a variety of effector molecules. It has been estimated that 50–60% of solid tumors contain areas of hypoxic and/or anoxic regions (Vaupel and Mayer 2007). Moreover, hypoxia correlates with poor patient outcome and the presence of metastases in multiple tumor types (Hockel et al. 1993, 1996; Brizel et al. 1996; Nordsmark et al. 1996; Fyles et al. 1998). Despite these correlations, it is not clear whether hypoxia provides the appropriate microenvironment promoting metastatic spread or metastasis is a simple consequence of chemo- and radio-resistance of cells in hypoxic regions. Hypoxic cells survive chemotherapy because many chemotherapeutic agents require proliferating cells to induce cytotoxicity and hypoxic cells divide slowly (Shannon et al. 2003). The second cause of failure is the need for vasculature to deliver drugs to cancer cells, which is absent or defective in hypoxic regions (Kizaka-Kondoh et  al. 2003). An additional contribution to chemotherapy resistance is through the hypoxic induction of multidrug transporters, leading to the transport of drugs out of the cell (Comerford et al. 2002; Wartenberg et al. 2003). Radiation therapy relies on molecular oxygen to form cytotoxic, double-strand breaks in the cancer cell DNA, which is limited under hypoxic conditions (Kallman and Dorie 1986; Hall 1994). The majority of studies now provide convincing evidence that hypoxia is not only capable to confer cells with chemo- and radio-resistance, but also shifts gene expression to promote the metastatic process (see below).

3  Hypoxia, Gene Expression, and Metastasis

hypoxia

45

intravasation angiogenic factor secretion

adh

esio

n

angiogenesis

extravasation

organ2

surv i

val

organ1

survival, proliferation

Fig. 3.1  Major steps of the metastatic process

Causes and Consequences of Tumor Hypoxia The physiologic range of oxygen for most tissues is between 4 and 8% oxygen (normoxia) in contrast to the 20% level of oxygen used for tissue culture experiments (Gardner and Corn 2008). For each tissue, the range of oxygen tensions can vary accordingly to need; thus, under physiologic conditions it is difficult to establish a universal value to define hypoxia. In this regard, the functional definition of hypoxia is the state when oxygen delivery does not meet the demands of the given tissue. Cellular hypoxia (<1–5% oxygen) is a common stress in normal development (Huang et al. 2004) and numerous pathological conditions, including cancer and ischemia (Giaccia et  al. 2003; Semenza 2003). Hypoxic areas in tumors develop as a result of an imbalance between oxygen supply and consumption (Papandreou et  al. 2005; Vaupel and Mayer 2007). The expanding tumor mass increases the distance of certain tumor areas from local blood vessels, limiting the oxygen supply. Additionally, at later stages of tumor growth, when a tumor establishes an extensive vasculature to increase oxygen supply, the defects in blood vessel structure contribute to hypoxia reoxygenation of certain tumor areas (Endrich et al. 1979; Grunt et al. 1985; Shah-Yukich and Nelson 1988; Dewhirst et al. 1989). The malformed tumor vasculature is twisted, tortuous with blind ends and can possess arterio-venous shunts. Moreover, tumor vessels are devoid of their normal multilayer complex structure and are leaky and predisposed to dilation and constriction, resulting in transient changes in tissue oxygenation (Carmeliet 2005). The main consequence of hypoxic exposure is the stabilization of hypoxiainducible factors (HIFs), which are critical regulators of gene transcription (Giaccia et al. 2003). The hypoxic microenvironment of the tumor leads to HIF stabilization, which induces gene expression changes in order to adapt to hypoxia (Semenza

46

O.V. Razorenova and A.J. Giaccia

2003; Lee et al. 2004); some of these changes, in turn, allow the acquisition of prometastatic properties of cells (invasiveness, migration, adhesion, and survival) (Giaccia et al. 2003; Semenza 2003). Indeed, HIF stabilization in solid tumors does not simply correlate with poor patient prognosis, but HIFs are active players in the metastatic process (Le et  al. 2004; Chan and Giaccia 2007; Rankin and Giaccia 2008; Qing and Simon 2009). There are three known HIF-a molecules (Rankin and Giaccia 2008): HIF-1a, HIF-2a, and HIF-3a, which are all hypoxia inducible with a similar mechanism of regulation by oxygen. HIF-1a is ubiquitously expressed, while HIF-2a expression is restricted to specific cell types: endothelial cells, glial cells, type II pneumocytes, cardiomyocytes, kidney fibroblasts, interstitial cells of the pancreas and duodenum, and hepatocytes (Wiesener et al. 2003). Our knowledge of HIF-3a is currently very limited and some studies suggest that it might act as a dominantnegative regulator of HIF-1a and HIF-2a by generation of an alternatively spliced isoform (Makino et al. 2001). Both HIF-1a and HIF-2a are detected in a variety of human tumors, and elevated HIF-1a expression correlates with poor patient outcome in head and neck cancer, nasopharyngeal carcinoma, colorectal, pancreatic, breast, cervical, osteosarcoma, endometrial, ovarian, bladder, glioblastoma, and gastric carcinomas, while elevated HIF-2a expression correlates with poor patient outcome in hepatocellular, colorectal carcinoma, melanoma, ovarian, and nonsmall cell lung cancers (Rankin and Giaccia 2008). Hypoxic conditions are stress conditions that trigger the cell to make a decision to live or to die. In response to hypoxia, as a pro-survival signal, HIF-a stability and activity increases (Giaccia et al. 2003; Semenza 2003). HIF-mediated gene expression allows cells to respond to a low oxygen environment by increasing oxygen delivery and metabolically to adapt to decreased oxygen availability. HIFs regulate expression of multiple target genes that are involved in glycolysis, angiogenesis, proliferation, migration, etc., by binding to hypoxia-responsive elements (HREs) (5¢RCGTG3¢) in their genomic regulatory regions (Semenza 2003; Le et al. 2004; Chan and Giaccia 2007).

HIF Regulation by Oxygen In untransformed cells, HIF-a activity is oxygen dependent and is tightly regulated (Fig. 3.2). Under normoxia, HIF-1a is hydroxylated on two conserved proline residues, ­prolines 402 and 564, by prolyl-4-hydroxylases 1–3 (PHDs 1–3) (Ivan et al. 2001; Jaakkola et  al. 2001). This hydroxylation reaction requires oxygen and a-ketoglutarate (a-KG) as substrates, and iron and ascorbate as cofactors. When hydroxylated, HIF-1a is ­recognized by the pVHL protein, which allows the formation of a complex with Elongins B and C, cullin 2, and RING-H2 protein ring-box1. This complex functions as E3 ubiquitin ligase, and ubiquitinated HIF-1a gets degraded in the proteasome (Pause et al. 1999). Under hypoxic conditions, hydroxylation of HIF-1a decreases in an oxygen-dependent manner as oxygen serves as a substrate in the hydroxylation reaction. HIF-1a without proline hydroxyl marks is

3  Hypoxia, Gene Expression, and Metastasis

Normoxia Presence of oxygen Cofactors: ascorbate Fe2+

Hypoxia Absence of oxygen pVHL

pVHL PHDs

HIFα

47

OH

O2 α-KG

OH

Ub Ub Ub

HIFα

HIFα

FIH

p300/ CBP

p300/ CBP

OH

26S

I

F

α

HIFα HIFβ

transcription of HIF target genes

5’RCGTG 3’YGCAC

metabolism angiogenesis invasion migration proliferation

H

Fig. 3.2  Oxygen-dependent HIF regulation. R, purine; Y, pyrimidine

not recognized by pVHL, becomes stabilized, and dimerizes with its constitutively expressed partner HIF-1b (also known as arylhydrocarbon receptor nuclear translocator, ARNT). HIF-a/HIF-1b; heterodimer acts in the nucleus and activates transcription by recruiting the transcriptional co-activators such as p300/CBP (Giaccia et al. 2003; Semenza 2003). The interaction between HIF and p300/CBP is also oxygen dependent and is ­regulated by factor inhibiting HIF (FIH) (Mahon et al. 2001). Under normoxia, FIH hydroxylates asparagine 803 in HIF-1a transactivation domain and prevents p300/CBP binding. Thus, HIF activation under hypoxia consists of HIF-a stabilization and effective p300/CBP binding.

HIF Regulation by Genetic Alterations of Upstream Regulators Many studies have shown that hypoxia is not the only mechanism of HIF activation. HIF-a stability and activity are tightly controlled in primary cells by multiple signaling pathways. Not surprisingly cancer cells acquire mutations in genes, involved

48

O.V. Razorenova and A.J. Giaccia

into HIF regulation, allowing its activation in tumors under normoxic conditions. As described above, pVHL plays a central role in regulating HIF stability (Fig. 3.2). Genetic inactivation of the VHL gene results in HIF-a stabilization and activation, leading to upregulation of target gene expression under normoxia (Iliopoulos et al. 1996; Maxwell et al. 1999; Giaccia et al. 2003; Kaelin 2008). Germ-line mutations in VHL result in VHL disease, also known as familial Von Hippel–Lindau syndrome (Latif et al. 1993). Affected individuals are born with one wild-type (WT) allele and another mutant allele. In these individuals the remaining, WT VHL allele, is eliminated by a loss of heterozygosity (LOH). Von Hippel–Lindau syndrome is characterized by the high frequency of development of several cancer types, including central nervous system (CNS) and retinal hemangioblastomas, clear-cell renal cell carcinomas (CC-RCCs), and pheochromocytomas. VHL is also inactivated in the majority of sporadic RCCs and hemangioblastomas (up to 80% of cases) (Gnarra et al. 1993; Sprenger et al. 2001; Maynard and Ohh 2004). Families at risk for developing pheochromocytoma almost invariably harbor VHL missense ­mutations, in contrast to families without pheochromocytomas, which frequently harbor VHL deletions or truncation mutations (Chen et al. 1995; Zbar et al. 1996). Most likely genotype–phenotype correlations in VHL disease reflect the degree of impairment of pVHL function by different mutations (Hoffman et al. 2001; Li et al. 2007). Thus, VHL is a bona fide tumor suppressor gene. One of the best studied functions of HIF-1a is the regulation of glucose metabolism by glycolytic enzymes. Importantly, cell metabolic pathways feedback to HIF to allow tight control of energy generating process. Thus, mutations in isocitrate dehydrogenase 1 (IDH1), which occur in certain types of brain tumors, were found to impair a-KG synthesis. Because a-KG is a substrate in HIF-a hydroxylation reaction (Fig. 3.2), impairment of IDH1 function leads to increased HIF-a stability (Zhao et al. 2009). Importantly, IDH1 acts as a homodimer and mutations in IDH1 not only inactivate the gene product of the mutated allele, but confirm the mutated protein with the dominant-negative activity toward the WT allele (Zhao et al. 2009). This fact explains why the LOH for IDH1 is not observed in tumors. Thus, IDH1 appears to be another tumor suppressor gene involved in HIF-a destabilization. Activation of phosphoinositide 3 kinase/Akt (PI-3K/Akt) and mitogen-activated protein kinase (MAPK) signaling pathways is frequently observed in multiple types of cancer and it was shown that they can induce HIF activity (Berra et al. 2000; Blancher et al. 2001; Jiang et al. 2001; Sodhi et al. 2001). The PI-3K pathway is activated in cancer by enhanced growth factor signaling (Zelzer et al. 1998), activation of oncogenes (positive regulators) or inactivation of PTEN (Zundel et  al. 2000), or TSC2 (Brugarolas et al. 2003) (negative regulators). To date, the detailed mechanism of PI-3K/Akt-mediated HIF regulation is unknown, but several studies show it to be mTOR dependent, where mTOR contributes to increased HIF protein synthesis (Hudson et al. 2002; Majumder et al. 2004). On the other hand, MAPK can directly phosphorylate HIF-a and increase its stabilization. In summary, HIF activity can be regulated at multiple levels: oxygen concentration, a-KG synthesis (substrates in hydroxylation reaction), and VHL mutation

3  Hypoxia, Gene Expression, and Metastasis

49

(impairment of HIF-a ubiquitination). One can envision additional ways to stabilize HIF, including mutations in prolines 402 and 564, proteasome inhibition, and treatment with iron chelators. Indeed, these are widely used as artificial means of HIF stabilization; and HIF mutations are rarely found in tumors.

HIF Target Genes Involved in the Metastatic Process Hypoxia promotes metastasis through the regulation of key genes involved in ­different steps of metastatic process, including angiogenesis, proliferation, ­migration, invasion, motility, adhesion, ECM remodeling, and survival (list of hypoxia-regulated genes can be found elsewhere) (Giaccia et al. 2003; Semenza 2003; Le et al. 2004; Maynard and Ohh 2007). It is important to note that approximately 1–1.5% of the genome is transcriptionally regulated by hypoxia, and many of these genes have been shown to be HIF regulated (Harris 2002; Semenza 2003, 2007). For some solid tumors, angiogenesis is a critical step in the metastatic process, whereas in other cancers (e.g. ovarian cancer and scirrhous adenocarcinoma), the vasculature is not involved in the metastatic spread. In these tumors, cells from the primary tumor shed to the peritoneal cavity, where they establish secondary lesions. In this case, the group of genes involved in angiogenesis is not relevant for the metastatic spread of these cancer types. The role of individual hypoxia-inducible genes in metastasis has just started to be investigated. This lack of direct in vivo evidence about their role in the metastatic process comes from the fact that many of hypoxia-regulated genes are affecting primary tumor growth, which complicates studies on their roles in metastatic spread. It is believed that the larger the primary tumor, the more cells from that tumor can colonize the secondary organs to form lesions. On the other hand, the establishment of metastasis from one cell requires extensive cell growth. Moreover many hypoxia-inducible genes are multifunctional, involved in several processes, like VEGF, which stimulates cancer cell migration (process not affecting tumor growth) and angiogenesis (affects tumor growth). Furthermore, HIFs activate a diverse array of target genes, which sometimes have opposite functions, like proapoptotic (transcriptional induction of BNIP3, BNIP3L, RTP801 and NIX, and stabilization of p53) and antiapoptotic (transcriptional induction of ADM, IGF2, IGF-BP1, 2, 3, TGF-a, VEGF, Bcl-w like, Pim1, 2). Thus, a shift in the balance of opposing signals as well as the mutation repertoire of a given cancer cell will affect tumor cell behavior under hypoxic conditions. An additional complexity comes from the observation that HIFs not only directly upregulate gene expression, by binding to target promoters, but also indirectly upregulate and downregulate gene expression through downstream signaling. The most ­significant example of indirect downregulation of gene expression is the E-cadherin (see below). The epithelial–mesenchymal transition (EMT) is one of the primary cellular mechanisms known to be involved in the metastatic process (Berx et  al. 2007; Guarino et al. 2007; Tse and Kalluri 2007; Gavert and Ben-Ze’ev 2008). Epithelial

50

O.V. Razorenova and A.J. Giaccia

cells form cell–cell contacts through adherens junctions, whereas the mesenchymal phenotype is characterized by reduced cell–cell contacts, lost cell polarity, and increased interaction with extracellular matrix via integrins and focal adhesion kinase (FAK) (Avizienyte and Frame 2005; McLean et al. 2005). Epithelial cells, which undergo EMT, show a dedifferentiated morphology and become more motile and invasive. Epithelial cells can be reprogrammed to mesenchymal state by hypoxia and HIF that involve suppression of E-cadherin and increased cell motility (Cannito et al. 2008; Sahlgren et al. 2008; Yang et al. 2008). E-cadherin is a cellular adhesion molecule that regulates cell–cell adhesion (for review, see Hanahan and Weinberg 2000). E-cadherin inactivation enhances metastatic potential and its overexpression in cancer inhibits metastasis (Hanahan and Weinberg 2000). HIF has recently been described as a critical factor for the regulation of E-cadherin expression in ovarian carcinoma (Imai et al. 2003) and VHL-deficient CC-RCC (Esteban et al. 2006; Evans et al. 2007). The HIF-mediated upregulation of E-cadherin repressors Twist, Snail, Slug, and SIP1 serves as a mechanism of E-cadherin downregulation (Evans et al. 2007; Vesuna et al. 2008; Yoshida et al. 2009). Twist, a basic helix-loop-helix transcription factor, is a master regulator of gastrulation and mesoderm specification during embryonic development. HIF-1a was shown to regulate the expression of Twist through the HRE located in the Twist proximal promoter (Yang and Wu 2008; Yang et al. 2008). Co-expression of HIF1a, Twist, and Snail in primary tumors of head and neck squamous cell carcinoma patients correlates with the development of metastases and a poor prognosis (Yang et al. 2008). Importantly, the key role of Twist in metastasis was demonstrated by manipulation of Twist expression in highly metastatic mammary carcinoma cells. The downregulation of Twist expression in these cells led to the inhibition of their ability to metastasize from the mammary fat pad to the lung (Yang et  al. 2004, 2006). In human breast cancers, high levels of Twist expression are correlated with invasive lobular carcinoma, a highly infiltrating tumor type associated with loss of E-cadherin expression (Yang et al. 2004). Lysyl oxidase (LOX), an amine oxidase involved in extracellular matrix formation, was shown to be regulated by hypoxia and HIFs (Erler et al. 2006; Erler and Giaccia 2006). Expression of LOX is associated with hypoxia in breast cancer as well as head and neck tumors, and correlates with poor patient outcome, including lower metastasis-free and overall survival rates (Erler et al. 2006; Le et al. 2009). Inhibition of LOX pharmacologically, genetically, or using a neutralizing antibody, in an orthotopic breast cancer model suppressed the metastatic potential of the tumor cells without affecting primary tumor growth (Erler et al. 2006). In vitro LOX regulated cell invasion and migration through regulating the activity of FAK. Moreover, LOX is critical for “premetastatic niche” formation in organs, where it relocates after being secreted by hypoxic primary tumors, crosslinks collagen IV in the basement membrane, and promotes CD11b+ myeloid cell recruitment (Erler et al. 2009). CD11b+ cells adhere to crosslinked collagen IV and produce matrix metalloproteinase-2, which cleaves collagen, enhancing the invasion and recruitment of bone marrow derived cells and metastasizing tumor cells.

3  Hypoxia, Gene Expression, and Metastasis

51

Hypoxia, Cancer Stem Cells, and Metastasis In the recent years, our view of cancer has been transformed by the “cancer stem cell” (CSC) theory (Reya et  al. 2001; Pardal et  al. 2003). Multiple studies have identified CSC subpopulations of cells within leukemias (Lapidot et  al. 1994; Bonnet and Dick 1997), gliomas (Singh et al. 2003, 2004), breast (Al-Hajj et al. 2003), colon (Dalerba et al. 2007; O’Brien et al. 2007; Ricci-Vitiani et al. 2007), pancreatic (Li et  al. 2007), bladder (Chan et  al. 2009), melanoma (Boiko et al. 2010) cancers, head and neck squamous cell carcinoma (Prince et al. 2007), which have high self-renewal capacity, as well as more differentiated cancer cells (nonCSCs), which have very limited proliferative potential, making them incapable to drive tumor growth. The presence of CSC in tumors along with non-CSC makes tumor a complex hierarchically organized tissue. Moreover, it appears that many cancer therapies, while killing the bulk of tumor cells, fail to eliminate CSCs, which are resistant to chemo- and radiation treatment (Bao et al. 2006; Hambardzumyan et al. 2006; Li et al. 2008; Liu et al. 2006; Todaro et al. 2007; Wulf et al. 2001), allowing them to regenerate new tumors. Interestingly, many parallels can be drawn between CSCs and cancer cells residing in the hypoxic microenvironment. There is a striking correlation between chemo- and radiation resistance of CSCs and cancer cells in hypoxic regions. Increasing evidence suggests that hypoxia has the potential to inhibit tumor cell differentiation (Jogi et al. 2002; Helczynska et al. 2003), and thus plays a direct role in the maintenance of CSCs (Keith and Simon 2007; Hill et al. 2009). The role of hypoxia in general and HIFs in particular in maintaining the undifferentiated and pluripotent state of normal stem cells (SCs) is well documented (Cipolleschi et al. 1993, 2000; Lee et  al. 2001; Ezashi et  al. 2005; Parmar et  al. 2007). Embryonic stem cells (ESCs) express the transcription factor Oct4 regulated by hypoxia during development (Covello et al. 2006). Moreover, the activity of Oct4, Sox2, and c-Myc used to reprogram fibroblasts to induced pluripotent stem (iPS) cell state (Takahashi and Yamanaka 2006; Takahashi et al. 2007; Wernig et al. 2007) is modulated by hypoxia. Oct4 is a HIF-2 target gene (Covello et al. 2006), Sox2 is hypoxia inducible in CD133+ neural CSCs and human ES cells (Forristal et al. 2009; McCord et al. 2009), and activity of c-Myc is induced by HIF-2 and inhibited by HIF-1. HIF-2 potentiates c-Myc activity by enhancing its physical association with Sp1, Miz1, and Max, although the precise mechanisms regulating these events are not yet fully understood (Gordan et al. 2007). HIF-1 antagonizes c-Myc activity by competing for binding to the transcription factor Sp1 under hypoxic conditions resulting in inhibition of c-Myc-dependent cell-cycle progression (Koshiji et al. 2004). Hypoxic regulation of Oct4, Sox2, and c-Myc simultaneously in one cell type needs to be investigated, opening the possibility that hypoxic exposure of fibroblasts or forced HIF-2 overexpression might lead to the iPS phenotype. In addition, Notch was shown to maintain a dedifferentiated state in multiple cell types under hypoxia (Gustafsson et al. 2005), where the formation of HIF-1a/Notch ICD domain complex enhanced Notch signaling (Sahlgren et al. 2008).

52

O.V. Razorenova and A.J. Giaccia

Interestingly, PTEN appears to be the negative regulator of SC pool (Groszer et al. 2001; Zheng et al. 2008), because deletion of PTEN results in an increase of neural stem cells (NSCs) numbers along with tumor expansion. Thus, the normal NSCs are finding a way to circumvent PTEN tumor suppressor function in order to become malignant. As discussed above, PTEN is a documented negative regulator of HIF, acting through the PI3K/Akt pathway (Zundel et al. 2000). In recent years, direct links between hypoxia, HIFs, and CSCs have emerged. Growing CD133+ neural CSCs at 7% oxygen compared to 20% reduced their doubling time and increased their self-renewal potential as reflected by clonogenicity (McCord et  al. 2009). Growth of neural CSCs at 7% oxygen also resulted in an increase in the expression levels of the NSC markers CD133 and nestin as well as the SC markers Oct4 and Sox2. In addition, HIF-1 activity was not changed in CD133+ CSCs grown at 7%, and HIF-2 was expressed at higher levels (McCord et al. 2009). Another study showed that HIF-2a and multiple HIF-regulated genes are preferentially expressed in glioma stem cells (GSCs) in comparison to nonGSCs and normal neural progenitors (Li et  al. 2009). Targeting HIFs in GSCs inhibited self-renewal, proliferation, and survival in  vitro, and attenuated tumor initiation potential of GSCs in vivo. The third study showed that hypoxia (1% oxygen) promoted the self-renewal capacity and maintained the undifferentiated phenotype of CD133+ human glioma-derived CSCs, accompanied with expansion of cells bearing CXCR4 (CD184), CD44 (low), and A2B5 surface markers (Soeda et  al. 2009). Knockdown of HIF-1a abrogated the hypoxia-mediated CD133-positive CSC expansion. On the other hand, mTOR signal and HIF-1 were shown to negatively regulate CD133 expression in cancer cell lines under 0.1% oxygen (which is close to anoxia) (Matsumoto et  al. 2009). In support of the hypothesis that cells might be shifted into the iPS state by forced HIF expression, glioblastoma nonCSCs were unexpectedly found to possess plasticity under hypoxic conditions (2% oxygen), which helped to maintain self-renewal of glioblastoma CSCs and promoted reprogramming of glioblastoma non-CSCs toward a CSC phenotype (Heddleston et al. 2009). Thus, it looks like the oxygen concentration in the normal and CSC niche is very important and it has to be determined whether physiologic oxygen conditions or conditions of hypoxia/anoxia favor CSC and normal SC maintenance. In this case, the apparent discrepancy between the two reports about CD133 expression under hypoxia can be explained by differential effects on HIF-1 and 2 and by different oxygen levels. Because CSCs are capable of self-renewal, and are responsible for the long-term maintenance of tumor growth, it has been predicted that they might be also involved in metastases (Dalerba and Clarke 2007; Hermann et al. 2007). Thus, tumor initiating CD133+ pancreatic CSC population can be subdivided into CXCR4- (nonmetastatic) and CXCR4+ (metastatic) subpopulations. Bulk CD133+ cells (containing a mixture of CD133+/CXCR4+ and CD133+/CXCR4- cells) and individual CD133+/ CXCR4+ and CD133+/CXCR4- populations show similar tumor growth abilities. However, CD133+/CXCR4- tumors do not form spontaneous metastases during the time of experiment. A similar effect was obtained by pharmacological inhibition of

3  Hypoxia, Gene Expression, and Metastasis

53

CXCR4. CXCR4 is the receptor for the CXCL12/SDF-1 chemokine, which is involved in the control of leukocyte trafficking under physiological conditions. In cancer, CXCR4 was shown to regulate migration and metastasis (Muller et  al. 2001; Balkwill 2004); notably, its ligand SDF-1 is highly expressed at common sites of metastasis, including the lung, bone marrow, and liver, and might direct CXCR4-expressing tumor cells to secondary metastatic sites (Kucia et  al. 2005; Arya et al. 2007). An interesting twist in this story is that CXCR4 (Staller et  al. 2003; Phillips et al. 2005; Zagzag et al. 2006) as well as its ligand SDF-1 (Ceradini et al. 2004) are hypoxia-inducible genes and HIF is a potent inducer of CXCR4 in VHL-deficient RCCs, nonsmall cell lung cancer, glioblastomas, and SDF-1 in endothelial cells (Staller et  al. 2003; Ceradini et  al. 2004; Phillips et  al. 2005; Zagzag et al. 2006).

Conclusion The metastatic process is a complex multistep process requiring programming of cancer cells toward mesenchymal undifferentiated phenotype, with dominating promigratory and pro-survival signals. This phenotype is largely achieved through the tumor microenvironment, where cancer cells reside. The growing amount of evidence now shows that hypoxia creates a powerful niche, capable of shifting gene expression in cancer cells through stimulation of HIF activity toward pro-metastatic phenotype and protecting cells from chemo- and radiation treatment. Development of drugs, specifically targeting hypoxic cells, grows rapidly.

References Al-Hajj M, Wicha MS et al (2003) Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci USA 100(7):3983–3988. Arya M, Ahmed H et al (2007) Clinical importance and therapeutic implications of the pivotal CXCL12-CXCR4 (chemokine ligand-receptor) interaction in cancer cell migration. Tumour Biol 28(3):123–131. Avizienyte E, Frame MC (2005) Src and FAK signalling controls adhesion fate and the epithelialto-mesenchymal transition. Curr Opin Cell Biol 17(5):542–547. Balkwill F (2004) The significance of cancer cell expression of the chemokine receptor CXCR4. Semin Cancer Biol 14(3):171–179. Bao S, Wu Q et al (2006) Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. Nature 444(7120):756–760. Berra E, Pages G et  al (2000) MAP kinases and hypoxia in the control of VEGF expression. Cancer Metastasis Rev 19(1–2):139–145. Berx G, Raspe E et al (2007) Pre-EMTing metastasis? Recapitulation of morphogenetic processes in cancer. Clin Exp Metastasis 24(8):587–597. Blancher C, Moore JW et al (2001) Effects of ras and von Hippel–Lindau (VHL) gene mutations on hypoxia-inducible factor (HIF)-1alpha, HIF-2alpha, and vascular endothelial growth factor

54

O.V. Razorenova and A.J. Giaccia

expression and their regulation by the phosphatidylinositol 3’-kinase/Akt signaling pathway. Cancer Res 61(19):7349–7355. Boiko AD, Razorenova OV et al (2010) Human melanoma-initiating cells express neural crest nerve growth factor receptor CD271. Nature 466:133–137. Bonnet D, Dick JE (1997) Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 3(7):730–737. Brizel DM, Scully SP et al (1996) Tumor oxygenation predicts for the likelihood of distant metastases in human soft tissue sarcoma. Cancer Res 56(5):941–943. Brugarolas JB, Vazquez F et  al (2003) TSC2 regulates VEGF through mTOR-dependent and -independent pathways. Cancer Cell 4(2):147–158. Cannito S, Novo E et  al (2008) Redox mechanisms switch on hypoxia-dependent epithelial– mesenchymal transition in cancer cells. Carcinogenesis 29(12):2267–2278. Carmeliet P (2005) VEGF as a key mediator of angiogenesis in cancer. Oncology 69(Suppl 3):4–10. Ceradini DJ, Kulkarni AR et al (2004) Progenitor cell trafficking is regulated by hypoxic gradients through HIF-1 induction of SDF-1. Nat Med 10(8):858–864. Chan DA, Giaccia AJ (2007) Hypoxia, gene expression, and metastasis. Cancer Metastasis Rev 26(2):333–339. Chan KS, Espinosa I et  al (2009) Identification, molecular characterization, clinical prognosis, and therapeutic targeting of human bladder tumor-initiating cells. Proc Natl Acad Sci USA 106(33):14016–14021. Chen F, Kishida T et  al (1995) Germline mutations in the von Hippel–Lindau disease tumor ­suppressor gene: correlations with phenotype. Hum Mutat 5(1):66–75. Cipolleschi MG, Dello Sbarba P et al (1993) The role of hypoxia in the maintenance of hematopoietic stem cells. Blood 82(7):2031–2037. Cipolleschi MG, Rovida E et al (2000) The expansion of murine bone marrow cells preincubated in hypoxia as an in  vitro indicator of their marrow-repopulating ability. Leukemia 14(4):735–739. Comerford KM, Wallace TJ et al (2002) Hypoxia-inducible factor-1-dependent regulation of the multidrug resistance (MDR1) gene. Cancer Res 62(12):3387–3394. Covello KL, Kehler J et  al (2006) HIF-2alpha regulates Oct-4: effects of hypoxia on stem cell function, embryonic development, and tumor growth. Genes Dev 20(5):557–570. Dalerba P, Clarke MF (2007) Cancer stem cells and tumor metastasis: first steps into uncharted territory. Cell Stem Cell 1(3):241–242. Dalerba P, Dylla SJ et  al (2007) Phenotypic characterization of human colorectal cancer stem cells. Proc Natl Acad Sci USA 104(24):10158–10163. Dewhirst MW, Tso CY et  al (1989) Morphologic and hemodynamic comparison of tumor and healing normal tissue microvasculature. Int J Radiat Oncol Biol Phys 17(1):91–99. Endrich B, Reinhold HS et al (1979) Tissue perfusion inhomogeneity during early tumor growth in rats. J Natl Cancer Inst 62(2):387–395. Erler JT, Giaccia AJ (2006) Lysyl oxidase mediates hypoxic control of metastasis. Cancer Res 66(21):10238–10241. Erler JT, Bennewith KL et al (2006) Lysyl oxidase is essential for hypoxia-induced metastasis. Nature 440(7088):1222–1226. Erler JT, Bennewith KL et al (2009) Hypoxia-induced lysyl oxidase is a critical mediator of bone marrow cell recruitment to form the premetastatic niche. Cancer Cell 15(1):35–44. Esteban MA, Tran MG et al (2006) Regulation of E-cadherin expression by VHL and hypoxiainducible factor. Cancer Res 66(7):3567–3575. Evans AJ, Russell RC et al (2007) VHL promotes E2 box-dependent E-cadherin transcription by HIF-mediated regulation of SIP1 and snail. Mol Cell Biol 27(1):157–169. Ezashi T, Das P et al (2005) Low O2 tensions and the prevention of differentiation of hES cells. Proc Natl Acad Sci USA 102(13):4783–4788. Forristal CE, Wright KL et al (2009) Hypoxia inducible factors regulate pluripotency and proliferation in human embryonic stem cells cultured at reduced oxygen tensions. Reproduction 139(1):85–97.

3  Hypoxia, Gene Expression, and Metastasis

55

Fyles AW, Milosevic M et  al (1998) Oxygenation predicts radiation response and survival in patients with cervix cancer. Radiother Oncol 48(2):149–156. Gardner LB, Corn PG (2008) Hypoxic regulation of mRNA expression. Cell Cycle 7(13): 1916–1924. Gavert N, Ben-Ze’ev A (2008) Epithelial–mesenchymal transition and the invasive potential of tumors. Trends Mol Med 14(5):199–209. Giaccia A, Siim BG et al (2003) HIF-1 as a target for drug development. Nat Rev Drug Discov 2(10):803–811. Gnarra JR, Glenn GM et al (1993) Molecular genetic studies of sporadic and familial renal cell carcinoma. Urol Clin North Am 20(2):207–216. Gordan JD, Bertout JA et al (2007) HIF-2alpha promotes hypoxic cell proliferation by enhancing c-myc transcriptional activity. Cancer Cell 11(4):335–347. Groszer M, Erickson R et al (2001) Negative regulation of neural stem/progenitor cell proliferation by the Pten tumor suppressor gene in vivo. Science 294(5549):2186–2189. Grunt TW, Lametschwandtner A et  al (1985) The vascular pattern of basal cell tumors: light microscopy and scanning electron microscopic study on vascular corrosion casts. Microvasc Res 29(3):371–386. Guarino M, Rubino B et al (2007) The role of epithelial–mesenchymal transition in cancer pathology. Pathology 39(3):305–318. Gustafsson MV, Zheng X et al (2005) Hypoxia requires notch signaling to maintain the undifferentiated cell state. Dev Cell 9(5):617–628. Hall EJ (1994) Molecular biology in radiation therapy: the potential impact of recombinant technology on clinical practice. Int J Radiat Oncol Biol Phys 30(5):1019–1028. Hambardzumyan D, Squatrito M et al (2006) Radiation resistance and stem-like cells in brain tumors. Cancer Cell 10(6):454–456. Hanahan D, Weinberg RA (2000) The hallmarks of cancer. Cell 100(1):57–70. Harris AL (2002) Hypoxia – a key regulatory factor in tumour growth. Nat Rev Cancer 2(1):38–47. Heddleston JM, Li Z et  al (2009) The hypoxic microenvironment maintains glioblastoma stem cells and promotes reprogramming towards a cancer stem cell phenotype. Cell Cycle 8(20): 3274–3284, Helczynska K, Kronblad A et al (2003) Hypoxia promotes a dedifferentiated phenotype in ductal breast carcinoma in situ. Cancer Res 63(7):1441–1444. Hermann PC, Huber SL et al (2007) Distinct populations of cancer stem cells determine tumor growth and metastatic activity in human pancreatic cancer. Cell Stem Cell 1(3):313–323. Hill RP, Marie-Egyptienne DT et  al (2009) Cancer stem cells, hypoxia and metastasis. Semin Radiat Oncol 19(2):106–111. Hockel M, Knoop C et  al (1993) Intratumoral pO2 predicts survival in advanced cancer of the uterine cervix. Radiother Oncol 26(1):45–50. Hockel M, Schlenger K et al (1996) Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cancer Res 56(19):4509–4515. Hoffman MA, Ohh M et al (2001) von Hippel–Lindau protein mutants linked to type 2C VHL disease preserve the ability to downregulate HIF. Hum Mol Genet 10(10):1019–1027. Huang ST, Vo KC et al (2004) Developmental response to hypoxia. FASEB J 18(12):1348–1365. Hudson CC, Liu M et  al (2002) Regulation of hypoxia-inducible factor 1alpha expression and function by the mammalian target of rapamycin. Mol Cell Biol 22(20):7004–7014.Iliopoulos O, Levy AP et al (1996) Negative regulation of hypoxia-inducible genes by the von Hippel– Lindau protein. Proc Natl Acad Sci USA 93(20):10595–10599. Imai T, Horiuchi A et al (2003) Hypoxia attenuates the expression of E-cadherin via up-regulation of SNAIL in ovarian carcinoma cells. Am J Pathol 163(4):1437–1447. Ivan M, Kondo K et  al (2001) HIFalpha targeted for VHL-mediated destruction by proline hydroxylation: implications for O2 sensing. Science 292(5516):464–468. Jaakkola P, Mole DR et al (2001) Targeting of HIF-alpha to the von Hippel–Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science 292(5516):468–472.

56

O.V. Razorenova and A.J. Giaccia

Jiang BH, Jiang G et al (2001) Phosphatidylinositol 3-kinase signaling controls levels of hypoxiainducible factor 1. Cell Growth Differ 12(7):363–369. Jogi A, Ora I et al (2002) Hypoxia alters gene expression in human neuroblastoma cells toward an immature and neural crest-like phenotype. Proc Natl Acad Sci USA 99(10):7021–7026. Kaelin WG Jr (2008) The von Hippel–Lindau tumour suppressor protein: O2 sensing and cancer. Nat Rev Cancer 8(11):865–873. Kallman RF, Dorie MJ (1986) Tumor oxygenation and reoxygenation during radiation therapy: their importance in predicting tumor response. Int J Radiat Oncol Biol Phys 12(4):681–685. Keith B, Simon MC (2007) Hypoxia-inducible factors, stem cells, and cancer. Cell 129(3):465–472. Kizaka-Kondoh S, Inoue M et  al (2003) Tumor hypoxia: a target for selective cancer therapy. Cancer Sci 94(12):1021–1028. Koshiji M, Kageyama Y et al (2004) HIF-1alpha induces cell cycle arrest by functionally counteracting Myc. EMBO J 23(9):1949–1956. Kucia M, Reca R et al (2005) Trafficking of normal stem cells and metastasis of cancer stem cells involve similar mechanisms: pivotal role of the SDF-1-CXCR4 axis. Stem Cells 23(7):879–894. Lapidot T, Sirard C et al (1994) A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature 367(6464):645–648. Latif F, Tory K et  al (1993) Identification of the von Hippel–Lindau disease tumor suppressor gene. Science 260(5112):1317–1320. Le QT, Denko NC et al (2004) Hypoxic gene expression and metastasis. Cancer Metastasis Rev 23(3–4):293–310. Le QT, Harris J et al (2009) Validation of lysyl oxidase as a prognostic marker for metastasis and survival in head and neck squamous cell carcinoma: Radiation Therapy Oncology Group trial 90–03. J Clin Oncol 27(26):4281–4286. Lee YM, Jeong CH et al (2001) Determination of hypoxic region by hypoxia marker in developing mouse embryos in vivo: a possible signal for vessel development. Dev Dyn 220(2):175–186. Lee JW, Bae SH et  al (2004) Hypoxia-inducible factor (HIF-1)alpha: its protein stability and biological functions. Exp Mol Med 36(1):1–12. Li C, Heidt DG et  al (2007) Identification of pancreatic cancer stem cells. Cancer Res 67(3):1030–1037. Li L, Zhang L et al (2007) Hypoxia-inducible factor linked to differential kidney cancer risk seen with type 2A and type 2B VHL mutations. Mol Cell Biol 27(15):5381–5392. Li X, Lewis MT et al (2008) Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy. J Natl Cancer Inst 100(9):672–679. Li Z, Bao S et al (2009) Hypoxia-inducible factors regulate tumorigenic capacity of glioma stem cells. Cancer Cell 15(6):501–513. Liu G, Yuan X et al (2006) Analysis of gene expression and chemoresistance of CD133+ cancer stem cells in glioblastoma. Mol Cancer 5:67. Mahon PC, Hirota K et al (2001) FIH-1: a novel protein that interacts with HIF-1alpha and VHL to mediate repression of HIF-1 transcriptional activity. Genes Dev 15(20):2675–2686. Majumder PK, Febbo PG et al (2004) mTOR inhibition reverses Akt-dependent prostate intraepithelial neoplasia through regulation of apoptotic and HIF-1-dependent pathways. Nat Med 10(6):594–601. Makino Y, Cao R et al (2001) Inhibitory PAS domain protein is a negative regulator of hypoxiainducible gene expression. Nature 414(6863):550–554. Matsumoto K, Arao T et al (2009) mTOR signal and hypoxia-inducible factor-1 alpha regulate CD133 expression in cancer cells. Cancer Res 69(18):7160–7164. Maxwell PH, Wiesener MS et  al (1999) The tumour suppressor protein VHL targets hypoxiainducible factors for oxygen-dependent proteolysis. Nature 399(6733):271–275. Maynard MA, Ohh M (2004) Von Hippel–Lindau tumor suppressor protein and hypoxia-inducible factor in kidney cancer. Am J Nephrol 24(1):1–13. Maynard MA, Ohh M (2007) The role of hypoxia-inducible factors in cancer. Cell Mol Life Sci 64(16):2170–2180.

3  Hypoxia, Gene Expression, and Metastasis

57

McCord AM, Jamal M et  al (2009) Physiologic oxygen concentration enhances the stem-like properties of CD133+ human glioblastoma cells in vitro. Mol Cancer Res 7(4):489–497. McLean GW, Carragher NO et al (2005) The role of focal-adhesion kinase in cancer – a new therapeutic opportunity. Nat Rev Cancer 5(7):505–515. Muller A, Homey B et al (2001) Involvement of chemokine receptors in breast cancer metastasis. Nature 410(6824):50–56. Nguyen DX, Bos PD et al (2009) Metastasis: from dissemination to organ-specific colonization. Nat Rev Cancer 9(4):274–284. Nordsmark M, Hoyer M et al (1996) The relationship between tumor oxygenation and cell proliferation in human soft tissue sarcomas. Int J Radiat Oncol Biol Phys 35(4):701–708. O’Brien CA, Pollett A et al (2007) A human colon cancer cell capable of initiating tumour growth in immunodeficient mice. Nature 445(7123):106–110. Papandreou I, Powell A et al (2005) Cellular reaction to hypoxia: sensing and responding to an adverse environment. Mutat Res 569(1–2):87–100. Pardal R, Clarke MF et al (2003) Applying the principles of stem-cell biology to cancer. Nat Rev Cancer 3(12):895–902. Parmar K, Mauch P et  al (2007) Distribution of hematopoietic stem cells in the bone marrow according to regional hypoxia. Proc Natl Acad Sci USA 104(13):5431–5436. Pause A, Peterson B et al (1999) Studying interactions of four proteins in the yeast two-hybrid system: structural resemblance of the pVHL/elongin BC/hCUL-2 complex with the ubiquitin ligase complex SKP1/cullin/F-box protein. Proc Natl Acad Sci USA 96(17): 9533–9538. Phillips RJ, Mestas J et  al (2005) Epidermal growth factor and hypoxia-induced expression of CXC chemokine receptor 4 on non-small cell lung cancer cells is regulated by the phosphatidylinositol 3-kinase/PTEN/AKT/mammalian target of rapamycin signaling pathway and activation of hypoxia inducible factor-1alpha. J Biol Chem 280(23):22473–22481. Prince ME, Sivanandan R et al (2007) Identification of a subpopulation of cells with cancer stem cell properties in head and neck squamous cell carcinoma. Proc Natl Acad Sci USA 104(3): 973–978. Qing G, Simon MC (2009) Hypoxia inducible factor-2alpha: a critical mediator of aggressive tumor phenotypes. Curr Opin Genet Dev 19(1):60–66. Rankin EB, Giaccia AJ (2008) The role of hypoxia-inducible factors in tumorigenesis. Cell Death Differ 15(4):678–685. Reya T, Morrison SJ et  al (2001) Stem cells, cancer, and cancer stem cells. Nature 414(6859):105–111. Ricci-Vitiani L, Lombardi DG et al (2007) Identification and expansion of human colon-cancerinitiating cells. Nature 445(7123):111–115. Sahlgren C, Gustafsson MV et al (2008) Notch signaling mediates hypoxia-induced tumor cell migration and invasion. Proc Natl Acad Sci USA 105(17):6392–6397. Semenza GL (2003) Targeting HIF-1 for cancer therapy. Nat Rev Cancer 3(10):721–732. Semenza GL (2007) Hypoxia and cancer. Cancer Metastasis Rev 26(2):223–224. Shah-Yukich AA, Nelson AC (1988) Characterization of solid tumor microvasculature: a threedimensional analysis using the polymer casting technique. Lab Invest 58(2):236–244. Shannon AM, Bouchier-Hayes DJ et al (2003) Tumour hypoxia, chemotherapeutic resistance and hypoxia-related therapies. Cancer Treat Rev 29(4):297–307. Singh SK, Clarke ID et  al (2003) Identification of a cancer stem cell in human brain tumors. Cancer Res 63(18):5821–5828. Singh SK, Hawkins C et al (2004) Identification of human brain tumour initiating cells. Nature 432(7015):396–401. Sodhi A, Montaner S et al (2001) MAPK and Akt act cooperatively but independently on hypoxia inducible factor-1alpha in rasV12 upregulation of VEGF. Biochem Biophys Res Commun 287(1):292–300. Soeda A, Park M et al (2009) Hypoxia promotes expansion of the CD133-positive glioma stem cells through activation of HIF-1alpha. Oncogene 28(45):3949–3959.

58

O.V. Razorenova and A.J. Giaccia

Sprenger SH, Gijtenbeek JM et  al (2001) Characteristic chromosomal aberrations in sporadic cerebellar hemangioblastomas revealed by comparative genomic hybridization. J Neurooncol 52(3):241–247. Staller P, Sulitkova J et  al (2003) Chemokine receptor CXCR4 downregulated by von Hippel– Lindau tumour suppressor pVHL. Nature 425(6955):307–311. Takahashi K, Yamanaka S (2006) Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126(4):663–676. Takahashi K, Tanabe K et al (2007) Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131(5):861–872. Todaro M, Alea MP et al (2007) Colon cancer stem cells dictate tumor growth and resist cell death by production of interleukin-4. Cell Stem Cell 1(4):389–402. Tse JC, Kalluri R (2007) Mechanisms of metastasis: epithelial-to-mesenchymal transition and contribution of tumor microenvironment. J Cell Biochem 101(4):816–829. Vaupel P, Mayer A (2007) Hypoxia in cancer: significance and impact on clinical outcome. Cancer Metastasis Rev 26(2):225–239. Vesuna F, van Diest P et al (2008) Twist is a transcriptional repressor of E-cadherin gene expression in breast cancer. Biochem Biophys Res Commun 367(2):235–241. Wartenberg M, Ling FC et  al (2003) Regulation of the multidrug resistance transporter P-glycoprotein in multicellular tumor spheroids by hypoxia-inducible factor (HIF-1) and reactive oxygen species. FASEB J 17(3):503–505. Wernig M, Meissner A et  al (2007) In vitro reprogramming of fibroblasts into a pluripotent ES-cell-like state. Nature 448(7151):318–324. Wiesener MS, Jurgensen JS et al (2003) Widespread hypoxia-inducible expression of HIF-2alpha in distinct cell populations of different organs. FASEB J 17(2):271–273. Wulf GG, Wang RY et al (2001) A leukemic stem cell with intrinsic drug efflux capacity in acute myeloid leukemia. Blood 98(4):1166–1173. Yang J, Mani SA et al (2004) Twist, a master regulator of morphogenesis, plays an essential role in tumor metastasis. Cell 117(7):927–939. Yang J, Mani SA et  al (2006) Exploring a new twist on tumor metastasis. Cancer Res 66(9):4549–4552. Yang MH, Wu KJ (2008) TWIST activation by hypoxia inducible factor-1 (HIF-1): implications in metastasis and development. Cell Cycle 7(14):2090–2096. Yang MH, Wu MZ et al (2008) Direct regulation of TWIST by HIF-1alpha promotes metastasis. Nat Cell Biol 10(3):295–305. Yoshida J, Horiuchi A et  al (2009) Changes in the expression of E-cadherin repressors, Snail, Slug, SIP1, and Twist, in the development and progression of ovarian carcinoma: the important role of Snail in ovarian tumorigenesis and progression. Med Mol Morphol 42(2):82–91. Zagzag D, Lukyanov Y et al (2006) Hypoxia-inducible factor 1 and VEGF upregulate CXCR4 in glioblastoma: implications for angiogenesis and glioma cell invasion. Lab Invest 86(12): 1221–1232. Zbar B, Kishida T et al (1996) Germline mutations in the Von Hippel–Lindau disease (VHL) gene in families from North America, Europe, and Japan. Hum Mutat 8(4):348–357. Zelzer E, Levy Y et al (1998) Insulin induces transcription of target genes through the hypoxiainducible factor HIF-1alpha/ARNT. EMBO J 17(17):5085–5094. Zhao S, Lin Y et al (2009) Glioma-derived mutations in IDH1 dominantly inhibit IDH1 catalytic activity and induce HIF-1alpha. Science 324(5924):261–265. Zheng H, Ying H et al (2008) p53 and Pten control neural and glioma stem/progenitor cell renewal and differentiation. Nature 455(7216):1129–1133. Zundel W, Schindler C et al (2000) Loss of PTEN facilitates HIF-1-mediated gene expression. Genes Dev 14(4):391–396.

Chapter 4

Molecular Mechanisms Regulating Expression and Function of Cancer-Associated Carbonic Anhydrase IX Jaromir Pastorek and Silvia Pastorekova

Abstract  Hypoxia and acidosis are typical physiological factors of tumor microenvironment acting in concert to support invasive and metastatic propensity of tumor cells through rearrangement of their gene expression profile and remodeling of their phenotype. Induction of carbonic anhydrase IX (CA IX) has recently been recognized as one of the important molecular events observed in tumor cells exposed to hypoxia. CA IX can serve as a “tag” of hypoxic areas in many tumor types and thereby provides a diagnostic tool as well as a target for selective delivery of anticancer immunotherapy. Moreover, CA IX actively contributes to pH regulation which protects tumor cells from acidosis and thereby provides a target for anticancer therapy based on reduction of tumor cell survival via inhibition of CA IX enzyme activity. Both aspects attract considerable attention but their full appraisal and further development require deeper understanding of the mechanisms behind CA IX expression and functioning. Here, we summarize and discuss the “state of the art” of CA IX field in context of its relevance for tumor microenvironment.

Carbonic Anhydrases Acid–base balance mediated by equilibrium between carbon dioxide and carbonic acid is essential for proper functioning of metabolically active cells, tissues, and organs in virtually all living organisms. Although reversible conversion of carbon dioxide to carbonic acid followed by dissociation of carbonic acid to bicarbonate ion and a proton can proceed spontaneously, turnover of this reaction is not fast enough to satisfactorily supply physiological processes and therefore, it needs to be catalyzed by zinc-binding enzymes traditionally called carbonic anhydrases (CAs).

J. Pastorek (*) Department of Molecular Medicine, Institute of Virology, Slovak Academy of Sciences, Dubravska cesta 9, 845 05 Bratislava, Slovak Republic e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_4, © Springer Science+Business Media, LLC 2010

59

60

J. Pastorek and S. Pastorekova

CAs are present all over the kingdoms of animals, plants, fungi, and bacteria (including archeabacteria), and are encoded by five genetically unrelated families (a–e), each containing diverse isoforms (Supuran 2008). Human CAs belong to a-CA family, which is characterized by a deep and wide active site pocket that contains three histidine residues coordinating catalytic zinc at the bottom of the active site cleft and the fourth histidine residue functioning as a proton shuttle between the active site and solution shifting between inward and outward orientation (Zheng et  al. 2008). Twelve out of fifteen human a-CA ­isoforms are catalytically active (CA I–IV, VA, VB, VI, VII, IX, XII–XIV). In addition to histidine residues, they contain also other conserved amino acids involved in formation of inner and outer layers of the active site cavity. The CA isoezymes ­differ by their catalytic activity that ranges from extremely high (in case of CA II and CA IX that belong to most efficient enzymes in general) through high/intermediate (CA IV, V, VI, VII, XII, XIV) to very low (CA I, III), as shown in Fig. 4.1. Three acatalytic isoforms (CA VIII, X, XI) lack one of three critical zinc-coordinating histidines. Moreover, there are two CA-related proteins that function as receptor protein tyrosine phosphatases (RPTPb/g). Their incompletely conserved, catalytically silent CA domain pocket serves as a ligand-binding site for neuronal adhesion molecule contactin (Peles et al. 1995).

Fig. 4.1  Schematic illustration of the human CA isoforms, their domain composition, enzyme activity, and subcellular localization. Cytoplasmic and mitochondrial CAs consist only of the CA domain, secreted CA contains a short C-terminal extension, membrane-associated CAs in addition have a transmembrane anchor and, except for CA IV, also the cytoplasmic tail. CA IX is the only isoform containing the N-terminal PG-like domain. Structure of the CA IX isoforms is depicted in more detail on the left side. It includes signal peptide (SP), proteoglycan-like segment (PG), carbonic anhydrase domain (CA), transmembrane anchor (TM), and intracytoplasmic tail (IC). Conserved zinc-binding histidines are indicated by black circles, cysteines are shown as white squares, and intramolecular disulfidic bond is drawn as dotted line. Numbers designate amino acids encompassed within the individual domains

4  Molecular Mechanisms Regulating Expression

61

Evolutionarily oldest human CA isoenzymes contain an extracellular catalytic domain and are either transmembrane (CA IX, XII, XIV), or membrane-associated via GPI anchor (CA IV) or secreted (CA VI). The transmembrane CAs also contain a hydrophobic region and a relatively conserved intracellular tail. In addition, CA IX possesses an N-terminal proteoglycan-like region. The younger CAs (I, II, III, V, VII) are intracellular isoenzymes containing a sole CA domain. Except CA VA and VB that reside in the mitochondrion, all intracellular CAs are located in the cytoplasm. Finally, the evolutionarily youngest cluster includes inactive ­cytoplasmic isoforms mentioned above. CAs are spread throughout the human body. Some of them are ubiquitous (CA II) or broadly distributed (CA IV, XII, XIII), others are confined to only few tissues/cell types such as muscle and adipocytes (CA III), liver (CA VA), and saliva and milk (CA VI). Cells/organs of the most abundant CA expression include the red blood cells, brain, lung, gut, liver, kidney, and testis, where various concurrently present CA isoforms cooperatively contribute to breathing, synaptic transmission, CSF secretion, production of ocular fluid, bone resorption, secretion of gastric acid, urine acidification, sperm maturation, etc. In metabolically active tissues such as kidney or stomach epithelia, the active CAs located in various subcellular compartments or on the opposite sites of the plasma membrane can interact with bicarbonate transporters or biosynthetic enzymes to form spatially and functionally coordinated protein complexes called metabolons (McMurtrie et al. 2004). In bicarbonate transport metabolon, CAs can improve the transmembrane movement of bicarbonate ion, which is a membrane-impermeable molecule transported by integral membrane bicarbonate transporters and exchangers (Casey 2006). The role of CAs is to increase the local concentration of bicarbonate in order to accelerate its flux across the membrane and thereby regulate intracellular and extracellular pH (McMurtrie et al. 2004). CAs are essential components of physiological processes and disbalance in their expression is associated with several diseases. These include on the one hand osteopetrosis and mental retardation due to CA II deficiency and, on the other hand, glaucoma, diabetic retinopathy, epilepsy, obstructive pulmonary disease, obesity and various neurological diseases linked with abnormally high expression of CA II, III, XII, and other isoforms. Some of these diseases are successfully treated with inhibitors of CA activity used as primary or supplementary therapy, for the others, new derivatives of CA inhibitors are under development as promising drugs (Pastorekova et al. 2004b). Although importance of CAs in cancer has been demonstrated relatively recently, inspections into their expression in tumor cells and tissues have been initiated a long time ago. Several independent studies revealed reduced activity and/or decreased levels of cytosolic isoforms CA I, II, III, XIII in colorectal, hepatocellular, and lung tumors (when compared to their normal counterparts) indicating that these isoforms are predominantly needed for differentiated cells/tissues (Chiang et al. 2002; Kivela et al. 2001; Kummola et al. 2005; Kuo et al. 2003; Mori et al. 1993; Yoshiura et al. 2005). This conclusion has, however, few exceptions, including CA II expression in endothelium of malignant melanomas and meningiomas and a recently discovered

62

J. Pastorek and S. Pastorekova

extremely high expression of CA II in gastrointestinal stromal tumors (Haapasalo et  al. 2007; Yoshiura et  al. 2005; Parkkila et  al. 2010). Relationship of the mitochondrial CA VA and VB isoforms to cancer remains unexplored, although they might be relevant for biosynthesis of metabolic intermediates in proliferating tumor cells. Finally, out of four CA isoforms localized at the cell surface, only CA IX and XII showed consistent link with tumor tissues and are systematically investigated as promising targets for anticancer therapy (Pastorekova et al. 2008). Especially CA IX, which is almost exclusively expressed in tumors, attracts considerable interest with expectation that unraveling its regulation and functional contribution to tumor phenotype will provide clues for development of new anticancer strategies.

Molecular Features of CA IX CA IX (initially named MN antigen) was identified in HeLa cervical carcinoma cells as a density-induced cell surface protein associated with tumor phenotype of human cell lines and tissue specimens (Pastorekova et al. 1992; Zavada et al. 1993). Primary structure of the MN antigen was derived from cDNA and genomic sequences and was found to contain a well-conserved carbonic anhydrase domain located in the large extracellular part of the molecule (Pastorek et  al. 1994) (Fig. 4.1). This CA domain shows a significant identity to secreted isozyme CA VI (40.8%) and cytosolic isozyme CA II (35.8%), respectively, and possesses all important amino acids required for the catalytic activity. It is N-glycosylated by 4 kDa high mannose sugar chain attached to N346 and is catalytically active. At the time of MN protein discovery, this was the ninth mammalian CA identified and thus it was renamed to CA IX (Hewett-Emmett and Tashian 1996). In addition to catalytic domain, CA IX (as the only CA isoform) contains an N-terminal PG-like region homologous to keratan sulfate attachment domain of a large proteoglycan aggreccan (Opavsky et  al. 1996). According to mass spectrometric analysis, this part of the molecule appears to contain a keratan sulfate glycosaminoglycan chain of about 3.5 kDa that is presumably attached to T115, in the article numbered T78 (without signal peptide) (Hilvo et al. 2008). Presence of such modification might have important biological implications analogous to keratan sulfate chain of CD44 metastasis-related protein, which mediates interactions with constituents of extracellular matrix. However, experimental proof of this concept has not been given for CA IX so far. The PG-like region has also another important feature – it predominantly contains negatively charged acidic amino acids that are mostly arranged within four identical and three imperfect GEEDLP repeats. This results in remarkably acidic character of the PG-like region that seems to have a significant impact on the catalytic activity of CA IX as an intrinsic buffer optimizing CO2 hydration at acidic pH values characteristic for solid tumors (Innocenti et al. 2009). Thus, activity of the entire extracellular portion of CA IX (including PG and CA domains) expressed in baculovirus system exceeds that of CA II

4  Molecular Mechanisms Regulating Expression

63

(the best known catalyst) and reaches maximum at pH values around 6.5 in contrast to other CA isoforms which have optimum catalytic efficiency around pH 7. C-terminal part of the catalytic domain is extended with a little juxtamembrane fragment, followed by a hydrophobic transmembrane (TM) region and a short intracellular tail (IC). The IC tail is well conserved among the transmembrane CA isoforms (CA IX, XII, XIV) and its removal leads to loss of cell surface location of CA IX despite preserved TM region suggesting that IC mediates protein–protein interactions that maintain CA IX at its proper position (Hulikova et al. 2009). In vitro mutagenesis of the submembrane cluster of basic amino acids revealed that IC tail participates in inside-out signaling that affects functions of the extracellular CA IX portion (Hulikova et al. 2009). In addition, the IC tail contains three potential phosphorylation sites (T443, S448, andY449). While Y449 participates in EGFRinduced signal transduction to Akt kinase (Dorai et al. 2005), phosphorylated T443 is critical for modulation of the pH regulatory function of CA IX (manuscript in preparation). Molecular weight of CA IX analyzed under reducing conditions and visualized on Western blot corresponds to 58/54 kDa. Existence of two forms of the protein remains unexplained and the data attributing the double band to different parts of CA IX molecule are still controversial. Interestingly, in non-reducing conditions, CA IX is capable to form oligomers that were initially considered for trimers (or tetramers). This supposition was based on the molecular weight of about 153 kDa and on other indices including the capability of CA IX to fit into envelope of the vesicular stomatitis virus and form the virus pseudotypes (Pastorekova et al. 1992; Zavada et  al. 1993). The latter aspect was shown to require trimerization or tetramerization (Kreis and Lodish 1986). In fact, CA IX molecule contains four cysteine residues, three of them located in the catalytic domain and one in the region just above the plasma membrane. While C156 and C336 are placed at the positions analogous to intra-molecular S–S bond-forming cysteines in other CAs, the additional two residues C174 and C409 might be theoretically available for intermolecular S–S bonds. Biochemical and crystallographic studies confirmed the proposed intramolecular S–S bond, but they also suggested that CA domain of CA IX forms dimer stabilized by homologous S–S bridge between C174 corresponding to C41 according to CA I numbering (Alterio et al. 2009). Although results of the biochemical and crystallographic analyses do not support the concept of CA IX trimerization, it is still possible that the discrepancy results from the absence of the regions flanking the catalytic domain, as it was observed also in other cases, such as with Fas receptor or TRAF6 (Scott et al. 2009; Yin et al. 2009). Crystallographic analysis brought very important structural data on the CA IX catalytic domain that is similar to other a-CAs (Alterio et al. 2009). It has a compact globular appearance with a large conical cavity containing zinc ion at the bottom. The entrance to the active site is occupied by three arginine residues R95, R97, and R167, which are not conserved in other CA isoforms and have been proposed to mediate crosstalk of CA domain with acidic PG-like region presumably implicated in steric control of the substrate accession or the proton-transfer reaction. The active site and the N-terminus of the catalytic domain face the extracellular space, whereas

64

J. Pastorek and S. Pastorekova

its C-terminus is oriented toward the plasma membrane. This arrangement is fully compatible with the role of CA IX in extracellular acidification of tumor microenvironment.

CA IX Tissue Distribution CA IX protein expression has been studied in a broad range of human tissues. Most of these studies were performed using the monoclonal antibody M75 that recognizes a repetitive linear epitope in the PG-like domain of both native and denatured CA IX (Zavada et al. 2000) and is therefore suitable for virtually all immunodetection methods including immunohistochemistry on both paraffinembedded and frozen sections (even those archived for years), immunoblotting, immunoprecipitation, flow cytometry, etc. Additional antibodies produced by our team (Zat’ovicova et  al. 2003) are mostly directed to CA domain and bind to conformational epitopes (thus they do not work in immunoblotting) similarly to well-known monoclonal antibody G250 that was generated even before the discovery of CA IX against a renal cell carcinoma antigen later shown to be identical to CA IX (Oosterwijk 2008a, b). New human monoclonal antibodies described recently display excellent CA IX binding properties (Ahlskog et al. 2009b). Some of these antibodies have a great potential as diagnostic and therapeutic tools. Noteworthy, a commercial antibody NB100 was advertised to give the same staining pattern as M75, but later demonstration of its significant non-specific reactivity against tubulin considerably reduces its value for detection of CA IX (Li et al. 2009). Overall expression pattern of CA IX is characterized by limited expression in normal tissues and broad distribution in different tumors. Physiological expression of CA IX is linked with epithelial tissues of the gastrointestinal tract with the highest CA IX level in all principal cell types of the glandular gastric mucosa and in the gallbladder epithelium. CA IX is also present in the small intestine and proximal colon, but its level decreases toward rectum (Pastorekova et al. 1997). Intestinal CA IX is expressed in the cryptal epithelial cells that possess the greatest proliferative activity (Saarnio et al. 1998). CA IX is also abundant in the gallbladder mucosa, while the pancreatic ducts show only weak expression. In all gastrointestinal ­epithelia, CA IX is present in the basolateral membranes, suggesting its possible involvement in intercellular communication, maintenance of tissue integrity, and regulation of basolateral ion transport (Pastorekova et al. 1997). Based on gastric phenotype of CA IX-deficient mice that includes hyperplasia, numerous cysts, and abnormal lineage development but no change in gastric pH, CA IX plays a role in gastric morphogenesis and control of cell proliferation and/or differentiation (Gut et al. 2002). CA IX is also expressed in some other healthy tissues, albeit at much lower levels compared to alimentary tract. These include linings of the body cavity, reproductive tract, ventricular linings of CNS, and choroid plexus (Ivanov et al. 2001). The other normal human tissues are CA IX negative.

4  Molecular Mechanisms Regulating Expression

65

In contrast, CA IX expression is readily detected at significant levels in diverse tumors, including carcinomas of the uterine cervix, kidney, brain, head and neck, esophagus and lung, colon, breast, ovaria, endometrium, vulva, bladder, etc. (Bartosova et al. 2002; Haapasalo et al. 2006; Ivanov et al. 2001; Jarvela et al. 2008; Kowalewska et al. 2005; Liao et al. 1994, 1997; McKiernan et al. 1997; Niemela et al. 2007; Saarnio et al. 1998; Turner et al. 1997; Vermylen et al. 1999). From an unknown reason, CA IX is only occasionally expressed in the prostate carcinomas (Smyth et al. 2009).

Regulation of CA IX Expression Presence of CA IX in cancerous tissues of different origin suggests that it represents a phenomenon common to all solid tumors. In accord with this supposition, it has been proven that transcription of CA IX is principally driven by hypoxia that occurs in variable extent in virtually all tumor tissues. Hypoxia often develops in growing tumors due to insufficient supply of oxygen by irregular and functionally inadequate vasculature. This creates a selective pressure favoring those tumor cells that are capable to induce adaptative responses including shift to anaerobic glycolysis, angiogenesis, reduced proliferation and cell adhesion, increased pH regulation, etc. Consequently, hypoxic cells gain oncogenic metabolism, resistance to conventional anticancer treatment, and more aggressive phenotype with metastatic potential (Harris 2002). At the molecular level, hypoxia leads to reprogramming of the cell’s transcriptional profile, which is principally governed by a master transcription factor HIF, i.e. hypoxia inducible factor (Coleman and Ratcliffe 2007), illustrated in Fig. 4.2. HIF is a heterodimer consisting of an oxygen-regulated a subunit (existing in three isoforms) and a constitutive b subunit. In normoxia, HIF-a is modified by oxygendependent prolyl hydroxylases (PHDs) and an asparaginyl hydroxylase (factor inhibiting HIF, named FIH), enzymes that belong to the Fe(II) and 2-oxoglutarate dioxygenase superfamily. FIH hydroxylates N803 in the C-terminal activation domain of human HIF-1a and thereby prevents its interaction with transcriptional coactivators. Prolyl hydroxylases hydoxylate P564 and P402 in the oxygen-dependent degradation domain, thus leading to HIF-a recognition by pVHL tumor suppressor protein, followed by its rapid ubiquitilation and proteasome degradation (Hirota and Semenza 2005). In hypoxia and/or following oncogenic activation or loss of VHL function, HIF-a escapes modifications by PHDs and FIH, enters the nucleus, dimerizes with the constitutive b subunit of HIF, interacts with coactivators, and forms an active transcriptional complex (Ruas and Poellinger 2005). This complex binds DNA at sites containing the hypoxia-response element (HRE; 5¢-RCGTG-3¢), which is present in the promoters of a wide spectrum of target genes encoding functionally diverse proteins involved in angiogenesis (VEGF, VEGF receptor, transforming growth factor TGF-b), vascular tonus (nitric oxide synthase), erythropoiesis (erythropoietin-1),

66

J. Pastorek and S. Pastorekova

Fig. 4.2  Hypoxia-induced HIF-1-mediated transcriptional activation of CA9 gene. The left side shows the situation in normoxia. When oxygen supply is sufficient, prolyl hydroxylases (PHD) and factor inhibiting HIF (FIH) modify a-subunit of hypoxia-inducible factor 1 (HIF-a). Hydroxylation of conserved prolines in HIF-a molecule leads to its recognition by von Hippel– Lindau tumor suppressor protein (VHL), which triggers HIF-a ubiquitilation and degradation. Hydroxylaion of N-terminal asparagine blocks binding of co-activators. This principally results in the absence of active HIF-a from normoxic cells. The right part shows the situation in hypoxia. Lack of oxygen keeps hydroxylases inactive, HIF-a is not modified and remains protected from recognition by VHL. This results in stabilization and accumulation of HIF-a, translocation to nucleus, dimerization with HIF-b, interaction with co-activators, and formation of active transcription complex, which induces expression of target genes, including CA9. This induction can be modified by additional pathways not shown in the figure and leads to exposure of CA IX protein on the surface of hypoxic cells

glycolysis (glucose transporters GLUT-1,3, lactate dehydrogenase LDH-A, pyruvate dehydrogenase kinase PDK1), ion transport and pH regulation (natrium-proton exchanger NHE1, monocarboxylate transporter MCT4), remodeling of extracellular matrix (matrix metalloproteinase MMP-2, cathepsin D), adhesion and migration (hepatocyte growth factor receptor c-Met, transcription factors Snail/Slug), cell ­proliferation and survival (insulin growth factor-binding protein), drug resistance (P glycoprotein), and additional cellular phenomena linked with tolerance of tumor cells to hypoxia. Genomic position of HRE differs among the transcriptional targets of HIF, in some cases representing enhancer rather than promoter element (Camenisch et al. 2001). In case of the CA9 gene, HRE is an important component of the core promoter and its position adjacent to transcription initiation site (–3/–10) makes it a key regulatory element principally determining CA9 gene expression (Wykoff et al. 2000). This HRE element binds HIF-1 (rather than HIF-2) and cooperates with a neighboring consensus sequence binding SP1 transcription factor (Kaluz et  al. 2002). The HRE-SP1 module is critical for transcriptional activation of the CA9 gene both in hypoxia and in high cell density (creating pericellular hypoxia) and its mutagenesis leads to reduction or complete loss of the promoter activity (Kopacek et al. 2005). Additional regulatory elements present within the CA9 promoter, including AP-1 binding site and position/orientation-independent

4  Molecular Mechanisms Regulating Expression

67

silencer, play secondary roles as they just modulate the outcome of the transcriptional activation triggered by HIF-1-SP1 (Kaluz et al. 1999). As recently discussed, CA9 transcription profile primarily reflects the transcriptional activity (not the level) of HIF-1 (Kaluz et  al. 2009). In accord, different pathways that affect HIF-1 can also increase the expression of CA9 gene, such as MAPK and PI3 kinase pathways, Notch pathway, HBx oncoprotein of hepatitis B virus, and v-Src and c-Src oncoproteins (Holotnakova et al. 2010; Kopacek et al. 2005; Sansone et al. 2007; Takacova et  al. 2010). Moreover, hypoxia-induced CA9 gene expression requires an intact unfolded protein response (UPR) pathway, which operates via ATF4 transcription factor that binds to CA9 promoter (van den Beucken et  al. 2009a, b). Under normoxic conditions, expression of CA9 gene is down-regulated by pVHL tumor suppressor protein, which negatively regulates HIF-1a. However, CA9 expression is not suppressed by the mutated variants of pVHL lacking the elongin binding domain which is required for the interaction of pVHL with elongins B/C and integration of pVHL within a ubiquitin ligase complex (Ivanov et al. 1998). This explains why CA IX protein is overexpressed in a high percentage of renal cell carcinomas (RCC) that frequently carry an inactivating mutation in the VHL gene (Mandriota et al. 2002). Extracellular acidosis, which frequently occurs in tumor microenvironment, contributes to CA9 transcription at least in some tumor cell types, such as those derived from glioblastoma (Ihnatko et al. 2006). CA9 induction by acidosis can be observed in both absence and presence of hypoxia. Promoter analysis revealed involvement of SP1 and HRE promoter elements and showed that acidosis can modulate CA9 expression only when basic CA9 activation through MAPK and/or PI3K pathways is preserved. However, inconsistent effects of acidosis on CA9 expression were obtained in other cell types possibly due to their differential sensitivity to acidosis-associated stress (Willam et al. 2006). Finally, transcription of CA9 gene can be negatively controlled by methylation of CA9 promoter and by treatment with xenobiotics (Ashida et al. 2002; Cho et al. 2000; Jakubickova et al. 2005; Takacova et al. 2009). CA9 is also a subject of post-transcriptional regulation via alternative splicing that produces two types of transcripts – a full-length transcript encoding a plasma membrane-localized and functionally competent CA IX protein, and a truncated mRNA lacking exons 8/9 and producing a cytoplasmic/secreted form of CA IX protein with decreased enzyme activity (Barathova et  al. 2008). The longer transcript is regulated by hypoxia and linked to tumor phenotype, whereas the shorter one is expressed at low level independently of hypoxia and is therefore detectable also under normoxia and in the normal tissues. This is very important for the correct design of primers used in the clinical studies of cancer- and/or hypoxia-related expression of CA9, where the alternatively spliced variant can provide a falsepositive signal. As recently demonstrated in non-small cell lung carcinoma, fulllength CA9 transcript is the most accurate surrogate of hypoxic stress and represents the only variant with a prognostic role (Malentacchi et al. 2009). These data indicate the importance of a separate measurement of the two isoforms in cancer and

68

J. Pastorek and S. Pastorekova

the need of an accurate re-evaluation of most studies on the clinical role of CA9 mRNA in cancer diagnosis. At the posttranslational level, CA IX protein shows high stability, with the halflife of about 38 h (as determined in reoxygenated HeLa cells), and therefore represents a marker of both present and past hypoxia (Rafajova et al. 2004). The final amount of CA IX protein is regulated by shedding of its ectodomain (ECD) to culture medium and to body fluids of tumor patients (Zatovicova et  al. 2005; Zavada et al. 2003). Constitutive shedding of CA IX ECD releases approximately 10% of the cell-associated protein and is sensitive to metalloproteinase-inhibitor batimastat. Hypoxia maintains the normal rate of the basal ECD release, thus leading to a concomitant increase in cell-associated and extracellular levels of CA IX. The CA IX ECD shedding can be several fold induced by treatment with phorbol12-myristate-13-acetate and pervanadate. This activated shedding is mediated by TNF-a-converting enzyme (TACE/ADAM17). Thus, cleavage of CA IX ectodomain appears to be a regulated process that responds to signal transduction-related stimuli and may contribute to adaptive changes in the protein composition of tumor cells and of their microenvironment. In this context, it is noteworthy that the extracellular domain of CA IX shed to body fluids is a promising circulating cancer marker for patient selection, and monitoring of treatment outcome (Carney 2007). Finally, posttranslational modifications, such as phosphorylation, regulate the CA IX-mediated signal transduction (Dorai et al. 2005) as well as the functional status of CA IX (the latter aspect has just been experimentally proven). Taken all data together, it can be concluded that control of CA9 gene expression is a complex process, which affects biogenesis of CA IX protein at several levels, including the functional activation (as discussed in the text below). It is also evident that hypoxic pathway is a major regulator of CA IX.

Role of CA IX in Cancer Hypoxic regulation of CA9 gene indicates that it is a part of the adaptive mechanisms of tumor cells exposed to hypoxia (and associated acidosis) and that the cells need the CA IX protein to survive these stresses typical for tumor microenvironment. In accord with this supposition, it has been demonstrated that CA IX protein is functionally involved in pH regulation and in cell adhesion–migration–invasion, which are important phenomena contributing to tumor progression. The need for increased pH regulation is related to metabolic changes triggered by hypoxia. Insufficient oxygen delivery to hypoxic tumor areas restricts energy production by oxidative phosphorylation and induces metabolic shift of hypoxic cells to glycolysis, which can produce ATP in the absence of oxygen. For this purpose, HIF-1 coordinates up-regulation of virtually all glycolytic enzymes and glucose transporters, as well as lactate dehydrogenase isoforms LDH-A and LHD5, which convert pyruvate to lactate, and also pyruvate dehydrogenase kinase 1, which prevents entry of pyruvate into the Krebs cycle and reinforces the glycolytic

4  Molecular Mechanisms Regulating Expression

69

metabolism (Brahimi-Horn and Pouyssegur 2007). Although glycolysis is less efficient in energy production than respiration, its metabolic intermediates can be utilized for biosynthetic reactions and this makes it useful for proliferating tumor cells in both anaerobic and aerobic conditions, in accord with the classical Warburg’s observation (Gatenby and Gillies 2004). For that reason, tumor cells depend on oncogenic metabolism, in which glycolysis predominates over respiration to extent depending on oxygen availability and activation of different oncogenic pathways. Glycolysis generates lactate as an end-product, but the oncogenic metabolism produces also excess of protons and CO2 (Helmlinger et  al. 2002). These acidic products have to be eliminated from tumor cells to maintain neutral intracellular pH and preserve proliferation and survival. The acidic waste then accumulates in the extracellular space because of its insufficient removal by inadequate tumor vasculature and causes extracellular acidosis (Stubbs et al. 2000). Acidosis develops especially in tumor regions that suffer from hypoxia. This is supported by correlation between the mean profiles of partial oxygen pressure and intratumoral pH values (Helmlinger et al. 1997), as well as by VHL/HIF pathway controlled expression of several pH regulating molecules (Karumanchi et  al. 2001). Importantly, acidic extracellular pH (pHe) has been associated with tumor progression via multiple effects including up-regulation of angiogenic factors and proteases, increased invasion, and impaired immune functions (Fukumura and Jain 2007; Raghunand et  al. 2003). In addition, it can influence the uptake of anticancer drugs and modulate the response of tumor cells to conventional therapy (Teicher 2009). Tumor cells utilize two main mechanisms to regulate their pH – lactate/proton extrusion and bicarbonate uptake (Fig. 4.3). Lactic acid and protons are extruded mostly by the H+/monocarboxylate transporters (MCT4 and MCT1) and the Na+/H+ exchanger (NHE1). Bicarbonate ions are imported by anion exchangers (e.g. AE2) or Na+/bicarbonate co-transporters (NBC1) and react with intracellular protons, thereby increasing the intracellular production of CO2, which diffuses across the plasma membrane to extracellular space. However, origin of the extracellular bicarbonate ions in an acidic pericellular milieu was unclear until CA IX was described as a component of hypoxic tumor phenotype (Stubbs et al. 2000). CA IX seems to be “designed” for this purpose, because its high catalytic activity is insensitive to lactate and reaches maximum at pH 6.5 typical for acidic tumor microenvironment (both aspects contrasting with other CA isoforms) (Innocenti et al. 2005, 2009). In addition, the catalytic domain of CA IX is localized on the cell surface and its entrance is exposed to extracellular space in the position suitable for accommodating CO2 substrate (Alterio et al. 2009). Moreover, CA IX activity is inhibited by bicarbonate suggesting that it operates mainly in catalysis of CO2 hydration to produce bicarbonate rather than to utilize it (Innocenti et  al. 2005). Finally, CA IX spatially and functionally cooperates with bicarbonate exchangers/ transporters in bicarbonate transport metabolon (Morgan et al. 2007). According to a recent concept, bicarbonate locally produced by CA IX from extracellular CO2 is directly consumed for transport and neutralization of intracellular pH in tumor cells, whereas proton produced in the same reaction remains

70

J. Pastorek and S. Pastorekova

Fig. 4.3  Role of CA IX as a component of pH regulatory machinery in cancer cell. Hypoxia induces expression of many proteins involved in metabolism and pH regulation including GLUT-1 and glycolytic enzymes. This causes shift to glycolysis that produces excess of lactate and protons. To preserve neutral intracellular pH, these acidic products are extruded by HIF-1-regulated monocarboxylate transporter (MCT4) and the Na+/H+-exchanger (NHE1) and acidify the extracellular microenvironment. Tumor cells also produce CO2 that diffuses across the plasma membrane. Extracellular CO2 is hydrated in a reaction catalyzed by CA IX, which produces protons and bicarbonate ions. Protons remain outside of cell and further enhance acidosis. Bicarbonate ions are delivered directly to bicarbonate transporters (BT) for accelerated inward transport to cytoplasm where they buffer protons and contribute to intracellular pH neutralization. This produces CO2 that leaves the cell by diffusion and may enter new round of hydration

outside of cells and increases extracellular acidosis (Fig. 4.3). In accord with this concept, CA IX role in pH regulation was demonstrated using different cellular models either with constitutive overexpression of CA IX or with suppression of CA IX by RNA interference. The first proof was obtained using MDCK cells grown in monolayer, in which CA IX could acidify the culture medium in hypoxia but not in normoxia, suggesting that hypoxia activates the catalytic performance of CA IX (Svastova et al. 2004). Additional evidence was obtained in three-dimensional multicellular spheroids generated from transfected HCT116 cells (Swietach et  al. 2008, 2009). There, the ectopic expression of CA IX resulted in less prominent core intracellular acidity and in enhanced core extracellular acidity. The data obtained in spheroids and interpreted in the framework

4  Molecular Mechanisms Regulating Expression

71

of tumor microenvironment indicate that CA IX facilitates CO2 excretion from cells and that this effect depends on the cellular CO2/lactic acid emission ratio set by local oxygenation and bicarbonate uptake (Swietach et al. 2009). Finally, in hypoxic LS174Tr tumor cells, CA IX was shown to contribute to extracellular acidification in response to a CO2 load and to maintenance of a neutral-alkaline intracellular pH (Chiche et  al. 2009b). This latest study also demonstrated that CA IX preserves tumor cell survival in a range of acidic extracellular pH (6.0– 6.8). Moreover, CA IX silencing led to a 40% reduction in xenograft tumor volume supporting the view that hypoxia-induced CA IX is a significant tumor pro-survival pH-regulating enzyme. Interestingly, parallel suppression of CA IX and XII caused dramatic 85% reduction in tumor growth, suggesting cooperation/ compensation of these transmembrane CA isoforms (although suppression of CA XII alone had no effect). The second function of CA IX is linked to cell adhesion and related processes, such as motility, migration, and invasion, which facilitate tumor progression. Again, the insight into this role of CA IX was obtained in the study of E-cadherin-mediated cell–cell adhesion in CA IX-transfected MDCK cells (Svastova et al. 2003). Ectopic expression of CA IX resulted in decreased intercellular contacts, reduced aggregation, and increased dissociation. Similarly to some oncoproteins, such as EFGR and Muc-1, CA IX acts via competitive interaction with b-catenin, thereby disconnecting E-cadherin from adhesion contacts with actin cytoskeleton. The capability of CA IX to interfere with the function of E-cadherin in those tumors that express both proteins might contribute to acquisition of more aggressive phenotype. Since decreased cell adhesion is a prerequisite of motile/migratory phenotype, it was tempting to examine possible involvement of CA IX. Recent experiments using various cell models have shown that expression of CA IX leads to increased scattering, better wound healing, and improved transwell migration both with and without HGF stimulation (manuscript in preparation). On the other hand, the suppression of CA IX by RNA interference decreases the wound healing capacity of tumor cells (Chiche et al. 2010). In this context, it is important to note that motility and migration require not only coordinated formation and release of cell adhesion contacts, digestion and reorganization of the extracellular matrix, but also intense local ion and water transport across the plasma membrane. All of these operations depend on intracellular and extracellular pH regulation. Thus, cell migration is significantly affected by mechanisms that protect tumor cells from hypoxia and acidosis (Stock and Schwab 2009). This knowledge is integrated into a model, in which protruding edge of migrating cell is occupied and driven by different components of ion transport and pH regulating machinery (including CA IX). On this basis, it is conceivable that the catalytic activity plays a considerable role in CA IX-induced de-adhesion, motility, and migration and that the PG domain, implicated in these processes, requires a cooperation with the CA domain (and possibly also with intracellular tail) of CA IX. Initial experimental data favor this idea (unpublished data).

72

J. Pastorek and S. Pastorekova

Clinical Value of CA IX Association with different solid tumors and up-regulation by hypoxia predisposes CA IX to serve as an intrinsic marker of hypoxia and prognostic indicator (Potter and Harris 2004). Moreover, expression pattern and functional involvement of CA IX in hypoxic tumor phenotype make it a suitable target for anticancer treatment (Pastorekova et  al. 2006). Both aspects already found a considerable support in clinical studies, in vivo models, and even in the clinical trials. When considering CA IX as a marker of hypoxia, it is important to take into account temporal parameters of its induction and stability. Based on cell culture data, CA IX protein appears on the cell surface several hours after exposure of cell culture to hypoxia, reaching maximum level at around 16 h (Rafajova et al. 2004). Due to its high stability, CA IX remains detectable for several days also in reoxygenated cells. Moreover, CA9 gene is responsive to a broad range of oxygen concentrations representing severe to moderate hypoxia, with optimum between 1 and 2% atmospheric O2 in culture conditions. These parameters differ from those of HIF-1a, which is both quickly induced and short-lived protein as well as from metabolic labels, such as pimonidazole, which accumulates only in severely hypoxic cells (as indicated by narrow perinecrotic incorporation pattern). Furthermore, HIF-1a staining only detects protein levels, but not the transactivation status, which actually determines induction of transcriptional targets. All these facts can explain why there is incomplete overlap between HIF-1a staining and CA IX staining in tumor tissue specimens, and why there are areas with presence of CA IX in absence of HIF-1a and vice versa. From the similar reasons, partial discrepancy exists also between CA IX and other HIF-1 targets, such as VEGF (which is secreted and considerably regulated by non-hypoxic pathways), GLUT-1 (which shows the highest response at 0.1% O2), and surely also other HIF-regulated molecules that all differ with respect to time and magnitude of induction, stability, and other characteristics (Lendahl et al. 2009). Such enormous variability suggests that there is no ideal intrinsic hypoxia marker, since each hypoxia-induced molecule or metabolic label represents only certain facet of the hypoxic response. Indeed, this can be taken as an advantage – for example, comparison of the spatial distribution of CA IX and pimonidazole allows for the discrimination between hypoxic and reoxygenated areas, which might have different clinical impact (Shin et al. 2007). Nevertheless, based on numerous studies it appears that CA IX is the best available indicator of chronic hypoxia and as such, it also seems to have prognostic and/or predictive value in certain tumor types, particularly derived from brain, lung, breast, colon, cervix, ovaria, etc. Although some groups reported absence of significant association with cancer prognosis, most of the published papers state that CA IX expression can be utilized as an independent predictor of poor outcome after therapy and indicator of worse cancer development (summarized in Table 4.1). In this respect, renal cell carcinomas (particularly of clear cell type) represent a special category of tumors, in which CA IX expression is frequent and high due to genetic defect in VHL gene, and absent or non-functional pVHL protein.

Independent prognostic factor for both progression-free survival and overall survival in high-risk early-stage cervical cancer Elevated expression associated with more frequent distant metastases in early-stage cervical cancer Strong correlation to metastasis, important indicator for more aggressive systemic therapy Up-regulated in hypoxic human cervical tumors and ssociated with poor prognosis

Associated with aggressive phenotype and chemo-resistance of basal-like breast tumors Independent prognostic factor with respect to relapse-free survival Correlates with poor prognosis in invasive breast carcinoma In both univariate and multivariate analysis, survival is significantly inferior in patients with invasive breast tumors expressing CA IX Independent prognostic marker in premenopausal patients with one to three positive lymph nodes and a putative marker of radiation resistance Negative predictor of treatment efficacy in ER-positive patients on the adjuvant tamoxifen after primary chemo-endocrine therapy Associated with a significantly shorter disease-free survival in patients with HIF-a-negative breast tumors Perinecrotic HIF-a overexpression with strong expression of CA IX and GLUT-1 associated with poor prognosis in invasive breast cancer Associated with worse relapse-free and overall survival in an unselected cohort of patients with invasive breast cancer

Cervical carcinoma

Breast carcinoma

Chia et al. (2001) (continued)

Vleugel et al. (2005)

Generali et al. (2006a)

Generali et al. (2006b)

Brennan et al. (2006)

Tan et al. (2009) Crabb et al. (2008) Trastour et al. (2007) Hussain et al. (2007)

Kim et al. (2006) Loncaster et al. (2001)

Kirkpatrick et al. (2008)

Liao et al. (2010)

Table 4.1  Overview of clinical studies on relationship of CA IX to cancer progression, prognosis, and/or treatment outcome (non-RCC tumors) Tumor type Proposed clinical value of CA IX Reference

4  Molecular Mechanisms Regulating Expression 73

Proposed clinical value of CA IX

Co-expression of CA IX and HIF-a associated with poor prognosis in oral squamous cell carcinoma patients Combined CA IX and HIF-a significantly predictive of overall survival in patients with advanced head-and-neck cancer Better local control in radiotherapy-treated laryngeal carcinoma patients having tumors with non-hypoxic profile (low CA IX, low HIF-a) Significant association with postoperative recurrence and poor overall survival in surgically treated oral squamous cell carcinoma High CA IX and Ki-67 expression associated with poorer overall survival and shorter diseasefree survival in patients with squamous cell carcinoma of the tongue Independent correlation with prognosis when combined with GLUT-1

Poorer outcome in patients with oligodendroglial tumors Significant parameter for the survival of patients with grades II/III astrocytic gliomas Poor prognosis in astrocytic tumor patients

Expression by cancer-associated fibroblasts predicts lower survival rate in patients with lung adenocarcinoma Full-length CA IX transcript associated with lymph node involvement, tumor stage, and poor prognosis Indicates poor prognosis for patients with stages I+II lung adenocarcinoma The most reliable hypoxia marker for predicting tumor aggressiveness Indicator for poor disease-free survival in early-stage non-small-cell lung cancer Associated with a poor prognosis in non-small-cell lung cancer

Intensity predicts disease-free survival and disease-specific survival in patients with rectal cancer

Correlates with extremely poor prognosis in esophageal and gastric adenocarcinomas Has a tripartite role as a diagnostic, prognostic, and therapeutic molecular marker in urothelial carcinoma of the bladder Correlates with clinical outcome in transitional cell carcinoma of the bladder Predicts for poor prognosis in patients with deep, large, and high-grade soft tissue sarcoma

Head-and-neck cancer

Brain tumors

Lung carcinoma

Colorectal carcinoma

Other tumors

Table 4.1  (continued)

Tumor type

Hussain et al. (2004) Maseide et al. (2004)

Driessen et al. (2006) Klatte et al. (2009)

Korkeila et al. (2009)

Kon-no et al. (2006) Kim et al. (2005) Kim et al. (2004) Swinson et al. (2003)

Malentacchi et al. (2009)

Nakao et al. (2009)

Jarvela et al. (2008) Korkolopoulou et al. (2007) Haapasalo et al. (2006)

De Schutter et al. (2005)

Kim et al. (2007)

Choi et al. 2008)

Schrijvers et al. (2008)

Kappler et al. (2008)

Eckert et al. (2010)

Reference

74 J. Pastorek and S. Pastorekova

4  Molecular Mechanisms Regulating Expression

75

Resulting activation of HIF pathway leads to constitutive up-regulation of HIF targets including CA IX independently of physiological hypoxia (Mandriota et al. 2002). In contrast to other tumor types, CA IX expression in RCC decreases with increasing tumor grade and stage. This is apparently because CA IX is a target of HIF-1a that predominates early in VHL disease but is later displaced by tumor growth-promoting isoform HIF-2a (Raval et al. 2005). Correspondingly, decreased CA IX levels are independently associated with poor prognosis in advanced RCC (Bui et al. 2003). However, levels of CA IX can be upregulated by interferons (IFN-a and IFN-g), which can increase the therapeutic responses in 5–25% of patients with metastatic RCC (Brouwers et al. 2003b). Furthermore, high CA IX expression was suggested as an important predictor of better outcome in RCC patients receiving interleukin-2-based therapy (Atkins et al. 2005). In several recent reviews, CA IX has been proposed to be the most significant molecular marker described in kidney cancer (Pantuck et  al. 2008; Signoretti et al. 2008). Assessment of CA IX expression in biopsies offers a new means for stratification of cancer patients and rational selection of appropriate treatment strategy – an approach highly desirable in recent clinical oncology. However, in certain phases of cancer management, the patients would benefit from non-invasive monitoring of CA IX shed to blood circulation. Data published so far suggest that serum CA IX levels in RCC patients correlate with tumor size and stage of the disease, and support the use of soluble CA IX as a biomarker of progression and postoperative recurrence (Zhou et al. 2009). Furthermore, in vivo imaging of CA IX is recently coming to the scene as a very promising approach for pre-surgical diagnosis of RCC (Divgi et al. 2007; Zhang 2008). The imaging tool called REDECTANE® developed by WILEX AG company is based on radioactively labeled monoclonal antibody G250 originally generated by Oosterwijk and coworkers (Lam et al. 2005). The radiolabeled antibody WX-G250 targets clear cell renal cell carcinoma and accumulates in the tumor tissue via binding to CA IX followed by receptor-mediated internalization (Durrbach et  al. 1999). The G250 accumulation can be visualized by means of positron emission tomography (PET) (Divgi et al. 2007). A proof-of-concept study showed that REDECTANE® could predict clear cell RCC with a specificity of 100% and a sensitivity of 94%. Moreover, Phase II clinical study demonstrated that the combination of REDECTANE® with PET and computer tomography (CT) versus the standard use of CT alone could improve the diagnosis of renal masses (http://www.wilex.com).

CA IX Targeting Strategies CA IX usefulness as therapeutic target stems from the following facts: (a) strong association with different tumors and low or absent expression in normal tissues, (b) induction by hypoxia and correlation with aggressive phenotype, (c) localization

76

J. Pastorek and S. Pastorekova

on the cell surface allowing interaction with antibodies or drugs, and (d) direct functional involvement in hypoxic tumor phenotype. These facts create a basis for two principal strategies of targeting CA IX-expressing tumor cells – immunotherapy with specific antibodies inducing cytotoxicity and functional inhibition of the CA IX catalytic activity leading to disruption of pH control. The first approach has been consistently studied in pre-clinical and clinical experiments in patients with renal cell carcinomas. Although advanced RCC show reduced CA IX expression, the cut-off was set to 85% (i.e. the expression is considered low when less than 85% of cells display CA IX) and this is still a very high proportion of targetable cells. Again, the major tool here is the monoclonal antibody G250 (Brouwers et  al. 2003a; Mulders et  al. 2004). Chimeric G250 (composed of variable regions of murine G250 and constant regions derived from human IgG) is recently produced by WILEX AG under commercial name RENCAREX®. Its effector mechanism is ADCC (antibody-dependent cell cytotoxicity). In Phase I and II studies, RENCAREX® has shown good safety and tolerability and a promising efficacy profile. The purpose of the ongoing Phase III study (called ARISER) is to assess the effect of adjuvant treatment with RENCAREX® on disease-free survival and overall survival in RCC patients with a high risk of recurrence following surgery (nephrectomy). Additional monoclonal antibodies potentially suitable for immunotherapeutic purposes are under development. Some of them display similar binding properties and capacity to internalize as G250 and exhibit anticancer effects in animal models (Zatovicova et al. in press). The second approach – functional inhibition of cancer-related carbonic anhydrases – is only at early stage of investigation and is based on carbonic anhydrase inhibitors (sulfonamides and their derivatives) that are mostly produced and characterized by Supuran and coworkers, who were the first to introduce the concept of their use as anticancer drugs (Supuran 2007). Sulfonamides and related compounds generally show non-selective inhibitory activity toward different CA isoenzymes including CA IX, which can be inhibited by a broad range of compounds (see Table 4.2). Interestingly, efficient CA IX inhibition was observed also with celecoxib and valdecoxib, sulfonamide inhibitors of cyclooxygense 2 (COX-2, a key enzyme of arachidonic acid metabolism involved in colorectal carcinoma), and with inhibitors of steroid sulfatase (implicated in production of active steroids in breast cancer) (Weber et al. 2004; Winum et al. 2003a). This may suggest that the mode of action of these drugs could involve targeting of CAs. As demonstrated by several recent papers, increased selectivity toward CA IX can be successfully achieved by modulating the physical and chemical properties of the compounds via attachment of different side chains and other modifications. Certain alterations can introduce or improve the membrane impermeability so that the inhibitor can bind only or predominantly to extracellularly exposed CAs. The other changes can modify the size or surface topology that fits better into active site cavity of CA IX than into other isoforms. Some types of modifications increase the  efficiency of inhibition so that they work even at subnanomolar concentrations. This approach was used for development of several

Acetazolamide Aromatic and heterocyclic sulfonamides Halogenated sulfonamides Aliphatic sulfamates Sulfamates or bis-sulfamates (+ steroid sulfatase inhibitors) E7070 (indisulam) anticancer drug in clinical development Non-steroidal anti-inflammatory inhibitors of COX-2 Positively-charged pyridinium derivatives of sulfanilamides Polyfluorinated sulfonamides Sulfonamides incorporating 1,2,4-triazine moieties Sulfonamides derived from 4-isothiocyanato-benzolamide Carboxylates Bis-sulfamates Sulfonamides incorporating hydrazino moieties Boron-containing sulfamides, sulfonamides, and sulfamates

Abbate et al. (2004) Weber et al. (2004)

Potent inhibition of CA I, II, IX in the range of 15–31 nM Unexpected nanomolar inhibition of CA I, II, IV, IX by celecoxib and valdecoxib First selective, membrane-impermeant inhibitors targeting CA IX

Effective inhibitors of CA IX suitable for boron neutron capture therapy targeting hypoxic tumors

The first subnanomolar and selective CA IX inhibitor CA IX-selective inhibitors with selectivity ratios for CA IX over CA II inhibition in the range of 166–706 Excellent inhibitory properties against CA IX with inhibition constants in the range of 3.2–23 nM Lactate insensitivity of CA IX Low nanomolar CA IX inhibitors with good selectivity ratio Several low nanomolar CA IX inhibitors

Ilies et al. (2003) Winum et al. (2003b) Winum et al. (2003a)

CA IX inhibition profile different from CA I, II, IV Efficient CA IX inhibition in the range of 9–23 nM Efficient CA IX inhibition in the range of 18–63 nM

(continued)

Winum et al. (2005a)

Innocenti et al. (2005) Winum et al. (2005c) Winum et al. (2005b)

Cecchi et al. (2004)

Vullo et al. (2004) Garaj et al. (2004)

Pastorekova et al. (2004a)

Parkkila et al. (2000) Vullo et al. (2003)

Inhibition of in vitro migration of renal carcinoma cells Several nanomolar CA IX inhibitors

Table 4.2  Summary of sulfonamide inhibitors, their derivatives, and other compounds tested against different CA isoforms including CA IX Inhibitor type Inhibition properties Reference

4  Molecular Mechanisms Regulating Expression 77

Imatinib (Glivec/Gleevec) and nilotinib (Tasigna)

Inhibitor-coated gold nanoparticles Coumarin derivatives

Bioreductive nitro-containing sulfonamides Biphenylsulfonamides

Mercaptans increase the inhibitory power (52.8- to 243-fold) over the corresponding oxidized derivatives Several potent and selective inhibitors of CA IX

Aromatic sulfonamides activatable in hypoxic tumors Glycoconjugate benzene sulfonamides Antiepileptic drug sulthiame Glycosyl-thioureido-sulfonamides Potent inhibitor of CA II, VII, IX, XII (6–56 nM) Low nM binding to CA IX and XII, high selectivity ratios in the range of 107–955 Some CA IX inhibitors with excellent selectivity ratios (in the range of 10–1395) Inhibitors of CA IX with cytotoxic activity against human colon, lung, and breast cancer cell lines Selective inhibition of CA IX (membrane impermeant) Bind at the entrance of the active site cavity, one of them highly selective for CA IX Inhibited all 13 catalytically active CA isoforms in the range of 4.1 nM–20.2 mM

Inhibition properties

Table 4.2  (continued)

Inhibitor type

Parkkila et al. (2009)

Stiti et al. (2008) Maresca et al. (2010)

Morsy et al. (2009)

D’Ambrosio et al. (2008)

Temperini et al. (2007) Smaine et al. (2007)

Wilkinson et al. (2006)

Saczewski et al. (2006)

Reference

78 J. Pastorek and S. Pastorekova

4  Molecular Mechanisms Regulating Expression

79

compounds with a reasonable selectivity ratio favoring inhibition activities against CA IX compared to other isoforms, in particular CA II (Table 4.2). Another strategy was used to generate hypoxia-activated inhibitors. Different 2-mercapto-substituted-benzenesulfonamides and their disulfides/sulfones showed consistently increased inhibitory power (52.8- to 243-fold) over the corresponding oxidized (S–S type) derivatives (Saczewski et  al. 2006). The best representatives out of these differentially acting derivatives can serve as lead compounds for further development of CA IX -specific inhibitors with therapeutic potential against cancer. Since the inhibition data were obtained using the purified recombinant catalytic domain of CA IX, experiments with native protein in cellular context are needed to determine their effective concentrations and consequences of their application. Noteworthy, fluorescein-conjugated carbonic anhydrase inhibitor thioureidohomosulfanilamide (FITC-CAI) was found to bind only to hypoxic cells expressing CA IX, but neither to CA IX-positive normoxic cells nor the CA IX-negative controls (Cecchi et al. 2005; Svastova et al. 2004). Moreover, the FITC-CAI binding was perturbed by reoxygenation (Dubois et al. 2007). This interesting finding suggests that CA IX active site is accessible to sulfonamide only under hypoxic conditions. Results of in  vivo imaging confirm previous in vitro data demonstrating that CAI binding and retention requires exposure to hypoxia (Dubois et al. 2009). Thus, labeled sulfonamides may provide a powerful tool to visualize hypoxic tumors. This may represent a promising approach to patient selection for CA IX-directed therapies based on the functional inhibition of CA IX. CA inhibitors can be used as anticancer drugs to compromise the survival of tumor cells due to perturbed pH control, or in combination with conventional chemotherapeutics to improve their uptake and efficacy due to modulated pH gradient. The first option is based on significance of the proper pH control in tumor cells, particularly in those that suffer from hypoxia. It is well established that extracellular acidosis supports cancer progression, whereas neutral intracellular pH is required for survival of hypoxic cancer cells (Stubbs et  al. 2000). Inhibition or reduction of extracellular acidosis by carbonic anhydrase inhibitors thus offers a possible therapeutic approach to retardation of tumor progression (Pouyssegur et al. 2006). On the other hand, inhibition of CA activity can lead to inability of tumor cells to neutralize their intracellular pH and adapt to metabolic stress. In fact, several sulfonamide derivatives were capable of reducing an extracellular acidification and intracellular neutralization mediated by CA IX, suggesting that these compounds interfere with the activity of CA IX in cell culture model (Svastova et al. 2004; Swietach et al. 2009). In addition, changes in extracellular pH can affect therapeutic outcome of conventional chemotherapy by influencing the uptake of weakly electrolytic anticancer drugs (Kozin et al. 2001; Vukovic and Tannock 1997). In particular, reduction of extracellular acidosis can increase the uptake and cytotoxic effects of weakly basic drugs, including doxorubicin. Interestingly, chronic ingestion of sodium bicarbonate solution was shown to enhance the capacity of doxorubicin to decrease the tumor size (Raghunand et al. 2003).

80

J. Pastorek and S. Pastorekova

In line with these suppositions, older but forward-looking study showed that acetazolamide, a classical CA inhibitor, reduced in  vivo growth of tumor when given alone and produced additive tumor growth delays when administered in combination with various chemotherapeutic compounds (Teicher et al. 1993). This was recently supported using more sophisticated approach, based on two acetazolamide derivatives containing either a charged fluorophore or an albumin-binding moiety, which restrict binding to carbonic anhydrase IX and XII present on tumor cells (Ahlskog et al. 2009a). In vivo studies showed the preferential targeting of tumor cells by the fluorescent acetazolamide derivative and the ability of the albuminbinding acetazolamide derivative to cause tumor retardation in a SK-RC-52 (RCC) xenograft model of cancer. Additional in vivo experiments with promising preliminary results are underway.

Conclusion Our present knowledge on CA IX clearly suggests that this hypoxia-induced molecule is an important component of the tumor microenvironment and that many tumors need CA IX to survive microenvironmental stresses. This addiction of tumor cells to CA IX makes it a useful marker of hypoxia, prognostic indicator, and therapeutic target. Although main strategies of diagnostic and therapeutic exploitation of CA IX have been rationally designed, their concepts were experimentally proven, practically evaluated (in part), and several specific tools are available; further preclinical and clinical studies are necessary to translate all proposed applications to the benefit of cancer patients.

References Abbate F, Casini A, Owa T, Scozzafava A, Supuran CT (2004) Carbonic anhydrase inhibitors: E7070, a sulfonamide anticancer agent, potently inhibits cytosolic isozymes I and II, and transmembrane, tumor-associated isozyme IX. Bioorg Med Chem Lett 14:217–223 Ahlskog JK, Dumelin CE, Trussel S, Marlind J, Neri D (2009a) In vivo targeting of tumorassociated carbonic anhydrases using acetazolamide derivatives. Bioorg Med Chem Lett 19:4851–4856 Ahlskog JK, Schliemann C, Marlind J, Qureshi U, Ammar A, Pedley RB, Neri D (2009b) Human monoclonal antibodies targeting carbonic anhydrase IX for the molecular imaging of hypoxic regions in solid tumours. Br J Cancer 101:645–657 Alterio V, Hilvo M, Di Fiore A, Supuran CT, Pan P, Parkkila S, Scaloni A, Pastorek J, Pastorekova S, Pedone C, Scozzafava A, Monti SM, De Simone G (2009) Crystal structure of the catalytic domain of the tumor-associated human carbonic anhydrase IX. Proc Natl Acad Sci USA 106:16233–16238 Ashida S, Nishimori I, Tanimura M, Onishi S, Shuin T (2002) Effects of von Hippel-Lindau gene mutation and methylation status on expression of transmembrane carbonic anhydrases in renal cell carcinoma. J Cancer Res Clin Oncol 128:561–568

4  Molecular Mechanisms Regulating Expression

81

Atkins M, Regan M, McDermott D, Mier J, Stanbridge E, Youmans A, Febbo P, Upton M, Lechpammer M, Signoretti S (2005) Carbonic anhydrase IX expression predicts outcome of interleukin 2 therapy for renal cancer. Clin Cancer Res 11:3714–3721 Barathova M, Takacova M, Holotnakova T, Gibadulinova A, Ohradanova A, Zatovicova M, Hulikova A, Kopacek J, Parkkila S, Supuran CT, Pastorekova S, Pastorek J (2008) Alternative splicing variant of the hypoxia marker carbonic anhydrase IX expressed independently of hypoxia and tumour phenotype. Br J Cancer 98:129–136 Bartosova M, Parkkila S, Pohlodek K, Karttunen TJ, Galbavy S, Mucha V, Harris AL, Pastorek J, Pastorekova S (2002) Expression of carbonic anhydrase IX in breast is associated with malignant tissues and is related to overexpression of c-erbB2. J Pathol 197:314–321 Brahimi-Horn MC, Pouyssegur J (2007) Hypoxia in cancer cell metabolism and pH regulation. Essays Biochem 43:165–178 Brennan DJ, Jirstrom K, Kronblad A, Millikan RC, Landberg G, Duffy MJ, Ryden L, Gallagher WM, O’Brien SL (2006) CA IX is an independent prognostic marker in premenopausal breast cancer patients with one to three positive lymph nodes and a putative marker of radiation resistance. Clin Cancer Res 12:6421–6431 Brouwers AH, Buijs WC, Oosterwijk E, Boerman OC, Mala C, De Mulder PH, Corstens FH, Mulders PF, Oyen WJ (2003a) Targeting of metastatic renal cell carcinoma with the chimeric monoclonal antibody G250 labeled with (131)I or (111)In: an intrapatient comparison. Clin Cancer Res 9:3953S–3960S Brouwers H, Frielink C, Oosterwijk E, Oyen WJ, Corstens FH, Boerman OC (2003b) Interferons can upregulate the expression of the tumor associated antigen G250-MN/CA IX, a potential target for (radio)immunotherapy of renal cell carcinoma. Cancer Biother Radiopharm 18:539–547 Bui MH, Seligson D, Han KR, Pantuck AJ, Dorey FJ, Huang Y, Horvath S, Leibovich BC, Chopra S, Liao SY, Stanbridge E, Lerman MI, Palotie A, Figlin RA, Belldegrun AS (2003) Carbonic anhydrase IX is an independent predictor of survival in advanced renal clear cell carcinoma: implications for prognosis and therapy. Clin Cancer Res 9:802–811 Camenisch G, Stroka DM, Gassmann M, Wenger RH (2001) Attenuation of HIF-1 DNA-binding activity limits hypoxia-inducible endothelin-1 expression. Pflugers Arch 443:240–249 Carney WP (2007) Circulating oncoproteins HER2/neu, EGFR and CAIX (MN) as novel cancer biomarkers. Expert Rev Mol Diagn 7:309–319 Casey JR (2006) Why bicarbonate? Biochem Cell Biol 84:930–939 Cecchi A, Hulikova A, Pastorek J, Pastorekova S, Scozzafava A, Winum JY, Montero JL, Supuran CT (2005) Carbonic anhydrase inhibitors. Design of fluorescent sulfonamides as probes of tumor-associated carbonic anhydrase IX that inhibit isozyme IX-mediated acidification of hypoxic tumors. J Med Chem 48:4834–4841 Cecchi A, Winum JY, Innocenti A, Vullo D, Montero JL, Scozzafava A, Supuran CT (2004) Carbonic anhydrase inhibitors: synthesis and inhibition of cytosolic/tumor-associated carbonic anhydrase isozymes I, II, and IX with sulfonamides derived from 4-isothiocyanatobenzolamide. Bioorg Med Chem Lett 14:5775–5780 Coleman ML, Ratcliffe PJ (2007) Oxygen sensing and hypoxia-induced responses. Essays Biochem 43:1–15 Crabb SJ, Bajdik CD, Leung S, Speers CH, Kennecke H, Huntsman DG, Gelmon KA (2008) Can clinically relevant prognostic subsets of breast cancer patients with four or more involved axillary lymph nodes be identified through immunohistochemical biomarkers? A tissue microarray feasibility study. Breast Cancer Res 10:R6 D’Ambrosio K, Vitale RM, Dogne JM, Masereel B, Innocenti A, Scozzafava A, De Simone G, Supuran CT (2008) Carbonic anhydrase inhibitors: bioreductive nitro-containing sulfonamides with selectivity for targeting the tumor associated isoforms IX and XII. J Med Chem 51: 3230–3237 De Schutter H, Landuyt W, Verbeken E, Goethals L, Hermans R, Nuyts S (2005) The prognostic value of the hypoxia markers CA IX and GLUT 1 and the cytokines VEGF and IL 6 in head and neck squamous cell carcinoma treated by radiotherapy +/– chemotherapy. BMC Cancer 5:42

82

J. Pastorek and S. Pastorekova

Divgi CR, Pandit-Taskar N, Jungbluth AA, Reuter VE, Gonen M, Ruan S, Pierre C, Nagel A, Pryma DA, Humm J, Larson SM, Old LJ, Russo P (2007) Preoperative characterisation of clear-cell renal carcinoma using iodine-124-labelled antibody chimeric G250 (124I-cG250) and PET in patients with renal masses: a phase I trial. Lancet Oncol 8:304–310 Dorai T, Sawczuk IS, Pastorek J, Wiernik PH, Dutcher JP (2005) The role of carbonic anhydrase IX overexpression in kidney cancer. Eur J Cancer 41:2935–2947 Driessen A, Landuyt W, Pastorekova S, Moons J, Goethals L, Haustermans K, Nafteux P, Penninckx F, Geboes K, Lerut T, Ectors N (2006) Expression of carbonic anhydrase IX (CA IX), a hypoxia-related protein, rather than vascular-endothelial growth factor (VEGF), a pro-angiogenic factor, correlates with an extremely poor prognosis in esophageal and gastric adenocarcinomas. Ann Surg 243:334–340 Dubois L, Douma K, Supuran CT, Chiu RK, van Zandvoort MA, Pastorekova S, Scozzafava A, Wouters BG, Lambin P (2007) Imaging the hypoxia surrogate marker CA IX requires expression and catalytic activity for binding fluorescent sulfonamide inhibitors. Radiother Oncol 83:367–373 Dubois L, Lieuwes NG, Maresca A, Thiry A, Supuran CT, Scozzafava A, Wouters BG, Lambin P (2009) Imaging of CA IX with fluorescent labelled sulfonamides distinguishes hypoxic and (re)-oxygenated cells in a xenograft tumour model. Radiother Oncol 92:423–428 Durrbach A, Angevin E, Poncet P, Rouleau M, Chavanel G, Chapel A, Thierry D, Gorter A, Hirsch R, Charpentier B, Senik A, Hirsch F (1999) Antibody-mediated endocytosis of G250 tumor-associated antigen allows targeted gene transfer to human renal cell carcinoma in vitro. Cancer Gene Ther 6:564–571 Eckert AW, Lautner MH, Schutze A, Bolte K, Bache M, Kappler M, Schubert J, Taubert H, Bilkenroth U (2010) Co-expression of Hif1alpha and CAIX is associated with poor prognosis in oral squamous cell carcinoma patients. J Oral Pathol Med 39:313–317 Fukumura D, Jain RK (2007) Tumor microenvironment abnormalities: causes, consequences, and strategies to normalize. J Cell Biochem 101:937–949 Garaj V, Puccetti L, Fasolis G, Winum JY, Montero JL, Scozzafava A, Vullo D, Innocenti A, Supuran CT (2004) Carbonic anhydrase inhibitors: synthesis and inhibition of cytosolic/ tumor-associated carbonic anhydrase isozymes I, II, and IX with sulfonamides incorporating 1,2,4-triazine moieties. Bioorg Med Chem Lett 14:5427–5433 Gatenby RA, Gillies RJ (2004) Why do cancers have high aerobic glycolysis? Nat Rev Cancer 4:891–899 Generali D, Berruti A, Brizzi MP, Campo L, Bonardi S, Wigfield S, Bersiga A, Allevi G, Milani M, Aguggini S, Gandolfi V, Dogliotti L, Bottini A, Harris AL, Fox SB (2006a) Hypoxiainducible factor-1alpha expression predicts a poor response to primary chemoendocrine therapy and disease-free survival in primary human breast cancer. Clin Cancer Res 12: 4562–4568 Generali D, Fox SB, Berruti A, Brizzi MP, Campo L, Bonardi S, Wigfield SM, Bruzzi P, Bersiga A, Allevi G, Milani M, Aguggini S, Dogliotti L, Bottini A, Harris AL (2006b) Role of carbonic anhydrase IX expression in prediction of the efficacy and outcome of primary epirubicin/ tamoxifen therapy for breast cancer. Endocr Relat Cancer 13:921–930 Gut MO, Parkkila S, Vernerova Z, Rohde E, Zavada J, Hocker M, Pastorek J, Karttunen T, Gibadulinova A, Zavadova Z, Knobeloch KP, Wiedenmann B, Svoboda J, Horak I, Pastorekova S (2002) Gastric hyperplasia in mice with targeted disruption of the carbonic anhydrase gene Car9. Gastroenterology 123:1889–1903 Haapasalo J, Nordfors K, Jarvela S, Bragge H, Rantala I, Parkkila AK, Haapasalo H, Parkkila S (2007) Carbonic anhydrase II in the endothelium of glial tumors: a potential target for therapy. Neuro Oncol 9:308–313 Haapasalo JA, Nordfors KM, Hilvo M, Rantala IJ, Soini Y, Parkkila AK, Pastorekova S, Pastorek J, Parkkila SM, Haapasalo HK (2006) Expression of carbonic anhydrase IX in astrocytic tumors predicts poor prognosis. Clin Cancer Res 12:473–477 Harris AL (2002) Hypoxia – a key regulatory factor in tumour growth. Nat Rev Cancer 2:38–47

4  Molecular Mechanisms Regulating Expression

83

Helmlinger G, Sckell A, Dellian M, Forbes NS, Jain RK (2002) Acid production in glycolysisimpaired tumors provides new insights into tumor metabolism. Clin Cancer Res 8: 1284–1291 Helmlinger G, Yuan F, Dellian M, Jain RK (1997) Interstitial pH and pO2 gradients in solid tumors in  vivo: high-resolution measurements reveal a lack of correlation. Nat Med 3: 177–182 Hewett-Emmett D, Tashian RE (1996) Functional diversity, conservation, and convergence in the evolution of the alpha-, beta-, and gamma-carbonic anhydrase gene families. Mol Phylogenet Evol 5:50–77 Hilvo M, Baranauskiene L, Salzano AM, Scaloni A, Matulis D, Innocenti A, Scozzafava A, Monti SM, Di Fiore A, De Simone G, Lindfors M, Janis J, Valjakka J, Pastorekova S, Pastorek J, Kulomaa MS, Nordlund HR, Supuran CT, Parkkila S (2008) Biochemical characterization of CA IX, one of the most active carbonic anhydrase isozymes. J Biol Chem 283:27799–27809 Hirota K, Semenza GL (2005) Regulation of hypoxia-inducible factor 1 by prolyl and asparaginyl hydroxylases. Biochem Biophys Res Commun 338:610–616 Holotnakova T, Tylkova L, Takacova M, Kopacek J, Petrik J, Pastorekova S, Pastorek, J (2010) Role of the HBx oncoprotein in carbonic anhydrase 9 induction. J Med Virol 82:32–40 Hulikova A, Zatovicova M, Svastova E, Ditte P, Brasseur R, Kettmann R, Supuran CT, Kopacek J, Pastorek J, Pastorekova S (2009) Intact intracellular tail is critical for proper functioning of the tumor-associated, hypoxia-regulated carbonic anhydrase IX. FEBS Lett 583:3563–3568 Hussain SA, Ganesan R, Reynolds G, Gross L, Stevens A, Pastorek J, Murray PG, Perunovic B, Anwar MS, Billingham L, James ND, Spooner D, Poole CJ, Rea DW, Palmer DH (2007) Hypoxia-regulated carbonic anhydrase IX expression is associated with poor survival in patients with invasive breast cancer. Br J Cancer 96:104–109 Hussain SA, Palmer DH, Ganesan R, Hiller L, Gregory J, Murray PG, Pastorek J, Young L, James ND (2004) Carbonic anhydrase IX, a marker of hypoxia: correlation with clinical outcome in transitional cell carcinoma of the bladder. Oncol Rep 11:1005–1010 Chia SK, Wykoff CC, Watson PH, Han C, Leek RD, Pastorek J, Gatter KC, Ratcliffe P, Harris AL (2001) Prognostic significance of a novel hypoxia-regulated marker, carbonic anhydrase IX, in invasive breast carcinoma. J Clin Oncol 19:3660–3668 Chiang WL, Chu SC, Yang SS, Li MC, Lai JC, Yang SF, Chiou HL, Hsieh YS (2002) The aberrant expression of cytosolic carbonic anhydrase and its clinical significance in human non-small cell lung cancer. Cancer Lett 188:199–205 Chiche J, Ilc K, Brahimi-Horn MC, Pouyssegur J (2010) Membrane-bound carbonic anhydrases are key pH regulators controlling tumor growth and cell migration. Adv Enzyme Regul 50:20–33 Chiche J, Ilc K, Laferriere J, Trottier E, Dayan F, Mazure NM, Brahimi-Horn MC, Pouyssegur J (2009b) Hypoxia-inducible carbonic anhydrase IX and XII promote tumor cell growth by counteracting acidosis through the regulation of the intracellular pH. Cancer Res 69:358–368 Cho M, Grabmaier K, Kitahori Y, Hiasa Y, Nakagawa Y, Uemura H, Hirao Y, Ohnishi T, Yoshikawa K, Ooesterwijk E (2000) Activation of the MN/CA9 gene is associated with hypomethylation in human renal cell carcinoma cell lines. Mol Carcinog 27:184–189 Choi SW, Kim JY, Park JY, Cha IH, Kim J, Lee S (2008) Expression of carbonic anhydrase IX is associated with postoperative recurrence and poor prognosis in surgically treated oral squamous cell carcinoma. Hum Pathol 39:1317–1322 Ihnatko R, Kubes M, Takacova M, Sedlakova O, Sedlak J, Pastorek J, Kopacek J, Pastorekova S (2006) Extracellular acidosis elevates carbonic anhydrase IX in human glioblastoma cells via transcriptional modulation that does not depend on hypoxia. Int J Oncol 29:1025–1033 Ilies MA, Vullo D, Pastorek J, Scozzafava A, Ilies M, Caproiu MT, Pastorekova S, Supuran CT (2003) Carbonic anhydrase inhibitors. Inhibition of tumor-associated isozyme IX by halogenosulfanilamide and halogenophenylaminobenzolamide derivatives. J Med Chem 46:2187–2196 Innocenti A, Pastorekova S, Pastorek J, Scozzafava A, De Simone G, Supuran CT (2009) The proteoglycan region of the tumor-associated carbonic anhydrase isoform IX acts as anintrinsic

84

J. Pastorek and S. Pastorekova

buffer optimizing CO2 hydration at acidic pH values characteristic of solid tumors. Bioorg Med Chem Lett 19:5825–5828 Innocenti A, Vullo D, Scozzafava A, Casey JR, Supuran C (2005) Carbonic anhydrase inhibitors. Interaction of isozymes I, II, IV, V, and IX with carboxylates. Bioorg Med Chem Lett 15:573–578 Ivanov S, Liao SY, Ivanova A, Danilkovitch-Miagkova A, Tarasova N, Weirich G, Merrill MJ, Proescholdt MA, Oldfield EH, Lee J, Zavada J, Waheed A, Sly W, Lerman MI, Stanbridge EJ (2001) Expression of hypoxia-inducible cell-surface transmembrane carbonic anhydrases in human cancer. Am J Pathol 158:905–919 Ivanov SV, Kuzmin I, Wei MH, Pack S, Geil L, Johnson BE, Stanbridge EJ, Lerman MI (1998) Down-regulation of transmembrane carbonic anhydrases in renal cell carcinoma cell lines by wild-type von Hippel-Lindau transgenes. Proc Natl Acad Sci USA 95:12596–12601 Jakubickova L, Biesova Z, Pastorekova S, Kettmann R, Pastorek J (2005) Methylation of the CA9 promoter can modulate expression of the tumor-associated carbonic anhydrase IX in dense carcinoma cell lines. Int J Oncol 26:1121–1127 Jarvela S, Parkkila S, Bragge H, Kahkonen M, Parkkila AK, Soini Y, Pastorekova S, Pastorek J, Haapasalo H (2008) Carbonic anhydrase IX in oligodendroglial brain tumors. BMC Cancer 8:1 Kaluz S, Kaluzova M, Chrastina A, Olive PL, Pastorekova S, Pastorek J, Lerman MI, Stanbridge EJ (2002) Lowered oxygen tension induces expression of the hypoxia marker MN/carbonic anhydrase IX in the absence of hypoxia-inducible factor 1 alpha stabilization: a role for phosphatidylinositol 3¢-kinase. Cancer Res 62:4469–4477 Kaluz S, Kaluzova M, Liao SY, Lerman M, Stanbridge EJ (2009) Transcriptional control of the tumor- and hypoxia-marker carbonic anhydrase 9: a one transcription factor (HIF-1) show? Biochim Biophys Acta 1795:162–172 Kaluz S, Kaluzova M, Opavsky R, Pastorekova S, Gibadulinova A, Dequiedt F, Kettmann R, Pastorek J (1999) Transcriptional regulation of the MN/CA 9 gene coding for the tumorassociated carbonic anhydrase IX. Identification and characterization of a proximal silencer element. J Biol Chem 274:32588–32595 Kappler M, Taubert H, Holzhausen HJ, Reddemann R, Rot S, Becker A, Kuhnt T, Dellas K, Dunst J, Vordermark D, Hansgen G, Bache M (2008) Immunohistochemical detection of HIF-1alpha and CAIX in advanced head-and-neck cancer. Prognostic role and correlation with tumor markers and tumor oxygenation parameters. Strahlenther Onkol 184:393–399 Karumanchi SA, Jiang L, Knebelmann B, Stuart-Tilley AK, Alper SL, Sukhatme VP (2001) VHL tumor suppressor regulates Cl-/HCO3- exchange and Na+/H+ exchange activities in renal carcinoma cells. Physiol Genomics 5:119–128 Kim JY, Shin HJ, Kim TH, Cho KH, Shin KH, Kim BK, Roh JW, Lee S, Park SY, Hwang YJ, Han IO (2006) Tumor-associated carbonic anhydrases are linked to metastases in primary cervical cancer. J Cancer Res Clin Oncol 132:302–308 Kim SJ, Rabbani ZN, Dewhirst MW, Vujaskovic Z, Vollmer RT, Schreiber EG, Oosterwijk E, Kelley MJ (2005) Expression of HIF-1alpha, CA IX, VEGF, and MMP-9 in surgically resected non-small cell lung cancer. Lung Cancer 49:325–335 Kim SJ, Rabbani ZN, Vollmer RT, Schreiber EG, Oosterwijk E, Dewhirst MW, Vujaskovic Z, Kelley MJ (2004) Carbonic anhydrase IX in early-stage non-small cell lung cancer. Clin Cancer Res 10:7925–7933 Kim SJ, Shin HJ, Jung KY, Baek SK, Shin BK, Choi J, Kim BS, Shin SW, Kim YH, Kim JS, Oosterwijk E (2007) Prognostic value of carbonic anhydrase IX and Ki-67 expression in squamous cell carcinoma of the tongue. Jpn J Clin Oncol 37:812–819 Kirkpatrick JP, Rabbani ZN, Bentley RC, Hardee ME, Karol S, Meyer J, Oosterwijk E, Havrilesky L, Secord AA, Vujaskovic Z, Dewhirst MW, Jones EL (2008) Elevated CAIX expression is associated with an increased risk of distant failure in early-stage cervical cancer. Biomark Insights 3:45–55 Kivela AJ, Saarnio J, Karttunen TJ, Kivela J, Parkkila AK, Pastorekova S, Pastorek J, Waheed A, Sly WS, Parkkila TS, Rajaniemi H (2001) Differential expression of cytoplasmic carbonic

4  Molecular Mechanisms Regulating Expression

85

anhydrases, CA I and II, and membrane-associated isozymes, CA IX and XII, in normal mucosa of large intestine and in colorectal tumors. Dig Dis Sci 46:2179–2186 Klatte T, Seligson DB, Rao JY, Yu H, de Martino M, Kawaoka K, Wong SG, Belldegrun AS, Pantuck AJ (2009) Carbonic anhydrase IX in bladder cancer: a diagnostic, prognostic, and therapeutic molecular marker. Cancer 115:1448–1458 Kon-no H, Ishii G, Nagai K, Yoshida J, Nishimura M, Nara M, Fujii T, Murata Y, Miyamoto H, Ochiai A (2006) Carbonic anhydrase IX expression is associated with tumor progression and a poor prognosis of lung adenocarcinoma. Lung Cancer 54:409–418 Kopacek J, Barathova M, Dequiedt F, Sepelakova J, Kettmann R, Pastorek J, Pastorekova S (2005) MAPK pathway contributes to density- and hypoxia-induced expression of the tumorassociated carbonic anhydrase IX. Biochim Biophys Acta 1729:41–49 Korkeila E, Talvinen K, Jaakkola PM, Minn H, Syrjanen K, Sundstrom J, Pyrhonen S (2009) Expression of carbonic anhydrase IX suggests poor outcome in rectal cancer. Br J Cancer 100:874–880 Korkolopoulou P, Perdiki M, Thymara I, Boviatsis E, Agrogiannis G, Kotsiakis X, Angelidakis D, Rologis D, Diamantopoulou K, Thomas-Tsagli E, Kaklamanis L, Gatter K, Patsouris E (2007) Expression of hypoxia-related tissue factors in astrocytic gliomas. A multivariate survival study with emphasis upon carbonic anhydrase IX. Hum Pathol 38:629–638 Kowalewska M, Radziszewski J, Kulik J, Barathova M, Nasierowska-Guttmajer A, Bidzinski M, Pastorek J, Pastorekova S, Siedlecki JA (2005) Detection of carbonic anhydrase 9-expressing tumor cells in the lymph nodes of vulvar carcinoma patients by RT-PCR. Int J Cancer 116:957–962 Kozin SV, Shkarin P, Gerweck LE (2001) The cell transmembrane pH gradient in tumors enhances cytotoxicity of specific weak acid chemotherapeutics. Cancer Res 61:4740–4743 Kreis TE, Lodish HF (1986) Oligomerization is essential for transport of vesicular stomatitis viral glycoprotein to the cell surface. Cell 46:929–937 Kummola L, Hamalainen JM, Kivela J, Kivela AJ, Saarnio J, Karttunen T, Parkkila S (2005) Expression of a novel carbonic anhydrase, CA XIII, in normal and neoplastic colorectal mucosa. BMC Cancer 5:41 Kuo WH, Chiang WL, Yang SF, Yeh KT, Yeh CM, Hsieh YS, Chu SC (2003) The differential expression of cytosolic carbonic anhydrase in human hepatocellular carcinoma. Life Sci 73:2211–2223 Lam JS, Pantuck AJ, Belldegrun AS, Figlin RA (2005) G250: a carbonic anhydrase IX monoclonal antibody. Curr Oncol Rep 7:109–115 Lendahl U, Lee KL, Yang H, Poellinger L (2009) Generating specificity and diversity in the transcriptional response to hypoxia. Nat Rev Genet 10:821–832 Li Y, Wang H, Oosterwijk E, Selman Y, Mira JC, Medrano T, Shiverick KT, Frost SC (2009) Antibody-specific detection of CAIX in breast and prostate cancers. Biochem Biophys Res Commun 386:488–492 Liao SY, Aurelio ON, Jan K, Zavada J, Stanbridge EJ (1997) Identification of the MN/CA9 protein as a reliable diagnostic biomarker of clear cell carcinoma of the kidney. Cancer Res 57: 2827–2831 Liao SY, Brewer C, Zavada J, Pastorek J, Pastorekova S, Manetta A, Berman ML, DiSaia PJ, Stanbridge EJ (1994) Identification of the MN antigen as a diagnostic biomarker of cervical intraepithelial squamous and glandular neoplasia and cervical carcinomas. Am J Pathol 145: 598–609 Liao SY, Darcy KM, Randall LM, Tian C, Monk BJ, Burger RA, Fruehauf JP, Peters WA, Stock RJ, Stanbidge EJ (2010) Prognostic relevance of carbonic anhydrase-IX in high-risk, earlystage cervical cancer: A Gynecologic Oncology Group study. Gynecol Oncol 116:452–458 Loncaster JA, Harris AL, Davidson SE, Logue JP, Hunter RD, Wycoff CC, Pastorek J, Ratcliffe PJ, Stratford IJ, West CM (2001) Carbonic anhydrase (CA IX) expression, a potential new intrinsic marker of hypoxia: correlations with tumor oxygen measurements and prognosis in locally advanced carcinoma of the cervix. Cancer Res 61:6394–6399

86

J. Pastorek and S. Pastorekova

Malentacchi F, Simi L, Nannelli C, Andreani M, Janni A, Pastorekova S, Orlando C (2009) Alternative splicing variants of carbonic anhydrase IX in human non-small cell lung cancer. Lung Cancer 64:271–276 Mandriota SJ, Turner KJ, Davies DR, Murray PG, Morgan NV, Sowter HM, Wykoff CC, Maher ER, Harris AL, Ratcliffe PJ, Maxwell PH (2002) HIF activation identifies early lesions in VHL kidneys: evidence for site-specific tumor suppressor function in the nephron. Cancer Cell 1:459–468 Maresca A, Temperini C, Vu H, Pham NB, Poulsen SA, Scozzafava A, Quinn RJ, Supuran CT (2009) Non-zinc mediated inhibition of carbonic anhydrases: coumarins are a new class of suicide inhibitors (#). J Am Chem Soc 131:3057–3062 Maseide K, Kandel RA, Bell RS, Catton CN, O’Sullivan B, Wunder JS, Pintilie M, Hedley D, Hill RP (2004) Carbonic anhydrase IX as a marker for poor prognosis in soft tissue sarcoma. Clin Cancer Res 10:4464–4471 McKiernan JM, Buttyan R, Bander NH, Stifelman MD, Katz AE, Chen MW, Olsson CA, Sawczuk IS (1997) Expression of the tumor-associated gene MN: a potential biomarker for human renal cell carcinoma. Cancer Res 57:2362–2365 McMurtrie HL, Cleary HJ, Alvarez BV, Loiselle FB, Sterling D, Morgan PE, Johnson DE, Casey JR (2004) The bicarbonate transport metabolon. J Enzyme Inhib Med Chem 19:231–236 Morgan PE, Pastorekova S, Stuart-Tilley AK, Alper SL, Casey JR (2007) Interactions of transmembrane carbonic anhydrase, CAIX, with bicarbonate transporters. Am J Physiol Cell Physiol 293:C738–C748 Mori M, Staniunas RJ, Barnard GF, Jessup JM, Steele GD Jr, Chen LB (1993) The significance of carbonic anhydrase expression in human colorectal cancer. Gastroenterology 105: 820–826 Morsy SM, Badawi AM, Cecchi A, Scozzafava A, Supuran CT (2009) Carbonic anhydrase inhibitors. Biphenylsulfonamides with inhibitory action towards the transmembrane, tumor-associated isozymes IX possess cytotoxic activity against human colon, lung and breast cancer cell lines. J Enzyme Inhib Med Chem 24:499–505 Mulders P, Bleumer I, Debruyne F, Oosterwijk E (2004) Specific monoclonal antibody-based immunotherapy by targeting the RCC-associated antigen carbonic anhydrase-IX(G250/MN). Urologe A 43(Suppl 3):S146–S147 Nakao M, Ishii G, Nagai K, Kawase A, Kenmotsu H, Kon-No H, Hishida T, Nishimura M, Yoshida J, Ochiai A (2009) Prognostic significance of carbonic anhydrase IX expression by cancer-associated fibroblasts in lung adenocarcinoma. Cancer 115:2732–2743 Niemela AM, Hynninen P, Mecklin JP, Kuopio T, Kokko A, Aaltonen L, Parkkila AK, Pastorekova S, Pastorek J, Waheed A, Sly WS, Orntoft TF, Kruhoffer M, Haapasalo H, Parkkila S, Kivela AJ (2007) Carbonic anhydrase IX is highly expressed in hereditary nonpolyposis colorectal cancer. Cancer Epidemiol Biomarkers Prev 16:1760–1766 Oosterwijk E (2008a) Carbonic anhydrase IX/G250/MN: a molecule too good to be true? BJU Int 101:527–528 Oosterwijk E (2008b) Carbonic anhydrase IX: historical and future perspectives. BJU Int 101(Suppl 4):2–7 Opavsky R, Pastorekova S, Zelnik V, Gibadulinova A, Stanbridge EJ, Zavada J, Kettmann R, Pastorek J (1996) Human MN/CA9 gene, a novel member of the carbonic anhydrase family: structure and exon to protein domain relationships. Genomics 33:480–487 Pantuck AJ, Klatte T, Seligson D, Atkins M, Belldegrun A (2008) Carbonic anhydrase IX as a predictive biomarker for clear cell renal cell carcinoma. J Clin Oncol 26:3105–3107; author reply 3107–3109 Parkkila S, Innocenti A, Kallio H, Hilvo M, Scozzafava A, Supuran CT (2009) The protein tyrosine kinase inhibitors imatinib and nilotinib strongly inhibit several mammalian alphacarbonic anhydrase isoforms. Bioorg Med Chem Lett 19:4102–4106 Parkkila S, Rajaniemi H, Parkkila AK, Kivela J, Waheed A, Pastorekova S, Pastorek J, Sly WS (2000) Carbonic anhydrase inhibitor suppresses invasion of renal cancer cells in vitro. Proc Natl Acad Sci USA 97:2220–2224

4  Molecular Mechanisms Regulating Expression

87

Parkkila S, Lasota J, Fietcher JA, Ou WB, Kivelä AJ, Nuorva K, Parkkila AK, Ollikainen J, Sly WS, Waheed A, Pastorekova S, Pastorek J, Isola J, Miettinen M (2010). Carbonic anhydrase II. A novel biomarker for gastrointestinal stromal tumors. Mod Pathol 23:743–750 Pastorek J, Pastorekova S, Callebaut I, Mornon JP, Zelnik V, Opavsky R, Zat’ovicova M, Liao S, Portetelle D, Stanbridge EJ et al (1994) Cloning and characterization of MN, a human tumorassociated protein with a domain homologous to carbonic anhydrase and a putative helixloop-helix DNA binding segment. Oncogene 9:2877–2888 Pastorekova S, Casini A, Scozzafava A, Vullo D, Pastorek J, Supuran CT (2004a) Carbonic anhydrase inhibitors: the first selective, membrane-impermeant inhibitors targeting the tumorassociated isozyme IX. Bioorg Med Chem Lett 14:869–873 Pastorekova S, Parkkila S, Parkkila AK, Opavsky R, Zelnik V, Saarnio J, Pastorek J (1997) Carbonic anhydrase IX, MN/CA IX: analysis of stomach complementary DNA sequence and expression in human and rat alimentary tracts. Gastroenterology 112:398–408 Pastorekova S, Parkkila S, Pastorek J, Supuran CT (2004b) Carbonic anhydrases: current state of the art, therapeutic applications and future prospects. J Enzyme Inhib Med Chem 19: 199–229 Pastorekova S, Parkkila S, Zavada J (2006) Tumor-associated carbonic anhydrases and their clinical significance. Adv Clin Chem 42:167–216 Pastorekova S, Zatovicova M, Pastorek J (2008) Cancer-associated carbonic anhydrases and their inhibition. Curr Pharm Des 14:685–698 Pastorekova S, Zavadova Z, Kostal M, Babusikova O, Zavada J (1992) A novel quasi-viral agent, MaTu, is a two-component system. Virology 187:620–626 Peles E, Nativ M, Campbell PL, Sakurai T, Martinez R, Lev S, Clary DO, Schilling J, Barnea G, Plowman GD, Grumet M, Schlessinger J (1995) The carbonic anhydrase domain of receptor tyrosine phosphatase beta is a functional ligand for the axonal cell recognition molecule contactin. Cell 82:251–260 Potter C, Harris AL (2004) Hypoxia inducible carbonic anhydrase IX, marker of tumour hypoxia, survival pathway and therapy target. Cell Cycle 3:164–167 Pouyssegur J, Dayan F, Mazure NM (2006) Hypoxia signalling in cancer and approaches to enforce tumour regression. Nature 441:437–443 Rafajova M, Zatovicova M, Kettmann R, Pastorek J, Pastorekova S (2004) Induction by hypoxia combined with low glucose or low bicarbonate and high posttranslational stability upon reoxygenation contribute to carbonic anhydrase IX expression in cancer cells. Int J Oncol 24: 995–1004 Raghunand N, Gatenby RA, Gillies RJ (2003) Microenvironmental and cellular consequences of altered blood flow in tumours. Br J Radiol 76 Spec No 1:S11–S22 Raval RR, Lau KW, Tran MG, Sowter HM, Mandriota SJ, Li JL, Pugh CW, Maxwell PH, Harris AL, Ratcliffe PJ (2005) Contrasting properties of hypoxia-inducible factor 1 (HIF-1) and HIF-2 in von Hippel-Lindau-associated renal cell carcinoma. Mol Cell Biol 25:5675–5686 Ruas JL, Poellinger L (2005) Hypoxia-dependent activation of HIF into a transcriptional regulator. Semin Cell Dev Biol 16:514–522 Saarnio J, Parkkila S, Parkkila AK, Haukipuro K, Pastorekova S, Pastorek J, Kairaluoma MI, Karttunen TJ (1998) Immunohistochemical study of colorectal tumors for expression of a novel transmembrane carbonic anhydrase, MN/CA IX, with potential value as a marker of cell proliferation. Am J Pathol 153:279–285 Saczewski F, Slawinski J, Kornicka A, Brzozowski Z, Pomarnacka E, Innocenti A, Scozzafava A, Supuran CT (2006) Carbonic anhydrase inhibitors. Inhibition of the cytosolic human isozymes I and II, and the transmembrane, tumor-associated isozymes IX and XII with substituted aromatic sulfonamides activatable in hypoxic tumors. Bioorg Med Chem Lett 16: 4846–4851 Sansone P, Storci G, Giovannini C, Pandolfi S, Pianetti S, Taffurelli M, Santini D, Ceccarelli C, Chieco P, Bonafe M (2007) p66Shc/Notch-3 interplay controls self-renewal and hypoxia survival in human stem/progenitor cells of the mammary gland expanded in  vitro as mammospheres. Stem Cells 25:807–815

88

J. Pastorek and S. Pastorekova

Scott FL, Stec B, Pop C, Dobaczewska MK, Lee JJ, Monosov E, Robinson H, Salvesen GS, Schwarzenbacher R, Riedl SJ (2009) The Fas-FADD death domain complex structure unravels signalling by receptor clustering. Nature 457:1019–1022 Shin KH, Diaz-Gonzalez JA, Russell J, Chen Q, Burgman P, Li XF, Ling CC (2007) Detecting changes in tumor hypoxia with carbonic anhydrase IX and pimonidazole. Cancer Biol Ther 6:70–75 Schrijvers ML, van der Laan BF, de Bock GH, Pattje WJ, Mastik MF, Menkema L, Langendijk JA, Kluin PM, Schuuring E, van der Wal JE (2008) Overexpression of intrinsic hypoxia markers HIF1alpha and CA-IX predict for local recurrence in stage T1-T2 glottic laryngeal carcinoma treated with radiotherapy. Int J Radiat Oncol Biol Phys 72:161–169 Signoretti S, Regan M, Atkins M (2008) Carbonic anhydrase IX as a predictive biomarker of response to kidney cancer therapy. BJU Int 101(Suppl 4):31–35 Smaine FZ, Winum JY, Montero JL, Regainia Z, Vullo D, Scozzafava A, Supuran CT (2007) Carbonic anhydrase inhibitors: Selective inhibition of the extracellular, tumor-associated isoforms IX and XII over isozymes I and II with glycosyl-thioureido-sulfonamides. Bioorg Med Chem Lett 17:5096–5100 Smyth LG, O’Hurley G, O’Grady A, Fitzpatrick JM, Kay E, Watson RW (2010) Carbonic anhydrase IX expression in prostate cancer. Prostate Cancer Prostatic Dis 13:178–181 Stiti M, Cecchi A, Rami M, Abdaoui M, Barragan-Montero V, Scozzafava A, Guari Y, Winum JY, Supuran CT (2008) Carbonic anhydrase inhibitor coated gold nanoparticles selectively inhibit the tumor-associated isoform IX over the cytosolic isozymes I and II. J Am Chem Soc 130: 16130–16131 Stock C, Schwab A (2009) Protons make tumor cells move like clockwork. Pflugers Arch 458:981–992 Stubbs M, McSheehy PM, Griffiths JR, Bashford CL (2000) Causes and consequences of tumour acidity and implications for treatment. Mol Med Today 6:15–19 Supuran CT (2007) Carbonic anhydrases as drug targets – an overview. Curr Top Med Chem 7:825–833 Supuran CT (2008) Carbonic anhydrases: novel therapeutic applications for inhibitors and activators. Nat Rev Drug Discov 7:168–181 Svastova E, Hulikova A, Rafajova M, Zat’ovicova M, Gibadulinova A, Casini A, Cecchi A, Scozzafava A, Supuran CT, Pastorek J, Pastorekova S (2004) Hypoxia activates the capacity of tumor-associated carbonic anhydrase IX to acidify extracellular pH. FEBS Lett 577: 439–445 Svastova E, Zilka N, Zat’ovicova M, Gibadulinova A, Ciampor F, Pastorek J, Pastorekova S (2003) Carbonic anhydrase IX reduces E-cadherin-mediated adhesion of MDCK cells via interaction with beta-catenin. Exp Cell Res 290:332–345 Swietach P, Patiar S, Supuran CT, Harris AL, Vaughan-Jones RD (2009) The role of carbonic anhydrase 9 in regulating extracellular and intracellular ph in three-dimensional tumor cell growths. J Biol Chem 284:20299–20310 Swietach P, Wigfield S, Supuran CT, Harris AL, Vaughan-Jones RD (2008) Cancer-associated, hypoxia-inducible carbonic anhydrase IX facilitates CO2 diffusion. BJU Int 101 (Suppl 4): 22–4 Swinson DE, Jones JL, Richardson D, Wykoff C, Turley H, Pastorek J, Taub N, Harris AL, O’Byrne KJ (2003) Carbonic anhydrase IX expression, a novel surrogate marker of tumor hypoxia, is associated with a poor prognosis in non-small-cell lung cancer. J Clin Oncol 21: 473–482 Takacova M, Holotnakova T, Vondracek J, Machala M, Pencikova K, Gradin K, Poellinger L, Pastorek J, Pastorekova S, Kopacek J (2009) Role of aryl hydrocarbon receptor in modulation of the expression of the hypoxia marker carbonic anhydrase IX. Biochem J 419: 419–425 Takacova M, Holotnakova T, Barathova M, Pastorekova S, Kopacek J, Pastorek J (2010). Src induces expression of carbonic anhydrase IX via hypoxia-inducible factor 1. Oncol Rep 23:869–874

4  Molecular Mechanisms Regulating Expression

89

Tan EY, Yan M, Campo L, Han C, Takano E, Turley H, Candiloro I, Pezzella F, Gatter KC, Millar EK, O’Toole SA, McNeil CM, Crea P, Segara D, Sutherland RL, Harris AL, Fox SB (2009) The key hypoxia regulated gene CAIX is upregulated in basal-like breast tumours and is associated with resistance to chemotherapy. Br J Cancer 100:405–411 Teicher BA (2009) Acute and chronic in  vivo therapeutic resistance. Biochem Pharmacol 77: 1665–1673 Teicher BA, Liu SD, Liu JT, Holden SA, Herman TS (1993) A carbonic anhydrase inhibitor as a potential modulator of cancer therapies. Anticancer Res 13:1549–1556 Temperini C, Innocenti A, Mastrolorenzo A, Scozzafava A, Supuran CT (2007) Carbonic anhydrase inhibitors. Interaction of the antiepileptic drug sulthiame with twelve mammalian isoforms: kinetic and X-ray crystallographic studies. Bioorg Med Chem Lett 17:4866–4872 Trastour C, Benizri E, Ettore F, Ramaioli A, Chamorey E, Pouyssegur J, Berra E (2007) HIF1alpha and CA IX staining in invasive breast carcinomas: prognosis and treatment outcome. Int J Cancer 120:1451–1458 Turner JR, Odze RD, Crum CP, Resnick MB (1997) MN antigen expression in normal, preneoplastic, and neoplastic esophagus: a clinicopathological study of a new cancer-associated biomarker. Hum Pathol 28:740–744 van den Beucken T, Koritzinsky M, Niessen H, Dubois L, Savelkouls K, Mujcic H, Jutten B, Kopacek J, Pastorekova S, van der Kogel AJ, Lambin P, Voncken W, Rouschop KM, Wouters BG (2009a) Hypoxia-induced expression of carbonic anhydrase 9 is dependent on the unfolded protein response. J Biol Chem 284:24204–24212 van den Beucken T, Ramaekers CH, Rouschop K, Koritzinsky M, Wouters BG (2009b) Deficient carbonic anhydrase 9 expression in UPR-impaired cells is associated with reduced survival in an acidic microenvironment. Radiother Oncol 92:437–442 Vermylen P, Roufosse C, Burny A, Verhest A, Bosschaerts T, Pastorekova S, Ninane V, Sculier JP (1999) Carbonic anhydrase IX antigen differentiates between preneoplastic malignant lesions in non-small cell lung carcinoma. Eur Respir J 14:806–811 Vleugel MM, Greijer AE, Shvarts A, van der Groep P, van Berkel M, Aarbodem Y, van Tinteren H, Harris AL, van Diest PJ, van der Wall E (2005) Differential prognostic impact of hypoxia induced and diffuse HIF-1alpha expression in invasive breast cancer. J Clin Pathol 58: 172–177 Vukovic V, Tannock IF (1997) Influence of low pH on cytotoxicity of paclitaxel, mitoxantrone and topotecan. Br J Cancer 75:1167–1172 Vullo D, Franchi M, Gallori E, Pastorek J, Scozzafava A, Pastorekova S, Supuran CT (2003) Carbonic anhydrase inhibitors: inhibition of the tumor-associated isozyme IX with aromatic and heterocyclic sulfonamides. Bioorg Med Chem Lett 13:1005–1009 Vullo D, Scozzafava A, Pastorekova S, Pastorek J, Supuran CT (2004) Carbonic anhydrase inhibitors: inhibition of the tumor-associated isozyme IX with fluorine-containing sulfonamides. The first subnanomolar CA IX inhibitor discovered. Bioorg Med Chem Lett 14: 2351–2356 Weber A, Casini A, Heine A, Kuhn D, Supuran CT, Scozzafava A, Klebe G (2004) Unexpected nanomolar inhibition of carbonic anhydrase by COX-2-selective celecoxib: new pharmacological opportunities due to related binding site recognition. J Med Chem 47:550–557 Wilkinson BL, Bornaghi LF, Houston TA, Innocenti A, Supuran CT, Poulsen SA (2006) A novel class of carbonic anhydrase inhibitors: glycoconjugate benzene sulfonamides prepared by “click-tailing”. J Med Chem 49:6539–6548 Willam C, Warnecke C, Schefold JC, Kugler J, Koehne P, Frei U, Wiesener M, Eckardt KU (2006) Inconsistent effects of acidosis on HIF-alpha protein and its target genes. Pflugers Arch 451: 534–543 Winum JY, Cecchi A, Montero JL, Innocenti A, Scozzafava A, Supuran CT (2005a) Carbonic anhydrase inhibitors. Synthesis and inhibition of cytosolic/tumor-associated carbonic anhydrase isozymes I, II, and IX with boron-containing sulfonamides, sulfamides, and sulfamates: toward agents for boron neutron capture therapy of hypoxic tumors. Bioorg Med Chem Lett 15:3302–3306

90

J. Pastorek and S. Pastorekova

Winum JY, Dogne JM, Casini A, de Leval X, Montero JL, Scozzafava A, Vullo D, Innocenti A, Supuran CT (2005b) Carbonic anhydrase inhibitors: synthesis and inhibition of cytosolic/ membrane-associated carbonic anhydrase isozymes I, II, and IX with sulfonamides incorporating hydrazino moieties. J Med Chem 48:2121–2125 Winum JY, Pastorekova S, Jakubickova L, Montero JL, Scozzafava A, Pastorek J, Vullo D, Innocenti A, Supuran CT (2005c) Carbonic anhydrase inhibitors: synthesis and inhibition of cytosolic/tumor-associated carbonic anhydrase isozymes I, II, and IX with bis-sulfamates. Bioorg Med Chem Lett 15:579–584 Winum JY, Vullo D, Casini A, Montero JL, Scozzafava A, Supuran CT (2003a) Carbonic anhydrase inhibitors. Inhibition of cytosolic isozymes I and II and transmembrane, tumor-associated isozyme IX with sulfamates including EMATE also acting as steroid sulfatase inhibitors. J Med Chem 46:2197–2204 Winum JY, Vullo D, Casini A, Montero JL, Scozzafava A, Supuran CT (2003b) Carbonic anhydrase inhibitors: inhibition of transmembrane, tumor-associated isozyme IX, and cytosolic isozymes I and II with aliphatic sulfamates. J Med Chem 46:5471–5477 Wykoff CC, Beasley NJ, Watson PH, Turner KJ, Pastorek J, Sibtain A, Wilson GD, Turley H, Talks KL, Maxwell PH, Pugh CW, Ratcliffe PJ, Harris AL (2000) Hypoxia-inducible expression of tumor-associated carbonic anhydrases. Cancer Res 60:7075–7083 Yin Q, Lin SC, Lamothe B, Lu M, Lo YC, Hura G, Zheng L, Rich RL, Campos AD, Myszka DG, Lenardo MJ, Darnay BG, Wu H (2009) E2 interaction and dimerization in the crystal structure of TRAF6. Nat Struct Mol Biol 16:658–666 Yoshiura K, Nakaoka T, Nishishita T, Sato K, Yamamoto A, Shimada S, Saida T, Kawakami Y, Takahashi TA, Fukuda H, Imajoh-Ohmi S, Oyaizu N, Yamashita N (2005) Carbonic anhydrase II is a tumor vessel endothelium-associated antigen targeted by dendritic cell therapy. Clin Cancer Res 11:8201–8207 Zat’ovicova M, Tarabkova K, Svastova E, Gibadulinova A, Mucha V, Jakubickova L, Biesova Z, Rafajova M, Ortova Gut M, Parkkila S, Parkkila AK, Waheed A, Sly WS, Horak I, Pastorek J, Pastorekova S (2003) Monoclonal antibodies generated in carbonic anhydrase IX-deficient mice recognize different domains of tumour-associated hypoxia-induced carbonic anhydrase IX. J Immunol Methods 282:117–134 Zatovicova M, Sedlakova O, Svastova E, Ohradanova A, Ciampor F, Arribas J, Pastorek J, Pastorekova S (2005) Ectodomain shedding of the hypoxia-induced carbonic anhydrase IX is a metalloprotease-dependent process regulated by TACE/ADAM17. Br J Cancer 93:1267–1276 Zatovicova M, Jeleuska L, Hulikova A, Csaderova L, Ditte P, Goliasova T, Pastorek J, Pastorekova S (in press). Carbonic anhydrase IX as an anticancer target: preclinical evaluation of internalizing monoclonal antibody directed to catalytic domain. Curr Pharm Design Zavada J, Zavadova Z, Pastorek J, Biesova Z, Jezek J, Velek J (2000) Human tumour-associated cell adhesion protein MN/CA IX: identification of M75 epitope and of the region mediating cell adhesion. Br J Cancer 82:1808–1813 Zavada J, Zavadova Z, Pastorekova S, Ciampor F, Pastorek J, Zelnik V (1993) Expression of MaTu-MN protein in human tumor cultures and in clinical specimens. Int J Cancer 54: 268–274 Zavada J, Zavadova Z, Zat’ovicova M, Hyrsl L, Kawaciuk I (2003) Soluble form of carbonic anhydrase IX (CA IX) in the serum and urine of renal carcinoma patients. Br J Cancer 89: 1067–1071 Zhang J (2008) Recent advances in preoperative imaging of renal tumors. Curr Opin Urol 18: 111–115 Zheng J, Avvaru BS, Tu C, McKenna R, Silverman DN (2008) Role of hydrophilic residues in proton transfer during catalysis by human carbonic anhydrase II. Biochemistry 47: 12028–12036 Zhou GX, Ireland J, Rayman P, Finke J, Zhou M (2010) Quantification of carbonic anhydrase IX expression in serum and tissue of renal cell carcinoma patients using enzyme-linked immunosorbent assay: prognostic and diagnostic potentials. Urology 75:257–261

Chapter 5

Glycolytic Pathway as a Target for Tumor Inhibition Weiqin Lu and Peng Huang

Abstract  It has been known for decades that cancer cells exhibit elevated aerobic glycolysis, a phenomenon first observed by Otto Warburg. The imaging technique 18 fluoro-deoxyglucose-positron emission tomography (FDG-PET) widely used in clinical diagnosis of cancer is based on the increased glucose uptake by cancer cells, likely due to a significant increase in glucose flow into the glycolytic pathway to generate ATP and conversion to other metabolic intermediates and building blocks for cell growth and proliferation. Furthermore, the increased glucose uptake in cancer tissues seems correlated with tumor aggressiveness and poor prognosis. Such profound metabolic alterations suggest that cancer cells may prefer to use glycolysis for their proliferative and survival advantage. Recent studies have begun to elucidate the molecular mechanisms underlying these metabolic alterations, and provide important new insights into the mechanistic links between oncogenic signals and metabolic regulation. Importantly, targeted inhibition of glycolysis and its regulatory pathways may provide exciting opportunities for the development of therapeutic strategies to preferentially kill cancer cells. This chapter will summarize our current understanding of glycolytic alterations in cancer cells and the relevant regulatory mechanisms, and discuss possible therapeutic strategies that exploit the metabolic abnormalities in cancer to preferentially kill malignant cells.

Introduction Glucose is a major source of fuel for ATP generation. It can be converted into two molecules of pyruvate in the cytosol with a net gain of 2 ATP and a reduction of 2 NAD+ to 2 NADH per glucose through a sequence of reactions known as glycolysis.

P. Huang (*) Department of Molecular Pathology, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030, USA e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_5, © Springer Science+Business Media, LLC 2010

91

92

W. Lu and P. Huang Glucose GLUT

Glucose

ROS

HK

PPP (oxidative)

G6P PFK2

NADPH anabolic process

G6PD

GPI

Ribose-5-P

F6P

F2,6-BP

TKL/TAL

PFK1

F1,6-BP Aldolase

Dihydroxyacetone-P

Nucleotide synthesis

PPP (non-oxidative) GA3P

Mitochondria

GAPDH

1,3-BPG 3-PG 2-PG

PDH Enolase

PEP

ATP Succinate

OAA

TCA cycle SDH

Pyr PK

Malate Pyr

Lactate

NADH

Ac-CoA

PGM

ATP

MRC

NAD+ Citrate

PGK

Fumarate Fumerase

LDH

MCT

Lactate

Fig. 5.1  Overview of glycolysis, PPP, and the TCA cycle. Glucose is metabolized through the glycolytic pathway and the TCA cycle in the mitochondria in conjunction with mitochondrial respiratory to generate ATP. Glucose can also be metabolized through the PPP (in dashed line) to generate ribose-5-phosphate (ribose-5-P) as a precursor for the synthesis of nucleic acids, and NADPH for anabolic synthesis of fat acids or as a reducing equivalent to combat ROS. Notes: Dihydroxyacetone-P, Dihydroxyacetone-phosphate; F6P, fructose-6-phosphate; F1,6-BP, fructose1,6-bisphosphate; G6PD, glucose-6-phosphate dehydrogenase; GA3P, glyceraldehyde-3-phosphate; G6P, glucose-6-phoaphate; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GLUT, glucose transporter; GPI, glucose-phosphate isomerase; LDH, lactate dehydrogenase; HK, hexokinases; MCT, monocarboxylate transporter; MRC, mitochondrial respiratory chain (complexes I-V); PFK1, phosphofructokinase 1; PFK2, phosphofructokinase 2; PGK, phosphoglycerate kinase; PGM, phosphoglycerate mutase; PK, pyruvate kinase; Pyr, pyruvate; Ribose-5-P, ribose-5-phosphate; Ac-CoA, acetyl-CoA; OAA, oxaloacetate; PDH, pyruvate dehydrogenase; PPP, pentose phosphate pathway; ROS, reactive oxygen species; SDH, succinate dehydrogenase; TAL, transaldolases; TKL, transketolase

As illustrated in Fig. 5.1, glucose can be actively transported from blood stream into cells by Na+-dependent glucose specific transporters (GLUTs). Once it enters the cell, glucose is phosphorylated into glucose-6-phosphate (G6P) by a family of enzymes called hexokinases (HKs), the first rate-limiting enzyme of the glycolysis. G6P is a key metabolic intermediate at the branching point of two major metabolic pathways, namely glycolysis and pentose phosphate pathway (PPP). G6P can be

5  Glycolytic Pathway as a Target for Tumor Inhibition

93

dehydrogenated by glucose 6-phosphate dehydrogenase (G6PD) in the PPP to generate ribose 5-phosphate and NADPH when the cellular ratio of NADP+/ NADPH is increased. If cellular ATP level decreases, G6P will be actively converted to fructose-6-phosphate (F6P) by glucose phosphate isomerase (GPI) and thus favoring the glycolytic pathway. With ATP energy expenditure, F6P is then converted to fructose-1,6-bisphosphate (F1,6-BP) catalyzed by another rate-limiting enzyme phosphofructokinase 1 (PFK1). This reaction is energetically favored. The glycolytic process appears irreversible beyond this point, thus this is the key regulatory step in glycolysis. Aldolase further splits F1,6-BP into two triose sugars, dihydroxyacetone phosphate and glyceraldehyde-3-phosphate (GA3P). The former can be converted to GA3P by triosephosphate isomerase and proceeds further into glycolysis. GA3P is then dehydrogenated to 1, 3-bisphosphoglycerate (1,3-BPG) by an oxidoreductase glyceraldehyde dehydrogenase (GAPDH). This step requires NAD+ as a hydrogen carrier accepting H+ to generate NADH. Thus, the presence of sufficient amount of NAD+ is important to support a sustained level of glycolysis. 1,3BPG is then further metabolized by phosphoglycerate kinase (PGK) to 3-phosphoglycerate (3-PG), which is converted to 2-phosphoglycerate (2-PG) by phosphoglycerate mutase (PGM) and further converted to phosphoenolpyruvate (PEP) by enolase. Finally, pyruvate is produced from PEP by pyruvate kinase, one of the three key rate-limiting enzymes in glycolysis. If pyruvate is not further utilized in the mitochondria, it will be converted into lactate by lactate dehydrogenase (LDH) in the cytosol, generating NAD+ as substrate for the GAPDH-catalyzed reaction to allow continued glycolysis. NAD+/NADH ratio is important in regulating LDH activity. Because cells can obtain ATP supply through the oxygen-independent glycolytic pathway in cytosol and the oxygen-dependent oxidative phosphorylation (OXPHOS) pathway in mitochondria, the availability of oxygen and the mitochondrial functional status of the cells play important roles in determining the proportion of each pathway in ATP generation. In the presence of oxygen and functional mitochondria, pyruvate can be transported into the mitochondria and decarboxylated into acetyl CoA by the pyruvate dehydrogenase (PDH) complex in the mitochondria. The oxidation of acetyl CoA is achieved by the tricarboxylic acid (TCA) cycle generating CO2, H2O, GTP and other bioenergetic products including NADH and FADH2 (Krebs 1970a,b). The electrons from NADH and FADH2 are transported through the respiratory chain complexes (I–IV) and donated to oxygen to generate water, and the proton gradient generated through this process is used as the energy source for ATP production in complex V, where the ATP generation is highly efficient (Mitchell and Moyle 1967; Skulachev 1994). Under normal physiological conditions, mitochondria play an important role in OXPHOS generating most of the ATP for cellular function (Newmeyer and Ferguson-Miller 2003). However, in cancer cells the glucose metabolism seems significantly altered, and such alterations may serve as a ­biochemical basis to selectively kill malignant cells. A major goal of chemotherapy is to selectively kill cancer cells without significant toxicity to normal cells. Tremendous progress has been made recently in our

94

W. Lu and P. Huang

understanding of the biochemical and metabolic characteristics of cancer cells and the molecular mechanisms that govern cancer development and progression. It has been recognized that cancer cells exhibit a high degree of heterogeneity, adaptability, and genetic instability, making it difficult to eliminate the entire cancer cell population using target-specific agents. However, despite the different genetic or epigenetic alterations in cancer cells, aerobic glycolysis seems to be a common feature of the malignant cells as originally observed by Otto Warburg (1956a). Compared to normal adult cells, which mainly use OXPHOS to generate ATP, cancer cells have a higher tendency to generate ATP by converting glucose to lactate through the glycolytic pathway even in the presence of adequate oxygen (an aerobic glycolytic process known as the Warburg effect). It was estimated that more than 90% of metastatic tumors tested are highly glycolytic (Czernin and Phelps 2002). Such widespread metabolic alterations in cancer suggests that the malignant cells may prefer to use aerobic glycolysis rather than the highly efficient OXPHOS pathway to generate energy for their survival advantages, and perhaps facilitating tumor invasion and metastasis through acidification of the tumor microenvironment that in turn select the cancer cells with more aggressive malignant behaviors (Gatenby and Gawlinski 1996; Gatenby et al. 2006; Schornack and Gillies 2003; Wallace 2005). The increase in glucose uptake by cancer cells has been used as a metabolic basis in FDG-PET scan, which uses the glucose analogue 18 F-deoxyglucose as a tracer, for diagnosis of cancer and monitoring of tumor progression or therapeutic response (Czernin and Phelps 2002; Seemann 2004). The sustained activation of glycolysis observed in cancer cells may play a crucial role in cancer cell transformation and promotion of malignant cell behaviors (Gatenby 1995). With further understanding of the tumor micro-environment related to hypoxia and hypoxia-inducible factor (HIF) function, and the roles of Myc, Akt, Ras, p53, and AMPK in the glycolytic regulation, the metabolic diffe­ rences between cancer and normal cells have gained broad attention recently, and study in this area is likely to have profound influence on our understanding on cancer biology and cancer therapeutics.

Alterations of Glucose Metabolism in Cancer More than 80 years ago, Warburg pioneered the research on mitochondrial respiratory chain defects in the context of cancer, and was the first to recognize that tumors are highly glycolytic. His decades of work in this area was perhaps best illustrated in his landmark publication in Science more than 50 years ago (Warburg 1956b). It is estimated that over 50% of ATP in many tumor cells is derived from glycolysis to meet the needs for rapid cell growth and proliferation, although some of the highly proliferating tumor cells do not have obvious defects in oxidative metabolism. FDG-PET scan of tumor tissues has unequivocally demonstrated the increase in glucose uptake as a hallmark of human cancer, showing a profound alteration of glucose metabolism in tumors (Haberkorn et al. 1991; Weber 2006).

5  Glycolytic Pathway as a Target for Tumor Inhibition

95

It should be pointed out, however, that an increase in glucose uptake dose not necessarily indicate that this increase is solely due to increased glycolysis. The increase in glycolysis in cancer cells is likely to provide proliferative and survival advantages. In addition to ATP generation, the glycolysis also produces important metabolic intermediates for the biosynthesis of macromolecules and avoids ROS generation. Some of the glycolytic enzymes also have anti-apoptotic function. The metabolic switch or “reprogramming” from aerobic mitochondrial metabolism to glycolysis in tumor cells might be achieved through a number of putative mechanisms, including up-regulation of the glycolytic enzymes, suppression of the mitochondrial function due to mitochondrial mutation and/or defects in respiratory chain function at various points of the electron transport complexes, decreased pyruvate utilization by mitochondria, impaired TCA cycle, activation of oncogenes, mutation or loss of tumor suppressors, and adaptation to hypoxic tumor environment. Mitochondrial dysfunction may promote cancer cell survival and drug resistance through alterations in cell survival pathways such as activation of the Akt pathway and attenuation of the mitochondria-mediated apoptotic process (Carew and Huang 2002; Ohta 2006).

Overexpression of Glycolytic Enzymes in Cancer Favors Aerobic Glycolysis A large body of evidence suggests that tumor cells have enhanced enzyme activities or/and increased transcription of genes of glycolytic enzymes such as HKII, PFKs, LDH-A, and GLUTs when compared with non-tumorigenic cells (Brown et  al. 2002; Mathupala et  al. 1997; Minchenko et  al. 2002; Smith 2000). Interestingly, some of the elevated glycolytic enzymes were found to have anti-apoptotic function and may confer resistance to anticancer agents. GLUT1 The entry of glucose into the cells relies on a family of membrane transporters that facilitate the cellular uptake of glucose. GLUTs are an important family of membraneassociated proteins with isoform-specific tissue distribution, some of which are over-expressed in cancer cells. Among the 12 isoforms of GLUTs (GLUT1-GLUT12), GLUT1, 2, and 5 are found to be expressed in tumor cells (Godoy et  al. 2006). GLUT1, which transfers glucose from bloodstream into the cells, seems to play an important role in tumor development. It has a high affinity for glucose and transports glucose at a high rate. Compared to normal tissues, tumor tissues having higher 18fluoro-deoxyglucose (FDG) uptake by PET scan are associated with a higher GLUT1 expression. HIF-1 was found to bind to the promoter of GLUT1 in human HepG2 cells to stimulate glucose uptake, and deregulation of GLUT1 by c-Myc was also reported (Dombrauckas et  al. 2005). Activated AKT can also

96

W. Lu and P. Huang

enhance GLUT function. Thus, it is likely that GLUT1 up-regulation and increased glucose uptake in cancer are associated with oncogene regulation. HKII HKs, the first rate-limiting enzymes in glycolysis, catalyze the conversion of ­glucose to G6P. There are four isoforms of HKs with different tissue expression patterns and regulation. Besides catalyzing the phosphorylation of glucose, HKII has been suggested to promote cell survival by preventing apoptosis through association with the mitochondrial permeability transition pore component. It has been shown that HKII may interact with the voltage-dependent anion channel (VDAC) to induce permeability transition pore closure, and prevents BAX/BAK oligomerization on the mitochondrial membrane, blocking cytochrome c release from the mitochondria (GolshaniHebroni and Bessman 1997; Mathupala et  al. 2006). As such, the mitochondriaassociated HK (HKII) is regarded as a guardian of the mitochondria to prevent apoptosis (Robey and Hay 2005). Enhanced transcription of HKII occurs in many cancers, especially in breast cancer, suggesting a possibility of HKII promoter activation during tumorigenesis (Kim and Dang 2006). The promoter region of HKII has putative binding sites for p53, HIF, and c-Myc. Interestingly, p53 was found to downregulate HKII while c-Myc and HIF could upregulate HKII mRNA transcription (Mathupala et al. 1997; Pastorino et al. 2005). Therefore, mutation or loss of p53, activation of c-Myc, and/or stabilization of HIF may lead to upregulation of HKII expression and result in hyperpolarization of the mitochondria and suppression of apoptosis. AKT activation can also signal HKII to bind to the mitochondrial membrane and remove the G6P inhibitory regulation (Majewski et al. 2004). Thus, it is conceivable that increased HKII due to oncogene activation or tumor suppressor dysfunction can stimulate glycolysis and promote cancer cell survival and drug resistance. Previous study in mitochondrial defective cells (r0 cells) showed that inhibition of glycolysis using the HKII inhibitor 3-bromopyruvate was effective in killing cancer cells with high glycolytic activity or under hypoxic conditions (Xu et al. 2005b). PFK1 PFK1, a rate-limiting enzyme under a tight regulation during glycolysis, catalyzes the phosphorylation of F6P to F1,6BP. Interestingly, fructose-2,6-biphosphate (F2,6,BP) is an extremely potent allosteric activator of PFK1 and thus exerts a major regulatory control over the rate of glycolysis and glucose utilization. The enzyme 6-phosphofructo-2-kinase/fructose-2,6-biphosphase (PFKBF) catalyzing the reversible reaction to produce F2,6,BP plays an essential role in the regulation of glycolysis. F1,6-BP level and PFK1 activity are significantly increased in most tumors (Sanchez-Martinez and Aragon 1997). Through the transcriptional regulation of TIGAR (TP53-induced glycolysis and apoptosis regulator), p53 inhibits F2,6,BP production and therefore lowers PFK1 activity and downregulates

5  Glycolytic Pathway as a Target for Tumor Inhibition

97

g­ lycolysis (Bensaad et  al. 2006). AKT was shown to phosphorylate and activate PFK1 to upregulate glycolysis (Rathmell et al. 2003). Several other oncoproteins such as Myc and Ras have been reported to activate PFK1 in immortalized cells (Kole et al. 1991; Osthus et al. 2000). PKM2 Pyruvate kinase, another key rate-limiting enzyme in glycolysis, has four isoforms (L, R, M1, and M2). Proliferating cancer cells of various histological types seem to mainly express the human embryonic M2 isoform PKM2 (Altenberg and Greulich 2004). When PKM2 forms a highly active tetramer, this enzyme favors the formation of pyruvate and ATP. However, when PKM2 is in a less active dimer form, it impairs pyruvate formation and causes an accumulation of the upstream intermediate F1,6BP. PKM2 can selectively bind to tyrosine-phosphorylated peptides, resulting in the release of its allosteric activator F1,6BP and thus induces a diversion of glucose metabolites from energy production to the non-oxidative PPP pathway leading to anabolic processes. A recent study showed that a knockdown of PKM2 in tumor cells by shRNAs and a replacement with isoform PKM1 led to reduced lactate production, increased oxygen consumption, and reduced tumor formation in a mouse xenograft model (Christofk et  al. 2008). In proliferating cancer cells, expression of PKM2 enhances the cells’ ability to use glucose for anabolic processes and this metabolic phenotype seems to provide a selective growth advantage for the tumor cells (Vander Heiden et al. 2009). LDH-A LDH catalyzes the conversion of pyruvate to lactate and provides NAD+ to support glycolysis. LDH-A is an enzyme that favors the conversion of pyruvate to lactate in anaerobic condition, while LDH-B seems to favor the conversion of lactate to pyruvate in aerobic condition. Since NAD+ is required for the glycolytic reaction catalyzed by GAPDH, regeneration of NAD+ by LDH is important to maintain glycolysis, especially when mitochondrial respiration is compromised. It is reported that cancer cells produce more lactate and have increased LDH-A activity than normal cells (Hove et al. 1993; Schwickert et al. 1995). LDH-A plays an important role in cancer cell survival, invasion, and metastasis not only by regenerating NAD+ to maintain glycolysis but also by causing cellular acidosis due to the generation of the metabolic product lactate, which may exit cancer cells through a lactate carrier monocarboxylate transporter 4 (MCT4), and induce tissue acidosis. LDH-A gene transcription can be induced by hypoxia through the activation of HIF-1. Interestingly, MCT4 also appears to be a downstream target of HIF-1. Lactate was reported to stimulate HIF-1 expression regardless of hypoxia and lactate-induced acidosis often precedes angiogenesis (Lu et al. 2002; Shi et al. 2001). Thus, through transcriptional activation of LDH-A and MCT4, HIF1 stimulates glycolysis and lactate exit out of the cells. Lactate generation

98

W. Lu and P. Huang

due to an increased LDH-A activation in turn stimulates HIF1 expression and creates a cycle of sustained enhancement of glycolysis. LDH-A is also a transcriptional target of c-Myc and its overexpression is required for c-Myc-mediated transformation, since lowering its expression by LDH-A-antisense reduces the transformation capacity of c-Myc (Shim et  al. 1997). A recent study showed that a knockdown of LDH-A in tumor cells by shRNAs led to a stimulation of mitochondrial OXPHOS, a decrease of cell proliferation under hypoxia, and a suppression of tumorigenicity (Fantin et  al. 2006). Acidification of tumor microenvironment by lactate may facilitate a selection of more malignant cancer cells with highly aggressive behaviors, and thus promote tumor development, invasion, and metastasis (Gatenby and Gillies 2004).

The PPP The biological roles of glucose are not only to function as a major fuel source for ATP generation but also to serve as a metabolic precursor to provide carbon source for the biosynthesis of macromolecules in proliferating cells. For the fast-growing tumor cells, production of sufficient amount of such metabolic intermediates is particularly important for the timely replication of their genome, synthesis of amino acids, membrane lipids, and the package of daughter cells. As such, only a portion of the glucose utilized by the tumor cells is used to convert to lactate during aerobic glycolysis, while a significant portion of glucose is metabolized through the PPP (Kuo et  al. 2000; Poulsen and Frederiksen 1981). PPP is a major branch of metabolic flow derived from glycolysis, with G6PD as the rate-limiting enzyme locating at the branching point of PPP (Fig. 5.1). This pathway is essential for generating metabolic intermediates pentose-5-phosphate for nucleotide synthesis and the reducing equivalent NADPH for the reductive biosynthesis of lipids and fats. Significantly, NADPH is also an important reducing equivalent that is essential for the maintenance of redox homeostasis (Tian et al. 1999). The sugar moiety of the nucleotide precursor ribose-5-phosphate can be synthesized either from the oxidative or non-oxidative part of PPP (for review, see Tong et al. 2009). Both the oxidative portion (including steps catalyzed by G6PD and 6-phosphogluconate dehydrogenase enzymes) and the non-oxidative steps (including those catalyzed by transketolase-like-1 and transaldolases) of PPP have been reported to be activated or over-expressed in cancer cells (Langbein et al. 2006; Ramos-Montoya et al. 2006). Transketolase and transaldolase seem to play an important role in the inter-conversion of the metabolic intermediates between PPP and glycolysis. When the requirement for NADPH exceeds the need for pentose-phosphate in biosynthesis, the excess amount of pentose-phosphate can be converted to the glycolytic intermediates F6P and GA3P by the metabolic steps catalyzed by transaldolase and transketolase. On the other hand, when pentose-phosphate is in high demand due to fast cell growth and proliferation, the generation of a large amount of pentosephosphate can be met by conversion from F6P and GA3P catalyzed by transaldolase and transketolase. Targeted inhibition of PPP is reported to be effective in preventing tumor metastasis (Langbein et al. 2008). The exact roles of oxidative PPP and nonoxidative PPP in different stages of tumorigenesis remain to be elucidated.

5  Glycolytic Pathway as a Target for Tumor Inhibition

99

Mitochondrial Dysfunction and Increased Glycolysis in Cancer Mitochondria consume the majority of the oxygen utilized in cellular metabolism, and are responsible for aerobic ATP generation from glucose through OXPHOS (Newmeyer and Ferguson-Miller 2003; Wallace 1999). Embedded in the inner mitochondrial membrane, the mitochondrial respiratory chain is composed of five multi-enzyme complexes, including NADH dehydrogenase (complex I), succinate dehydrogenase (complex II), cytochrome bc1 complex (complex III), cytochrome c oxidase (complex IV), and the ATP synthase (complex V) (for review see Mandavilli et  al. 2002). One unique feature of mitochondria is that these cellular organelles have their own genetic materials, mitochondrial DNA (mtDNA). It should be noted that mtDNA lacks histone protection, has limited DNA repair capacity, and is located in a close proximity to the electron transport chain, the major site for ROS generation. As such, mtDNA is more vulnerable to damage compared to nuclear DNA. Since mitochondrial genome is devoid of introns, mutations in mtDNA are more likely to have significant functional consequences, leading to dysfunction in respiration and alterations in energy metabolism. Thus, mitochondrial genomic integrity is essential for maintaining normal ATP pro­duction and metabolic homeostasis (Wallace 1999). Because mitochondrial transcription is polycistronic, any deletion or insertion of a nucleo­ tide may cause a frame-shifting of the codon and thus affect the downstream genes encoding for the respiratory chain components (Zastawny et al. 1998). It was reported that mtDNA mutation rates are 10–20 times greater than nuclear DNA (Olgun et al. 2002), and increased mutations in mtDNA have been reported in various cancer cells (Verma et  al. 2003; Wallace 1999). Patients bearing mtDNA mutations show refractory to conventional therapeutic agents (Carew et al. 2003; Ohta 2006). ROS-induced mtDNA mutations may play a key role in many diseases including cancer, ageing, and neurodegenerative diseases. mtDNA mutations, which may directly cause defective mitochondrial respiration to generate ATP through the coupled electron-transport chain and OXPHOS, have been found to alter cellular energy metabolism and induce higher glycolysis in a large spectrum of human cancers, suggesting that mitochondrial respiratory dysfunction may be mechanistically linked with increased glycolysis in human cancer (Cavalli and Liang 1998; Halabe Bucay 2007). In cancer cells with altered mitochondrial metabolism, mitochondria seem to export a large amount of citrate, which may be utilized as a metabolic intermediate for the biosynthesis of cholesterol and fatty acids (Parlo and Coleman 1986). Interestingly, citrate has been observed to increase in certain subsets of cancers (Kuhajda et al. 1994; Ristow 2006). The impaired activity of mitochondrial aconitase in cancer cells may negatively affect the TCA cycle, leading to elevated fatty acid generation and a coupling of increased glycolysis with lipogenesis. In addition, impairment of the mitochondrial Fe-S complex activity and a decrease in OXPHOS can lead to altered energy metabolism and tumor formation (Thierbach et  al. 2005). ATP synthase, an enzyme required for OXPHOS, was found to be decreased in certain cancer types (Isidoro et al. 2004). Recently it was

100

W. Lu and P. Huang

reported that cells bearing mutations in mtDNA encoding for the NADH dehydrogenase complex component ND6 may cause defect in complex I, leading to elevated ROS production and increased metastatic potential (Ishikawa et al. 2008). The mitochondrial fumarate hydratase and succinate dehydrogenase (SDH), two enzymes involved in the TCA cycle and encoded by the nuclear DNA, have recently been reported to have tumor suppressor function (Baysal et  al. 2000; Tomlinson et  al. 2002). SDH is also an integral component of the electron transport chain complex II. Patients with inherited mutations in succinate dehydrogenase and fumarate hydratase are more likely to develop hereditary paraganglioma and papillary renal cell carcinoma (Baysal et al. 2000; Tomlinson et al. 2002). This may be due to the accumulation of succinate or fumarate, resulting in the inhibition of prolyl hydroxylase function that otherwise promotes HIF degradation. These tumors display high HIF activity even in the presence of oxygen (Isaacs et al. 2005; Pollard et al. 2005; Selak et al. 2005). Further study in fumarate hydratase-deficient cells showed an increase in LDH-A expression due to HIF-1a stabilization. Interestingly, inhibition of LDH-A in these cells led to ROS-mediated apoptosis. A knockdown of LDH-A resulted in a significant reduction in tumor growth in xenograft mouse model (Xie et al. 2009). Recently, it is reported that SDH5, a gene required for the interaction and flavination of SDH, is mutated in paraganglioma, a neuroendocrine tumor previously linked to mutations in genes encoding for SDH subunits (Hao et al. 2009). In addition, cells with increased HIF-1 stabilization or p53 mutation also show elevated aerobic glycolysis and reduced mitochondrial respiration. Such metabolic shift is characteristic of the Warburg effect and seems to play an important role in malignant transformation, not just a bystander in cancer (Frezza and Gottlieb 2009).

Tumor Microenvironment and Selection of Highly Glycolytic Cancer Cells HIF-1 and Glycolysis Cancer cells in a tumor mass often endure hypoxia, especially when the tumor mass has overgrown the capacity of blood supply. The ability of cancer cells to adapt and survive in hypoxic environment is a crucial step in tumor progression (Dang and Semenza 1999). Under hypoxic conditions, mitochondrial respiration may be limited and ATP generation through OXPHOS is decreased, forcing the cancer cells to utilize alternative mechanisms such as increased glycolysis to maintain sufficient cellular ATP for cell growth and proliferation. During adaptation to hypoxic microenvironment, the increase in HIF-1 plays an essential role (Sharp and Bernaudin 2004). HIF-1 is a heterodimer of HIF1a and HIF1b and functions as a transcription factor that regulates a number of genes critical for glycolysis, angiogenesis, cell proliferation, mitochondrial ­function, and cell sur-

5  Glycolytic Pathway as a Target for Tumor Inhibition

101

vival. Under physiologic conditions, the half-life (protein stability) and the transcriptional activity of HIF1a are negatively regulated by the O2-dependent prolyl and asparaginyl hydroxylation, respectively (Semenza 2003). Prolyl hydroxylase is a key enzyme capable of modifying HIF-1a at proline sites and marking it for Von Hippel Lindau protein (VHL)-mediated degradation by the 26S proteasome (Cockman et al. 2000; Thomas et al. 2003). The stabilized HIF-1a may be transported into the nucleus and forms a heterodimer with HIF-1b. The HIF1 heterodimer can then recruit p300/CBP and bind to the promoter region of the specific DNA sequence known as the hypoxia responsive element (HRE) to promote the expression of its targeted genes (Dang and Semenza 1999). Examples of HIF-1 downstream genes include a number of molecules involved in glycolysis such as GLUT1, HKII, aldolase A, enolase 1, LDHA, PFK, PGK, and pyruvate kinase (Dang and Semenza 1999; Denko 2008). The increased glycolytic gene expression due to HIF-1a stabilization in turn leads to enhanced glycolysis (Lu et al. 2002; Safran and Kaelin 2003). Clinically, cancer patients whose tumors with a greater hypoxic fraction seem to be associated with high HIF-1 expression and resistance to chemotherapy and radiotherapy (Hockel et  al. 1993; Hockel and Vaupel 2003). Furthermore, due to the lack of pVHL tumor suppressor in some clear cell renal carcinoma, HIF-1 is consistently expressed and its targeted glycolytic genes are upregulated even under normoxic condition, resulting in the increased glucose uptake, increased lactate production, and decreased mitochondrial respiration (Semenza 2007). HIF-1 can also be activated by other non-hypoxic stimuli such as oncogenes and ROS, leading to increased glycolysis.

HIF-1 and Mitochondria In addition to promoting glycolysis, HIF-1 can also suppress aerobic respiration by enhancing the expression of pyruvate dehydrogenase kinase 1 (PDK1), which inhibits the activity of PDH. As illustrated in Fig. 5.2, an increased expression of PDK1 promotes the phosphorylation of PDH, leading to its inactivation, and thus reduces the conversion of pyruvate to acetyl CoA. As such, an increase in HIF-1 due to hypoxic microenvironment may limit the entry of pyruvate into the TCA cycle in the mitochondria, and thus suppresses the utilization of glucose as the energy supply via OXPHOS pathway (Papandreou et al. 2006). This suppression of substrate entry into the TCA cycle would further enhance the dependence of cells on glycolysis for ATP generation. The ability of HIF-1 to enhance glycolysis has been observed in many tumor cells. As described above, glycolysis not only serves as a rapid energetic pathway to generate ATP but also provides the building blocks for the synthesis of other biomolecules in cancer cells. For instance, the metabolic intermediates of the glycolytic pathway can be converted to ribose for the synthesis of nucleic acids via the PPP. Thus, it is possible that

102

W. Lu and P. Huang Glucose GLUT1

p53

Glucose

AMPK

LKB -1

Bioenergetic stress

HK

TIGAR

AMP

G6P PFK2

H 2O

PGI

F6P

F2,6-BP

O2

SCO2 Respiration Chain

PFK1

F1,6-BP Aldolase

NADH FADH2

Dihydroxyacetone-P + GA3P GAPDH

Pyr

1,3-BPG

Myc

Ac-CoA PDH

TCA cycle Mito.

PGK

3-PG

PDK1

PGM

2-PG Enolase

PEP

PK-M2

Pyr Lactate

LDH-A

AKT HIF-1

PIP3 Ras

PTEN

PI3K

MCT 4

Lactate

Fig. 5.2  Potential metabolic regulatory mechanisms contributing to the Warburg effect. Several important molecules including the oncogenic molecule Myc, the tumor suppressor p53, the transcription factor HIF-1, and the kinase Akt play essential roles in the regulation of cellular metabolism and contribute to the development of Warburg phenomenon in cancer cells. Activation of Myc oncogene upregulates the expression of many glycolytic molecules including the glucose transporter GLUT1, HK, and LDH, leading to increased glycolysis (orange arrows). Similarly, stabilization of HIF-1 due to hypoxia in tumor microenvironment or oncogenic signaling (e.g. Ras) may transactivate many glycolytic genes including GLUT1, HKII, PK-M2, and LDH-A, and enhances glycolytic flux (red arrows). HIF-1 also activates PDK1 gene expression, which in turn inhibits PDH and thus limits pyruvate entry into the mitochondria as fuel to generate ATP. Akt promotes plasma membrane association of GLUT1 to increase glucose uptake, and enhance HK association with the mitochondria and prevent apoptosis (blue arrows). On the contrary, the tumor suppressor p53 inhibits glycolysis by regulating the expression of HKII, PGM, and TIGAR, and stimulates mitochondrial respiration through transcriptional upregulation of SCO2 (green arrows). The loss of p53 function in cancer cells will lead to an increase in glycolysis and a decrease in respiration. The AMP-dependent kinase (AMPK) regulates glycolysis in part through its activation of p53

even in cancer cells without significant mitochondrial structural defects, the abnormal increase of HIF-1 may block normal mitochondrial function through transcriptional activation of PDK1, leading to a decrease in OXPHOS and an increase in glycolysis (Lu et al. 2002; Safran and Kaelin 2003). As such, the suppression of PDH by elevated HIF-1 in hypoxic tissue microenvironment may

5  Glycolytic Pathway as a Target for Tumor Inhibition

103

significantly contribute to the decrease in mitochondrial function and increase in glycolysis in cancer cells, characteristic of the Warburg effect.

Mutations of Tumor Suppressor Genes and Metabolic Alterations p53 Regulation of Glycolysis and Mitochondrial Respiration The tumor suppressor p53 plays major roles in a number of important cellular processes including stress responses (Vogelstein et al. 2000). Through transcriptional activation of its target genes, p53 involves in the regulation of cell cycle, DNA damage response, apoptosis, senescence, and metabolism. Through the transcriptional regulation of HK, PGM, TIGAR, and SCO2, p53 seems to prevent a switch from aerobic respiration to glycolysis (Green and Chipuk 2006; Shaw 2006). As illustrated in Fig. 5.2, HKII is the first rate-limiting enzyme of the glycolysis. The wildtype p53 protein can bind to the HKII promoter region and down-regulate HKII gene expression, limiting the entry of glucose into the glycolytic pathway. Similarly, p53 down-regulates the function of PGM, a glycolytic enzyme that converts 3-PG to 2-PG, and thus modulates the overall flux of glycolysis (Kondoh et  al. 2005; Smith et al. 2006). The ability of p53 to modulate glycolysis has been further substantiated by the demonstration that p53 promotes the expression of TIGAR, a novel isoform of phosphofructose kinase (PFK2) that only contains the FBPase domain but lacks the kinase domain. It functions to lower the intracellular levels of fructose-2,6-bisphosphate. Because fructose-2,6-bisphosphate is a key allosteric activator of PFK1 (a key rate-limiting glycolytic enzyme), the elevated expression of TIGAR in cells with functional p53 would have low fructose-2, 6-bisphosphate levels and thus a lower flow of glucose metabolism through the step catalyzed by PFK1 (Bensaad et al. 2006). This causes a shift toward the PPP shunt, leading to increased production of NADPH, the major reducing equivalent for the generation of reduced glutathione and other cellular antioxidants, which in turn promote the scavenging of ROS and prevent mutational events associated with cancer. PPP can also produce ribose-5-phosphate, which could contribute to nucleotide biosynthesis and DNA repair. Therefore, through upregulation of TIGAR, p53 sets a glycolytic checkpoint to suppress glycolysis, to enhance ROS-scavenging capacity, and to promote genomic stability (Green and Chipuk 2006). A loss of p53 function in cancer cells, on the other hand, would promote glycolysis, leading to the Warburg phenomenon. Recently, p53 has been found to directly regulate mitochondrial energy production through transcriptional regulation of the synthesis of cytochrome c oxidase 2 (SCO2), a regulator of mitochondrial respiration and the oxygen utilization in the eukaryotic cells. SCO2 encodes a copper-binding protein required for the assembly of cytochrome c oxidase subunit II into the mitochondrial respiratory chain complex IV. Disruption of the SCO2 gene in human cancer cells with wild-type p53 ­recapitulated

104

W. Lu and P. Huang

the metabolic switch toward glycolysis similar to that observed in p53-deficient cells (Bensaad et al. 2006; Matoba et al. 2006). Therefore, p53 is able to regulate glycolysis and mitochondrial respiration in the concerted manner through HK, PGM, TIGER, and SCO2. Because p53 mutations and deletion are common in human cancer, it is likely that the loss of p53 function may contribute significantly to the switch from aerobic respiration to glycolysis observed in many cancer types (Shaw 2006), and provide a molecular link between genetic alterations and the Warburg effect. AMPK and Glycolytic Regulation AMP-activated protein kinase (AMPK) is a master metabolic regulator that plays an important role in cellular energy homeostasis. AMPK is a heterotrimeric protein complex consisting of an a catalytic subunit and b, g regulatory subunits (Kahn et al. 2005). It functions as a serine–threonine kinase at a regulatory pathway that integrates signals from energy metabolism (Fig. 5.2). As cellular ATP decreases and AMP level increases, AMPK is allosterically activated by increased cellular AMP levels, a condition that occurs under energetic stress such as cellular ATP depletion or hypoxia. Through binding to the regulatory g-subunit of the AMPK complex, AMP triggers a conformational change of the g-subunit, leading to the exposure of the active site Thr-172 on the a catalytic subunit. This in turn inhibits dephosphorylation and facilitates phosphorylation at Thr-172 by the upstream kinases (Suter et al. 2006). LKB-1, functioning as a tumor suppressor, is a primary upstream kinase of AMPK (Hardie 2007). LKB1 is required for AMPK activation under bioenergetic stress such as glucose withdrawal (Hawley et al. 2003; Shaw et al. 2004b; Woods et al. 2003). Activation of AMPK promotes phosphorylation of p53 at Ser-15 to regulate glycolysis. AMPK can also stimulate GLUT1,3mediated glucose uptake, and therefore enhances glycolytic pathway (Cidad et al. 2004). In addition, AMPK can phosphorylate the TSC1-TSC2 complex or/and block TORC1 activity to inhibit mammalian target of rapamycin (mTOR)-dependent protein translation in an LKB-dependent manner (Inoki et al. 2003). AMPK has been reported to down-regulate lipid synthesis through inhibition of acetylCoA carboxylase-1 (ACC1) (Hardie and Pan 2002). Mutations in LKB-1 have been associated with Peutz–Jeghers Syndrome (Scott et al. 2008). Mice with heterozygous disruption of LKB-1 develop neoplasia similar to Peutz–Jeghers Syndrome (Bardeesy et  al. 2002). Loss of LKB-1 results in unchecked cell growth and the development of aggressive cancer (Gurumurthy et al. 2008; Ji et al. 2007).

Activation of Oncogenes and Increased Glycolysis c-Myc c-Myc is an oncogenic transcription factor with the helix-loop-helix leucine zipper structure. Through dimerization with its partner Max, c-Myc binds to specific DNA sequences known as E box, and stimulates the transcription of many genes

5  Glycolytic Pathway as a Target for Tumor Inhibition

105

involved in cell proliferation, metabolism, and apoptosis (Dang 1999). Max can form heterodimer with the Mad family proteins to repress transcription and therefore antagonize c-Myc function and achieve tight c-Myc regulation. Deregulated expression of the c-Myc gene is a common feature of human malignancies (Little et  al. 1983). c-Myc expression enhances aerobic glycolysis by directly upregulating the expression of glycolytic genes such as enolase A, HK II, LDH-A, PFK, GAPDH, and GLUT1 independent of hypoxia (Kim et  al. 2004). Many of the genes regulated by c-Myc are known to be regulated by HIF-1 as well (Dang et al. 2008). This suggests a possibility that in normal cells the low expression of glycolytic genes may be due to the tightly controlled Myc and HIF-1 expression, while in cancer cells, the increased expression of glycolytic enzymes may be attributed to the constitutive activation of c-Myc and/or elevated HIF-1 in response to the hypoxic tumor microenvironment. Genes of mitochondrial biogenesis and function such as cytochrome c and TFAM (transcription factor A of mitochondria) constitute another group of molecules upregulated by c-Myc overexpression (Dang et al. 2005). It is suggested that imbalanced expression of nuclear encoded versus mitochondrial encoded proteins could result in defective respiratory chain complex as noted in prostate cancer progression (Herrmann et  al. 2003). The ability of c-Myc to upregulate glycolysis in the normoxic condition suggests that Myc may play a role in a switch to glycolytic metabolism during tumorigenesis.

Ras Ras is a guanine nucleotide triphosphatase that functions as a molecular switch in a large network of intracellular signaling pathways. Two major downstream signal pathways, MAPK and PI3K/Akt, are found to be upregulated in many cancer cells and may collaboratively promote the malignant phenotype of cancer cells. In particular, it has been shown that PI3K is required for both the development and maintenance of certain tumors driven by mutant H-Ras (Lim and Counter 2005). Alteration in the Ras-PI3K-Akt signal transduction pathway can result in the upregulation of HKII and elevation of glycolytic activity. Ras activation can also upregulate HIF1, NF-kB, PFK1 and thus enhances tumor glycolysis (Kole et al. 1991). PI3K/Akt Pathway Phosphatidylinositol 3-kinase (PI3K) is a heterodimer consisting of a catalytic subunit p110 and a regulatory subunit p85. Binding of growth factor molecules to their receptor tyrosine kinase activates PI3K, resulting in the phosphorylation of phosphatidylinositol on the plasma membrane and the activation of its downstream pathways such as Akt and mTOR. Tumor suppressor phosphatase and tensin homolog (PTEN) is a critical negative counterpart of PI3K by dephospho-

106

W. Lu and P. Huang

rylation PIP3 ­(phosphatidylinositol 3,4,5-triphosphate) to PIP2, leading to an inhibition of the Akt signal pathway, which is a serine/threonine kinase regulated by PIP3. Akt activation may result from PI3K activation or loss of PTEN function. Other oncogenes, such as Ras, HER2/neu, and Bcr-Abl, can also activate Akt. The PI3K/Akt kinase pathway is a central regulator of cell metabolism, proliferation, and survival and is deregulated in many cancer types. A recent paper reports that Akt stimulates glucose consumption in transformed cells and Akt-expressing cells are more susceptible to death after glucose withdrawal than control cells (Elstrom et al. 2004). Activation of the PI3K/Akt pathway results in coordinated localization of glucose transporters to the cell surface and upregulation of glycolytic enzymes such as PFK2 and enhances glucose capture through glucose phosphorylation by HK (Pozuelo Rubio et al. 2003). Akt can also inactivate p53 through stabilization of Mdm2 (Gottlieb et al. 2002). Thus, Akt activation seems to be able to cause a metabolic switch from OXPHOS to aerobic glycolysis in cancer cells. As such, Akt is considered as “Warburg kinase” by some investigators (Robey and Hay 2009). In addition to its role in glycolysis, Akt is also involved in other metabolic processes. For instance, Akt phosphorylates and inhibits glycogen synthase kinase-3, leading to the dephosphorylation of glycogen synthase by the glycogenassociated form of protein phosphatase 1, and a stimulation of glycogen synthesis. Activation of Akt stimulates the transport of both glucose and amino acids, which in turn support the mTOR-dependent promotion of protein translation. On the contrary, the LKB-1 tumor suppressor negatively regulates mTOR signaling (Shaw et  al. 2004a). Many human tumors display an increased expression of lipogenic enzymes including ATP citrate lyase (ACL, an enzyme that converts cytosolic citrate to acetyl-CoA for fatty acid synthesis), fatty acid synthase, and acetyl-CoA carboxylase (Milgraum et al. 1997; Rossi et al. 2003; Swinnen et al. 2000; Yahagi et al. 2005). Besides, some tumors have a high rate of de-novo fatty acid synthesis from carbohydrate precursors as shown by 14C-glucose studies (Kuhajda 2000; Ookhtens et al. 1984). Such an increase of lipogenic activity in cancer cells may in part be attributed to Akt activation, which stimulates the synthesis and nuclear accumulation of a key lipogenic transcription factor sterol regulatory element binding protein-1, resulting in the upregulation of lipogenic enzymes (Porstmann et al. 2005). Akt also directly phosphorylates and activates ACL (Swinnen et al. 2006). Thus, it appears that the PI3K/Akt/mTOR axis is a major regulatory pathway that governs glycolysis and cellular biosynthesis (DeBerardinis et al. 2008a).

Glycolytic Pathway as a Target for Tumor Inhibition Compelling evidence shows that cancer cells have profound alterations in their energy metabolism, with an increase in aerobic glycolysis as one of the most prominent metabolic changes. As described above, multiple factors including mitochondrial dysfunction, oncogenic signals, loss of tumor suppressor function, and aberrant expression of certain regulatory molecules and enzymes can all contribute

5  Glycolytic Pathway as a Target for Tumor Inhibition

107

to the highly glycolytic phenotype of cancer cells. Several key regulators of cancer metabolism and their influence on the major metabolic steps are illustrated in Fig. 5.2. Although whether an increase in glycolysis is a cause or a “symptom” of malignant transformation still remains as a matter of debate, the profound metabolic alterations in cancer cells are likely to have significant therapeutic implications. In particular, the increased dependency of cancer cells on the glycolysis for energy supply and for metabolic intermediates as the building blocks for cell proliferation provides a biochemical basis to preferentially inhibit tumor growth by targeting the glycolytic pathway. Recent studies suggest that it is indeed possible to use small compounds or siRNA to abrogate certain key steps in the glycolytic pathway by suppression of the key enzymes such as HK and LDH, leading to selective killing of cancer cells (Fantin et al. 2006; Pelicano et al. 2006; Xu et al. 2005b). Theoretically, many of the enzymes and regulatory molecules shown in Figs. 5.1 and 5.2 may serve as potential targets for such a metabolic intervention. The following sections provide a brief summary of several important compounds that have been shown to inhibit glycolysis with potential use in cancer therapeutics.

2-Deoxyglucose 2-Deoxyglucose is a glucose molecule which has the 2-hydroxyl group replaced by hydrogen. Upon transport into the cells, 2-deoxyglucose is phosphorylated by HK to 2-deoxyglucose-phosphate, which cannot be further metabolized by phosphohexose isomerase. Thus, 2-deoxyglucose-phosphate accumulates in the cells and interferes with glycolytic metabolism by competing with glucose for glycolytic enzymes such as HK and phosphoglucose isomerase. Because of the presence of abundant glucose in the cell culture medium and in the blood circulation, high concentrations of 2-deoxyglucose in the millimolar range are required to be effective to inhibit cancer cell growth. An in  vivo study showed that this compound alone exhibited limited anticancer activity, but could significantly enhance the therapeutic efficacy of doxorubicin and taxol in animal tumor models (Maschek et al. 2004). This compound has been evaluated clinically in phase I trials to determine its safety profile and pharmacokinetics in 34 patients with solid tumors who had relapsed after chemotherapy with other agents. It appears that a combination of 2-deoxyglucose and docetaxel was safe with no evidence of pharmacokinetic interaction (http://www.thresholdpharm.com/sec/pipe_2deoxyglucose). The clinical therapeutic activity of this compound, alone or in combination with other anticancer agents, still remains unclear and requires further evaluation.

3-Bromopyruvate The pyruvate analog 3-bromopyruvate (3-BrPA) has been shown to inhibit HK and cause severe depletion of cellular ATP, leading to massive cancer cell death, especially in cells with mitochondrial respiratory dysfunction or under hypoxic conditions

108

W. Lu and P. Huang

(Geschwind et al. 2004; Ko et al. 2001; Xu et al. 2005b). Although depletion of cellular ATP is thought to be a key mechanism by which 3-BrPA exerts its cytotoxic effect, a recent study also showed that this compound could covalently modify HKII protein and cause its dissociation from mitochondria. Dissociation of HKII from mitochondria could in turn cause the release of apoptosis-inducing factor and subsequent cell death (Chen et al. 2009). It is worth noting that cancer cells with mitochondrial dysfunction or in hypoxic environment usually become less sensitive to many anticancer drugs currently used in clinic, due in part to the accumulation of HIF-1 and activation of Akt survival pathways. These cancer cells, however, are particularly sensitive to the cytotoxic action of 3-BrPA because of their high dependency on glycolysis. Furthermore, ATP depletion by 3-BrPA may disable the ATPdependent drug-exporting pump and thus overcome multi-drug resistance associated with increased expression of MDR pump (Xu et  al. 2005b). Thus, the unique mechanism of action of 3-BrPA makes this compound potentially useful in killing cancer cells in hypoxic tumor and overcoming multi-drug resistance. Interestingly, combination of glycolytic inhibition with the mTOR inhibitor rapamycin seems to have synergistic anticancer effect (Xu et al. 2005a). It should be pointed out, however, that 3-BrPA is an alkylating agent, which may potentially attack other cellular molecules, and the cytotoxic action of this compound may not be entirely due to its inhibition of HKII. Although there is compelling evidence showing that 3-BrPA has promising anticancer activity both in  vitro and in  vivo in several animal tumor models, this compound is still in its pre-clinical development, and has not yet entered clinical trials.

Lonidamine Lonidamine (1-(2,4-dichlorobenzyl)-1H-indazole-3-carboxylic acid) is a derivative of indazole-3-carboxylic acid known to inhibit glycolysis in tumor cells. This compound is an orally administered small molecule that has been evaluated in phase II/III clinical trials for the treatment of various solid tumors (Di Cosimo et  al. 2003). The mechanism by which lonidamine suppresses glycolysis appears through the inhibition of the mitocondrially bound HK (Floridi et al. 1998). Interestingly, studies using murine cells showed that this compound inhibits aerobic glycolysis in tumor cells but enhances glycolysis in the normal cells (Floridi et al. 1998). The mechanism responsible for this differential effect on glycolysis in normal and cancer cells still remains unclear. Further studies by the same group showed that lonidamine may also inhibit mitochondrial respiration, as evidenced by the suppression of oxygen consumption in both normal and cancer cells (Floridi et  al. 1998). The anticancer activity of this compound may be ascribed to its overall ability to abrogate energy metabolism in cancer cells. In clinical trials, lonidamine appeared to be able to enhance the therapeutic activity of several major anticancer agents, including paclitaxel and cisplatin in the treatment of advanced ovarian cancer (De Lena et al. 2001).

5  Glycolytic Pathway as a Target for Tumor Inhibition

109

Oxythiamine and 6-Aminonicotinamide The PPP is important for cancer cell survival because this pathway provides the metabolic intermediates ribose-5-phosphate for the synthesis of nucleic acids and NADPH for scavenging of reactive oxygen species (ROS). In addition, the non-oxidative phase of PPP also converts 5-carbon metabolic intermediates to 6-carbon and 3-carbon compounds (e.g. F6P and glyceraldehydes-3-phosphate), which can then be used in the glycolytic pathway. Thus, PPP is an important sugar metabolic pathway that generates the key building blocks for cell proliferation, reducing equivalents for redox regulation, and metabolic intermediates for energy metabolism. The enzymes transketolase and transaldolase are essential for the non-oxidative phase of PPP, and seem to be increased in tumor cells (Coy et al. 2005; Heinrich et al. 1976). Oxythiamine is a thiamine antagonist that inhibits transketolase and PDH, which use thiamine pyrophosphate (TPP) as a cofactor for their enzymatic activities. Oxythiamine has been shown to have anticancer activity both in  vitro and in  vivo (Comin-Anduix et  al. 2001; Rais et al. 1999). Specific inhibition of TKTL1 by RNAi can inhibit cancer cell proliferation and compromise cell viability (Hu et al. 2007; Zhang et al. 2007). The compound 6-aminonicotinamide (6-AN) is an inhibitor of G6PD, an enzyme that catalyzes the conversion of G-6-P to 6-phosphogluconolactone at the first step of PPP. This compound exhibits anticancer activity, causes oxidative stress, and sensitizes cells to anticancer agents (Budihardjo et al. 1998; Varshney et al. 2003). However, potential neurotoxicity is a concern for using 6-AN as a therapeutic agent.

Dichloroacetate Dichloroacetate (DCA) is a small compound capable of penetrating most tissues after oral administration, and has been used in the treatment of human lactic acidosis and inherited mitochondrial diseases. Although this compound could lower blood lactate levels, it showed no significant clinical benefit in a controlled clinical trial (Stacpoole and Greene 1992). However, a recent study showed that DCA exhibited significant anticancer activity by inhibiting pyruvate dehydrogenase kinase and thus activating PDH and enhancing the flux of pyruvate into the mitochondria (Bonnet et al. 2007). As such, DCA is able to decrease aerobic glycolysis and promote glucose oxidation and mitochondrial function, leading to a reversion of the Warburg effect in cancer cells. Through promoting cytochrome c release and ROS generation, DCA was shown to induce apoptosis in various human cancer cells (Bonnet et  al. 2007; Cao et  al. 2008), and enhance the effectiveness of hypoxiaspecific cytotoxic chemotherapy in solid tumor (Cairns et al. 2007). Administration of DCA decreases lactate level, activates PDH activity, and decreases tumor growth in xenograft models (Bonnet et  al. 2007; Parolin et  al. 2000). With its apparent advantages in drug delivery and anticancer activity/selectivity, DCA is currently in phase I trial in Canada for the treatment of recurrent or metastatic solid tumors.

110

W. Lu and P. Huang

Other Metabolic Modulators Other metabolic modulators, which either involve in the direct inhibition of glycolysis or target the relevant regulatory pathways, are also under investigation in pre-clinical and clinical settings (Chen et al. 2007; Lopez-Lazaro 2008; Pelicano et  al. 2006). For instance, metformin is an inhibitor of gluconeogenesis and has been used for clinical treatment of diabetes. This compound is able to activate AMPK and attenuate the mTOR pathway, thus impacting glycolysis and inhibiting tumor growth. Interestingly, a large case–control study suggests that the use of metformin may significantly reduce the risk of developing pancreatic cancer (Li et al. 2009). A potent inhibitor of the glycolytic enzyme PGM was identified as an effective agent against breast cancer in a chemical screening study (Evans et  al. 2005). Selective LDH inhibitors such as dihydroxynaphthoic acids have been developed (Yu et al. 2001), and their potential anticancer activity should be further tested. The MCT, which functions to export lactate out of the cells, is upregulated in various human tumor and may be considered as a potential target for tumor inhibition (Sonveaux et al. 2008). NVP-BEZ235, a small molecule that inhibits PI3K and mTOR, is able to decrease glucose uptake and cause marked tumor regression as measured by FDG-PET (Engelman et al. 2008). Disruption of glucose transport may also be a strategy to inhibit glucose metabolism in cancer. Imatinib, a BCRABL tyrosine kinase inhibitor, was found to reduce the expression of GLUT1 and decrease glycolysis in leukemic cells (Gottschalk et al. 2004).

Summary Increased aerobic glycolysis is a hallmark of cancer metabolism. Recent studies suggest that this metabolic alteration may confer proliferative and survival advantages to cancer cells, and help the selection of aggressive malignant cells in the tumor microenvironment. Although OXPHOS in the mitochondria is highly efficient in ATP generation, glycolysis can rapidly generate ATP and, in conjunction with the PPP, can provide the building bricks for the synthesis of nucleic acids, lipids, and proteins to meet the needs of cancer cell proliferation (Bui and Thompson 2006). On the contrary, normal cells mainly use mitochondrial OXPHOS to oxidize glucose to CO2 and H2O and generate a large amount of ATP to fulfill the energy needs of cellular activity with little or low requirements for proliferation in welldifferentiated adult tissues. This profound metabolic difference between cancer and normal cells provides a biochemical basis for therapeutic intervention to preferentially kill cancer cells by inhibition of glycolysis. In addition, beyond their roles in catalyzing glucose breakdown and providing metabolic intermediates for the anabolic reactions, the elevated glycolytic enzymes in cancer cells may also contribute to tumor adaptation and survival through their non-glycolytic functions. For example, HKII exerts its anti-apoptotic function through association with the mitochondria

5  Glycolytic Pathway as a Target for Tumor Inhibition

111

and preventing cytochrome c release. Sustained aerobic glycolysis may also protect cancer cell from ROS-induced cell damage (Ahmad et  al. 2005). The metabolic transformation of cancer cells can be achieved by upregulation of the key glycolytic enzymes and glucose transporters through various regulatory mechanisms, including oncogene activation and loss of tumor suppressors. Cancer cells are able to overproduce lactic acid aerobically, whereas normal cells undergo anaerobic glycolysis only when deprived of oxygen. The increase in aerobic glycolysis and elevation of lactate may confer cancer cells an advantage to invade surrounding tissues and facilitate their invasion and metastasis. Thus, inhibition of glycolysis may also be effective in suppressing cancer metastasis. It should be pointed out, however, that cancer cells are genetically unstable with a high degree of heterogeneity in gene regulation and metabolism. Some cancer cells may be more dependent on glutamine metabolism instead of glycolysis for energy supply (Deberardinis et al. 2008b), and inhibition of glycolysis may not be effective in killing these cancer cells. Furthermore, because certain normal tissues such as heart and brain also use glucose as an important fuel source, inhibition of glucose utilization in these tissues might compromise their normal functions. Thus, it is extremely important to understand the exact differences between normal and cancer cells in their metabolic regulation, and specifically target the unique metabolic alterations in cancer cells to improve therapeutic selectivity. With the new discoveries of metabolic regulatory mechanisms and novel glycolytic inhibitors with improved pharmacokinetic and pharmacodynamic properties, it is possible to develop more effective therapeutic strategies with high selectivity against cancer.

References Ahmad IM, Aykin-Burns N, Sim JE, Walsh SA, Higashikubo R, Buettner GR, Venkataraman S, Mackey MA, Flanagan SW, Oberley LW et al (2005) Mitochondrial O2*- and H2O2 mediate glucose deprivation-induced stress in human cancer cells. J Biol Chem 280:4254–4263. Altenberg B, Greulich KO (2004) Genes of glycolysis are ubiquitously overexpressed in 24 cancer classes. Genomics 84:1014–1020. Bardeesy N, Sinha M, Hezel AF, Signoretti,S, Hathaway NA, Sharpless NE, Loda,M, Carrasco DR, DePinho RA (2002) Loss of the Lkb1 tumour suppressor provokes intestinal polyposis but resistance to transformation. Nature 419:162–167. Baysal BE, Ferrell RE, Willett-Brozick JE, Lawrence EC, Myssiorek D, Bosch A, van der Mey A, Taschner PE, Rubinstein WS, Myers EN et al (2000) Mutations in SDHD, a mitochondrial complex II gene, in hereditary paraganglioma. Science (New York, NY) 287:848–851. Bensaad K, Tsuruta A, Selak MA, Vidal MN, Nakano K, Bartrons R, Gottlieb E, Vousden KH (2006) TIGAR, a p53-inducible regulator of glycolysis and apoptosis. Cell 126:107–120. Bonnet S, Archer SL, Allalunis-Turner J, Haromy A, Beaulieu C, Thompson R, Lee CT, Lopaschuk GD, Puttagunta L, Bonnet S et  al (2007) A mitochondria-K+ channel axis is suppressed in cancer and its normalization promotes apoptosis and inhibits cancer growth. Cancer Cell 11: 37–51. Brown RS, Goodman TM, Zasadny KR, Greenson JK, Wahl RL (2002) Expression of hexokinase II and Glut-1 in untreated human breast cancer. Nuclear Med Biol 29:443–453.

112

W. Lu and P. Huang

Budihardjo II, Walker DL, Svingen PA, Buckwalter CA, Desnoyers S, Eckdahl S, Shah GM, Poirier GG, Reid JM, Ames MM et al (1998) 6-Aminonicotinamide sensitizes human tumor cell lines to cisplatin. Clin Cancer Res 4:117–130. Bui T, Thompson CB (2006) Cancer’s sweet tooth. Cancer Cell 9:419–420. Cairns RA, Papandreou I, Sutphin PD, Denko NC (2007) Metabolic targeting of hypoxia and HIF1 in solid tumors can enhance cytotoxic chemotherapy. Proc Natl Acad Sci USA 104: 9445–9450. Cao W, Yacoub S, Shiverick KT, Namiki K, Sakai Y, Porvasnik S, Urbanek C, Rosser CJ (2008) Dichloroacetate (DCA) sensitizes both wild-type and over expressing Bcl-2 prostate cancer cells in vitro to radiation. Prostate 68:1223–1231. Carew JS, Huang P (2002) Mitochondrial defects in cancer. Mol Cancer 1:9. Carew JS, Zhou Y, Albitar M, Carew JD, Keating MJ, Huang P (2003) Mitochondrial DNA mutations in primary leukemia cells after chemotherapy: clinical significance and therapeutic implications. Leukemia 17:1437–1447. Cavalli LR, Liang BC (1998) Mutagenesis, tumorigenicity, and apoptosis: are the mitochondria involved? Mutat Res 398:19–26. Chen Z, Lu W, Garcia-Prieto C, Huang P (2007) The Warburg effect and its cancer therapeutic implications. J Bioenerg Biomembr 39:267–274. Chen Z, Zhang H, Lu W, Huang P (2009) Role of mitochondria-associated hexokinase II in cancer cell death induced by 3-bromopyruvate. Biochim et Biophys Acta 1787:553–560. Christofk HR, Vander Heiden MG, Wu N, Asara JM, Cantley LC (2008) Pyruvate kinase M2 is a phosphotyrosine-binding protein. Nature 452:181–186. Cidad P, Almeida A, Bolanos JP (2004) Inhibition of mitochondrial respiration by nitric oxide rapidly stimulates cytoprotective GLUT3-mediated glucose uptake through 5’-AMP-activated protein kinase. Biochem J 384:629–636. Cockman ME, Masson N, Mole DR, Jaakkola P, Chang GW, Clifford SC, Maher ER, Pugh CW, Ratcliffe PJ, Maxwell PH (2000) Hypoxia inducible factor-alpha binding and ubiquitylation by the von Hippel-Lindau tumor suppressor protein. J Biol Chem 275:25733–25741. Comin-Anduix B, Boren J, Martinez S, Moro C, Centelles JJ, Trebukhina R, Petushok N, Lee WN, Boros LG, Cascante M (2001) The effect of thiamine supplementation on tumour proliferation. A metabolic control analysis study. Eur J Biochem/FEBS 268:4177–4182. Coy JF, Dressler D, Wilde J, Schubert P (2005) Mutations in the transketolase-like gene TKTL1: clinical implications for neurodegenerative diseases, diabetes and cancer. Clin Lab 51:257–273. Czernin J, Phelps ME (2002) Positron emission tomography scanning: current and future applications. Annu Rev Med 53:89–112. Dang CV (1999) c-Myc target genes involved in cell growth, apoptosis, and metabolism. Mol Cell Biol 19:1–11. Dang CV, Kim JW, Gao P, Yustein J (2008) The interplay between MYC and HIF in cancer. Nat Rev 8:51–56. Dang CV, Li F, Lee LA (2005) Could MYC induction of mitochondrial biogenesis be linked to ROS production and genomic instability? Cell Cycle 4:1465–1466. Dang CV, Semenza GL (1999) Oncogenic alterations of metabolism. Trends Biochem Sci 24:68–72. De Lena M, Lorusso V, Latorre A, Fanizza G, Gargano G, Caporusso L, Guida M, Catino A, Crucitta E, Sambiasi D et al (2001) Paclitaxel, cisplatin and lonidamine in advanced ovarian cancer. A phase II study. Eur J Cancer 37:364–368. DeBerardinis RJ, Lum JJ, Hatzivassiliou G, Thompson CB (2008a) The biology of cancer: metabolic reprogramming fuels cell growth and proliferation. Cell Metab 7:11–20. Deberardinis RJ, Sayed N, Ditsworth D, Thompson CB (2008b) Brick by brick: metabolism and tumor cell growth. Curr Opin Genet Dev 18:54–61. Denko NC (2008) Hypoxia, HIF1 and glucose metabolism in the solid tumour. Nat Rev 8: 705–713. Di Cosimo S, Ferretti G, Papaldo P, Carlini P, Fabi A, Cognetti F (2003) Lonidamine: efficacy and safety in clinical trials for the treatment of solid tumors. Drugs Today (Barc) 39:157–174.

5  Glycolytic Pathway as a Target for Tumor Inhibition

113

Dombrauckas JD, Santarsiero BD, Mesecar AD (2005) Structural basis for tumor pyruvate kinase M2 allosteric regulation and catalysis. Biochemistry 44:9417–9429. Elstrom RL, Bauer DE, Buzzai M, Karnauskas R, Harris MH, Plas DR, Zhuang H, Cinalli RM, Alavi A, Rudin CM et al (2004) Akt stimulates aerobic glycolysis in cancer cells. Cancer Res 64:3892–3899. Engelman JA, Chen L, Tan X, Crosby K, Guimaraes AR, Upadhyay R, Maira M, McNamara K, Perera SA, Song Y et al (2008) Effective use of PI3K and MEK inhibitors to treat mutant Kras G12D and PIK3CA H1047R murine lung cancers. Nat Med 14:1351–1356. Evans MJ, Saghatelian A, Sorensen EJ, Cravatt BF (2005) Target discovery in small-molecule cell-based screens by in situ proteome reactivity profiling. Nat Biotechnol 23:1303–1307. Fantin VR, St-Pierre J, Leder P (2006) Attenuation of LDH-A expression uncovers a link between glycolysis, mitochondrial physiology, and tumor maintenance. Cancer Cell 9:425–434. Floridi A, Bruno T, Miccadei S, Fanciulli M, Federico A, Paggi MG (1998) Enhancement of doxorubicin content by the antitumor drug lonidamine in resistant Ehrlich ascites tumor cells through modulation of energy metabolism. Biochem Pharmacol 56:841–849. Frezza C, Gottlieb E (2009) Mitochondria in cancer: not just innocent bystanders. Semin Cancer Biol 19:4–11. Gatenby RA (1995) The potential role of transformation-induced metabolic changes in tumor-host interaction. Cancer Res 55:4151–4156. Gatenby RA, Gawlinski ET (1996) A reaction-diffusion model of cancer invasion. Cancer Res 56:5745–5753. Gatenby RA, Gawlinski ET, Gmitro AF, Kaylor B, Gillies RJ (2006) Acid-mediated tumor invasion: a multidisciplinary study. Cancer Res 66:5216–5223. Gatenby RA, Gillies RJ (2004) Why do cancers have high aerobic glycolysis? Nat Rev 4: 891–899. Geschwind JF, Georgiades CS, Ko YH, Pedersen PL (2004) Recently elucidated energy catabolism pathways provide opportunities for novel treatments in hepatocellular carcinoma. Expert Rev Anticancer Ther 4:449–457. Godoy A, Ulloa V, Rodriguez F, Reinicke K, Yanez AJ, Garcia Mde L, Medina RA, Carrasco M, Barberis S, Castro T et al (2006) Differential subcellular distribution of glucose transporters GLUT1-6 and GLUT9 in human cancer: ultrastructural localization of GLUT1 and GLUT5 in breast tumor tissues. J Cell Physiol 207:614–627. Golshani-Hebroni SG, Bessman SP (1997) Hexokinase binding to mitochondria: a basis for proliferative energy metabolism. J Bioenerg Biomembr 29:331–338. Gottlieb TM, Leal JF, Seger R, Taya Y, Oren M (2002) Cross-talk between Akt, p53 and Mdm2: possible implications for the regulation of apoptosis. Oncogene 21:1299–1303. Gottschalk S, Anderson N, Hainz C, Eckhardt SG, Serkova NJ (2004) Imatinib (STI571)-mediated changes in glucose metabolism in human leukemia BCR-ABL-positive cells. Clin Cancer Res 10:6661–6668. Green DR, Chipuk JE (2006) p53 and metabolism: inside the TIGAR. Cell 126:30–32. Gurumurthy S, Hezel AF, Sahin E, Berger JH, Bosenberg MW, Bardeesy N (2008) LKB1 deficiency sensitizes mice to carcinogen-induced tumorigenesis. Cancer Res 68:55–63. Haberkorn U, Strauss LG, Reisser C, Haag D, Dimitrakopoulou A, Ziegler S, Oberdorfer F, Rudat V, van Kaick G (1991) Glucose uptake, perfusion, and cell proliferation in head and neck tumors: relation of positron emission tomography to flow cytometry. J Nucl Med 32: 1548–1555. Halabe Bucay A (2007) The biological significance of cancer: mitochondria as a cause of cancer and the inhibition of glycolysis with citrate as a cancer treatment. Med Hypotheses 69: 826–828. Hao HX, Khalimonchuk O, Schraders M, Dephoure N, Bayley JP, Kunst H, Devilee P, Cremers CW, Schiffman JD, Bentz BG et al (2009) SDH5, a gene required for flavination of succinate dehydrogenase, is mutated in paraganglioma. Science (New York, NY) 325:1139–1142. Hardie DG (2007) AMP-activated/SNF1 protein kinases: conserved guardians of cellular energy. Nat Rev Mol Cell Biol 8:774–785.

114

W. Lu and P. Huang

Hardie DG, Pan DA (2002) Regulation of fatty acid synthesis and oxidation by the AMP-activated protein kinase. Biochem Soc Trans 30:1064–1070. Hawley SA, Boudeau J, Reid JL, Mustard KJ, Udd L, Makela TP, Alessi DR, Hardie DG (2003) Complexes between the LKB1 tumor suppressor, STRAD alpha/beta and MO25 alpha/beta are upstream kinases in the AMP-activated protein kinase cascade. J Biol 2:28. Heinrich PC, Morris HP, Weber G (1976) Behavior of transaldolase (EC 2.2.1.2) and transketolase (EC 2.2.1.1) activities in normal, neoplastic, differentiating, and regenerating liver. Cancer Res 36:3189–3197. Herrmann PC, Gillespie JW, Charboneau L, Bichsel VE, Paweletz CP, Calvert VS, Kohn EC, Emmert-Buck MR, Liotta LA, Petricoin EF, 3rd (2003) Mitochondrial proteome: altered cytochrome c oxidase subunit levels in prostate cancer. Proteomics 3:1801–1810. Hockel M, Knoop C, Schlenger K, Vorndran B, Baussmann E, Mitze M, Knapstein PG, Vaupel P (1993) Intratumoral pO2 predicts survival in advanced cancer of the uterine cervix. Radiother Oncol 26:45–50. Hockel M, Vaupel P (2003) Oxygenation of cervix cancers: impact of clinical and pathological parameters. Adv Exp Med Biol 510:31–35. Hove H, Rye Clausen M, Brobech Mortensen P (1993) Lactate and pH in faeces from patients with colonic adenomas or cancer. Gut 34:625–629. Hu LH, Yang JH, Zhang DT, Zhang S, Wang L, Cai PC, Zheng JF, Huang JS (2007) The TKTL1 gene influences total transketolase activity and cell proliferation in human colon cancer LoVo cells. Anticancer Drugs 18:427–433. Inoki K, Zhu T, Guan KL (2003) TSC2 mediates cellular energy response to control cell growth and survival. Cell 115:577–590. Isaacs JS, Jung YJ, Mole DR, Lee S, Torres-Cabala C, Chung YL, Merino M, Trepel J, Zbar B, Toro J et al (2005) HIF overexpression correlates with biallelic loss of fumarate hydratase in renal cancer: novel role of fumarate in regulation of HIF stability. Cancer Cell 8:143–153. Ishikawa K, Takenaga K, Akimoto M, Koshikawa N, Yamaguchi A, Imanishi H, Nakada K, Honma Y, Hayashi J (2008) ROS-generating mitochondrial DNA mutations can regulate tumor cell metastasis. Science (New York, NY) 320:661–664. Isidoro A, Martinez M, Fernandez PL, Ortega AD, Santamaria G, Chamorro M, Reed JC, Cuezva JM (2004) Alteration of the bioenergetic phenotype of mitochondria is a hallmark of breast, gastric, lung and oesophageal cancer. Biochem J 378:17–20. Ji H, Ramsey MR, Hayes DN, Fan C, McNamara K, Kozlowski P, Torrice C, Wu MC, Shimamura T, Perera SA et  al (2007) LKB1 modulates lung cancer differentiation and metastasis. Nature 448:807–810. Kahn BB, Alquier T, Carling D, Hardie DG (2005) AMP-activated protein kinase: ancient energy gauge provides clues to modern understanding of metabolism. Cell Metab 1:15–25. Kim JW, Dang CV (2006) Cancer’s molecular sweet tooth and the Warburg effect. Cancer Res 66: 8927–8930. Kim JW, Zeller KI, Wang Y, Jegga AG, Aronow BJ, O’Donnell KA, Dang CV (2004) Evaluation of myc E-box phylogenetic footprints in glycolytic genes by chromatin immunoprecipitation assays. Mol Cell Biol 24:5923–5936. Ko YH, Pedersen PL, Geschwind JF (2001) Glucose catabolism in the rabbit VX2 tumor model for liver cancer: characterization and targeting hexokinase. Cancer Lett 173:83–91. Kole HK, Resnick RJ, Van Doren M, Racker E (1991) Regulation of 6-phosphofructo-1-kinase activity in ras-transformed rat-1 fibroblasts. Arch Biochem Biophys 286:586–590. Kondoh H, Lleonart ME, Gil J, Wang J, Degan P, Peters G, Martinez D, Carnero A, Beach D (2005) Glycolytic enzymes can modulate cellular life span. Cancer Res 65:177–185. Krebs HA (1970a) The history of the tricarboxylic acid cycle. Perspect Biol Med 14:154–170. Krebs HA (1970b) Rate control of the tricarboxylic acid cycle. Adv Enzyme Regul 8: 335–353. Kuhajda FP (2000) Fatty-acid synthase and human cancer: new perspectives on its role in tumor biology. Nutrition 16:202–208.

5  Glycolytic Pathway as a Target for Tumor Inhibition

115

Kuhajda FP, Jenner K, Wood FD, Hennigar RA, Jacobs LB, Dick JD, Pasternack GR (1994) Fatty acid synthesis: a potential selective target for antineoplastic therapy. Proc Natl Acad Sci USA 91:6379–6383. Kuo W, Lin J, Tang TK (2000) Human glucose-6-phosphate dehydrogenase (G6PD) gene transforms NIH 3T3 cells and induces tumors in nude mice. Int J Cancer 85:857–864. Langbein S, Frederiks WM, Hausen AZ, Popa J, Lehmann J, Weiss C, Alken P, Coy JF (2008) Metastasis is promoted by a bioenergetic switch: new targets for progressive renal cell cancer. Int J Cancer 122:2422–2428. Langbein S, Zerilli M, Zur Hausen A, Staiger W, Rensch-Boschert K, Lukan N, Popa J, Ternullo MP, Steidler A, Weiss C et al (2006) Expression of transketolase TKTL1 predicts colon and urothelial cancer patient survival: Warburg effect reinterpreted. Br J Cancer 94:578–585. Li D, Yeung SC, Hassan MM, Konopleva M, Abbruzzese JL (2009) Antidiabetic therapies affect risk of pancreatic cancer. Gastroenterology 137:482–488. Lim KH, Counter CM (2005) Reduction in the requirement of oncogenic Ras signaling to activation of PI3K/AKT pathway during tumor maintenance. Cancer Cell 8:381–392. Little CD, Nau MM, Carney DN, Gazdar AF, Minna JD (1983) Amplification and expression of the c-myc oncogene in human lung cancer cell lines. Nature 306:194–196. Lopez-Lazaro M (2008) The warburg effect: why and how do cancer cells activate glycolysis in the presence of oxygen? Anticancer Agents Med Chem 8:305–312. Lu H, Forbes RA, Verma A (2002) Hypoxia-inducible factor 1 activation by aerobic glycolysis implicates the Warburg effect in carcinogenesis. J Biol Chem 277:23111–23115. Majewski N, Nogueira V, Bhaskar P, Coy PE, Skeen JE, Gottlob K, Chandel NS, Thompson CB, Robey RB, Hay N (2004) Hexokinase-mitochondria interaction mediated by Akt is required to inhibit apoptosis in the presence or absence of Bax and Bak. Mol Cell 16:819–830. Mandavilli BS, Santos JH, Van Houten B (2002) Mitochondrial DNA repair and aging. Mutat Res 509:127–151. Maschek G, Savaraj N, Priebe W, Braunschweiger P, Hamilton K, Tidmarsh GF, De Young LR, Lampidis TJ (2004) 2-Deoxy-d-glucose increases the efficacy of adriamycin and paclitaxel in human osteosarcoma and non-small cell lung cancers in vivo. Cancer Res 64:31–34. Mathupala SP, Ko YH, Pedersen PL (2006) Hexokinase II: cancer’s double-edged sword acting as both facilitator and gatekeeper of malignancy when bound to mitochondria. Oncogene 25:4777–4786. Mathupala SP, Rempel A, Pedersen PL (1997) Aberrant glycolytic metabolism of cancer cells: a remarkable coordination of genetic, transcriptional, post-translational, and mutational events that lead to a critical role for type II hexokinase. J Bioenerg Biomembr 29:339–343. Matoba S, Kang JG, Patino WD, Wragg A, Boehm M, Gavrilova O, Hurley PJ, Bunz F, Hwang PM (2006) p53 regulates mitochondrial respiration. Science 312:1650–1653. Milgraum LZ, Witters LA, Pasternack GR, Kuhajda FP (1997) Enzymes of the fatty acid synthesis pathway are highly expressed in in situ breast carcinoma. Clin Cancer Res 3:2115–2120. Minchenko A, Leshchinsky I, Opentanova I, Sang N, Srinivas V, Armstead V, Caro J (2002) Hypoxia-inducible factor-1-mediated expression of the 6-phosphofructo-2-kinase/fructose-2, 6-bisphosphatase-3 (PFKFB3) gene. Its possible role in the Warburg effect. J Biol Chem 277:6183–6187. Mitchell P, Moyle J (1967) Chemiosmotic hypothesis of oxidative phosphorylation. Nature 213:137–139. Newmeyer DD, Ferguson-Miller S (2003) Mitochondria: releasing power for life and unleashing the machineries of death. Cell 112:481–490. Ohta S (2006) Contribution of somatic mutations in the mitochondrial genome to the development of cancer and tolerance against anticancer drugs. Oncogene 25:4768–4776. Olgun A, Akman S, Serdar MA, Kutluay T (2002) Oxidative phosphorylation enzyme complexes in caloric restriction. Exp Gerontol 37:639–645. Ookhtens M, Kannan R, Lyon I, Baker N (1984) Liver and adipose tissue contributions to newly formed fatty acids in an ascites tumor. Am J Physiol 247:R146–R153.

116

W. Lu and P. Huang

Osthus RC, Shim H, Kim S, Li Q, Reddy R, Mukherjee M, Xu Y, Wonsey D, Lee LA, Dang CV (2000) Deregulation of glucose transporter 1 and glycolytic gene expression by c-Myc. J Biol Chem 275:21797–21800. Papandreou I, Cairns RA, Fontana L, Lim AL, Denko NC (2006) HIF-1 mediates adaptation to hypoxia by actively downregulating mitochondrial oxygen consumption. Cell Metab 3:187–197. Parlo RA, Coleman PS (1986) Continuous pyruvate carbon flux to newly synthesized cholesterol and the suppressed evolution of pyruvate-generated CO2 in tumors: further evidence for a persistent truncated Krebs cycle in hepatomas. Biochim Biophys Acta 886:169–176. Parolin ML, Spriet LL, Hultman E, Matsos MP, Hollidge-Horvat MG, Jones NL, Heigenhauser GJ (2000) Effects of PDH activation by dichloroacetate in human skeletal muscle during exercise in hypoxia. Am J Physiol 279:E752–E761. Pastorino JG, Hoek JB, Shulga N (2005) Activation of glycogen synthase kinase 3beta disrupts the binding of hexokinase II to mitochondria by phosphorylating voltage-dependent anion channel and potentiates chemotherapy-induced cytotoxicity. Cancer Res 65:10545–10554. Pelicano H, Martin DS, Xu RH, Huang P (2006) Glycolysis inhibition for anticancer treatment. Oncogene 25:4633–4646. Pollard PJ, Briere JJ, Alam NA, Barwell J, Barclay E, Wortham NC, Hunt T, Mitchell M, Olpin S, Moat SJ et  al (2005) Accumulation of Krebs cycle intermediates and over-expression of HIF1alpha in tumours which result from germline FH and SDH mutations. Hum Mol Genet 14:2231–2239. Porstmann T, Griffiths B, Chung YL, Delpuech O, Griffiths JR, Downward J, Schulze A (2005) PKB/Akt induces transcription of enzymes involved in cholesterol and fatty acid biosynthesis via activation of SREBP. Oncogene 24:6465–6481. Poulsen HS, Frederiksen P (1981) Glucose-6-phosphate dehydrogenase activity in human breast cancer. Lack of association with oestrogen receptor content. Acta Pathol Microbiol Scand 89:263–270. Pozuelo Rubio M, Peggie M, Wong BH, Morrice N, MacKintosh C (2003) 14-3-3s regulate fructose-2,6-bisphosphate levels by binding to PKB-phosphorylated cardiac fructose-2,6-bisphosphate kinase/phosphatase. EMBO J 22:3514–3523. Rais B, Comin B, Puigjaner J, Brandes JL, Creppy E, Saboureau D, Ennamany R, Lee WN, Boros LG, Cascante M (1999) Oxythiamine and dehydroepiandrosterone induce a G1 phase cycle arrest in Ehrlich’s tumor cells through inhibition of the pentose cycle. FEBS Lett 456:113–118. Ramos-Montoya A, Lee WN, Bassilian S, Lim S, Trebukhina RV, Kazhyna MV, Ciudad CJ, Noe V, Centelles JJ, Cascante M (2006) Pentose phosphate cycle oxidative and nonoxidative balance: a new vulnerable target for overcoming drug resistance in cancer. Int J Cancer 119: 2733–2741. Rathmell JC, Fox CJ, Plas DR, Hammerman PS, Cinalli RM, Thompson CB (2003) Akt-directed glucose metabolism can prevent Bax conformation change and promote growth factorindependent survival. Mol Cell Biol 23:7315–7328. Ristow M (2006) Oxidative metabolism in cancer growth. Curr Opin Clin Nutr Metab Care 9:339–345. Robey RB, Hay N (2005) Mitochondrial hexokinases: guardians of the mitochondria. Cell cycle (Georgetown, TX) 4:654–658. Robey RB, Hay N (2009) Is Akt the “Warburg kinase”?-Akt-energy metabolism interactions and oncogenesis. Semin Cancer Biol 19:25–31. Rossi S, Graner E, Febbo P, Weinstein L, Bhattacharya N, Onody T, Bubley G, Balk S, Loda M (2003) Fatty acid synthase expression defines distinct molecular signatures in prostate cancer. Mol Cancer Res 1:707–715. Safran M, Kaelin WG, Jr (2003) HIF hydroxylation and the mammalian oxygen-sensing pathway. J Clin Invest 111:779–783. Sanchez-Martinez C, Aragon JJ (1997) Analysis of phosphofructokinase subunits and isozymes in ascites tumor cells and its original tissue, murine mammary gland. FEBS Lett 409: 86–90.

5  Glycolytic Pathway as a Target for Tumor Inhibition

117

Schornack PA, Gillies RJ (2003) Contributions of cell metabolism and H+ diffusion to the acidic pH of tumors. Neoplasia (New York, NY) 5:135–145. Schwickert G, Walenta S, Sundfor K, Rofstad EK, Mueller-Klieser W (1995) Correlation of high lactate levels in human cervical cancer with incidence of metastasis. Cancer Res 55: 4757–4759. Scott R, Crooks R, Meldrum C (2008) Gene symbol: STK11. Disease: Peutz-Jeghers syndrome. Hum Genet 124:300. Seemann MD (2004) PET/CT: fundamental principles. Eur J Med Res 9:241–246. Selak MA, Armour SM, MacKenzie ED, Boulahbel H, Watson DG, Mansfield KD, Pan Y, Simon MC, Thompson CB, Gottlieb E (2005) Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-alpha prolyl hydroxylase. Cancer Cell 7:77–85. Semenza GL (2003) Targeting HIF-1 for cancer therapy. Nat Rev Cancer 3:721–732. Semenza GL (2007) HIF-1 mediates the Warburg effect in clear cell renal carcinoma. J Bioenerg Biomembr 39:231–234. Sharp FR, Bernaudin M (2004) HIF1 and oxygen sensing in the brain. Nat Rev Neurosci 5: 437–448. Shaw RJ (2006) Glucose metabolism and cancer. Curr Opin Cell Biol 18:598–608. Shaw RJ, Bardeesy N, Manning BD, Lopez L, Kosmatka M, DePinho RA, Cantley LC (2004a) The LKB1 tumor suppressor negatively regulates mTOR signaling. Cancer Cell 6:91–99. Shaw RJ, Kosmatka M, Bardeesy N, Hurley RL, Witters LA, DePinho RA, Cantley LC (2004b) The tumor suppressor LKB1 kinase directly activates AMP-activated kinase and regulates apoptosis in response to energy stress. Proc Natl Acad Sci USA 101:3329–3335. Shi Q, Le X, Wang B, Abbruzzese JL, Xiong Q, He Y, Xie K (2001) Regulation of vascular endothelial growth factor expression by acidosis in human cancer cells. Oncogene 20: 3751–3756. Shim H, Dolde C, Lewis BC, Wu CS, Dang G, Jungmann RA, Dalla-Favera R, Dang CV (1997) c-Myc transactivation of LDH-A: implications for tumor metabolism and growth. Proc Natl Acad Sci USA 94:6658–6663. Skulachev VP (1994) Chemiosmotic concept of the membrane bioenergetics: what is already clear and what is still waiting for elucidation? J Bioenerg Biomembr 26:589–598. Smith TA (2000) Mammalian hexokinases and their abnormal expression in cancer. Br J Biomed Sci 57:170–178. Smith TA, Sharma RI, Thompson AM, Paulin FE (2006) Tumor 18F-FDG incorporation is enhanced by attenuation of P53 function in breast cancer cells in  vitro. J Nucl Med 47: 1525–1530. Sonveaux P, Vegran F, Schroeder T, Wergin MC, Verrax J, Rabbani ZN, De Saedeleer CJ, Kennedy KM, Diepart C, Jordan BF et  al (2008) Targeting lactate-fueled respiration selectively kills hypoxic tumor cells in mice. J Clin Investig 118:3930–3942. Stacpoole PW, Greene YJ (1992) Dichloroacetate. Diabetes Care 15:785–791. Suter M, Riek U, Tuerk R, Schlattner U, Wallimann T, Neumann D (2006) Dissecting the role of 5’-AMP for allosteric stimulation, activation, and deactivation of AMP-activated protein kinase. J Biol Chem 281:32207–32216. Swinnen JV, Brusselmans K, Verhoeven G (2006) Increased lipogenesis in cancer cells: new ­players, novel targets. Curr Opin Clin Nutr Metab Care 9:358–365. Swinnen JV, Vanderhoydonc F, Elgamal AA, Eelen M, Vercaeren I, Joniau S, Van Poppel H, Baert L, Goossens K, Heyns W et al (2000) Selective activation of the fatty acid synthesis pathway in human prostate cancer. Int J Cancer 88:176–179. Thierbach R, Schulz TJ, Isken F, Voigt A, Mietzner B, Drewes G, von Kleist-Retzow JC, Wiesner RJ, Magnuson MA, Puccio H et al (2005) Targeted disruption of hepatic frataxin expression causes impaired mitochondrial function, decreased life span and tumor growth in mice. Hum Mol Genet 14:3857–3864. Thomas DD, Miranda KM, Colton CA, Citrin D, Espey MG, Wink DA (2003) Heme proteins and nitric oxide (NO): the neglected, eloquent chemistry in NO redox signaling and regulation. Antioxid Redox Signal 5:307–317.

118

W. Lu and P. Huang

Tian WN, Braunstein LD, Apse K, Pang J, Rose M, Tian X, Stanton RC (1999) Importance of glucose-6-phosphate dehydrogenase activity in cell death. Am J Physiol 276:C1121–C1131. Tomlinson IP, Alam NA, Rowan AJ, Barclay E, Jaeger EE, Kelsell D, Leigh I, Gorman P, Lamlum H, Rahman S et al (2002) Germline mutations in FH predispose to dominantly inherited uterine fibroids, skin leiomyomata and papillary renal cell cancer. Nat Genet 30:406–410. Tong X, Zhao F, Thompson CB (2009) The molecular determinants of de novo nucleotide biosynthesis in cancer cells. Curr Opin Genet Dev 19:32–37. Vander Heiden MG, Cantley LC, Thompson CB (2009) Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science (New York, NY) 324:1029–1033. Varshney R, Adhikari JS, Dwarakanath BS (2003) Contribution of oxidative stress to radiosensitization by a combination of 2-DG and 6-AN in human cancer cell line. Ind J Exp Biol 41:1384–1391. Verma M, Kagan J, Sidransky D, Srivastava S (2003) Proteomic analysis of cancer-cell mitochondria. Nat Rev 3:789–795. Vogelstein B, Lane D, Levine AJ (2000) Surfing the p53 network. Nature 408:307–310. Wallace DC (1999) Mitochondrial diseases in man and mouse. Science (New York, NY) 283:1482–1488. Wallace DC (2005) Mitochondria and cancer: Warburg addressed. Cold Spring Harb Symp Quant Biol 70:363–374. Warburg O (1956a) On respiratory impairment in cancer cells. Science 124:269–270. Warburg O (1956b) On the origin of cancer cells. Science (New York, NY) 123:309–314. Weber WA (2006) Positron emission tomography as an imaging biomarker. J Clin Oncol 24:3282–3292. Woods A, Johnstone SR, Dickerson K, Leiper FC, Fryer LG, Neumann D, Schlattner U, Wallimann T, Carlson M, Carling D (2003) LKB1 is the upstream kinase in the AMP-activated protein kinase cascade. Curr Biol 13:2004–2008. Xie H, Valera VA, Merino MJ, Amato AM, Signoretti S, Linehan WM, Sukhatme VP, Seth P (2009) LDH-A inhibition, a therapeutic strategy for treatment of hereditary leiomyomatosis and renal cell cancer. Mol Cancer Ther 8:626–635. Xu RH, Pelicano H, Zhang H, Giles FJ, Keating MJ, Huang P (2005a) Synergistic effect of targeting mTOR by rapamycin and depleting ATP by inhibition of glycolysis in lymphoma and leukemia cells. Leukemia 19:2153–2158. Xu RH, Pelicano H, Zhou Y, Carew JS, Feng L, Bhalla KN, Keating MJ, Huang P (2005b) Inhibition of glycolysis in cancer cells: a novel strategy to overcome drug resistance associated with mitochondrial respiratory defect and hypoxia. Cancer Res 65:613–621. Yahagi N, Shimano H, Hasegawa K, Ohashi K, Matsuzaka T, Najima Y, Sekiya M, Tomita S, Okazaki,H, Tamura,Y et al (2005) Co-ordinate activation of lipogenic enzymes in hepatocellular carcinoma. Eur J Cancer 41:1316–1322. Yu Y, Deck JA, Hunsaker LA, Deck LM, Royer RE, Goldberg E, Vander Jagt DL (2001) Selective active site inhibitors of human lactate dehydrogenases A4, B4, and C4. Biochem Pharmacol 62:81–89. Zastawny TH, Dabrowska M, Jaskolski T, Klimarczyk M, Kulinski L, Koszela A, Szczesniewicz M, Sliwinska M, Witkowski P, Olinski R (1998) Comparison of oxidative base damage in mitochondrial and nuclear DNA. Free Radical Biol Med 24:722–725. Zhang S, Yang JH, Guo CK, Cai PC (2007) Gene silencing of TKTL1 by RNAi inhibits cell proliferation in human hepatoma cells. Cancer Lett 253:108–114.

Part II

Malignant Cells

Chapter 6

Aberrant DNA Methylation in Cancer Cells Toshikazu Ushijima

Abstract  DNA methylation is inherited upon somatic cell division, and methylation of a promoter CpG island silences its downstream gene. Methylation alterations in cancer cells are characterized by aberrant methylation of specific promoter CpG islands, which can cause inactivation of both tumor-suppressor genes (driver) and other genes (passenger), and by global hypomethylation. Recent studies revealed that aberrant methylation is present even in non-cancerous tissues, and its level is associated with cancer risk (an epigenetic field for cancerization). Quantification of methylation revealed that aberrant methylation can be induced much more frequently than mutations, and it has already been indicated that methylation alterations are involved in epithelial-mesenchymal transition. The high frequency of methylation alterations also suggested that they could be involved in phenotypic changes of stromal cells, and thus formation of cancer microenvironment. DNA methylation alterations are likely to be important players not only in transformation of epithelial cells but also in the formation of cancer microenvironment by stromal cells.

Introduction Epigenetic modifications are responsible for long-term cellular memory of gene expression, and play critical roles in development, differentiation, and reprogramming (Bird 2007). To play these roles, epigenetic modifications must be faithfully replicated upon DNA replication. DNA methylation and histone modifications are known to have such characteristics, and DNA methylation is especially known for its high fidelity in replication (Ushijima et al. 2003; Laird et al. 2004; Riggs and Xiong 2004). It is generally considered that, during developmental pro­cesses, epigenetic modifications of the genome (epigenome) are established as a T. Ushijima (*) National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo, Japan e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_6, © Springer Science+Business Media, LLC 2010

121

122

T. Ushijima

c­ onsequence of interactions between cellular lineage and environmental input (Gan et al. 2007). It was previously believed that aberrant DNA methylation occurs rarely, but that, once induced in critical genes, it can be causally involved in development of diseases, such as cancers. However, recent studies indicate that DNA methylation alterations of specific genes can be induced relatively frequently under specific conditions. Epigenetic changes were shown to be physiologically involved in determination of lymphocyte differentiation (Kioussis and ­Geor­gopoulos 2007). When gastric epithelial cells are exposed to a specific pathological condition, several to several 10% of them can have aberrant methylation of  specific genes (Maekita et al. 2006). These widespread occurrences of epigenetic changes in somatic cells indicate a possibility that they can be responsible for formation of tumor microenvironments, although cells involved in it are polyclonal. This chapter will introduce the basic characteristics of DNA methylation and methylation alterations in cancer cells, and finally discuss possible involvement of altered methylation in formation of tumor microenvironments.

Characteristics of DNA Methylation DNA is physiologically methylated at the 5 position of cytosines at some CpG sites (Fig. 6.1a). This physiological methylation is different from DNA methylations at O6 and N7 positions of guanines, which are abnormal adducts produced by alkylating agents. DNA methylation is characterized by its stable main­ tenance upon somatic cell division, and critical roles in regulation of gene transcription.

Maintenance of DNA Methylation Statuses When a CpG site is methylated, cytosines on both strands are methylated (Fig. 6.1b). At DNA replication, a newly synthesized DNA strand does not contain methyl groups, and hemi-methylated CpG sites are temporarily formed. However, a maintenance methylase, DNA methyltransferase 1 (DNMT1), associated with a replication fork (Hermann et al. 2004b), restores those hemi-methylated CpG sites into fully methylated CpG sites. DNMT1 has much lower activity on unmethylated CpG sites, and unmethylated CpG sites are kept unmethylated. Therefore, DNA methylation patterns are replicated at somatic DNA replication with a high fidelity (~99.9%), especially in CpG islands (Ushijima et al. 2003; Laird et al. 2004; Riggs and Xiong 2004).

6  Aberrant DNA Methylation in Cancer Cells

a

123

NH2 N

O

3

NH2

1 N

N

O

dR

dR

Cytosine

b 5' 3'

CH3

N

5

5-methylcytosine

M CG

M CG

GC M

GC M

CG

CG

3'

GC

GC

5'

DNA replication

5'

M CG

M CG

CG

CG

3'

3'

GC

GC

GC

GC

5'

5'

CG

CG

CG

CG

3'

3'

GC M

GC M

GC

GC

5'

DNMT1

5' 3'

5' 3'

M CG

M CG

GC M M CG GC M

CG

CG

3'

GC M M CG

GC

GC

5'

CG

CG

3'

GC M

GC

GC

5'

Fig. 6.1  DNA methylation and its maintenance upon DNA replication. (a) Chemical structures of cytosine and 5-methylcytosine. (b) Maintenance of methylated and unmethylated statuses of CpG sites. Upon DNA replication, hemimethylated CpG sites are temporarily formed, but DNMT1 at replication fork restores hemimethylated CpG sites specifically into fully methylated CpG sites

124

T. Ushijima

Regulation of Gene Transcription by DNA Methylation DNA methylation of a CpG island in a gene promoter region has been known to be consistently associated with transcriptional repression of its downstream gene (Baylin and Ohm 2006; Ushijima 2005). Also, in recent genome-wide analysis of DNA methylation and gene expression, the consistent association was clearly observed (Weber et al. 2007; Rauch et al. 2009; Yamashita et al. in press). In contrast, methylation of CpG islands in gene body regions is often, but not consistently, associated with increased gene transcription (Hellman and Chess 2007; Rauch et al. 2009; Yamashita et al. in press). The consistent association between promoter methylation and transcriptional repression is now known to be due to the presence of a cause–consequence relationship, mediated by several mechanisms. First, an approximately 200-bp region just upstream of a transcription start site usually lacks a nucleosome, designated as a nucleosome-free region (NFR) (Fig. 6.2a) (Lee et al. 2004; Li et al. 2007; Ozsolak et al. 2007). However, if the NFR is methylated, a nucleosome is formed, and transcription from the region is severely repressed (Fig. 6.2b) (Lin et al. 2007). Second, DNA methylation is recognized by methylated DNA binding proteins, such as MeCP2 and MBDs, and these proteins recruit histone deacetylases (Richards and Elgin 2002) and a histone methyltransferase, SUV39H1 (Fujita et  al. 2003). Deacetylated histones are known to be positively charged and to associate tightly with DNA, inhibiting accession of transcription complexes to DNA. SUV39H1 is a

b

Unmethylated CpG site Methylated CpG site

Fig. 6.2  Transcriptional repression by DNA methylation of a CpG island in a promoter region. (a) A NFR is present when DNA is kept unmethylated. (b) Nucleosome is formed when DNA in the NFR is methylated, and transcription is strongly repressed

6  Aberrant DNA Methylation in Cancer Cells

125

known to be involved in formation of a heterochromatin structure (Stewart et  al. 2005). Thirdly, even in naked DNA, some transcription factors, such as CTCF, are known to have much less binding capacity with methylated CpG sites (Bird 2002).

Maintenance and De Novo DNA Methylases DNA methyltransferases are essential machineries to establish and maintain DNA methylation, and individual DNA methyltransferases are known as maintenance methylase, DNMT1, and de novo methylases, DNMT3A, and DNMT3B, respectively (Hermann et al. 2004a). As mentioned above, DNMT1 plays a major role in maintaining DNA methylation upon DNA replication. Homozygous knockout of Dnmt1 is lethal in midgestation (Li et  al. 1992). Dnmt3a and Dnmt3b have the activity of methylating CpG sites without any preference for hemimethylated CpG sites (Okano et  al. 1998). While Dnmt3a cannot methylate nucleosomal DNA, Dnmt3b can (Takeshima et  al. 2006). Homozygous knockout of Dnmt3a causes lethality after birth (Okano et al. 1999), and Dnmt3a is essential in establishment of genomic imprinting (Kaneda et al. 2004b). Homozygous knockout of Dnmt3b causes lethality before birth, and germline mutations of DNMT3B cause a recessive inherited disorder, ICF syndrome, in humans (Okano et al. 1999).

Methylation Alterations in Cancer Cells DNA methylation patterns of the genome (methylome) in normal cells are precisely regulated according to developmental stages (Meissner et  al. 2008; Rauch et  al. 2009). In brief, the vast majority of CpG islands are kept unmethylated, and repetitive sequences, which consist of more than 40% of the genome (Lander et al. 2001), are heavily methylated. In cancer cells, the presence of altered DNA methylation has been long known, and the alteration is characterized by “genome-overall hypomethylation and regional hypermethylation” (Fig. 6.3).

Genome-Overall Hypomethylation The “genome-overall” hypomethylation was discovered in the early 1980s, and the phenomenon is now observed in almost all types of cancer cells (Feinberg and Tycko 2004). The decrease of 5-methylcytosine content in the genome is mainly due to hypomethylation of repetitive sequences (Feinberg and Tycko 2004; Kaneda et al. 2004a), but can involve demethylation of normally methylated CpG islands. Demethylation of normally methylated promoter CpG islands leads to aberrant transcription of their downstream genes, and a well known example is aberrant

126

T. Ushijima Normal cell Cancer cell Unmethylated Methylated Repetitive sequence CpG island

Fig. 6.3  DNA methylation changes in cancer cells. In normal cells, the vast majority of CpG islands are kept unmethylated, and repetitive sequences are kept methylated. In cancer cells, a limited number of CpG islands are aberrantly methylated, and some repetitive sequences are demethylated

expression of cancer-testis antigen genes, such as melanoma antigen genes (MAGEs) (de Smet et al. 1999). Also, hypomethylation of a differentially methylated region (DMR) of IGF2 is known for its loss of imprinting, leading to its increased expression and tumor development (Cui et al. 2002). Despite the long history, there are still limited findings on the causal role of genome-overall hypomethylation in carcinogenesis. A mouse strain with genomeoverall hypomethylation, due to introduction of a hypomorphic allele of Dnmt1, demonstrated increased rates of chromosomal loss (Chen et al. 1998). The mouse strain showed increased incidences of lymphomas, colonic microadenomas and liver tumors (Chen et al. 1998; Eden et al. 2003; Yamada et al. 2005). At the same time, the systemic low activity of Dnmt1 led to suppression of macroscopic tumors of the intestine (Laird et al. 1995; Yamada et al. 2005).

Aberrant DNA Methylation of CpG Islands “Regional hypermethylation” denotes methylation of CpG islands that are normally unmethylated. If such methylation is induced in promoter CpG islands of tumor-suppressor genes, it leads to their permanent inactivation, and is causally involved in cancer development and progression (Baylin and Ohm 2006; Jones and Baylin 2007). After the RB tumor-suppressor gene was shown to be inactivated by its promoter methylation (Ohtani-Fujita et  al. 1993), many tumorsuppressor genes involved in various cellular processes, such as WNT signalling (SFRP family and CDH1), DNA repair (MLH1 and MGMT), and cell cycle regulation (CDKN2A), were shown to be inactivated by promoter methylation (Baylin and Ohm 2006). In some cancer types, such as gastric cancers, tumor-suppressor genes are inactivated more frequently by promoter methylation than by mutations (Ushijima and Sasako 2004). CpG islands aberrantly methylated in cancers are not limited to CpG islands in promoter regions, and methylation of CpG islands in gene bodies are often associated with increased gene expression (Ushijima 2005; Rauch et al. 2009; Yamashita

6  Aberrant DNA Methylation in Cancer Cells

127

et al. in press). However, the biological meaning of physiological and pathological DNA methylation of gene bodies remains unclear.

Driver Methylation and Passenger Methylation Recent genome-wide screening revealed that several hundred to one thousand promoter CpG islands are methylated in cancer cells (Rauch et  al. 2009; Yamashita et al. in press). Most of the genes methylated in cancers are not expressed or have only low expression in normal counterpart cells (Takeshima et  al. 2009). These clearly show that promoter CpG islands aberrantly methylated in cancers are not always causally involved in cancer development or progression. According to tradition in the field of mutations, methylation causally involved in cancer development and progression is designated as “driver methylation”, and methylation that simply accompanies the process is designated as “passenger methylation”. Although the distinction between driver and passenger is conceptually clear, actual distinctions between the two types are often difficult, especially when methylation is present in a promoter CpG island of a gene. Although most genes with low or no transcription in normal cells have limited functions, tumor suppressor genes can have only low transcription levels in normal cells when not induced. Although an important tumor-suppressor gene is methylated with a high incidence in specific types of cancers, we cannot conclude that genes with methylation at low incidences are not important. At least, to demonstrate that a gene with promoter methylation in cancers is a driver, its expression in normal counterpart cells and gene function at physiological expression levels should be examined. If a gene is inactivated by mutations, the gene is likely to be a tumor-suppressor gene.

Possible Involvement of Altered Methylation in Tumor Microenvironments Traditionally, aberrant DNA methylation was considered to be similar to mutations in terms of its frequency, target genes, and inducers. However, it was recently indicated that DNA methylation of specific genes can be induced in a significant fraction of cells even in polyclonal tissue, and there is a possibility that altered methylation statuses can be involved in formation of tumor microenvironments.

Unique Natures of Aberrant DNA Methylation, in Contrast with Mutations Recent findings showed unique natures of DNA methylation in contrast with mutations. First, as described above, aberrant methylation of promoter CpG islands, which inactivate their downstream genes, is induced in several hundred to

128

T. Ushijima

one thousand genes in cancer cells (Rauch et al. 2009; Yamashita et al. in press) while a cancer cell is estimated to harbor mutations of 90 genes (Sjoblom et al. 2006). Second, aberrant methylation of a specific gene can be present in several 10% of cells in non-cancerous tissues (Maekita et al. 2006; Ushijima 2007) while mutations of a specific gene are quite infrequent, being less than 1 out of 105 cells (Nagao et  al. 2001). Third, aberrant methylation is induced in specific genes (Nakajima et al. 2009; Oka et al. 2009) while mutations are considered to occur in random genes. Fourth, chronic inflammation is an important inducer of aberrant methylation, while mutagenic chemicals and radiations induce mutations.

Field for Cancerization and DNA Methylation The presence of aberrant DNA methylation in non-cancerous tissues is important from the viewpoint of the field cancerization. It has long been known that some cancers tend to have multiple independent occurrences, suggesting that there is a field for cancerization produced by exposure to environmental and other carcinogenic stimuli (Braakhuis et  al. 2003). It was recently clarified that the degree of accumulation of aberrant methylation in gastric mucosae correlates with risk of developing gastric cancers (Maekita et  al. 2006; Nakajima et  al. 2006). Similar associations between methylation in non-cancerous tissues and cancer development were obtained in various cancers, such as liver, colorectal, breast, and urothelial cancers (Kondo et al. 2000; Shen et al. 2005; Yan et al. 2006; Arai et al. 2006), and such accumulation is considered to represent an “epigenetic field for cancerization” (epigenetic field defect) (Fig. 6.4) (Ushijima 2007). The epigenetic field for cancerization can be utilized for cancer risk diagnosis that takes account of an individual’s own life history.

Epithelial-Mesenchymal Transition and DNA Methylation Epithelial-mesenchymal transition (EMT) refers to a phenomenon in which epithelial cells lose their epithelial characteristics and acquire mesenchymal features (Polyak and Weinberg 2009). EMT has been known to be critically involved in embryonic development and tumor progression (Polyak and Weinberg 2009; Dumont et al. 2008). In addition, it was recently demonstrated that epithelial cells acquire stem cell features when they undergo EMT, and that stem-like cells in normal epithelium and cancer tissues have EMT markers (Mani et al. 2008). Physiological EMT, which takes place in a group of cells and cannot be attributed to a mutation in a cell, is likely to be due to epigenetic changes (Polyak and Weinberg 2009). Also, EMT involved in tumor progression, which can be due to a mutation or an epigenetic alteration, was recently shown to be mediated by DNA methylation (Dumont et al. 2008; Yang et al. 2009).

6  Aberrant DNA Methylation in Cancer Cells

Noncancerous predisposed tissue

Normal tissue 1

cells

2 3

129

Cancer development

Exposure to carcinogenic factors

4 5 6

Methylated CpG island Unmethylated CpG island

Fig. 6.4  Epigenetic field for cancerization. In a normal tissue, the vast majority of CpG islands are unmethylated. When it is exposed to some carcinogenic stimulus, such as Helicobacter pylori infection, various but specific genes, most of which are passengers, are methylated in various cells. If further genetic or epigenetic alterations take place, clinical cancer will develop. The amount of aberrant DNA methylation accumulated in a tissue can be used as a cancer risk marker

Tumor Microenvironments and DNA Methylation Not only is aberrant DNA methylation present in non-cancerous tissues, but also the degree can be severe, reaching several 10%. If a specific gene is inactivated in a significant fraction of cells in a tissue, it can affect the overall function of the tissue. Functional change of a tissue due to the presence of somatic mutations has been considered very unlikely because the population of cells with mutation of a specific gene is very small. This unique nature of aberrant DNA methylation suggests that it can be involved in pathological conditions without any clonal expansion, including formation of cancer microenvironments and any diseases that involve persistent gene repression. In cancer stromal cells, persistent changes in gene expression have been already reported (Finak et al. 2008; Hu et al. 2005; Lin et al. 2008), and involvement of mutations in such changes was shown to be unlikely (Qiu et al. 2008). Moreover, epigenetic silencing of specific genes due to changes in histone modification has been reported in tumor endothelial cells (Hellebrekers et  al. 2007). Mesenchymal cells produced by EMT of tumor cells may contribute to tumor

130

T. Ushijima

microenvironments. Clarification of DNA methylation changes in cancer stromal cells is becoming an important issue.

Epilogue Unique natures of DNA methylation, in sharp contrast with mutations, indicate that altered DNA methylation is involved in formation of tumor microenvironments. Although I did not describe technological aspects of methylation analysis in this chapter, all the technologies for genome-wide screening are not capable of identifying methylation changes present only in a fraction of cells. Therefore, innovation in technology or approach seems to be necessary to identify critical DNA methylation alterations in cancer stromal cells. Even without such an innovation, sensitive analysis of candidate genes might greatly contribute to clarification of genes critically involved in tumor microenvironments. Acknowledgments  The author is grateful to Dr. Eriko Okochi for her critical reading of the manuscript. The author’s research described here was supported by a Grant-in-Aid for the 3rd-term Comprehensive Cancer Control Strategy from Ministry of Health, Labor, and Welfare, Japan.

References Arai E, Kanai Y, Ushijima S et  al (2006) Regional DNA hypermethylation and DNA methyl­ transferase (DNMT) 1 protein overexpression in both renal tumors and corresponding nontumorous renal tissues. Int J Cancer 119:288–296 Baylin SB, Ohm JE (2006) Epigenetic gene silencing in cancer – a mechanism for early oncogenic pathway addiction? Nat Rev Cancer 6:107–116 Bird A (2002) DNA methylation patterns and epigenetic memory. Genes Dev 16:6–21 Bird A (2007) Perceptions of epigenetics. Nature 447:396–398 Braakhuis BJ, Tabor MP, Kummer JA et al (2003) A genetic explanation of Slaughter’s concept of field cancerization: evidence and clinical implications. Cancer Res 63:1727–1730 Chen RZ, Pettersson U, Beard C et al (1998) DNA hypomethylation leads to elevated mutation rates. Nature 395:89–93 Cui H, Onyango P, Brandenburg S et al (2002) Loss of imprinting in colorectal cancer linked to hypomethylation of H19 and IGF2. Cancer Res 62:6442–6446 de Smet C, Lurquin C, Lethe B et al (1999) DNA methylation is the primary silencing mechanism for a set of germ line- and tumor-specific genes with a CpG-rich promoter. Mol Cell Biol 19: 7327–7335 Dumont N, Wilson MB, Crawford YG et al (2008) Sustained induction of epithelial to mesenchymal transition activates DNA methylation of genes silenced in basal-like breast cancers. Proc Natl Acad Sci USA 105:14867–14872 Eden A, Gaudet F, Waghmare A et al (2003) Chromosomal instability and tumors promoted by DNA hypomethylation. Science 300:455 Feinberg AP, Tycko B (2004) The history of cancer epigenetics. Nat Rev Cancer 4:143–153 Finak G, Bertos N, Pepin F et  al (2008) Stromal gene expression predicts clinical outcome in breast cancer. Nat Med 14:518–527

6  Aberrant DNA Methylation in Cancer Cells

131

Fujita N, Watanabe S, Ichimura T et al (2003) Methyl-CpG binding domain 1 (MBD1) interacts with the Suv39h1-HP1 heterochromatic complex for DNA methylation-based transcriptional repression. J Biol Chem 278:24132–24138 Gan Q, Yoshida T, McDonald OG et al (2007) Concise review: epigenetic mechanisms contribute to pluripotency and cell lineage determination of embryonic stem cells. Stem Cells 25:2–9 Hellebrekers DM, Melotte V, Vire E et al (2007) Identification of epigenetically silenced genes in tumor endothelial cells. Cancer Res 67:4138–4148 Hellman A, Chess A (2007) Gene body-specific methylation on the active X chromosome. Science 315:1141–1143 Hermann A, Gowher H, Jeltsch A (2004a) Biochemistry and biology of mammalian DNA methyltransferases. Cell Mol Life Sci 61:2571–2587 Hermann A, Goyal R, Jeltsch A (2004b) The Dnmt1 DNA-(cytosine-C5)-methyltransferase methylates DNA processively with high preference for hemimethylated target sites. J Biol Chem 279:48350–48359 Hu M, Yao J, Cai L et al (2005) Distinct epigenetic changes in the stromal cells of breast cancers. Nat Genet 37:899–905 Jones PA, Baylin SB (2007) The epigenomics of cancer. Cell 128:683–692 Kaneda A, Tsukamoto T, Takamura-Enya T et al (2004a) Frequent hypomethylation in multiple promoter CpG islands is associated with global hypomethylation, but not with frequent promoter hypermethylation. Cancer Sci 95:58–64 Kaneda M, Okano M, Hata K et  al (2004b) Essential role for de  novo DNA methyltransferase Dnmt3a in paternal and maternal imprinting. Nature 429:900–903 Kioussis D, Georgopoulos K (2007) Epigenetic flexibility underlying lineage choices in the adaptive immune system. Science 317:620–622 Kondo Y, Kanai Y, Sakamoto M et al (2000) Genetic instability and aberrant DNA methylation in chronic hepatitis and cirrhosis – a comprehensive study of loss of heterozygosity and microsatellite instability at 39 loci and DNA hypermethylation on 8 CpG islands in microdissected specimens from patients with hepatocellular carcinoma. Hepatology 32:970–979 Laird PW, Jackson-Grusby L, Fazeli A et al (1995) Suppression of intestinal neoplasia by DNA hypomethylation. Cell 81:197–205 Laird CD, Pleasant ND, Clark AD et al (2004) Hairpin-bisulfite PCR: assessing epigenetic methylation patterns on complementary strands of individual DNA molecules. Proc Natl Acad Sci USA 101:204–209 Lander ES, Linton LM, Birren B et  al (2001) Initial sequencing and analysis of the human genome. Nature 409:860–921 Lee CK, Shibata Y, Rao B et al (2004) Evidence for nucleosome depletion at active regulatory regions genome-wide. Nat Genet 36:900–905 Li E, Bestor TH, Jaenisch R (1992) Targeted mutation of the DNA methyltransferase gene results in embryonic lethality. Cell 69:915–926 Li B, Carey M, Workman JL (2007) The role of chromatin during transcription. Cell 128: 707–719 Lin JC, Jeong S, Liang G et al (2007) Role of nucleosomal occupancy in the epigenetic silencing of the MLH1 CpG island. Cancer Cell 12:432–444 Lin HJ, Zuo T, Lin CH et al (2008) Breast cancer-associated fibroblasts confer AKT1-mediated epigenetic silencing of Cystatin M in epithelial cells. Cancer Res 68:10257–10266 Maekita T, Nakazawa K, Mihara M et  al (2006) High levels of aberrant DNA methylation in Helicobacter pylori-infected gastric mucosae and its possible association with gastric cancer risk. Clin Cancer Res 12:989–995 Mani SA, Guo W, Liao MJ et al (2008) The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell 133:704–715 Meissner A, Mikkelsen TS, Gu H et al (2008) Genome-scale DNA methylation maps of pluripotent and differentiated cells. Nature 454:766–770 Nagao M, Ochiai M, Okochi E et al (2001) LacI transgenic animal study: relationships among DNA-adduct levels, mutant frequencies and cancer incidences. Mutat Res 477:119–124

132

T. Ushijima

Nakajima T, Maekita T, Oda I et al (2006) Higher methylation levels in gastric mucosae significantly correlate with higher risk of gastric cancers. Cancer Epidemiol Biomarkers Prev 15:2317–2321 Nakajima T, Yamashita S, Maekita T et al (2009) The presence of a methylation fingerprint of Helicobacter pylori infection in human gastric mucosae. Int J Cancer 124:905–910 Ohtani-Fujita N, Fujita T, Aoike A et al (1993) CpG methylation inactivates the promoter activity of the human retinoblastoma tumor-suppressor gene. Oncogene 8:1063–1067 Oka D, Yamashita S, Tomioka T et al (2009) The presence of aberrant DNA methylation in noncancerous esophageal mucosae in association with smoking history: a target for risk diagnosis and prevention of esophageal cancers. Cancer 115:3412–3426 Okano M, Xie S, Li E (1998) Cloning and characterization of a family of novel mammalian DNA (cytosine-5) methyltransferases. Nat Genet 19:219–220 Okano M, Bell DW, Haber DA et  al (1999) DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 99:247–257 Ozsolak F, Song JS, Liu XS et al (2007) High-throughput mapping of the chromatin structure of human promoters. Nat Biotechnol 25:244–248 Polyak K, Weinberg RA (2009) Transitions between epithelial and mesenchymal states: acquisition of malignant and stem cell traits. Nat Rev Cancer 9:265–273 Qiu W, Hu M, Sridhar A et al (2008) No evidence of clonal somatic genetic alterations in cancerassociated fibroblasts from human breast and ovarian carcinomas. Nat Genet 40:650–655 Rauch TA, Wu X, Zhong X et al (2009) A human B cell methylome at 100-base pair resolution. Proc Natl Acad Sci USA 106:671–678 Richards EJ, Elgin SC (2002) Epigenetic codes for heterochromatin formation and silencing: rounding up the usual suspects. Cell 108:489–500 Riggs AD, Xiong Z (2004) Methylation and epigenetic fidelity. Proc Natl Acad Sci USA 101: 4–5 Shen L, Kondo Y, Rosner GL et al (2005) MGMT promoter methylation and field defect in sporadic colorectal cancer. J Natl Cancer Inst 97:1330–1338 Sjoblom T, Jones S, Wood LD et al (2006) The consensus coding sequences of human breast and colorectal cancers. Science 314:268–274 Stewart MD, Li J, Wong J (2005) Relationship between histone H3 lysine 9 methylation, transcription repression, and heterochromatin protein 1 recruitment. Mol Cell Biol 25:2525–2538 Takeshima H, Suetake I, Shimahara H et al (2006) Distinct DNA methylation activity of Dnmt3a and Dnmt3b towards naked and nucleosomal DNA. J Biochem 139:503–515 Takeshima H, Yamashita S, Shimazu T et al (2009) The presence of RNA polymerase II, active or stalled, predicts epigenetic fate of promoter CpG islands. Genome Res 19:1974–1982 Ushijima T (2005) Detection and interpretation of altered methylation patterns in cancer cells. Nat Rev Cancer 5:223–231 Ushijima T (2007) Epigenetic field for cancerization. J Biochem Mol Biol 40:142–150 Ushijima T, Sasako M (2004) Focus on gastric cancer. Cancer Cell 5:121–125 Ushijima T, Watanabe N, Okochi E et al (2003) Fidelity of the methylation pattern and its variation in the genome. Genome Res 13:868–874 Weber M, Hellmann I, Stadler MB et al (2007) Distribution, silencing potential and evolutionary impact of promoter DNA methylation in the human genome. Nat Genet 39:457–466 Yamada Y, Jackson-Grusby L, Linhart H et al (2005) Opposing effects of DNA hypomethylation on intestinal and liver carcinogenesis. Proc Natl Acad Sci USA 102:13580–13585 Yamashita S, Hosoya K, Gyobu K et al (2009) Development of a novel output value for quantitative assessment in methylated DNA immunoprecipitation-CpG island microarray analysis. DNA Research 16:275–286 Yan PS, Venkataramu C, Ibrahim A et al (2006) Mapping geographic zones of cancer risk with epigenetic biomarkers in normal breast tissue. Clin Cancer Res 12:6626–6636 Yang X, Pursell B, Lu S et al (2009) Regulation of beta 4-integrin expression by epigenetic modifications in the mammary gland and during the epithelial-to-mesenchymal transition. J Cell Sci 122:2473–2480

Chapter 7

DNA Repair and Redox Signaling Mark R. Kelley, Millie M. Georgiadis, and Melissa L. Fishel

Abstract  DNA repair pathways have long been recognized as important in protecting the genome from damage caused by endogenous and exogenous DNA damaging agents. However, the role of reduction–oxidation (redox) signaling as a regulatory mechanism in these pathways is a comparatively recent discovery. We have attempted to cover the general aspects of DNA repair, while focusing more attention on the interface between redox regulation and DNA repair activity and response in mammalian cells. The primary focus of this chapter is on APE1, the only DNA repair protein currently known to serve a dual role as a DNA repair enzyme and redox signaling factor. APE1 has been shown to regulate a number of downstream transcription factors (TF) such as HIF-1a, Egr-1, AP-1, NFkB, CREB, p53 and an increasing number of other TFs and other proteins. Additionally, this redox signaling function of APE1 indirectly alters the activity of other DNA repair pathways through its regulation of TF binding to DNA. Included in this chapter is also an overview of general redox systems. We also discuss the role small molecule inhibitors play in the modulation of APE1 redox activity and the potential for chemotherapeutic development via targeting redox regulation of DNA repair. Abbreviations 8-oxoG AAG AGT Ang II AP AP-1

8 oxoguanine alkyladenine DNA glycosylase O-6-alkylguanine-DNA methyltransferase angiotension II apurinic/apyrimidinic activator protein 1

M.R. Kelley (*) Department of Pediatrics, Section of Hematology/Oncology, Herman B Wells Center for Pediatric Research, Indiana University, Indianapolis, IN 46202, USA e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_7, © Springer Science+Business Media, LLC 2010

133

134

M.R. Kelley et al.

APE1 apurinic/apyrmidinic endonuclease 1 ATF activating transcription factor ATFa activating transcription factor a ATF2 activating transcription factor 2 ATF3 activating transcription factor 3 ATF4 activating transcription factor 4 ATM ataxia telangiectasia mutated protein BER base excision repair BKca Large-conductance Ca2+-activated K+channels BLM Bloom’s syndrome gene product BRCA1 breast cancer 1, early onset BRCA2 breast cancer 2, early onset BSO buthionine sulfoximine CD40L CD40 ligand CHO Chinese hamster ovary cells CREB cyclic AMP response-element binding protein CSA Cockayne’s syndrome protein A CSB Cockayne’s syndrome protein B CTL cytotoxic T lymphocyte DDB1 DNA damage binding protein 1 DDB2 DNA damage binding protein 2 DNA-PKcs DNA-dependent kinase catalytic subunit DSBs double-strand breaks DR direct repair dRP deoxyribophosphate dRPase deoxyribophosphodiesterase E3330 2E-3-[5-(2, 3 dimethoxy-6-methyl-1, 4-benzoquinolyl)]-2-nonyl-2propenoic acid Egr-1 early growth response protein 1 ERCC1 excision repair cross-complementation group 1 ERCC3 excision repair cross-complementation group 3 ERCC5 excision repair cross-complementation group 5 flavin adenine dinucleotide dehydrogenase FADH2 FEN1 flap endonuclease 1 Flk-1/KDR fetal liver kinase 1/kinase insert domain receptor FosB FBJ murine osteosarcoma viral oncogene homolog B FPG formamidopyrimidine DNA glycosylase Fra-1 fos related antigen 1 Fra-2 fos related antigen 2 GAPDH glyceraldehyde-3-phosphate dehydrogenase GGR global genome repair GRX glutaredoxin GSH glutathione GSSG glutathione disulfide GTPase Guanosine-5’-triphosphate

7  DNA Repair and Redox Signaling

HEK293 HeLa HIF-1 HIF-1a HR HRE HR32B HUVES IL-2 LP-BER Lys6 Lys7 MeSeH MGMT MiTF MLH1 MMR MMS MPG MRN MSH2 MSH3 MSH6 MSI MYH NAC NADPH Nbs1 NEIL NER NFkB NHEJ NK NTH O6-meG OGG1 P2Y PARP1 PCEC PCNA PEBP-2 PMS2 PRX Pol b Pol d

Human embryonic kidney 293 cells Henrietta Lacks cells hypoxia-inducible factor-1 hypoxia-inducible factor-1 alpha homologous recombination hypoxia response elements human homologue of the yeast RAD23 protein human umbilical vein endothelial cells interleukin 2 long-patch base excision repair lysine 6 lysine 7 methylselenol O6-methylguanine-DNA methyltransferase Microphthalmia-associated transcription factor MutL homolog 1 mismatch repair methyl methane sulfonate N-methylpurine DNA glycosylase Mre11/Rad50/Nbs1 MutS homolog 2 MutS homolog 3 MutS homolog 6 microsatellite instability MutY homolog N-acetyl-L-cysteine Nicotinamide adenine dinucleotide phosphate Nijmegen breakage syndrome 1 fgp/nei family DNA glycosylase nucleotide excision repair nuclear factor kappa B non-homologous end joining natural killer homolog of E. coli endonuclease III (nth) O6 methyl guanine oxoguanine glycosylase 1 purinergic receptor poly ADP ribose polymerase 1 pancreatic cancer-associated endothelial cells proliferating cell nuclear antigen phosphatidylethanolamine-binding protein 2 postmeiotic segregation increased 2 peroxiredoxin DNA polymerase beta DNA polymerase delta

135

136

M.R. Kelley et al.

Pol e DNA polymerase epsilon PMS1 post-meiotic-segregation increased-1 protein PMS2 post-meiotic-segregation increased-2 protein PTEN Phosphatase and tensin homolog deleted on chromosome 10 Ref-1 redox effector factor 1 RF-C replication factor C rRNA ribosomal RNA RNA Pol II RNA polymerase II ROS reactive oxygen species RPA replication protein A SeMet selenomethionine SETMAR SET domain and mariner transposase fusion siRNA small interfering RNA SMUG1 mammalian 5-formyluracil DNA glycosylase SP-BER short-patch base excision repair ssDNA single-strand DNA TCR transcription-coupled repair TDG thymine-DNA glycosylase TFIIH transcription factor IIH TNF-a tumor necrosis factor alpha Trx thioredoxin TrxR thioredoxin reductase TSH thyroid stimulating hormone TTF-1 thyroid transcription factor 1 UNG uracil-DNA glycosylase UV-DDB UV-damaged DNA-binding protein WRN Werner protein, deficient in Werner’s syndrome XPB xeroderma pigmentosum complementary group B protein XPC xeroderma pigmentosum complementary group C protein XPD xeroderma pigmentosum complementary group D protein XPF1 xeroderma pigmentosum complementary group F protein XPG xeroderma pigmentosum group XRCC1 X-ray repair complementing defective repair in Chinese hamster cells 1 XRCC4 X-ray repair complementing defective repair in Chinese hamster cells 4

Introduction Genomic stability relies on efficient, effective methods of repairing DNA. Unrepaired damage can cause cell cycle arrest, apoptosis or accumulation of mutations (Eker et al. 2009; Hofmann et al. 2002). To protect DNA and keep the genome intact, a number of DNA repair pathways exist. Although much has been written about these pathways, the role of redox regulation in DNA repair is a relatively recent discovery. This chapter discusses the mechanisms for human DNA repair,

7  DNA Repair and Redox Signaling

137

redox signaling and its regulation of DNA repair transcription factors, the unique APE1 protein that participates in both categories of activities, how DNA repair differs in tumor cells and implications for emerging cancer therapies.

Overview of DNA Repair Pathways DNA repair is a never-ending task of all eukaryotic cells, as endogenous and exogenous agents constantly assault DNA. Damage from endogenous agents include oxidation by reactive oxygen species (ROS) from normal metabolic processes, alkylation by agents such as S-adenosylmethionine, adduct formation caused by reactive carbonyl species created during lipid peroxidation, hydrolytic depurination that forms abasic sites and deamination of bases, primarily cytidine (and, to a lesser extent, adenine) (De Bont and van Larebeke 2004). Environmental insults from exogenous agents include chemicals, carcinogens and ultraviolet light, as well as chemotherapeutic agents and ionizing radiation damage (Christmann et al. 2003; Fleck and Nielsen 2004). Failure to repair DNA damage can cause a host of problems that ultimately lead to genomic instability (Arcangeli et al. 1997; Kelley and Fishel 2008). For example, improper DNA damage response in mitotic cells can halt cell cycle progress. In post-mitotic cells, DNA damage may result in cell cycle activation without proper coordination of cell cycle machinery, leading to deleterious events, including potentially lethal DNA double-strand breaks (Kruman et al. 2004; Schwartz et al. 2007). However, eukaryotic cells possess a cadre of DNA repair pathways to correct the damage: direct repair (DR), base excision repair (BER), mismatch repair (MMR), nucleotide excision repair (NER), non-homologous end joining (NHEJ) and homologous recombination (HR) (Hoeijmakers 2001, 2001) (Fig. 7.1). The number of known DNA repair proteins and other factors involved in the cellular response to DNA damage keeps growing as we learn more details about each repair pathway and induction of regulatory networks when the damage persists (Wood et al. 2005). The type of DNA damage dictates which pathway is activated to perform repairs. However, some overlap exists among the pathways. Although they are not redundant, certain pathways can function as “backup systems” to others. In addition, some mechanisms overlap. (See Table 7.1 for examples.) These overlaps and interactions provide the most efficient and effective defense net for preserving the cell genome, whereas reduced repair capacity can lead to genomic modification.

DR The DR pathway, sometimes called direct reversal, repairs damaged bases by removing the alteration to the base – as opposed to removing the damaged base itself. This unique mechanism may be the most efficient of all repair pathways

138

M.R. Kelley et al.

Direct Repair

Base Excision

MGMT

Glycosylase

Mismatch

MSH2/6 MSH2/3

TCR

RNA PoLII, CSA, CSB, XPAB2

APE1

Long Patch

Nucleotide Excision

Short Patch

Non-homologous End Joining (NHEJ)

Homologous Recombination (HR)

GGR

DDB-XPE, XPC, HR23β

Ku70, Ku80

ATM

MLH1-PMS2 MLH1-PMS1 DNA-PK RPA, XPA, XPCTFIIH, XPB, XPC, XPD, XPG, XPFERCC1

EX01, RFC, PCNA, RPA, NA Polδ, Ligase

RFC, PCNA, Polδ/ε, Ligase I

MRN

MRN, Rad51, Rad52, Rad54, BRCA1, BRCA2, PCNA, Polδ/ε, Ligase I

DNA Polµ, XRCC4, Ligase 4, Artemis

Fig. 7.1  Schematic overview of DNA repair pathway. Several DNA repair pathways are involved in maintaining cell genomic stability; these include DR, BER, NER, MMR, HR, and NHEJ. More than 150 proteins are involved. Only selected genes of each pathway are shown here. (Adapted from: Fishel et al. (Fishel and Kelley 2007))

Table 7.1  Examples of overlap in DNA repair pathways Overlap Mechanism DR/MMR In the DR pathway, if O6-methylguanine-DNA methyltransferase (AGT) is unsuccessful in removing O6-methylguanine, the MMR pathway can recognize and fix O6-meG mispairs (Fishel and Wilson 1997) BER/NER The BER pathway is primarily responsible for repairing oxidative DNA damage, but NER can serve as a backup for repairing some minor damage (Farmer et al. 2005) BER/HR If BER does not repair single-strand DNA breaks (SSBs), they may lead to double-strand breaks (DSBs), which HR can repair HR/NHEJ HR also can repair DNA DSBs that the NHEJ pathway fails to process (Essers et al. 2000)

(Kaina et al. 2007). O6-alkylguanine-DNA methyltransferase, or AGT, removes an alkyl group from the O6position of guanine on the damaged base, the alkyl group is transferred to the active site of AGT and AGT is subsequently degraded. AGT performs this function to a lesser extent on the O4position of thymine. This is a stoichiometric reaction, as AGT can work only once before being inactivated. It is essential to repair O6-meG adducts, as they cause thymine mispairing errors during replication, leading to G:C-to-A:T transitions or a strand break.

7  DNA Repair and Redox Signaling

139

Clinical application: small-molecule inhibitors of AGT Indication for utility Increased AGT levels in tumors correlate with tumor resistance How it works

A chemotherapy agent generates lesions at the O 6 position of guanine, causing interstrand crosslink formation, which blocks DNA replication. An AGT inhibitor would allow crosslink formation by a chemotherapeutic agent by blocking AGT’s removal of the O6lesion NOTE: AGT was one of the first DNA repair targets identified for cancer therapeutics

Anti-AGT agents in use

• BG (O6-benzylguanine) • PARP1 inhibitors

Additional applications

A mutant AGT protein is in development to protect the bone marrow from myelosuppression while inhibiting tumors with anti-AGT therapies

BER The BER pathway is responsible for repairing DNA damage caused by oxidation, alkylation, deamination and ionizing radiation (Christmann et al. 2003; Altieri et al. 2008; Evans et  al. 2000). This pathway repairs single-base lesions including N-alkylated purines (N3-methyladenine, N7-methylguanine and N3-methylguanine), 8-oxo-7,8-dihydroguanine(8-OxoG), thymine glycols, 5-OH and 6-OH dihydrothymine, uracil glycol, 5-hydroxycytosine, urea residues and other adducts (Aiyer et al. 2008; Bindra and Glazer 2007; Chen and Olkowski 1994). The process starts when a DNA glycosylase recognizes and excises the damaged base, leaving an abasic site. The type of DNA lesion dictates which glycosylase performs this activity. For example, methylpurine DNA glycosylase (MPG, AAG), 8-oxoguanine DNA glycosylase (OGG1), uracil DNA glycosylase (UNG), thymine DNA glycosylase (TDG) and others remove specific lesions as their names suggest (Essers et al. 2000). In addition to acting on a specific target, glycosylases may be monofunctional or bifunctional. Monofunctional glycosylases excise the damaged base to generate an apurinic/apyrimidinic (AP) or abasic site. Bifunctional glycosylases do the same, but also perform a lyase function (Chen et al. 1991; David and Williams 1998) that nicks the phosphodiester backbone 3¢ to the AP site. Monofunctional glycosylases repair alkylated DNA damage, while bifunctional glycosylases repair oxidative DNA damage. In contrast to the plethora of glycosylases that can remove damaged bases in BER, only one enzyme processes BER abasic sites: APE1/Ref-1 (apurinic apyrimidinic endonuclease redox effector factor-1, or APE1). APE1 hydrolyzes the phosphodiester backbone 5¢ to the AP site, creating a normal 3¢ hydroxyl terminus and an abasic 5¢ deoxyribose phosphate (5¢ dRP) terminus. In addition, APE1 recruits other proteins to assist with the repair, which includes processing by various polymerases and ligases. At this stage, DNA repair can progress via the short-patch BER (SP-BER) pathway or the long-patch BER (LP-BER) pathway. These branches of BER differ in the proteins involved, as well as the length of DNA synthesized during the repair process. The mechanisms by which one pathway is chosen over the other are still

140

M.R. Kelley et al.

being investigated; however, the type of damage and the glycosylase that removes the damage appear to be two contributing factors. The predominant BER pathway is SP-BER. SP-BER preferentially repairs normal AP sites, while LP-BER preferentially repairs oxidized and reduced AP sites (Bapat et al. 2009). SP-BER removes single-base damage by DNA polymerase b (Pol b) acting upon APE1’s endonuclease activity. Pol b removes the 5¢ dRP moiety and inserts the correct base at the 3¢ hydroxyl termimus that APE1 creates. DNA ligase III/ XRCC1 closes the nick to finish the repair. LP-BER repairs damage to two or more bases. In LP-BER, flap endonuclease 1 (FEN1) displaces a segment of 3–8 nucleotides surrounding the AP site. A DNA polymerase (d, e, or b), along with the DNA clamp PCNA (proliferating cell nuclear antigen) and replication factor-C (RF-C) inserts the correct nucleotides; then DNA ligase I finishes the repair (Bapat et al. 2009). APE1 is responsible for 95% of the endonuclease activity in the cell; in addition, its ability to recruit repair proteins to the site is critical to both BER sub-pathways (Demple and Harrison 1994; Doetsch and Cunningham 1990) (See Fig. 7.3). APE1’s reduction/oxidation signaling also influences DNA repair indirectly via modulation of transcription factors, as discussed later in this chapter. APE1’s multifunctional abilities are unique, and there is effectively no backup to any of APE1’s intracellular activities. Clinical application: small-molecule inhibitors of APE1 Indication for APE1 is responsible for almost all endonuclease activity in both branches utility of the BER pathway. Because it performs direct DNA repair functions and also influences DNA transcription proteins through redox regulation, APE1 is an attractive target for chemotherapeutics How it works Potentially two ways: • Inhibiting APE1 can sensitize tumor cells to alkylating and oxidizing agents • Some chemotherapeutic agents, including platinum-based drugs and anthracyclines, generate ROS as a by-product, which could inhibit APE1’s redox functions. Keeping transcription factors in an oxidized state adversely affects gene transcription Anti-BER agents None yet in use A dozen APE1-inhibiting compounds are under investigation. However, many challenges exist in creating a usable inhibitory APE1 agent. Some compounds have proven to be non-specific, binding to the AP site instead of to APE1; others have difficulty penetrating the cell membrane or reaching APE1’s location in the nucleus. To date, E3330 is the sole compound that can selectively inhibit APE1’s redox functions only

MMR The MMR pathway recognizes and repairs single base-pair mismatches (A-G, T-C) or misaligned short nucleotide repeats (e.g., small-loop insertions). Base–base mismatches can be due to endogenous causes: spontaneous deamination of 5-methylcytosine to

7  DNA Repair and Redox Signaling

141

thymine that causes a guanine-to-thymine mismatch, cytosine deamination to uracil resulting in a guanine-to-uracil mismatch, damage to the cellular nucleotide pool or incorrect incorporation by a DNA polymerase. The process starts when MSH2 (MutS homolog 2) combines with MSH3 or MSH6, depending on which type of mismatch is present. MSH2:MSH6 recognizes both insertion–deletion mispairs and single-base mismatches; the MSH2:MSH3 dimer recognizes only insertion–deletion mispairs. After identifying the type of mismatch, MSH proteins recruit MutL homolog 1 (MLH1) and its binding partners, post-meiotic-segregation increased-1 protein (PMS1) and PMS2, to determine the specific strand error. An exonuclease removes the DNA lesion; a DNA polymerase (d or e) synthesizes a new strand, and a DNA ligase completes the repair (Jiricny 2006; Lopez de Saro et al. 2006). Clinical application: small-molecule inhibitors of MMR Indication for Unclear. Genetic diseases linked to MMR defects (e.g., hereditary utility non-polyposis colon cancer; HNPCC) are deterrents to researching MMR inhibitors. Defective MMR repair functions actually increase tumor resistance to cisplatin How it works Inhibiting MMR enzymes is probably not an effective approach for developing new chemotherapeutics. However, reactivation of MMR enzymes may re-sensitize tumors to chemotherapy Anti-MMR agents None yet in use Experimentally, MLH1 reactivation by decitabine can increase cisplatin sensitivity – but the actual mechanism by which it does this is unknown and may not be specific to MLH reactivation

NER The NER pathway is responsible for repairing bulky DNA lesions: cyclobutane– pyrimidine dimers induced by ultraviolet light, adducts induced by polycyclic aromatic hydrocarbons and other adducts induced by cross-linking agents and basedamaging chemical carcinogens. The NER and MMR pathways are presumed to be the major pathways for repairing cross-link damage, which has significant implications for tumor survival. More than 25 proteins participate in the NER pathway, which can be divided into two sub-pathways: global genome repair (GGR) and transcription-coupled repair (TCR). The distinction is predicated on which complex recognizes the helix-distorting damage (Altieri et al. 2008; Friedberg 2001). After the initial recognition step, both sub-pathways operate the same way. A multi-protein complex binds at the damaged site. Then incisions are made on either side of the damaged strand, several nucleotides away from the actual damage. Both the 5¢ and 3¢ sides are cut. The damage-containing oligonucleotide is removed; a DNA polymerase fills the gap, and a ligase completes the repair. In TCR, recognition occurs when RNA polymerase II (RNA Pol II) stalls at a site of DNA damage. TCR-specific factors, including the Cockayne’s Syndrome

142

M.R. Kelley et al.

proteins, CSA and CSB, are recruited at the site of transcription arrest. The GGR sub-pathway starts when the heterodimer XPC/HR23B recognizes the damage, binds to the damaged DNA (distorting it so other complexes can bind) and recruits other proteins to the site. In addition, the UV-DNA damage binding protein (UV-DDB) is required for recruitment when NER repairs UV-induced cyclobutane pyrimidine dimers. Following recognition, both TCR and GGR use the same proteins to repair the damaged DNA. Transcription factor IIH (TFIIH) is recruited to the site of DNA damage. This large complex includes two helicases, XPB and XPD (xeroderma pigmentosum group B and D proteins), which use ATP energy to unwind the DNA strand on either side of the damage. The helicases work in complementary fashion: XPB unwinds DNA in the 3¢ to 5¢ direction; XPD unwinds DNA in the opposite direction. Then, XPA and RPA (replication protein A) stabilize the exposed single-strand DNA. XPA fine-tunes the location of normal versus abnormal DNA, while RPA protects both of the separated strands in the open complex. Endonucleases XPG, ERCC1 (excision repair cross-complementing group 1), XPF1 and XPC cleave the 27- to 30-nucleotide fragment and 3¢ and 5¢ of the lesion. Finally, DNA polymerase d or e, along with PCNA, RPA and replication factor C (RFC), fill the gap, using the undamaged DNA strand as a template (Kelley and Fishel 2008). Research in clinical applications for exploiting various steps in the NER pathway to develop new chemotherapeutics points to the fact that we still have much to learn about pathway signaling proteins. As we elucidate the actions of each step and the implications of inhibiting selected proteins, the opportunity for developing ­targeted therapies will become more apparent.

Clinical application: small-molecule inhibitors of NER Indication for utility Unclear. Genetic diseases linked to NER defects (xeroderma pigmentosum, Cockayne’s Syndrome) are a deterrent to researching NER inhibitors How it works Efforts in modulating NER response primarily have focused on potentiating the effects of platinum-based drugs, so any smallmolecule inhibitors developed would be chemotherapeutic adjuncts. Inhibitors to two NER proteins (XPA, RPA) and a NER enzyme (ERCC1) are currently under investigation • A global cyclin-dependent kinase inhibitor (UCN-01; 7-hydroxystaurosporine) blocks the interaction between XPA and ERCC1, which can affect the damage-recognition and incision steps of NER • An RPA inhibitor targets RPA’s ability to bind to single-strand DNA • ERCC1 expression may be modulated by a number of small molecules, including gefitinib (an epidermal growth factor receptor tyrosine kinase) and vascular epidermal growth factor (VEGF). Whether their action is specific to NER is unclear Anti-NER agents in use None yet

7  DNA Repair and Redox Signaling

143

NHEJ Repair NHEJ is the main repair pathway for DNA double-stranded breaks (DSBs) in mammalian cells. DSBs are the most severe form of DNA damage, as they can cause deletion or translocation of chromosomal DNA. Ionizing radiation, chemotherapeutic drugs, cleavage during V(D)J-recombination, meiotic recombination or the collapse of replication forks can cause DSBs. NHEJ rejoins DSBs at the ends of the broken DNA, regardless of sequence homology. In NHEJ, the break ends are ligated directly without requiring a homologous template – hence, the name of the pathway. Furthermore, NHEJ can repair DNA double-strand breaks during any phase of the cell cycle (Kelley and Fishel 2008). NHEJ begins when Ku70 and Ku80 proteins bind to the ends of the DNA breaks. Then, the Ku proteins recruit the catalytic subunit of DNA-dependent protein kinase (DNA-PKcs), which creates a nick to mark the spot for repair. The Ku/DNAPKcscomplex recruits additional proteins to make the actual repair. DNA Ligase IV and XRCC4 function as a complex to ligate the nick and complete the process (Burma et al. 2006). Discovery of new proteins involved in NHEJ includes Metnase (also known as SETMAR, histone–lysine N-methyltransferase, SET domain and mariner transposase fusion gene-containing protein) (Lee et al. 2005), which has been shown to interact with DNA Ligase IV to enhance the efficiency and accuracy of NHEJ (Hromas et al. 2008). Because repairing DNA DSBs is critical for tumor cell survival, the search for inhibitors to this pathway is an ongoing pursuit (Altieri et  al. 2008; Aiyer et  al. 2008; Akamatsu et al. 1997; Akterin et al. 2006). Currently, DNA-PKcsis the most highly researched target in the NHEJ pathway. Clinical application: small-molecule inhibitors of NHEJ Indication for utility DNA-protein kinase is a member of the phosphatidylinostitol 3-kinase (PI3-K) superfamily (Shinohara et al. 2005), which plays a key role in repairing non-homologous doublestranded DNA breaks How it works Inhibition of DNA-PKcsmakes cells hypersensitive to ionizing radiation and numerous DNA cross-linking agents by blocking re-ligation of DNA DSBs Anti-NHEJ agents in use None yet Several compounds in development (NU7441, IC60211, IC86621, AMA37 and others) are being tested in animal models

HR HR is also involved in repairing DNA DSBs. In contrast to NHEJ, HR operates only during the cell cycle’s S and G2 phases during DNA replication. The HR pathway is aptly named, as a sister chromatid must be available nearby to provide a repair template.

144

M.R. Kelley et al.

Repair starts when the MRN complex (Mre11/Rad50/Nbs1) recognizes the DSB and recruits ATM (ataxia telangiectasia mutated protein) to monitor the cell cycle’s checkpoints and analyze the break (Aiyer et  al. 2008). MRN’s nuclease activity creates single-strand DNA (ssDNA). When ATM sees the block in replication, it activates numerous downstream proteins via phosphorylation to create a complex that protects the ends of the ssDNA and enables the repair process. ATM-activated BRCA1 (breast cancer 1, early onset) attracts BRCA2 and RAD51 to bind to the ssDNA ends, allowing the RAD52/RAD54 complex to join to and form a larger complex with BLM (Bloom’s syndrome gene product) and WRN (Werner) proteins. These large protein complexes attach to the strand break, direct the pairing of processed DNA with a homologous region on the sister chromatid and initiate strand exchange (Sung and Klein 2006). Clinical application: small-molecule inhibitors of HR Indication for utility ATM regulates cell cycle checkpoints. It interrupts the cell cycle when it detects DNA damage, especially double-strand breaks How it works Inhibiting signaling through ATM can impair the cells’ response to DNA damage, which can make cells more sensitive to chemotherapeutic agents. Ionizing radiation, ROS and (indirectly) alkylating agents can easily kill ATM-deficient cells Anti-HR agents in use None yet, as previously investigated. ATM inhibitors were too non-specific and produced too many off-target effects. Under investigation is KU55933, which blocks phosphorylation of numerous ATM targets and CP466722. Early evidence indicates that it can potentiate the effects of camptothecin, doxorubicin and etoposide

Overview of Redox Signaling Oxidation–reduction (redox) signaling maintains cellular homeostasis and regulates numerous cellular processes, including metabolism, growth, reproductive biosynthesis and apoptosis. Two systems are primarily responsible for general redox regulation: the thioredoxin (Trx) and glutaredoxin/glutathione (GRX/GSH) systems. Both perform their redox and cellular regulation functions through a thiol– redox exchange mechanism (Holmgren 1995; Nakamura et al. 1997). Thiol-based redox reactions rely on binding activities of cysteine (Cys) residues. Special properties of Cys allow them to adopt 10 different sulfur oxidation states from +6 to -2 (the latter being the fully reduced state) (Giles et  al. 2003; Jacob et al. 2003), enabling them to exist in many forms (Jacob et al. 2003). The thiol/disulfide configurations in Cys drive human redox system functions in general (Giles et al. 2003). In a thiol/disulfide exchange, a reduced redox protein recognizes an oxidized protein containing a disulfide bond; the redox protein donates two hydrogen molecules to create an intermediate mixed disulfide bond on the oxidized protein.

7  DNA Repair and Redox Signaling

145

In doing so, the redox protein becomes oxidized. A reductase restores the redox protein to a reduced state, using cofactors such as FADH2and NADPH (Giles et al. 2003; Jacob et al. 2003). A specific Cys acts as a nucleophile in this type of reaction (Giles et al. 2003; Jacob et al. 2003). Because the proper folding and functionality of many proteins relies in part on the reversible formation of inter- or intramolecular disulfide bonds, redox systems have far-reaching effects on protein-mediated cellular activities (Berndt et al. 2008).

The Thioredoxin (Trx) System This system includes thioredoxin (Trx), thioredoxin reductase (TrxR) and NADPH (Jacob et  al. 2003; Holmgren 1989). Thioredoxins (Trxs) are a large family of structurally conserved oxidoreductases that reduce disulfide bonds in a variety of proteins through a thiol/disulfide exchange mechanism (Nakamura 2005). The system has a built-in “recycle-and-reuse” mechanism: thioredoxin reductase, a flavoprotein containing a selenocysteine, reduces oxidized thioredoxin in a reaction using NADPH as an electron donor. Thus, thioredoxin may be used repeatedly (Jacob et al. 2003). Although all thioredoxins contain a similar active-site motif of Cys-X-X-Cys, their most defining structural feature is the “Trx fold:” a four-stranded b-sheet surrounded by three a-helices (Holmgren 1995; Lillig and Holmgren 2007; Powis and Montfort 2001). Cys-X-X-Cys is located on the loop connecting b-sheet 1 and a-helix 1. The N-terminal Cys32 residue in the active site of human Trx is surface exposed and has a low estimated pKavalue of 6.3 (Forman-Kay et al. 1992), while the C-terminal Cys is buried in the molecule and has a much higher pKavalue. The low pKavalue of the N-terminal Cys may be due to the partial positive charge from the dipole moment associated with a-helix 2 (Hol 1985), or its hydrogen bond to the C-terminal Cys (Weichsel et al. 1996). Regardless of the cause, the low pKaof the thiolate group of the Cys increases its selective reactivity. The proposed reaction mechanism of disulfide reduction by thioredoxin is as follows: the N-terminal cysteine thiolate of Trx acts as a nucleophile and attacks the target disulfide on the oxidized protein, resulting in a transient mixed disulfide intermediate. This, in turn, is reduced by the C-terminal active site Cys residue, generating a dithiol in the target protein and a disulfide in thioredoxin (Holmgren 1995; Lillig and Holmgren 2007; Kondo et al. 2006). The resulting disulfide in the active site of Trx can be reduced by TrxR. The mechanism by which it regenerates Trx back to the dithiol involves the formation of a selenylsulfide on the active site of Trx (Datta et al. 2004). A second redox active site located in the other subunit of the dimeric TrxR contains two thiols that reduce the selenylsulfide back to a thiol and selenol, resulting in the formation of a disulfide bond. Electron transfer from FADH2reduces this disulfide; the resulting FAD is then reduced by electron transfer from NADPH (Jacob et al. 2003) to complete the catalytic cycle.

146

M.R. Kelley et al.

The Glutaredoxin/Glutathione (GRX/GSH) System The glutaredoxin system includes glutaredoxin, glutathione, NADPH and the flavoprotein glutathione reductase (Holmgren 1989; Kondo et al. 2006; Lillig et al. 2008). Like the Trx system, this system also works through a cascade of disulfide oxidation and reduction. Glutaredoxins (GRXs) are small redox enzymes (100 amino acids) that use glutathione as a cofactor. Structurally, glutaredoxins are very similar to thioredoxins, exhibiting the same fold and active sites. However, the active site of GRXs includes Cys-X-X-Cys or Cys-X-X-Ser. GRXs catalyze the reversible reduction of substrate protein disulfides, resulting in the oxidation of the GRXs. Oxidized GRXs are reduced non-enzymatically by glutathione (glutamyl-cysteinyl-glycine, or GSH). Then, the oxidized glutathionine disulfide group (GSSG) is returned to a reduced state by the action of glutathione reductase, at the expense of NADPH (Lillig et al. 2008). Although the thioredoxin system and glutaredoxin system share many functions, they are not redundant systems. Trx and GRX act on different substrates (Lillig et al. 2008; Fernandes and Holmgren 2004). For example, Trx – but not GRX – is implicated in the reduction of APE1, a protein exhibiting both redox and DNA repair functions (Akamatsu et  al. 1997; Hirota et  al. 1997; Huang and Adamson 1993; Tell et al. 2005).

Roles of Redox Systems The thioredoxin and glutaredoxin systems scavage ROS to reduce intracellular oxidative stress. A reduced state is important for optimizing many cell functions including growth and differentiation, biosynthesis and apoptosis. The systems’ collective efforts to maintain a reduced intracellular environment not only facilitate correct protein folding and efficient work of protein chaperones but they also prevent oxidative modifications to DNA, including base changes, changes to the deoxyribose backbone and direct breakage of a strand (Berndt et al. 2008). Both the DNA damage response and the wide-ranging influence of redox systems include activation and control of numerous transcription proteins (Powis and Montfort 2001; Arner and Holmgren 2000; Yamawaki and Berk 2005). The “dark side” of the two redox systems is that they can contribute to the pathophysiology of various human diseases, including cancers, viral infections, neurodegenerative diseases (including Alzheimer’s) and others (Akterin et al. 2006; Arner and Holmgren 2006; Burke-Gaffney et  al. 2005; Raffel et  al. 2003). For example, APE1 plays a role in the accumulation of amyloid b protein in Alzheimer’s plaques (Bhakat et al. 2008). Thus, inhibition of redox systems is a vital drug target for many therapeutic areas (Biaglow and Miller 2005; Chew et al. 2008; Liu et al. 2009; Liu 2008; Mukherjee and Martin 2008).

7  DNA Repair and Redox Signaling

147

Clinical application: small-molecule inhibitors of Trx (Tonissen and Di Trapani 2009) Indication for utility Increased levels of Trx and TrxR are present in many cancers. This also correlates with increased resistance to chemotherapeutics The Trx system may assist early cancer development via its growth-promoting and anti-apoptotic functions. Later, the Trx system may influence angiogeneis and metastasis How it works All redox-inhibiting agents ultimately force cancer cells to die by creating an oxidative intracellular environment that exceeds the cell’s redox damage–repair abilities; this, in turn, triggers one of the apoptotic pathways. Redox-inhibiting agents often target the selenothiol of TrxR to accomplish this Anti-Trx agents in use • SAHA (a histone deactylase inhibitor): upregulates an endogenous inhibitor of Trx • Platinum compounds: Although they are most well known for forming irreversible DNA-platinum adducts, these compounds also affect both the Trx and GRX redox systems • ATO (arsenic trioxide): inhibits TrxR via a largely unknown mechanism of action Emerging treatments • PX-12 (1-methyl-propyl-2-imidazolyl disulfide): binds to Cys73 of Trx, irreversibly alkylating it, so that TrxR cannot regenerate it. Also decreases expression of VEGF (which Trx indirectly influences)· Gold compounds: inhibit TrxR in a mechanism still being elucidated. Mode of action may be similar to that of platinum compounds • MGd (motexafin gadolinium): inhibits Trx and generates ROS, including superoxides and hydrogen peroxide. Also inhibits an essential enzyme for DNA synthesis • DCNB (nitroaromatic compounds): inhibits TrxR and induces superoxide generation • Curcumin & flavonoids: convert TrxR to a pro-oxidant form that stimulates ROS production

The Redox Activity of APE1 APE1 is the only DNA repair protein known to have a role in redox regulation, which affects the expression of a number of DNA transcription factors. The protein’s dual functions are reflected in the history behind its full designation: APE1/Ref-1. The multifunctional protein was first identified as being responsible for reducing transcription factor AP-1; therefore, it was named “redox effector factor 1,” or “Ref-1” (Xanthoudakis and Curran 1992, 1996). Since this initial discovery, further studies revealed that APE1 is responsible for regulating many transcription factors including NFkB, HIF-1a, p53, PAX and others (Hirota et al. 1997; Cao et al. 2002; Ema et al. 1999; Hirota et al. 1999; Lando et al. 2000; Tell et al. 2000; Ueno et al. 1999) (Fig. 7.3a). And, as discussed in the first section of this chapter, APE1 performs critical DNA endonuclease repair functions. That spawned its other designation: “APE1,” or apurinic/apyrimidinic endonuclease. Today, the complete name of this protein (APE1/Ref-1) is usually truncated to “APE1.”

148

M.R. Kelley et al.

How APE1 Performs Its Redox Functions APE1’s uniqueness is manifested both in its breadth of function and how it functions, which is still somewhat of a mystery (Georgiadis et al. 2008; Walker et al. 1993). Animal and bacterial APE1 proteins contribute to DNA repair but lack any redox abilities. However, human APE1 does both, making it unique in that regard. How APE1 performs its redox activity remains a controversy, as only the Cys65 residue in APE1 has been identified as critical for its redox reactions (Table 7.1). Cys65 is buried near the N-terminus on the first beta strand in the fold, which is part of a beta sheet in the core of the protein. Normally, two Cys residues are required for thiol/disulfide exchange reactions; however, APE1 does not contain two Cys residues within a C-X-X-C motif that are structurally close enough (~2.2Å) to form the disulfide bond required to drive a redox reaction. APE1’s Cys residues that are in closest proximity to each other are buried in APE1’s core and are on opposite sides of the beta sheet (Fig. 7.3a, b) (Luo et al. 2009). Given these apparent obstacles, four possibilities have been proposed regarding the mechanism for APE1’s redox abilities. The first speculation is that a yet undiscovered Cys residue is involved. A second possibility is that an undiscovered Ser reside is involved, as is found in glutaredoxins. However, to function properly, the Ser would require a pKathat is unusually low for its proposed location. A third postulation is that APE1 undergoes a conformational change to restore its reduced state. This type of activity is seen in another group of redox proteins called peroxiredoxins, which are responsible for catalyzing intracellular hydrogen peroxide (Hofmann et al. 2002). Peroxiredoxins lack a C-X-X-C motif, but they do include two Cys residues that are required for activity. Peroxiredoxins exist as fully folded dimeric structures, where the nucleophilic thiolate is sequestered and the resolving thiolate is located near the C-terminus of the molecule. One Cys from each monomer forms the active site (Wood et al. 2003). Being approximately 9Å from each other, one would wonder how they could form an active site; however, a local unfolding places them in close proximity to detoxify hydrogen peroxide. APE1 could have a similar requirement of a conformational change to complete its catalytic cycle. Further support for this theory can be found in the way that GAPDH interacts with APE1. GADPH acts upon APE1’s Cys152 active site, causing a conformational change that enables APE1 to perform its endonuclease activity (Luo et al. 2009). A fourth possibility exists for explaining APE1’s redox functionality: the involvement of two APE1 molecules. This would provide the second Cys or Ser residue needed to resolve the redox reaction. Further studies are needed to elucidate this important feature of APE1, as it has many implications for future clinical therapeutics.

APE1-Regulated Transcription Factors and Their Link to DNA Damage Repair Because APE1 is involved in both DNA damage repair and the redox regulation of a number of stress-inducible transcription factors, APE1 clearly plays pivotal roles

7  DNA Repair and Redox Signaling

149

in both redox signaling that modulates DNA damage response and direct DNA repair functions. APE1 is a reductive activator for several transcription factors, including p53, AP-1 and HIF1-a (Fig. 7.3a, b). Analysis of APE1’s effector action on these proteins underscores additional unique features about APE1. These transcription factors are unrelated multidomain proteins; that is, APE1 does not recognize any common structural motif when acting upon them. Furthermore, the redox-regulated DNA-binding domains of these factors are structurally distinct: p53, a single immunoglobulin-like domain; NFkB (discussed here in conjunction with p53), a dimer of two-domain immunoglobulin-like subunits; AP-1, a heterodimeric bZip family protein; and HIF-1a, a basic helix-loop-helix domain. Redox regulation of transcription factors is one of several mechanisms that control sequence-specific DNA binding and thereby, gene expression. APE1’s influence on p53, AP-1 and HIF-1a affects the DR, BER, HR, MMR and GGR pathways, as discussed on the following pages (Table 7.1 and Fig. 7.2).

p53 Often called “the guardian of the genome” (Zaky et al. 2008), p53 is a transcription regulatory protein that helps preserve genomic integrity by its participation in stress–response pathways and DNA repair pathways (Lando et al. 2000; Helton and Chen 2007; Sengupta and Harris 2005; Zurer et  al. 2004). More than half of all cancers contain mutant or inactive p53, underscoring its importance as a tumor suppressor. Two overall strategies that p53 employs in maintaining the genome are to arrest the cell cycle or to induce apoptosis. For example, hypoxia in tumor cells prompts p53 to induce apoptosis (Jayaraman et  al. 1997). Under other circumstances, p53 can interrupt the cell cycle to allow sufficient time to complete DNA repairs (Jiang et al. 2009). The far-reaching influence of p53 also includes complex interactions with other proteins and effector genes. Initial investigation of redox regulation of p53 was based on the finding that oxidized p53 has little binding affinity for DNA (Hainaut and Milner 1993). This led to the discovery that APE1 was responsible for enhancing the DNA-binding activity of wild-type p53 by reducing it (Jayaraman et al. 1997). The mechanisms by which APE1 enhances this activity of p53 are still being discovered, but they appear to include both a redox-dependent activation of the DNAbinding domain (through direct reduction of a disulfide bond), as well as a redox-independent mechanism (Jayaraman et al. 1997). APE1 reduces p53 (Jayaraman et al. 1997). This is essential, but not sufficient, for optimizing p53’s DNA-binding function (Jayaraman et al. 1997). Recent evidence shows that APE1 also facilitates tetramerization of p53 by interacting with its C-terminal regulatory domain in a redox-independent manner, producing p53’s optimal configuration for binding to DNA. Thus, APE1 appears to enhance sequence-specific DNA-binding activity of p53 through two independent, coordinated mechanisms (Hanson et al. 2005). Redox signaling involving p53 depends upon APE1 plus a general redox factor as well. Trx has been shown to enhance stimulation of p53-dependent expression

150

M.R. Kelley et al.

APE1 has two major functions: redox regulatory and DNA repair activity

DNA damage

Redox Role

(reduced) APE1 (oxidized)

T A

AP 1 p53 NFkB HLF HIF 1a CREB ATF Egr 1 NF Y PAX

(oxidized)

(reduced)

DNA Repair G C

A T

T A

C G

T G A T C A C T T A G

*

T A

G A T C C T T A G

T A

G C

APE1

5’ P

T G A C

b

T

Glycosylase (Ogg1, Nth1)

Redox control of Transcripon factors

DNA Repair gene expression altered

Alteraon of gene expression

A

A T C T T A G

A

T A

C G

G A A T C T T A

C G

T

A T

DNA Ligase T A

Transcription Factor

H2N

S-S

Oxidized AP-1, HIF-1α, NFkB

COOH

DBD

APE1RED

DNA

H2N

APE1OXIDIZED

COOH HS

DBD

SH

Target gene expression

Regulatory region

Fig. 7.2  Role of APE1 in regulating expression of other DNA repair proteins and pathway through its redox signaling function

7  DNA Repair and Redox Signaling

151

of p21 by APE1, suggesting a link between Trx and APE1 in cellular response to oxidative stress (Ueno et al. 1999). Because p53 influences many DNA repair pathways and APE1 constitutively influences p53, a short discussion of their effects on each pathway is warranted. p53’s regulation of DNA repair is complex, involving both transactivation-dependent and -independent mechanisms that affect the HR, BER, NER, MMR and NHEJ pathways (Sengupta and Harris 2005). The BER pathway, the main mechanism for repairing oxidative and alkylating base modifications, is initiated by highly specialized DNA glycosylases that cleave the DNA base to create an AP site. p53 regulation of BER varies according to the type of genotoxic stress placed on cells. For example, exposure to nitric oxide species causes p53 to down-regulate the specific glycosylase that would begin repairs in response to this threat. Interestingly, NO does not alter APE1 activity. In this respect, p53 acts as a “brake,” stopping repairs to prevent the possibility of engendering a mutagenic phenotype. However, in response to g-ray treatment, p53 promotes activation of AAG (Zurer et al. 2004). The choice of response also may be specific to the cell line involved. Recent studies with human colorectal cancer line HCT116 show that wild-type p53 (but not mutant p53) downregulates APE1 by indirect recruitment of its promoter, possibly blocking binding of SP1 (Zaky et al. 2008). Thus, in this example, APE1 regulates DNA binding of p53; and p53, in turn, regulates both expression of APE1 as well as certain proteins that bind to the promoter of APE1. This system of checks and balances maintains equilibrium between p53’s pro-apoptopic activity and APE1’s pro-survival activity. In addition, p53 regulates DNA polymerase b (Pol b), which replaces excised bases in the SP-BER pathway. The single-nucleotide replacement mechanism of Pol b requires dRP lyase activity for the repair (Allinson et al. 2001). This lyase activity occurs after APE1 activity and is often rate limiting in BER. Pol b expression is down-regulated in p53-deficient cells (Cao et al. 2002), which increases the likelihood of the accumulation of cytotoxic repair intermediates that effectively can halt DNA replication. Additionally, in vitro studies with various cell lines show that CREB (which recruits and stabilizes Pol b) can upregulate Pol b gene expression in response to DNA alkylating exposure (He et al. 2003). CREB (cAMP response element binding) is under redox control for its DNA binding and activating transcription. Other proteins in the DNA repair arsenal are under redox control, and this control regulates DNA repair response (Fig. 7.2). p53 differentially affects the two NER sub-pathways, TCR and GGR. Several studies found that p53 selectively affects GGR, but not TCR. In GGR, sublethal UV radiation causes p53 to arrest the cell cycle in G1, to gain the needed time for DNA repair before progressing to S phase (Smith and Seo 2002; Li et al. 2001). In addition, p53 also regulates GGR downstream effectors, DDB2 and XPC, the main proteins involved in DNA damage recognition (Adimoolam and Ford 2002; Hwang et al. 1999; Tan and Chu 2002). Inactive or absent p53 results in defective GGR repair activity and appears to lead to genomic instability (Xu and Morris 1999). Studies of knockout mice support this: in one study, 100% of XPC-/-mice developed lung cancer and 100% of DDB2-/-mice developed skin tumors (Hollander

152

M.R. Kelley et al.

et al. 2005). PCNA, a progressivity factor in the NER pathway, is also under p53 regulation (Xu and Morris 1999). The MMR pathway removes and repairs DNA mismatches because of the DNA polymerase errors. MSH2, which must be complexed with MSH6 or MSH3 to recognize DNA damage, is under p53 regulation. Low levels of p53 prevent MSH2:MSH6 recognition of single-base mismatches and short insertion/deletion mispairs, and MSH2:MSH3 recognition of larger loops of unpaired nucleotides. MSH2, MLH1, and PMS2 all are under p53 regulation, similar to the way that DDB2 and XPC are in NER (Chen and Sadowski 2005; Scherer et al. 2000). Both the HR and NHEJ pathways address reparation of double-strand breaks. DSBs are the most severe threats to genomic stability because they facilitate deletion and/or translocation of chromosomal DNA. As a result, both pathways are highly regulated. Once again, p53 appears as an important player in the repair of DSBs through regulation of both the HR and NHEJ repair pathways. Mice that are deficient in wild-type p53 exhibit increased levels of HR activity (Bishop et  al. 2003; Lu et al. 2003). p53 inhibits HR through repression of RAD51 expression (Arias-Lopez et  al. 2006). (Conversely, absent or mutated p53 increases RAD51 DNA repair activity, which contributes to chemotherapeutic resistance.) In the NHEJ pathway, p53 downregulates expression of WRN and RecQ4, two RecQ helicases (Sengupta et al. 2005; Yamabe et al. 1998). Finally, in the DR pathway, studies on murine fibroblast cell lines have demonstrated that AGT is under p53 regulation (Grombacher et  al. 1998; Rafferty et  al. 1996). Interactions between p53 and AGT are complex and depend on many variables, including the type of cellular stress present. For example, p53 appears to induce AGT in response to ionizing radiation. AGT also appears to be regulated by NFkB, a transcription factor mediating immune and inflammatory responses. For example, overexpression of NFkB in HEK293 cells (human embryonic kidney 293 cells) increases AGT expression (Lavon et al. 2007). NFkB is under redox control by APE1. All these studies point to the possible clinical relevance of redox-controlled DNA repair responses through p53. If reduced p53 is required to bind DNA and either activate or repress the transcription of DNA repair genes, then it follows that APE1 plays an important role not only in the redox modulation of p53, but also in the regulation of DNA repair.

AP-1 Activator protein-1 (AP-1) refers to a family of proteins that recognize AP-1 sites. The AP-1 family is also referred to as tetradecanoylphorbol-13-acetate (TPA)responsive elements. AP-1 mediates diverse cellular functions including proliferation, differentiation, apoptosis and transformation; it also responds to environmental changes including stress, radiation or growth factor signals (Hess et al. 2004). The AP-1 family consists of structurally and functionally related basic leucine zipper

7  DNA Repair and Redox Signaling

153

proteins (bZIP) that intermix to form heterodimeric sequence-specific DNAbinding proteins. These include Jun proteins (c-Jun, JunB and JunD), Fos proteins (c-Fos, FosB, Fra-1 and Fra-2) and some ATF family members (ATFa, ATF-2 and ATF-3) (Hess et al. 2004). Studies of AP-1 activity provided the backdrop for early investigations into APE1’s redox activity. In fact, APE1 was first identified as a result of efforts to determine the nuclear factor responsible for reducing AP-1 (c-Jun/c-Fos) and thereby enhancing its DNA-binding activity (Xanthoudakis and Curran 1992; Abate et al. 1990). Redox regulation of AP-1 was found to result from oxidation/ reduction of conserved Cys residues within the basic DNA-binding domains of c-Jun and c-Fos, which enables them to dimerize (Abate et al. 1990). APE1 copurifies with AP-1 (Xanthoudakis and Curran 1992), which is in stark contrast to APE1’s interaction with p53. The former is a stable interaction; the latter is transient. Trx has also been identified as a factor modulating transcriptional activation of AP-1 through its reduction of APE1 (Hirota et al. 1997). AP-1 is rapidly induced in response to many cellular stimuli, and its activation regulates the expression of several proteins involved primarily in NER and MMR DNA repair pathways. Numerous studies of AP-1’s influence on DNA repair proteins have revealed more than 20 genes whose promoters are bound following cisplatin treatment. The proteins that AP-1 regulates in repairing DNA–cisplatin adducts include RAD23B, XPA, ERCC3, XPF, ERCC1 and ERCC3 in the NER pathway and MSH2, MSH6, MLH1 and PMS2 in the MMR pathway. Phosphorylated c-Jun or ATF2 activates these proteins to increase their DNA-binding capacity (Luo et al. 2009; Luo et al. 2008). Because AP-1 must be converted from an oxidized to a reduced state in order to bind to its target sequence, redox control of this protein has significant implications. APE1 is implicated as a potential point of control for regulating the DNAbinding activity of AP-1, thus modulating the expression of DNA repair proteins (Table 7.1and Fig. 7.2).

HIF-1a Hypoxia inducible factor-1 (HIF-1), as its name suggests, responds to low-oxygen conditions in the cell. Increasing evidence reveals that hypoxic stress in the tumor microenvironment can cause genetic instability in cancer cells. HIF-1 is a heterodimeric transcription factor composed of two subunits; the HIF1a subunit is crucial for regulating cellular response to hypoxia and is frequently overexpressed in human cancers. Under hypoxic conditions, HIF-1a translocates to the nucleus and dimerizes with HIF-1b, forming HIF-1. HIF-1, along with coactivators, binds hypoxia response elements (HRE) within promoters and regulates the expression of their downstream genes, including vascular endothelial growth factor (VEGF) (Forsythe et al. 1996; Jiang et al. 1996). Redox-dependent stabilization of the HIF-1a protein is required for activation of HIF-1 (Huang et  al. 1996), and

154

M.R. Kelley et al.

redox signaling regulates the DNA-binding activity of HIF-1. These combined activities suggest a role for APE1 as a redox regulator of HIF-1a. Furthermore, APE1 has been implicated in its role of helping to form the hypoxia-inducible transcriptional complex that includes HIF-1 and transcriptional coactivators p300 and CREB. More recently, HIF-1a has been shown to play a role in down-regulating mRNA and APE1 levels under hypoxic conditions in human microvascular endothelial cells (Loboda et al. 2009). Thus, HIF-1a regulates the expression levels of APE1 and is itself regulated by APE1. Hypoxia downregulates the expression of key genes within two DNA repair pathways. In HR, several critical mediators including BRCA1, BRCA2 and RAD51 are inhibited, resulting in significant genetic instability. In addition, HIF-1a decreases the expression of NBS1, a member of the Mre11-RAD50-NBS1 (MRN) complex that initially recognizes DSBs (To et  al. 2006). In MMR, MLH1 and MSH2 are inhibited (Bindra and Glazer 2007; Bindra et  al. 2004; Bindra et  al. 2005, 2007; Bindra and Glazer 2007; Bristow and Hill 2008; Koshiji et al. 2005; Meng et al. 2005; Mihaylova et al. 2003). The MMR genes appear to be repressed through a mechanism involving c-Myc (Bindra and Glazer 2007). Interactions between HIF-1a and these genes are still being determined, but some studies show that HIF-1a represses MSH2:MSH6 in a p53-dependent manner (Koshiji et  al. 2005). Regulation of DNA repair is an integral part of the hypoxic response; and HIF1a, a critical response mediator, influences the DNA repair response (Fig. 7.2). As APE1 is required for redox regulation of HIF-1a, it then also plays a role in the regulation of DNA repair genes controlled by HIF-1a. APE1 also exhibits redox-independent transcriptional regulatory functions that are subject to post-translational modifications including phosphorylation, acetylation and nitrosation. The purposes of these modifications are still being elucidated. For example, acetylation of APE may influence its subcellular localization or transacting functions (Bhakat et al. 2008). Review articles by Bhakat and others describe these implications further (Bhakat et  al. 2008; Tell et  al. 2009).

Other Global Influences of APE1 APE1 is a critical “master regulator” for maintaining genomic stability (Fig. 7.3). Earlier sections of this chapter discussed APE’1s apurinic/apyrimidinic endonuclease activity that is essential to BER’s DNA repair activities (Bhakat et al. 2008). In addition, APE1 performs redox functions that include modulation of transcription factors. APE1 controls the redox status of either ubiquitous (AP-1, Egr-1, NF-kB, p53, CREB, HIF-1a) or tissue-specific transcription proteins (PEBP-2, Pax-5 and -8 TTF-1) (Fig. 7.3a) (Akamatsu et  al. 1997; Huang and Adamson 1993; Xanthoudakis and Curran 1992; Cao et  al. 2002; Ema et  al. 1999; Hirota et al. 1999; Lando et al. 2000; Tell et al. 2000; Ueno et al. 1999; Xanthoudakis et al. 1992).

7  DNA Repair and Redox Signaling

155

Fig. 7.3  APE1 has dual roles in redox and DNA repair. APE1 possesses two major functions: Redox regulatory/signaling and DNA repair. Through its redox function, APE1 regulates gene expression by modifying the redox status of some transcription factors which are involved in variety of cancer processes. (Adapted from: Luo et al. (Luo et al. 2008))

Cell Survival APE1 is also essential for cell survival and organism development, as evidenced by the fact that no viable knockout model of APE1 exists. The absence of APE1 in mice causes post-implantation embryonic lethality between days E5 and E9 (Xanthoudakis et al. 1996). Conditional knockout and knock-down strategies also confirm the crucial role of this protein in cells (Fishel et al. 2008; Fung and Demple 2005; Izumi et al. 2005). Studies have demonstrated that altering APE1 levels leads to changes in cell growth, survival and sensitivity (Chen and Olkowski 1994; Fishel et al. 2008; Bobola et al. 2005; Fishel et al. 2007; Herring et al. 1998; Lau et al. 2004; McNeill and Wilson 2007; Ono et al. 1994; Walker et al. 1994; Wang et al. 2004). Experiments with normal cells show that siRNA directed against APE1 results in decreased cell proliferation, increased AP sites and increased apoptosis (Fishel et al. 2007). While the extent of APE1’s activities for cell survival and overall growth are still unknown, recent studies show that APE1 acts as a quality control monitor of RNA. During the S phase of the cell cycle (when rRNA is synthesized), APE1 localizes in the nucleus and enhances de  novo protein synthesis and cell proliferation by methods only partially understood. The N-terminus of APE1 binds to proteins involved in generating ribosomes and processing RNA (Barnes et al. 2009; Vascotto et al. 2009). APE1 also performs an RNA “cleansing” process that includes cleaving abasic RNA. Perhaps RNA’s single-strand construction makes it more susceptible to oxidative insults than DNA; therefore, APE1’s “cleansing” function affects overall cellular viability. APE1’s nuclear localization corresponds to increased

156

M.R. Kelley et al.

κ

α

β α

κ

Fig. 7.4  Redox control of DNA repair. In its role as a redox factor, APE1 reduces a number of transcription factors including AP-1, Egr-1, NF-kB, p53, CREB, and HIF-1a. Thus, APE1 controls the redox status of several transcription factors that in turn regulate expression of APE1

intracellular resistance to ROS, g radiation, and a number of other cytotoxic agents (Bhakat et al. 2008) Fig. 7.4. Other recent studies propose that APE1’s intracellular redox control may also affect the cell cycle by inducing nucleus–cytoplasm redistribution of p21, which regulates the G1 phase (Merluzzi et al. 2008).

Angiogenesis A mosaic of APE1’s redox influence on angiogenesis can be seen by observing its known functions in normal endothelial cells, plus examining in vitro and preclinical studies of its inhibition. In normal endothelial cells, APE1’s redox activity helps (1) maintain cell differentiation, (2) maintain vascular tone by regulating NO levels (Zou et al. 2009), (3) enable differentiation of angiogenic progenitor cells and (4) prevent apoptosis by repressing pro-apoptotic TNFa signaling and upregulating pro-survival NFkB signaling. Intracellular hypoxia is a feature of many cancers, including pancreatic and prostate carcinomas. Thus, decreased activity of APE1 influences downstream genes that induce endothelial cell death (Bhakat et al. 2009). In vitro studies hint at additional influences of APE1 on endothelial cells. Studies show that inhibition of APE1 attenuates the formation of retinal vascular endothelial cells (RVECs), and, to a lesser extent, their proliferation (Luo et  al. 2008). In a

7  DNA Repair and Redox Signaling

157

d­ ifferent realm of endothelial health, Tat-mediated protein transduction of APE1 decreases vascular cell adhesion molecule-1 expression, blunting the response of monocyte adhesion factors to atherosclerotic injury (Song et al. 2008). Preclinical studies of E3330, a unique quinone that selectively inhibits APE1’s redox functions, allude to APE1’s influence on endothelial cells in the tumor microenvironment. Tumors will not grow larger than a few millimeters unless they have constant access to a blood supply. Migration of endothelial progenitor cells to the tumor microenvironment is essential for neoangiogenesis. Preclinical utility studies with E3330 demonstrate that it inhibits the growth of pancreatic cancer cell lines (Zou and Maitra 2008) as well as in pancreatic cancer-associated endothelial (PCECs) and endothelial progenitor cells (Zou et al. 2009). Because intracellular hypoxia is a common feature of many cancers, E3330’s influence on HIF-1 has been studied as well. E3330 inhibits HIF-1’s DNA-binding activity, which is consistent with E3330’s ability to inhibit APE1’s redox functions. Thus, inhibition of APE1 induces endothelial cell growth arrest via a cascade of signaling upstream from HIF-1a. Additional studies of E3330’s effect on human bone marrow cells, pancreatic cancer cells and human umbilical vein endothelial cells show that E3330 can reduce tumor endothelial VEGF secretion – and, at the same time, downregulate levels of the cognate receptor Flk-1/KDR on PCECs, blocking a potentially critical angiogenic ligand–receptor interaction in the tumor microenvironment. In addition, E3330 can block the differentiation of bone-marrow hemangioblasts (Zou et al. 2009).

Inflammation Long-standing inflammation is a risk factor for tumorgenesis. For example, patients who live with ulcerative colitis (UC) for more than 10 years have a 20- to 30-fold increased risk of developing colorectal cancer. Chromosomal instabilities can be detected even in early-stage dysplastic UC tissues (Guo et al. 2008). Sources of inflammation such as ROS and toxic agents transiently increase intracellular levels of APE1 (Evans et al. 2000; Luo et al. 2009; Tell et al. 2009). Prolonged stimulation and response of APE to intracellular stresses from inflammation can cause genomic instability. It is well known that many DNA repair intermediates are cytotoxic and potentially mutagenic – including abasic sites and single-strand breaks (Bapat et  al. 2009). If DNA repairs stall or are performed incorrectly, they can lead to microsatellite instability (MSI) – which is present in the aforementioned dysplastic colon cells. Inflamed areas of UC lesions contain significantly greater amounts of APE1 and AAG than tissues in non-inflamed areas, suggesting that long-term adaptive increases in these proteins contribute to the production of MSI. Many other sporadic tumors also contain MSI, which provides motivation to study BER and other DNA repair pathways in other chronic inflammatory diseases. Overexpression of

158

M.R. Kelley et al.

AAG and APE1 is likely just one example of how long-term dysregulation of a DNA repair pathway can contribute to tumorgenicity (Guo and Loeb 2003).

DNA Repair in the Tumor Microenvironment Little is known about DNA repair in the tumor microenvironment, except that it is abnormal. If DNA repair is imperfect or incomplete, it can result in DNA damage that is genomically unstable, which is a key step for initiation and progression of cancer (Kruman et al. 2004; Schwartz et al. 2007). Tumors can be expected to be defective or deficient in one DNA repair pathway, but they have an astonishing ability to compensate and adapt by using other repair mechanisms. Crossover, interaction and compensation within and among DNA repair pathways often allow cancer cells to avoid or escape apoptosis (Reed et al. 2009). For example, cancer cells deficient in the proteins of the HR pathway may compensate for this deficiency by repairing DSBs through the NHEJ or BER pathway (Christmann et al. 2003; Hess et al. 2004; Beckman and Loeb 2006; Handel et al. 1995). A tumor’s DNA repair abilities are directly related to its unchecked cellular proliferation and its intrinsic or acquired cellular resistance to clinical DNAdamaging agents (Fishel and Wilson 1997; Holmgren 1989; Arner and Holmgren 2006; Bravard et al. 2009). So, in theory, the very defect that makes cancer cells highly mutable should be able to be exploited – to create a drug that makes the original defect lethal to cancer cells without harming normal cells. Therefore, inhibiting specific proteins in DNA damage repair pathways is a promising strategy for developing targeted cancer treatments (Fishel and Wilson 1997; Hromas et al. 2008; Holmgren 1989; Arner and Holmgren 2006; Bravard et al. 2009). It may sound paradoxical to inhibit DNA repair pathway proteins because cancer promotion and deregulated cellular growth are aided by deficient DNA repair pathways. However, a fine balance exists between the induction of DNA damage and its efficient repair (Hromas et al. 2008; Arner and Holmgren 2006). Inhibiting specific DNA repair proteins can selectively sensitize cancer cells to chemotherapeutic agents (Hromas et al. 2008; Holmgren 1989). Great clinical interest exists for combining DNA repair inhibition with current chemotherapy regimens (Beckman and Loeb 2006) to develop targeted therapies. Redox regulation plays an important role in DNA repair, and increasing attention is being focused on redox activities as potential targets for new chemotherapeutic agents. APE1 is a major player in redox signaling to downstream transcription factors that regulate stress and DNA repair responses. Increased levels of APE1 have been linked to tumor promotion, progression, poorer prognosis, shorter time to progression and poorer outcomes from chemotherapy (Hanson et al. 2005). Abnormal APE1 levels are speculated to be one cause of resistance to chemotherapy. Although APE1 largely is restricted to the nucleus in normal cells, its increased level in the cytoplasm has been associated with more aggressive lung, thyroid, breast and ovarian cancers (Zaky et al. 2008; Di Maso et al. 2007). The cause of

7  DNA Repair and Redox Signaling

159

this phenomenon is yet unknown, but it may provide another clue in understanding how APE1 redox activity is altered in the tumor microenvironment. Ample data demonstrate that downregulation or inhibition of APE1 using RNA interference and DNA antisense oligonucleotide techniques can sensitize cancer cells to laboratory and clinical chemotherapeutic agents (Bapat et al. 2009; Lillig et  al. 2008; Hainaut and Milner 1993; Herring et  al. 1998; Barnes et  al. 2009; Langie et  al. 2007; Madhusudan and Hickson 2005; Mathers et  al. 2007). Thus, regulation of APE1 is emerging as a major player in manipulating the tumor microenvironment and modulating immune response. Redox regulation clearly plays an important role in DNA repair. APE1’s multifunctional abilities in DNA repair and redox signaling of transcription proteins are unique; no backup exists for its critically important activities. Therefore, identifying selective inhibitors of APE1 make it an excellent target for both single-agent and combination therapeutic approaches. Future research in this area will bring additional insights into the importance of redox regulation in DNA repair and will likely result in the identification of other key redox proteins.

Modulating APE1’s Activities as a Cancer Therapeutic Approach Elevated APE1 levels are typically associated with aggressive proliferation, increased resistance to therapeutic agents and poor prognosis (Evans et al. 2000; Wang et al. 2004; Moore et al. 2000; Xu et al. 1997) in various cancers. Altered or elevated levels of APE1 have been observed in breast and ovarian cancers, gliomas, sarcomas (osteosarcomas and rhabdomyosarcomas) and multiple myelomas, among others (Chen and Olkowski 1994; Hainaut and Milner 1993; Hanson et  al. 2005; Koshiji et  al. 2005; Langie et  al. 2007; Madhusudan et  al. 2005). Decreasing APE1 levels can block tumor cell growth and increase cellular sensitivity to DNA-damage agents (Fishel et al. 2008, 2007; Lau et al. 2004; Ono et al. 1994; Wang et al. 2004). However, current methods for modulating APE1 alter all its functions. Because of APE1’s diversity, it is important to distinguish and characterize which of APE1’s function(s) are involved in different biological events – especially those that may differ in normal versus pathologic cells such as cancer cells. The relative importance of APE1’s redox vs. repair function in cancer is unknown. Efforts to determine the differential therapeutic effects of inhibiting one or the other are ongoing. Blocking of the redox function of APE1 would be expected to impact the activity of a number of downstream transcription factors and the genes products that they regulate. Exploration of novel targets clearly merits pursuit, particularly in cancers for which current treatments are ineffective. Many small-molecule inhibitors in development are specific to proteins in DNA damage–response. Designed as potentiators of traditional chemotherapeutics, their goal is to inhibit targeted proteins while adding minimal toxicity to the treatment (Reed et al. 2009).

160

M.R. Kelley et al.

There has been considerable debate in the literature regarding the wisdom of inhibiting essential DNA repair enzymes (Kelley and Fishel 2008; Bapat et  al. 2009; Madhusudan and Hickson 2005; Damia and D’Incalci 2007; Fishel and Kelley 2007). Targeting a specific protein (particularly one that plays an important role in cellular response to stress) by chemical knockout through the use of a smallmolecule inhibitor may have unintended consequences. However, small-molecule inhibitors have been identified for several DNA repair enzymes including MGMT (AGT), poly ADP-ribose polymerase (PARP1), ataxia-telangiectasia mutated kinase (ATM kinase), APE1 and DNA PKcs (Madhusudan and Hickson 2005; Damia and D’Incalci 2007; Fishel and Kelley 2007). The potential for small-molecule inhibitors to attenuate one or more of APE1’s functions offers potential for reversing drug resistance (Bhakat et al. 2008). Targeting the redox function of APE1 represents a novel approach in the development of a cancer therapeutic agent, as selectively blocking this function seems to lead to cytostatic growth without causing cell death. Finding specific small-molecule inhibitors that can block either of APE1’s repair or redox functions, but not both, will help determine how to best modify its function in treating different diseases. A unique compound called E3330 has single-agent cancer-cell killing abilities in cancer cell lines including ovarian, colon, lung, breast, brain, pancreatic, prostate and multiple myeloma cancers. It is significant that E3330 does not produce cell killing in normal cells, which implies a particular redox role of APE1 in cancer survival, but not “normal” cell survival. This suggests that a “therapeutic window” exists, as the results of the studies of E3330 demonstrate that it has a greater effect on differentiated endothelial tumor cells versus bone marrow stem cells (Zou et al. 2009). The collective data on E3330 suggest that APE1’s redox function is a promising target for cancer treatment Tables 7.2 and 7.3. Table 7.2  APE1’s relationship between redox activities and DNA repair APE1’s Influence on Redox Regulation of DNA Repair Pathways PATHWAY WHAT APE1 ACTS UPON DR Reduces p53, which dictates AGT’s level of activity, depending on the type of cytotoxic stress present Reduces NFkB, which appears to regulate AGT BER Reduces p53 to optimize its DNA-binding ability; which, in turn, influences DNA polymerase b activity NER Reduces p53 so it can activate DDB2 and XPC to recognize DNA damage in the GGR sub-pathway Reduces AP-1 so it can regulate the expression of several proteins, including those that repair DNA-cisplatin adducts MMR Reduces p53 so it can enable MSH2 to complex and recognize DNA damage Reduction of p53 also enables it to activate MLH1 and PMS2 Reduces AP-1 so it can regulate the expression of several proteins, including those that repair DNA-cisplatin adducts Stabilizes HIF-1a so it can activate HIF-1, which, in turn, inhibits NLH1 and MSH2 during hypoxia HR Reduces p53 so it can modulate RAD51 expression Stabilizes HIF-1a so it can activate HIF-1, which, in turn, inhibits BRCA1, BRCA2, and RAD51 during hypoxia NHEJ Reduces p53 so it can regulate the expression of RecQ helicases WRN and RecQ4

Is reduced by

Reduce(s)

Proximity of residues How reaction resolution occurs

TrxR+FADH2+NADPH

N-terminal cys is the nucleophile; attacks the target protein disulfide; creates a transient mixed disulfide intermediate, this is reduced by the C-terminal active site Cys, generat-ing a dithiol in the target protein and a disulfide in thioredoxin Disulfide bonds of signaling and regulatory proteins

Unknown

Disulfide bonds of signaling and regulatory proteins Thioredoxin (for redox activity; GADPH reduces APE1 for its endonuclease activity)

Close

Far apart

Table 7.3  Comparison of APE1’s redox features to other redox systems Redox system Feature APE1 Trx Cys65 Cys32+Cys35 Residues required for activity Location of active Near the N-terminus; Surface accessible; near the C-terminus residue(s) buried

Structural unfolding in the area of the thiolate near the C-terminus places it in close proximity to the nucleophilic thiolate

Hydrogen peroxide

Reduces the target protein disulfide; creates a transient mixed disulfide intermediate, GRX is oxidized; dithiol created in the protein Disulfide bonds of signaling and regulatory proteins Glutathione+glutathione reductase+NADPH

Sulfiredoxin

1 near the N-terminus; other near the C-terminus Far apart

N-terminus

Close

Peroxiredoxins 1 or 2 Cys (depends on which Prx type)

GRX/GSH Cys22+Cys25

7  DNA Repair and Redox Signaling 161

162

M.R. Kelley et al.

Conclusions Redox regulation has an important role in DNA repair, not only directly with the APE1 protein but also through other redox signaling and cellular redox status changes. APE1 has been a major focus of this chapter since it is clearly the DNA repair protein most directly involved in redox control and signaling. Additionally, we discuss recent data demonstrating a new role of APE1 with its involvement in angiogenesis and potential role in the tumor microenvironment. We expect that future research in this area will bring additional insights into the importance or redox regulation in DNA repair, particularly as more interest if brought to bear on this topic. These new studies will likely result in the identification of other proteins that also play important roles in redox regulation and DNA repair. Acknowledgments  Financial support for this work was provided by the National Institutes of Health, National Cancer Institute CA106298, CA114571 and CA121168 to M.R.K., CA114571 to M.M.G. and the Riley Children’s Foundation to M.R.K., and National Cancer Institute CA122298 and to M.L.F.

References Abate C, et al (1990) Expression and purification of the leucine zipper and DNA-binding domains of Fos and Jun: both Fos and Jun contact DNA directly. Proc Natl Acad Sci USA 87(3):1032–1036. Abate C, et  al (1990) Redox regulation of fos and jun DNA-binding activity in  vitro. Science 249(4973):1157–1161. Adimoolam S, Ford JM(2002) p53 and DNA damage-inducible expression of the xeroderma pigmentosum group C gene. Proc Natl Acad Sci USA 99(20):12985–12990. Aiyer HS, Kichambare S, Gupta RC (2008) Prevention of oxidative DNA damage by bioactive berry components. Nutr Cancer 60(Suppl 1):36–42. Akamatsu Y, et  al (1997) Redox regulation of the DNA binding activity in transcription factor PEBP2. J Biol Chem 272:14497–14500. Akterin S, et al (2006) Involvement of glutaredoxin-1 and thioredoxin-1 in beta-amyloid toxicity and Alzheimer’s disease. Cell Death Differ 13(9):1454–1465. Allinson SL, Dianova II, Dianov GL (2001) DNA polymerase beta is the major dRP lyase involved in repair of oxidative base lesions in DNA by mammalian cell extracts. EMBO J 20(23):6919–6926. Altieri F, et al (2008) DNA damage and repair: from molecular mechanisms to health implications. Antioxid Redox Signal 10(5):891–938. Arcangeli L, et al (1997) Site- and strand-specific mismatch repair of human H-ras genomic DNA in a mammalian cell line. Carcinogenesis 18(7):1311–1318. Arias-Lopez C, et al (2006) p53 modulates homologous recombination by transcriptional regulation of the RAD51 gene. EMBO Rep 7(2):219–224. Arner ES, Holmgren A (2000) Physiological functions of thioredoxin and thioredoxin reductase. Eur J Biochem 267(20):6102–6109. Arner ES, Holmgren A (2006) The thioredoxin system in cancer. Semin Cancer Biol 16(6):420–426. Bapat A, et al (2009) Going ape as an approach to cancer therapeutics. Antioxid Redox Signal 11(3):651–668.

7  DNA Repair and Redox Signaling

163

Barnes T, et  al (2009) Identification of Apurinic/apyrimidinic endonuclease 1 (APE1) as the endoribonuclease that cleaves c-myc mRNA. Nucl Acids Res 37(12):3946–3958. Beckman RA, Loeb LA (2006) Efficiency of carcinogenesis with and without a mutator mutation. Proc Natl Acad Sci USA 103(38):14140–14145. Berndt C, Lillig CH, Holmgren A (2008) Thioredoxins and glutaredoxins as facilitators of protein folding. Biochim Biophys Acta (BBA): Mol Cell Res 1783(4):641–650. Bhakat KK, Mantha AK, Mitra S (2009) Transcriptional regulatory functions of mammalian AP-endonuclease (APE1/Ref-1), an essential multifunctional protein. Antioxid Redox Signal 11:621–638. Biaglow JE, Miller RA (2005) The thioredoxin reductase/thioredoxin system: novel redox targets for cancer therapy. Cancer Biol Ther 4(1):6–13. Bindra RS, Glazer PM (2007) Repression of RAD51 gene expression by E2F4/p130 complexes in hypoxia. Oncogene 26(14):2048–2057. Bindra RS, Glazer PM (2007), Co-repression of mismatch repair gene expression by hypoxia in cancer cells: role of the Myc/Max network. Cancer Lett 252(1):93–103. Bindra RS, et al (2004) Down-regulation of Rad51 and decreased homologous recombination in hypoxic cancer cells. Mol Cell Biol 24(19):8504–8518. Bindra RS, et al (2005) Alterations in DNA repair gene expression under hypoxia: elucidating the mechanisms of hypoxia-induced genetic instability. Ann N Y Acad Sci 1059:184–195. Bindra RS, Crosby ME, Glazer PM (2007) Regulation of DNA repair in hypoxic cancer cells. Cancer Metastasis Rev 26(2):249–260. Bishop AJ, et  al (2003) Atm-, p53-, and Gadd45a-deficient mice show an increased frequency of homologous recombination at different stages during development. Cancer Res63(17):5335–5343. Bobola MS, et al (2005) Apurinic/apyrimidinic endonuclease activity is associated with response to radiation and chemotherapy in medulloblastoma and primitive neuroectodermal tumors. Clin Cancer Res 11(20):7405–7414. Bravard A, et al (2009) Oxidation status of human OGG1-S326C polymorphic variant determines cellular DNA repair capacity. Cancer Res 69(8):3642–3649. Bristow RG, Hill RP (2008) Hypoxia and metabolism. Hypoxia, DNA repair and genetic instability. Nat Rev Cancer 8(3):180–192. Burke-Gaffney A, Callister ME, Nakamura H (2005) Thioredoxin: friend or foe in human disease? Trends Pharmacol Sci 26(8):398–404. Burma S, Chen BP, Chen DJ (2006) Role of non-homologous end joining (NHEJ) in maintaining genomic integrity. DNA Repair (Amst) 5(9–10):1042–1048. Cao X, et al (2002) Oxidoreductive modification of two cysteine residues in paired domain by Ref-1 regulates DNA-binding activity of Pax-8. Biochem Biophys Res Commun 297(2):288–293. Chen DS, Olkowski ZL (1994) Biological responses of human apurinic endonuclease to radiationinduced DNA damage. Ann N Y Acad Sci 726:306–308. Chen DS, Herman T, Demple B (1991) Two distinct human DNA diesterases that hydrolyze 3’-blocking deoxyribose fragments from oxidized DNA. Nucl Acids Res 19(21):5907–5914. Chen J, Sadowski I (2005) Identification of the mismatch repair genes PMS2 and MLH1 as p53 target genes by using serial analysis of binding elements. Proc Natl Acad Sci USA 102(13):4813–4818. Chew EH, et al (2008) Thioredoxin reductase inhibition by antitumor quinols: a quinol pharmacophore effect correlating to antiproliferative activity. FASEB J 22(6):2072–2083. Christmann M, et  al (2003) Mechanisms of human DNA repair: an update. Toxicology 193(1–2):3–34. Damia G, D’Incalci M (2007) Targeting DNA repair as a promising approach in cancer therapy. Eur J Cancer 43(12):1791–1801. Datta K, et  al (2004) Sensitizing glioma cells to cisplatin by abrogating the p53 response with antisense oligonucleotides. Cancer Gene Ther 11(8):525–531. David SS, Williams SD (1998) Chemistry of glycosylases and endonucleases involved in baseexcision repair. Chem Rev 98(3):1221.

164

M.R. Kelley et al.

De Bont R, van Larebeke N (2004) Endogenous DNA damage in humans: a review of quantitative data. Mutagenesis 19(3):169–185. Demple B, Harrison L (1994) Repair of oxidative damage to DNA: enzymology and biology. Ann Rev Biochem 63:915–948. Di Maso V, et al (2007) Subcellular localization of APE1/Ref-1 in human hepatocellular carcinoma: possible prognostic significance. Mol Med 13(1–2):89–96. Doetsch PW, Cunningham RP (1990) The enzymology of apurinic/apyrimidinic endonucleases. Mutat Res 236(2–3):173–201. Eker AP, et al (2009) DNA repair in mammalian cells: Direct DNA damage reversal: elegant solutions for nasty problems. Cell Mol Life Sci 66(6):968–980. Ema M, et al (1999) Molecular mechanisms of transcription activation by HLF and HIF1alpha in response to hypoxia: their stabilization and redox signal-induced interaction with CBP/p300. EMBO J 18(7):1905–1914. Essers J, et al (2000) Homologous and non-homologous recombination differentially affect DNA damage repair in mice. EMBO J 19(7):1703–1710. Evans AR, Limp-Foster M, Kelley MR (2000) Going APE over ref-1. Mutat Res 461(2):83–108. Farmer H, et al (2005) Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature 434(7035):917–921. Fernandes AP, Holmgren A (2004) Glutaredoxins: glutathione-dependent redox enzymes with functions far beyond a simple thioredoxin backup system. Antioxid Redox Signal 6(1):63–74. Fishel ML, Kelley MR (2007) The DNA base excision repair protein Ape1/Ref-1 as a therapeutic and chemopreventive target. Mol Aspects Med 28(3–4):375–395. Fishel ML, et al (2007) Manipulation of base excision repair to sensitize ovarian cancer cells to alkylating agent temozolomide. Clin Cancer Res 13(1):260–267. Fishel ML, et al (2008) Knockdown of the DNA repair and redox signaling protein Ape1/Ref-1 blocks ovarian cancer cell and tumor growth. DNA Repair (Amst) 7(2):177–186. Fishel R, Wilson T (1997) MutS homologs in mammalian cells. Curr Opin Genet Dev 7(1):105–113. Fleck O and O. Nielsen (2004) DNA repair. J Cell Sci 117(Pt 4):515–517. Forman-Kay JD, Clore GM, Gronenborn AM (1992) Relationship between electrostatics and redox function in human thioredoxin: characterization of pH titration shifts using two-dimensional homo- and heteronuclear NMR. Biochemistry 31(13):3442–3452. Forsythe JA, et al (1996) Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. Mol Cell Biol 16(9):4604–4613. Friedberg EC (2001) How nucleotide excision repair protects against cancer. Nat Rev Cancer 1(1):22–33. Fung H, Demple B (2005) A vital role for Ape1/Ref1 protein in repairing spontaneous DNA damage in human cells. Mol Cell 17(3):463–470. Georgiadis M, et al.(2008) Evolution of the redox function in mammalian apurinic/apyrimidinic. Mutat Res 643(1–2):54–63. Giles NM, Giles GI, Jacob C (2003) Multiple roles of cysteine in biocatalysis. Biochem Biophys Res Commun 300(1):1–4. Grombacher T, Eichhorn U, Kaina B (1998) P53 is involved in regulation of the DNA repair gene O-6-methylguanine-DNA methyltransferase (MGMT) by DNA damaging agents. Oncogene 17(7):845–851. Guo HH, Loeb LA (2003) Tumbling down a different pathway to genetic instability. J Clin Invest 112(12):1793–1795. Guo Y, et al (2008) Granzyme K degrades the redox/DNA repair enzyme Ape1 to trigger oxidative stress of target cells leading to cytotoxicity. Mol Immunol 45(8):2225–2235. Hainaut P, Milner J (1993) Redox modulation of p53 conformation and sequence-specific DNA binding in vitro. Cancer Res 53(19):4469–4473. Handel ML, et al (1995) Inhibition of AP-1 binding and transcription by gold and selenium involving conserved cysteine residues in Jun and Fos. Proc Natl Acad Sci USA 92(10):4497–4501. Hanson S, Kim E, Deppert W (2005) Redox factor 1 (Ref-1) enhances specific DNA binding of p53 by promoting p53 tetramerization. Oncogene 24(9):1641–1647.

7  DNA Repair and Redox Signaling

165

He F, et  al (2003) DNA polymerase beta gene expression: the promoter activator CREB-1 is upregulated in Chinese hamster ovary cells by DNA alkylating agent-induced stress. Biol Chem 384(1):19–23. Helton ES, Chen X (2007) p53 modulation of the DNA damage response. J Cell Biochem 100(4):883–896. Herring CJ, et al (1998) Levels of the DNA repair enzyme human apurinic/apyrimidinic endonuclease (APE1, APEX, Ref-1) are associated with the intrinsic radiosensitivity of cervical cancers. Br J Cancer 78(9):1128–1133. Hess J, Angel P, Schorpp-Kistner M (2004) AP-1 subunits: quarrel and harmony among siblings. J Cell Sci 117(Pt 25):5965–5973. Hirota K, Matsui M, Yodoi J (1997) AP-1 transcriptional activity is regulated by a direct association between thioredoxin and Ref-1. Proc Natl Acad Sci USA 94(8):3633–3638. Hirota K, et al (1999) Distinct roles of thioredoxin in the cytoplasm and in the nucleus. A two-step mechanism of redox regulation of transcription factor NF-kappaB. J Biol Chem 274(39):27891–27897. Hoeijmakers JH (2001) DNA repair mechanisms. Maturitas 38(1):17–22; discussion 22–23. Hoeijmakers JH (2001) Genome maintenance mechanisms for preventing cancer. Nature 411(6835):366–374. Hofmann B, Hecht HJ, Flohe L (2002) Peroxiredoxins. Biol Chem 383(3–4):347–364. Hollander MC, et al (2005) Deletion of XPC leads to lung tumors in mice and is associated with early events in human lung carcinogenesis. Proc Natl Acad Sci USA 102(37):13200–13205. Holmgren A (1989) Thioredoxin and glutaredoxin systems. J Biol Chem 264(24):13963–13966. Holmgren A (1995) Thioredoxin structure and mechanism: conformational changes on oxidation of the active-site sulfhydryls to a disulfide. Structure 3(3):239–243. Hol WG (1985) The role of the alpha-helix dipole in protein function and structure. Prog Biophys Mol Biol 45(3):149–195. Hromas R, et  al (2008) The human set and transposase domain protein Metnase interacts with DNA Ligase IV and enhances the efficiency and accuracy of non-homologous end-joining. DNA Repair (Amst) 7(12):1927–1937. Huang LE, et  al (1996) Activation of hypoxia-inducible transcription factor depends primarily upon redox-sensitive stabilization of its alpha subunit. J Biol Chem 271(50):32253–32259. Huang RP, Adamson ED (1993) Characterization of the DNA-binding properties of the early growth response-1 (Egr-1) transcription factor: evidence for modulation by a redox mechanism. DNA & Cell Biol 12(3):265–273. Hwang BJ, et al (1999) Expression of the p48 xeroderma pigmentosum gene is p53-dependent and is involved in global genomic repair. Proc Natl Acad Sci USA 96(2):424–428. Izumi T, et al (2005) Two essential but distinct functions of the mammalian abasic endonuclease. Proc Natl Acad Sci USA 102(16):5739–5743. Jacob C, et al (2003) Sulfur and selenium: the role of oxidation state in protein structure and function. Angew Chem Int Ed Engl 42(39):4742–4758. Jayaraman L, et al (1997) Identification of redox/repair protein Ref-1 as a potent activator of p53. Genes and Develop 11:558–570. Jiang BH, et  al (1996) Dimerization, DNA binding, and transactivation properties of hypoxiainducible factor 1. J Biol Chem 271(30):17771–17778. Jiang Y, et al (2009) Role of APE1 in differentiated neuroblastoma SH-SY5Y cells in response to oxidative stress: use of APE1 small molecule inhibitors to delineate APE1 functions. DNA Repair 8:1273–1282. Jiricny J (2006) The multifaceted mismatch-repair system. Nat Rev Mol Cell Biol 7(5):335–346. Kaina B, et  al (2007) MGMT: key node in the battle against genotoxicity, carcinogenicity and apoptosis induced by alkylating agents. DNA Repair (Amst) 6(8):1079–1099. Kelley MR, Fishel ML (2008) DNA repair proteins as molecular targets for cancer therapeutics. Anticancer Agents Med Chem 8(4):417–425. Kondo N, et al (2006) Redox regulation of human thioredoxin network. Antioxid Redox Signal 8(9–10):1881–1890.

166

M.R. Kelley et al.

Koshiji M, et al (2005) HIF-1alpha induces genetic instability by transcriptionally downregulating MutSalpha expression. Mol Cell 17(6):793–803. Kruman, II, et al (2004) Cell cycle activation linked to neuronal cell death initiated by DNA damage. Neuron 41(4):549–561. Lando D, et al (2000) A redox mechanism controls differential DNA binding activities of hypoxiainducible factor (HIF) 1alpha and the HIF-like factor. J Biol Chem 275(7):4618–4627. Langie SA, et  al (2007) The role of glutathione in the regulation of nucleotide excision repair during oxidative stress. Toxicol Lett 168(3):302–309. Lau JP, et al (2004) Effects of gemcitabine on APE/ref-1 endonuclease activity in pancreatic cancer cells, and the therapeutic potential of antisense oligonucleotides. Br J Cancer 91(6):1166–1173. Lavon I, et al (2007) Novel Mechanism whereby nuclear factor {kappa}B mediates DNA damage repair through regulation of O6-methylguanine-DNA-methyltransferase. Cancer Res 67(18):8952–8959. Lee SH, et  al (2005) The SET domain protein Metnase mediates foreign DNA integration and links integration to nonhomologous end-joining repair. Proc Natl Acad Sci USA 102(50):18075–18080. Lillig CH, Holmgren A (2007) Thioredoxin and related molecules – from biology to health and disease. Antioxid Redox Signal 9(1):25–47. Lillig CH, Berndt C, Holmgren A (2008) Glutaredoxin systems. Biochim Biophys Acta 1780(11):1304–1317. Li SL, et al (2001) Transcriptional regulation of the p67phox gene: role of AP-1 in concert with myeloid-specific transcription factors. J Biol Chem 276(42):39368–39378. Liu F, Fu Y, Meyskens FL, Jr (2008) MiTF regulates cellular response to reactive oxygen species through transcriptional regulation of APE-1//Ref-1. J Invest Dermatol 129:422–431. Liu Z, et al (2009) Inhibition of thioredoxin reductase by mansonone F analogues: Implications for anticancer activity. Chem Biol Interact 117(1):48–57. Loboda A, et al (2009) HIF-1 attenuates Ref-1 expression in endothelial cells: Reversal by siRNA and inhibition of geranylgeranylation. Vascul Pharmacol 51(2–3):133–139. Lopez de Saro FJ, et al (2006) The beta sliding clamp binds to multiple sites within MutL and MutS. J Biol Chem 281(20):14340–14349. Luo M, et al (2008) Role of the multifunctional DNA repair and redox signaling protein Ape1/ Ref-1 in cancer and endothelial cells: Small molecule inhibition of Ape1’s redox function. Antioxid Redox Signal 10(11):1853–1867. Luo M, et al (2010) Redox regulation of DNA repair: implications for human health and cancer therapeutic development. Antioxid Redox Signal 12:1247–1269. Lu X, Lozano G, Donehower LA (2003) Activities of wildtype and mutant p53 in suppression of homologous recombination as measured by a retroviral vector system. Mutat Res 522(1–2):69–83. Madhusudan S, Hickson ID (2005) DNA repair inhibition: a selective tumour targeting strategy. Trends Mol Med 11(11):503–511. Madhusudan S, et al (2005) Isolation of a small molecule inhibitor of DNA base excision repair. Nucleic Acids Res 33(15):4711–4724. Mathers J, Coxhead J, Tyson J (2007) Nutrition and DNA Repair – potential molecular mechanisms of action. Curr Cancer Drug Targets 7(5):425–431. McNeill DR, Wilson DM 3rd (2007) A dominant-negative form of the major human abasic endonuclease enhances cellular sensitivity to laboratory and clinical DNA-damaging agents. Mol Cancer Res 5(1):61–70. Meng AX, et al (2005) Hypoxia down-regulates DNA double strand break repair gene expression in prostate cancer cells. Radiother Oncol 76(2):168–176. Merluzzi S, et al (2008) APE/Ref-1 makes fine-tuning of CD40-induced B cell proliferation. Mol Immunol 45(14):3731–3739. Mihaylova VT, et al (2003) Decreased expression of the DNA mismatch repair gene Mlh1 under hypoxic stress in mammalian cells. Mol Cell Biol 23(9):3265–3273.

7  DNA Repair and Redox Signaling

167

Moore DH, et al (2000) Alterations in the expression of the DNA repair/redox enzyme APE/ref-1 in epithelial ovarian cancers. Clin Cancer Res 6(2):602–609. Mukherjee A, Martin SG (2008) The thioredoxin system: a key target in tumour and endothelial cells. Br J Radiol 81(1):S57–S68. Nakamura H (2005) Thioredoxin and its related molecules: update 2005. Antioxid Redox Signal 7(5–6):823–828. Nakamura H, Nakamura K, Yodoi J (1997) Redox regulation of cellular activation. Annu Rev Immunol 15:351–369. Ono Y, et al (1994) Stable expression in rat glioma cells of sense and antisense nucleic acids to a human multifunctional DNA repair enzyme, APEX nuclease. Mutat Res 315(1):55–63. Powis G, Montfort WR (2001) Properties and biological activities of thioredoxins. Annu Rev Pharmacol Toxicol 41:261–295. Raffel J, et al (2003) Increased expression of thioredoxin-1 in human colorectal cancer is associated with decreased patient survival. J Lab Clin Med 142(1):46–51. Rafferty JA, et al (1996) Induction of murine O6-alkylguanine-DNA-alkyltransferase in response to ionising radiation is p53 gene dose dependent. Oncogene 12(3):693–697. Reed AM, Fishel ML, Kelley MR (2009) Small-molecule inhibitors of proteins involved in base excision repair potentiate the anti-tumorigenic effect of existing chemotherapeutics and irradiation. Future Oncol 5(5):713–726. Scherer SJ, et al (2000) p53 and c-Jun functionally synergize in the regulation of the DNA repair gene hMSH2 in response to UV. J Biol Chem 275(48):37469–37473. Schwartz EI, et al (2007) Cell cycle activation in postmitotic neurons is essential for DNA repair. Cell Cycle 6(3):318–329. Sengupta S, Harris CC (2005) p53: traffic cop at the crossroads of DNA repair and recombination. Nat Rev Mol Cell Biol 6(1):44–55. Sengupta S, et  al (2005) Tumor suppressor p53 represses transcription of RECQ4 helicase. Oncogene 24(10):1738–1748. Shinohara ET, et al (2005) DNA-dependent protein kinase is a molecular target for the development of noncytotoxic radiation-sensitizing drugs. Cancer Res 65(12):4987–4992. Smith ML, Seo YR (2002) p53 regulation of DNA excision repair pathways. Mutagenesis 17(2):149–156. Song YJ, et al (2008) Tat-APE1/ref-1 protein inhibits TNF-[alpha]-induced endothelial cell activation. Biochem Biophys Res Commun 368(1):68–73. Sung P, Klein H (2006) Mechanism of homologous recombination: mediators and helicases take on regulatory functions. Nat Rev Mol Cell Biol 7(10):739–750. Tan T, Chu G (2002) p53 Binds and activates the xeroderma pigmentosum DDB2 gene in humans but not mice. Mol Cell Biol 22(10):3247–3254. Tell G, et  al (2000) An ‘environment to nucleus’ signaling system operates in B lymphocytes: redox status modulates BSAP/Pax-5 activation through Ref-1 nuclear translocation. Nucleic Acids Res 28(5):1099–1105. Tell G, et al (2005) The intracellular localization of APE1/Ref-1: more than a passive phenomenon? Antioxid Redox Signal 7(3–4):367–384. Tell G, et al (2009) The many functions of APE1/Ref-1: not only a DNA repair enzyme. Antioxid Redox Signal 11(3):601–620. To KK, et  al (2006) The phosphorylation status of PAS-B distinguishes HIF-1alpha from HIF2alpha in NBS1 repression. EMBO J 25(20):4784–4794. Tonissen KF, Di Trapani G (2009) Thioredoxin system inhibitors as mediators of apoptosis for cancer therapy. Mol Nutr Food Res 53(1):87–103. Ueno M, et al (1999) Thioredoxin-dependent redox regulation of p53-mediated p21 activation. J Biol Chem 274(50):35809–35815. Vascotto C, et al (2009) APE1/Ref-1 interacts with NPM1 within nucleoli and plays a role in the rRNA quality control process. Mol Cell Biol 29(7):1834–1854.

168

M.R. Kelley et al.

Walker LJ, et  al (1993) Identification of residues in the human DNA repair enzyme HAP1 (Ref-1) that are essential for redox regulation of Jun DNA binding. Mol Cell Biol 13(9):5370–5376. Walker LJ, et al (1994) A role for the human DNA repair enzyme HAP1 in cellular protection against DNA damaging agents and hypoxic stress. Nucl Acids Res 22(23):4884–4889. Wang D, Luo M, Kelley MR (2004) Human apurinic endonuclease 1 (APE1) expression and prognostic significance in osteosarcoma: enhanced sensitivity of osteosarcoma to DNA damaging agents using silencing RNA APE1 expression inhibition. Mol Cancer Ther 3(6):679–686. Weichsel A, et al (1996) Crystal structures of reduced, oxidized, and mutated human thioredoxins: evidence for a regulatory homodimer. Structure 4(6):735–751. Wood RD, Mitchell M, Lindahl T (2005) Human DNA repair genes, 2005. Mutat Res 577(1–2):275–283. Wood ZA, Poole LB, Karplus PA (2003) Peroxiredoxin evolution and the regulation of hydrogen peroxide signaling. Science 300(5619):650–653. Xanthoudakis S, Curran T (1992) Identification and characterization of Ref-1, a nuclear protein that facilitates AP-1 DNA-binding activity. EMBO J 11(2):653–665. Xanthoudakis S, Curran T (1996) Redox regulation of Ap-1: a link between transcription factor signaling and DNA repair. Adv Exp Med Biol 387:387. Xanthoudakis S, et al (1992) Redox activation of Fos-Jun DNA binding activity is mediated by a DNA repair enzyme. EMBO J 11(9):3323–3335. Xanthoudakis S, et al (1996) The redox/DNA repair protein, Ref-1, is essential for early embryonic development in mice. Proc Natl Acad Sci USA 93(17):8919–8923. Xu J, Morris GF (1999) p53-mediated regulation of proliferating cell nuclear antigen expression in cells exposed to ionizing radiation. Mol Cell Biol 19(1):12–20. Xu Y, et  al (1997) The apurinic/apyrimidinic endonuclease (APE/ref-1) DNA repair enzyme is elevated in premalignant and malignant cervical cancer. Anticancer Res 17(5B):3713–3719. Yamabe Y, et al (1998) Sp1-mediated transcription of the Werner helicase gene is modulated by Rb and p53. Mol Cell Biol 18(11):6191–6200. Yamawaki H, Berk BC (2005) Thioredoxin: a multifunctional antioxidant enzyme in kidney, heart and vessels. Curr Opin Nephrol Hypertens 14(2):149–153. Zaky A, et al (2008) Regulation of the human AP-endonuclease (APE1/Ref-1) expression by the tumor suppressor p53 in response to DNA damage. Nucl Acids Res 36(5):1555–1566. Zou G-M, Maitra A (2008) Small-molecule inhibitor of the AP endonuclease 1/REF-1 E3330 inhibits pancreatic cancer cell growth and migration. Mol Cancer Ther 7(7):2012–2021. Zou GM, et  al (2009) The Ape-1/Ref-1 redox antagonist E3330 inhibits the growth of tumor endothelium and endothelial progenitor cells: therapeutic implications in tumor angiogenesis. J Cell Physiol 219(1):209–218. Zurer I, et  al (2004) The role of p53 in base excision repair following genotoxic stress. Carcinogenesis 25(1):11–19.

Chapter 8

Cancer Stem Cells and Microenvironment Mario Federico and Antonio Giordano

Abstract  The theory of the cancer stem cell (CSC) is fairly recent and has both challenged and disrupted the previous understandings of cancer biology. From the initial findings of cancer-driving cellular sub-populations, the interest in the CSC theory has flourished. Here we discuss the biology behind both embryonic and adult stem cells and how this biology is the basis for our understanding of CSCs. Furthermore, we elaborate on findings demonstrating the importance of the microenvironment on the stem cell and how changes in the microenvironment alone may direct stem cell differentiation or possibly tumorigenesis. Cancer research has taken a new fork in the road towards the involvement of CSCs and this road has been productive. However, the challenge remains in understanding and identifying CSCs independently from the embryonic and adult stem cell models of the past. Moving beyond the preconceived understanding of stem cells will allow researchers to be able to fully explore the complexity and role of CSCs in cancer and hopefully provide the knowledge necessary to combat cancer from that first initiating cell.

Introduction The traditional model explaining tumor progression assumes that the tumor cell, through a series of stochastic events, acquires increased proliferative capability because of the deregulation of oncogenes and/or tumor suppressor genes. Once the cancerous phenotype is established, every cell within the tumor has the same capacity to proliferate extensively with no differences in the intrinsic tumorigeneity. The occurrence of additional stochastic genetic events (conferring to the tumor cell an increased malignant potential) will lead to the substitution of the former cancer clone with a more aggressive clone, otherwise known as clonal selection. A. Giordano (*) Biology Department, Sbarro Health Research Organization, Temple University, Philadelphia, PA, USA e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_8, © Springer Science+Business Media, LLC 2010

169

170

M. Federico and A. Giordano

In 1997 Bonnet and Dick (1997) challenged this paradigm. They isolated a subpopulation of CD34+/CD38− leukemic precursors from human AML and demonstrated that only those cells were able to initiate leukemia when engrafted into NOD/SCID mice, whereas the other fractions of leukemic blasts representing the vast majority of all leukemic cells (CD34+, CD38+, or CD34−) were unable to do so. Moreover, they found that engrafted CD34+/CD38− leukemic precursors could be serially transplanted into secondary recipients, thus providing evidence of an intrinsic self-renewal capability. After this observation, several other groups also found tumorigenic self-renewal stem-like populations in a wide spectrum of solid tumors, and nice reviews have been published so far (Kakarala and Wicha 2008; Dirks 2008; Maitland and Collins 2008; Sell and Leffert 2008; Boman and Huang 2008; Takaishi et  al. 2008; Peacock and Watkins 2008; Huff and Matsui 2008; Lee et al. 2008; Prince and Ailles 2008). Self-renewal and differentiation capability are the hallmarks of staminality, a concept that traditionally has been associated with totipotent embryonic cells (embryonic stem cells, ESCs), which possess characteristics such as asymmetric cellular division and the ability to differentiate into all three germ layers originating all the different cell types within the body. Stem-like cellular populations also exist in adult organisms and are defined adult stem cells (ASCs). ASCs are immature, undifferentiated, and functionally unspecialized cells, which exist in small numbers in nearly all post-embryonic tissues. ASC populations are thought to give rise to multiple, more differentiated cell types, maintaining a long-term capacity for selfrenewal, and so regulating homeostatic tissue regeneration. In contrast to ESCs, ASCs lack the capacity to differentiate into the three germ layers, are capable of regenerating a cellular population of a specific tissue type, and maintain asymmetric cellular division. ASCs are characterized as being in a state of relative proliferative quiescence, with a high dependence on the microenvironment in which they reside: the so-called stem cell niche. Under the proper conditions, they can exit from the quiescent state to obtain the proliferative potential necessary for tissue regeneration or wound repair. Bonnet and Dick sustained the hypothesis that tumors, similar to normal adult tissues, are constituted with a heirarchy of cell types (Fig. 8.1). Stem or stem-like cells (heirin referred to as cancer stem cells or CSCs) are a relatively small population within this heirarchy and are the only population that maintains the ability to initiate tumorigenesis by undergoing self-renewal and differentiation. The differentiation of this specialized population leads to the loss in self-renewal capacity and physically creates the vast majority of cells found within a tumor. This change of landscape is a radical shift from the traditional stochastic model. First of all because it assumes that not any cell can originate a tumor but just a stem cell, which through an acquired deregulation of the self-renewal process becomes tumorigenic. The second reason is that it postulates a histological hierarchy within the tumor, further implying (by analogy with ESCs and ASCs) a series of mechanisms that control CSCs along their differentiation path. Since stem cells generally have a higher resistance to chemo- and radio-therapy than committed cells, CSC

8  Cancer Stem Cells and Microenvironment

171

Fig. 8.1  The CSC hypothesis

theory has been advocated to explain clinical relapses after cancer therapy. In fact, chemo- and radio-therapy normally produce a significant shrinkage of the tumor mass acting mainly on the “mature” cancer cells with limited effect on CSCs or tumor progenitors. The latter is postulated to be the engine of tumor re-formation, leading to subsequent clinical relapse. This hypothesis has been recently validated in a breast cancer clinical trial, in which patients receiving cytotoxic chemotherapy displayed a marked increase in breast CSC frequency in the residual tumor, while a targeted anti-CSC therapeutic stabilized the CSC population (Li et  al. 2008). Finally, from the cellular biology point of view, the CSC paradigm challenges our knowledge of the molecular mechanisms that underlie tumorigeneity. What are the molecular mechanisms behind self-renewal and how is this maintained over time? What stimuli are necessary for differentiation to occur and how are these stimuli limited to only a portion of the population? And how does the environment around the stem cell population affect these events? In the last few years, an exponentially growing number of publications have addressed these issues; however, the majority of knowledge on the general biological mechanisms of staminality still derives from the studies on ESCs and ASCs. Even if we can “phenomenologically” define CSCs through their attitude toward self-renewal and differentiation into a “mature” cancer phenotype, we are still dependent upon the ESC and ASC knowledge to guide us through this novel research field in oncology.

172

M. Federico and A. Giordano

ESCs: A Prototype Model of Stem Cell Biology Thomson and Gearhart isolated and characterized the initial human ESCs (hESCs) from the inner cell mass (ICM) of the blastocyst, noting the self-renewing and differentiation capacity of the cells in  vitro (Shamblott et  al. 1998; Thomson et  al. 1998). These cells, through further characterization, were found to express cell surface markers typical of undifferentiated nonhuman primate ESCs (pESCs) and human embryonic carcinoma cells as was originally described (Shamblott et  al. 1998; Thomson et al. 1998; Martin and Evans 1974, 1975). These specific markers included stage-specific embryonic antigen (SSEA)-1, SSEA-3, SSEA-4, TRA-1– 60, TRA-1–81, alkaline phosphatase activity, and high levels of telomerase activity. Telomerase is a ribonucleoprotein enzyme that preserves the telomeric regions at the ends of chromosomes by de  novo oligonucleotide synthesis (Greider and Blackburn 1987) and is not present in normal diploid somatic cells, which incur shorted telomeres with age, leading to replicative senescence after a finite number of replications (Hayflick 1965; Hayflick and Moorhead 1961; Harley et al. 1990; Allsopp et al. 1992). It has been shown that TRA-1–60 and TRA-1–81 are specific epitopes of a larger membrane-bound protein podocalyxin, which undergoes retinoic acid modification when ESCs differentiate losing its reactivity with the TRA-1–60 and TRA-1–81 antibodies (Schopperle and DeWolf 2007). These characterizations remain to be used to identify stem cells today, along with the expression of the intrinsic transcription factor Oct-4 and in mouse ESCs (mESCs) the constitutive ability to receive extrinsic signals from the cytokine leukemia inhibitory factor (LIF) (Niwa et al. 1998; Nichols et al. 1998; Fuchs and Segre 2000). The specific characteristic of ESCs is to differentiate during development into many different “mature” phenotypes, realizing a sophisticated balance between pluripotency and commitment that allows the timely and spatially appropriate embryo development. Although not completely elucidated, such a unique characteristic relies upon at least four general different mechanisms: a distinctive transcriptional hierarchy, a balanced epigenetic state, a particular cell cycle regulation, and a tight crosstalk with the microenvironment. Transcriptional regulators contribute to the maintenance of pluripotency and three major players have been identified so far in ESCs: POU5FI, NANOG, and SOX2. POU5FI is a member of the POU transcription factor family and is expressed in the mouse embryo from the four to the eight cells stage until the epiblast starts to differentiate. Its presence is essential for the initial development of pluripotency in the ICM and high levels of this transcription factor are restricted to pluripotent lineages (Wobus and Boheler 2005). Mouse embryos lacking POU5FI protein die following implantation because of the lack of ICM (Nichols et al. 1998). POU5FI represses differentiation to the trophoblast; however, appropriate levels are necessary to maintain pluripotency. In fact, a twofold increase in its expression causes differentiation into primitive endoderm and mesoderm, while loss of POU5FI induces the formation of trophectoderm and concomitant loss of pluripotency (Niwa et al. 2000). NANOG homeodomain protein in pre-implantation embryos is restricted to the inner cells of the compacted morula and blastocyst

8  Cancer Stem Cells and Microenvironment

173

and conversely is absent in differentiated cells (Chambers et  al. 2003). High expression levels of NANOG confer constitutive self-renewal capacity to mESCs in the absence of LIF stimulation and can be propagated in serum-free media in the absence of otherwise needed stimuli, such as bone morphogenetic protein (Ying et  al. 2003). NANOG knockout mice develop to the blastocyst stage, but when cultured, the ICM differentiates to parietal endoderm-like cells. From these findings, NANOG appears to be involved not only in pluripotency maintenance but also in the inhibition of the transition to primitice endoderm. Finally, the SOX2 gene is expressed in the pluripotent lineages of early mouse embryos and functions as a transcription factor with an HMG-DNA-binding domain. Sox2−/− mice fail to develop beyond implantation (Avilion et al. 2003) and down regulation of Sox2 is associated with differentiation. A ChIP-on-chip analysis revealed that all three of these factors co-occupy a substantial portion of target genes (Boyer et al. 2005) with an evident overlapping function. The generation of induced pluripotent stem cells using PUI5FI, SOX2, NANOG, and LIN28 along with the generation of pluripotent cells using POU5FI and SOX2 alone with the histone deacetylase inhibitor, valproic acid (Huangfu et al. 2008), have further demonstrated the role of these transcription factors in maintaining pluripotency. The second crucial mechanism that regulates ESC development relies on a precise epigenetic control of gene expression patterns. The term epigenetics specifically refers to heritable genotypic changes that are caused by mechanisms other than changes in the primary DNA sequence. Central to epigenetic control is chromatin, a complex of DNA, histones, and other proteins that make up chromosomes (Misteli 2004, 2007; Meshorer 2007). Histones are subject to several post-translational modifications that occur primarily in their amino terminal tails including methylation, acetylation, phosphorylation, sumoylation, and ubiquitination (Kouzarides 2007). During differentiation, genes regulating pluripotency and self-renewal become progressively silenced, whereas genes controlling cell-type specific functions are gradually turned on. These alterations in gene expression patterns are orchestrated by defined epigenetic programs that are capable of indirectly altering the expression of hundreds to thousands of genes, proving the efficiency of this type of control in establishing cellular identities (Lunyak and Rosenfeld 2008; Cheng et al. 2005; Shukla et al. 2008; Wu and Sun 2006). Based on a genome-wide analyses of chromatin patterns of multiple cell types, several observations have emerged: acetylation of lysine residues on histones H3 and H4 and methylation of lysine 4 of histone H3 (H3K4me), in general, are associated with transcriptionally active chromatin, whereas methylation of histone H3 lysine 9 (H3K9me) and 27 (H3K27me) are hallmarks of transcriptionally silenced chromatin (Kouzarides 2007). Several studies demonstrated that the differentiation of ESCs is accompanied with an increase in histone H3K9me3 and decreased acetylation of histone H3 and H4 (Lee et al. 2004; Meshorer et al. 2006). However, further studies will be necessary to elucidate the role of epigenetics in stem cell regulation. The third major difference between stem cells and somatic cells is found in the basic regulation of the cell cycle. In somatic cells, the cell cycle is controlled mainly by Rb-E2F family complexes, cyclin–cyclin-dependent kinases (Cdks), and

174

M. Federico and A. Giordano

Cdk inhibitors through the INK4a/ARF pathway. Undulations in expression and post-translational modifications of the proteins involved in these pathways result in the control and regulation of the cell cycle. Similarly, mutations or de-regulation of these proteins can lead to uncontrolled cell proliferation, aneuploidy, and genomic instability (Haas et al. 1997; Mumberg et al. 1996). The cell cycle regulatory mechanisms, which differ between somatic cells and ESCs, have been determined using the mESC model in combination with mESCs, representing a pluripotent lineage (mEPLC) (Rathjen et al. 1999). mESCs of late pre-implantation and early post-implantation embryos proliferate at an unusually rapid rate (Solter et  al. 1971). Between 4.5 and 6.0 dpc (days post coitum), the epiblast expands with a generation time of approximately 10 h (Hogan et al. 1994). This increases between 6.5 and 7.0 dpc, where mean generation times are found to be approximately 4.4 h (Hogan et al. 1994; Power and Tam 1993). The cell cycle in mESCs and mEPLCs has been found to curtail G1 and G2 phases with an increased proportion of the cycle, approximately 50–60%, spent in S phase (Savatier et al. 1994; Stead et al. 2002). Under normal somatic cell cycle conditons, Rb/p105, in the hypophosphorylated state, interacts with E2F transcription factors inhibiting the transcription of genes necessary for the progression of the cell cycle through the restriction point (R point). The phosphorylation levels of Rb/p105 are dependent upon the CDK activity present in the cell. Mitogen signaling through the Ras/Raf/mitogen activated protein kinase (MAPK) pathway activates the cyclin D–CDK4/6 complexes, which are believed to initially activate Rb/p105 activity by hypophosphorylating the unphosphorylated protein. To pass the R point of the cell cycle, cyclin E/CDK2 hyperphosphorylates Rb/p105 inhibiting the protein from binding to E2F transcription factors, thus initiating the transcription of genes required in the S phase of the cell cycle. To obtain a cell cycle that is less influenced by mitogen variations, the stem cells appear to adopt a different regulation mechanism as depicted in Fig. 8.2 (Giacinti and Giordano 2006; Kasten 1998). Along with shortened gap phases in the ESC cell cycle, the R point does not seem to regulate the G1–S transition. Stead and collegues found that in both mESCs and mEPLCs, there was a precocious cell cycle-independent expression of CDK2, cyclin A and cyclin E kinase activity (Kasten and Giordano 1998). Furthermore, when CDK2 was suppressed, they found a significant decrease in cell proliferation rate. Instead of CDC2, cyclin B, essential to G2 – M transition, was the only CDK activity that was found to be cell cycle-dependent and E2F target genes were constitutively expressed throughout the cell cycle (Stead et al. 2002). Evidence has also shown a lack in hypophosphorylated Rb/p105, instead findings support the presence of hyperphosphorylated Rb/p105 in mESCs and mEPLCs (Savatier et al. 1994; Fraichard et al. 1995). of cell cycle. There is cell cycle-independent expression of cyclin E – CDK2 maintaining the hyperphosphorylated levels of Rb family member proteins. This results in cell cycle-independent expression of E2F-regulated genes. Cyclin B – CDC2 is the only CDK activity that appears to be regulated by the cell cycle. ESCs have shortened gap phases and an elongated S phase of the cell cycle, with an apparent lack in the R point for G1-S transition

8  Cancer Stem Cells and Microenvironment

175

Fig. 8.2  Specific characteristics of pluripotent cells a) Interplay between the key transcriptional regulators POU5F1, NANOG and SOX2 and their effect on the formation of primitive endoderm and trophectoderm. b) Schematic representation of a nucleosome model and major post-translational modifications which play essential roles in gene expression regulation. c) Cell cycle in somatic cells vs. ESCs. Cell cycle regulation in somatic cells: mitogen signaling through MAPK pathway activates cyclin D – CDK4/6 kinase activity hypophosphorylating Rb family member proteins. Hypophosphorylated Rb family member proteins bind to E2F transcription factors blocking the transcription of E2F-regulated genes. To surpass the R point cyclin E – CDK2 kinase activity is activated hyperphosphorylating Rb family member proteins. Hyperphosphorylated Rb family member proteins are unable to interact with E2F factors, allowing them to activate transcription of genes necessary in the progression of cell cycle. Cell cycle regulation in ESCs as is currently understood: Mitogen signaling through MAPK pathways seems to be irrelevant in the progression

176

M. Federico and A. Giordano

Given the cell-cycle independent expression of cyclin E and CDK2, it would be logical that Rb/p105 would be found in the hyperphosphorylated state, further supporting the absence of the R point in ESC cell cycle progression (Fig. 8.1). Mitogen signaling through the MAPK pathway normally stimulates cell division in somatic cells; however, this signaling when prolonged is a potent inducer of differentiation. mESCs appear to avoid this stimulation by maintaining low levels of cyclin D expression and almost no detectable CDK4 kinase activity (Fluckinger et  al. 2006). This corresponds to the lack in hypophosphorylated Rb/p105 levels previously detected in mESCs. These findings support the absence of early G1 in mESCs, allowing them to avoid the differentiation-inducing effects of MAPK signaling as is found in other cell types. Although the majority of studies thus far have been performed in mESCs, hESCs similarly show a truncation of the G1 phase of the cell cycle; however, not much else is known about cell cycle regulation in hESCs. Interestingly, pESCs behave similarly to mESCs in having cell cycle-independent expression of cyclin E, constitutive hyperphosphorylation of Rb/p105 and serum and MAPK-independent cell cycle progression (Fluckinger et  al. 2006; Becker et  al. 2006). Therefore, it could be inferred that through conserved evolution, hESCs may regulate the cell cycle in a similar fashion. Taken together, these data lead to the hypothesis that the ESC cell cycle is rate-dependent upon high levels of CDK activity, is not regulated by Rb/p105 or E2F gene expression, and lacks the G1 check point and the traditional periodicity found during the somatic cell cycle.

Stem Cells and Microenvironment Soon after the initial isolations and characterizations of hESCs, interest shifted toward understanding the factors involved in their differentiation. For example, if all cells are derived from initial progenitor cells, what directs the timely and spatially appropriate differentiation toward various mature cell types (i.e., glial cells versus adipocytes)? In this sense, it was immediately clear that a crucial role in differentiation was played by the microenvironment. Brüstle et  al. (1999) were among the first to demonstrate in vitro controlled differentiation of hESCs using a series of growth factor combinations, which successfully elicited a reactivity to a monoclonal antibody specific for a membrane epitope typically found on the membranes of glial precursors. They initially grew ES cells in a media that favored the growth of neural precursors. They then exposed cells to the following series of growth factors: (1) basic fibroblast growth factor (FGF2), (2) FGF2 and epidermal growth factor (EGF), and (3) FGF2 and platelet-derived growth factor (PDGF) (Brüstle et al. 1999). The cells maintained in the final growth factor-supplemented media could be stored and kept in culture without further differentiation for many passages. However, as growth factors were removed, cells further differentiated into more specific neural cell types such as oligodendrocytes and astrocytes (Brüstle et  al. 1999). Today, it is well established that ESC microenvironment

8  Cancer Stem Cells and Microenvironment

177

provides a dynamic regulatory niche for stem cell function constituting a finely regulated network of cell–cell communication that is critical to cell behavior and fate determination throughout embryogenesis. The stem cell niche can be defined as the production of local signal(s) and spatial organization of cells receptive to those signals to generate location-dependent control over contrasting cell-fate decisions such as self-renewal and differentiation. The concept of niche is also crucial in the regulation of ASCs as demonstrated by the pioneering studies on Drosophila (Deng and Lin 1997). Deng showed that the female fly has cells named “cap cells” at the tip of the ovary to which germ line stem cells adhere. During mitosis, daughter stem cells divided in such a way that they no longer adhered to the cap cells and started differentiating and growing into cystoblasts. On the contrary, daughter stem cells that stayed attached to the cap cells after cell division remained germ line stem cells. These results suggested that the cap cells provided a special environment for the germ line stem cells and played a role as a stem cell niche in the fly ovary. Today, human ASC niches have been characterized in several body regions; however, how they function as stem cell niches is far from being elucidated. In ESCs molecular cues in the embryonic milieu, originating from various cell types, possess various signaling characteristics with diverse effects on specific target cells or tissues. Certain stimulatory factors such as Activin, Nodal, and FGF maintain pluripotency in ESCs, while other factors, such as ngn3, Notch, and ASK-1 (endocrine, hematopoietic, and neuronal lineages, respectively), dictate a differentiated fate (Serafimidis et  al. 2007; Radtke et  al. 2004; Elmi et  al. 2007; Vallier et  al. 2005). These signals regulate the phenotype of stem cells or their progeny, are at times antagonistic and may be separated in a spatial and/or temporal fashion. These processes are extremely complex, and their sophistication is far to be fully understood (Constantinescu 2000; Lensch et al. 2006). It has to be underlined that the stem niche environment is not only determined by soluble endocrine or paracrine growth factors through cell–cell cross-talk but also on the contrary is regulated by a more complex interplay of different factors such as cell–matrix adhesion and the physical and physical–chemical environment. Peerani et al. (2007) in a series of sophisticated experiments demonstrated that the modulation of ESCs’ density per se can be used to generate distinct patterns of selfrenewal and differentiation. It has been demonstrated that in the osteoblastic niche, hematopoietic stem cells are predominantly located at the lowest end of an oxygen gradient in the bone marrow (Parmar et al. 2007); therefore, it is reasonable to think that physiochemical properties of the niche (pH, oxygen abundance, and size) are important factors regulating stem cell fate. Moreover, an increasing number of experimental evidence suggests that physical forces acting on stem cells through extracellular matrix or physical–chemical parameters, such as hypoxia (pH), can commit their fate. For example, the shear forces exerted by the blood flow have been found to direct isolated ESCs through an endothelial differentiation (Yamamoto et al. 2005) and slight mechanical deformations can induce mesenchymal stem cells (MSCs) to differentiate into smooth muscle cells (Kurpinski et al. 2006). A wide range of in vitro studies confirm the relevance of the so-called mechanobiology to

178

M. Federico and A. Giordano

stem cell differentiation. Undifferentiated MSCs grown on polymer gels mimicking the extracellular matrix elasticity of a given tissue were found to express precursor proteins for the cell type typically present in that tissue in the absence of specific growth factors. Cells grown on gels of elasticity comparable to brain express neuronal markers and their phenotypes and are not contractile. Conversely, MSCs cultured on more rigid gels (mimicking muscular tension) begin to express early skeletal muscle proteins. Finally, more rigid substrates that mimic pre-calcified bone yield MSCs that resemble bone cells and exert significant traction on their environment (Reilly and Engler 2009). Recently, to highlight the contribution of extracellular matrix to stem cell behavior, the physical parameters of the microenvironment have been proposed to act also as homing signals to migrating stem cells. As seen from the use of various gel matrices with controlled elasticity and non-limiting ligand density, most, if not all, cells are found to adhere and to anchor more strongly to stiff substrates compared with soft substrates. In a gradient of elasticity, cells therefore accumulate on stiffer substrates in a process called durotaxis, which might constitute a biophysical basis for which MSCs home to sites of injury, scar formation, and fibrosis (Disher et al. 2009).

CSCs and the Microenvironment Some of the general mechanisms, mentioned above, that regulate stem cell selfrenewal and differentiation are also crucial in CSC biology. The analogies are so many that maybe cancer could be considered a disease of unregulated self-renewal, in which mutations convert normal stem cell self-renewal pathways into the promoters of neoplastic proliferation (Pardal et  al. 2003; Al-Hajj and Clarke 2004; Beachy et al. 2004). The Bmi-1 (B-cell specific Moloney murineleukemia virus integration site 1), Notch, Wnt, and Sonic hedgehog pathways (Park et al. 2003; Lessard and Sauvageau 2003; Duncan et  al. 2005; Bhardwaj et  al. 2001/2002) were initially identified to play a role in carcinogenesis and were later related to stem cell self-renewal (Taipale and Beachy 2001; Weng and Aster 2004). Studies on these pathways have revealed close links between cancer cells and normal ASCs (Pardal et al. 2003; Reya et al. 2001; Reya and Clevers 2005). Uncontrolled activation of these pathways may result in specific cancers, possibly as an attempt to recapitulate normal embryonic organogenesis (Molofsky et al. 2004; Ruiz i Altaba et al. 2004). Aberrant Notch signaling has been detected in several cancers and has recently been strongly connected to T-cell acute lymphoblastic leukemia (Grabher et al. 2006). Aberrant activation of the Wnt pathway has been found in various human tumors and is strongly associated with colorectal cancer (Taipale and Beachy 2001). Increased Hedgehog signaling has been linked not only to a small subset of tumors in the brain, skin, and muscle but also recently to cancers in the lungs, gastrointestinal tract, and pancreas (Pasca di Magliano and Hebrok 2003). Bmi-1 is an oncogene that is found overexpressed in several human cancers (Valk-Lingbeek et al. 2004) as for instance in the majority

8  Cancer Stem Cells and Microenvironment

179

of medulloblastomas (Leung et al. 2004; Marino 2005). As for ESCs and ASCs, also for CSCs the spatial definition of these signaling pathways in a particular microenvironment is crucial and the concept of CSC niche is rapidly emerging. A study of stem cells in brain tumors has made significant progress in this area. Calabrese and colleagues showed that brain tumor cells co-expressing Nestin and CD133 (the fraction believed to contain the CSCs) were found near the capillaries in the brain tumor (Calabrese et al. 2007). When these cells were co-cultured, the CSCs selectively adhered to the endothelial cells, similar to the cap cells in drosophila. This suggested that endothelial cells secrete factors necessary to maintain brain tumor CSCs. Moreover, it has been shown that the CD133-positive cells derived from human medulloblastoma together with endothelial cells developed brain tumors when xenografted into the brain of a recipient nude mouse. Comparably, Krause and colleagues (2006) showed that there was impaired induction of chronic myeloid leukemia (CML)-like myeloproliferative disease among recipient mice that received transplanted BCR–ABL1transduced CML progenitors from CD44-mutant donors. These results indicate that cell–cell interactions are crucial in providing a homing signal to CSCs; however, they are also able to interfere with engraftment in their niche within the recipient organism. Tumor hypoxia is a well-established factor of aggressiveness that since the 1980s and recent studies on ASC and CSC microenvironments offer a molecular explanation of the former clinical observation (Overgaard 1989). In fact, as for some ASC populations, hypoxia is thought to play a critical role also in CSC regulation. Soeda colleagues (2009) clearly demonstrated that human-derived CSCs when cultured in hypoxic conditions are promped toward self-renewal and strongly retain the undifferentiated phenotype, whereas CSCs cultured in normoxia do not. Moreover, it has been recently reported (Li et  al. 2009) that physiological oxygen levels differentially induce hypoxia inducible factor-2a (HIF2a) levels in CSCs. HIF1a is known to function in proliferation and survival of all cancer cells, while HIF2 seems to be essential only in CSCs and was not expressed by normal neural progenitors. Heddleston et al. (2009) extended these studies to examine the role of hypoxia in regulating tumor cell plasticity and found that hypoxia promotes the self-renewal capability of the stem and non-stem populations. Furthermore, hypoxia induces a more stem-like phenotype in the non-stem population with a marked increased in neurosphere formation as well as upregulation of important stem cell factors, such as OCT4, NANOG, and c-MYC. The importance of HIF2a was further supported as forced expression of non-degradable HIF2a-induced expression of a CSC marker and augmented the tumorigenic potential of the non-stem population. It has to be underlined that the influence of the microenvironment on the tumor progenitor is eventually even greater. Several groups have reported the ability of the microenvironment to be able to dramatically reprogram the cancer phenotype. In the early seventies, Mintz and Illmensee reported that teratocarcinoma cells injected into murine blastocysts aquired totipotency and a reversion to a normal phenotype (Mintz and Illmensee 1975; Illmensee and Mintz 1976). This model clearly established a non-mutational basis for the transformation of these teratocarcinoma cells

180

M. Federico and A. Giordano

into a malignant phenotype, and demonstrated the unique ability of an embryonic microenvironment to reprogram these cells to a normal phenotype. In more recent years, other researchers (Lee et al. 2005; Kulesa et al. 2006) demonstrated that the chick and zebrafish embryonic microenvironments not only repress tumor formation when implanted with aggressive human melanoma cells but induce lineagespecific genes and migratory behaviors associated with a more normal phenotypes, representative of an epigenetic reprogramming event. The same group with an elegant in vitro model showed that apart from any cell–cell interaction, the extracellular matrix when conditioned by specific cell types can clearly influence the phenotype of aggressive cancers (melanoma) in a reversible (epigenetic) way (Abbott et al. 2008). To further demonstrate a specific influence of tumor microenvironment on stem cells, Taylor et al. (2008) combined human ES cells with cancer-associated fibroblasts (CAFs) from patients with prostate cancer. When grafted into the kidney capsule of host mice, human ES cells form teratomas that rapidly enlarge and kill the host. When combined with CAFs, the teratoma formation is significantly abrogated, demonstrating that the stromal tumor niche is able to influence human ES cell differentiation. Furthermore, the CAF-derived niche differs from that derived from nonmalignant prostatic fibroblasts that were unable to alter or impede teratoma formation. Further data in support of the relevance of the environmental influence on CSC plasticity comes from Knutson et al. (2006; Santisteban et al. 2009), who demonstrated that in vivo the induction of an immune response against an epithelial breast cancer led to the T-cell-dependent outgrowth of a tumor, the cells of which had undergone epithelial to mesenchymal transition (EMT). The resulting mesenchymal tumor cells had a CD24/low/CD44+ phenotype, consistent with breast CSCs. EMT induced by CD8 T cells produced tumors that had characteristics of breast CSCs, including potent tumorigenicity, ability to re-establish an epithelial tumor, and enhanced resistance to drugs and radiation. These data in contrast to the hierarchical CSC hypothesis suggest that breast cancer arises from the transformation of a resident tissue stem cell and that EMT can produce the breast CSC phenotype. These findings have two important implications: first, the observed phenotypic plasticity suggests that a dynamic equilibrium may exist between CSCs and nonCSCs within tumors; second, this equilibrium may be shifted in one direction or another by contextual signals within the tumor microenvironment influencing the probability for there to be interchanged between the CSC and non-CSC compartments and thus regulating the overall tumor homeostasis.

Concluding Remarks CSCs represent a budding field of research that has yet to emerge on its own. We are still completely reliant on our understanding of embryonic and ASC biology to explain the role of stem cells in cancer. However, it is clear from the initial studies that CSCs play a critical role in tumorigenisis and resistance to cancer therapies;

8  Cancer Stem Cells and Microenvironment

181

therefore, our continued efforts are essential in elucidating the biological role of these tumor sub-populations. Here we have given an overview of what is known so far in CSC biology, from the initial documentations of CSCs to the role of specific growth factors and transcriptional regulators in CSC biology. We have also discussed major findings in the ESC and ASC research fields, which we retain necessary for the proper physiological understanding of CSCs. It is important to note that these correlations are assumed because of the similarities between CSCs and ESCs or ASCs, respectively. It will be imparitive in the future to better characterize CSCs on their own to understand their origin and their unique role in cancer biology.

References Abbott DE, Bailey CM, Postovit LM et al. (2008) The epigenetic influence of tumor and embryonic microenvironments: how different they are? Cancer Microenvironment 1:13–21. Al-Hajj M, Clarke MF (2004) Self-renewal and solid tumor stem cells. Oncogene 23:7274–7282. Allsopp RC, Vaziri H, Patterson C, Goldstein S, Younglai EV, Futcher B, Greider CW, Harley CB (1992) Telomere length predicts replicative capacity of human fibroblasts. Proc Natl Acad Sci U S A 89:10114–10118. Avilion AA, Nicolis SK, Pevny LH, Perez L, Vivian N, Lovell-Badge R (2003) Multipotent cell lineages in early mouse development depend on SOX2 function. Genes Dev 17:126–140. Beachy PA, Karhadkar SS, Berman DM (2004) Tissue repair and stem cell renewal in carcinogenesis. Nature 432 :324–331, 639–648. Becker KA, Ghule PN, Therrien JA, Lian JB, Stein JL, van Wijnen AJ, Stein GS (2006) Selfrenewal of human embryonic stem cells is supported by a shortened G1 cell cycle phase. J Cell Physiol 209:883–893. Bhardwaj G, Murdoch B, Wu D et al. (2001/2002) Sonic hedgehog induces the proliferation of primitive human hematopoietic cells via BMP regulation. Nat Immunol 2:172–180, 542–545. Boman BM, Huang E (2008) Human colon cancer stem cells: a new paradigm in gastrointestinal oncology. J Clin Oncol June:2828–2838. Bonnet D, Dick J (1997) Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 3:730–737. Boyer LA, Lee TI, Cole MF et al. (2005) Core transcriptional regulatory circuitry in human embryonic stem cells. Cell 122:947–956. Brüstle O, Jones KN, Learish RD, Karram K, Choudhary K, Wiestler OD, Duncan ID, McKay RDG (1999). Embryonic stem cell-derived glial precursors: a source of myelinating transplants. Science 285:754–756. Calabrese C, Poppleton H, Kocak M et al. (2007) A perivascular niche for brain tumor stem cells. Cancer Cell 11:69–82. Cheng LC, Tavazoie M, Doetsch F (2005) Stem cells: from epigenetics to microRNAs. Neuron 46:363–367. Constantinescu SN (2000) Stem cell generation and choice of fate: role of cytokines and cellular microenvironment. J Cell Mol Med 4(4):233–248. Deng W, Lin H (1997) Spectrosomes and fusomes anchor mitotic spindles during asymmetric germ cell divisions and facilitate the formation of a polarized microtubule array for oocyte specification in Drosophila. Dev Biol 189:79–94. Dirks PB (2008) Brain tumor stem cells: bringing order to the chaos of brain. J Clin Oncol June:2916–2924.

182

M. Federico and A. Giordano

Disher D, Mooney DJ, Zandstra PW (2009) Growth factors, matrices and forces combine and control stem cells. Science 324:1673–1677. Duncan AW, Rattis FM, Dimascio LN, Congdon KL et al. (2005) Integration of Notch and Wnt signaling in hematopoietic stem cell maintenance. Nat Immunol 6:314–322. Elmi M et al. (2007) Mechanism of MASH1 induction by ASK1 and ATRA in adult neural progenitors. Mol Cell Neurosci 36(2):248–259. Fluckinger AC, Marcy G, Marchand M, Négre D, Cosset FL, Mitalipov S, Wolf D, Savatier P, Dehay C (2006) Cell cycle features of primate embryonic stem cells. Stem Cells 24:547–556. Fraichard A, Chassande O, Bilbaut G, Dehay C, Savatier P, Samarut J (1995) In vitro differentiation of embryonic stem cells into glial cells and functional neurons. J Cell Sci 108:3181–3188. Fuchs E, Segre JA (2000) Stem cells: a new lease on life. Cell 100:143–155. Giacinti C, Giordano A (2006) RB and cell cycle progression. Oncogene 25:5220–5227. Grabher C, von Boehmer H, Look AT (2006) Notch 1 activation in the molecular pathogenesis of T-cell acute lymphoblastic leukaemia. Nat Rev Cancer 6:347–359. Greider CW, Blackburn EH (1987) The telomere terminal transferase of tetrahymena is a ribonucleoprotein enzyme with two kinds of primer specificity. Cell 51:887–898. Haas K, Johannes C, Geisen C, Schmidt T, Karsunky H, Blass-Kampmann S, Obe G, Möröy T (1997). Malignant transformation by cyclin E and Ha-Ras correlates with lower sensitivity towards induction of cell death but requires functional Myc and CDK4. Oncogene 15:2615–2623. Harley CB, Futcher AB, Greider CW (1990) Telomeres shorten during aging of human fibroblasts. Nature 345:458–460. Hayflick L (1965) The limited in  vitro lifetime of human diploid cell strains. Exp Cell Res 37:614–636. Hayflick L, Moorhead PS (1961) The serial cultivation of human diploid cell strains. Exp Cell Res 25:585–621. Heddleston JM, Li Z, McLendon RE et al. (2009) The hypoxic microenvironment maintains glioblastoma stem cells and promotes reprogramming towards a cancer stem cell phenotype. Cell Cycle 8(20):3274–3284. Hogan BLM, Beddington RSB, Constantini F, Lacy E (1994) Manipulating the mouse embryo, 2nd ed. Cold Spring Harbor Laboratory Press, New York. Huff CA, Matsui W (2008) Multiple myeloma cancer stem cells. J Clin Oncol 2895–2900. Huangfu DW, Osafune K et al. (2008) Induction of pluripotent stem cells from primary human fibroblasts with only Oct4 and Sox2. Nat Biotechnol 26:1269–1275. Illmensee K, Mintz B (1976) Totipotency and normal differentiation of single teratocarcinoma cells cloned by injection into blastocysts. Proc Natl Acad Sci U S A 73(2,9):549–553, 1998–2006. Kakarala M, Wicha MS (2008) Implications of the cancer stem-cell hypothesis for breast cancer prevention. J Clin Oncol June:2813–2820. Kasten MM, Giordano A (1998) pRB and the Cdks in apoptosis and the cell cycle. Cell Death Differ 5:132–140. Knutson KL, Lu H, Stone B et  al. (2006). Immunoediting of cancers may lead to epithelial to mesenchymal transition. J Immunol 177:1526–1533. Kouzarides T (2007) Chromatin modifications and their function. Cell 128:693–705. Krause DS, Lazarides K, von Andrian UH et  al. (2006) Requirement for CD44 in homing and engraftment of BCR-ABL-expressing leukemic stem cells. Nat Med 12:1175–1180. Kulesa PM, Kasemeier-Kulesa JC, Teddy JM et al. (2006) Reprogramming metastatic melanoma cells to assume a neural crest cell-like phenotype in an embryonic microenvironment. Proc Natl Acad Sci U S A 103(10):3752–3757. Kurpinski K, Chu J, Hashi C et  al. (2006) Anisotropic mechanosensing by mesenchymal stem cells. Proc Natl Acad Sci U S A 103:16095. Lee CJ, Dosch J, Simeone DM (2008) Pancreatic cancer stem cells. J Clin Oncol June:2806–2812.

8  Cancer Stem Cells and Microenvironment

183

Lee JH, Hart SR, Skalnik DG (2004) Histone deacetylase activity is required for embryonic stem cell differentiation. Genesis 38:32–38. Lee LM, Seftor EA, Bonde G et al. (2005) The fate of human malignant melanoma cells transplanted into zebrafish embryos: assessment of migration and cell division in the absence of tumor formation. Dev Dyn 233(4):1560–1570. Lensch MW, Daheron L, Schlaeger TM (2006) Pluripotent stem cells and their niches. Stem Cell Rev 2(3):185–201, 773–789. Lessard J, Sauvageau G (2003) Bmi-1 determines the proliferative capacity of normal and leukaemic stem cells. Nature 423:255–260. Leung C, Lingbeek M, Shakhova O (2004). Bmi1 is essential for cerebellar development and is overexpressed in human medulloblastomas. Nature 428:337–341. Li X, Lewis MT, Huang J et al. (2008) Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy. J Natl Cancer Inst 100:672–679. Li Z, Bao S, Wu Q et al. (2009) Hypoxia inducible factors regulate tumorigenic capacity of glioma stem cells. Cancer Cell 15:501–513. Lunyak VV, Rosenfeld MG (2008) Epigenetic regulation of stem cell fate. Hum Mol Genet 17:28–36. Maitland NJ, Collins AT (2008) Prostate cancer stem cells: a new target for therapy. J Clin Oncol June:2862–2870. Marino S (2005) Medulloblastoma: developmental mechanisms out of control. Trends Mol Med 11:17–22. Martin GR, Evans MJ (1974) The morphology and growth of a pluripotent teratocarcinoma cell line and its derivatives in tissue culture. Cell 2:163–172. Martin GR, Evans MJ (1975) Differentiation of clonal lines of teratocarcinoma cells: formation of embryoid bodies in vitro. Proc Natl Acad Sci U S A 72:1441–1445. Meshorer E (2007) Chromatin in embryonic stem cell neuronal differentiation. Histol Histopathol 22:311–319. Meshorer E, Yellajoshula D, George E, Scambler PJ, Brown DT, Misteli T (2006) Hyperdynamic plasticity of chromatin proteins in pluripotent embryonic stem cells. Dev Cell 10: 105–116. Mintz B, Illmensee K (1975) Normal genetically mosaic mice produced from malignant teratocarcinoma cells. Proc Natl Acad Sci U S A 72(9):3585–3589. Misteli T (2004) Spatial positioning: a new dimension in genome function. Cell 119:153–156. Misteli T (2007) Beyond the sequence: cellular organization of genome function. Cell 128:787–800. Molofsky AV, Pardal R, Morrison SJ (2004) Diverse mechanisms regulate stem cell self-renewal. Curr Opin Cell Biol 16:700–707. Mumberg D, Haas K, Möröy T, Niedenthal R, Hegemann JH, Funk M, Müller R (1996) Uncoupling of DNA replication and cell cycle progression by human cyclin E. Oncogene 13:2493–2497. Nichols J, Zevnik B, Anastassiadis K, Niwa H, Klewe-Nebenius D, Chambers I, Schöler H, Smith A (1998) Formation of pluripotent stem cells in the mammalian embryo depends on the POU transcription factor Oct4. Cell 95:379–391. Niwa H, Burdon T, Chambers I, Smith A (1998) Self-renewal of pluripotent embryonic stem cells is mediated via activation of STAT3. Genes Dev 12:2048–2060. Overgaard J (1989) Sensitization of hypoxic tumour cells; clinical experience. Int J Radiat Biol 56(5):801–811. Pardal R, Clarke MF, Morrison SJ (2003) Applying the principles of stem-cell biology to cancer. Nat Rev Cancer 3:895–902. Park IK, Qian D, Kiel M et al. (2003) Bmi-1 is required for maintenance of adult self-renewing haematopoietic stem cells. Nature 423:302–305. Parmar K, Mauch P, Vergilio JA et al. (2007) Distribution of hematopoietic stem cells in the bone marrow according to regional hypoxia. PNAS 104(13):5431–5436.

184

M. Federico and A. Giordano

Pasca di Magliano M, Hebrok M (2003) Hedgehog signalling in cancer formation and maintenance. Nat Rev Cancer 3:903–911. Peacock CD, Watkins DN (2008) Cancer stem cells and the ontogeny of lung cancer. J Clin Oncol June:2883–2889. Peerani R, Balaji MR, Bauwens C et al. (2007) Niche-mediated control of human embryonic stem cell self-renewal and differentiation. EMBO J 26:4744–4755. Power MA, Tam PP (1993) Onset of gastrulation, morphogenesis and somitogenesis in mouse embryos displaying compensatory growth. Anat Embryol (Berl) 187:493–504. Prince MEP, Ailles LE (2008) Cancer stem cells in head and neck squamous cell cancer. J Clin Oncol June:2871–2875. Radtke F et al. (2004) Notch regulation of lymphocyte development and function. Nat Immunol 5(3):247–253. Rathjen J, Lake JA, Bettess MD, Washington JM, Chapman G, Rathjen PD (1999) Formation of a primitive ectoderm like cell population, EPL cells, from ES cells in response to biologically derived factors. J Cell Sci 112:601–612. Reilly GC, Engler AJ (2010) Intrinsic extracellular matrix properties regulate stem cell differentiation. J Biomechanics. 43:55–62. Reya T, Clevers H (2005) Wnt signalling in stem cells and cancer. Nature 434:843–850. Reya T, Morrison SJ, Clarke MF et al. (2001) Stem cells, cancer and cancer stem cells. Nature 1; 414 (6859):105–11. Risbridger GP, Taylor RA (2008) Minireview: regulation of prostatic stem cells by stromal niche in health and disease. Endocrinology 149(9):4303–4306. Ruiz i Altaba A, Stecca B, Sanchez P (2004) Hedgehog–Gli signaling in brain tumors: stem cells and paradevelopmental programs in cancer. Cancer Lett 204:145–157. Santisteban M, Reiman MJ, Asiedu MK et al. (2009) Immune-induced epithelial to mesenchymal transition in vivo generates breast cancer stem cells. Cancer Res 69(7):2887–2895. Savatier P, Huang S, Szekely L, Wiman KG, Smaraut J (1994) Contrasting patterns of retinoblastoma protein expression in mouse embryonic stem cells and embryonic fibroblasts. Oncogene 9:809–818. Schopperle WM, DeWolf WC (2007) The TRA-1–60 and TRA-1–81 human pluripotent stem cell markers are expressed on podocalyxin in embryonal carcinoma. Stem Cells 25:723–730. Sell S, Leffert HL (2008) Liver cancer stem cells. J Clin Oncol June:2800–2805. Serafimidis I et al. (2007) Novel effectors of directed and ngn3 mediated differentiation of mouse embryonic stem cells into endocrine pancreas progenitors. Stem Cells 26:3–16. Shamblott MJ, Axelman J, Wang S, Bugg EM, Littlefield JW, Donovan PJ, Blumenthal PD, Huggins GR, Gearhart JD (1998). Derivation of pluripotent stem cells from cultured human primordial germ cells. Proc Natl Acad Sci U S A 95:13726–13731. Shukla V, Vaissiere T, Herceg Z (2008) Histone acetylation and chromatin signature in stem cell identity and cancer. Mutat Res 637:1–15. Soeda A, Park M, Lee D et al. (2009) Hypoxia promotes expansion of the CD133-positive glioma stem cells through activation of HIF1a. Oncogene. 28:3949–3459. Solter D, Skreb N, Damjanov I (1971) Cell cycle analysis in the mouse EGG-cylinder. Exp Cell Res 64:331–334. Stead E, White J, Faast R, Conn S, Goldstone S, Rathjen J, Dhingra U, Rathjen P, Walker D, Dalton S (2002) Pluripotent cell division cycles are driven by ectopic Cdk2, cyclin A/E and E2F activities. Oncogene 21:8320–8333. Taipale J, Beachy PA (2001) The Hedgehog and Wnt signalling pathways in cancer. Nature 411:349–354. Takaishi S, Okomura T, Wang TC (2008) Gastric cancer stem cells. J Clin Oncol June:2876–2882. Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, Jones JM (1998) Embryonic stem cell lines derived from human blastocysts. Science 282:1145–1147. Valk-Lingbeek ME, Bruggeman SW, van Lohuizen M (2004) Stem cells and cancer; the polycomb connection. Cell 118:409–418.

8  Cancer Stem Cells and Microenvironment

185

Vallier L, Alexander M, Pedersen RA (2005) Activin/Nodal and FGF pathways cooperate to maintain pluripotency of human embryonic stem cells. J Cell Sci 118(Pt 19)(17):4495–4509, 6322–6327. Weng AP, Aster JC (2004) Multiple niches for Notch in cancer: context is everything. Curr Opin Genet Dev 14:48–54. Wu H, Sun YE (2006) Epigenetic regulation of stem cell differentiation. Pediatr Res 59:21–25. Wobus AM, Boheler KR (2005) Embryonic stem cells: Prospects for developmental biology and cell therapy. Physiol Rev 85:635–678. Yamamoto K, Sokabe T, Watanabe T et  al. (2005) Fluid shear stress induces differentiation of Flk-1-positive embryonic stem cells into vascular endothelial cells in vitro. Am J Physiol Heart Circ Physiol 288:H1915.

Chapter 9

Epithelial–Mesenchymal Transition in Development and Diseases Yadi Wu and Binhua P. Zhou

Abstract  Epithelial–mesenchymal transition (EMT) is critical for appropriate embryo implantation, embryogenesis, wound healing, tissue regeneration, and organ development. During this EMT process, epithelial cells lose the adherent and tight junctions. They gain a mesenchymal cell phenotype that enables them to invade and migrate over long distances and resist apoptosis. A similar process is also detected in tumor metastasis, suggesting that the tumor cells hijack this developmental pathway for tumor progression. Here, we review three different types of EMT in physiological and pathological conditions with special focus on the new development on type 3 EMT in metastasis. We summarize the recent new findings on tumor microenvironment, signaling pathways, and mechanisms underlying the regulation of EMT at metastasis. Understanding the biology of EMT will open new avenues for controlling fibrosis and cancer progression.

Overview of EMT Epithelial–mesenchymal transition (EMT) is a phenotypic conversion during embryonic development when tissue remodeling and cell migration shape the future organism, such as those in embryonic development and wound healing. In pathological condition, a similar EMT process is also observed in fibrotic diseases and metastasis. During EMT, epithelial cells lose the adherent and tight junctions that keep them in contact with their neighbors; they gain a mesenchymal cell phenotype enabling them to break through the basal membrane and migrate over a long distance owing to profound changes in their cytoskeleton

B.P. Zhou (*) Markey Cancer Center, University of Kentucky School of Medicine, Lexington, KY, USA and Department of Molecular and Cellular Biochemistry, University of Kentucky School of Medicine, Lexington, KY, USA e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_9, © Springer Science+Business Media, LLC 2010

187

188

Y. Wu and B.P. Zhou Type 1: formations of Mesoderm, Neural Crest, Cardiac Valve and Secondary palate

EMT types

Type 2: Fibrosis and re-epithelialization of wounded skin Type 3: tumor metastasis

Fig. 9.1  EMT is classified into three different subtypes (please see text for more detail)

architecture and gene expression profile (Kalluri and Neilson 2003). This concept was pioneered by the seminal study from Elizabeth Hay using chick primitive streak formation as a model in 1967 (Hay 1995). Hay realized that an epithelial phenotypic conversion was of crucial importance during gastrulation and cell migration in the early vertebrate embryo. She proposed that differentiated epithelial cells could undergo dramatic “transformation” into mesenchymal cells (Hay 1995; Greenburg and Hay 1998). However, since this “transformation” is reversible and mesenchymal cells can revert back to epithelial cells through a reverse process of mesenchymal–epithelial transition (MET), the term of “transition” is now used. The field of EMT has vastly expanded recently in both scope and understanding. EMT was divided into three subtypes based on the biological context under which they occur (Kalluri 2009; Kalluri and Weinberg 2009; Zeisberg and Neilson 2009) (Fig. 9.1). Type 1 EMT describes the transition events of epithelial cells to become motile mesenchymal cells during implantation, embryo formation, gastrulation, and neural crest migration. These primary mesenchymal cells can revert back and form secondary epithelia in mesodermal and endodermal organs through MET. Type 2 EMT involves the program of generating tissue fibroblasts from epithelial or endothelial cells during injury and chronic inflammation. This type of EMT will eventually lead to tissue or organ fibrosis and organ malfunction if inflammation persists. Type 3 EMT describes how tumor cells at the invasive front of primary tumors undergo a phenotypic transitioning in order to invade and metastasize through the blood stream or lymph node system and form a metastatic lesion through MET at distant tissues or organs. Although these three subtypes of EMT represent distinct biological processes, the genetic element and mechanism of regulation may be similar and well conserved. Studies of EMT often involve various model systems ranging from different epithelial cell types to assorted stimulations, it is important to use validated biomarkers to examine the phenotypic conversion during all three classes of EMT. Common biomarkers include cell-surface and extracellular molecules, cytoskeletal proteins, and specific transcription factors (Table 9.1). For example, downregulation of E-cadherin is a hallmark of EMT and loss of E-cadherin expression facilitates the induction of EMT (Huber et  al. 2005). E-cadherin is a cell–cell adhesion molecule that participates in homotypic, calcium-dependent interactions to form epithelial adherent junctions (Cowin et  al. 2005; Junghans et  al. 2005).

9  Epithelial–Mesenchymal Transition in Development and Diseases

189

Table 9.1  EMT markers Cell-surface proteins

Cytoskeketal molecules

ECM proteins

Transcription factors (common)

Acquired markers N-cadherin ob-cadherin a5b1 integrin aVb6 integrin Syndecan-1 FSP1 a-SMA Vimentin b-catenin a1(I) collagen a1(III) collagen Fibronectin Laminin 5 Snail Slug Twist SIP1/ZEB2

Attenuated markers E-cadherin ZO-1 Claudins Occludin Desmoplakin Cytokeratin Plakophilin Crumbs3 HUGHL2 a1(IV) collagen Laminin 1

The cytoplasmic tail of E-cadherin links to cytoskeleton through a- and b-catenin to modulate cell–cell adhesion and cell migration. Cadherin switching from E-cadherin to N-cadherin is commonly found in EMT. N-cadherin is expressed in mesenchymal cells, fibroblasts, cancer cells, and neural tissue, and N-cadherin levels in these cells are closely associated with their invasiveness, motility, and metastasis. In addition, E-cadherin repressors, such as Snail, Slug, Twist, and ZEB1/2, are commonly used as EMT markers. Snail is the first described E-cadherin repressor and is also the common downstream target of various signaling pathways leading to EMT. Vimentin, an intermediate filament that is mainly expressed in fibroblasts, endothelial cells, and hematopoietic cells, is also commonly used as an indicator for type 3 EMT, as expression of vimentin in tumor cells correlates with their invasiveness and metastatic potential. Furthermore, differential expression of integrin is also used as a biomarker of EMT, as integrins modulate the interaction of cells with extracellular matrix (ECM). For example, increased expression of a5 integrin is commonly found in type 2 and type 3 EMT (Qian et al. 2005; Davidson et al. 2006; White et al. 2007).

Type 1 EMT in the Formation of Mesoderm and Neural Crest EMT is crucially important to the tissue morphogenetic events during embryonic development, such as in the mesoderm formation, neural crest formation, heart valve development, and secondary palate formation. In the absence of EMT,

190

Y. Wu and B.P. Zhou

development cannot proceed through the blastula stage. Mesoderm formation and neural crest development represent the major EMT programs that occur during early embryonic development; the resulting mesenchymal and neural crest cells act as progenitors and further differentiate into various cell types via MET program. For example, gastrulation EMT produces the mesoderm, giving rise to muscle, bone, and connective tissues, whereas neural crest delamination EMT gives rise to glial and neuronal cells, adrenal glandular tissues, pigment-containing cells of the epidermis, and skeletal and connective tissues. The heart valve development and secondary palate formation occur in relatively well-differentiated epithelial cells that are destined to become defined mesenchymal cells types.

Mesoderm Formation The classic example of EMT is the formation of mesoderm from the primitive ectoderm during gastrulation. Gastrulation is observed in all metazoans and is accompanied by drastic morphogenic changes from a single epithelial layer (the epiblast) into three embryonic germ layers, the ectoderm, mesoderm, and endoderm, to form a complex three-dimensional multilayered embryo (Shook and Keller 2003). In chicken and mouse embryo, Wnt and TGFb signaling provide the initial induction signals for EMT while the FGF signal is necessary to maintain the EMT regulatory network during mesoderm formation. All these signaling events activate the expression of Snail, which represses the expression of E-cadherin and other tight junction components (such as claudins, occludins, and Crumbs) and promotes cell migration. In Snail knock-out mice, although mesoderm specification is not affected, the cells that form are unable to migrate. Gastrulation in Drosophila melanogaster initiates at the completion of the celluarization step in a stripe of ventral cells. The newly formed mesenchymal cells migrate dorsally along the ectoderm before differentiating into visceral and somatic mesoderm. Snail and Twist are expressed in the presumptive mesoderm and in the invaginating cells. These two molecules are under the control of Dorsal (p65 subunit of nuclear factor-kB), which is involved in the establishment of the dorso-ventral axis. Snail represses the transcription of Shotgun (the orthologue of vertebrate E-cadherin) in mesodermal precursor cells that undergo invagination at the ventral furrow (Oda et al. 1998). More importantly, Snail and Twist are involved in the control of rapid and orderly transition from the ventral epithelium into a mesenchymal state (Dawes-Hoang et al. 2005; Kolsch et al. 2007). Therefore, Dorsal, Twist and Snail act in concert to induce EMT and bring the internalization and formation of the Drosophila mesoderm. The zebrafish (Dabio rerio) organizer is the site for EMT of axial mesendoderm cells. Liv1, a seven-spanning transmembrane receptor that is involved in zinc transport is a downstream of signal transducer and activator of transcription 3 (STAT3) (Yamashita et al. 2004). STAT3 is essential for gastrulation movements in the organizer during zebrafish gastrulation. Interestingly, LIV1 is essential for the

9  Epithelial–Mesenchymal Transition in Development and Diseases

191

nuclear localization and repressive activity of Snail, which inhibits E-cadherin expression and promotes the up-regulation of N-Cadherin.

Neural Crest Formation Neural crest formation is another example of Type I EMT in embryogenesis where premigratory neural crest cells form at the border of the neural plate and non-neural ectoderm as a result of signals emanating from these two tissues. Interestingly, similar signaling pathways that operate during EMT at gastrulation are utilized in the neural crest formation. Indeed, a combination of Wnt, FGF, and TGFb (mainly BMP) induce the expression of Snail, Sox, and forkhead box D3 transcription factors (Villanueva et al. 2002). In addition, experimental evidence shows that Notch signaling plays an important role in neural crest formation through induction of Slug in frog and chick embryo (Nieto 2002). The combination of these transcription factors generates the full spectrum of phenotypic changes associated with EMT and primes the precursor cells to become migratory neural crest cells. These neural crest cells are equipped with the ability to migrate over extraordinarily long distances in the embryo prior to their reaggregation via MET for further differentiation.

MET EMT is a reversible process during the embryogenesis, tissue construction and tumor progression, and metastasis. The converted mesenchymal cells can revert to an epithelial cell state by passing through MET. This suggests that inter-conversion between epithelial and mesenchymal cell states may also occur under certain pathological conditions and be affected by signals from microenvironment. In fact, following metastatic spread and colonization, an MET often converts the disseminated mesenchymal cancer cells back to a more differentiated epithelial cell state to form secondary metastatic lesions.

Type 2 EMT in Tissue and Organ Fibrosis Implications of EMT in Fibrosis Re-epithelization, tissue regeneration, and organ fibrosis constitute type 2 EMT. Organ fibrosis is mediated by inflammatory cells and fibroblasts which deposit collagens, elastin, tenacin, and other matrix molecules. Fibrosis associated type 2 EMT specifically occurs in kidney, liver, lung and intestine (Zeisberg et al. 2007).

192

Y. Wu and B.P. Zhou

A series of typical experiments has shown that EMT is an important process during tissue injury leading to organ fibrosis. In terms of the EMT proteomes, fibroblastspecific protein 1 (FSP1, also known as S100A4 and MTS-1), a-SMA (smooth muscle actin), and collagen I are reliable markers to characterize the mesenchymal products generated by the EMTs in the development of fibrosis in various organs. TGFb1, as the major pro-fibrotic cytokine, induces many of the central processes involved in fibrosis, including fibroblast to myfibroblasat differentiation, ECM deposition and EMT. TGFb not only contributes to pulmonary and hepatic fibrosis but also plays a key role in cardiac fibrosis (Gressner et al. 2002; Willis and Borok 2007; Zeisberg et  al. 2007). TGFb induces EMT via both a Smad2/3-dependent pathway and a MAPK-dependent pathway. The relevance of TGFb-induced EMT for progression of organ fibrosis was recently further elucidated using BMP-7 as an intracellular competitor of TGFb signaling in mouse models of kidney, liver, billiard tract, lung, and intestinal fibrosis (Zeisberg et al. 2003a, b). The function of TGFb in fibrosis is highlighted by the finding that Smad3–/– mice are resistant to the induction of several fibrotic diseases (Flanders 2004). TGFb levels are also overproduced and are associated with functional impairment in patients with fibrotic pulmonary diseases such as idiopathic pulmonary fibrosis (Salez et  al. 1998). Clinical studies also demonstrated the correlation between fibrosis and EMT (Rastaldi et al. 2002). Using immunohistochemistry and in situ hybridization, an EMT was demonstrated with the expression of several markers of tubular phenotype transition such as cytokeratin, vimentin, a-SMA, and zona occludens (ZO-1) in 133 kidney biopsies (Rastaldi et  al. 2002). Similarly, in patients with Crohn disease, an expression pattern of EMT was found in areas of fibrosis in the colon (Bataille et al. 2008).

Re-epithelialization of Wounded Skin Re-epithelialization recapitulates several aspects of EMT. Re-epithelialization requires epithelial cells at the edge of the wounded tissue to loosen their cell–cell and cell–ECM contacts and assume a migratory phenotype which are reminiscent of EMT. Slug has a crucial role in wound-healing which is expressed in keratinocytes at the boundary of wounds. Importantly, epithelial cell outgrowth from skin explants was markedly reduced in Slug knock-out mice, whereas overexpression of Slug in cultured human keratinocytes results in increased cell spreading and desmosomal disruption (Savagner et al. 2005). Arnoux et al. further found that EGF can activate Erk5 which specifically enhances Slug promoter activity and controls wound healing in keratinocyte-derived HacaT cells in vitro (Arnoux et al. 2008). However, it should be noted that not all features of EMT are seen. First, the migrating keratinocytes remain part of a cohesive cell sheet since they retain some intercellular junction. Second, the epithelial cells do not actually become mesenchymal, i.e., interstitial cells. They remain epithelial in characteristics. Once wound closure is completed, the involved epithelial cells revert to their tissue-specific differentiated state.

9  Epithelial–Mesenchymal Transition in Development and Diseases

193

Type 3 EMT in Cancer Metastasis Cancer metastasis is believed to consist of four distinct steps: invasion, intravasation, extravasation, and metastatic colonization (Chambers et  al. 2002; Pantel and Brakenhoff 2004). During invasion, tumor cells lose cell–cell adhesion, gain mobility, and leave the site of the primary tumor to invade adjacent tissues. In intravasation, tumor cells penetrate through the endothelial barrier and enter the systemic circulation through blood and lymphatic vessels. In extravasation, cells that survive the anchorage-independent growth conditions in the bloodstream attach to vessels at distant sites and leave the bloodstream. Finally, in metastatic colonization, tumor cells form macrometastases in the new host environment (Chambers et  al. 2002; Pantel and Brakenhoff 2004). All of these steps, from initial breakdown of tissue structure, through increased invasiveness, and ultimately distribution and colonization throughout the body, are characteristics of the developmental process at EMT/ MET. The similarity of genetic controls and biochemical mechanisms underlying the acquisition of the invasive phenotype and the subsequent systemic spread of the cancer cells highlights that tumor cells usurp this developmental pathway for their metastatic dissemination. We will discuss in detail on the environmental stimuli, the signaling pathways, and the mechanisms of regulation that control type 3 EMT in metastasis.

EMT Stimuli from Tumor Microenvironment EMT is a dynamic process that is controlled by signals that cells receive from their microenvironment. Consistent with this notion, EMT commonly occurs at the invasive front (tumor–stromal boundary) of many invasive carcinomas (Franci and Christofori 2006). These observations indicate that EMT is triggered by cellular signals from the tumor microenvironment. The tumor microenvironment as a complex system consisting of extracellular matrix (ECM) (such as collagen and hyaluronic acid), stromal cells (such as fibroblasts of various phenotypes/myofibroblasts, endothelial cells, pericytes and smooth-muscle cells), and infiltrated inflammatory cells (such as neutrophils, myeloid-derived suppressor cells (MDSC), lymphocytes, and macrophages). These infiltrated immune cells secrete cytokines, chemokines, and growth factors, such as TNFa, TGFb, IL-6, FGF, epidermal growth factor (EGF), and HGF (Joyce and Pollard 2009) and play an essential role for supporting tumor progression and metastasis. It is conceivable that the migratory and invasive ability of tumor cells at the invasive front is initiated and propelled by an inflammatory microenvironment through the induction of EMT. Consistent with this notion, a high content of inflammatory cells, particularly tumor-associated macrophages, is commonly found at the invasive fronts of advanced carcinoma (Condeelis and Pollard 2006). Macrophages are key cells in chronic inflammation. Tumor-associated macrophages (TAMs) are important

194

Y. Wu and B.P. Zhou

component of infiltrated leukocytes in most malignant tumors. They modulate inflammation and adaptive immunity and promote cell proliferation by producing growth factors and by enhancing angiogenesis, tissue remodeling, and repair. Macrophage directly influences the behavior and function of tumor cells and has been regarded as an “obligate partner for tumor-cell migration, invasion and metastasis” (Condeelis and Pollard 2006). For example, in PyMT-induced mammary tumors, macrophages are present in the areas of basement membrane breakdown during the development of “early-stage” metastatic lesions and systemic depletion of macrophages results in reduced formation of lung metastasis (Lin et  al. 2001). Inflammatory macrophages increase peritoneal dissemination and metastatic spread of tumor cells both in  vitro and in  vivo in an ovarian cancer model (Mantovani et al. 2008). Clinical studies also indicate a correlation between TAM density and poor prognosis (Pollard 2004). TAMs produce a wide variety of growth factors and cytokines to stimulate the growth, motility, and invasiveness of tumor cells. Interestingly, TAM often cluster around blood vessels and the tumor margin and produce many proteases, ranging from uPA to a variety of matrix metalloproteinases (MMPs) to degrade the basement membrane for creating a channel for tumor cell invasion. TAMs have been shown to be a major source of MMP9 and urokinase-type plasminogen activator. In our recent study, we found that macrophage increase tumor invasive capacity in a TNFa–NF-kB–Snail dependent manner. TNFa-stabilized Snail is critical for inflammation-induced invasion of breast cancer cells. Knockdown of Snail expression significantly inhibited cell migration and invasion induced by inflammatory cytokines and suppressed inflammation-mediated breast cancer metastasis in an animal model (Wu et al. 2009). Thus, macrophages, the major inflammatory component of the stroma in malignancies, facilitate angiogenesis, extracellular matrix breakdown, invasion, and metastasis through multiple mechanisms. In addition to macrophages, fibroblasts/myofibroblasts comprise another major component of tumor stroma. The presence of large numbers of fibroblasts and myofibroblasts in tumor is a hallmark of advanced carcinomas. Tumor fibroblasts are derived not only from the mesenchyme that surrounds the tumor but also from the epithelial cells through EMT process (Radisky et  al. 2007). Recent studies have demonstrated that cancer-activated fibroblasts (CAFs) are important in tumor cell migration and invasion. CAFs isolated from metastatic breast cancer produce elevated levels of IL-6 and enhance cancer cell invasiveness (Studebaker et  al. 2008). In agreement with this, fibroblasts isolated from surgical colon cancer specimens can stimulate the invasive growth of breast and colon cancer cells (De Wever et  al. 2004). Another study that systematically evaluated the effect of fibroblasts found that only melanoma cells that metastasized were influenced by fibroblasts (Cornil et al. 1991). This indicates that fibroblasts create a niche to promote metastasis. In addition, CAFs in pancreatic ductal adenocarcinoma are responsible for a poorly vascularized architecture that imposes a barrier for drug delivery and spurs metastasis (Olive et al. 2009). Furthermore, fibroblasts promote tumor cell proliferation and metastasis through the production of exogenous EMT stimuli

9  Epithelial–Mesenchymal Transition in Development and Diseases

195

such as growth factors, cytokines, chemokines, and MMPs. MMPs derived from tumor cells and stromal components are regarded as major players in assisting the metastasis of tumor cells. For example, transgenic expression of MMP3 stimulates expression of Snail through the increased cellular reactive oxygen species and thus induces down-regulation of E-cadherin and increased tumor progression (Radisky et al. 2005). Myeloid-derived suppressor cells (MDSC) are present in many cancer patients and mice with transplanted or spontaneously developed tumors (Young and Lathers 1999; Almand et al. 2001). MDSCs, characterized as CD11b+ Gr-1+ in mice, can be recruited and activated by multiple factors, such as VEGF, IL-1b, and IL-6, many of which are associated with chronic inflammation (Gabrilovich and Nagaraj 2009). Recruitment of MDSCs in turn further produces these pro-inflammatory factors, resulting in the amplification of the pro-inflammatory response. MDSCs not only suppress the adaptive immune responses but also regulate innate immune responses by modulating the cytokine production of macrophages (Sinha et  al. 2007) and thus directly facilitate metastasis. Inflammatory chemo-attractants S100A8 and S100A9 regulate the accumulation of MDSC to pre-metastatic niche (Hiratsuka et  al. 2006, 2008). Recent studies have shown a close correlation between the level of MDSCs and cancer stage, metastatic tumor burden, and responsiveness to chemotherapy (Diaz-Montero et al. 2009). MDSCs from mammary carcinoma can promote tumor invasion and metastasis (Bunt et al. 2007). In Tgfbr2-decificent mice, MDSCs are concentrated at the invasive tumor front and facilitate tumor cell invasion and metastasis through chemokine receptors CXCR2 and CXCR4 (Yang et al. 2008). The mesenchymal stem cells (MSCs) are a prominent component of the tumor microenvironment which can differentiate into osteoblasts, chondrocytes, adipocytes, and other cells of mesenchymal origin. MSCs can communicate with macrophage through secreting prostaglandin E2 (PGE2) to down-modulate inflammatory cytokines. PGE2 can also potentiate Wnt signaling during wound repair in various organs (Stappenbeck and Miyoshi 2009). The presence of MSCs at the primary site modulates the behavior of neighboring cancer cells (Karnoub et  al. 2007). Co-mingling of bone-marrow-derived human MSCs with weakly metastatic human breast carcinoma cells significantly increases the dissemination of the cancer cells to the lung on a subcutaneous xenograft mouse model. The enhanced metastatic ability is reversible and is dependent on a paracrine loop involving MSC-supplied CCL5 and its receptor CCR5 on the breast cancer cell. The finding that MSCs concentrate in tumors from the circulation raises the possibility of using them as a cellular vehicle to deliver anticancer drug into the tumor. Other inflammatory cells, including lymphocytes, neutrophils, mast cells, T-regulatory cells, and platelets, contribute to the metastatic phenotype of cancers. All of these infiltrated inflammatory cells secrete different cytokines, chemokines, and other factors to promote the tumor cell migration and invasion, and affect the ability of tumor cells to colonize in the metastatic niche.

196

Y. Wu and B.P. Zhou

Molecular Regulation of EMT The hallmark of EMT is the loss of E-cadherin expression, an important caretaker of the epithelial phenotype. Loss of E-cadherin expression is often correlated with the tumor grade and stage because it results in the disruption of the cell–cell adhesion and an increase in the nuclear b-catenin (Cowin et al. 2005; Junghans et al. 2005). Several transcription factors have been implicated in the regulation of EMT, including zinc finger proteins of the Snail/Slug family, the basic helix– loop–helix factor Twist, E12/E47, Goosecoid, dEF1/ZEB1, and SIP1 (Nieto 2002; Yang et al. 2004; Hartwell et al. 2006). These factors act as a molecular switch of EMT program by repressing a subset of common genes that encode cadherins, claudins, integrins, mucins, plakophilin, occluding, and ZO1 to induce EMT. For example, the expression of Snail is associated with E-cadherin repression in metastasis and correlates with tumor recurrence and poor prognosis in various cancers (Elloul et al. 2005; Moody et al. 2005; Bruyere et al. 2009). In addition, extensive crosstalk among these transcription factors forms a signaling network that is responsible for establishing and maintaining mesenchymal cell phenotypes. Furthermore, some of these transcription factors, including Snail, play an important part in overcoming oncogene-induced senescence (Ansieau et al. 2008), inhibiting tumor immunosuppression (Kudo-Saito et al. 2009), and generating tumorigenic cancer stem cells (Mani et al. 2008). These transcription factors communicate and respond to the extracellular signals such as growth factors, cytokines and hypoxia from their microenvironment to induce EMT (Fig. 9.2).

κ

Fig. 9.2  An overview of signaling pathways and transcription factors involved in EMT

α

9  Epithelial–Mesenchymal Transition in Development and Diseases

197

Signaling Pathways TGFb is a primary inducer of EMT. It not only contributes to EMT during embryonic development but also induces EMT during tumor progression in vivo (Zavadil and Bottinger 2005). Overexpression of Smad2 and Smad3 results in increased EMT in a mammary epithelial model (Valcourt et  al. 2005). Knockout of Smad3 blocks TGF-b-induced EMT in primary tubular epithelial cells; and the reduction of Smad2 and Smad3 function are associated with the decreased metastatic potential of breast cancer cell lines in a xenograft model (Zavadil et al. 2004). It is interesting that SMAD3 and SMAD4 interact and form a complex with Snail and target to the promoters of CAR (a tight-junction protein) and E-cadherin during TGFb-inducing EMT in breast epithelial cells (Vincent et al. 2009). Bos et al. (2009) identified that TGFb primed cancer cells for lung metastasis through angiopoietin-like 4 via Smad signaling pathway. On contrast, inhibition of TGFb or TGFb receptor reduces the invasive and metastatic activities of cancer cells. TGFb can also downregulate various epithelial molecules, including E-cadherin, ZO-1, and several specific keratins; and it also upregulates certain mesenchymal proteins such as fibronectin, fibroblast specific protein 1, a-smooth muscle actin and vimentin. In addition, TGFb cooperates with numerous kinases such as RAS, MAPK, p38MAP to promote EMT (Zavadil and Bottinger 2005; Buijs et al. 2007). More specifically, p38 MAPK and RhoA mediate an autocrine TGFb-induced EMT in NMuMG mouse mammary epithelial cells (Bhowmick et al. 2001). ECM molecules, such as integrin b1 and Fibulin-5, augment TGFb-induced EMT in a MAPK-dependent mechanism (Bhowmick et al. 2001; Lee et al. 2008). Constitutive activation of Raf enhances the function of TGFb in inducing EMT via MAPK in MDCK cells (Janda et al. 2002). TGFb also induces EMT through changes in the expression of certain cell polarity molecules. For example, TGFb can induce phosphorylation of Par6, which in turn stimulates binding of Par6 to E3 ligase Smurf1. The Par6–Smurf1 complex then mediates the localized ubiquination of RhoA to disrupt tight junctions during EMT (Ozdamar et al. 2005). TGFb can also downregulate the Par3 expression to destroy the cell polarity (Wang et al. 2008). It is interesting to note that Abl can inhibit TGFb-mediated EMT in normal and metastatic mammary epithelial cells (MECs) (Allington et al. 2009). Furthermore, TGFb can cooperate with other oncogenic pathways such as Notch, Wnt/b-catenin, and NF-kB to maintain the mesenchymal phenotype of invasive/metastatic tumor cells (Nawshad et al. 2005; Zavadil and Bottinger 2005; Neth et al. 2007). The Wnt/b-catenin pathway has a particularly tight link with EMT (Li et al. 2000). On one hand, b-catenin is an essential component of adherent junctions, where it provides the link between E-cadherin and a-catenin and modulates cell– cell adhesion and cell migration. On the other hand, b-catenin also functions as a transcription cofactor with T cell factor (TCF). Nuclear translocation of b-catenin can activate the expression of Slug and thus induces EMT. Expression of b-catenin in oocyte induces a premature EMT in the epiblast, concomitant with Snail transcription. Interestingly, Snail is a highly unstable protein and is dually regulated

198

Y. Wu and B.P. Zhou

by protein stability and cellular location. We showed that GSK-3b binds and phosphorylates Snail at two consensus motifs to dually regulate the function of this protein: phosphorylation at the first motif regulates its ubiquitination mediated by b-Trcp, whereas phosphorylation at the second motif controls its subcellular localization (Zhou et  al. 2004). Thus, Wnt can suppress the activity of GSK-3b and stabilizes the protein level of Snail and b-catenin to induce EMT and cancer metastasis (Yook et  al. 2005, 2006). Meanwhile, Snail can functionally interact with b-catenin to increase Wnt-dependent target gene expression to promote EMT (Stemmer et al. 2008). Increasing evidence indicates that Wnt signaling is strongly associated with human basal-like breast cancer. Inhibiting Wnt signaling through LRP6 reduces the capacity of cancer cells to self-renew and colonize in vivo, and results in the re-expression of breast epithelial markers and repression of EMT transcription factors Slug and Twist (DiMeo et al. 2009). How the synergistic activation of Snail and b-catenin by Wnt signaling pathway enhances EMT and metastasis remains further defined. Notch is an evolutionarily conserved signaling pathway that regulates cell fate specification, self-renewal and differentiation in embryonic and postnatal tissues. Four Notch (Notch 1−4) and five ligands (Jagged1, 2 and Deltalike1, 3, 4) have been identified. Notch signaling is normally activated followed by ligand–receptor binding between two neighboring cells, Notch undergoes intramembrane cleavage by g-secretase and its intracellular domain (NICD) is released and translocates to the nucleus to activate gene transcription by associating with Mastermind-like 1 (MAM) and histone acetyltransferase p300/CBP. Alteration of Notch signaling has been associated with various types of cancer in which Notch can act as an oncogene or as a tumor suppressor depending on the cellular context. The first observation that Notch pathway is required for EMT was derived from cardiac valve and cushion formation at heart development (Timmerman et al. 2004). This implies that Notch, acting through a similar mechanism, induces EMT during tumor progression and converts polarized epithelial cells into motile and invasive ones (Grego-Bessa et al. 2004). Indeed, overexpression of Jagged1 and Notch1 induces the expression of Slug and correlates with poor prognosis in various human cancers (Leong et al. 2007). Slug is essential for Notch-mediated EMT by repressing E-cadherin expression, which results in b-catenin activation and resistance to anoikis. Inhibition of Notch signaling in xenografted Slug-positive/E-cadherin-negative breast tumors promotes apoptosis and inhibits tumor growth and metastasis (Leong et al. 2007). In addition, Notch signaling deploys two distinct mechanisms that act in synergy to control the expression of Snail (Sahlgren et al. 2008). First, Notch directly upregulates Snail expression by recruitment of the Notch intracellular domain to the Snail promoter, and second, Notch potentiates hypoxia-inducible factor 1a (HIF-1a) recruitment to the lysyl oxidase (LOX) promoter and elevates the hypoxia-induced upregulation of LOX, which stabilizes the Snail protein. Thus, Notch signaling is required to convert the hypoxic stimulus into EMT and increases the invasiveness of tumor cells. In addition, Notch signaling pathway is involved in the acquisition of EMT phenotype of gemcitabine-resistant (GR) cells in pancreatic cancer (Wang et al. 2009). Downregulation of Notch signaling is associated with decreased invasive behavior of GR

9  Epithelial–Mesenchymal Transition in Development and Diseases

199

cells. Moreover, Notch signaling leads to the increased expression of vimentin, ZEB1, Slug, Snail, and NF-kB and results in EMT. Thus, inhibition of Notch signaling by novel therapeutic strategies can be clinically important in overcoming drug resistance and EMT phenotype of tumor cells. The Hedgehog (Hh) signaling pathway was first identified in a large screen for Drosophila genes required for patterning of the early embryo (Hooper and Scott 2005; Jacob and Lum 2007). The Hh ligands, Sonic-, Desert-, and Indian Hh in vertebrates and Hh in Drosophila, are secreted proteins that undergo several posttranslational modifications to gain full activity. Key effectors of Hh signaling include zinc-finger proteins of the Gli1-3 transcription factors. Hh signaling can initiate cell growth, cell division, lineage specification, and axon guidance and can also function as a survival factor. Activation of Hh signaling also leads to EMT. In mouse epidermal cells or in rat kidney epithelial cells immortalized with adenovirus E1A, Gli1 rapidly induces transcription of Snail and promotes EMT (Li et al. 2006, 2007). Targeted expression of Gli1 in the epithelial cells of mammary gland of mice induces the expression of Snail and thus results in the disruption of the mammary epithelial network and alveologenesis during pregnancy (Fiaschi et  al. 2007). In addition, Hedgehog signals induce JAG2 up-regulation for Notch-CSLmediated Snail expression; on the other hand, Hedgehog induces TGFb1 secretion to induce ZEB1 and ZEB2 expression through TGFb and NF-kB pathways. Conversely, blockade of Hedgehog signaling by inhibitor cyclopamine suppresses pancreatic cancer invasion and metastasis through inhibiting EMT (Feldmann et al. 2007). The crosstalk between the Hh and EMT also presents in human esophageal squamous cell carcinoma (ESCC) (Isohata et  al. 2009). Hh and EMT signaling genes are co-expressed on the undifferentiated esophageal epithelial cells and most ESCCs. These findings suggest that the mesenchymal gene expression is maintained or strengthened through Hh signaling in cancer cells.

Cytokines TNFa, a key inflammatory cytokine, plays a central role in the tumor progression. Constitutive expression of the TNFa from tumor microenvironment is a characteristic of many malignant tumors and its presence is often associated with poor prognosis. Several lines of evidence point to the tumor-promoting effects of TNFa in inflammation-driven tumorigenesis. First, overexpression of TNFa confers migratory and invasive properties of many tumor cell lines (Rosen et al. 1991). Second, TNFa and TNFa receptor 1 (TNFR1) knock-out mice are resistant to chemicalinduced carcinogenesis in skin and liver metastasis in an experimental colon cancer model (Knight et  al. 2000; Arnott et  al. 2004). Third, various tumor-promoting effects of TNFa are further confirmed in enhancing tumor cell motility, activating oncogenic pathways, and triggering EMT. TNFa can also promote breast cancer cell migration through up-regulation LOX (Liang et al. 2007). Endogenous TNFa contributes to the growth and invasiveness of primary pancreatic ductal adenocarcinoma

200

Y. Wu and B.P. Zhou

and anti-TNFa inhibit metastasis of these tumors (Egberts et al. 2008). Using RNA interference technology, Kulbe et  al. (2007) demonstrated that tumor growth and dissemination were significantly inhibited when TNFa production was blocked. In addition, TNFa can up-regulate SELECTIN and VCAM1 on endothelial cells that promote tumor cell adhesion and migration (Mannel et  al. 1994; Stoelcker et  al. 1995). Furthermore, TNFa enhances the invasive property of cancer cells by inducing EMT through Snail or ZEB1/ZEB2 (Chua et al. 2007; Chuang et al. 2008). In our recent study, we found that inflammatory cytokine TNFa is the major signal to induce Snail stabilization and EMT induction (Wu et  al. 2009). We showed that TNFa greatly enhanced the migration and invasion of tumor cells by inducing EMT program through NF-kB-mediated Snail stabilization. Knockdown of Snail expression not only inhibits TNFa-induced cancer cell migration and invasion in vitro but also suppresses LPS-mediated metastasis in vivo. Furthermore, knockdown of Snail expression not only blocks metastasis that is intrinsic to the metastatic breast cancer cells but also greatly suppresses inflammation-accelerated metastasis. Collectively, our study indicates that Snail stabilization and EMT induction mediated by the inflammatory cytokine TNFa are critical for metastasis. Our study provides a plausible molecular mechanism for tumor cell dissemination and invasion at the tumor invasive front. IL-6 is another important inflammatory cytokine linking inflammation and cancer. IL-6 transmits its signal through a common signaling receptor, gp130, expressed on many cell types. IL-6 binds to the sIL-6R receptor (gp80, present either on the cell surface or in solution), which then induces dimerization of gp130 chains resulting in activation of the associated Janus kinases (JAKs). JAKs phosphorylate gp130, leading to the recruitment and activation of the STAT3 and STAT1 transcription factors as well as other molecules (SHP2, Ras-MAPK, and PI3K) (Mumm and Oft 2008). The role of IL-6 in accelerating tumorigenesis is becoming clear as exogenous administration of IL-6 to mice during tumor initiation results in an increase in tumor burden and multiplicity (Grivennikov et al. 2009). IL-6 also enhances tumor proliferation in tumor-initiating intestinal epithelial cells (IECs) through NF-kB-IL-6-STAT3 cascade (Bollrath et  al. 2009; Bromberg and Wang 2009; Grivennikov et al. 2009). IL-6 can also act as an inducer of EMT in breast cancer cells. Ectopic expressing of IL-6 in breast adenocarcinoma cells exhibits an EMT phenotype characterized by suppressing E-cadherin expression and inducting vimentin, N-cadherin, Snail, and Twist (Sullivan et al. 2009). In addition, IL-6 also synergizes with EGF in inducing EMT through the activation JNK2/STAT3 in ovarian carcinomas (Colomiere et al. 2009). The interleukin-1 (IL-1) also promotes inflammatory processes and augments metastasis. There are two forms of IL-1 protein, IL-1a and IL-1b, and one antagonistic protein IL-1 receptor antagonist (IL-1ra). IL-1b is active solely in its secreted form, whereas IL-1b is active mainly as an intracellular precursor. IL-1 is abundant at tumor sites, where it affects the process of carcinogenesis, tumor growth and invasiveness, and the patterns of tumor–host interactions (Apte et al. 2006). Genetic ablation of IL-1b in mice results in the absence of metastatic tumors in  vivo (Voronov et al. 2003). Liver metastasis is almost completely inhibited in mice with

9  Epithelial–Mesenchymal Transition in Development and Diseases

201

deletion of interleukin-1b converting enzyme, which is required for the processing of IL-1b (Vidal-Vanaclocha et al. 2000). IL-1b also directly induces uPA expression and NF-kB activation, which results in the migration of A549 cells (Cheng et al. 2009).

Hypoxia A hypoxic condition has been detected in many human solid malignancies and it provides a strong selective advantage for tumor growth and survival by enhancing angiogenesis and metastasis. Under hypoxic conditions, cells respond by stabilizing hypoxia-inducible factor (HIF1a) that, in turn, forms the transcriptional complex with HIF1b and activates the transcription of downstream target genes through direct binding to the hypoxic response element located in the promoter of those genes that control glucose metabolism, angiogenesis, survival, and invasion. The well-recognized mechanism involved in hypoxia-induced metastasis is the activation of VEGF by HIF1a to promote angiogenesis and facilitate tumor metastasis. In addition, HIF1a can induce the expression of the chemokine receptor CXCR4 in renal cell carcinoma to promote organ specific metastatic dissemination (Staller et al. 2003). Although hypoxia and EMT are considered as crucial events favoring invasion and metastasis of many cancer cells, the two processes are long thought to involve few common molecular mechanisms. The link between hypoxia and EMT has been established recently as hypoxia induces the expression and coordinates the interplay of several EMT regulators. For example, HIF1a can regulate the expression of several EMT inducers such as Snail, Slug, Twist, ZEB1, and SIP1, directly or indirectly, and thus leads to the induction of EMT and enhances metastasis under hypoxic condition (Imai et  al. 2003; Krishnamachary et  al. 2003; Davidson and Sukumar 2005; Hotz et al. 2007; Yang et al. 2008). Interestingly, tumor hypoxia (measured by the expression of HIF1a) significantly correlated with both Snail and Twist expression and co-expression of these molecules correlated with the highest probability of metastasis and the worst prognosis in head and neck cancer (Yang et  al. 2008). Hypoxia also induces the expression of uPA, uPAR, MMP-2, and MMP-9 to increase metastatic invasion both in vitro and in vivo (Lunt et al. 2009). Furthermore, hypoxia also induces the expression of lysyl oxidas (LOX), which promotes Snail stabilization and activity (Peinado et al. 2005a, b). LOX is required for the maturation of newly synthesized collagen fibrils and promotes metastasis through changes in focal adhesion kinase activity. These results indicate that Snail not only can repress E-cadherin but also cooperates with LOX to induce EMT under hypoxic condition. Interestingly, EMT mainly occurs at the tumor invasion front (tumor–stroma boundary) while hypoxia mainly operates in the intratumoral areas where the pO2 level is below 10 mmHg. How these two events communicate with each other in  vivo remains to be further investigated. We speculate that the alterations within the tumor microenvironment as a consequence of hypoxia connect these two processes together.

202

Y. Wu and B.P. Zhou

EMT Generates Cancer Stem Cells In addition to its ability of enhancing cell migration during tissue morphogenesis and tumor metastasis, EMT creates cells that act as progenitors of many different tissues during development. For example, type 1 EMT generates mesoderm, which gives rise to a wide range of cell types, including muscle, bone, and connective tissues; type 1 EMT also creates neural crest delamination, which gives rise to glial and neuronal cells, adrenal glandular tissues, pigment-containing cells of the epidermis, and skeletal and connective tissues. These findings indicate that the genetic program of EMT possesses the ability to generate various cancer stem cells (CSC) in solid tumors. Interestingly, EMT characteristics are enriched in a subtype of basal-like breast cancer, which is highly metastatic and has CSC phenotype with the expression of high level of stem cell marker CD44+/CD24–. In addition, CD44+/ CD24– cells purified from normal and malignant breast cancer tissue exhibit features of an EMT, such as reduced expression of E-cadherin, significantly increased expression of fibronectin and vimentin, and robust levels of FOXC2, Twist, Snail, and Slug. Weinberg’s group demonstrated recently that EMT confers tumor cells with stem cell-like properties (Mani et al. 2007, 2008). Ectopic expression of Snail or Twist greatly increases the tumor cells’ ability to form tumorsphere, a property associated with mammary epithelial stem cells. In addition, TGFb, a potent inducer of EMT, also enhances the expression of CD44+/CD24– in tumor cells, implying that the surrounding tumor microenvironment is most likely the critical factor for metastasis by influencing the tumor cells undergoing EMT. Furthermore, breast tumors resistant to chemo- and endocrine therapies are enriched with cells bearing CSC signature and EMT markers in the post-treatment specimens. The findings that EMT is often seen at the invading edge of tumors and that EMT confers tumor cells with “metastatic cancer stem cell” traits highlight a notion that EMT is responsible for invasion and metastasis. Consistent with this notion, a distinct subset of highly migratory pancreatic stem cells co-expressing CD133+ and CXCR4 (CD133+/ CXC4+) found at the invasive edge of pancreatic carcinomas is essential for metastasis (Hermann et al. 2007). Interestingly, several signaling pathways that mediate stem cell self-renewal such as Wnt, Hh, and Notch also induce EMT. Thus, EMT endows cancer cells with stem cell properties for metastatic dissemination at primary tumor and self-renewal properties needed for initiating secondary tumor.

Genetic and Epigenetic Control of EMT Both EMT and epigenetic modification (DNA methylation and histone modifications) are dynamic and efficient processes during development, differentiation and carcinogenesis. Several lines of evidences indicate that gene expression involved in EMT is altered in tumors through genetic and epigenetic mechanisms. First, E-cadherin can be inactivated by genetic mutation or DNA hyper-methylation

9  Epithelial–Mesenchymal Transition in Development and Diseases

203

(epigenetic). Second, several EMT transcriptional factors are hyper-methylated in more differentiated tumor cells. For example, Bloushtain-Qimron et  al. isolated stem cell-like CD44+/CD24– and more differentiated epithelial CD44–/CD24+ cells from either normal or neoplastic breast tissues and compared their methylation profiles. They found that transcriptional factors (such as FOXC2) implicated in the induction of EMT and stem cell functions are hypomethylated and highly expressed in CD44+/CD24– cells compared with their CD44–/CD24+ counterparts (Bloushtain-Qimron et  al. 2008). In addition, sustained TGFb stimulation of immortalized human mammary epithelial cells induces EMT, which accompanies de novo DNA methylation on E-cadherin and estrogen receptor alpha (ERa) promoters, a prominent feature in human basal-like breast cancer (Dumont et  al. 2008). Furthermore, TGFb induces de novo DNA methylation on the promoter of E-cadherin and b4 integrin, a receptor for laminins of the basement membrane in mammary gland cells. These DNA methylations are accompanied by the decrease in histone activation marks (H3K9Ac and H3K4me3) and an increased histone repression modification on H3K27me3 (Yang et al. 2009). Together, these studies indicate that signals from the tumor microenvironment can modulate the induction of EMT through an epigenetic mechanism.

Micro RNA for EMT MicroRNA appears as a powerful component of the cellular signaling regulator of EMT (Table 9.2). MicroRNAs are small 20- to 22-nucleotide long noncoding RNAs that modulate gene expression at the post-transcriptional level (Stadler and Ruohola-Baker 2008; Stefani and Slack 2008). MicroRNAs have been implicated in regulating diverse cellular pathways, such as cell differentiation, proliferation, and programmed cell death, and are commonly dysregulated in human cancers. Recent findings suggest that microRNAs also contribute to EMT as either an enhancer or a suppressor. For example, Twist induces microRNA-10b, which inhibits the translation of HOXD10 and results in elevated expression of RhoC, and thus facilitates metastasis of breast tumor cells (Ma et al. 2007). Acting in the opposite direction, miR-31 inhibits breast cancer metastasis via coordinated repression of a cohort of metastasis-promoting genes, including RhoA (Valastyan et  al. 2009). Evidence for specific microRNAs in EMT regulation is further provided in that microRNA-200s, consisting of five members, are markedly downregulated in cells that have undergone EMT in response to TGFb (Burk et al. 2008; Gregory et al. 2008; Korpal et al. 2008). Because microRNA-200 directly targets the mRNA of ZEB1 and ZEB2/SIP1, expression of miR-200 induces upregulation of E-cadherin in cancer cell lines and suppresses their motility. Consistent with their role in regulating EMT, loss of microRNA-200 is commonly found in invasive breast cancer cell lines with mesenchymal phenotype and in regions of metaplastic breast cancer specimens lacking E-cadherin expression. Several recent reports overwhelmingly link the microRNA-200 family with EMT. These studies show that expression of

204

Y. Wu and B.P. Zhou

Table 9.2  MicroRNAs in EMT Name Function Targets miR-10b enhance HOXD10 miR-21 enhance PDCD4 TPM1 nmiR-29a enhance tristetraprolin miR-373 enhance CD44 miR-520c

enhance

CD44

miR-155 miR-126 miR-335 miR-205 miR-31 miR-146a miR-200

enhance inhibit inhibit inhibit inhibit inhibit inhibit

RhoA Crk, PI3K SOX4 and tenascin C ZEB1 and SIP1 RhoA Bcl-xl and STAT3 ZEB1 and SIP1

miR-138

inhibit

miR-221 miR-23b

inhibit inhibit

uPA and c-met

Cancer types Breast cancer Breast cancer Breast cancer Prostate, breast, and esophageal cancers, testicular germ cell tumors Prostate, breast, and esophageal cancers, testicular germ cell tumors Breast cancer Lung and colon cancer Breast cancer Breast cancer Breast cancer Breast, ovarian, bladder, pancreatic, and prostate cancers Head and neck squamous cell carcinoma Prostate cancer Hepatocellular carcinoma

microRNA-200 overcomes the resistance to EGFR therapy in bladder cancer cells and gemcitabine-resistant pancreatic cancer cells (Adam et al. 2009; Li et al. 2009). In addition, an EMT specific microRNA miR-21 is found in TGFb-induced EMT in human keratinocytes, a model of epithelial cell plasticity for epidermal injury and skin carcinogenesis (Zavadil et al. 2007). MiR-21 is abundantly expressed and associates with carcinogenesis. It targeted two tumor suppressors, tropomyosin 1 (TPM1) and programmed cell death-4 (PDCD4) to promote cell proliferation, microfilament organization, and anchorage-independent growth (Zhu et  al. 2007; Asangani et  al. 2008). MiR-155 also plays an important role in TGFb-induced EMT and cell migration and invasion through targeting RhoA (Kong et al. 2008).

Perspective Significant progress has been made in recent years regarding our understanding of EMT in development, fibrosis, and metastasis. Converging evidence from cell culture studies, transgenic mouse model, and documented immunohistochemical staining on human tumors, points out that EMT, amongst other ways, is the most important mechanism implicated in tumor metastasis and recurrence. It is clear that the induction of EMT is dependent on the signals that cells received from their microenvironment. Activation of several different EMT transcriptional factors, depending on the origin of the cells and the signals that cells received, induce a

9  Epithelial–Mesenchymal Transition in Development and Diseases

205

common genetic program of EMT. A recent link between EMT and cancer stem cells has sparked considerable interests and excitements, because it is widely accepted that only a minor population of tumor cells can initiate and support the development of tumors and that highly aggressive tumor cells share many of the characteristics of embryonic progenitor cells. The finding that EMT confers tumor cells with the CSC traits provides plausible molecular mechanism for tumor metastasis and recurrence. However, several challenges remain to be addressed regarding the fundamental mechanism and regulation of EMT. First, EMT provides a considerable degree of plasticity and reversibility for cells to differentiate or de-differentiate between cancer epithelial cells and mesenchymal cells (or CSC): why tumor cells have more plasticity than the normal differentiated epithelial cells? Do the mutation of tumor suppressors and/or activation of oncogenes during carcinogenesis endow tumor cells more intrinsic plasticity so that they are more sensitive to EMT inductive signals? Alternatively, does this plasticity come from stem or progenitor cells that actually initiate the tumor? And how the plasticity is regulated during tumor progress? Second, many studies are focused on EMT, however, the reversible event, the mesenchymal–epithelial transition, remains unclear. Uncovering the mechanism and signaling pathways of MET will help us to better understand the biology of EMT in development and metastasis. Third, EMT and MET are transient and dynamic events, particularly at the tumor invasive front, and a powerful imaging system that can capture this process will reveal the mystery of EMT. In summary, the association of EMT with tumor invasive front, disseminated cells, traits of CSC, and resistance to conventional therapies will place type 3 EMT in the spotlight for those wishing to control metastasis. A better understanding of this event will ultimately lead to new and improved therapies for patients with cancers. Acknowledgment  We apologize to the many contributors to this field whose works are important while we are unable to cite here. Our study is supported by the grants from NIH (RO1CA125454), the Susan G Komen Foundation (KG081310), and the Mary Kay Ash Foundation (to B.P. Zhou).

References Adam L, et  al (2009) miR-200 expression regulates epithelial-to-mesenchymal transition in bladder cancer cells and reverses resistance to epidermal growth factor receptor therapy. Clin Cancer Res 15(16):5060–5072. Allington TM, Galliher-Beckley AJ, Schiemann WP (2009) Activated Abl kinase inhibits oncogenic transforming growth factor-{beta} signaling and tumorigenesis in mammary tumors. FASEB J 23:4231–4243. Almand B, et  al (2001) Increased production of immature myeloid cells in cancer patients: a mechanism of immunosuppression in cancer. J Immunol 166(1):678–689. Ansieau S, et al (2008) Induction of EMT by twist proteins as a collateral effect of tumor-promoting inactivation of premature senescence. Cancer Cell 14(1):79–89. Apte RN, et al (2006) Effects of micro-environment- and malignant cell-derived interleukin-1 in carcinogenesis, tumour invasiveness and tumour-host interactions. Eur J Cancer 42(6): 751–759.

206

Y. Wu and B.P. Zhou

Arnott CH, et al (2004) Expression of both TNF-alpha receptor subtypes is essential for optimal skin tumour development. Oncogene 23(10):1902–1910. Arnoux V, et al (2008) Erk5 controls Slug expression and keratinocyte activation during wound healing. Mol Biol Cell 19(11):4738–4749. Asangani IA, et  al (2008) MicroRNA-21 (miR-21) post-transcriptionally downregulates tumor suppressor Pdcd4 and stimulates invasion, intravasation and metastasis in colorectal cancer. Oncogene 27(15):2128–2136. Bataille F, et al (2008) Evidence for a role of epithelial mesenchymal transition during pathogenesis of fistulae in Crohn’s disease. Inflamm Bowel Dis 14(11):1514–1527. Bhowmick NA, et al (2001) Transforming growth factor-beta1 mediates epithelial to mesenchymal transdifferentiation through a RhoA-dependent mechanism. Mol Biol Cell 12(1):27–36. Bloushtain-Qimron N, et  al (2008) Cell type-specific DNA methylation patterns in the human breast. Proc Natl Acad Sci U S A 105(37):14076–14081. Bollrath J, et al (2009) gp130-mediated Stat3 activation in enterocytes regulates cell survival and cell-cycle progression during colitis-associated tumorigenesis. Cancer Cell 15(2):91–102. Bos PD, et al (2009) Genes that mediate breast cancer metastasis to the brain. Nature 459(7249): 1005–1009. Bromberg J, Wang TC (2009) Inflammation and cancer: IL-6 and STAT3 complete the link. Cancer Cell 15(2):79–80. Bruyere F, et al (2009) Snail expression is an independent predictor of tumor recurrence in superficial bladder cancers. Urol Oncol Epub ahead of print. Buijs JT, et al (2007) TGF-beta and BMP7 interactions in tumour progression and bone metastasis. Clin Exp Metastasis 24(8):609–617. Bunt SK, et al (2007) Reduced inflammation in the tumor microenvironment delays the accumulation of myeloid-derived suppressor cells and limits tumor progression. Cancer Res 67(20): 10019–10026. Burk U, et al (2008) A reciprocal repression between ZEB1 and members of the miR-200 family promotes EMT and invasion in cancer cells. EMBO Rep 9(6):582–589. Chambers AF, Groom AC, MacDonald IC (2002) Dissemination and growth of cancer cells in metastatic sites. Nat Rev Cancer 2(8):563–572. Cheng CY, et  al (2009) IL-1 beta induces urokinase-plasminogen activator expression and cell migration through PKC alpha, JNK1/2, and NF-kappaB in A549 cells. J Cell Physiol 219(1):183–193. Christofori G (2006) New signals from the invasive front. Nature 441(7092):444–450. Chua HL, et  al (2007) NF-kappaB represses E-cadherin expression and enhances epithelial to mesenchymal transition of mammary epithelial cells: potential involvement of ZEB-1 and ZEB-2. Oncogene 26(5):711–724. Chuang MJ, et  al (2008) Tumor-derived tumor necrosis factor-alpha promotes progression and epithelial-mesenchymal transition in renal cell carcinoma cells. Cancer Sci 99(5):905–913. Colomiere M, et al (2009) Cross talk of signals between EGFR and IL-6R through JAK2/STAT3 mediate epithelial-mesenchymal transition in ovarian carcinomas. Br J Cancer 100(1): 134–144. Condeelis J, Pollard JW (2006) Macrophages: obligate partners for tumor cell migration, invasion, and metastasis. Cell 124(2):263–266. Cornil I, et  al (1991) Fibroblast cell interactions with human melanoma cells affect tumor cell growth as a function of tumor progression. Proc Natl Acad Sci U S A 88(14):6028–6032. Cowin P, Rowlands TM, Hatsell SJ (2005) Cadherins and catenins in breast cancer. Curr Opin Cell Biol 17(5):499–508. Davidson LA, et al (2006), Integrin alpha5beta1 and fibronectin regulate polarized cell protrusions required for Xenopus convergence and extension. Curr Biol 16(9):833–844. Davidson NE, Sukumar S (2005) Of Snail, mice, and women. Cancer Cell 8(3):173–174. Dawes-Hoang RE, et al (2005) Folded gastrulation, cell shape change and the control of myosin localization. Development 132(18):4165–4178.

9  Epithelial–Mesenchymal Transition in Development and Diseases

207

De Wever O, et  al (2004) Critical role of N-cadherin in myofibroblast invasion and migration in  vitro stimulated by colon-cancer-cell-derived TGF-beta or wounding. J Cell Sci 117(Pt 20):4691–4703. Diaz-Montero CM, et al (2009) Increased circulating myeloid-derived suppressor cells correlate with clinical cancer stage, metastatic tumor burden, and doxorubicin-cyclophosphamide chemotherapy. Cancer Immunol Immunother 58(1):49–59. DiMeo TA, et al (2009) A novel lung metastasis signature links Wnt signaling with cancer cell self-renewal and epithelial-mesenchymal transition in basal-like breast cancer. Cancer Res 69(13):5364–5373. Dumont N, et al (2008) Sustained induction of epithelial to mesenchymal transition activates DNA methylation of genes silenced in basal-like breast cancers. Proc Natl Acad Sci U S A 105(39): 14867–14872. Egberts JH, et al (2008) Anti-tumor necrosis factor therapy inhibits pancreatic tumor growth and metastasis. Cancer Res 68(5):1443–1450. Elloul S, et al (2005) Snail, Slug, and Smad-interacting protein 1 as novel parameters of disease aggressiveness in metastatic ovarian and breast carcinoma. Cancer 103(8):1631–1643. Feldmann G, et al (2007) Blockade of hedgehog signaling inhibits pancreatic cancer invasion and metastases: a new paradigm for combination therapy in solid cancers. Cancer Res 67(5): 2187–2196. Fiaschi M, et al (2007) Targeted expression of GLI1 in the mammary gland disrupts pregnancyinduced maturation and causes lactation failure. J Biol Chem 282(49):36090–36101. Flanders KC (2004) Smad3 as a mediator of the fibrotic response. Int J Exp Pathol 85(2):47–64. Franci C, et  al (2006) Expression of Snail protein in tumor-stroma interface. Oncogene 25(37):5134–5144. Gabrilovich DI, Nagaraj S (2009) Myeloid-derived suppressor cells as regulators of the immune system. Nat Rev Immunol 9(3):162–174. Greenburg G, Hay ED (1998) Cytoskeleton and thyroglobulin expression change during transformation of thyroid epithelium to mesenchyme-like cells. Development 102(3):605–622. Grego-Bessa J, et  al (2004) Notch and epithelial-mesenchyme transition in development and tumor progression: another turn of the screw. Cell Cycle 3(6):718–721. Gregory PA, et al (2008) The miR-200 family and miR-205 regulate epithelial to mesenchymal transition by targeting ZEB1 and SIP1. Nat Cell Biol 10(5):593–601. Gressner AM, et al (2002) Roles of TGF-beta in hepatic fibrosis. Front Biosci 7:d793–d807. Grivennikov S, et al (2009) IL-6 and Stat3 are required for survival of intestinal epithelial cells and development of colitis-associated cancer. Cancer Cell 15(2):103–113. Hartwell KA, et al (2006) The Spemann organizer gene, Goosecoid, promotes tumor metastasis. Proc Natl Acad Sci U S A 103(50):18969–18974. Hay ED (1995) An overview of epithelio-mesenchymal transformation. Acta Anat (Basel), 154(1):8–20. Hermann PC, et al (2007) Distinct populations of cancer stem cells determine tumor growth and metastatic activity in human pancreatic cancer. Cell Stem Cell 1(3):313–323. Hiratsuka S, et al (2006) Tumour-mediated upregulation of chemoattractants and recruitment of myeloid cells predetermines lung metastasis. Nat Cell Biol 8(12):1369–1375. Hiratsuka S, et al (2008) The S100A8-serum amyloid A3-TLR4 paracrine cascade establishes a pre-metastatic phase. Nat Cell Biol 10(11):1349–1355. Hooper JE, Scott MP (2005) Communicating with Hedgehogs. Nat Rev Mol Cell Biol 6(4): 306–317. Hotz B, et al (2007) Epithelial to mesenchymal transition: expression of the regulators snail, slug, and twist in pancreatic cancer. Clin Cancer Res 13(16):4769–4776. Huber MA, Kraut N, Beug H (2005) Molecular requirements for epithelial-mesenchymal transition during tumor progression. Curr Opin Cell Biol 17(5):548–558. Imai T, et al (2003) Hypoxia attenuates the expression of E-cadherin via up-regulation of SNAIL in ovarian carcinoma cells. Am J Pathol 163(4):1437–1447.

208

Y. Wu and B.P. Zhou

Isohata N, et al (2009) Hedgehog and epithelial-mesenchymal transition signaling in normal and malignant epithelial cells of the esophagus. Int J Cancer 125(5):1212–1221. Jacob L, Lum L (2007) Deconstructing the hedgehog pathway in development and disease. Science 318(5847):66–68. Janda E, et al (2002) Ras and TGF[beta] cooperatively regulate epithelial cell plasticity and metastasis: dissection of Ras signaling pathways. J Cell Biol 156(2):299–313. Joyce JA, Pollard JW (2009) Microenvironmental regulation of metastasis. Nat Rev Cancer 9(4): 239–252. Junghans D, Haas IG, Kemler R (2005) Mammalian cadherins and protocadherins: about cell death, synapses and processing. Curr Opin Cell Biol 17(5):446–452. Kalluri R (2009) EMT: when epithelial cells decide to become mesenchymal-like cells. J Clin Invest 119(6):1417–1419. Kalluri R, Neilson EG (2003) Epithelial-mesenchymal transition and its implications for fibrosis. J Clin Invest 112(12):1776–1784. Kalluri R, Weinberg RA (2009) The basics of epithelial-mesenchymal transition. J Clin Invest 119(6):1420–1428. Karnoub AE, et al (2007) Mesenchymal stem cells within tumour stroma promote breast cancer metastasis. Nature 449(7162):557–563. Knight B, et al (2000) Impaired preneoplastic changes and liver tumor formation in tumor necrosis factor receptor type 1 knockout mice. J Exp Med 192(12):1809–1818. Kolsch V, et al (2007) Control of Drosophila gastrulation by apical localization of adherens junctions and RhoGEF2. Science 315(5810):384–386. Kong W, et al (2008) MicroRNA-155 is regulated by the transforming growth factor beta/Smad pathway and contributes to epithelial cell plasticity by targeting RhoA. Mol Cell Biol 28(22):6773–6784. Korpal M, et al (2008) The miR-200 family inhibits epithelial-mesenchymal transition and cancer cell migration by direct targeting of E-cadherin transcriptional repressors ZEB1 and ZEB2. J Biol Chem 283(22):14910–14914. Krishnamachary B, et al (2003) Regulation of colon carcinoma cell invasion by hypoxia-inducible factor 1. Cancer Res 63(5):1138–1143. Kudo-Saito C, et al (2009) Cancer metastasis is accelerated through immunosuppression during Snail-induced EMT of cancer cells. Cancer Cell 15(3):195–206. Kulbe H, et al (2007) The inflammatory cytokine tumor necrosis factor-alpha generates an autocrine tumor-promoting network in epithelial ovarian cancer cells. Cancer Res 67(2):585–592. Lee YH, et al (2008) Fibulin-5 initiates epithelial-mesenchymal transition (EMT) and enhances EMT induced by TGF-beta in mammary epithelial cells via a MMP-dependent mechanism. Carcinogenesis 29(12):2243–2251. Leong KG, et  al (2007) Jagged1-mediated Notch activation induces epithelial-to-mesenchymal transition through Slug-induced repression of E-cadherin. J Exp Med 204(12):2935–2948. Liang M, Zhang P, Fu J (2007) Up-regulation of LOX-1 expression by TNF-alpha promotes transendothelial migration of MDA-MB-231 breast cancer cells. Cancer Lett 258(1):31–37. Lin EY, et  al (2001) Colony-stimulating factor 1 promotes progression of mammary tumors to malignancy. J Exp Med 193(6):727–740. Li X, et al (2006) Snail induction is an early response to Gli1 that determines the efficiency of epithelial transformation. Oncogene 25(4):609–621. Li X, et al (2007) Gli1 acts through Snail and E-cadherin to promote nuclear signaling by betacatenin. Oncogene 26(31):4489–4498. Li Y, et al (2009) Up-regulation of miR-200 and let-7 by natural agents leads to the reversal of epithelial-to-mesenchymal transition in gemcitabine-resistant pancreatic cancer cells. Cancer Res 69(16):6704–6712. Li Y, Hively WP, Varmus HE (2000) Use of MMTV-Wnt-1 transgenic mice for studying the genetic basis of breast cancer. Oncogene 19(8):1002–1009. Lunt SJ, Chaudary N, Hill RP (2009) The tumor microenvironment and metastatic disease. Clin Exp Metastasis 26(1):19–34.

9  Epithelial–Mesenchymal Transition in Development and Diseases

209

Ma L, Teruya-Feldstein J, Weinberg RA (2007) Tumour invasion and metastasis initiated by microRNA-10b in breast cancer. Nature 449(7163):682–688. Mani SA, et al (2007) Mesenchyme Forkhead 1 (FOXC2) plays a key role in metastasis and is associated with aggressive basal-like breast cancers. Proc Natl Acad Sci U S A 104(24): 10069–10074. Mani SA, et al (2008) The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell 133(4):704–715. Mannel DN et al (1994) Mechanisms involved in metastasis enhanced by inflammatory mediators. Circ Shock 44(1):9–13. Mantovani A, et al (2008) Cancer-related inflammation. Nature 454(7203):436–444. Moody SE, et al (2005) The transcriptional repressor Snail promotes mammary tumor recurrence. Cancer Cell 8(3):197–209. Mumm JB, Oft M (2008) Cytokine-based transformation of immune surveillance into tumorpromoting inflammation. Oncogene 27(45):5913–5919. Nawshad A, et al (2005) Transforming growth factor-beta signaling during epithelial-mesenchymal transformation: implications for embryogenesis and tumor metastasis. Cells Tissues Organs 179(1–2):11–23. Neth P, et al (2007) The Wnt signal transduction pathway in stem cells and cancer cells: influence on cellular invasion. Stem Cell Rev 3(1):18–29. Nieto MA (2002) The snail superfamily of zinc-finger transcription factors. Nat Rev Mol Cell Biol 3(3):155–166. Oda H, Tsukita S, Takeichi M (1998) Dynamic behavior of the cadherin-based cell-cell adhesion system during Drosophila gastrulation. Dev Biol 203(2):435–450. Olive KP, et al (2009) Inhibition of Hedgehog signaling enhances delivery of chemotherapy in a mouse model of pancreatic cancer. Science 324(5933):1457–1461. Ozdamar B, et al (2005) Regulation of the polarity protein Par6 by TGFbeta receptors controls epithelial cell plasticity. Science 307(5715):1603–1609. Pantel K, Brakenhoff RH (2004) Dissecting the metastatic cascade. Nat Rev Cancer 4(6):448–456. Peinado H, et al (2005a) A molecular role for lysyl oxidase-like 2 enzyme in snail regulation and tumor progression. EMBO J 24(19):3446–3458. Peinado H, Portillo F, Cano A (2005b) Switching on-off Snail: LOXL2 versus GSK3beta. Cell Cycle 4(12):1749–1752. Pollard JW (2004) Tumour-educated macrophages promote tumour progression and metastasis. Nat Rev Cancer 4(1):71–78. Qian F, et al (2005) Interaction between integrin alpha(5) and fibronectin is required for metastasis of B16F10 melanoma cells. Biochem Biophys Res Commun 333(4):1269–1275. Radisky DC, et al (2005) Rac1b and reactive oxygen species mediate MMP-3-induced EMT and genomic instability. Nature 436(7047):123–127. Radisky DC, Kenny PA, Bissell MJ (2007) Fibrosis and cancer: do myofibroblasts come also from epithelial cells via EMT? J Cell Biochem 101(4): 830–839. Rastaldi MP, et al (2002) Epithelial-mesenchymal transition of tubular epithelial cells in human renal biopsies. Kidney Int 62(1):137–146. Rosen EM, et al (1991) Tumor necrosis factor stimulates epithelial tumor cell motility. Cancer Res 51(19):5315–5321. Sahlgren C, et al (2008) Notch signaling mediates hypoxia-induced tumor cell migration and invasion. Proc Natl Acad Sci U S A 105(17):6392–6397. Salez F, et  al (1998) Transforming growth factor-beta1 in sarcoidosis. Eur Respir J 12(4): 913–919. Savagner P, et al (2005) Developmental transcription factor slug is required for effective re-epithelialization by adult keratinocytes. J Cell Physiol 202(3):858–866. Shook D, Keller R (2003) Mechanisms, mechanics and function of epithelial-mesenchymal transitions in early development. Mech Dev 120(11):1351–1383. Sinha P, et al (2007) Cross-talk between myeloid-derived suppressor cells and macrophages subverts tumor immunity toward a type 2 response. J Immunol 179(2):977–983.

210

Y. Wu and B.P. Zhou

Stadler BM, Ruohola-Baker H (2008) Small RNAs: keeping stem cells in line. Cell 132(4): 563–566. Staller P, et al (2003) Chemokine receptor CXCR4 downregulated by von Hippel-Lindau tumour suppressor pVHL. Nature 425(6955):307–311. Stappenbeck TS, Miyoshi H (2009) The role of stromal stem cells in tissue regeneration and wound repair. Science 2009. 324(5935):1666–9. Stefani G, Slack FJ (2008) Small non-coding RNAs in animal development. Nat Rev Mol Cell Biol 9(3):219–230. Stemmer V, et  al (2008) Snail promotes Wnt target gene expression and interacts with betacatenin. Oncogene 27(37):5075–5080. Stoelcker B, et al (1995) Role of adhesion molecules and platelets in TNF-induced adhesion of tumor cells to endothelial cells: implications for experimental metastasis. J Inflamm 46(3): 155–167. Studebaker AW, et al (2008) Fibroblasts isolated from common sites of breast cancer metastasis enhance cancer cell growth rates and invasiveness in an interleukin-6-dependent manner. Cancer Res 68(21):9087–9095. Sullivan NJ, et al (2009) Interleukin-6 induces an epithelial-mesenchymal transition phenotype in human breast cancer cells. Oncogene 28:2940–2947. Timmerman LA, et al (2004) Notch promotes epithelial-mesenchymal transition during cardiac development and oncogenic transformation. Genes Dev 18(1): p. 99–115. Valastyan S, et  al (2009) A pleiotropically acting microRNA, miR-31, inhibits breast cancer metastasis. Cell 137(6):1032–1046. Valcourt U, et  al (2005) TGF-beta and the Smad signaling pathway support transcriptomic reprogramming during epithelial-mesenchymal cell transition. Mol Biol Cell 16(4): 1987–2002. Vidal-Vanaclocha F, et al (2000) IL-18 regulates IL-1beta-dependent hepatic melanoma metastasis via vascular cell adhesion molecule-1. Proc Natl Acad Sci U S A 97(2):734–739. Villanueva S, et al (2002) Posteriorization by FGF, Wnt, and retinoic acid is required for neural crest induction. Dev Biol 241(2):289–301. Vincent T, et al (2009) A SNAIL1-SMAD3/4 transcriptional repressor complex promotes TGFbeta mediated epithelial-mesenchymal transition. Nat Cell Biol 11:943–950. Voronov E, et al (2003) IL-1 is required for tumor invasiveness and angiogenesis. Proc Natl Acad Sci U S A 100(5):2645–2650. Wang X, et al (2008) Downregulation of Par-3 expression and disruption of Par complex integrity by TGF-beta during the process of epithelial to mesenchymal transition in rat proximal epithelial cells. Biochim Biophys Acta 1782(1):51–59. Wang Z, et al (2009) Acquisition of epithelial-mesenchymal transition phenotype of gemcitabineresistant pancreatic cancer cells is linked with activation of the notch signaling pathway. Cancer Res 69(6):2400–2407. White LR, et al (2007) The characterization of alpha5-integrin expression on tubular epithelium during renal injury. Am J Physiol Renal Physiol 292(2):F567–F576. Willis BC, Borok Z (2007) TGF-beta-induced EMT: mechanisms and implications for fibrotic lung disease. Am J Physiol Lung Cell Mol Physiol 293(3):L525–L534. Wu Y, et al (2009) Stabilization of snail by NF-kappaB is required for inflammation-induced cell migration and invasion. Cancer Cell 15(5):416–428. Yamashita S, et  al (2004) Zinc transporter LIVI controls epithelial-mesenchymal transition in zebrafish gastrula organizer. Nature 429(6989):298–302. Yang J, et al (2004) Twist, a master regulator of morphogenesis, plays an essential role in tumor metastasis. Cell 117(7):927–939. Yang L, et  al (2008) Abrogation of TGF beta signaling in mammary carcinomas recruits Gr-1+CD11b+ myeloid cells that promote metastasis. Cancer Cell 13(1):23–35. Yang MH, et al (2008) Direct regulation of TWIST by HIF-1alpha promotes metastasis. Nat Cell Biol 10(3):295–305.

9  Epithelial–Mesenchymal Transition in Development and Diseases

211

Yang X, et al (2009) Regulation of {beta}4-integrin expression by epigenetic modifications in the mammary gland and during the epithelial-to-mesenchymal transition. J Cell Sci 122(Pt 14): 2473–2480. Yook JI., et al (2005) Wnt-dependent regulation of the E-cadherin repressor snail. J Biol Chem 280(12):11740–11748. Yook JI, et al (2006) A Wnt-Axin2-GSK3beta cascade regulates Snail1 activity in breast cancer cells. Nat Cell Biol 8(12):1398–1406. Young MR, Lathers DM (1999) Myeloid progenitor cells mediate immune suppression in patients with head and neck cancers. Int J Immunopharmacol 21(4):241–252. Zavadil J, Bottinger EP (2005) TGF-beta and epithelial-to-mesenchymal transitions. Oncogene 24(37):5764–5774. Zavadil J, et al (2004) Integration of TGF-beta/Smad and Jagged1/Notch signalling in epithelialto-mesenchymal transition. EMBO J 23(5):1155–1165. Zavadil J, et  al (2007) Transforming growth factor-beta and microRNA:mRNA regulatory networks in epithelial plasticity. Cells Tissues Organs 185(1–3):157–161. Zeisberg EM, et al (2007) Endothelial-to-mesenchymal transition contributes to cardiac fibrosis. Nat Med 13(8):952–961. Zeisberg M, et  al (2003a) BMP-7 counteracts TGF-beta1-induced epithelial-to-mesenchymal transition and reverses chronic renal injury. Nat Med 9(7):964–968. Zeisberg M, et al (2003b) Bone morphogenic protein-7 inhibits progression of chronic renal fibrosis associated with two genetic mouse models. Am J Physiol Renal Physiol 285(6):F1060–F1067. Zeisberg M, Neilson EG (2009) Biomarkers for epithelial-mesenchymal transitions. J Clin Invest 119(6):1429–1437. Zhou BP, et al (2004) Dual regulation of Snail by GSK-3beta-mediated phosphorylation in control of epithelial-mesenchymal transition. Nat Cell Biol 6(10):931–940. Zhu S, et  al (2007) MicroRNA-21 targets the tumor suppressor gene tropomyosin 1 (TPM1). J Biol Chem 282(19):14328–14336.

Chapter 10

Invasion and Metastasis Douglas M. Noonan, Giuseppina Pennesi, and Adriana Albini

Abstract  In 1889, Stephen Paget recognized the key role that the local microenvironment must play in the formation of metastases by malignant tumor cells. Later, the interactions of the metastatic cell with the various barriers and within the final target organ were likened to a tumor cell decathlon, where the final metastasis was the result of the ability of a single tumor cell to perform series of steps needed to lodge and grow in a distant site. As the clinical oncologist finds metastatic disease much more difficult to manage than the primary tumor, these cells were considered to have attained additional mutations that confirmed a unique invasive and metastatic phenotype. Recent studies have challenged that hypothesis, thus what was once an accepted dogma is now on uncertain terms. The role of the tumor microenvironment has now come to the forefront in understanding the metastasis; here we review the mechanisms by which invasive cells may disseminate out of the local tumor microenvironment to colonize distant tissues.

Tumors as Tissues Numerous studies on the tumor microenvironment suggest that we need to revise the definition of the term “cancer”, currently defined in classical terms as a malignancy derived from hyperproliferating tumor cells, to that of a disease of a tissue. To control cancer, we need to regard carcinogenesis and tumors inherently as phenomena that occur within the whole tissue, not in individual cancer cells. In this new perspective, the microenvironment becomes an integral, essential part of the cancer and its metastasis. The tumor microenvironment is the complex society of many cell types, including the “stroma”, fibroblasts of various phenotypes,

A. Albini (*) Oncology Research, Science and Technology Park, IRCCS MultiMedica, Milan and Castellanza, Italy e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_10, © Springer Science+Business Media, LLC 2010

213

214

D.M. Noonan et al.

myofibroblasts, endothelial cells and their precursors, pericytes, smooth muscle cells, and immune cells: neutrophils and other granulocytes (eosinophils and basophils), mast cells, T, B and natural killer lymphocytes, and antigen-presenting cells such as macrophages and dendritic cells. All these cells can participate in tumor progression (Balkwill and Mantovani 2001; Brigati et al. 2002; Coussens and Werb 2002; De Visser et  al. 2006; Hanahan and Weinberg 2000; Noonan et  al. 2008; Orimo et al. 2005). If the process of carcinogenesis and its end result, invasive and metastatic cancer, are viewed as a maladaptive response of an entire tissue or organ to both genetic and epigenetic stress, then knowledge and control of the immediate microenvironment within a developing tumor become as important as the corresponding knowledge and control of the dysfunctional epithelial cells within that tumor (Albini and Sporn 2007; Balkwill and Mantovani 2001; Hanahan and Weinberg 2000).

Metastatic Disease Any clinical oncologist knows that, with current therapy, patients rarely die from their primary tumors. Benign tumors are easily cured by surgery, and a competent surgeon can treat most confined tumor masses. It is when surgery is not an option that cancer does become problematic. Local invasion, recurrence and, most dreaded, metastasis are also associated with resistance to chemotherapy, difficulty or inability in surgical management, and ensuing disease progression that is still often fatal. Traditionally, invasion and metastasis have been viewed from a tumor cell-centric aspect. The problem was considered to be the tumor cell itself; it was able to detach from the primary tumor, invade locally, access lymphatic or hematic vessels, and use these as a highway to distant sites (Fig. 10.1a). The role of the tumor microenvironment in favoring invasion and metastasis has returned into focus only more recently; it is now recognized as a key mechanism controlling invasion and metastasis, as well as a therapeutic and prevention target.

Metastatic Cascades The traditional view of the metastasis process was often likened to a tumor cell decathlon (Crissman 1986; Fidler 1978). The transformed cell had to acquire additional mutations that permitted it to perform all the various tasks associated with the road to malignancy at levels that were sufficient for each task. The end result was thought to be a malignant cell with extensive additional mutagenesis, often then associated with enhanced drug resistance as well. The decathlon largely consisted of discrete steps that could be assayed individually in vitro. These are (1) uncontrolled growth, (2) loss of cell–cell adhesion, (3) acquisition of migratory potential, (4) acquisition of invasive capacity, (5) exit from the primary tumor mass,

10  Invasion and Metastasis

a Classic

215

b Stromal

c Social

Tumor Cell

Tumor Cell

MMP producing Metastatic Tumor Cell or Cancer Stem Cell

MMP producing Stroma and Inflammatory Cells

2a

2b

2a

4

2a

4

Mobilized liz d Bone o e Marrow w Cells ell

1 2a

3

3

4

VEGF, GMCSF, IL-6…

2b

2b

3

5

Tumor Cell

1 2b

3

5

Proliferative Tumor T m r Cell C Migratory rat ry Tumor o Cell ll

1

1

d Remote Control

5

4

5

6

Fig. 10.1  Potential mechanisms of metastases: (a) Classic metastatic cascade. (b) The involvement of stromal tissues in facilitating invasion, typical of epithelial-derived tumors. (c) Social networks where highly proliferative but poorly invasive tumor cells cooperate with highly invasive but poorly proliferative tumor cells. (d) The involvement of generation of the pre-metastatic niche that facilitates tumor cell extravasation and organ colonization. In the four scenarios the primary tumors are proliferating, apoptosis-resistant cells. (1) The acquisition of the ability to detach from the primary tumor and migrate. Outlined in blue: cells producing matrix proteinases, in particular metalloprotinases, which permit tumor cells to breach the vessel basement membranes and enter into either the hematic (2a) or lymphatic (2b) systems, often moving as “cords” or emboli (3). (4) The tumor cells extravasate, again using proteinases (blue). (5) Tumor cells then proliferate at the distant site to form a metastasis. In some cases (5) the first cells to arrive are not tumor cells but local inflammation (b), specifically invasive but poorly proliferative tumor cells (c, 6) or the premetastatic niche is formed by bone marrow cells mobilized by products released from the primary tumor that act on bone-marrow-derived cells to form foci in specific organs (d). In many cases, tumor cells may extravasate but lack sufficient stimuli to form proliferating metastasis (6)

(6) breaching of the vascular basement membrane, (7) intravasation into a vessel, (8) survival of turbulence and immune surveillance in the circulatory system, (9) extravasation at a distant site to then (10) invade into the local tissue to reconstruct the uncontrolled tissue growth cycle (Fig. 10.1a). Each of these steps was considered to be a key point along the cascade and the ability of any one individual or clonal tumor cell to perform any one step could be measured, a deficiency at any point was to be considered a block of malignant progression. Assays were developed to measure each of these steps individually in vitro (Albini and Benelli 2007; Albini et al. 1987b), thus permitting analysis of mechanisms and testing of agents targeted the specific steps. These assays include (1) cell proliferation assays, (2) colony formation in soft agar to show resistance to anoikis, (3) adhesion assays to demonstrate the capacity to interact with specific, or complex, extracellular matrix components,

216

D.M. Noonan et al.

(4) migration assays to demonstrate motility, in particular directional motility (chemotaxis), and (5) invasion assays to demonstrate the acquisition of the ability to degrade and breach extracellular matrix barriers. These assays permitted assessment of the tumor cells’ capacity to perform the various tasks within the malignancy marathon. Further, this also permitted rapid development of agents targeting these individual steps that could potentially be employed in therapeutic approaches, theoretically preventing the oncology patient from developing metastatic disease. Recently, the infinite replication paradigm of tumor cells has been considered within the context of several cell death pathways, which sometimes counterbalance, but other times enhance malignancy. Resistance to apoptosis and apoptotic signals is a characteristic of most tumor cells (Hanahan and Weinberg 2000). In normal cells, when the replication machinery detects excessive or un-repairable errors in the genome, death by apoptosis is engaged; this system is frequently disengaged in tumor cells (Letai 2008). Tumor cells are also under constant cellular stress. Hypoxia is almost always present in tumors, and plays an important role in induction of vascularization as well as enhancing the plasticity of cancer cells (Pouyssegur et al. 2006). This is associated with alterations in metabolism (dependence on glycolysis) (Kroemer and Pouyssegur 2008), redox imbalances, and acidification of the microenvironment (Tosetti et al. 2009). Autophagy is one of the normal cellular reactions to adverse environments and lack of nutrients, the cell literally eats its own organelles to conserve energy and remove damaged organelles, it is also activated in cancer cells and is a potential therapeutic target (White and DiPaola 2009). A major focus has also been devoted to tumor cell resistance to anoikis, a sort of death by loneliness evoked when normal cells detach from their substrate (Simpson et al. 2008).

Migration, Invasion, and Metastasis Tumor cells adhere to many different extracellular matrices through numerous integrin receptors; a summary of this area is beyond the scope of this review, and will be dealt with in more detail in other chapters. We were among the first to report the enhanced chemotaxis of tumor cells (Mensing et al. 1984), as well as involvement of integrin engagement in migration (Albini et  al. 1987a), concepts now widely accepted. A number of factors have been associated with tumor cell migration, from early “autocrine motility factors”, to extracellular matrix and coagulation cascade components and their fragments, to many growth factors, most of which also induce migration at even lower concentrations than that needed to induce cell proliferation (Albini and Benelli 2007). Finally, more recently the chemokines, a class of immune system migration factors, have also been implicated as key elements determining organ preferences in tumor metastasis (Balkwill 2004; Singh et al. 2007). Invasion is considered a key element in that it is a critical step in malignancy, has relative tumor cell specificity, and is potentially useful as a pharmacological

10  Invasion and Metastasis

217

target. Over 25 years ago, we developed one of the still most widely used invasion assays based on a tumor-derived matrix barrier (Matrigel or equivalents) within a porous membrane migration chamber (Albini et al. 1987b). One of the first molecular mechanisms postulated (Liotta et  al. 1982), and later demonstrated (StetlerStevenson et  al. 1989), was the existence of proteinases capable of degrading specific matrix barriers, such as the matrigel barrier, that would be associated with tumor cell invasion. Of these the matrix metalloproteinases (MMPs), a large family with over 20 members, and in particular those which are able to degrade type IV collagen – MMP2 and MMP9 – received a great deal of attention (Egeblad and Werb 2002), and will also be the focus of other chapters. Blockage of the MMPs with either endogenous inhibitors or, later, pharmacological MMP inhibitors demonstrated that this could drastically curb tumor cell invasiveness (Egeblad and Werb 2002). One of the “natural” inhibitors of MMPs and cell invasion is endogenous TIMP (Albini et al. 1991). These observations led to a campaign to develop MMP inhibitors for the clinic. Interestingly, it was precisely the MMPs that shed light on the role of the microenvironment in regulating tumor cell invasion and metastasis. The discovery that chemopreventive drugs, such as green tea flavonoids and N-acetyl-cystine (NAC), are metalloprotease inhibitors led to the initial suggestions of “angioprevention” mechanisms (Albini and Sporn 2007; Albini et  al. 2005; Tosetti et al. 2002); see below. The extracellular matrix and the stromal cells provide tumor cells with the tensile scaffold necessary for appropriate assembly into three-dimensional macroscopic structures. In addition, it behaves as a reservoir of growth factors and immunomodulatory cytokines, for instance the numerous heparin-binding factors able to interact with heparan sulfate proteoglycans embedded within the matrix. So the matrix supplies to cancer cells both positional cues and signals to determine cellular fate, including growth, differentiation, survival, and movement (Egeblad and Werb 2002; Tasso et al. 2009). Several different scenarios of the metastatic cascade can be envisioned (Fig. 10.1), which will be discussed here at different points. The original is that of the “classic” metastatic cascade (Fig. 10.1a). As a small benign tumor expands, tumor cells acquire secondary “hits” or genetic alterations that improve the “fitness” of the cell for undergoing detachment from the primary mass, invasion, and metastatic dissemination (Fidler 1990). This cell must be able to (1) migrate and make matrix proteinases, in particular metalloprotinases, which permit it to (2) breach the vessel basement membranes and enter into either the lymphatic or hematic system. Recent studies have suggested that tumor cells migrate in “cords” (Friedl and Gilmour 2009; Friedl and Wolf 2008), and this “collective migration” may also contribute to the formation of emboli that contain coagulation factors, platelets, and other vehicles as well (2). The formation of these “emboli” may help the tumor cell to survive the turbulence found within the vascular system (3), as well as help hide it from eventual immune cells. Once in the distant site, the tumor cell adheres to the endothelium or an area of exposed vascular basement membrane, and reverses the intravasation process to extravasate into the target tissue (4). There the cell resumes proliferation to form a new tumor as a metastasis (5). Thus one could also assume that the metastasis,

218

D.M. Noonan et al.

c­ lonally derived from a highly invasively “fit” cell, would show a distinct molecular profile from that of the primary tumor, which in contrast was selected mostly on the basis of proliferative capacity alone. This also led to the hunt for genes that characterized metastatic cells, as well as the anti-metastatic genes that prevented cellular invasion and diffusion (Eccles and Welch 2007; Lee et al. 1996; Steeg et al. 1988). This scenario of a metastatic cascade seemed to fit particularly well to tumors derived from mesenchymal cells such as fibrosarcomas, as well as other tumors such as melanomas. Interestingly, these appear to use principally hematic metastasis routes. This also fit well with most cell lines, which may have undergone selection for survival in this processes in the in vitro and in vivo models used to measure invasiveness. The tumor-cell-centric view often overlooked the fact that tumors are complex tissues of numerous cell types closely interacting, the proliferating tumor cell is only one of these, and may not even be the dominant cell type within the tumor mass. Tumors are tissues that undergo constant remodeling, and as such have a chronic inflammatory component and activation of mesenchymal cell support systems resulting in an altered biology and tissue homeostasis. The contrasts with the tumor-cell-centric views first came when examining the metastatic scenario in the context of epithelial tumors, such as breast, ovarian, and intestinal tumors. One surprise was that although MMPs were clearly made within the epithelial-derived tumors, in situ hybridization showed that the cells containing high levels of mRNAs for MMPs were the stromal cells; fibroblasts and inflammatory cells within the tumors, rather than the tumor cells themselves (Autio-Harmainen et al. 1993; Soini et al. 1993). Further studies demonstrated that the tumor cells were able to induce MMP production by other cells within the tumor microenvironment, and that they could also capture the MMPs on their cell surfaces to use for invasion (Egeblad and Werb 2002). Thus the second scenario (Fig. 10.1b) is stromal dependent; these cells make the MMPs, in particular, MMP2 (primarily from fibroblasts) and MMP9 (primarily from inflammatory cells), which allow the tumor cells to acquire these on their cell surfaces and to enter the vasculature. Moreover, it has been demonstrated that breast cancer cells prompt mesenchymal cells within the tumor stroma to produce a cytokine, CCL5, which acts on cancer cells to promote their invasion and metastasis. Remarkably, these stimulated cancer cells do not acquire a stable metastatic phenotype; rather they revert to their pre-malignant state in the absence of other stimuli from the microenvironment (Karnoub et  al. 2007). Interestingly, most epithelial-derived tumors show this pattern of stromal production of MMPs, and tend to metastasize via lymphatic routes. Unfortunately, the clinical trials of several MMP inhibitors did not do well in the clinic (Egeblad and Werb 2002). There were unexpected collateral effects (in particular, joint pain), and no benefit for the patients. This was likely due to several causes, including a much more complex biology than that anticipated (Egeblad and Werb 2002) and the lack of experience with combination therapy, the mode in which bevacizumab gained approval. A more detailed discussion of MMPs will be provided in other chapters. We also noted that the invasive tumor cell shares many of its properties with endothelial cells that are activated to enter into the angiogenic process (Fig. 10.2).

10  Invasion and Metastasis

219

Tumor Cells

Endothelial Cells

Proliferation Invasion n i

Proliferation rol e t on

Migration to

Basement se t Membrane Me a Target rg t Tissue u Morphogenesis orpho e i Differentiation

Chemokines Growth factors Altered matrix

VEGF, other Angiogenic factors Altered matrix

Fig. 10.2  Functional similarities in the processes of metastasis and angiogenesis

An endothelial cell in response to an angiogenic factor must undergo a series of steps that are similar to those utilized by metastatic cells (Fig. 10.3): (1) Entry into proliferation and migration toward the stimulus, both usually involving angiogenic growth factors such as VEGF and the relative tyrosine kinase receptor on the endothelial cell surface. (2) Adhesion to matrix components in the switch from a stable quiescent cell to a migratory cell, often involving the integrin avb3. (3) Activation of matrix degradation, mostly through release of MMP2. All these steps reflect those discussed for metastatic tumor cells (Figs. 10.2 and 10.3). In addition, endothelial cells undergo (4) reorganization and activation of a differentiation phase, and (5) recruitment of supporting cells (pericytes) returning to quiescence and eventual removal via apoptosis of collateral vessels in excess. Both tumor and endothelial cells can recruit fibroblasts and inflammatory cells as part of the invasive process (Figs. 10.2 and 10.3), and endothelial cells also take advantage of inflammation to promote angiogenesis (Albini et al. 2005; Balkwill and Mantovani 2001; De Visser et al. 2006; Noonan et al. 2008). Early on we found that the invasion assay we developed could be adapted to the assessment of induction of a migratory and invasive phenotype in endothelial cells (Thompson et al. 1991). This permitted the use of an in vitro system to both identify new angiogenic agents (for example, see Albini et al. 1995, 1994, 1996; Bussolino et  al. 1995) and as a screen for angiogenesis inhibitors (for example, see Albini et al. 1999; Cai et al. 1999; Iurlaro et al. 1998; Valente et al. 1998). Angiogenesis also appears to be a factor linked to metastatic dissemination. In many cancer types, high vessel density is a negative prognostic factor. The effects of angiogenesis are likely to be multiple: (1) Angiogenesis facilitates tumor expansion, thus acquisition of new genetic/epigenetic events favoring invasive capacity. (2) Angiogenesis is closely linked with inflammation, improving leukocyte recruitment as an indicator

220

D.M. Noonan et al.

Fibroblasts

TK K ligands an s

MMP-2

MMP-2 MMP- 9

αβ β

αβ

Cy to

kin

es

MMP-9

FGF2

Neutrophils Macrophages

VEGF GF

Flg-1 Fl

PlGF P F

β αβ

VEGFR2

VEGFR1 E F

MMP-2 MMP- 9

MMPs

αVβ3

activated endothelial cell αVβ3

Fig. 10.3  Mechanistic similarities and interactions in tumor and endothelial cell invasion. Tumor cells produce a wide variety of autocrine and paracrine growth factors that stimulate tyrosine kinases and other cell surface receptors. Endothelial cells are activated in response to stimuli arising from tumor cells or inflammatory cells. Tumors can either produce matrix proteases, such as MMP2 and to a lesser extent, MMP9, or induce the production of these factors by surrounding fibroblasts or immune cells, respectively. Endothelial cells produce MMP2 and, to a lesser extent, MMP9, and may take advantage of the proteinases produced by inflammation as well. Activation of integrins on the tumor and endothelial cells permits directional migration toward the stimuli, resulting in breaching of cellular compartments

of an effective pro-tumorigenic “M2” inflammatory environment associated with malignancy (Albini et al. 2005; Balkwill and Mantovani 2001). (3) Angiogenesis is associated with highly increased vessel permeability, hemorrhages, and open channels that allow even minimally invasive cells’ ready access to entry into the vascular system. In some cases, tumor cells have been found to line vessels in what is known as “vascular mimicry” (Maniotis et al. 1999) and can even acquire endothelial-like functions and characteristics (Pezzolo et al. 2007). Angiogenesis and lymphangiogenesis are often associated, and similar events in lymphangiogenesis provide access to this vascular compartment as well (Alitalo et al. 2005). (4) Not only the factors associated with angiogenesis, in particular VEGF, but also hypoxia-derived signals and chemokines, released at high levels, systemically mobilize bone marrow components that may either directly participate in vessel formation within the tumors (Grunewald et  al. 2006; Jin et  al. 2006), or, as discussed below, directly favor formation of a metastatic niche. Counter-intuitively, however, recent studies suggest that at least in some experimental cases blocking VEGF through

10  Invasion and Metastasis

221

a­ nti-angiogenic therapy can promote invasion and metastasis (Loges et al. 2009), thus complicating therapeutic approaches. Another intriguing aspect is that of dormant metastases. Clinically in some tumors such as breast cancer, metastatic disease can manifest even decades after surgery and remission of the primary disease. What triggers these cells to awaken is not known. Studies have suggested that extravasation is in fact a relatively efficient process, whereas that of formation of metastatic foci is much less so (Cameron et al. 2000). Addition of an angiogenic stimuli has been found to convert dormant experimental tumors to actively growing cancer (Almog et  al. 2009; Indraccolo et al. 2006; Moserle et al. 2009; Naumov et al. 2006a, b); inflammation may also be a trigger having similar effects.

Rethinking Metastasis As current theories of multistep carcinogenesis predicted that the metastatic cell was an evolution of the primary transformed cell to a genetically altered cell capable of performing the additional tasks associated with invasion and colonization, it was assumed that these cells would show a transcriptional profile distinct from that of the primary tumor. Further, it was also expected that since the hurdles along the metastatic decathlon were similar, the genetic responses would also be similar, thus metastases should show distinct profiles that were similar to each other. The use of microarrays to investigate these hypotheses in an unbiased manner rapidly created a crisis (Webb 2003). The transcriptome profiles of metastases were more similar to the primary tumor from which they were derived than to each other (Pomeroy et al. 2002; Singh et al. 2002; Van ’t Veer et al. 2002; Van De Vijver et al. 2002). One interpretation of this is that the tumors were simply “born” with metastatic potential; the transformation pathway of the cell led to acquisition of both tumorigenic potential and invasive/metastatic potential together (Webb 2003). This conclusion was in clear contrast to the established dogma of multistep carcinogenesis, and led to extensive rethinking and new investigations into if and how exactly to do tumors produce metastases (Lonneborg et  al. 2009; Lonning et  al. 2005; Pfeffer et al. 2003, 2009). Clinically, in some cases the available observations suggest that the “born metastatic” hypothesis could be active; many patients who present with a primary tumor often already have metastatic disease, even on occasion deriving from a relatively small tumor. On the other hand, in oncology the greatest benefit in clinical terms over recent years has been largely due to early detection through specific screening programs; clear examples are breast, prostate, colon, and melanoma. When detected early and removed, the patients have a much greater probability of disease-free survival, indicating that tumor removal indeed prevents tumor progression toward dissemination and metastasis. So what really is going on? One possible answer to the microarray dilemma was that by the time the tumor was detected clinically, it had already undergone much of the progression needed to become metastatic, thus the cells with metastatic potential could predominate in the

222

D.M. Noonan et al.

primary tumor when analyzed, therefore reflecting closely the transcriptome of the metastasis disseminated from it (Pfeffer et al. 2003, 2009). Recent investigations suggest that metastasis is likely to be much more complicated than a well-ordered decathlon. Among the key contributing factors may well be the genotype of the host itself (Hunter 2006; Hunter and Alsarraj 2009). Experimental studies using diverse murine strains have indicated that the host contributes extensively to the metastatic capacity derived from a specific mutational process (Hunter 2006; Hunter and Alsarraj 2009). This produces a theory that suggests that in one individual, a specific series of mutations may produce a primary tumor, but this tumor might lack sufficient interactions with the environment to disseminate. In another host, the same specific series of mutations instead produces a tumor that interacts with the microenvironment in a manner that favors metastatic dissemination. Potential points of interactions include local components such as fibroblasts and vascular cells, inflammation, immune skewing and immunosuppression. Other aspects may be systemic, involving the reaction of bone marrow components, dissemination of precursors, and systemic immune-suppression. In any case, the transformed primary tumors are similar in propensity for eventual metastases, while host components determine the metastatic destiny. Apart from the intense interactions within the constantly remodeling tumor tissue, there may be tumor cell subsets that show fitness for specific activities. The cancer “stem” cell, or “cancer initiating cell,” may be one of these (Gupta et al. 2009; Kakarala and Wicha 2008; Wicha et al. 2006). These cells may be either poorly differentiated cells that are able to remain in a more slowly proliferating stem cell niche which gives rise to the rapidly proliferating, partially differentiated cells that make up the majority of the tumor mass. Or they may represent the capacity of tumor cells to undergo reprogramming or “de-differentiation” toward more primitive precursor cell types. It is presumed either way that the successful metastasis must recruit these cells. One of the predictions of the multistep carcinogenesis model is that the “metastatic phenotype” might be a characteristic of only a very small subpopulation of the cells within the primary tumor and that the metastasizing cells acquired additional mutations that are not present in the primary tumor. These features would escape from detection by expression profiling (or any other genomic technique such as comparative genome hybridization on arrays) unless one analyzes each potentially unique cell population within a tumor, an essentially impossible enterprise. Recent theories suggest that meta­ stasis arises from cancer “stem cells” or “cancer-initiating cells” that might stay latent in the primary tumor and escape to form metastasis when local conditions induce their scattering, for instance in hypoxia. Once settled in their new location, they go back to giving rise to partially differentiated, highly replicative cells which may resemble that of the parental tumor. If we take cancer stem cells into account, the success of metastasis prediction by expression profiling of the primary tumor suggests that the primary and metastatic stem cells give rise to similar populations of partially differentiated tumor cells. The stem cell also appears to be particularly sensitive to the microenvironment. The tumor stroma is a facilitator of tumor progression, providing an environment in which stem cells can

10  Invasion and Metastasis

223

either stay quiescent or enter into asymmetric division, grow, and metastasize (Comoglio and Trusolino 2005). Another aspect is that tumors may build social networks. For example, what if one tumor cell type is not sufficient to produce a tumor? Different derivatives of tumor cells may interact with each other, and in some cases the diverse differentiation pathways taken during transformation may produce cells that do poorly alone but do well together (Bidard et al. 2008). In a simplified scenario as in Fig. 10.1c, we show two tumor cell types, one rapidly proliferating, capable of providing growth factors and a favorable environment also to the other cell type, which is specialized in invasion. The invasive cells may colonize tissues, but they would remain dormant in the absence of the proliferative cells. The proliferative cells enter the vasculature through the vessel alterations produced during the angiogenesis/ lymphangiogenesis process, where they can exit in areas preconditioned by the migratory/invasive cell subtype. Together they form a tumor colony of similar composition to the parental tumor. The ability of these two cells to meet in a distant site may be limited, potentially explaining the propensity of some tumors to produce metastatic disease from dormant micrometastases many years after removal of the primary tumor and extensive disease-free survival. Finally, recent experimental studies in vivo have suggested that tumors may send emissaries to prepare the ground for colonization (Kaplan et al. 2005; Steeg 2005). This concept has been named the “pre-metastatic niche” (Kaplan et al. 2006a, b; Wels et  al. 2008), and is one area under intense investigation. It is based on the observation of mobilization of bone marrow cells by tumors that localized in the organs that will later be colonized by the tumor. These precursors could be found in the tumor target tissues very early, before the tumor cells themselves arrived, as well as with the tumor cells at later time within micrometastases. Further, the tissues colonized were specific to those targeted by the tumor cells, and cell-free supernatants could essentially reproduce a similar colonization by the premetastatic bone-marrow-derived cells. Thus the scenario becomes one of remote control (Fig. 10.1d); the tumor produces soluble factors that mobilizes bone marrow precursors and then colonizes the target tissue. These produce specific alterations that then promote tumor cell colonization. The specific alterations could include deposition of novel matrix components, degradation of matrix barriers to facilitate tumor cell entry, and production of chemoattractant/chemokine profiles that draw the tumor cell through chemotaxis. As cancer is a highly heterogeneous disease, the mechanisms leading to invasion and metastasis are likely to be as heterogeneous. If one could predict that all of the potential scenarios are clinically relevant, the task then becomes an attempt to identify which process may be occurring, and whether this can be circumvented therapeutically. Apart from early detection through screening programs, another potentially successful approach to reduction of metasatic disease could be application of prevention, in particular, angioprevention. Angioprevention is a term that refers to the tendency of chemoprevention agents to target angiogenesis (Tosetti et al. 2002), thus slowing tumor growth and progression. Bland agents with no side effects could be used in the general population before cancer appears; aspirin and

224

D.M. Noonan et al.

other COX inhibitors have shown clinical benefit in those using these agents (Cuzick et al. 2009; Harris 2009; Kraus and Arber 2009; Wang and Dubois 2009). Diet is one aspect that can be readily modified (Albini and Sporn 2007; Chambers 2009) in this population (Berrino et al. 2007; Coleman et al. 2008). Somewhat more aggressive and efficacious agents may be warranted for higher risk populations, taking into account the risk–risk balance to keep patient benefit in mind (Albini and Sporn 2007). Finally, in oncologic patients, in remission following therapy, antiangiogenesis and anti-tumor approaches may be efficacious (Albini and Sporn 2007). However, as noted above, single target anti-angiogenic agents may in fact favor metastasis (Loges et al. 2009), although the effects on dormant disease remain to be determined. One simple therapeutic approach that is often overlooked, largely because of limited economic incentive for large clinical studies, is metronomic therapy (Kerbel 2007; Kerbel and Kamen 2004). Metronomic therapy, the use of low dose, high-frequency chemotherapy to target normal cell populations (endothelial cells, inflammation) within the tumor microenvironment (Browder et al. 2000; Kerbel and Folkman 2002), is advancing in phase II trials where it seems to be effective at prolonging disease-free survival while being well tolerated (for example, see Briasoulis et al. 2009; Brizzi et al. 2009; Dellapasqua et al. 2008; GarciaSaenz et  al. 2008). It is also relatively inexpensive and would act as a broad anti-angiogenic approach, potentially circumventing the pitfalls of single agent targeting. Combination of metronomic therapy with anti-inflammatory agents would also be of interest; however, substantial funding is needed for clinical trials to demonstrate clear benefit in those patients in remission following primary therapy.

Conclusions The process of metastasis remains an enigma of great clinical importance. Resolving this enigma would provide a key stride in prevention and management of cancer. Our knowledge of the mechanisms involved seems to have expanded faster than our capacity to comprehend fully the information available. However, the increasingly open-minded approach, and the realization that numerous pathways may exist and may even interact between them (some are summarized in Fig. 10.1) in reality holds great promise. The realization of the involvement of the tumor microenvironment has provided further stimuli in understanding the complex and multiple mechanisms. Novel approaches with transcriptome and proteome analyses, including microRNAs, and host genetics as well as more refined biological approaches will provide windows into how cancer cells spread, how to evaluate these clinically, and what may be the combined approaches to block the process, even before it starts. Acknowledgments  These studies were funded by the AIRC (Associazione Italiana per la Ricerca sul Cancro), the Ministero della Salute, the Compagnia di San Paolo, the Ministero dell’Università e della Ricerca, and the Università degli Studi dell’Insubria.

10  Invasion and Metastasis

225

References Albini A, Benelli R (2007) The chemoinvasion assay: a method to assess tumor and endothelial cell invasion and its modulation. Nat Protocols 2:504–511 Albini A, Sporn MB (2007) The tumour microenvironment as a target for chemoprevention. Nat Rev Cancer 7:139–147 Albini A, Allavena G, Melchiori A et al. (1987a) Chemotaxis of 3t3 and sv3t3 cells to fibronectin is mediated through the cell attachment site in fibronectin and a fibronectin cell surface receptor. J Cell Biol 105:1867–1872 Albini A, Iwamoto Y, Kleinman HK et  al. (1987b) A rapid in  vitro assay for quantitating the invasive potential of tumor cells. Cancer Res 47:3239–3245 Albini A, Melchiori A, Santi L et al. (1991) Tumor cell invasion inhibited by timp-2. J Natl Cancer Inst 83:775–779 Albini A, Fontanini G, Masiello L et al. (1994) Angiogenic potential in vivo by kaposi sarcoma cell-free supernatants and hiv1-tat product: inhibition of ks-like lesions by timp-2. AIDS 8:1237–1244 Albini A, Barillari G, Benelli R et al. (1995) Angiogenic properties of human immunodeficiency virus type 1 tat protein. Proc Natl Acad Sci U S A 92:4838–4842 Albini A, Soldi R, Giunciuglio D et al. (1996) The angiogenesis induced by hiv-1 tat is mediated by the flk-1/kdr receptor on vascular endothelial cells. Nat Med 2:1371–1375 Albini A, Florio T, Giunciuglio D et  al. (1999) Somatostatin controls kaposi’s sarcoma tumor growth through inhibition of angiogenesis. FASEB J 13:647–655 Albini A, Tosetti F, Benelli R et al. (2005) Tumor inflammatory angiogenesis and its chemoprevention. Cancer Res 65:10637–10641 Alitalo K, Tammela T, Petrova TV (2005) Lymphangiogenesis in development and human disease. Nature 438:946–953 Almog N, Ma L, Raychowdhury R et al. (2009) Transcriptional switch of dormant tumors to fastgrowing angiogenic phenotype. Cancer Res 69:836–844 Autio-Harmainen H, Karttunen T, Hurskainen T et al. (1993) Expression of 72 kilodalton type IV collagenase (gelatinase a) in benign and malignant ovarian tumors. Lab Invest 69:312–321 Balkwill F (2004) Cancer and the chemokine network. Nat Rev Cancer 4:540–550 Balkwill F, Mantovani A (2001) Inflammation and cancer: back to virchow? Lancet 357:539–545 Berrino F, De Angelis R, Sant M et al. (2007) Survival for eight major cancers and all cancers combined for european adults diagnosed in 1995–99: results of the Eurocare-4 Study. Lancet Oncol 8:773–783 Bidard FC, Pierga JY, Vincent-Salomon A et al. (2008) A “class action” against the microenvironment: do cancer cells cooperate in metastasis? Cancer Metastasis Rev 27:5–10 Briasoulis E, Pappas P, Puozzo C et al. (2009) Dose-ranging study of metronomic oral vinorelbine in patients with advanced refractory cancer. Clin Cancer Res 15:6454–6461 Brigati C, Noonan DM, Albini A et  al. (2002) Tumors and inflammatory infiltrates: friends or foes? Clin Exp Metastasis 19:247–258 Brizzi MP, Berruti A, Ferrero A et al. (2009) Continuous 5-fluorouracil infusion plus long acting octreotide in advanced well-differentiated neuroendocrine carcinomas. A phase II trial of the piemonte oncology network. BMC Cancer 9:388 Browder T, Butterfield CE, Kraling BM et al. (2000) Antiangiogenic scheduling of chemotherapy improves efficacy against experimental drug-resistant cancer. Cancer Res 60:1878–1886 Bussolino F, Arese M, Montrucchio G et al. (1995) Platelet activating factor produced in vitro by kaposi’s sarcoma cells induces and sustains in vivo angiogenesis. J Clin Invest 96:940–952 Cai T, Fassina G, Morini M et al. (1999) N-acetyl cysteine inhibits endothelial cell invasion and angiogenesis. Lab Invest 79:1151–1159 Cameron MD, Schmidt EE, Kerkvliet N et al. (2000) Temporal progression of metastasis in lung: cell survival, dormancy, and location dependence of metastatic inefficiency. Cancer Res 60:2541–2546

226

D.M. Noonan et al.

Chambers AF (2009) Influence of diet on metastasis and tumor dormancy. Clin Exp Metastasis 26:61–66 Coleman MP, Quaresma M, Berrino F et al. (2008) Cancer survival in five continents: a worldwide population-based study (concord). Lancet Oncol 9:730–756 Comoglio PM, Trusolino L (2005) Cancer: the matrix is now in control. Nat Med 11:1156–1159. Coussens LM, Werb Z (2002) Inflammation and cancer. Nature 420:860–867 Crissman JD (1986) Tumor-host interactions as prognostic factors in the histologic assessment of carcinomas. Pathol Annu 21(Pt 1):29–52 Cuzick J, Otto F, Baron JA et al. (2009) Aspirin and non-steroidal anti-inflammatory drugs for cancer prevention: an international consensus statement. Lancet Oncol 10:501–507 De Visser KE, Eichten A, Coussens LM (2006) Paradoxical roles of the immune system during cancer development. Nat Rev Cancer 6:24–37 Dellapasqua S, Bertolini F, Bagnardi V et al. (2008) Metronomic cyclophosphamide and capecitabine combined with bevacizumab in advanced breast cancer. J Clin Oncol 26:4899–4905 Eccles SA, Welch DR (2007) Metastasis: recent discoveries and novel treatment strategies. Lancet 369:1742–1757 Egeblad M, Werb Z (2002) New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer 2:161–174 Fidler I (1978) Tumor heterogeneity and the biology of cancer invasion and metastasis. Cancer Res 38:2651–2660 Fidler IJ (1990) Critical factors in the biology of human cancer metastases: 28th G. H. Clowes Memorial Award lecture. Cancer Res 50:613–618 Friedl P, Gilmour D (2009) Collective cell migration in morphogenesis, regeneration and cancer. Nat Rev Mol Cell Biol 10:445–457 Friedl P, Wolf K (2008) Tube travel: the role of proteases in individual and collective cancer cell invasion. Cancer Res 68:7247–7249 Garcia-Saenz JA, Martin M, Calles A et al. (2008) Bevacizumab in combination with metronomic chemotherapy in patients with anthracycline- and taxane-refractory breast cancer. J Chemother 20:632–639 Grunewald M, Avraham I, Dor Y et al. (2006) Vegf-induced adult neovascularization: recruitment, retention, and role of accessory cells. Cell 124:175–189 Gupta PB, Chaffer CL, Weinberg RA (2009) Cancer stem cells: mirage or reality? Nat Med 15:1010–1012 Hanahan D, Weinberg RA (2000) The hallmarks of cancer. Cell 100:57–70 Harris RE (2009) Cyclooxygenase-2 (cox-2) blockade in the chemoprevention of cancers of the colon, breast, prostate, and lung. Inflammopharmacology 17:55–67 Hunter K (2006) Host genetics influence tumour metastasis. Nat Rev Cancer 6:141–146 Hunter KW, Alsarraj J (2009) Gene expression profiles and breast cancer metastasis: a genetic perspective. Clin Exp Metastasis 26:497–503 Indraccolo S, Stievano L, Minuzzo S et  al. (2006) Interruption of tumor dormancy by a transient angiogenic burst within the tumor microenvironment. Proc Natl Acad Sci U S A 103:4216–4221 Iurlaro M, Benelli R, Masiello L et al. (1998) Beta interferon inhibits HIV-1 TAT induced angiogenesis; synergism with 13-cis retinoic acid. Eur J Cancer 34:570–576 Jin DK, Shido K, Kopp HG et al. (2006) Cytokine-mediated deployment of SDF-1 induces revascularization through recruitment of CXCR4+ hemangiocytes. Nat Med 12:557–567 Kakarala M, Wicha MS (2008) Implications of the cancer stem-cell hypothesis for breast cancer prevention and therapy. J Clin Oncol 26:2813–2820 Kaplan RN, Riba RD, Zacharoulis S et al. (2005) VEGFR1-positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche. Nature 438:820–827 Kaplan RN, Psaila B, Lyden D (2006a) Bone marrow cells in the ‘pre-metastatic niche’: within bone and beyond. Cancer Metastasis Rev 25:521–529 Kaplan RN, Rafii S, Lyden D (2006b) Preparing the “Soil”: the premetastatic niche. Cancer Res 66:11089–11093

10  Invasion and Metastasis

227

Karnoub AE, Dash AB, Vo AP et  al. (2007) Mesenchymal stem cells within tumour stroma promote breast cancer metastasis. Nature 449:557–563 Kerbel R, Folkman J (2002) Clinical translation of angiogenesis inhibitors. Nat Rev Cancer 2:727–739 Kerbel RS (2007) Improving conventional or low dose metronomic chemotherapy with targeted antiangiogenic drugs. Cancer Res Treat 39:150–159 Kerbel RS, Kamen BA (2004) The anti-angiogenic basis of metronomic chemotherapy. Nat Rev Cancer 4:423–436 Kraus S, Arber N (2009) Inflammation and colorectal cancer. Curr Opin Pharmacol 9:405–410 Kroemer G, Pouyssegur J (2008) Tumor cell metabolism: cancer’s achilles’ heel. Cancer Cell 13:472–482 Lee JH, Miele ME, Hicks DJ et al. (1996) Kiss-1, a novel human malignant melanoma metastasissuppressor gene. J Natl Cancer Inst 88:1731–1737 Letai AG (2008) Diagnosing and exploiting cancer’s addiction to blocks in apoptosis. Nat Rev Cancer 8:121–132 Liotta LA, Thorgeirsson UP, Garbisa S (1982) Role of collagenases in tumor cell invasion. Cancer Met Rev 1:277 Loges S, Mazzone M, Hohensinner P et  al. (2009) Silencing or fueling metastasis with VEGF inhibitors: antiangiogenesis revisited. Cancer Cell 15:167–170 Lonneborg A, Aaroe J, Dumeaux V et al. (2009) Found in transcription: gene expression and other novel blood biomarkers for the early detection of breast cancer. Expert Rev Anticancer Ther 9:1115–1123 Lonning PE, Sorlie T, Borresen-Dale AL (2005) Genomics in breast cancer-therapeutic implications. Nat Clin Pract Oncol 2:26–33 Maniotis AJ, Folberg R, Hess A et al. (1999) Vascular channel formation by human melanoma cells in vivo and in vitro: vasculogenic mimicry. Am J Pathol 155:739–752 Mensing H, Albini A, Krieg T et  al. (1984) Enhanced chemotaxis of tumor-derived and virustransformed cells to fibronectin and fibroblast-conditioned medium. Int J Cancer 33:43–48 Moserle L, Amadori A, Indraccolo S (2009) The angiogenic switch: implications in the regulation of tumor dormancy. Curr Mol Med 9:935–941 Naumov GN, Akslen LA, Folkman J (2006a) Role of angiogenesis in human tumor dormancy: animal models of the angiogenic switch. Cell cycle 5:1779–1787 Naumov GN, Bender E, Zurakowski D et  al. (2006b) A model of human tumor dormancy: an angiogenic switch from the nonangiogenic phenotype. J Natl Cancer Inst 98:316–325 Noonan DM, De Lerma Barbaro A, Vannini N et al. (2008) Inflammation, inflammatory cells and angiogenesis: decisions and indecisions. Cancer Metastasis Rev 27:31–40 Orimo A, Gupta PB, Sgroi DC et al. (2005) Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated sdf-1/cxcl12 secretion. Cell 121:335–348 Pezzolo A, Parodi F, Corrias MV et al. (2007) Tumor origin of endothelial cells in human neuroblastoma. J Clin Oncol 25:376–383 Pfeffer U, Noonan D, Albini A (2003) Re: Microarray studies challenge theories of metastasis. J Natl Cancer Inst 95:829 Pfeffer U, Romeo F, Noonan DM et al. (2009) Prediction of breast cancer metastasis by genomic profiling: where do we stand? Clin Exp Metastasis 26:547–558 Pomeroy SL, Tamayo P, Gaasenbeek M et al. (2002) Prediction of central nervous system embryonal tumour outcome based on gene expression. Nature 415:436–442 Pouyssegur J, Dayan F, Mazure NM (2006) Hypoxia signalling in cancer and approaches to enforce tumour regression. Nature 441:437–443 Simpson CD, Anyiwe K, Schimmer AD (2008) Anoikis resistance and tumor metastasis. Cancer Lett 272:177–185 Singh D, Febbo PG, Ross K et al. (2002) Gene expression correlates of clinical prostate cancer behavior. Cancer Cell 1:203–209 Singh S, Sadanandam A, Singh RK (2007) Chemokines in tumor angiogenesis and metastasis. Cancer Metastasis Rev 26:453–467

228

D.M. Noonan et al.

Soini Y, Paakko P, Autio-Harmainen H (1993) Genes of laminin b1 chain, alpha 1 (IV) chain of type IV collagen, and 72-kD type IV collagenase are mainly expressed by the stromal cells of lung carcinomas. Am J Pathol 142:1622–1630 Steeg PS (2005) Cancer biology: emissaries set up new sites. Nature 438:750–751 Steeg PS, Bevilacqua G, Kopper L et al. (1988) Evidence for a novel gene associated with low tumor metastatic potential. J Natl Cancer Inst 80:200–204 Stetler-Stevenson W, Krutzsch H, Wacher M et al. (1989) The activation of human type IV collagenase proenzyme. J Biol Chem 264:1353–1356 Tasso R, Augello A, Carida M et  al. (2009) Development of sarcomas in mice implanted with mesenchymal stem cells seeded onto bioscaffolds. Carcinogenesis 30:150–157 Thompson E, Nakamura S, Shima T et  al. (1991) Supernatants of acquired immunodeficiency syndrome-related kaposi’s sarcoma cells induce endothelial cell chemotaxis and invasiveness. Cancer Res 51:2670–2676 Tosetti F, Ferrari N, De Flora S et al. (2002) Angioprevention’: angiogenesis is a common and key target for cancer chemopreventive agents. FASEB J 16:2–14 Tosetti F, Noonan DM, Albini A (2009) Metabolic regulation and redox activity as mechanisms for angioprevention by dietary phytochemicals. Int J Cancer 125:1997–2003 Valente P, Fassina G, Melchori A et  al. (1998) Timp-2 over-expression reduces invasion and angiogenesis and protects b16f10 melanoma cells from apoptosis. Int J Cancer 75:246–253 Van ’T Veer LJ, Dai H, Van De Vijver MJ et al. (2002) Gene expression profiling predicts clinical outcome of breast cancer. Nature 415:530–536 Van De Vijver MJ, He YD, Van’t Veer LJ et al. (2002) A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med 347:1999–2009 Wang D, Dubois RN (2010) The role of cox-2 in intestinal inflammation and colorectal cancer. Oncogene 29:781–788 Webb T (2003) Microarray studies challenge theories of metastasis. J Natl Cancer Inst 95:350–351 Wels J, Kaplan RN, Rafii S et al. (2008) Migratory neighbors and distant invaders: tumor-associated niche cells. Genes Dev 22:559–574 White E, Dipaola RS (2009) The double-edged sword of autophagy modulation in cancer. Clin Cancer Res 15:5308–5316 Wicha MS, Liu S, Dontu G (2006) Cancer stem cells: an old idea – a paradigm shift. Cancer Res 66:1883–1890; discussion 1895–1886

Chapter 11

Dormancy of Disseminated Tumor Cells: Reciprocal Crosstalk with the Microenvironment Paloma Bragado, Aparna C. Ranganathan, and Julio A. Aguirre-Ghiso

Abstract  Majority of cancer patients will die of metastases that develop from disseminated tumor cells (DTCs), months, years, or even decades after treatment. This pause in cancer progression suggests that, during these disease-free periods, DTCs may stop proliferating and survive in a dormant state. The mechanisms that determine whether tumor cells, after disseminating to target organs, will continue to proliferate, die or enter a protracted state of dormancy are poorly understood. Here, we review the different manifestations of dormancy and the experimental and clinical evidence supporting that the target organ microenvironment where DTCs lodge might influence the choice to enter dormancy. We also review the available animal models to study DTCs dormancy. This information is important to design strategies to maintain dormancy of DTCs or eradicate DTCs before they progress to overt metastasis. Such information would lead to anti-metastatic and/or metastasis preventive therapies which are urgently needed.

General Concepts on Tumor Cell Dormancy in the Context of Cancer Progression One of the hallmarks of cancer progression is the ability of tumor cells to move out of the primary site, invade adjacent tissues, and travel to distant sites to form discrete metastasis. Despite advances in early detection and diagnosis of cancer, as well as significant improvements in surgical techniques, general patient care and

J.A. Aguirre-Ghiso (*) Division of Hematology and Oncology, Departments of Medicine and Department of Otolaryngology, Tisch Cancer Institute, Mount Sinai School of Medicine, One Gustave L Levy Place, Box 1079, New York, NY 10029, USA e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_11, © Springer Science+Business Media, LLC 2010

229

230

P. Bragado et al.

local and adjuvant therapies, mortality rates from cancer are still high. Particularly, almost 90% of all cancer-related deaths are due to distant metastases (Slade et al. 2005; Weigelt et al. 2005). In traditional models of metastases, metastatic cells are extremely rare, and are thought to arise only during later stages of tumor progression when the primary tumor cells are deemed “metastasis capable” having accumulated all the characteristics (genetic and epigenetic changes) required to complete the metastatic process (Fidler 2003). However, this model has been challenged by recent results from a number of laboratories that have shown that cells with metastatic potential can disseminate early from the primary tumor, even in a premalignant stage, suggesting that dissemination and progression to metastatic growth can occur in parallel to the primary tumor expansion. Accordingly, Kang et  al. (2003) performed a global gene expression analysis of primary breast cancer and identified a group of genes that predicted for the development of distant metastases. These studies suggest that the genes and therefore the ability to metastasize are already present in sufficient abundance to be detected by gene arrays in primary tumors. In addition to this, recent studies have shown that oncogene activation in mouse mammary cells triggered the activation of twist, and in other studies this gene can induce dissemination of pre-invasive mammary tumor cells to lungs and bone marrow prior to the appearance of overt primary mammary tumors (Ansieau et al. 2008a, b; Husemann et al. 2008). The consequences of this new view of the metastatic cascade are that dissemination is no longer regarded as a selected mutational event and it is indeed considered an early event during cancer progression (Husemann et al. 2008). However, this does not negate that more progressed tumors will have disseminating cells and perhaps with enhanced efficiency. In fact 20–45% of patients with breast or prostate cancer will relapse years or even decades after treatment of the primary tumor, even with advanced tumors such as stage 3 breast cancer. The early dissemination theory is supported by findings from the Klein laboratory showing that DTCs can be detected in patients without metastasis and with lesions considered non-invasive (i.e., ADH, DCIS). These cells have fewer chromosomal aberrations than primary tumors or DTCs from patients with metastasis (Schmidt-Kittler et al. 2003). This suggests that in breast cancer, DTCs acquire the genomic aberrations typical of metastatic growth after having disseminated. Supporting evidence came from studies in non-invasive atypical ductal hyperplasia (ADH) and ductal carcinoma in situ (DCIS), which are considered to be ‘early’ in progression, where dissemination has already occurred (Klein 2008; Stoecklein et al. 2008). Why is this important for dormancy? These studies suggest that a part of the dormancy time is the time it takes for these DTCs to accumulate genetic alterations at distant sites and eventually give rise to metastasis. The study of the functional properties of these cells in  vitro, along with their molecular analysis, can provide valuable information about the mechanisms regulating dormancy of these cells. Solakoglu et  al. (2002) study chromosomal aberrations in BM-DTCs isolated from patients with breast cancer, and found that the genetic aberrations were different among different patients. This suggests that there is a high genetic heterogenicity in BM-DTCs of patients. This is supported by studies from Klein laboratory where they studied the genetic variation of disseminated

11  Dormancy of Disseminated Tumor Cells

231

tumor cells. They screened 525 bone-marrow, blood, and lymph-node samples from 474 patients with breast, prostate, and gastrointestinal cancers. What they found is that irrespective of the cancer type, there was a huge genetic difference between DTCs from MRD patients and DTC from patients with metastasis (Klein 2002; Klein et al. 2002). What is the clinical relevance of DTCs? As mentioned earlier, evidences suggest that dissemination to distant sites occurs even before primary tumor diagnosis (Husemann et  al. 2008). Due to their scarcity, these disseminated tumor cells (DTCs) usually remain undetected, however, sensitive immunocytochemical techniques and molecular assays have permitted specific detection of occult disseminated tumor cells even at the single cell stage (Pantel et  al. 2009; Wikman et al. 2008). These technologies allow monitoring tumor cell dissemination to the blood and bone marrow, two sites that can be easily sampled. The detection of these cells is associated with poor prognosis, and is predictive of metastases to the lung, liver, and BM (Pantel et al. 2009). Various clinical studies have given evidence that there is an association between the presence of DTC at the time of primary tumor surgery and the metastatic relapse in patients with breast, prostrate, lung, and gastrointestinal cancer among others (Alix-Panabieres et al. 2005; Bidard et al. 2008; Braun and Naume 2005). It has been proposed that these DTCs that constitute the minimal residual disease (MRD) may remain in a dormant state over the years until the conditions are propitious to resume proliferation (Aguirre-Ghiso 2007; Pantel et al. 2009; Pantel and Brakenhoff 2004). In fact, after diagnosis of the primary lesions, patients spend more time harboring residual disease than expanding lesions (metastases once overt usually grow rapidly) (Fig. 11.1). This highlights the importance of understanding the biology of residual disease, as it is an unmet opportunity to prevent the development of metastasis. Studying dormancy and residual disease is in its deepest sense the clearer avenue to the development of metastasis preventive strategies. Thus, maintaining dormancy of DTC or more simply eradicating these dormant tumor cells would represent a significant clinical breakthrough. Tumor dormancy can be broadly separated into two categories: One type is the dormancy of micrometastasis and therefore of a tumor mass and the other is the dormancy of solitary disseminated tumor cells. Detailed descriptions of these types of dormancy have been thoroughly reviewed elsewhere (Aguirre-Ghiso 2007). Here, we provide a brief updated overview.

Dormancy of Micrometastasis Angiogenic Dormancy This type of dormancy also known as tumor mass dormancy is defined by a balance between proliferation and apoptosis of micrometastasis. As a result there is no net gain in tumor mass. But this does not imply that the tumor cells are non-proliferative, in fact it is the opposite. One such mechanism, known as pre-angiogenic dormancy,

232

P. Bragado et al.

Fig. 11.1  Tumor dormancy during cancer progression. Different manifestations of dormancy. Tumor cells that have accumulated genetic and epigenetic changes that provide a growth advantage will form a primary tumor (a). These cells will invade the surrounding tissue, intravasate either in the blood or lymphatic vessels, and disseminate via circulation to distant sites (a – bottom). Tumor cells that survive dissemination lodge in the target organs (e.g., BM, liver, lungs). After treatment, the primary tumor will regress and/or be removed; however, DTCs will still be detectable in the BM of patients during long periods of time and they constitute the minimal residual disease (b, c). The crosstalk between the DTCs and the new microenvironment will determine the fate of the DCTs. They can either die from apoptosis (d – red), enter dormancy, either as solitary DTCs (e) or as a micrometastatic lesion (f ), where there is a balance between proliferation

11  Dormancy of Disseminated Tumor Cells

233

occurs because of the inability of the cells to recruit new blood vessels (Fig. 11.1). Studies from Folkman’s and other laboratories have indicated that for primary tumors to grow beyond a size of 1–2 mm, they require the recruitment of new blood vessels via angiogenesis. This mechanism is known as the “angiogenic switch” (Folkman 2002; Gimbrone et al. 1972; Holmgren et al. 1995). However, in some tumors sometimes proliferation competent tumor cells may be incapable of recruiting new blood vessels resulting in a balance between the fraction of dying cells and dividing ones. Under these circumstances, these avascular micrometastases remain viable but with no net increase in tumor mass resulting now in the entire tumor mass becoming dormant. Studies have demonstrated that such angiogenic dormancy can result from a decrease in the expression of pro-angiogenic factors such as VEGF and PDGF (platelet-derived growth factor) and/or an increase in the expression of angiogenic inhibitors such as thrombspondin, endostatin, etc. (Folkman 2002; Folkman and Hanahan 1991; Naumov et al. 2006). However such pre-angiogenic micrometastases can emerge from dormancy if they acquire the ability to become vascularized either by down-regulating angiogenesis inhibitors or by upregulating proangiogenic factors (Fig. 11.1) (Folkman 2002; Folkman and Hanahan 1991). Alternatively, this loss of dormancy can also be induced by activating or inactivating pathways that promote or suppress angiogenesis. For instance, tumor suppressors such as p53, PTEN, and Smad4 were shown to inhibit angiogenesis by regulating thrombospondin synthesis (Dameron et al. 1994a, b; Schwarte-Waldhoff et  al. 2000; Su et  al. 2003; Volpert et  al. 1997; Wen et  al. 2001). Furthermore, expression of activated oncogene c-Ha-Ras in osteosarcoma and gastric cancer cells

Fig. 11.1 (continued) (pink) and apoptosis (red) and thus no net gain in tumor mass, or they can resume proliferation and give rise to a secondary tumor. (e) Cellular dormancy represents dormancy of single disseminated tumor cells, which enter a growth arrest (G0/G1) upon arriving to a new microenvironment. The majority of DTCs studies show that these cells lack markers of proliferation such as Ki67, suggesting a deep G0–G1 arrest. Other factors that can also define prolonged quiescence and dormancy. For example, HES1, a transcription factor that appears to be required for induction of quiescence but prevents the induction of senescence and/or differentiation in several cellular systems, could favor cellular dormancy. Alternatively, the activation of genes known as metastasis suppressor genes, could contribute to dormancy. For instance, KISS1 has been shown to inhibit metastasis by inducing cellular dormancy. (g) Angiogenic dormancy occurs because proliferation competent DTCs are not able to recruit new blood vessels which leads to a balance between proliferating (pink) and dying (red) cells, and as a result the entire tumor mass becomes dormant. Angiogenic dormancy can result from a decrease in the expression of pro-angiogenic factors such as VEGF and PDGF (platelet-derived growth factor) and/or an increase in the expression of anti-angiogenic factors such as thrombospondin. However, these micrometastases can undergo an angiogenic switch and emerge from dormancy if they acquire the ability to become vascularized either by down-regulating angiogenesis inhibitors or by upregulating proangiogenic factors. (h) Immunesurveillance prevents the expansion of proliferating DTCs cells via cytotoxic T-cell-mediated cell death, thereby maintaining a constant tumor mass. In order to escape this state of dormancy, DTCs can downregulate specific tumor-associated antigens or express co-stimulatory molecules that induce apoptosis of cytotoxic CD8+ T lymphocytes (Zhang et al. 2008)

234

P. Bragado et al.

induced the loss of angiogenic dormancy in otherwise non-angiogenic cell lines (Arbiser et al. 1997; Udagawa et al. 2002). This loss of dormant mass phenotype correlated with a Ras-induced VEGF expression (Rak et al. 1995). Indirectly Ras has also been shown to stimulate angiogenesis by down-regulating angiogenesis inhibitors such as thrombospondin (Sheibani and Frazier 1996; Watnick et al. 2003) While many of the above studies were carried out on primary tumor masses, it is unclear whether angiogenic dormancy of micrometastasis occurs in patients as clinical evidence for this is weak. More importantly, if such a state of dynamic proliferation and apoptosis can be sustained by one or multiple small micrometastatic masses for years or decades has not been proven. It is tempting to speculate that given the high degree of genetic heterogeneity and the epigenetic changes that these dividing cells might experience, the angiogenic switch should occur rather quickly. Alternatively, micrometastasis can also be kept in a dormant state by an active immune response. Immunity-Driven Dormancy of Micrometastasis Another form of tumor mass dormancy is mediated by the immune system through a mechanism known as immunesurveillance or cancer editing. This form of control of dormancy is indeed microenvironmental in nature as the immune system and its influence on stromal cells can maintain tumors from expanding (see below). In this case the immune system prevents the expansion of proliferating tumor cells via cytotoxic T-cell-mediated cell death, thereby maintaining a constant tumor mass (Fig. 11.1) (Quesnel 2008). For instance, studies found that in patients with nonHodgkin’s lymphoma, residual lymphoma cells were still present in a few of the patients 3–8 years after they underwent a tumor-targeted monoclonal antibody therapy (Davis et  al. 1998). Similarly, a significant number of non-proliferating residual leukemic cells were found several years after follow-up in patients with CML who had received INFg or allogenic bone marrow transplants (Chomel et al. 2000). Furthermore in several cases, cancers of donor origin have been reported in organ transplant recipients years and decades after the donor was in complete remission. One interpretation is that the tumor cells may have remained dormant in the donor and that the immunesuppression of the transplant recipients may have enabled these cells to escape from dormancy and proliferate (MacKie et al. 2003). Another evidence for the role of immune response in maintaining tumor dormancy is the observation that in colorectal cancer patients there was a strong positive correlation between immune infiltration in the tumor and the extent of tumor dissemination and clinical outcome (Galon et al. 2007). While the above clinical findings indicate a strong correlation between an immune-mediated control and tumor dormancy, the evidence is indirect and direct mechanistic evidence for this is lacking in the clinic. Current knowledge of the mechanisms of immune-mediated tumor dormancy is derived exclusively from experimental animal models. Using a mouse model of chemically induced sarcomas, Koebel et  al. (2007) demonstrated that adaptive immunity maintains occult tumors containing slow cycling cells at an

11  Dormancy of Disseminated Tumor Cells

235

equilibrium with cells undergoing apoptosis. Evidence for immune-mediated tumor dormancy was supported by the finding that anti-CD4/CD8, anti-IFNg, or anti-IL-12 induced escape from dormancy (Koebel et al. 2007). In the BCL1 lymphoma mouse model, about 70% of mice develop dormant tumors in the spleen (Uhr et al. 1991; Yefenof et  al. 1993a; Yefenof et  al. 1993b). This occurs via an anti-idiotype immune response mediated by CD8+ T cells and IFNg in collaboration with the humoral immunity that usually induces cell cycle arrest and apoptosis (Farrar et al. 1999). Additionally, Schirmacher et  al. showed that tumor-associated antigen (TAA) from residual dormant tumor cells is involved in maintaining high frequencies of long-term surviving memory CD8(+) T cells that help maintain mouse lymphoma cells at a low number in the bone marrow (Mahnke et al. 2005). In addition to its cytotoxic role in the promotion of tumor dormancy, the immune system can also maintain tumor cells in an occult or dormant state by affecting the tumor microenvironment. This occurs through anti-angiogenic mechanisms as many of the immune effectors such as IFN-g and IL-12 are potent angiogenesis inhibitors (Sidky and Borden 1987; Voest et al. 1995). Further, studies in PPAR-g knock-out mice have shown that the tumor-infiltrating neutrophils express high levels of thrombospondin that promoted tumor dormancy characterized by reduced vascularity and a balance between proliferation and apoptosis (Kaipainen et  al. 2007). Establishment of tumor dormancy by immunesurveillance can not only be mediated by the direct cytotoxic and cytolytic effects of CTLs on tumor cells, but can also be promoted indirectly by the T-cell-mediated destruction of stromal cells that present antigen released from the tumor cells in the tumor microenvironment. Recent studies from the Schreiber laboratory have elegantly demonstrated that a single adoptive transfer of T cells solely resulted in the destruction of only tumor antigen-presenting stromal myeloid derived suppressor cells (MDSCs), without killing the tumor cells (Zhang et al. 2008). Such an eradication of MDSCs, which can otherwise cause CTL inhibition and promote tumor growth, resulted in a longterm equilibrium between the host stromal compartment and the tumor cells, which subsequently lead to a long-term maintenance of a residual tumor mass (Zhang et  al. 2008). However, additional studies are needed to determine whether such immunesurveillance mechanisms of tumor mass dormancy are in fact operational in patients. In contrast to an efficient immune response in maintaining micrometastatic mass dormancy, solitary dormant disseminated tumor cells can persist by escaping the immune system. For instance, in the DA1-3b mouse acute myeloid leukemia model, dormant tumor cells evade the cytotoxic effects of the immune system, by overexpressing B7-H1 (also known as PD-L1 or CD274), which inhibits T-cell activation and CTL-mediated lysis (Quesnel 2006; Saudemont and Quesnel 2004). Additionally, tumor cells can escape the immune effectors and persist in a dormant state by decreasing the expression of adhesion molecules, required for interaction with CTLs that lead to tumor cell death (Mahnke et al. 2005) or by silencing tumor antigen expression either through the loss of expression or alteration in MHC molecules (Yamshchikov et al. 2005). Alternatively, dormant tumor cells can evade the immune system by resisting immune-mediated apoptosis (Fig. 11.1). Studies in the

236

P. Bragado et al.

DA1-3b model showed that methylation of SOCS1 gene resulted in progressive loss of expression and deregulation of the JAK-STAT pathway, leading to enhanced resistance to CTL-mediated apoptosis (Saudemont et al. 2007). Another mechanism of apoptosis resistance was observed in the BCL1 lymphoma dormancy model, wherein the cross-linking of membrane IgM by anti-idiotypic antibodies not only resulted in the induction of apoptosis but also in cell cycle arrest via overexpression of p21waf1, which could protect the dormant tumor cells from apoptosis (Marches et al. 1998). Thus, the interaction of dormant tumor cells with the immune system could have a bivalent effect – while on one hand an effective immune system could promote tumor dormancy, on the other, the ability of dormant DTCs to counteract the immune effectors could result in the persistence of dormant DTCs. An important question about DTCs undergoing cellular dormancy (see below) is if there is anything specific about quiescent tumor cells in their ability to regulate antigen presentation or be immunemodulatory or whether it is independent of their prolonged growth arrest.

Cellular Dormancy Unlike angiogenic or immune-mediated dormancy, cellular dormancy represents dormancy of single disseminated tumor cells, which enter a growth arrest (G0/G1) upon arriving in a new microenvironment (Fig. 11.1). The possibility that a prolonged but reversible G0–G1 arrest (quiescence) might explain DTC dormancy is due to the fact that majority of DTC studies across different cancers show that these cells lack markers of proliferation such as Ki67, suggesting a deep G0 or G1 arrest. Work by Luzzi et al. (1998) and Cameron et al. (2000) demonstrated that most of the tumor cells arrive at the secondary site survive but fail to proliferate. Such cells can remain in a non-proliferative dormant state even in the presence of a rich vascular network. However, some cells appear to survive only if in direct proximity to a blood vessel (Kienast et al. 2010). Unlike angiogenic dormancy and immunesurveillance, cellular dormancy or quiescence occur at the level of individual cells. As mentioned above such BM-DTCs are negative for proliferation markers, have been described to survive chemotherapy and hormonal therapy (Pantel et  al. 2009), and can persist in the bone marrow (BM) over many years post-surgery (Braun et al. 2009). Therefore, they are thought to be the precursors of metastasis. Understanding their biology is one of the most important challenges for the future of cancer research, because these cells carry the information for the development of the disease. The reasons why a DTC that arrives at a secondary site will either die, start growing, or enter a protracted state of dormancy are unknown, but it has been proposed that the stress caused by the treatment, an unfavorable microenvironment or, the dissemination process itself may force these cells to enter a reversible growth arrest as a mechanism to survive until the conditions favor growth. An important question that remains is what factors define prolonged quiescence and dormancy in addition to a G0–G1 arrest? While much

11  Dormancy of Disseminated Tumor Cells

237

work is needed to define these variables across different cancers, we could speculate on some generalities based on the available data. A prerequisite is that the mechanism drives a long-term growth arrest that is reversible. Given that senescence, differentiation and apoptosis are for the most part irreversible states, one would speculate that dormant tumor cells might be highly resistant to induction of these terminal cell fates. HES1 is a transcription factor that appears to be required for induction of quiescence and it prevents the induction of senescence and differentiation in several cellular systems (Sang L et al., 2008). Thus, molecular mechanisms dependent on this or other transcription factors or signaling pathways that results in a similar state of quiescence might help to functionally and molecularly define cellular dormancy (see also below). Upon dissemination, tumor cells entering the new microenvironment can also enter into a state of cellular dormancy via the activation of genes known as metastasis suppressor genes (MSGs). These genes by definition suppress metastatic growth by inhibiting different stages of the metastatic process. They were originally identified by their reduced expression in metastatic cancer cells as compared to their non-metastatic counterparts (Horak et al. 2008). Of the 23 MSGs characterized, four of these genes KISS1, Kai1, MKK4/7, and Nm23-H1 have been shown to inhibit metastasis by inducing cellular dormancy. KISS1 encodes for secreted polypeptides known as kisspeptins and were shown to induce prolonged dormancy of DTCs in an experimental melanoma model (Nash et  al. 2007). However, KISS1 induced dormancy in melanoma cells was found to be independent of its receptor grp54, suggesting that it is signaling via an alternate receptor. Kai1 or CD82, a tetraspanin protein was also shown to induce prolonged cellular dormancy of prostate cancer cells by binding to and interacting with the endothelial cell receptor DARC (Duffy antigen receptor for chemokines) (Bandyopadhyay et  al. 2006). These studies demonstrated that interaction of Kai1 with DARC suppressed metastasis in  vitro by inducing an irreversible senescence-like growth arrest that was associated with an upregulation in p21, a cell cycle inhibitor and a decrease in TBX2, an inhibitor of senescence (Prince et  al. 2004). Since dormancy implies reversibility whether a reversible form of this senescence could be operational in vivo needs to be determined. The other MSGs, Nm23-H1, and MKK4/7 were also shown to induce dormancy of micrometastasis by inhibiting Ras signaling or as in the case of MKK4 by activating JNK and p38 stress pathways (Steeg et  al. 2003) through yet unknown mechanisms. While strong activation of p38 pathways can induce apoptosis, more moderate levels of activation can induce growth arrest and/or survival mechanisms that favor subsequent tumor regrowth (see below). Experimental evidence indicates that Kai1 and KISS1 can promote cellular dormancy, meanwhile. MKK4/7 and Nm23-H1 were shown to induce dormancy of micrometastasis. However, this does not preclude the possibility that Kai1 and KISS1 may also suppress the growth of micrometastasis and MKK4/7 and Nm23-H1 can promote dormancy in single disseminated tumor cells. Therefore, identifying the pathways regulated by these metastasis suppressor genes could offer potential therapeutic targets to prevent metastatic growth.

238

P. Bragado et al.

Cellular Dormancy and the Microenvironment The tumor microenvironment plays a very critical role in modulating the tumor cell behavior. The fact that patients with advanced stage cancers can be free of disease for years and that they harbor DTCs suggests that these cells, despite having more genetic alterations, are still responsive to inhibitory signals present in the target organ microenvironment. Thus, such a tissue might have an instructive role in the induction of dormancy. It is also important to consider that dormancy might ensue if (1) there is a lack of appropriate signals for the tumor cells to resume growth (i.e., non-permissive microenvironment), (2) growth suppressive signals are encountered by DTCs that are responsive to these inhibitory cues, and (3) that DTCs might lack the appropriate surface receptors or their genes to integrate signals from a growth permissive microenvironment (Wikman et al. 2008). It is appreciated that the cross-talk between the surface receptors of these solitary tumor cells and the tumor microenvironment most likely dictates whether the newly seeded tumor cells will die, proliferate, or enter a state of cellular dormancy. For example, stromal fibroblasts from prostate tumors can induce tumor formation of non-transformed immortalized prostate epithelial cells when injected orthotopicaly in mice (Olumi et al. 1999). Moreover, transplantation of mammary epithelial cells bearing a weakly tumorigenic mutant-p53 from a non-irradiated to an irradiated mammary fat pad resulted in significant enhancement in tumor growth (Barcellos-Hoff and Ravani 2000). While these studies suggest that paracrine tumor–stroma interactions are critical components of tumor proliferation, they did not address its role in promoting exit from quiescence of DTCs and subsequent metastasis. Further, although these mechanisms are related to transformation (i.e., early cancer progression) it is possible that the stromal fibroblasts might influence entry or exit from dormancy. Researchers have demonstrated that reduced or aberrant signaling arising from deregulated interaction between adhesion molecules on the tumor cell and the stromal ECM could indeed promote tumor dormancy (Fig. 11.2) (White et  al. 2006). For example, studies from the Bissell laboratory have shown that the inhibition of b-1 integrin and EGF signaling in transformed human breast cancer cells induced a dormancy-like state in three-dimensional cell culture in vitro and a proliferation block in vivo upon transplantation into nude mice (Weaver et al. 1997). As opposed to the disorganized aggregates formed by these transformed cells in three-dimensional cell cultures in vitro, this revertant population induced by b1integrin blockade formed well-organized acinar structures associated with a reassembled basement membrane and reestablished cell–cell adhesions and the down-regulation of cyclin D1 and upregulation of p21 expression (Weaver et  al. 1997). In this study, a quasi-state of differentiation was achieved and because it was not shown to be reversible it is not clear if it represented a dormancy state. Furthermore, genetic ablation of b1 integrin in a transgenic mouse model to study the role of PyMT antigen or ERBB2 signaling did not result in cell death but rather induced a non-proliferating dormancy-like state (White et al. 2004).

Fig. 11.2  Potential mechanisms driving cellular dormancy in response to microenvironmentderived cues. Following intravasation (blood-borne or lymphatic routes), DTCs disseminate through circulation to target organs. Upon arrival at the secondary sites, a permissive microenvironment (Scenario a) in co-ordination with appropriate tumor cell surface receptors could allow metastatic cells to adapt and remodel their microenvironment to integrate growth-promoting signals and proliferate. (Scenario a) As an example (1) signals derived from fibronectin (FN) and transduced by the uPAR (metastasis associated urokinase receptor)–a5b1–integrin complex, focal adhesion kinase (FAK) and the epidermal growth factor receptor (EGFR) can result in extracellular signal-regulated kinase (ERK) activation and p38 inactivation thereby promoting metastatic proliferation. Transcriptional induction and activation of c-Jun and FoxM1 might mediate the G1 exit signals activated by ERK (124). In addition, circulating hormones present in the microenvironment (e.g., estrogen) (2) can also activate proliferative signals upon interaction and activation of hormone receptors (e.g., estrogen receptor on breast cancer cells) on the DTCs. (Scenario b) In contrast, loss of a surface uPAR, a5b1 integrin or EGFR (3) or inactivation of FAK (4) that transduce proliferative signals (e.g., from fibronectin) or interaction with a non-growth permissive ligand such as collagen-I (5) could result in stress signaling (low FAK–Ras–ERK, and high p38 activity) which in turn might lead to dormancy. This could in part be mediated by transcriptional activation of BHLHB3 and p53R213Q. This specific p53 mutation might enable growth arrest but avoid the induction of apoptosis (Adam et al. 2009). Additionally, collagen-mediated activation of DDR2 can lead to subsequent p16- and p21-mediated tumor cell growth arrest (6). However, whether the latter mechanism could lead to dormancy of DTC in  vivo needs to be examined. Increased levels of TGFb in the microenvironment could have either a tumor growth promoting (7 Scenario a) or growth suppressive role (7 Scenario b) depending on the genetics of the DTCs. For instance, in DTCs that lodge in the BM microenvironment (tissue with high TGFb levels), TGFb-mediated inhibition and activation of cycD1 and p21 and p15, respectively, could lead to DTC growth arrest and subsequent cellular dormancy. Alternatively, increase mitogenic signaling via TGFb-SMAD pathway or inhibition of metastasis suppressor genes such as BHLHB3 in co-operation with specific genetic alterations (p53R175H) could lead to metastatic proliferation (Adorno et al. 2009). It is likely that other unidentified pathways are also involved in determining the DTC cell fate upon arrival at the secondary site

240

P. Bragado et al.

Additional evidence for the role of deregulated ligand-dependent signaling through b1 integrin in promoting cellular dormancy came from our laboratory, where we examined the behavior of HEp3 head and neck squamous carcinoma in vivo. In this model, the cross-talk between the fibronectin receptor a5b1 integrin and uPAR sets up a positive feedback loop that leads to the recruitment of FAK and EGFR, which resulted in a strong and persistent activation of the mitogenic MEK-ERK MAPK pathway thereby promoting rapid cell proliferation in  vivo (Aguirre-Ghiso et al. 2001). In contrast, downregulation of uPAR lead to the inactivation of a5b1 and disruption of the uPA, uPAR, a5b1, FAK, and EGFR complex resulting in the deactivation of the ERK pathway and subsequent arrest of the cells in G0/G1 phase of cell cycle in vivo, triggering a cellular dormancy state that can last for months (Fig. 11.2) (Aguirre Ghiso et al. 1999; Kook et al. 1994). Such a state of dormancy could also be achieved by blocking signal transduction from the ECM and integrins via the interruption of FAK mitogenic signaling. Studies by Aguirre Ghiso (2002) showed that expression of FRNK, a dominant negative mutant of FAK in HEp3 cells resulted in inhibition of ERK activity and severely impaired their ability to proliferate in vivo without a subsequent increase in apoptosis and induction of dormancy. Later, a similar phenomenon of tumor dormancy was also observed upon mammary-specific ablation of FAK, which blocked progression of mammary hyperplasias to full carcinomas (Lahlou et al. 2007). Such FAK-deficient hyperplasias were still detectable in late-stage tumor samples resembling the induction of a cellular dormancy program (Lahlou et al. 2007). The inability of these cells to proliferate in  vivo highlights the central role of FAK activation and b1-FAK signaling complex in tumor growth (Fig. 11.2). Recently, the Green laboratory has elegantly shown that MLCK and cytoskeleton dynamics are essential to promote escape from dormancy (Barkan et al. 2008). They showed that tumor cells from different cancers that exhibited either a dormant or metastatic phenotype in  vivo however proliferated normally in two-dimensional culture conditions. However, upon culturing these cells in three-dimensional basement membrane cultures they displayed growth characteristics that recapitulated their in  vivo dormant or proliferative phenotype. They demonstrated that the dormant cells entered a cell cycle arrest that was associated with increased p16 and p27 levels. Furthermore, the switch from dormant to proliferative state was associated with fibronectin production, signaling through b1 integrin MLC phosphorylation that lead to cytoskeleton reorganization and actin stress fiber formation (Barkan et  al. 2008). A corroborative study of those presented above showed, using an experimental model of lung colony formation, that downregulation of FAK or b1 expression in mouse mammary carcinoma cells D2A1, resulted in a significant inhibition of proliferation of cancer cell colonies in the lungs obtained by direct intravenous injection (Shibue and Weinberg 2009). It became important to understand how this state of dormancy could be sustained for months and whether just the reduction of mitogenic signaling was sufficient to induce dormancy or whether other pathways had to be engaged. Further examination of the biochemical pathways downstream of the uPAR–integrin–EGFR surface complex showed that the establishment of the dormancy program in HEp3 cells

11  Dormancy of Disseminated Tumor Cells

241

required not only a powerful inhibition of ERK signaling but also the simultaneous activation of p38MAPK stress signaling pathway and robust induction of p21 (Aguirre-Ghiso et al. 2001). Activation of p38a was essential for persistent inhibition of ERK activation and induction of the expression of a number of transcription factors that lead to a prolonged and programmatic but reversible growth arrest. Activation of p38 resulted in the transcriptional downregulation of c-Jun and FOXM1, while it induced the TFs BHLHB3 (see below) and p53. Interestingly, in this model, p53 displays an R213Q mutation that appears to allow cells to switch of the apoptotic capacity of this TF while maintaining the induction of p21 (Fig. 11.2) (Pan and Haines 2000; and our unpublished data). These studies provided a novel insight into the opposing roles of both ERK and p38 signaling in the induction and maintenance of dormancy in HEp3 cells. While a low ERK to p38 activity ratio predicts for tumor dormancy, an inversion in the balance between these two opposing pathways predisposes the tumor cells to active proliferation (Aguirre-Ghiso et al. 2001). This, ratio or balance was found to be predictive of the in vivo behavior of various tumor types including prostate, breast, and fibrosarcoma indicating that it is not limited to the HEp3 model (Aguirre-Ghiso et al. 2003). Furthermore, targeted inhibition of EGFR with Gefitinib was shown to inhibit ERK and activate p38 (Sharma et al. 2006), suggesting that such a dormancy-inducing ERK/p38 ratio can also be achieved by using clinically relevant therapies. Further studies are needed to determine if the above-discussed mechanisms of cellular dormancy are in fact operational in disseminated tumor cells in patients. Nevertheless, some of these genes such as BHLHB3, FOXM1 (see below), and uPAR have important prognostic value and/or have been directly studied in DTCs. For example, expression of uPAR was found to be associated with dissemination of tumor cells to the bone marrow over time and the presence of uPAR expressing DTC in BM correlated with an unfavorable prognosis for patients with prostate and gastric cancer (Heiss et  al. 1995; Thomas et al. 2009). In addition, the presence of uPAR expressing cells in the primary tumor positively correlated with the presence of DTCs in the bone marrow of breast cancer patients (Hemsen et al. 2003; Pierga et al. 2005). Therefore, targeted therapies aimed at disrupting uPAR and/or integrin signaling complexes (Chaurasia et al. 2009) in DTCs could potentially force these cells into a dormancy like program and improve patient outcome.

Stroma-Associated Factors and Tumor Cell Dormancy Collagen Matrix Signaling The above data demonstrated that if tumor cells lack appropriate surface molecules and signals transducers for interpreting the ECM, they can activate pathways that lead to prolonged dormancy. But, could ECM molecules other than fibronectin for example have an active role in inducing a state of quiescence/dormancy? Research from DeClerck’s group found that melanoma cells plated on fibrillar collagen-I

242

P. Bragado et al.

entered a dormancy-like cell cycle arrest. This was mediated by the interaction and activation of discoidin domain receptor 2 (DDR2), a specific collagen tyrosine kinase receptor, by fibrillar collagen-I and subsequent increase in p15 and p21 levels (Henriet et al. 2000; Wall et al. 2005; Wall et al. 2007). Also, signals mediated by the collagen receptor a2b1 integrin specifically activated the p38a isoform leading to the upregulation of collagen gene transcription (Ivaska et  al. 1999). Moreover, induction of DDR2 by hypoxia was shown to be dependent on p38 activation (Chen et al. 2008). In agreement with these findings, we identified in our HEp3 tumor dormancy model, collagen-I and DDR2 as targets of p38SAPK-induced network of genes required for the induction of tumor cell quiescence. Therefore, although it remains to be tested it is possible that in addition to the above mechanisms, activation of p38 by collagen receptors could also activate a positive feedback loop that contributes to the quiescence as in the HEp3 dormancy model (Fig. 11.2).

Hormone Depletion and Dormancy Certain types of cancers such as breast, prostate, ovary, and endometrial cancer depend on the presence of circulating hormones in the microenvironment for their proliferation and survival. Therefore, strategies such as pharmacological or surgical depletion of endocrine hormones could promote a prolonged quiescence of tumor cells. In fact, such targeted therapies aimed at ablating primary tumors, do not completely eliminate residual tumor cells that can persist for prolonged periods of time. For example, reduction in estrogen receptor signaling can promote a dramatic regression of ER-positive tumors. In support of this, in studies done using invasive mammary carcinoma EMR-86 rat mammary carcinoma model, Wijsman et al. (1991) showed that the tumor outgrowth and metastasis of these cells in vivo was completely dependent on the presence of estrogen. Upon removal of estrogen albeit a rapid regression in the tumors, small dormant tumor remnants persisted in these animals, which could be restimulated to grow upon retreatment with the hormone (Wijsman et al. 1991). They further demonstrated that the in vivo tumor dormancy was not due to the residual tumor cells entering a growth arrest, but rather due to a balance between proliferation and apoptosis resulting in a tumor mass dormancy (Wijsman et al. 1991). The authors proposed that preexisting cancer stem cells within the growing tumor that rapidly proliferates in the presence of hormone could enter a state of dormancy upon hormone depletion. Another potentially important factor in endocrine-related dormancy is the active role of the tumor stroma in supporting growth. MMTV-induced pregnancy-dependent mammary tumors enter a prolonged state of dormancy of up to 300 days when transplanted into virgin female mice (Gattelli et al. 2004). Unlike the main mechanisms of dormancy described above, these tumor cells are dormant mainly due to a slow proliferation rate. These cells however can emerge from dormancy upon hormonal treatment. In an analogous situation, treatment of prostate cancer by androgen

11  Dormancy of Disseminated Tumor Cells

243

depletion also results in a significant decrease in tumor burden, but a few cancer cells could persist in a dormant quiescent state for extended lengths of time (Kusumi et al. 2008; Mostaghel et al. 2009). However, recurrences from these dormant breast and prostate cancers, are usually non-responsive to endocrine therapy and resistant to conventional chemotherapy. Several mechanisms have been proposed to account for these hormone-independent recurrences. These include: alterations in the expression or function of the hormone receptor (Santen et  al. 2005); genetic changes such as inactivating mutations in the hormone receptor; amplification of the receptor or epigenetic changes in gene expression could also lead to such hormone independence (Iwase et al. 1998; van Agthoven et al. 1994), or development of clones that can adapt and grow in a low hormone environment (Mostaghel et al. 2009). These studies suggest that in addition to the role of adhesion molecules on the tumor cells, the nature of the ligands and soluble factors within the stroma that interacts with the tumor cells can also influence the phenotypic fate of disseminated tumor cells.

TGFb Signaling TGFb is a member of a large family of cytokines that are involved in the regulation of different cellular process such as cell proliferation, morphogenesis, migration, extracellular matrix production, angiogenesis, immune suppression, cytokine secretion, and apoptosis (see Chap. 31 in this book). TGFb is known to have a dual role in cancer. During early stages of tumorigenesis, TGFb exerts antiproliferative effects and functions as a tumor suppressor (Barcellos-Hoff and Akhurst 2009; Derynck et al. 2001). For instance, TGFb has been shown to be a potent anticancer agent that prohibits the uncontrolled proliferation of epithelial, endothelial, and hematopoietic cells (Tian and Schiemann 2009). Moreover, TGFb signaling has been shown to be required to maintain normal stem cell quiescence in the BM (Tang et  al. 2007). However, at later stages during cancer progression, TGFb is considered to be pro-invasive and to induce epithelial to mesenchymal transitions (EMT) during metastatic cancer progressions (Seoane 2006; Thiery 2003) Therefore, TGFb can have a dual functional role as tumor suppressor and as metastatic promoter depending on the stage of cancer progression. But this also suggests that perhaps some tumors even having the ability to disseminate might still be negatively influenced by TGFb signaling in organs where it is abundant, such as the BM. The role of TGFb in inducing dormancy of DTCs is not well established, but since dissemination is an early event in tumorigenesis, it might be possible that DTCs still responsive to TGFb arriving at distant places like the BM or the liver remain arrested due to the high levels of TGFb on these tissues. It is well known, for example, that in breast cancer the detection of BM-DTCs is much higher (>50% of patients) (Klein 2008; Wikman et al. 2008) than the rate of metastasis in the BM.

244

P. Bragado et al.

Therefore, it has been proposed that the bone marrow microenvironment may be able to induce dormancy. BM stores a variety of growth factors including IGFs, TGFb, PDGF, and BMP (Mundy 1997; Mundy 2002; Roodman 2004). In addition, there is experimental evidence showing that TGFb is released from the bone matrix by bone osteoclasts (Guise and Chirgwin 2003). Moreover, majority of breast tumors, including their metastases, are positive for nuclear phosphorylated Smad2, which suggests they have an actively signaling TGFb pathway (Kang et al. 2005; Xie et al. 2002). In addition, Salm et al. (2005) have showed that TGFb is necessary to maintain dormancy of normal prostatic stem cells in the proximal region of ducts. Thus, TGFb is another potential signaling pathways that might participate in the induction of dormancy of DTCs in the BM. TGFb has been shown to exert its anti-proliferative activities by regulating epithelial cell and fibroblasts, which synthesize and secrete a variety of cytokines, growth factors, and extracellular matrix (ECM) proteins that mediate tissue homeostasis and inhibit cancer cell proliferation (Alexandrow and Moses 1995; Huang and Huang 2005). For example, studies show that TGFb stimulation of epithelial cells induces a G1 cell cycle arrest via the transcriptional upregulation of the cyclindependent kinase (CDK) inhibitors p21Cip1/WAF1 and p15Ink4b (Florenes et al. 1996). In addition, TGFb also causes transcriptional repression of transcription factors that regulate proliferation such as c-Myc and the inhibitors of differentiation Id1, Id2, and Id3 (Wu et al. 2009). Therefore, it is possible that TGFb induces quiescence of DTC through any of these mechanisms (Fig. 11.2). TGFb growth response can be inhibited by changes in active signaling networks or by the availability of transcriptional co-repressors or co-activators that regulate the Smad pathway. For instance, c-Ski is an important corepressor of TGF-b signaling that binds to and represses the activity of the Smad proteins. In breast and lung cancer cells, TGF-b induces Ski degradation through the ubiquitin-dependent proteasome and suppresses its ability to inhibit tumor metastasis (Deheuninck and Luo 2009; Reed et al. 2001). It has also been shown that in melanoma cells Id2 upregulation suppresses TGFb induction of p15, and this way circumvent TGFb-mediated inhibition of proliferation (Schlegel et al. 2009). The discovery that TGF-b-induced growth inhibition requires p53 activity suggests that mutational inactivation of p53 may contribute to the development of resistance to TGF-b antiproliferative effects by tumor cells (Atfi and Baron 2008; Cordenonsi et  al. 2003). Recently, Adorno et al. (2009) showed that TGFb-dependent metastasis are empowered by mutantp53 and opposed by p63. They showed that TGFb collaborates with Ras and mutant-p53 to induce the assembly of a mutant-p53/p63 protein complex and inhibit p63 functions. They also have identified two candidate metastasis suppressor genes regulated by TGFb and associated with metastasis risk in a large cohort of breast cancer patients. Thus, it is possible that some of the above described regulatory mechanisms could be operational in the BM microenvironment and convert dormant DTC into proliferative active DTCs that may be even able to migrate to other tissues or grow in the bone microenvironment like in breast and prostate cancer. One of the metastasis suppressor genes regulated by TGFb described by Adorno et  al. is BHLHB3, which is a transcriptional repressor involved in differentiation

11  Dormancy of Disseminated Tumor Cells

245

and a regulator of the circadian rhythm (Azmi et al. 2004; Fujimoto et al. 2007). In addition, in breast cancer, high BHLHB3 expression in primary tumors from patients correlates with good prognosis at 5-year survival (van de Vijver et al. 2002), suggesting that it influences behavior of residual disease. Work from our laboratory has recently showed that BHLHB3 is required for D-HEp3 cell quiescence in vivo and that its expression is repressed by ERK and induced by p38 (Adam et al. 2009). These results suggest that BHLHB3 is a novel negative regulator of tumor growth functionally linked to p38 signaling. Based on these results it is tempting to speculate that regulation of BHLHB3 through TGFb might limit expansion of DTCs into micrometastasis via cellular dormancy (i.e., quiescence) (Fig. 11.2). A study by Hoek et al. (2008) illustrates the dual role of TGFb. They found two different transcriptional signatures – one corresponding to a proliferative melanoma that was sensitive to TGFb inhibition and another corresponding to an invasive melanoma that was resistant to TGFb inhibition. They showed that upon transplantation in nude mice, the proliferative TGFb-sensitive cells produced growing tumors faster than the invasive variant. The authors concluded that there was a switch between both transcription signatures in vivo, probably regulated by environmental conditions (Hoek et al. 2008). Therefore, from all these studies we can conclude that there are two possible scenarios for TGFb regulation of DTCs fate. At early stages of progression, DTCs that lodge in tissues rich in TGFb like the BM, might be sensitive to its antiproliferative effect, and thus, TGFb may induce quiescence of DTC, that can be in that state for months, or even years. It is even possible that TGFB might be produced in an autocrine fashion to maintain dormancy. At a given point, these cells will become resistant to TGFb growth inhibitory effect, either because of changes in the microenvironment or DTCs newly acquired mutations, which will favor expansion.

Models to Study Dormancy Clinical dormancy has long been reported in breast, prostate, head and neck cancer, and melanoma (Fehm et al. 2008; Kell et al. 2000; Klein 2008; Morgan et al. 2007; Ossowski and Aguirre-Ghiso 2010). These dormant DTCs can be detected in the blood and bone marrow of these patients, months, years, and even decades after remission. Detection of these DTCs usually correlates with poor patient outcome (Allan et al. 2006; Braun et al. 2009; Ignatiadis et al. 2008; Janni et al. 2000). DTCs are usually found in the BM, therefore most of the studies regarding the biology of DTC are done in the BM, mainly because it is easier to obtain BM samples. However, there are other tissues where dormant DTC can also be found, for instance, studies in mice injected intravenously with melanoma and breast cancer cells, show a proportion of dormant cells in lungs and liver (Cameron et al. 2000; Logan et al. 2008). Most of the mouse models of cancer reproduce primary tumor development. Much less effort has been dedicated to model the progression of spontaneous MRD

246

P. Bragado et al.

and metastasis development despite the abundance of mouse transgenic models. Dormancy is a hidden state and thus is difficult to observe and study directly. In addition, characterization of disseminated tumor cells is very complicated because of their low abundance (one disseminated tumor cell per 106 bone marrow cells) (Braun et al. 2009). As we have discussed before, DTCs and primary tumor cells appear to be two different entities. Therefore, the models that we have classically used to study primary tumor growth cannot be used to extrapolate information about the DTCs. Because, metastasis is still the leading cause of mortality in patients and it arises from DTCs, we urgently need new models that better recapitulate MRD and DTC biology so that we can gather more information about the mechanisms that regulate their behavior. This will allow us to design new therapies specifically designed either to eradicate DTCs or to maintain them in a state of dormancy forever. The ideal model of solitary dormant tumor cells would consist of a population of cells which become dormant upon arrival to a secondary site, stay dormant for a period of time that is indicative of a prolonged inactive state (weeks to months), and then escape dormancy to grow. Importantly, it would be ideal that this occurs to all cells that disseminate. If only a subpopulation is dormant and other resume aggressive growth concomitantly in the same organ like in some of the above studies using i.v. injections, this argues that clonal heterogeneity is driving these differences. These may be informative, but if all cells that disseminate enter dormancy in a specific site it suggests more powerful microenvironmental mechanisms driving this phenotypic change independently of the genetics of those tumor cells. In order to study minimal residual disease, we need to have models that better recapitulate what happens in patients. The most commonly used metastatic models generate very rapid development of metastasis or rely on intravenous or intracardial injection and also, in the case of transgenic mice, are accompanied by exaggeratedly large primary tumor masses that shed enormous amounts of cells. Therefore, including surgery of the primary tumors that are smaller than 500 mm3, and then allowing the animals to recover and follow the residual cells might help better model MRD and dormancy. Bone marrow is the most important site to detect DTCs from epithelial tumors (Fehm et al.2008; Pantel et al. 2009; Riethdorf and Pantel 2008). DTCs are present in BM aspirate from 20 to 40% of breast and prostrate patients (Msaouel et  al. 2008; Zhang et al. 2009). Spontaneous bone metastasis in animals is uncommon. For instance, most spontaneous mammary carcinomas in mice do not metastasize to bone but readily do to lungs and lymph nodes. However, this does not mean that they do not disseminate to bone and stay dormant. In fact, Husemann et al. (2008) and Husemann and Klein (2009) used HER-2 and PyMT transgenic mice, transplanted with premalignant HER-2 transgenic glands, and showed that they displayed disseminated tumor cells and micrometastasis in bone marrow and lungs. MMTV-Neu cells disseminated to the bone marrow of mice fail to expand unless they are transplanted into recipient mice that were irradiated. This suggests that the bone microenvironment may delay cancer progression and thus prolong patient survival, but also that perturbations of this microenvironment might trigger growth.

11  Dormancy of Disseminated Tumor Cells

247

Apart from breast cancer, in prostate cancer, the main cause of death is bone metastasis and BM-DTC can be found in patients (Alix-Panabieres et  al. 2005; Morgan et  al. 2007). In fact, in a recent publication Morgan et  al. (2009) have showed that finding DTC in prostate cancer patients, who have undergone radical prostatectomy and do not show evidence of disease, predicts biochemical recurrence. BM-DTCs were detected in 72% of patients prior to radical prostatectomy. Significant effort has been taken on trying to characterize BM-DTCs. Orthotopic injection of human cancer cells into the prostate usually results in a low incidence of bone metastasis from late-stage cancers. For instance, Vesella and collaborators (2003) has developed a prostate cancer xenograft tumor cell dormancy model. They have used LuCaP 23.8 and LuCaP 35, hormone-sensitive CaP xenografts that express PSA and the wild-type androgen receptor and grew them orthotopically in SCID mice. Once the tumor volumes reached 250–500 mg, primary tumors were removed to allow micrometastases to grow. Using this procedure they were able to generate macroscopic metastases (>20 mg) in 71% (LuCaP 23.8) and 100% (LuCaP 35) of animals, respectively. Visible metastases in pelvic lymph nodes were observed in nearly half of the animals and microscopic lung metastases (1–40 cells) were observed in 80% of the animals. However, even using the prostatectomy procedure with the orthotopic implants, the authors were unable to generate bone metastasis. Due to the fact that they found that some bone marrow samples were PSA RT-PCR positive, which suggests the presence of CaP micrometastases in the bone, they propose that the lack of bone metastasis may be due to the time frame, and that if the observation time would have been longer maybe dormant BM-DTC might have developed into visible metastases. But this information also suggests that the timing to develop overt lesions is in part controlled by the target tissue microenvironment. Thus, the bone marrow microenvironment delays growth and this model can be used to study BM-DTC dormancy of CaP xenografts. In addition to this Shachaf et  al. has published that inactivation of the MYC oncogene is sufficient to induce sustained regression of invasive liver cancers. In liver cells, MYC inactivation uncovers stem cell properties and triggers differentiation, but some of the cells remain in a dormant state and can become cancerous again by MYC reactivation. They showed that in MYC-induced hepatocellular carcinoma, MYC inactivation initially induces differentiation and apoptosis of most of the tumor cells. However under this circumstances, reactivation of MYC causes these differentiated tumor cells to rapidly become tumorigenic again (Shachaf and Felsher 2005; Shachaf et  al. 2004 2008). These studies suggest that cancer cells retain the capacity to become biologically normal behaving cells, and that this change can be controlled by oncogene induction. Depending on the context oncogene reactivation can inhibit or activate the malignant phenotype. Based on our discussions above it is essential to include DTC as therapeutic targets. Further, molecular and functional information about these cells and the molecular mechanisms driving their behavior are necessary to design and develop more efficient therapies. Screening of BM for DTC could provide information about the therapeutic efficiency of a tested drug against DTCs. Identifying the specific

248

P. Bragado et al.

genetic and epigenetic changes, that DTC undergo during cancer progression, is essential to develop efficient therapies that target DTC, before they develop in overt metastasis. Acknowledgments  This work was supported by grants from the Samuel Waxman Cancer Research Foundation Tumor Dormancy Program (to JAG), NIH/NCI (CA109182 to JAG), NIEHS (ES017146 to JAG), New York State Stem Cell Science – NYSTEM (to JAG) and NIDCR (DE020121 to AR).

References Adam AP, George A, Schewe D et al (2009) Computational identification of a p38SAPK-regulated transcription factor network required for tumor cell quiescence. Cancer Res 69: 5664–5672 Adorno M, Cordenonsi M, Montagner M et al (2009) A Mutant-p53/Smad complex opposes p63 to empower TGFbeta-induced metastasis. Cell 137:87–98 Aguirre Ghiso JA (2002) Inhibition of FAK signaling activated by urokinase receptor induces dormancy in human carcinoma cells in vivo. Oncogene 21:2513–2524 Aguirre Ghiso JA, Kovalski K, Ossowski L (1999) Tumor dormancy induced by downregulation of urokinase receptor in human carcinoma involves integrin and MAPK signaling. J Cell Biol 147:89–104 Aguirre-Ghiso JA (2007) Models, mechanisms and clinical evidence for cancer dormancy. Nat Rev Cancer 7:834–846 Aguirre-Ghiso JA, Estrada Y, Liu D et al (2003) ERK(MAPK) activity as a determinant of tumor growth and dormancy; regulation by p38(SAPK). Cancer Res 63:1684–1695 Aguirre-Ghiso JA, Liu D, Mignatti A et al (2001) Urokinase receptor and fibronectin regulate the ERK(MAPK) to p38(MAPK) activity ratios that determine carcinoma cell proliferation or dormancy in vivo. Mol Biol Cell 12:863–879 Alexandrow MG, Moses HL (1995) Transforming growth factor beta and cell cycle regulation. Cancer Res 55:1452–1457 Alix-Panabieres C, Rebillard X, Brouillet JP et al (2005) Detection of circulating prostate-specific antigen-secreting cells in prostate cancer patients. Clin Chem 51:1538–1541 Allan AL, Vantyghem SA, Tuck AB et al (2006) Tumor dormancy and cancer stem cells: implications for the biology and treatment of breast cancer metastasis. Breast Dis 26:87–98 Ansieau S, Bastid J, Doreau A et al (2008a) Induction of EMT by twist proteins as a collateral effect of tumor-promoting inactivation of premature senescence. Cancer Cell 14:79–89 Ansieau S, Hinkal G, Thomas C et al (2008b) Early origin of cancer metastases: dissemination and evolution of premalignant cells. Cell Cycle 7:3659–3663 Arbiser JL, Moses MA, Fernandez CA et al (1997) Oncogenic H-ras stimulates tumor angiogenesis by two distinct pathways. Proc Natl Acad Sci U S A 94:861–866 Atfi A, Baron R (2008) p53 brings a new twist to the Smad signaling network. Sci Signal 1:pe33 Azmi S, Ozog A, Taneja R (2004) Sharp-1/DEC2 inhibits skeletal muscle differentiation through repression of myogenic transcription factors. J Biol Chem 279:52643–52652 Bandyopadhyay S, Zhan R, Chaudhuri A et  al (2006) Interaction of KAI1 on tumor cells with DARC on vascular endothelium leads to metastasis suppression. Nat Med 12:933–938 Barcellos-Hoff MH, Akhurst RJ (2009) Transforming growth factor-beta in breast cancer: too much, too late. Breast Cancer Res 11:202 Barcellos-Hoff MH, Ravani SA (2000) Irradiated mammary gland stroma promotes the expression of tumorigenic potential by unirradiated epithelial cells. Cancer Res 60:1254–1260

11  Dormancy of Disseminated Tumor Cells

249

Barkan D, Kleinman H, Simmons JL et al (2008) Inhibition of metastatic outgrowth from single dormant tumor cells by targeting the cytoskeleton. Cancer Res 68:6241–6250 Bidard FC, Vincent-Salomon A, Sigal-Zafrani B et al (2008) Time to metastatic relapse and breast cancer cells dissemination in bone marrow at metastatic relapse. Clin Exp Metastasis 25:871–875 Braun S, Auer D, Marth C (2009) The prognostic impact of bone marrow micrometastases in women with breast cancer. Cancer Invest 27:598–603 Braun S, Naume B (2005) Circulating and disseminated tumor cells. J Clin Oncol 23:1623–1626 Cameron MD, Schmidt EE, Kerkvliet N et al (2000) Temporal progression of metastasis in lung: cell survival, dormancy, and location dependence of metastatic inefficiency. Cancer Res 60:2541–2546 Chaurasia P, Mezei M, Zhou M-M et al (2009) Computer aided identification of small molecules disrupting uPAR/a5b1-integrin interaction: a new paradigm for metastasis prevention. PLoS ONE 4:e4617 Chen SC, Wang BW, Wang DL et al (2008) Hypoxia induces discoidin domain receptor-2 expression via the p38 pathway in vascular smooth muscle cells to increase their migration. Biochem Biophys Res Commun 374:662–667 Chomel JC, Brizard F, Veinstein A et al (2000) Persistence of BCR-ABL genomic rearrangement in chronic myeloid leukemia patients in complete and sustained cytogenetic remission after interferon-alpha therapy or allogeneic bone marrow transplantation. Blood 95:404–408 Cordenonsi M, Dupont S, Maretto S et al (2003) Links between tumor suppressors: p53 is required for TGF-beta gene responses by cooperating with Smads. Cell 113:301–314 Corey E, Quinn JE, Vessella RL (2003) A novel method of generating prostate cancer metastases from orthotopic implants. Prostate 56:110–114 Dameron KM, Volpert OV, Tainsky MA et al (1994a) Control of angiogenesis in fibroblasts by p53 regulation of thrombospondin-1. Science 265:1582–1584 Dameron KM, Volpert OV, Tainsky MA et  al (1994b) The p53 tumor suppressor gene inhibits angiogenesis by stimulating the production of thrombospondin. Cold Spring Harb Symp Quant Biol 59:483–489 Davis TA, Maloney DG, Czerwinski DK et al (1998) Anti-idiotype antibodies can induce longterm complete remissions in non-Hodgkin’s lymphoma without eradicating the malignant clone. Blood 92:1184–1190 Deheuninck J, Luo K (2009) Ski and SnoN, potent negative regulators of TGF-beta signaling. Cell Res 19:47–57 Derynck R, Akhurst RJ, Balmain A (2001) TGF-beta signaling in tumor suppression and cancer progression. Nat Genet 29:117–129 Farrar JD, Katz KH, Windsor J et al (1999) Cancer dormancy. VII. A regulatory role for CD8+ T cells and IFN-gamma in establishing and maintaining the tumor-dormant state. J Immunol 162:2842–2849 Fehm T, Muller V, Alix-Panabieres C et al (2008) Micrometastatic spread in breast cancer: detection, molecular characterization and clinical relevance. Breast Cancer Res 10(Suppl 1):S1 Fidler IJ (2003) The pathogenesis of cancer metastasis: the ‘seed and soil’ hypothesis revisited. Nat Rev Cancer 3:453–458 Florenes VA, Bhattacharya N, Bani MR et  al (1996) TGF-beta mediated G1 arrest in a human melanoma cell line lacking p15INK4B: evidence for cooperation between p21Cip1/WAF1 and p27Kip1. Oncogene 13:2447–2457 Folkman J (2002) Role of angiogenesis in tumor growth and metastasis. Semin Oncol 29:15–18 Folkman J, Hanahan D (1991) Switch to the angiogenic phenotype during tumorigenesis. Princess Takamatsu Symp 22:339–347 Fujimoto K, Hamaguchi H, Hashiba T et al (2007) Transcriptional repression by the basic helixloop-helix protein Dec2: multiple mechanisms through E-box elements. Int J Mol Med 19:925–932 Galon J, Fridman WH, Pages F (2007) The adaptive immunologic microenvironment in colorectal cancer: a novel perspective. Cancer Res 67:1883–1886

250

P. Bragado et al.

Gattelli A, Cirio MC, Quaglino A et  al (2004) Progression of pregnancy-dependent mouse mammary tumors after long dormancy periods. Involvement of Wnt pathway activation. Cancer Res 64:5193–5199 Gimbrone MA Jr, Leapman SB, Cotran RS et al (1972) Tumor dormancy in vivo by prevention of neovascularization. J Exp Med 136:261–276 Guise TA, Chirgwin JM (2003) Transforming growth factor-beta in osteolytic breast cancer bone metastases. Clin Orthop Relat Res (415 Suppl):S32–S38 Heiss MM, Allgayer, H, Gruetzner KU et  al (1995) Individual development and uPA-receptor expression of disseminated tumour cells in bone marrow: a reference to early systemic disease in solid cancer. Nat Med 1:1035–1039 Hemsen A, Riethdorf L, Brunner N et al (2003) Comparative evaluation of urokinase-type plasminogen activator receptor expression in primary breast carcinomas and on metastatic tumor cells. Int J Cancer 107:903–909 Henriet P, Zhong ZD, Brooks PC et al (2000) Contact with fibrillar collagen inhibits melanoma cell proliferation by up-regulating p27KIP1. Proc Natl Acad Sci U S A 97:10026–10031 Hoek KS, Eichhoff OM, Schlegel NC et al (2008) In vivo switching of human melanoma cells between proliferative and invasive states. Cancer Res 68:650–656 Holmgren L, O’Reilly MS, Folkman J (1995) Dormancy of micrometastases: balanced proliferation and apoptosis in the presence of angiogenesis suppression. Nat Med 1:149–153 Horak CE, Lee JH, Marshall JC et al (2008) The role of metastasis suppressor genes in metastatic dormancy. APMIS 116:586–601 Huang SS, Huang JS (2005) TGF-beta control of cell proliferation. J Cell Biochem 96:447–462 Husemann Y, Geigl JB, Schubert F et al (2008) Systemic spread is an early step in breast cancer. Cancer Cell 13:58–68 Husemann Y, Klein CA (2009) The analysis of metastasis in transgenic mouse models. Transgenic Res 18:1–5 Ignatiadis M, Georgoulias V, Mavroudis D (2008) Micrometastatic disease in breast cancer: clinical implications. Eur J Cancer 44:2726–2736 Ivaska J, Reunanen H, Westermarck J et al (1999) Integrin alpha2beta1 mediates isoform-specific activation of p38 and upregulation of collagen gene transcription by a mechanism involving the alpha2 cytoplasmic tail. J Cell Biol 147:401–416 Iwase H, Omoto Y, Iwata H et al (1998) Genetic and epigenetic alterations of the estrogen receptor gene and hormone independence in human breast cancer. Oncology 55(Suppl 1):11–16 Janni W, Rjosk D, Braun S (2000) Clinical relevance of occult metastatic cells in the bone marrow of patients with different stages of breast cancer. Clin Breast Cancer 1:217–225 Kaipainen A, Kieran MW, Huang S et  al (2007) PPARalpha deficiency in inflammatory cells suppresses tumor growth. PLoS One 2:e260 Kang Y, He W, Tulley S et al (2005) Breast cancer bone metastasis mediated by the Smad tumor suppressor pathway. Proc Natl Acad Sci U S A 102:13909–13914 Kang Y, Siegel PM, Shu W et al (2003) A multigenic program mediating breast cancer metastasis to bone. Cancer Cell 3:537–549 Kell MR, Winter DC, O’Sullivan GC et al (2000) Biological behaviour and clinical implications of micrometastases. Br J Surg 87:1629–1639 Kienast Y, von Baumgarten L, Fuhrmann M et  al (2010) Real-time imaging reveals the single steps of brain metastasis formation. Nat Med 16:116–122 Klein CA (2002) Direct molecular analysis of single disseminated cancer cells: a prerequisite for the development of adjuvant therapies? Acta Med Austriaca Suppl 59:10–13 Klein CA (2008) The direct molecular analysis of metastatic precursor cells in breast cancer: a chance for a better understanding of metastasis and for personalised medicine. Eur J Cancer 44:2721–2725 Klein CA, Blankenstein TJ, Schmidt-Kittler O et al (2002) Genetic heterogeneity of single disseminated tumour cells in minimal residual cancer. Lancet 360:683–689 Koebel CM, Vermi W, Swann JB et al (2007) Adaptive immunity maintains occult cancer in an equilibrium state. Nature 450:903–907

11  Dormancy of Disseminated Tumor Cells

251

Kook YH, Adamski J, Zelent A et al (1994) The effect of antisense inhibition of urokinase receptor in human squamous cell carcinoma on malignancy. EMBO J 13:3983–3991 Kusumi T, Koie T, Tanaka M et al (2008) Immunohistochemical detection of carcinoma in radical prostatectomy specimens following hormone therapy. Pathol Int 58:687–694 Lahlou H, Sanguin-Gendreau V, Zuo D et al (2007) Mammary epithelial-specific disruption of the focal adhesion kinase blocks mammary tumor progression. Proc Natl Acad Sci U S A 104: 20302–20307 Logan PT, Fernandes BF, Di Cesare S et al (2008) Single-cell tumor dormancy model of uveal melanoma. Clin Exp Metastasis 25:509–516 Luzzi KJ, MacDonald IC, Schmidt EE et al (1998) Multistep nature of metastatic inefficiency: dormancy of solitary cells after successful extravasation and limited survival of early micrometastases. Am J Pathol 153:865–873 MacKie RM, Reid R, Junor B (2003) Fatal melanoma transferred in a donated kidney 16 years after melanoma surgery. N Engl J Med 348:567–568 Mahnke YD, Schwendemann J, Beckhove P et al (2005) Maintenance of long-term tumour-specific T-cell memory by residual dormant tumour cells. Immunology 115:325–336 Marches R, Scheuermann RH, Uhr JW (1998) Cancer dormancy: role of cyclin-dependent kinase inhibitors in induction of cell cycle arrest mediated via membrane IgM Cancer Res 58:691–697 Morgan TM, Lange PH, Porter MP et  al (2009) Disseminated tumor cells in prostate cancer patients after radical prostatectomy and without evidence of disease predicts biochemical recurrence. Clin Cancer Res 15:677–683 Morgan TM, Lange PH, Vessella RL (2007) Detection and characterization of circulating and disseminated prostate cancer cells. Front Biosci 12:3000–3009 Mostaghel EA, Montgomery B, Nelson PS (2009) Castration-resistant prostate cancer: targeting androgen metabolic pathways in recurrent disease. Urol Oncol 27:251–257 Msaouel P, Pissimissis N, Halapas A et  al (2008) Mechanisms of bone metastasis in prostate cancer: clinical implications. Best Pract Res Clin Endocrinol Metab 22:341–355 Mundy GR (1997) Mechanisms of bone metastasis. Cancer 80:1546–1556 Mundy GR (2002) Metastasis to bone: causes, consequences and therapeutic opportunities. Nat Rev Cancer 2:584–593 Nash KT, Phadke PA, Navenot JM et al (2007) Requirement of KISS1 secretion for multiple organ metastasis suppression and maintenance of tumor dormancy. J Natl Cancer Inst 99:309–321 Naumov GN, Bender E, Zurakowski D et al (2006) A model of human tumor dormancy: an angiogenic switch from the nonangiogenic phenotype. J Natl Cancer Inst 98:316–325 Olumi AF, Grossfeld GD, Hayward SW et  al (1999) Carcinoma-associated fibroblasts direct tumor progression of initiated human prostatic epithelium. Cancer Res 59:5002–5011 Ossowski L, Aguirre-Ghiso JA (2010) Dormancy of metastatic melanoma. Pigment Cell Melanoma Res 23:41–56. Pan Y, Haines DS (2000) Identification of a tumor-derived p53 mutant with novel transactivating selectivity. Oncogene 19:3095–3100 Pantel K, Alix-Panabieres C, Riethdorf S (2009) Cancer micrometastases. Nat Rev Clin Oncol 6:339–351 Pantel K, Brakenhoff RH (2004) Dissecting the metastatic cascade. Nat Rev Cancer 4:448–456 Pierga JY, Bonneton C, Magdelenat H et al (2005) Real-time quantitative PCR determination of urokinase-type plasminogen activator receptor (uPAR) expression of isolated micrometastatic cells from bone marrow of breast cancer patients. Int J Cancer 114:291–298 Prince S, Carreira S, Vance KW et  al (2004) Tbx2 directly represses the expression of the p21(WAF1) cyclin-dependent kinase inhibitor. Cancer Res 64:1669–1674 Quesnel B (2006) Cancer vaccines and tumor dormancy: a long-term struggle between host antitumor immunity and persistent cancer cells? Expert Rev Vaccines 5:773–781 Quesnel B (2008) Tumor dormancy and immunoescape. APMIS 116:685–694 Rak J, Mitsuhashi Y, Bayko L et al (1995) Mutant ras oncogenes upregulate VEGF/VPF expression: implications for induction and inhibition of tumor angiogenesis. Cancer Res 55: 4575–4580

252

P. Bragado et al.

Reed JA, Bales E, Xu W et al (2001) Cytoplasmic localization of the oncogenic protein Ski in human cutaneous melanomas in vivo: functional implications for transforming growth factor beta signaling. Cancer Res 61:8074–8078 Riethdorf S, Pantel K (2008) Disseminated tumor cells in bone marrow and circulating tumor cells in blood of breast cancer patients: current state of detection and characterization. Pathobiology 75:140–148 Roodman GD (2004) Mechanisms of bone metastasis. N Engl J Med 350:1655–1664 Salm SN, Burger PE, Coetzee S et al (2005) TGF-{beta} maintains dormancy of prostatic stem cells in the proximal region of ducts. J Cell Biol 170:81–90 Sang L, Coller HA, Roberts JM. (2008) Control of the reversibility of cellular quiescence by the transcriptional repressor HES1. Science. 321(5892):1095–1100. Santen RJ, Song RX, Zhang Z et al (2005) Adaptive hypersensitivity to estrogen: mechanisms and clinical relevance to aromatase inhibitor therapy in breast cancer treatment. J Steroid Biochem Mol Biol 95:155–165 Saudemont A, Hamrouni A, Marchetti P et al (2007) Dormant tumor cells develop cross-resistance to apoptosis induced by CTLs or imatinib mesylate via methylation of suppressor of cytokine signaling 1. Cancer Res 67:4491–4498 Saudemont A, Quesnel B (2004) In a model of tumor dormancy, long-term persistent leukemic cells have increased B7-H1 and B7.1 expression and resist CTL-mediated lysis. Blood 104:2124–2133 Schlegel NC, Eichhoff OM, Hemmi S et  al (2009) Id2 suppression of p15 counters TGF-betamediated growth inhibition of melanoma cells. Pigment Cell Melanoma Res 22:445–453 Schmidt-Kittler O, Ragg T, Daskalakis A et  al (2003) From latent disseminated cells to overt metastasis: genetic analysis of systemic breast cancer progression. Proc Natl Acad Sci U S A 100:7737–7742 Schwarte-Waldhoff I, Volpert OV, Bouck NP et al (2000) Smad4/DPC4-mediated tumor suppression through suppression of angiogenesis. Proc Natl Acad Sci U S A 97:9624–9629 Seoane J (2006) Escaping from the TGFbeta anti-proliferative control. Carcinogenesis 27:2148–2156 Shachaf CM, Felsher DW (2005) Tumor dormancy and MYC inactivation: pushing cancer to the brink of normalcy. Cancer Res 65:4471–4474 Shachaf CM, Gentles AJ, Elchuri S et al (2008) Genomic and proteomic analysis reveals a threshold level of MYC required for tumor maintenance. Cancer Res 68:5132–5142 Shachaf CM, Kopelman AM, Arvanitis C et  al (2004) MYC inactivation uncovers pluripotent differentiation and tumour dormancy in hepatocellular cancer. Nature 431:1112–1117 Sharma SV, Gajowniczek P, Way IP et  al (2006) A common signaling cascade may underlie “addiction” to the Src, BCR-ABL, and EGF receptor oncogenes. Cancer Cell 10:425–435 Sheibani N, Frazier WA (1996) Repression of thrombospondin-1 expression, a natural inhibitor of angiogenesis, in polyoma middle T transformed NIH3T3 cells. Cancer Lett 107:45–52 Shibue T, Weinberg RA (2009) Integrin Œ£1-focal adhesion kinase signaling directs the proliferation of metastatic cancer cells disseminated in the lungs. Proc Natl Acad Sci U S A 106: 10290–10295 Sidky YA, Borden EC (1987) Inhibition of angiogenesis by interferons: effects on tumor- and lymphocyte-induced vascular responses. Cancer Res 47:5155–5161 Slade MJ, Singh A, Smith BM et al (2005) Persistence of bone marrow micrometastases in patients receiving adjuvant therapy for breast cancer: results at 4 years. Int J Cancer 114:94–100 Solakoglu O, Maierhofer C, Lahr G et al (2002) Heterogeneous proliferative potential of occult metastatic cells in bone marrow of patients with solid epithelial tumors. Proc Natl Acad Sci U S A 99:2246–2251 Steeg PS, Ouatas T, Halverson D et  al (2003) Metastasis suppressor genes: basic biology and potential clinical use. Clin Breast Cancer 4:51–62 Stoecklein NH, Hosch SB, Bezler M et al (2008) Direct genetic analysis of single disseminated cancer cells for prediction of outcome and therapy selection in esophageal cancer. Cancer Cell 13:441–453

11  Dormancy of Disseminated Tumor Cells

253

Su JD, Mayo LD, Donner DB et al (2003) PTEN and phosphatidylinositol 3¢ -kinase inhibitors up-regulate p53 and block tumor-induced angiogenesis: evidence for an effect on the tumor and endothelial compartment. Cancer Res 63:3585–3592 Tang B, Yoo N, Vu M et al (2007) Transforming growth factor-beta can suppress tumorigenesis through effects on the putative cancer stem or early progenitor cell and committed progeny in a breast cancer xenograft model. Cancer Res 67:8643–8652 Thiery JP (2003) Epithelial-mesenchymal transitions in development and pathologies. Curr Opin Cell Biol 15:740–746 Thomas C, Wiesner C, Melchior SW et  al (2009) Urokinase-plasminogen-activator receptor expression in disseminated tumour cells in the bone marrow and peripheral blood of patients with clinically localized prostate cancer. BJU Int 104:29–34 Tian M, Schiemann WP (2009) The TGF-beta paradox in human cancer: an update. Future Oncol 5:259–271 Udagawa T, Fernandez A, Achilles E-G et al (2002) Persistence of microscopic human cancers in mice: alterations in the angiogenic balance accompanies loss of tumor dormancy. FASEB J 16:1361–1370 Uhr JW, Tucker T, May RD et  al (1991) Cancer dormancy: studies of the murine BCL1 lymphoma. Cancer Res 51:5045s–5053s van Agthoven T, van Agthoven TL, Dekker A et al (1994) Induction of estrogen independence of ZR-75–1 human breast cancer cells by epigenetic alterations. Mol Endocrinol 8: 1474–1483 van de Vijver MJ, He YD, van’t Veer LJ et al (2002) A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med 347:1999–2009 Voest EE, Kenyon BM, O’Reilly MS et al (1995) Inhibition of angiogenesis in vivo by interleukin 12. J Natl Cancer Inst 87:581–586 Volpert OV, Dameron KM, Bouck N (1997) Sequential development of an angiogenic phenotype by human fibroblasts progressing to tumorigenicity. Oncogene 14:1495–1502 Wall SJ, Werner E, Werb Z et al (2005) Discoidin domain receptor 2 mediates tumor cell cycle arrest induced by fibrillar collagen. J Biol Chem 280:40187–40194 Wall SJ, Zhong ZD, DeClerck YA (2007) The cyclin-dependent kinase inhibitors p15INK4B and p21CIP1 are critical regulators of fibrillar collagen-induced tumor cell cycle arrest. J Biol Chem 282:24471–24476 Watnick RS, Cheng YN, Rangarajan A et al (2003) Ras modulates Myc activity to repress thrombospondin-1 expression and increase tumor angiogenesis. Cancer Cell 3:219–231 Weaver VM, Petersen OW, Wang F et al (1997) Reversion of the malignant phenotype of human breast cells in three-dimensional culture and in vivo by integrin blocking antibodies. J Cell Biol 137:231–245 Weigelt B, Peterse JL, van ‘t Veer LJ (2005) Breast cancer metastasis: markers and models. Nat Rev Cancer 5:591–602 Wen S, Stolarov J, Myers MP et al (2001) PTEN controls tumor-induced angiogenesis. Proc Natl Acad Sci U S A 98:4622–4627 White DE, Kurpios NA, Zuo D et al (2004) Targeted disruption of beta1-integrin in a transgenic mouse model of human breast cancer reveals an essential role in mammary tumor induction. Cancer Cell 6:159–170 White DE, Rayment JH, Muller WJ (2006) Addressing the role of cell adhesion in tumor cell dormancy. Cell Cycle 5:1756–1759 Wijsman JH, Cornelisse CJ, Keijzer R et  al (1991) A prolactin-dependent, metastasising rat mammary carcinoma as a model for endocrine-related tumour dormancy. Br J Cancer 64:463–468 Wikman H, Vessella R, Pantel K (2008) Cancer micrometastasis and tumour dormancy. APMIS 116:754–770 Wu S, Hultquist A, Hydbring P et al (2009) TGF-beta enforces senescence in Myc-transformed hematopoietic tumor cells through induction of Mad1 and repression of Myc activity. Exp Cell Res 315:3099–3111

254

P. Bragado et al.

Xie W, Mertens JC, Reiss DJ et al (2002) Alterations of Smad signaling in human breast carcinoma are associated with poor outcome: a tissue microarray study. Cancer Res 62:497–505 Yamshchikov GV, Mullins DW, Chang CC et al (2005) Sequential immune escape and shifting of T cell responses in a long-term survivor of melanoma. J Immunol 174:6863–6871 Yefenof E, Picker LJ, Scheuermann RH et al (1993a) Cancer dormancy: isolation and characterization of dormant lymphoma cells. Proc Natl Acad Sci U S A 90:1829–1833 Yefenof E, Picker LJ, Scheuermann RH et al (1993b) Induction of B cell tumor dormancy by antiidiotypic antibodies. Curr Opin Immunol 5:740–744 Zhang B, Zhang Y, Bowerman NA et al (2008) Equilibrium between host and cancer caused by effector T cells killing tumor stroma. Cancer Res 68:1563–1571 Zhang Y, Ma B, Fan Q (2009) Mechanisms of breast cancer bone metastasis. Cancer Lett 292:1–7

Part III

Vasculature And Stroma

Chapter 12

Impact of Endothelial Progenitor Cells on Tumor Angiogenesis and Outcome of Antiangiogenic Therapy: New Perspectives on an Ongoing Controversy Robert S. Kerbel, Francesco Bertolini, and Yuval Shaked Abstract  Tumor angiogenesis is driven not only by proliferation of differentiated endothelial cells in sprouting capillaries from pre-existing mature vessels, but also by mobilization of bone marrow derived circulating endothelial progenitor cells (EPCs). The latter are thought to home to the tumor site and incorporate into the lumen of newly growing blood vessels. Over the past several years, a growing number of reports have challenged the hypothesis concerning the involvement of EPCs in tumor angiogenesis, and instead suggest that such cells only have a minor role, if any at all, in the formation of tumor-associated blood vessels. Consequently, these studies implicate EPCs as a minor or negligible target for cancer therapy. In this review, we discuss the arguments for and against a significant role of EPCs in tumor angiogenesis and growth, and as possible surrogate markers of angiogenesis as well as valuable therapeutic targets.

Introduction Over the last decade, various types of bone marrow derived cell (BMDCs) have emerged as key players in tumor angiogenesis, growth, and metastasis (Shaked and Voest 2009; Kaplan et al. 2007); most such cell types originate from the hematopoietic lineage (Coffelt et  al. 2009; De Palma and Naldini 2006; Grunewald et  al. 2006; Shojaei et al. 2007). They can contribute to angiogenesis by various mechanisms including paracrine secretion of different growth factors mediated by perivascular residing BMDCs, and by contributing to the stability of the vasculature. Research into the development of antiangiogenic drugs to treat cancer has revealed that some BMDCs may compensate for vessel-targeting antiangiogenic effects, thus allowing for continued tumor growth. These bone marrow derived Y. Shaked (*) Department of Molecular Pharmacology, Rappaport Faculty of Medicine, Technion – Israel Institute of Technology, Haifa, Israel e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_12, © Springer Science+Business Media, LLC 2010

257

258

R.S. Kerbel et al.

tumor-associated stromal cells include various myeloid, macrophage, and monocytic cell populations. They all express the pan-hematopoietic surface marker, CD45, along with various endothelial cell/angiogenesis-related surface markers. However, a subset of BMDCs which are CD45-negative has been found to promote tumor angiogenesis by acting as an alternative source of endothelial cells; these cells are termed endothelial precursor or progenitor cells (EPCs). In contrast to all other proangiogenic BMDC types, EPCs are thought to merge with the wall of a growing blood vessel, where they differentiate into mature endothelial cells, and contribute to vessel growth. The various types of BMDC promoting angiogenesis are summarized in Fig. 12.1.

Fig. 12.1  Cellular players in tumor angiogenesis. Representation of various BMDC populations which have been shown to stimulate or amplify tumor angiogenesis. The various hematopoietic (CD45+) cell types appear to have a perivascular location with respect to the tumor neovasculature, whereas the CD45-negative endothelial progenitor cells can actually incorporate into the lumen of a growing vessel and differentiate into a mature endothelial cell. TEMs tie-2 expressing monocytes, HSC hematopoietic stem cell, ECs endothelial cell, TASC tumor associated stromal cell, HPC hematopoietic progenitor cell, VLC vascular leukocytes. Mature endothelial cells detach from the basal membrane following damage to tumor blood vessel, and in the blood flow become apoptotic circulating endothelial cells (CECs). Circulating endothelial progenitors (CEPs) that might derive from hemangioblast are mobilized from the bone marrow to incorporate in the vasculature as part of physiological tissue repair. In micro-metastatic lesions CEPs may play a significant role in the angiogenic switch, which promote metastatic growth. Reproduced from Martin-Padura and Bertolini (2008), with permission from the publisher

12  Impact of Endothelial Progenitor Cells on Tumor Angiogenesis and Outcome

259

There are a number of controversial issues regarding the origin, nature, and relevance of EPCs to angiogenesis, including tumor angiogenesis which are summarized below, along with some recent findings that may help resolve some of these controversies.

The Identification of EPCs In 1997, Isner’s group reported that post-natal vasculogenesis and angiogenesis can be driven by BMDCs circulating in peripheral blood which are home to sites of ongoing angiogenesis. The systemic process of de novo formation of capillaries and vessels from angioblasts of endothelial progenitor, previously shown to be involved in the formation of an embryonic vessel network, has been investigated in adults. Activated endothelial cells have been shown to express CD34 along with VEGFR1. In addition, CD34 is expressed by all hematopoietic stem cells but not mature hematopoietic cells. Using this distinguishing surface marker, Asahara et al. (1997) isolated CD34-negative and -positive cells and assessed their possible characteristics as endothelial cells, both morphologically and functionally. Injection of CD34+ cells labeled with the fluorescent dye Dil into ischemic hind-limb bearing mice resulted in massive homing of CD34+ but not CD34- cells to the mouse ischemic limb, and incorporation into the blood vessel wall. This study was the first to show that circulating CD34+ cells, obtained from human peripheral blood, can contribute to angiogenesis in adult tissues implicating the possibility that CD34+ cells are, in fact, EPCs. This discovery opened up new directions in angiogenesis research including tumor angiogenesis.

The Controversy Surrounding Functions of EPCs Following the first study by Asahara et al. (1997), implicating a role for EPCs in adult angiogenesis, many investigators reported results confirming the involvement of EPCs in tumor growth; various studies revealed highly contrasting results. Perhaps the most compelling evidence to support a role for EPCs in tumor angiogenesis emerged from studies using knockout mice bearing mutations in the Id family of transcriptional repressors. Id proteins are helix–loop–helix transcription factors that regulate various physiological processes including angiogenesis. These mice have been used to investigate the role of EPCs in tumor angiogenesis (Lyden et al. 1999). In vivo studies revealed that Id1-null mice exhibit decreased neovascularization as assessed by the Matrigel perfusion assay, compared to their wildtype counterparts, and this is regulated by the endogenous inhibitor of angiogenesis, thrombospondin 1 (TSP-1) (Volpert et al. 2002). It was shown that loss of Id1 in tumor endothelial cells resulted in downregulation of several proangiogenic genes, including those encoding a6 and b4 integrins, matrix metalloprotease-2, and fibroblast

260

R.S. Kerbel et al.

growth factor receptor-1 (Ruzinova et  al. 2003). Benezra and colleagues showed tumors growing in mice with a defect in the Id1 and Id3 genes (Id1+/–Id3–/–) are defective in angiogenesis, and hence growth. However, if such mice are irradiated and then given a transplant of wild-type bone marrow derived cells, tumor angiogenesis and progressive growth are restored (Lyden et al. 2001). Thus this in vivo data seemingly confirm the existence of EPCs, and suggest that these cells can contribute to overall tumor growth. In contrast to the in  vivo and ex vivo evidence regarding a significant role of EPCs in tumor growth, several recent studies have challenged the definition, function, and contribution of EPCs. Perhaps the major controversy is whether putative EPCs can truly incorporate into tumor blood vessels. Some investigators have raised doubts about this possibility, in part, due to the presence of other BMDCs that are found in close proximity to blood vessels, but which do not directly incorporate to the blood vessel wall. Such cells do not express VEGFR2 or CD31 surface markers like EPCs or mature endothelial cells, but express a variety of surface markers associated with angiogenesis such as Tie-2, CXCR4, and VEGFR1 (Grunewald et al. 2006; De Palma et al. 2005; Jin et al. 2006). These cells are found to colonize tumors, and contribute to their growth. Another factor which has led to this increasing controversy is the fact that some tumor models exhibit relatively high number of EPCs, whereas in others the number detected is minimal or even nonexistent: a range of 0–50% of tumor endothelium consisting of EPCs has been reported. Machein et  al. (2003) reported that tumor type is one such factor as this may significantly alter vasculogenesis and relative colonization of EPCs. They reported that the majority of bone marrow cells expressing hematopoietic and/or microglia markers, did not contribute to glioma vasculature. They also showed that overexpression of VEGF in glioma cells produced highly vascularized tumors, but the number of EPCs incorporated into the tumor vasculature was not increased. In a more recent study, Machein and colleagues reported that EPCs do not contribute at all to the growth of Lewis Lung carcinomas. The study evaluated the colonization of BMDCs tagged with GFP using several imaging techniques (Wickersheim et  al. 2009). Ziegelhoeffer et  al. (2004) reported that in mice bearing methylcholanthrene-induced fibrosarcoma (BFS-1), there were little, if any, BMDCs found to be incorporated into the tumor vasculature. In addition to tumor type, tumor grade has also been shown to alter the percentage of EPCs. Li et al. (2004) found that Id1 and Id3 transcription factors are highly expressed in endothelial cells of poorly differentiated prostate adenocarcinomas but not in the vasculature of well-differentiated tumors. Overall, these studies suggest that neovascularization in certain tumor types and of different tumor grade may be mediated by division and migration of differentiated endothelial cells from pre-existing vessels and that EPCs may not play a significant role in these circumstances. A clinical study by Peters et al. (2005) evaluated the number of BMDCs incorporated into the tumor blood vessel wall. They studied tumors obtained from patients who had previously undergone bone marrow transplantation obtained from individuals of the opposite sex. By using fluorescent In Situ hybridization (FISH)

12  Impact of Endothelial Progenitor Cells on Tumor Angiogenesis and Outcome

261

staining techniques and confocal laser scan microscopy the authors counted the number of endothelial cells in tumors which expressed mismatched sex chromosomes, indicating that they were derived from the transplanted donor bone marrow cells. They found that approximately 5% of tumor blood vessels consisting of bone marrow origin. These results indicate that there is a low number of EPCs incorporating into tumor blood vessels, and thus the possibility that they may function as endothelial cells contributing to the formation of a tumor vasculature. In recent studies, we have offered another possibility to explain the differences in the number of EPCs contributing to tumor growth reported by different investigators. We have shown that the genetic background of the host may play a significant role in the extent of adult vasculogenesis (Shaked et al. 2005a). Using a number of mouse strains, we evaluated the levels of EPCs circulating in the blood, and found that there are some strains, e.g., BALB/c, which exhibit high levels of EPCs in peripheral blood whereas other strains, e.g., C57Bl showed a low number of EPCs in up to a tenfold difference. We also showed that these levels strikingly correlated with the angiogenic responsiveness of the mouse strains to angiogenic stimuli measured by the corneal neovascular micropocket assay (Rohan et al. 2000) as well as in cases where mice were genetically manipulated to induce or reduce angiogenesis factors (Shaked et al. 2005a). These results indicated that in addition to tumor grade and type, the genetic background of the host may play a significant role in tumor neoangiogenesis. Only in the past few years several investigators have refuted the existence of EPCs in tumors. Purhonen et al. (2008) studied the mobilization and tumor homing of EPCs. They induced angiogenesis by using melanomas, APCmin adenomas, adenoviral VEGF delivery, and Matrigel plugs in four different genetically tagged universal or endothelial cell-specific chimeric mouse models. They failed to detect VEGFR2 cells mobilized from the bone marrow, nor incorporation directly into blood vessels. They thus concluded that EPC contribution is an extremely rare event, and perhaps might not even exist. Similarly, as noted above, Machein et al. (2003) who studied BMDCs expressing GFP, found that local VEGF induces a massive tumor infiltration of BMDCs, but no evidence of vessel wall integration of these cells. Furthermore, Kim et al. (2009) investigated acute and chronic phases of angiogenesis in wound healing assays, gel foam angiogenesis, in a parabiosis mouse model, and found that BMDCs expressing CD31 homing to sites of angiogenesis are macrophages (or monocytes) as they also express F4/80, a specific pan-macrophage marker. This approach was undertaken to avoid radiation of mice followed by a bone marrow transplant as such radiation might have affected the outcome of some of the prior studies using GFP+ bone marrow cells. These observations constitute yet another challenge to the EPC concept and the role of such cells in tumor angiogenesis, by suggesting that CD31 is expressed on angiogenic monocytes which do not differentiate into endothelial cells (Horrevoets 2009). In summary, these particular studies suggest that during tumor development and progression, new capillary formation is dependent only on “sprouting angiogenesis”, i.e., by migration and division of differential endothelial cells from pre-existing mature vessels. They clearly contradict the hypothesis of an involvement of EPCs in angiogenesis and tumor growth, and suggest that such cells might even be an

262

R.S. Kerbel et al.

experimental artifact. It should be noted, however, that several criticisms have been made regarding the study and conclusions reported by Purhonen et al. (2008). For example, APCmin mice develop only obstructive adenomas, rather than adenocarcinomas; therefore, it is an inappropriate model to study EPC incorporation in tumors (Kerbel et  al. 2008). Furthermore, the study of 6-month-old VEGF-Aloaded Matrigel plugs in mice is not possible because Matrigel plugs are degraded within 2 months, particularly when VEGF-A by itself is unlikely to induce neoangiogenesis in Matrigel plugs. A recent study by Madlambayan et al. (2009) attempted to resolve this ongoing controversy by comparing a variety of models of tumor angiogenesis in a single mouse background. These included the Lewis Lung carcinoma, B16 melanoma as well as a model of retinopathy. The authors evaluated a spectrum of BMDC types including Tie2 expressing monocytes (TEMs) (De Palma et al. 2005), tumor associated macrophages (Coffelt et al. 2009), EPCs, and hemangiocytes (Jin et al. 2006). They found that the contribution of systemic BMDCs to tumor angiogenesis is dependent primarily on the expression of SDF-1 in the model/site tested. Manipulation of SDF-1 expression in such sites revealed large differences in the number of BMDCs recruited. Since the majority of EPCs express the CXCR4 chemokine receptor (Pitchford et al. 2009), the results suggest that EPCs, among other bone marrow derived CXCR4+ cells, home to active sites of angiogenesis, and promote tumor growth. This study offers a resolution for the finding of different values of BMDCs, including EPCs contributing to tumor growth in different models, as reported by some investigators.

The Controversy About the Definition of EPCs In addition to the questions raised about the possible numbers of EPCs detected in tumor vessels and their functional importance, yet another controversy which has emerged concerns the precise definition and origin of such cells. It is likely that the difficulties associated with some of the techniques used to isolate, sort, enumerate, and quantitate EPCs have contributed to this situation. In peripheral blood, EPCs are measured by using a combination of antibodies recognizing various endothelial and hematopoietic surface markers, as detected by multiparametric flow cytometry techniques (Bertolini et al. 2006). The lack of a single definitive antigen to define and detect EPC populations has contributed to the debate regarding the evaluation of EPCs. Initial studies using both in vitro quantitative and flow cytometry characterization of EPCs revealed that VEGFR2 and CD34 are the “ultimate” surface markers of EPCs; however, it has been shown that AC133 (CD133), an orphan receptor that is expressed on some hemoatopoietic progenitor cells, and lost once they differentiate, is also considered as a marker of EPCs (Peichev et al. 2000; Rafii et al. 2002). Furthermore, CD146, for example, has been evaluated as a specific endothelial cell marker (Solovey et al. 2001), and is still used to detect EPCs circulating in peripheral blood. However, other studies have recently reported that mesenchymal cells

12  Impact of Endothelial Progenitor Cells on Tumor Angiogenesis and Outcome

263

(Elshal et al. 2005), and activated T lymphocytes (Duda et al. 2006) can also express CD146. Moreover, VE-cadherin (CD144), an endothelial-specific marker which was also thought to be expressed specifically by endothelial cells, has been reported to be expressed by some hematopoietic stem cells (Kim et al. 2005). Overall, currently, such results indicate that the most useful markers for identifying EPCs circulating in the blood are CD34, VEGFR2, and CD133, which appear to constitute a phenotypically distinct population of cells capable of differentiating into endothelial cells that can play a role in post-natal angiogenesis/vasculogenesis. To add complexity to the phenotypic characterization of EPCs (and thus to the confusion in the field) recent studies by Yoder, Ingram, and colleagues indicated that established commercially available kits used to repopulate and isolate EPCs from human peripheral blood by identifying methylcellulose endothelial cell colony forming units (CFU-ECs) do not represent truly putative EPCs (Case et al. 2007; Yoder et al. 2007). In a number of studies they identified another population of endothelial colony-forming cells (ECFCs), and tested the origin, proliferative potential, and differentiation capacity of ECFCs as well as CFU-ECs (obtained from human peripheral blood). They compared the function of CFU-ECs and ECFCs and determined that CFU-ECs are derived from the hematopoietic system, and possess myeloid progenitor cell activity, whereas ECFCs display robust proliferative potential, and form perfused vessels in  vivo (Yoder et  al. 2007). Overall, their results suggest that CFU-ECs, obtained from commercially available kits to isolate EPCs, are in fact not authentic EPCs, whereas ECFCs may actually represent a rare EPC population. These studies indicate that in order to improve quantitation and isolation of putative EPC populations, there is a clear need to discover a specific EPC marker which would constitute verification of EPCs in endothelial engraftment experiments in vivo. The development of monoclonal antibodies to detect endothelial cell-associated antigens has led to the extensive use of flow cytometry based assays to measure the number of EPCs circulating in peripheral blood. The use of multiparametric flow cytometry analysis on whole blood labeled simultaneously using a combination of conjugated antibodies has provided a useful tool to adequately measure rare populations in the blood. The combination of antibodies to detect EPCs in the blood usually consists of first excluding cells expressing the hematopoietic marker CD45, and then detecting cells expressing progenitor markers such as CD34, CD133, in addition to VEGFR2 as an endothelial cell marker. However, the lack of a standardized method to detect EPCs and mature endothelial cells circulating in the blood (CECs) has resulted in very different values of EPCs detected using flow cytometry techniques by different investigators, and this has caused increasing confusion about the nature of these cells, both clinically and preclinically. Since flow cytometry approaches require accurate sequential gating and are prone to operator and instrument-associated variability, a greater effort has been made recently to try and standardize the methodology for detecting EPCs and CECs in peripheral blood, especially for clinical studies (Bertolini et al. 2006; Mancuso et al. 2009). Thus, the definition and detection of EPCs requires further refinement and improvement in order to become widely accepted when used in different centers and by different operators/groups.

264

R.S. Kerbel et al.

EPCs as a Surrogate Biomarker for Antiangiogenic Therapy The identification of EPCs in peripheral blood has led to emerging data regarding the levels of EPCs in different pathologies including cardiovascular disease (Werner et al. 2005), diabetes (Cubbon et al. 2009), arthritis (Distler et al. 2009), eclampsia (Gammill et al. 2007), and cancer. With respect to cancer patients, some early studies evaluated the number of circulating endothelial cells (CECs) and their precursor subset (CEPs). It has been shown that levels of CECs, including viable CECs in cancer patients is higher than in normal healthy controls (Mancuso et  al. 2001). This is thought to occur by virtue of the fact that the production of proangiogenic factors by the tumor may promote CEC and CEP mobilization. In some of our early studies, we hypothesized that certain antiangiogenic drugs may target CEPs or at least may decrease their mobilization into the bloodstream, despite the production of angiogenic stimuli by the tumor. To test this hypothesis, we took advantage of a previous study by Rohan et al. (2000) who reported that angiogenic responsiveness measured by the corneal neovascular micropocket assay using bFGF, is markedly heterogeneous in different strains of mice. We found a striking correlation between the angiogenic responsiveness and the levels of CECs and CEPs in the same mouse strain (Shaked et al. 2005a). We next asked whether CEPs are targeted by antiangiogenic drugs. We showed that mice treated with DC101, a monoclonal antibody directed to mouse VEGFR2 (Prewett et al. 1999) or ABT-510, a peptide mimetic of TSP-1 (an endogenous angiogenic inhibitor (Lawler 2002)) (Yap et  al. 2005), caused a significant dose-dependent drop in levels of CEPs (defined by the phenotype CD45-,VEGFR2+,CD117+, and CD13+). We also found that the maximum decline (the “nadir”) in CEP levels correlated with the optimal biological dose (OBD) of the drug based on prior dose-response studies of anti-tumor activity (Shaked et al. 2005a). Similar results were obtained when we used another antiangiogenic treatment strategy, namely, metronomic low-dose chemotherapy (Klement et al. 2000; Hanahan et al. 2000; Shaked et al. 2005c), i.e., the administration of chemotherapy drugs using lower doses than the maximum tolerated dose (MTD), given at frequent, regular intervals with no long drug-free break periods (Klement et  al. 2000). These results not only provided evidence that some antiangiogenic drugs and treatment strategies target putative CEPs, but also suggested that CEPs can be used as a surrogate pharmacodynamic biomarker to determine the OBD of antiangiogenic therapies (Schneider et al. 2005), an approach which has been further evaluated in clinical studies (Mancuso et  al. 2006; Buckstein et  al. 2006; Bender et al. 2008). Mancuso et  al. (2006) have raised the possibility that apoptotic CECs may be used as a biomarker for antiangiogenic drug or treatment outcome. In their study of 121 breast cancer patients treated with metronomic cyclophosphamide and methrotraxate levels of CECs were evaluated. In contrast to the perception, at that time, that antiangiogenic drug therapy would reduce the levels of CECs and CEPs (Shaked et al. 2005a), the authors found that increases in CECs were associated with tumor response. When the authors distinguished between apoptotic and viable CECs, they

12  Impact of Endothelial Progenitor Cells on Tumor Angiogenesis and Outcome

265

found that in responding patients, the majority of CECs are in fact apoptotic (Mancuso et al. 2006). Supported by preclinical models of metronomic chemotherapy, the authors showed that the elevation in apoptotic CEC levels were likely tumor derived (i.e., derived from the tumor vasculature) and not host derived, since studies in non-tumor bearing mice indicated no detectable rise in apoptotic CECs levels. Comparable results were generated in additional clinical studies. For example, in a phase I clinical study of pediatric cancer patients undergoing bevacizumab treatment (Bender et al. 2008), it was reported that the ratio between the number of apoptotic CECs and total CECs was associated with the number of bevacizumab treatment cycles, indicating patients responding to therapy (Bender et al. 2008). In another clinical study, patients with advanced breast cancer received metronomic oral capecitabine and cyclophosphamide along with bevacizumab and levels of CECs were measured during the treatment; higher baseline CECs were found to be correlated with overall response, clinical benefit, and improved progression-free survival (Dellapasqua et al. 2008). Along with this analysis, the pharmacokinetics of CECs was also evaluated, and a pattern of decreased CECs and increased levels of various angiogenic growth factors have been reported in relapsed patients (Calleri et al. 2009). Overall, the aforementioned preclinical and clinical studies suggest that CECs and perhaps also CEPs may be used as surrogate biomarkers to determining the OBD of an antiangiogenic drug as well as predictive biomarkers of clinical outcome.

Therapy-Induced EPC Mobilization and Tumor Vessel Incorporation The lack of a specific antigen to define and detect EPCs, and the use of techniques to enumerate and quantitate EPCs in peripheral blood using at least four different surface markers have become extremely challenging technical issues for investigators in the field. However, in addition to studying CECs and CEPs as possible biomarkers for treatment outcome, our efforts have also been directed to studying the functional properties and effects of EPCs in different therapeutic circumstances. The results we have obtained provide an entirely new perspective on their impact and importance to angiogenesis and cancer therapy not only using antiangiogenic drugs, but also other therapies, including chemotherapy, both conventional and metronomic. We first investigated the possible impact of CEPs on tumor angiogenesis immediately after therapy with various cytotoxic agents. The rationale was based on our initial observations that MTD chemotherapy using cyclophosphamide rapidly increases the mobilization and viability of CEPs in NOD/SCID mice bearing human lymphomas whereas, in contrast, a low-dose metronomic regimen using the same chemotherapy drug, had the very opposite effect (Bertolini et al. 2003). These results suggested that MTD chemotherapy transiently induces angiogenesis (vasculogenesis) during the follow-up drug-free break periods which are usually instituted when administrating MTD cytotoxic therapy regimens, in order to allow recovery of

266

R.S. Kerbel et al.

patients (or animals) from the toxic side effects of the treatment, e.g., myelosuppression. In contrast, the chemotherapy may induce an antiangiogenic effect during the therapy, by targeting dividing endothelial cells (Kerbel 1991). Thus this sequence of events may be likened to an “action” (antiangiogenesis) followed by a “reaction” (pro-angiogenesis). As such, the “drug holidays” might promote, at least in part, tumor cell repopulation between successive courses of MTD chemotherapy. The surviving repopulating tumor cells are obviously one of the most important causes for treatment failure or limitations in the effects of such treatments (Kim and Tannock 2005). In this regard, we have recently found that shortly after treatment with any one of a number of different cytotoxic or cytotoxic-like drugs such as microtubule inhibiting vascular disrupting agents (VDAs), and certain chemotherapy drugs, e.g., paclitaxel, the levels of CEPs rise in peripheral blood within a matter of hours, after which they subsequently home to the tumor site and incorporate into or around the lumen of tumor blood vessels (Shaked et al. 2006b; Shaked et al. 2008). In the case of VDAs, these drugs cause rapid tumor vessel ischemia and occlusion followed by massive tumor hypoxia and necrosis. However, a viable and perfused rim tumor tissue usually remains from which tumor re-growth rapidly resumes (Tozer et al. 2005; Siemann et al. 2004), similar in some respects to rapid tumor cell repopulation observed after MTD chemotherapy-induced tumor shrinkage (Kim and Tannock 2005). Since the mechanisms which underline tumor cell repopulation after cytotoxic (or cytotoxic-like) drug therapy are poorly understood, we sought to investigate whether CEPs contribute to tumor re-growth from the viable tumor rim. We found that mice treated with VDAs, e.g., combretastatin A 4 phosphate (CA4-P) or OXi-4503, a second generation (more potent derivative) of CA4-P, exhibited substantial increases in CEPs within 4h. Subsequently, these cells home to the tumor site in large numbers and promote angiogenesis which accompanies the rapid tumor growth from the viable rim. An example of the massive homing and invasion of BMDCs, including EPCs (which incorporate into the lumen of the blood vessel wall), after therapy with a VDA is presented in Fig. 12.2. We also demonstrated that when an antiangiogenic treatment (i.e., DC101) which can block CEP mobilization (Shaked et  al. 2005a; Shaked et  al. 2006b) was coadministered with the VDA, or when tumor bearing mice that are deficient in CEP mobilization (i.e., Id-1+/-Id-3-/- mice) (Lyden et  al. 2001) are used, a minimal residual viable tumor rim was observed after OXi-4503 treatment. Importantly, this phenomenon was not restricted to VDAs, as it was also observed following administration of other (but not all) cytotoxic chemotherapeutic drug therapies. Thus we found that certain chemotherapy drugs administered near or at the MTD, e.g., paclitaxel can induce a rapid spike in CEP levels but, again, this can be largely blocked by DC101 which enhanced treatment impact. However, other chemotherapy drugs, e.g., gemcitabine, which does not induce such a CEP spike, the addition of DC101, had no additional therapeutic benefit (Bertolini et  al. 2006; Shaked et  al. 2008). Thus, these observations indicate that in addition to the anti-tumor activity of chemotherapy, the subsequent tumor growth that occurs may be facilitated by a rapid influx of CEPs. The administration of an antiangiogenic drug (or other treatment strategies (Shaked et al. 2005b)) concomitantly with chemotherapy that can block

12  Impact of Endothelial Progenitor Cells on Tumor Angiogenesis and Outcome

267

Fig. 12.2  Colonization of GFP± bone marrow cells in LLC tumors grown in mice treated with OXi-4503, DC101, or the combination of the two drugs. Lewis Lung carcinoma grown in lethally irradiated mice that were previously transplanted with GFP+ bone marrow tagged cells (green) were treated with DC101, the VEGFR-2 blocking antibody, OXi-4503 or a combination of the two drugs. Three days later, tumors were removed and stained for blood vessels using the CD31 endothelial cell marker (red). Untreated mice revealed some GFP+ BMDCs that did not directly incorporated into the tumor blood vessels. In contrast, massive invasion of BMDCs to the viable tumor rim was observed in mice treated with OXi-4503, some of which were incorporated into the tumor blood vessel lumen. The combination of DC101 and OXi-4503 resulted in the absence of some BMDCs reside at the tumor viable rim. Bottom panels demonstrate the incorporation of BMDCs into the tumor vasculature (co-localization of CD31 and GFP), but also a number of other BMDCs which are adjacent to the tumor vessels and may support tumor angiogenesis in a paracrine manner. Scale bars: Upper 50 µm; bottom 20 µm. Reproduced from Shaked et al. (2006b), with permission from the publisher

tumor cell repopulation mediated by rebound angiogenesis can result in prolongation of the duration of tumor response. These results may also suggest how antiangiogenic drugs may act to enhance the efficacy of standard chemotherapy (Shaked and Kerbal 2007; Kerbal 2006). So how do these results provide an entirely new perspective regarding the ongoing controversy of the contribution of CEPs to tumor angiogenesis? The answer is that all previously published studies showing very low or non-existent levels of EPC incorporation in tumor vessels were performed on untreated tumors, which were removed at various stages of tumor growth. In contrast, there are circumstances, e.g., after acute cytotoxic therapy, where our results show that CEPs may play a significant role in tumor angiogenesis and tumor (re)growth; thus, targeting these cells at these critical time points may lead to significant therapeutic benefits. The robust and acute CEP host responses induced by drugs such as VDAs or paclitaxel may be a consequence, at least in part, of an attempt to repair damage to the tumor vasculature caused by such drugs. We should stress, however, that many other BMDC populations may also contribute to the overall host response and tumor vessel repair process following cytotoxic drug therapy, i.e., this phenomenon

268

R.S. Kerbel et al.

in all likelihood is not restricted to CEPs. It will be of great interest to assess the possibility that cytotoxic therapies, including radiation, induce the mobilization and tumor homing of other BMDC populations such as monocytes and macrophages (Brown et al. 2010). Another circumstance which implicates an important role of EPCs in tumor growth and angiogenesis even when they are present in very limited numbers comes from a study by Mittal and colleagues (Gao et al. 2008). Since the development of macroscopic metastases from micrometastases also requires angiogenesis (Woodhouse et al. 1997), Gao et al. (2008) recently studied whether EPCs are crucial regulators of the “angiogenic switch” mediating progression of metastasis. By using two models of pulmonary metastasis in mice, they demonstrated that the vessels of macrometastases contained approximately 11% EPCs. They found that this was regulated by the Id1 transcription factor (Shaked et al. 2006b), since the blockade of Id1 by short hairpin RNA (shRNA) resulted in impaired angiogenesis, and lack of formation of macrometastases. Importantly, this study not only demonstrated a role of EPCs in metastatic growth, but also indicated that even a relatively small percentage of EPC incorporation into metastatic blood vessels (~11%), can nevertheless induce a major biologic effect in tumor biology. As such, pharmacologically targeting these relatively rare cells may be therapeutically promising. The rapid and robust increase in CEP levels following cytotoxic drug therapy has been recently confirmed in clinical studies. With respect to VDAs, in a phase I clinical study patients treated with cisplatinum in combination with the tubulin binding VDA, AVE8062, exhibited increases in the number of CEPs within the first 3 days (Farace et al. 2007). In this study, CEPs were evaluated by flow cytometry and were defined by the surface markers CD45(dim), VEGFR2+, and CD34+. Similar results were reported based on a phase I clinical study of the VDA, ZD6126, conducted by Beerepoot et al. (2006). The authors measured levels of CECs (but not CEPs), and found that levels of such cells are significantly elevated within 2–8h after therapy. The authors suggested that the rise in CECs levels is probably a consequence of endothelial cell shedding from the damaged tumor vasculature in response to VDA therapy. Although the authors’ conclusion focused on the levels of CECs, no precursor marker for CEP subset was evaluated at that time, and it is plausible that some of the cells detected by flow cytometry are in fact CEPs (Shaked et al. 2006). Furthermore, an additional clinical study testing the combination of a VDA with an antiangiogenic drug has been conducted in a phase I trial involving combretastatin A 4 phosphate (CA4P) and bevacizumab (Nathan et  al. 2008). Preliminary results have been presented by Nathan and colleagues (2008) who reported that the combination treatment revealed statistically significant reductions in tumor perfusion/vascular permeability which reversed after CA4P alone but which were sustained following bevacizumab treatment using DCE-MRI. In addition they observed increases in circulating CD34+ and CD133+ bone marrow progenitors following CA4P treatment as well as increases in VEGF and GCSF levels, in a similar manner to a recent preclinical study (Shaked et al. 2009). Nathan et al. (2008) concluded that CA4P in combination with bevacizumab appears safe and well tolerated with evidence of clinical activity. With respect to chemotherapy,

12  Impact of Endothelial Progenitor Cells on Tumor Angiogenesis and Outcome

269

we reported that cancer patients treated with paclitaxel based therapy but not gemcitabine- or doxorubicin-based therapies exhibited increases in CEP levels along with SDF-1 and other proangiogenic factors (Shaked et  al. 2008; Shaked et al. 2009). In addition, Voest and colleagues have also reported that MTD paclitaxel exhibit an immediate increase in CEP levels 4h after treatment was initiated, but these immediate changes did not correlate with response to therapy. However, sustained and consistent high levels of CEPs measured on day 7 (regardless of the chemotherapy drug used) were associated with an increase in tumor growth and could predict progression free and/or overall survival (Roodhart et  al. 2010). Overall, these results further reinforce the concept of rapid cytotoxic drug induced host CEP mobilization, and its blockade by an antiangiogenic drug which results in increased treatment efficacy.

Conclusions The role of EPCs in post-natal angiogenesis, and specifically during tumor growth remains controversial. Although many studies have shown vessel incorporation and a functional role of EPCs in tumor biology, other studies have failed to reproduce such results. We and others have shown that in certain circumstances EPCs may play a major role in tumor angiogenesis, mainly with respect to repair of therapy-induced tumor vascular damage induced by certain drugs, e.g., VDAs and MTD chemotherapy. In such circumstances we have found that EPCs rapidly home to the tumor site and promote an active angiogenesis process. Clinically, EPCs and CECs have been investigated as possible surrogate biomarkers to predict clinical outcome in patients undergoing antiangiogenic therapies. However, such measurements require further evaluation and validation using sophisticated flow cytometry technology. Overall, a consensus regarding the nature of EPCs and their relative functional importance in tumor angiogenesis and anti-cancer therapy remains to be determined.

References Asahara T, Murohara T, Sullivan A et al (1997) Isolation of putative progenitor endothelial cells for angiogenesis. Science 275:964–967. Beerepoot LV, Radema SA, Witteveen EO et  al (2006) Phase I clinical evaluation of weekly administration of the novel vascular-targeting agent, ZD6126, in patients with solid tumors. J Clin Oncol 24:1491–1498. Bender JL, Adamson PC, Reid JM et al (2008) Phase I trial and pharmacokinetic study of bevacizumab in pediatric patients with refractory solid tumors: a Children’s Oncology Group Study. J Clin Oncol 26:399–405. Bertolini F, Paul S, Mancuso P et al (2003) Maximum tolerable dose and low-dose metronomic chemotherapy have opposite effects on the mobilization and viability of circulating endothelial progenitor cells. Cancer Res 63:4342–4346.

270

R.S. Kerbel et al.

Bertolini F, Shaked Y, Mancuso P, Kerbel RS (2006) The multifaceted circulating endothelial cell in cancer: from promiscuity to surrogate marker and target identification. Nature Rev Cancer 6:835–845. Brown M et al (2010), J Clin Investig (in press). Buckstein R, Kerbel RS, Shaked Y et al (2006) High-Dose celecoxib and metronomic “low-dose” cyclophosphamide is an effective and safe therapy in patients with relapsed and refractory aggressive histology non-Hodgkin’s lymphoma. Clin Cancer Res 12:5190–5198. Calleri A, Bono A, Bagnardi V et al (2009) Predictive Potential of Angiogenic Growth Factors and Circulating Endothelial Cells in Breast Cancer Patients Receiving Metronomic Chemotherapy Plus Bevacizumab. Clin Cancer Res 15:7652–7657. Case J, Mead LE, Bessler WK et  al (2007) Human CD34+AC133+VEGFR-2+ cells are not endothelial progenitor cells but distinct, primitive hematopoietic progenitors. Exp Hematol 35:1109–1118. Coffelt SB, Hughes R, Lewis CE (2009) Tumor-associated macrophages: effectors of angiogenesis and tumor progression. Biochim Biophys Acta 1796:11–18. Cubbon RM, Kahn MB, Wheatcroft SB (2009) Effects of insulin resistance on endothelial progenitor cells and vascular repair. Clin Sci (Lond)117:173–190. Dellapasqua S, Bertolini F, Bagnardi V et  al (2008) Metronomic Cyclophosphamide and Capecitabine Combined With Bevacizumab in Advanced Breast Cancer. J Clin Oncol 26(30):4899-4905. De Palma M, Naldini L (2006) Role of haematopoietic cells and endothelial progenitors in tumour angiogenesis. Biochim Biophys Acta 1766:159–166. De Palma M, Venneri MA, Galli R et al (2005) Tie2 identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitors. Cancer Cell 8:211–226. Distler JH, Beyer C, Schett G, Luscher TF, Gay S, Distler O (2009) Endothelial progenitor cells: novel players in the pathogenesis of rheumatic diseases. Arthritis Rheum 60:3168–3179. Duda DG, Cohen KS, di Tomaso E et  al (2006) Differential CD146 expression on circulating versus tissue endothelial cells in rectal cancer patients: implications for circulating endothelial and progenitor cells as biomarkers for antiangiogenic therapy. J Clin Oncol 24:1449–1453. Elshal MF, Khan SS, Takahashi Y, Solomon MA, McCoy JPJr (2005) CD146 (Mel-CAM), an adhesion marker of endothelial cells, is a novel marker of lymphocyte subset activation in normal peripheral blood. Blood 106:2923–2924. Farace F, Massard C, Borghi E, Bidart JM, Soria JC (2007) Vascular disrupting therapy-induced mobilization of circulating endothelial progenitor cells. Ann Oncol 18:1421–1422. Gammill HS, Lin C, Hubel CA (2007) Endothelial progenitor cells and preeclampsia. Front Biosci 12:2383–2394. Gao D, Nolan DJ, Mellick AS, Bambino K, McDonnell K, Mittal V (2008) Endothelial progenitor cells control the angiogenic switch in mouse lung metastasis. Science 319:195–198. Grunewald M, Avraham I, Dor Y et al (2006) VEGF-induced adult neovascularization: recruitment, retention, and role of accessory cells. Cell 124:175–189. Hanahan D, Bergers G, Bergsland E (2000) Less is more, regularly: metronomic dosing of cytotoxic drugs can target tumor angiogenesis in mice. J Clin Invest 105:1045–1047. Horrevoets AJ (2009) Angiogenic monocytes: another colorful blow to endothelial progenitors. Am J Pathol 174:1594–1596. Jin DK, Shido K, Kopp HG et al (2006) Cytokine-mediated deployment of SDF-1 induces revascularization through recruitment of CXCR4+ hemangiocytes. Nat Med 12:557–567. Kaplan RN, Psaila B, Lyden D (2007) Niche-to-niche migration of bone-marrow-derived cells. Trends Mol Med 13:72–81. Kerbel RS (1991) Inhibition of tumor angiogenesis as a strategy to circumvent acquired resistance to anti-cancer therapeutic agents. BioEssays 13:31–36. Kerbel RS (2006) Antiangiogenic therapy: a universal chemosensitization strategy for cancer? Science 312:1171–1175.

12  Impact of Endothelial Progenitor Cells on Tumor Angiogenesis and Outcome

271

Kerbel RS, Benezra R, Lyden DC et al (2008) Endothelial progenitor cells are cellular hubs essential for neoangiogenesis of certain aggressive adenocarcinomas and metastatic transition but not adenomas. Proc Natl Acad Sci U S A 105:E54. Kim I, Yilmaz OH, Morrison SJ (2005) CD144 (VE-cadherin) is transiently expressed by fetal liver hematopoietic stem cells. Blood 106:903–905. Kim JJ, Tannock IF (2005) Repopulation of cancer cells during therapy: an important cause of treatment failure. Nat Rev Cancer 5:516–525. Kim SJ, Kim JS, Papadopoulos J et al (2009) Circulating monocytes expressing CD31: implications for acute and chronic angiogenesis. Am J Pathol 174:1972–1980. Klement G, Baruchel S, Rak J et  al (2000) Continuous low-dose therapy with vinblastine and VEGF receptor-2 antibody induces sustained tumor regression without overt toxicity. J Clin Invest 105:R15–R24. Lawler J (2002) Thrombospondin-1 as an endogenous inhibitor of angiogenesis and tumor growth. J Cell Mol Med 6:1–12. Li H, Gerald WL, Benezra R (2004)Utilization of bone marrow-derived endothelial cell precursors in spontaneous prostate tumors varies with tumor grade. Cancer Res 64:6137–6143. Lyden D, Hattori K, Dias S et al (2001) Impaired recruitment of bone-marrow-derived endothelial and hematopoietic precursor cells blocks tumor angiogenesis and growth. Nat Med 7:1194–1201. Lyden D, Young AZ, Zagzag D et al (1999) Id1 and Id3 are required for neurogenesis, angiogenesis and vascularization of tumour xenografts. Nature 401:670–677. Machein MR, Renninger S, Lima-Hahn E, Plate KH (2003) Minor contribution of bone marrowderived endothelial progenitors to the vascularization of murine gliomas. Brain Pathol 13:582–597. Madlambayan GJ, Butler JM, Hosaka K et al (2009) Bone marrow stem and progenitor cell contribution to neovasculogenesis is dependent on model system with SDF-1 as a permissive trigger. Blood 114:4310–4319. Mancuso P, Antoniotti P, Quarna J et al (2009) Validation of a standardized method for enumerating circulating endothelial cells and progenitors: flow cytometry and molecular and ultrastructural analyses. Clin Cancer Res 15:267–273. Mancuso P, Burlini A, Pruneri G, Goldhirsch A, Martinelli G, Bertolini F (2001) Resting and activated endothelial cells are increased in the peripheral blood of cancer patients. Blood 97:3658–3661. Mancuso P, Colleoni M, Calleri A et al (2006) Circulating endothelial cell kinetics and viability predict survival in breast cancer patients receiving metronomic chemotherapy. Blood 108:452–459. Martin-Padura I, Bertolini F (2008) The circulating endothelial cell in cancer: towards marker and target identification. Curr Pharm Des 14:3780–3789. Nathan PD, Judson I, Padhani A et  al (2008) A phase I study of combretastatin A4 phosphate (CA4P) and bevacizumab in subjects with advanced solid tumors. J Clin Oncol 26(15S, Abstract No. 3550), —20 May 2008. Peichev M, Naiyer AJ, Pereira D et al (2000) Expression of VEGFR-2 and AC133 by circulating human CD34(+) cells identifies a population of functional endothelial precursors. Blood 95:952–958. Peters BA, Diaz LA, Polyak K et al (2005) Contribution of bone marrow-derived endothelial cells to human tumor vasculature. Nat Med 11:261–262. Pitchford SC, Furze RC, Jones CP, Wengner AM, Rankin SM (2009) Differential mobilization of subsets of progenitor cells from the bone marrow. Cell Stem Cell 4:62–72. Prewett M, Huber J, Li Y et al (1999) Antivascular endothelial growth factor receptor (fetal liver kinase 1) monoclonal antibody inhibits tumor angiogenesis and growth of several mouse and human tumors. Cancer Res 59:5209–5218. Purhonen S, Palm J, Rossi D et al (2008) Bone marrow-derived circulating endothelial precursors do not contribute to vascular endothelium and are not needed for tumor growth. Proc Natl Acad Sci U S A 105:6620–6625.

272

R.S. Kerbel et al.

Rafii S, Lyden D, Benezra R, Hattori K, Heissig B (2002) Vascular and haematopoietic stem cells: novel targets for anti-angiogenesis therapy? Nat Rev Cancer 2:826–835. Rohan RM, Fernandez A, Udagawa T, Yuan J, D’Amato RJ (2000) Genetic heterogeneity of angiogenesis in mice. FASEB J 14:871–876. Roodhart JM, Langenberg MH, Vermaat JS et  al (2010) Late release of circulating endothelial cells and endothelial progenitor cells after chemotherapy predicts response and survival in cancer patients. Neoplasia 12:87–94. Ruzinova MB, Schoer RA, Gerald W et al (2003) Effect of angiogenesis inhibition by Id loss and the contribution of bone-marrow-derived endothelial cells in spontaneous murine tumors. Cancer Cell 4:277–289. Schneider M, Tjwa M, Carmeliet P (2005) A surrogate marker to monitor angiogenesis at last. Cancer Cell 7:3–4. Shaked Y, Bertolini F, Emmenegger U, Lee CR, Kerbel RS. (2006a) On the origin and nature of elevated levels of circulating endothelial cells after treatment with a vascular disrupting agent. J Clin Oncol 24:4040–4041. Shaked Y, Bertolini F, Man S et al (2005a) Genetic heterogeneity of the vasculogenic phenotype parallels angiogenesis: implications for cellular surrogate marker analysis of antiangiogenesis. Cancer Cell 7:101–111. Shaked Y, Ciarrocchi A, Franco M et al (2006b) Therapy-induced acute recruitment of circulating endothelial progenitor cells to tumors. Science 313:1785–1787. Shaked Y, Emmenegger U, Francia G et al (2005b) Low-dose metronomic combined with intermittent bolus-dose cyclophosphamide is an effective long-term chemotherapy treatment strategy. Cancer Res 65:7045–7051. Shaked Y, Emmengger U, Man S et al (2005c) The optimal biological dose of metronomic chemotherapy regimens is associated with maximum antiangiogenic activity. Blood 106:3058–3061. Shaked Y, Henke E, Roodhart JM et  al (2008) Rapid chemotherapy-induced acute endothelial progenitor cell mobilization: implications for antiangiogenic drugs as chemosensitizing agents. Cancer Cell 14:263–273. Shaked Y, Kerbel RS (2007) Antiangiogenic strategies on defense: on the possibility of blocking rebounds by the tumor vasculature after chemotherapy. Cancer Res 67:7055–7058. Shaked Y, Tang T, Woloszynek J et al (2009) Contribution of granulocyte colony-stimulating factor to the acute mobilization of endothelial precursor cells by vascular disrupting agents. Cancer Res 69:7524–7528. Shaked Y, Voest EE (2009) Bone marrow derived cells in tumor angiogenesis and growth: are they the good, the bad or the evil? Biochim Biophys Acta 1796:1–4. Shojaei F, Wu X, Zhong C et al (2007) Bv8 regulates myeloid-cell-dependent tumour angiogenesis. Nature 450:825–831. Siemann DW, Chaplin DJ, Horsman MR (2004) Vascular-targeting therapies for treatment of malignant disease. Cancer 100:2491–2499. Solovey AN, Gui L, Chang L, Enenstein J, Browne PV, Hebbel RP (2001) Identification and functional assessment of endothelial P1H12. J Lab Clin Med 138:322–331. Tozer GM, Kanthou C, Baguley BC (2005) Disrupting tumour blood vessels. Nat Rev Cancer 5:423–435. Volpert OV, Pili R, Sikder HA et  al (2002) Id1 regulates angiogenesis through transcriptional repression of thrombospondin-1. Cancer Cell 2:473–483. Werner N, Kosiol S, Schiegl T et al (2005) Circulating endothelial progenitor cells and cardiovascular outcomes. N Engl J Med 353:999–1007. Wickersheim A, Kerber M, de Miguel LS, Plate KH, Machein MR (2009) Endothelial progenitor cells do not contribute to tumor endothelium in primary and metastatic tumors. Int J Cancer 125:1771–1777. Woodhouse EC, Chuaqui RF, Liotta LA (1997) General mechanisms of metastasis. Cancer 80:1529–1537.

12  Impact of Endothelial Progenitor Cells on Tumor Angiogenesis and Outcome

273

Yap R, Veliceasa D, Emmenegger U et  al (2005) Metronomic low-dose chemotherapy boosts CD95-dependent antiangiogenic effect of the thrombospondin peptide ABT-510: a complementation antiangiogenic strategy. Clin Cancer Res 11:6678–6685. Yoder MC, Mead LE, Prater D et  al (2007) Redefining endothelial progenitor cells via clonal analysis and hematopoietic stem/progenitor cell principals. Blood 109:1801–1809. Ziegelhoeffer T, Fernandez B, Kostin S et al (2004) Bone marrow-derived cells do not incorporate into the adult growing vasculature. Circ Res 94:230–238.

Chapter 13

Bone Marrow Derived Mesenchymal Stem/Stromal Cells and Tumor Growth Pravin J. Mishra and Debabrata Banerjee

Abstract  Carcinoma associated fibroblasts (CAFs) play an important role in the growth of epithelial solid tumors. The origin of these tumor or CAFs has not been conclusively established. There is experimental evidence to suggest that part of the tumor or CAFs may arise from bone marrow derived mesenchymal stromal/stem cells or MSCs. It is well known that bone marrow derived MSCs can give rise to cells of different lineages: muscle, bone, fat, and cartilage. Based on recent work from our own laboratory and that of others, we now suggest that human BM-derived MSCs exposed to tumor-conditioned medium (TCM) over a prolonged period of time can give rise to cells that assume a CAF-like phenotype. Thus, MSCs may be a source of CAFs and can be used experimentally for modeling tumor-stroma interactions. Although the importance of the dialog between cancer cells and other components of the tumor milieu has been increasingly appreciated, it is as yet unclear whether the stromal cells themselves harbor cancer promoting mutations or changes. Activated stromal cells have been shown to promote tumor growth and metastasis in experimental models and we speculate on the possibility of increased activation of bone marrow derived MSCs by higher levels of chemokines under certain physiological situations and how this may impact tumor growth.

D. Banerjee (*) Department of Medicine, The Cancer Institute of New Jersey, Robert Wood Johnson Medical School, University of Medicine and Dentistry of New Jersey, 195 Little Albany Street, New Brunswick, NJ 08901, USA and Department of Pharmacology, The Cancer Institute of New Jersey, Robert Wood Johnson Medical School, University of Medicine and Dentistry of New Jersey, 195 Little Albany Street, New Brunswick, NJ 08901, USA e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_13, © Springer Science+Business Media, LLC 2010

275

276

P.J. Mishra and D. Banerjee

Introduction Most solid tumors are made up of two discernable compartments, tumor or neoplastic cells and the stromal platform on which the tumor cells are scattered. Besides these, there are many other cell types present in the tumor milieu including tumor associated macrophages (TAMs), lymphocytes, endothelial cells, fibroblasts, and pericytes that interact with each other either directly or via paracrine mechanisms and influence tumor growth. The origin and function of tumor stroma has been the subject of intense investigation ever since Virchow first observed that cancers were infiltrated with inflammatory and immune cells (Balkwill and Mantovani 2001). The presence of fibroblast populations within human tumors often referred to as carcinoma associated fibroblasts (CAFs) has been associated with poor outcome and an increase in metastatic potential (Tsujino et  al. 2007; Yazhou et  al. 2004). Experimental evidence from tumor models as well as from clinical specimens suggest that CAFs may promote tumor growth and metastasis. In this chapter we will discuss the likely source of these CAFs as well as their characteristics and hope to convince the readers that bone marrow derived mesenchymal stem/stromal cells (MSCs) may be a likely source of CAFs. We will also discuss how this can be an important model system to study tumor-stroma interaction. Lastly, we will speculate on the possibility of increased activation of bone marrow derived MSCs by higher levels of chemokines in circulation under certain physiological situations and how this may impact tumor growth.

Characteristics of CAFs Over the years, characterization of CAFs has been largely based on immunohistochemical staining and more recently by specific gene expression signatures (Allinen et al. 2004; Hu and Polyak 2008; Chang et al. 2005). These have relied upon identification and or isolation of the CAFs from tumor specimens. We and others have developed in vitro methods to generate CAFs and study them in model systems to understand the role of these CAFs in tumor growth and metastasis as well as to better define paracrine factors that participate in the tumor stroma dialog (Mishra et al. 2008; Spaeth et al. 2009). Advantages of an in vitro system include better control over cell type, identification and characterization of CAFs and generation of large number of cells in culture to carry out studies in various model systems including xenograft studies. In order to generate CAFs (also referred to as tumor associated fibroblasts or TAFs) in vitro, long term exposure to tumor cell conditioned medium has been the method of choice. Spaeth et al. (2009) have defined the tumor or CAFs involved in ovarian cancer growth (as well as breast and prostate) by expression of four qualifying factors: (1) fibroblast markers fibroblast-specific protein (FSP)

13  Bone Marrow Derived Mesenchymal Stem/Stromal Cells and Tumor Growth

277

and fibroblast activating protein (FAP); (2) genes associated with an increase of tumor aggression, including stromelysin-1 (SL-1), thrombospondin-1 (Tsp-1), and tenascin-C (Tn-C); (3) myofibroblast/provascularizing potential including desmin, a-smooth muscle actin (a-SMA), and vascular endothelial growth factor (VEGF); and lastly, (4) growth factors, transforming growth factor-beta (TGF-b), HGF/ scatter factor (SF), basic fibroblast growth factor (bFGF), and epidermal growth factor (EGF). Bone marrow derived MSCs have been suggested to play a major role in tissue regeneration and repair based on their multilineage differentiation potential (Pittenger et al. 1999; Picinich et al. 2007) as well as the propensity of these cells to migrate to areas of injury, inflammation and tumor (Fox et al. 2007). Studeny et al. (2002) originally demonstrated the capacity of MSCs, also known as multipotent stromal cells, to home to tumors and participate in tumor stroma formation, suggesting MSCs as a potential source of stroma. The observed tendency of MSCs to migrate to sites of injury, inflammation and tumors is attributed to common chemoattractants or factors present in the wound and tumor milieu leading to the concept that tumors are unhealed wounds (Dvorak 1986). Our own work on role of MSCs in wound healing has revealed remarkable similarities in the process of wound healing and tumor growth with some subtle differences such as in production of cytokines by MSCs in the wound bed or in the presence of keratinocytes, the principal epithelial cells in skin (Mishra et al. unpublished observations). The ability of MSCs to migrate to tumor sites has been exploited to deliver therapeutic genes to tumor sites using MSCs as carriers. (Studeny et al. 2002, 2004; Nakamizo et al. 2005; Hung et al. 2005).

Bone Marrow Derived MSCs as Source of CAFs Accumulating evidence suggests that CAFs or TAFs play an important role in the growth of epithelial solid tumors. It has been demonstrated that a significant fraction of the stroma in some breast cancers consists of fibroblasts (Bissell and Radisky 2001). More recent studies have revealed that CAFs from breast cancer specimens promote tumor cell growth as compared to fibroblasts obtained from non-neoplastic locations. In addition to tumor growth, tumor stroma has also been shown to be involved in other important processes such as angiogenesis and metastasis. Orimo and colleagues (2005) defined several important characteristics of breast CAFs including promotion of breast carcinoma cell growth, promotion of angiogenesis, and expression of myofibroblast traits. Expression of the chemokine stromal-derived factor 1 (SDF-1) has also been shown to be of paramount importance in the interaction between tumor cells and stromal fibroblasts (Orimo et al. 2005). Given the importance of tumor stroma in promoting tumor growth, it becomes necessary to identify the likely source(s) of cells that make up the tumor stroma, particularly the carcinoma associated fibroblasts or CAFs.

278

P.J. Mishra and D. Banerjee

It has been suggested that the likely source (s) of tumor stroma include (1) the locally resident tissue stem cells that are recruited to the tumor site; (2) epithelial to mesenchymal transition or EMT; (3) fibroblast recruitment into the tumor stroma; and (4) recruitment of bone marrow-derived cells from the circulation (Udagawa et al. 2006; Koyama et al. 2008; Jodele et al. 2005). We and others have demonstrated that bone marrow derived MSCs, like other bone marrow-resident cells, have the capacity to differentiate within the tumor microenvironment into myofibroblastic-like cells commonly referred to as; myofibroblasts, tumor-associated (myo)fibroblasts (TAF), CAF, fibrocytes or pericytes (Ogawa et al. 2006; Mishra et al. 2008; Spaeth et al. 2009) which have been shown to play an important role in tumor formation, growth and metastasis. Although the cell type of origin of myofibroblasts has not been conclusively established it has been shown that they may be bone marrow derived (Direkze et al. 2004). In a recent study we demonstrated that human bone marrow-derived mesenchymal stem cells (hMSCs) when exposed to tumor-conditioned medium (TCM) over a prolonged period of time assume a CAF-like phenotype (Mishra et al. 2008) and this was confirmed by Spaeth et  al. (2009). More importantly, these cells exhibit functional properties of CAFs including sustained expression of SDF-1 and the ability to promote tumor cell growth both in vitro and in an in vivo co-implantation model. These cells also express myofibroblast markers including a-SMA and FSP. Gene expression profiling revealed similarities between TCM exposed hMSCs and CAFs. This suggests that hMSCs are a source of CAFs and can be used in modeling tumor-stroma interactions (Mishra et al. 2008; Karnoub et al. 2007). Further evidence for a bone marrow source of CAFs comes from studies using a gastric cancer mouse model (Gan mice) in which prostaglandin E2 (PG E2) and Wnt signaling were simultaneously activated in the gastric mucosa (Guo et  al. 2008). Since PGE2 and Wnt pathways both play a role in human gastric tumorigenesis, the Gan mouse model may display important aspects of the molecular etiology of human gastric cancer. Bone marrow transplantation (BMT) experiments revealed that subsets of gastric myofibroblasts were derived from bone marrow (Guo et al. 2008). In a recent report, Worthley et  al. (2009) have shown for the first time that human gastrointestinal neoplasia associated myofibroblasts can develop from bone marrow derived cells following allogeneic BMT based on identification of SMA expressing, Y chromosome positive CD45 negative cells in the tumor milieu of female recipients of BMT who went on to develop solid tumors following allogeneic male bone marrow/stem cells transplants. This clinical study supports the previous in vitro as well as in vivo animal model studies that suggested bone marrow origin of CAFs (Ogawa et al. 2006). Although literature supports the bone marrow origin of CAFs, alternate sources including resident fibroblasts and adipose tissue cells are also sources of CAFs. In a recent report Zhang et al. (2009) demonstrate that white adipose tissue derived stromal cells are also recruited to tumor stroma and promote cancer progression in animal models of human cancers including Kaposi’s sarcoma, prostate and breast cancer models.

13  Bone Marrow Derived Mesenchymal Stem/Stromal Cells and Tumor Growth

279

Alterations in Tumor Associated Stromal Cells Although the importance of the dialog between cancer cells and other components of the tumor milieu has been increasingly appreciated, the question of whether the stromal cells themselves harbor cancer promoting mutations or changes is just beginning to be addressed. For example, Patocs et  al. (2007) hypothesized that mutational inactivation of the tumor-suppressor gene TP53 and genomic alterations in stromal cells of a tumor’s microenvironment may be correlated to clinical outcome. The authors, in an analysis of somatic and stromal cell mutations, demonstrated that stroma-specific loss of heterozygosity or allelic imbalance was associated with somatic p53 mutations and regional lymph-node metastases in sporadic breast cancer but not in hereditary breast cancer. A previous study in head and neck squamous cell carcinoma (SCC) by the same investigators looked at whether the apparently “normal” stroma surrounding the tumor epithelium can acquire genomic alterations and contribute to cancer initiation and or progression. Tumor-associated stroma of head and neck SCC from smokers was found to have a high degree of genomic alterations. The results indicated that stroma-specific genetic alterations may have been associated with smoking-related head and neck SCC genesis (Weber et al. 2007). Whether the identified mutations correlate with stromal changes or whether this is a reflection of methodological artifacts remains unresolved and larger studies on tumor associated stromal cells from a variety of tumor specimens may be helpful in settling this contentious issue (Eng et al. 2009; Campbell et al. 2009). It has become clear that the initiation and progression of carcinomas depend not only on alterations in tumor epithelial cells, but also on changes in their microenvironment. To study changes in stromal cells within the tumor milieu Hasawi et al. (2008) undertook to characterize CAFs and their tumor counterpart fibroblasts (TCFs) at the cellular and molecular level in a small subset of breast cancer patients using normal breast fibroblasts (NBFs) from plastic surgery as a control. The results suggested that the p53/p21 response to gamma-radiation was attenuated in 70% of CAFs, whereas it was normal in all the TCF and NBF cells. These results indicate that alterations in the p53 pathway can occur in breast CAFs and their corresponding adjacent counterparts, further pointing to the important role that stroma may play in breast carcinogenesis and treatment. Alterations in the tumor suppressor p53 have been reported in tumor-associated stromal cells, however, the consequence of these alterations are not well understood. Dudley et  al. (2008) have investigated p53 status and response to p53-activating drugs using tumor-associated stromal cells from A375 melanoma and PC3 prostate carcinoma xenografts, as well as from a spontaneous prostate tumor model (TRAMP). Unlike normal stromal cells, tumor associated stromal cells failed to arrest in G2 after etoposide treatment, failed to upregulate p53-inducible genes, and failed to undergo apoptosis after treatment with vincristine. Tumor-associated stromal cells were also found to be less sensitive to p53-activating drugs. Knockdown of p53 in normal stromal cells produced similar results strongly supporting the

280

P.J. Mishra and D. Banerjee

contention that there was loss of p53 response in tumor associated stromal cells. Investigations into the role of tumor associated stromal cells in supporting aerobic glycolysis (also called the Warburg effect) has revealed that activation of highly conserved mammalian uncoupling proteins through interaction with the tumor milieu may facilitate the Warburg effect in the absence of permanent respiratory impairment (Samudio et  al. 2008). Additionally, tumor associated stromal cells have been reported to cooperate with tumor cells by taking up the lactate produced by tumor cells and secreted into the tumor microenvironment. The stromal cells take up the secreted lactate via monocarboxylate transporters (MCT1 and MCT2) and after conversion to pyruvate secrete it back into the extracellular milieu where it can be used by local cell constituents for oxidative phosphorylation (Samudio et al. 2008; Koukourakis et al. 2006). Our own observations indicate that exposure of bone marrow derived MSCs to 15 mM lactate for periods as short as 5 min results in upregulation of MCT1 protein as determined by western blotting (Rattigan and Banerjee 2009, unpublished data). Although little is known regarding how changes in stromal gene expression affect epithelial tumor progression, it is becoming clear that cancer is influenced by signals emanating from tumor stroma. Finak et al. (2008) studied gene expression profiles of tumor stroma from 53 primary breast tumors obtained by laser capture microdissection and were able to generate a novel stroma-derived prognostic predictor (SDPP) that correlated with disease outcome independently of standard clinical prognostic factors. The SDPP gene expression signature when applied to several previously published tumor-derived expression data sets was able to identify poor-outcome individuals from amongst multiple clinical subtypes, including lymph node-negative tumors. The SDDP signature appears to be an improved prognostic predictor compared to previously published methods, particularly for HER2-positive breast tumors. Genes showing strong prognostic tendencies included those associated with differential immune responses and angiogenic and hypoxic responses underscoring the importance of stromal biology in tumor progression (Kroemer and Pouyssegur 2008). The plasticity of both the epithelial tumor cells and bone marrow derived MSCs and its impact on tumor biology remain a subject of intense investigation. On the one hand tumor cells are known to undergo epithelial mesenchymal transition (Salomon and Thiery 2003; Radisky 2005) while on the other hand, MSCs are also capable of mesenchymal epithelial transition (Chaffer et al. 2006) thus adding to the complexity of cell types in the microenvironment.

Implications of MSCs as a Source of CAFs: A Model to Study Tumor Stroma Interactions In vitro generation of CAFs following exposure to TCM as well as the admixed injection of tumor cells and bone marrow derived MSCs hold great promise as systems where tumor stroma interaction can be manipulated and studied. Additionally, this may also provide a cell culture method for generating one of the important cell types of the tumor stroma, the activated myofibroblasts.

13  Bone Marrow Derived Mesenchymal Stem/Stromal Cells and Tumor Growth

281

A better understanding of the dialog between various bone marrow derived cell types and tumor cells within the tumor milieu will be important for improved tumor therapy that takes into account influence of the tumor milieu on survival and growth of tumor cells. We have initiated studies to model the tumor-stroma interaction in vitro by culturing TCM exposed BMDMSCs (representing CAFs), reporter gene marked tumor cells (to report growth of tumor cells in the cell mix) and other components of the tumor milieu. It is now becoming clearer that CAFs and other cells in the tumor milieu actively participate in altering the growth and drug response of tumors (Iwamoto et al. 2007; Mantovani et al. 2006). For example, MSCs, which have very high levels of asparaginase expression, can protect leukemic cells from asparaginese cytotoxicity by providing increased concentrations of asparagine in the vicinity of leukemic cells. This system can be exploited to study the contribution of freshly harvested TAFs from dissected tumors on growth of a similar type of tumor cell. By establishing a panel of reporter gene marked tumor cells representing a variety of tumor types and subtypes, it may be possible to determine influence of TAFs on tumor growth. Specific reporters can also be generated for pathways that may be activated in different tumors, for example one can study whether androgen independent prostate cancer cells are still influenced by the tumor stroma. The relative ease with which the reconstituted tumor in its microenvironment can be transplanted as a xenograft may also permit improved drug studies in vivo. An important aspect of the reconstituted tumor microenvironment is the ability to evaluate chemopreventive measures in vitro and in vivo. By targeted pretreatment of critical components of the reconstituted tumor microenvironment, it may be possible to rank chemopreventive agents by potency as well as cell type specificity. Further applications of the reconstituted tumor milieu system include study of gap junctions and other direct communication means between tumor cells and other cell types. This type of experimental system may provide a better simulation of an in vivo solid tumor and may be a more realistic model for investigation of tumor biology and experimental therapeutics.

Activation of BMD MSCs and Growth of Tumors Speculation on Role of Chemokines on Activation of Circulating MSCs and Effect on Tumor Growth in African American Individuals with Breast Cancer Caucasian American women have higher age-adjusted breast cancer incidence rates compared with African American women (143 per 100,000 versus 119 per 100,000; Ries et  al. 2002). However, African American women have higher age-adjusted mortality rates from the disease. Breast cancer mortality rates among younger African American women are approximately twice that of younger Caucasian American women (Amend et al. 2006; Albain et al. 2009). African American breast

282

P.J. Mishra and D. Banerjee

cancer patients also more frequently develop basal like subset of breast cancer (triple negative; ER-, PR- and Her2 low; greater stromal involvement), which has a poorer prognosis (van de Vijver et al. 2002; Sorlie et al. 2001). Gene expression studies reveal that pathways related to tumor angiogenesis and chemotaxis may be upregulated in tumors from African American women as compared with women of Caucasian origin. Increased microvessel density and increased TAMs in tumors from African American patients have been reported (Martin et al. 2009) suggesting involvement of the tumor microenvironment. A majority of African Americans are negative (~68% as compared to 3.4% of Caucasian Americans) for Duffy antigen receptor for chemokines (DARC) and it has been suggested that absence of this non-signaling receptor can increase migration of several types of blood cells from circulation to tissues (Tournamille et al. 1995; Fukuma et al. 2003; Lentsch 2002; Pruenster et al. 2009) supporting the idea of increased migration of activated MSCs to the tumor milieu. A mutation in the erythroid specific promoter GATA-1 binding site in the promoter of the gene encoding DARC results in lack of expression of DARC on erythrocytes. The erythroid DARC clears ELR chemokines such as CXCL8 from circulation while the endothelial DARC may facilitate transport of chemokines from tissue to vascular lumen for presentation (Lee et  al. 2003). Overexpression of DARC in breast cancer cells correlates with a tumor and metastasis suppressor phenotype underscoring the importance of the DARC-negative phenotype in tumor progression. Wang et al. (2006) have shown a strong negative correlation between DARC expression and lymph node metastasis, ER status and microvessel density suggesting that low DARC expression correlated with aggressive disease in an analysis of 75 breast tumors. We have re-analyzed data compiled previously from two separate gene expression profiling studies and discovered that lower relative expression of DARC correlates with basal subtypes (unpublished data; analysis carried out in collaboration with Dr. S. Ganesan at CINJ from data published by Alexe et al. 2007; Wang et al. 2005). Our analysis suggests that lower relative expression of DARC correlates with basal subtypes of breast cancer. The basal subtypes, basal A1 and basal A2 had the lowest expression as compared with the luminal type as well as normal breast tissue. African American breast cancer patients more frequently develop basal subtype of breast cancer. Our analysis did not separate the population by ethnicity supporting an independent association between DARC negativity and the basal subtype. Lack of DARC expression in African American individuals correlates with elevated levels of free chemokines, such as CXCL8, and leads to pathological conditions as shown below. Lack of DARC Expression, Circulating Chemokines and Pathological Conditions It has been shown that DARC negative individuals have higher levels of CXCL8 in supernatants of red cell concentrates than DARC positive individuals suggesting that plasma levels of CXCL8 are higher in the former (Wadhwa et  al. 2000).

13  Bone Marrow Derived Mesenchymal Stem/Stromal Cells and Tumor Growth

283

Moreover, high levels of ELR chemokines such as CXCL8 are found in renal biopsies from patients with HIV associated renal neuropathies seen predominantly in African American children with HIV. It is postulated that individuals who lack the DARC sink for chemokines would have higher levels of circulating chemokines available for endothelial cells to bind and deliver to renal tissues and increase risk of renal injury (Liu et al. 1999). DARC Expression and Cancer in African American Men DARC has been found to act as a multispecific receptor for chemokines of both the C–C and C–X–C families. Using a transgenic model of prostate cancer with DARC-deficient mice, Shen et al. (2006) demonstrated that lack of DARC expression contributed to enhanced prostate tumor growth. Their studies showed that DARC functions to clear angiogenic CXC chemokines (ELR chemokines, the key one being CXCL8) from the prostate tumor microcirculation and that lack of DARC expression in 70% of African Americans may be a contributing factor to the increased mortality to prostate cancer in this population but not to tumor initiation. Although DARC and elevated chemokine levels were implicated, no mechanistic insight was provided by these studies. We have developed the idea further and have generated the novel hypothesis that increased circulating chemokines results in greater number of activated MSCs leading to aggressive tumor growth in DARC negative African American breast cancer patients. Recent data from Mayr et  al. (2007) demonstrates that individuals of African origin have higher levels of CXCL8 than individuals of Caucasian origin. In the analysis from Mayr et al. (2007) 41 Caucasians were of European origin, whereas the remaining 31 individuals of sub-Saharan African origin. CXCL-8 levels averaged 5.7 pg/mL (CI 4.4–7.0) in Africans and 3.3 pg/mL (CI 2.7–3.8) in Caucasians (P< 0.0008*). These levels are consistent with previously determined values in Caucasians. This shows that the lack of erythroid DARC expression correlates with elevated CXCL8 level in plasma.

Activation of Bone Marrow-Derived MSCs and Metastasis Karnoub et  al. (2007) demonstrated that bone-marrow-derived human MSCs increased metastasis potency of MDAMB231 (otherwise weakly metastatic human breast carcinoma cells), when this cell mixture was introduced subcutaneously in nude mice. The MSCs were stimulated by the breast cancer cells to secrete CCL5, which acted in a paracrine fashion on the cancer cells to enhance metastasis. Thus, MSCs as part of the tumor microenvironment promoted metastatic spread by eliciting reversible changes in the phenotype of cancer cells. The influence of MSCs on tumor growth and metastasis was recently described by Shinagawa et  al. (2009) who examined the role of human MSCs in the

284

P.J. Mishra and D. Banerjee

Experimental model systems for Support Tumor growth studying tumor stroma interactions

Bone Marrow

CAFs Injected with cancer cell Therapeutic Potential /Regenerative medicine CAFs

TCM-Activated

MSCs

Differentiation

In DARC negative individuals Increased levels of circulating bioactive molecules such as CXCL8

Aggressive tumor growth ?

Fig. 13.1  Mesenchymal stem cells and tumor growth. TCM tumor conditioned medium, CAF carcinoma associated fibroblasts

tumor stroma using an orthotopic nude mice model of KM12SM colon cancer. Systemically injected MSCs migrated to the stroma of colon tumors, both orthotopically implanted as well as metastatic liver tumors. Orthotopically implanted KM12SM cells mixed with MSCs resulted in larger tumors than KM12SM cells alone suggesting that MSCs promoted tumor growth. Survival rate was significantly lower in the group of animals receiving admixed MSCs, and liver metastasis was seen only in this group. Tumors resulting from admixed cells had a significantly higher proliferating cell nuclear antigen labeling index, a significantly greater proportion of microvessels and a significantly lower apoptotic index. Splenic injection of KM12SM cells admixed with MSCs resulted in greater number of liver metastases than when KM12SM cells were injected alone. MSCs incorporated into the stroma of primary and metastatic tumors and expressed markers of carcinoma-associated fibroblasts (CAFs) such as a-SMA and platelet-derived growth factor receptor-b. In in vitro experiments, KM12SM cells attracted MSCs and MSCs stimulated migration and invasion of tumor cells. The results of studies carried out by Shinagawa et al. (2009) suggest that MSCs migrate and differentiate to CAFs in tumor stroma, and promote growth and metastasis of human colon cancer in a nude mouse model.

13  Bone Marrow Derived Mesenchymal Stem/Stromal Cells and Tumor Growth

285

Conclusion Solid tumors are often surrounded by a variety of cell types including myofibroblasts also known as tumor TAFs or CAFs, tumor associated macrophages (TAMs), lymphocytes, and pericytes. We have discussed how BMD MSCs can be activated in vitro to generate myofibroblasts that resemble CAFs and the advantages of this system to model tumor stroma interaction. Bone marrow derived MSCs are suggested to be a source of tumor or CAFs and influence tumor growth and metastasis. There is emerging clinical data to suggest that in some instances tumor associated myofibroblasts may indeed be of bone marrow origin. We have also taken the liberty to speculate on the possibility that activation of MSCs derived from bone marrow under certain physiological conditions may lead to aggressive tumor growth such as seen in African American individuals with breast and prostate cancers (Fig. 13.1).

References Albain KS, Unger JM, Crowley JJ, Coltman CA Jr, Hershman DL (2009) Racial disparities in cancer survival among randomized clinical trials patients of the Southwest Oncology Group. J Natl Cancer Inst 101(14):984–992 Alexe G, Dalgin GS, Scanfeld D et al (2007) High expression of lymphocyte-associated genes in node-negative HER2+ breast cancers correlates with lower recurrence rates. Can Research 67:10669–10676 Allinen M, Beroukhim R et al (2004) Molecular characterization of the tumor microenvironment in breast cancer. Cancer Cell 6(1):17–32 Amend K, Hicks D, Ambrosone, CB (2006) Breast cancer in African-American women: differences in tumor biology from European American women. Cancer Res 66:8327–8330 Balkwill F, Mantovani A (2001) Inflammation and cancer: back to Virchow? Lancet 357:539–545 Bissell MJ, Radisky D. (2001) Putting tumors in context. Nature Rev Cancer 1:46–54 Campbell I, Polyak K, Haviv I (2009) Clonal mutations in the cancer associated fibroblasts: the case against genetic coevolution. Cancer Res 69:6765–6769 Chaffer CL, Brennan JP, Slavin JL et  al (2006) Mesenchymal-to-epithelial transition facilitates bladder cancer metastasis: role of fibroblast growth factor receptor-2. Cancer Res 66:11271–11278 Chang HY, Nuyten DS, Sneddon JB et al (2005) Robustness scalability and integration of a wound response gene expression signature in predicting breast cancer survival. Proc Natl Acad Sci U S A 102(10): 3738–3743 Direkze NC, Hodivala-Dilke K, Jeffery R et al (2004) Bone marrow contribution to tumor-associated myofibroblasts and fibroblasts. Cancer Res 64:8492–8495 Dudley AC, Shih SC, Cliffe AR et  al (2008) Attenuated p53 activation in tumor-associated stromal cells accompanies decreased sensitivity to etoposide and vincristine. Br J Cancer 99:118–125 Dvorak HF (1986) Tumors: wounds that do not heal. Similarities between tumor stroma generation and wound healing. N Engl J Med 315:1650–1659 Eng C, Leone G, Orloff MS, Ostrowski MC (2009) Genomic alterations in tumor stroma. Cancer Res 69:6759–6764 Finak G, Bertos N, Pepin F et  al (2008) Stromal gene expression predicts clinical outcome in breast cancer. Nat Med 14:518–527

286

P.J. Mishra and D. Banerjee

Fox J M, Chamberlain G, Ashton BA, Middleton J (2007) Recent advances into the understanding of mesenchymal stem cell trafficking. Br J Haematol 137:491–502 Fukuma N, Akimitsu N, Hamamoto H, Kusuhara H, Sugiyama Y, Sekimizu K (2003) A role of the Duffy antigen for the maintenance of plasma chemokine concentrations. Biochem Biophys Res Commun 303:137–139 Guo X, Oshima H, Kitmura T et al (2008) Stromal fibroblasts activated by tumor cells promote angiogenesis in mouse gastric cancer. J Biol Chem 283:19864–19871 Hawsawi NM, Ghebeh H, Hendrayani SF et al (2008) Breast carcinoma-associated fibroblasts and their counterparts display neoplastic-specific changes. Cancer Res 68:2717–2725 Hu M, Polyak K (2008) Molecular characterisation of the tumour microenvironment in breast cancer. Eur J Cancer 44(18):2760–2765 Hung SC, Deng WP, Yang WK et  al (2005) Mesenchymal stem cell targeting of microscopic tumors and tumor stroma development monitored by noninvasive in  vivo positron emission tomography imaging. Clin Cancer Res 11:7749–7756 Iwamoto S, Mihara K, Dowining JR, Pui C-H, Campana D (2007) Mesenchymal cells regulate the response of acute lymphoblastic leukemia to asparaginase. J Clin Invest 117:1049–1057 Jodele S, Chantrain CF, Blavier L, Lutzko C, Crooks GM et al (2005) The contribution of bone marrow-derived cells to the tumor vasculature in neuroblastoma is matrix metalloproteinase-9 dependent. Cancer Res 65:3200–3208 Karnoub AE, Dash AB, Vo AP et al (2007) Mesenchymal stem cells within tumor stroma promote breast cancer metastasis. Nature 449:557–563 Koukourakis MI, Giatromanolaki A, Harris AL et al (2006) Comparison of metabolic pathways between cancer cells and stromal cells in colorectal carcinomas: a metabolic survival role for tumor-associated stroma. Cancer Res 66:632–637 Koyama H, Kobayashi N, Harada M, Takeoka M, Kawai Y et  al (2008) Significance of tumor-associated stroma in promotion of intratumoral lymphangiogenesis: pivotal role of a hyaluronan-rich tumor microenvironment. Am J Pathol 172:179–193 Kroemer G, Pouyssegur J (2008) Tumor cell metabolism: cancer’s Achilles’ heel. Cancer Cell 13:472–482 Lee JS, Frevert CW, Wurfel MM et al (2003) Duffy antigen facilitates movement of chemokine across the endothelium in vitro and promotes neutrophil transmigration in vitro and in vivo. J Immunol 170(10):5244–5251 Lentsch AB (2002) The Duffy antigen/receptor for chemokines (DARC) and prostate cancer. A role as clear as black and white? The FASEB Journal 16:1093–1095 Liu XH, Hadley TJ, Xu L, Peiper SC, Ray PE (1999) Up-regulation of Duffy antigen receptor expression in children with renal disease. Kidney Int 55(4):1491–1500 Mantovani A, Schioppa T, Porta C, Allavena P, Sica A (2006) Role of tumor-associated macrophages in tumor progression and invasion. Cancer Metastasis Rev 25:315–322 Martin DN, Boersma BJ et  al (2009). Differences in the tumor microenvironment between African-American and European-American breast cancer patients. PloS One 4:1–14 Mayr FB, Spiel AO, Leitner JM, Firbas C, Kliegel T, Jilma B (2007) Ethnic differences in plasma levels of interleukin-8 (IL-8) and granulocyte colony stimulating factor (G-CSF). Transl Res 149(1):10–14 Mishra PJ, Mishra PJ, Humeniuk R et al (2008) Carcinoma-associated fibroblast-like differentiation of human mesenchymal stem cells. Cancer Res 68:4331–4339 Nakamizo A, Marini F, Amano T et al (2005) Human bone marrow-derived mesenchymal stem cells in the treatment of gliomas. Cancer Res 65:3307–3318 Ogawa M, LaRue AC, Drake CJ (2006) Hematopoietic origin of fibroblasts/myofibroblasts: its pathophysiologic implications. Blood 108:2893–2896 Orimo A, Gupta PB, Sgroi DC et al (2005) Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion. Cell 121:335–348

13  Bone Marrow Derived Mesenchymal Stem/Stromal Cells and Tumor Growth

287

Patocs A, Zhang L, Xu Y et al (2007) Breast-cancer stromal cells with TP53 mutations and nodal metastases. N Engl J Med 357:2543–2551 Picinich SC, Mishra PJ, Mishra PJ, Glod J, Banerjee D (2007) The therapeutic potential of mesenchymal stem cells. Expert Opin Biol Ther 7:965–973 Pittenger MF, Mackay AM, Beck SC et al (1999) Multilineage potential of adult human mesenchymal stem cells. Science 284:143–147 Pruenster M, Mudde L, Bombosi P et  al (2009) The Duffy antigen receptor for chemokines transports chemokines and supports their promigratory activity. Nat Immunol 10:101–108 Radisky DC (2005) Epithelial-mesenchymal transition. J Cell Sci 118:4325–4326 Ries L, Eisner M, Kosary M (2002) SEER Cancer Statistics Review, 1973–1999. National Cancer Institute, Bethesda, MD Salomon AV, Thiery JP. (2003) Host microenvironment in breast cancer development Epithelial– mesenchymal transition in breast cancer development. Breast Cancer Res 5:101–106 Samudio I, Fiegl M, McQueen T et al (2008) The Warburg effect in leukemia-stroma cocultures is mediated by mitochondrial uncoupling associated with uncoupling protein-2 activation. Cancer Res 68:5198–5205 Shen H, Schuster R, Stringer KF, Waltz SE, Lentsch AB (2006) The Duffy antigen/receptor for chemokines (DARC) regulates prostate tumor growth. FASEB J 20:59–64 Shinagawa K, Kiatadai Y, Ohara E et al (2009) Mesenchymal stem cells migrate to tumor stroma and enhance tumor growth in an orthotopic nude mouse model of colon cancer. Abstract 3156. p314 Proc of 100th Annual Meeting AACR, Denver,CO, 18–22 April 2009 Sorlie T, Perou CM, Tibshirani R et  al (2001) Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A 98: 10869–10874 Spaeth EL, Dembinski JL, Kate Sasser A et al (2009) Mesenchymals stem cell transition to tumor associated fibroblasts contributes to fibrovascular network expansion and tumor progression. PloS One 4:1–11 Studeny M, Marini FC, Champlin RE et al (2002) Bone marrow-derived mesenchymal stem cells as vehicles for interferon-beta delivery into tumors. Cancer Res 62:3603–3608 Studeny M, Marini FC, Dembinski JL et al (2004) Mesenchymal stem cells: potential precursors for tumor stroma and targeted-delivery vehicles for anticancer agents. J Natl Cancer Inst 96:1593–1603 Tournamille C, Colin Y, Cartron JP, Le Van Kim C (1995) Disruption of a GATA motif in the Duffy gene promoter abolishes erythroid gene expression in Duffy-negative individuals. Nat Genet 10:224–228 Tsujino T, Seshimo I, Yamamoto H, Ngan CY, Ezumi K et  al (2007) Stromal myofibroblasts predict disease recurrence for colorectal cancer. Clin Cancer Res 13:2082–2090 Udagawa T, Puder M, Wood M, Schaefer BC, D’Amato RJ (2006) Analysis of tumor-associated stromal cells using SCID GFP transgenic mice: contribution of local and bone marrow-derived host cells. FASEB J 20:95–102 van de Vijver MJ, He YD, van’t Veer LJ et al (2002) Gene expression profiling predicts clinical outcome of breast cancer. N Engl J Med 347(25):1999–2009 Wadhwa M, Seghatchian MJ, Dilger P, Contreras M, Thorpe R (2000) Cytokine accumulation in stored red cell concentrates: effect of buffy-coat removal and leucoreduction. Transfus Sci 23(1):7–16 Wang Y, Klijn JG, Zhang Y et al (2005) Gene-expression profiles to predict distant metastasis of lymph-node-negative primary breast cancer. Lancet 365:671–679 Wang J, Ou ZL, Hou YF, Luo JM, Shen ZZ, Ding J, Shao ZM (2006) Enhanced expression of Duffy antigen receptor for chemokines by breast cancer cells attenuates growth and metastasis potential. Oncogene 25:7201–7211 Weber F, Xu Y, Zhang L, Patocs A, Shen L, Platzer P, Eng C (2007) Microenvironmental genomic alterations and clinicopathological behavior in head and neck squamous cell carcinoma. JAMA 297:187–195

288

P.J. Mishra and D. Banerjee

Worthley DL, Ruszkiewicz A, Davies R et al (2009) Human Gastrointestinal neoplasia associated myofibroblasts can develop from bone marrow derived cells following allogeneic stem cell transplantation. Stem Cells 27:1463–1468 Yazhou C, Wenlv S, Weidong Z, Licun W (2004) Clinicopathological significance of stromal myofibroblasts in invasive ductal carcinoma of the breast. Tumor Biol 25:290–295 Zhang Y, Daquinag A, Traktuev DO et al (2009) White adipose tissue cells are recruited by experimental tumors and promote cancer progression in mouse models. Cancer Res 69:5259–5265

Chapter 14

Integrin Signaling in Lymphangiogenesis Barbara Garmy-Susini

Abstract  The lymphatic vasculature is essential for the maintenance of normal fluid balance and immune response. The lymphatic vessels are also crucially involved in pathological processes such as lymphedema and tumor metastasis. In the adult, lymphangiogenesis is only involved in pathological conditions. Induction of tumoral lymphangiogenesis allows tumor cells to enter the lymphatics to induce metastasis to the lymph nodes and distant loci. Dynamic interactions between lymphatic endothelial cells (LEC) and components of their surrounding extracellular matrix are necessary for the invasion, migration, and survival of LEC during lymphangiogenesis. A major part of these interactions are mediated by integrins, a cell matrix receptors family that initiate intracellular signaling cascades in response to binding to specific extracellular matrix molecules. Understanding the molecular events that define this subset of invasive LEC will facilitate the development of new treatment strategies. Therefore, understanding the role of integrin in tumor lymphangiogenesis remains crucial. Here, we review the role of integrins as major regulator of lymphangiogenesis.

Introduction Lymphatic system collects the extravasated fluid to maintain normal tissue fluid balance. In addition, lymphatic vessels absorb and transport fat released by enterocytes and represent an important part of immune surveillance by carrying immune cells and antigens. Furthermore to these physiologic tasks, lymphatic system participates to pathological conditions such as lymphedema, inflammatory diseases, and tumor metastasis. Many studies have demonstrated the existence of proliferative peri- and intratumoral lymphatic vessels (Skobe et  al. 2001; Alitalo et  al. 2005). Additionally, tumoral lymphangiogenesis correlates with an increase of

B. Garmy-Susini (*) Unité mixte Inserm U858, Institut de Médecine Moléculaire de Rangueil, IFR 150, 1, Avenue Jean Poulhès, BP 84225 31432, Toulouse Cedex 4, France e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_14, © Springer Science+Business Media, LLC 2010

289

290

B. Garmy-Susini

metastases (Williams et al. 2003; Mumprecht and Detmar 2009) and detection of lymphangiogenic growth factors is associated with poor prognosis in many human tumors (Stacker et al. 2004; Renyi-Vamos et al. 2005; Zeng et al. 2005). Development and remodeling of lymphatic system requires complex interactions between lymphatic endothelial cells (LECs) and extarcellular matrix (ECM). The molecular mechanisms leading to LEC–ECM interactions remain crucial for the understanding of lymphatic development by contributing to cell movement, proliferation, and survival. Integrins represent the most important family of receptors mediating cell adhesion to ECM. A large body of evidences has demonstrated that integrins are not only adhesion receptors, but influenc e the biological activity of several other molecular systems within the cell. Their role in angiogenesis has also been clearly established. Here, we reviewed emerging results highlighting new roles of integrins in lymphangiogenesis.

Lyphangiogenesis Lymphatic Vasculature Lymphangiogenesis, the outgrowth of novel lymphatic vessels, plays a central role in maintaining the interstitial fluid balance via transporting extravasated tissue fluid, macromolecules and cells back into the blood circulation. Lymphatic vessels also play a crucial role in promoting the immune functions by controlling cellular and antigen trafficking. Initial lymphatics combine to form larger vessels called precollectors and collectors, which in turn lead to four major groups of lymph nodes in the axillary and inguinal regions. Present in the skin and in most internal organs, the lymphatic vasculature is composed of vessels with distinct morphological features. Similar to blood capillaries, lymphatic capillaries consist of a single layer of thin-walled, nonfenestrated LECs, but they are not covered by pericytes or smooth muscle cells, and have an absent or poorly developed basement membrane. In addition, they lack tight junctions and adherens junctions, which allow easy access for fluid, macromolecules, and cells into the vessel lumen (Leak 1976). Endothelial cells of lymphatic capillaries are oak leaf shaped and are interconnected by specialized discontinuous button-like junctions, whereas collecting lymphatic vessels downstream have continuous zipperlike junctions found also in blood vessels (Baluk et al. 2007). Overlapping endothelial cell–cell contacts (also called primary valves) in initial lymphatic vessels prevent fluid escaping back into the interstitial space (Trzewik et al. 2001; Schmid-Schonbein 2003). The recent discovery of specific markers and growth factors for lymphatic endothelium and the establishment of genetic mouse models with impairment of lymphatic function have provided novel insights into the molecular control of the lymphatic system in physiology and in embryonic development. Based upon these findings, novel therapeutic strategies are currently being developed that aim at inhibiting or promoting the formation and function of lymphatic vessels in disease.

14  Integrin Signaling in Lymphangiogenesis

291

Lymphatic Makers Research into the role of the lymphatic system in cancer metastasis has been hampered by the lack of specific markers that distinguish lymphatic vessels from blood vessels, and also by the lack of known lymphatic-specific growth factors (Alitalo et al. 2005). However, novel lymphatic-specific markers have been recently identified such as lymphatic vascular endothelial-cell hyaluronan receptor-1 (LYVE-1) (Prevo et al. 2001) a new homologue of the CD44 glycoprotein, which is a lymphspecific receptor for hyaluronan; Prox1, a homeobox transcription factor that induces lymphatic lineage-specific differentiation and is essential for the embryonic development of the lymphatic system (Wigle and Oliver 1999; Hong et  al. 2002; Petrova et al. 2002) and Podoplanin, a transmembrane glycoprotein molecule (Wicki et al. 2006) expressed in the cardinal vein and in Prox1-positive lymphatic progenitor cells in the embryos and restricted to lymphatic endothelium during later development. VEGFR-3 (also known as FLT-4) has largely been described as an major marker of lymphatics (Karkkainen et al. 2000; Kilic et al. 2007) as its expression in the adult becomes restricted to the lymphatic endothelium (Makinen et  al. 2001; Bridenbaugh 2005; Breslin et al. 2007). However, recent finding have shown that VEGFR-3 is also upregulated on vascular endothelial cells in angiogenic sprouts and is present on vessels in tumors and wounds (Petrova et  al. 2008; Tammela et al. 2008).

Induction of Lymphangiogenesis The spread of cancer cells from the primary tumor to distant organs usually first occurs via the sentinel lymph node. Tumors can induce lymphangiogenesis in both primary tumor and lymph node via release of the lymphangiogenic growth factors vascular endothelial growth factor VEGF-C or VEGF-D, leading to increased rates of metastasis to the draining sentinel lymph nodes and beyond (Mandriota et  al. 2001; Skobe et al. 2001; Stacker et al. 2001) (Fig. 14.1). Several mediators of lymphangiogenesis have been identified. Fibroblast growth factor FGF-2 promotes lymphatic vessel growth in the mouse cornea (Kubo et al. 2002; Chang et al. 2004) and also stimulates proliferation and migration of LECs by binding to the receptor FGFR-3, which is upregulated by the transcription factor Prox1 in lymphatic endothelium (Shin et  al. 2006). Hepatocyte growth factor (HGF; also known as scatter factor) also induces proliferation, migration, and tube formation of LECs and increases lymphangiogenesis in vivo (Kajiya et al. 2005). Despite the growing number of novel potential lymphangiogenic factors, the VEGF family represent the most extensive lymphangiogenic stimuli in the majority of human and experimental cancers, in particular VEGF-C and VEGF-D. VEGF-C promotes lymphangiogenesis by activating VEGF receptor (VEGFR)-2 and

292

B. Garmy-Susini

Fig. 14.1  Mechanisms regulating tumor lymphangiogenesis. (1) Tumor cells near preexisting blood vessels secrete growth factors and chemokines such as VEGF-A and FGF-2 that stimulate quiescent vascular endothelium to enter the cell cycle. (2)Tumors also secrete factors such as VEGF-C, VEGF-D, VEGF-A, FGF-2 and HGF that stimulate the growth of new lymphatic vessels in the peritumoral space. (3) These growth factors activate or upregulate expression of integrins such as avb3, avb5, a4b1, and a5b1 on blood vessels and a1b1, a2b1, a4b1, and a9b1 on lymphatic vessels. (4) These integrins then promote lymphatic endothelial cell migration, proliferation and survival, resulting in the creation of new lymphatic vessels. (5) Tumor-derived VEGF-C also promotes new lymphatic vessel growth in sentinel lymph nodes. (6) Whereas the new blood vessels induce tumor growth providing oxygen and nutriments, new lymphatic vessels represent an easier way for tumor cells to metastasize due to their vascular wall more permeable. (7) Tumor cells form metastases to proximal draining lymph nodes and then to distant organs such as lung, liver or bone

VEGFR-3 on LECs (Makinen et al. 2001). VEGF-C-deficient mice fail to develop a functional lymphatic system (Karkkainen et al. 2004), and transgenic expression of soluble VEGFR-3 results in pronounced lymphedema (Makinen et  al. 2001). Recently, VEGF-A was identified as a strong lymphangiogenic mediator. Adenoviral delivery of murine VEGF-A to the skin of mice strongly promotes lymphatic vessel growth, and transgenic mice that overexpress murine VEGF-A, specifically in the skin, show increased lymphangiogenesis during wound healing and inflammation (Nagy et al. 2002; Hong et al. 2004; Hirakawa et al. 2005). Moreover both VEGF-C- and VEGF-A-overexpressing primary tumors induce lymphangiogenesis in sentinel lymph nodes before metastasizing to these tissues (Hirakawa et al. 2007) indicates that primary tumors can prepare their future metastatic site in advance of their arrival, partly by producing lymphangiogenic factors that mediate their transport. Thus, tumors might actively modify their future distant loci to make it more suitable for their further metastatic spread. This hypothesis is further supported by recent findings that attraction of bone marrow-derived hematopoietic progenitor

14  Integrin Signaling in Lymphangiogenesis

293

cells to premetastatic sites was related to tumor-specific upregulation of fibronectin in resident fibroblasts (Kaplan et al. 2005). It will be important to determine the exact mechanisms by which lymphangiogenesis induces tumor-cell migration, via the lymphatic vasculature, to the lymph nodes and other tissues. Systemic blockade of VEGFR-3 has been recently shown to block tumor metastasis not only to lymph nodes, but also to lungs in experimental models of breast cancer (Krishnan et al. 2003; Roberts et al. 2006). Treatment with a VEGF-C inhibitor could potentially prevent the systemic spread of an early or even advanced-stage malignancy, by inhibiting lymphangiogenesis in tumors and lymph nodes. Recent reports have shown that the overexpression of VEGF-C or VEGF-D induces tumor lymphangiogenesis and promotes lymphatic metastasis in mouse tumor models (Stacker and Achen 2008). However, few clinical studies have investigated the association between the expression of VEGF-C and VEGF-D, and lymphangiogenesis and lymphatic metastasis. VEGF-D increases lymphatic vessel growth and lymphatic metastasis (Stacker et al. 2001). Recent studies revealed that VEGF-C is more significantly correlated with lymph node metastasis than VEGF-D. These findings suggest that VEGF-D is less important in lymphatic metastasis than VEGF-C, but is still necessary for metastasis (Sugiura et al. 2009).

Lymphangiogenesis and Pathology In adult organisms, lymphangiogenesis takes place only in certain pathological conditions. Abnormal function of the lymphatics is implicated in the diseases, such as lymphedema, inflammation, immune diseases, and tumor metastasis. Lymphedema is a disorder of the lymphatic vascular system characterized by impaired lymphatic return and swelling of the extremities. When the lymphatic system has been damaged during surgery or radiation treatment, its capacity to absorb excess water and cells from the interstitial space is reduced. If the transport capacity of the lymphatic system is reduced enough so that it cannot handle this increase in lymphatic load, an insufficiency of the lymphatic system may occur. Lymphedema can be an unfortunate side effect of cancer treatment. It is a chronic condition that, if ignored, can lead to disfigurement, immobilization, and severe infections. Without treatment, the swelling may continue to increase. Inflammation is thought to contribute to the development and progression of various cancers, including lung (Ardies 2003), breast (Van der Auwera et al. 2004), gastrointestinal (Jaiswal et  al. 2001; Biarc et  al. 2004; Brower 2005), ovarian (Altinoz and Korkmaz 2004), prostate (Wang et  al. 2004), skin (Hussein and Ahmed 2005), and liver cancers (Bartsch and Nair 2004). Inflammatory breast cancer exhibits increased angiogenesis and lymphangiogenesis and has a higher metastatic potential than noninflammatory breast cancer (Angelo and Kurzrock 2007). Blocking lymphangiogenesis in chronic inflammatory diseases may become an important means of ameliorating the severity of some of these pathologies.

294

B. Garmy-Susini

The extent of lymph node metastasis is a major determinant for the staging and the prognosis of most human malignancies. Although the clinical significance of lymph node involvement is well documented, molecular mechanisms that promote tumor spread into the lymphatic or blood vascular systems and widespread dissemination are not well understood. Recent studies have shown a large body of evidence that newly visualized lymphatics facilitates formation of metastases. High tumor interstitial fluid pressure is thought to promote tumor cell entry into lymphatic vessels that have lower fluid pressure (Jain 1994; Padera et  al. 2002). Intratumoral lymphatic vessel growth often correlates with metastasis of human melanoma, breast, or head and neck cancers (Maula et al. 2003; Choi et al. 2005; Dadras et al. 2005), where tumor cells can be observed within lymphatic vessels, demonstrating that lymphatic vessel growth is important for tumor spread.

Integrins Many studies have implicated a number of endothelial cell integrins in the regulation of cell growth, survival and migration during angiogenesis; nevertheless, their role in lymphangiogenesis remains unclear. The role of cell adhesion molecules and extracellular matrix proteins (ECM) in various pathological processes including angiogenesis, thrombosis, apoptosis, cell migration, and proliferation are well documented. These processes can lead to both acute and chronic disease states such as ocular diseases, metastasis, myocardial infarction, stroke, osteoporosis, a wide range of inflammatory diseases, vascular remodeling, and neurodegenerative disorders. As several integrin-targeted therapeutic agents are in clinical trials for cancer therapy, future clinical studies will likely determine whether integrin inhibitors will be best used against select tumors, such as those in which tumor cells themselves express the targeted integrin. As integrins are clearly a family of critical and fundamental regulators of angiogenesis and lymphangiogenesis, the future of integrin antagonists in cancer therapy is promising.

Integrin Expression and Function Integrin induce several physiological processes including cell activation, migration, proliferation, differentiation, and many other processes require direct contact between cells or ECM proteins. Cell–cell and cell–matrix interactions are mediated through several different families of CAM including the selectins, the integrins, the cadherins, and the immunoglobulins. The integrin family is an extensive group of structurally related receptors for ECM proteins and immunoglobulin superfamily molecules. Integrins are heterodimeric membrane glycoproteins comprised of noncovalently associated a and b subunits that promote cell attachment and migration on the surrounding extracellular matrix. Eighteen a and eight b subunits can

14  Integrin Signaling in Lymphangiogenesis

295

Fig. 14.2  Integrin family of adhesion receptors. Each integrin receptor heterodimer binds a specific set of endogenous ligands, which may include ligands in the ECM, soluble ligands, and ligands on other cell surfaces. Integrins are divalent cation-dependent heterodimeric membrane glycoproteins comprised of noncovalently associated a and b subunits. Each integrin subunit consists of an extracellular domain, a single transmembrane region, and a short (approximately 30–40 amino acids) cytoplasmic region. Upon ligand binding, a series of intracellular signaling events is initiated. These pathways are associated with enhanced cell proliferation, migration, and survival. The role of integrins lymphangiogenesis is at an early stage of understanding. Nevertheless, recent studies have shown an important role integrins a1b1, a2b1, a4b1, and a9b1 in promoting lymphatic endothelial cell migration and survival (green)

associate to form 24 unique integrin heterodimers (Fig. 14.2). Each integrin subunit consists of an extracellular domain, a single transmembrane region, and a short (approximately 30–40 amino acids) cytoplasmic region (Hynes 2002). The main ligands for integrins in the extracellular space are extracellular matrix proteins, such as laminin and collagen, as well as cellular counter-receptors. Integrins are linked to the cytoskeleton through their cytoplasmic domains. Integrins modulate the cytoskeleton via various submembrane adaptor proteins and kinases (Zamir and Geiger 2001). They transduce signals across the plasma membrane in both directions. Most of the time, they are expressed on the cell surface in an inactive state. The activation of integrins leads to conformational changes in the extracellular domain allowing ligand binding. Integrin binding to its ligands requires its activation by inside-out signals. Conversely, integrin ligation triggers outside–in signals that regulate different aspects of cell behavior, including cell survival, control of transcription, cell proliferation, cell motility, and cytoskeletal organization (Hynes 2002). Integrin bidirectional signals are thought to be transmitted via protein–protein interactions but how the integrin cytoplasmic tail orchestrates the transmission of bidirectional signals is not well understood. Specific binding of the cytoskeletal protein to integrin b subunit cytoplasmic tails leads to the conformational rearrangements of integrin extracellular domains that increase their affinity (Tadokoro et al. 2003). Studies support a model in which intracellular protein interactions with the integrin b

296

B. Garmy-Susini

subunit tail play key roles for both inside–out and outside–in signal transduction (Arias-Salgado et al. 2003; Tadokoro et al. 2003; Arias-Salgado et al. 2005). Outside–in signal: During migration, cells are constantly making and breaking integrin contacts, activating intra-cellular signaling pathways. These signals are associated with growth factors signals and regulate cell invasion into their microenvironment (Hood and Cheresh 2002). They are mediated by intracellular signaling molecules such as FAK, SHC, and RHO family of small GTPases. Inside–out signal: Integrins not only send signals to the cell in response to the extracellular environment, but also respond to intracellular stimuli to modify the way in which they interact with extracellular matrix proteins (Hood and Cheresh 2002). This process modulates the affinity and the avidity of integrins for their ligands.

Role of Integrins in Promoting Endothelial Cells Migration, Proliferation, and Survival During tumorigenesis, a switch of integrin expression can be observed, in as much as growth-promoting and growth-attenuating integrins are up- and downregulated, respectively. ECM-ligand binding to an integrin initiates signals, which eradiating from the integrins are transmitted via different interconnecting pathways and elicit various cell functions, such as morphological changes, adhesion, migration, and gene activation. Any of these functions takes part in the metastatic cascade of tumor progression, such as epithelial-to-mesenchymal transition of carcinoma cells, tumor cell contact with the basement membrane, invasion into neighboring tissues as well as production and activation of ECM-degrading MMPs. Besides their direct involvement in tumor progression as cell surface molecules on tumor cells, integrins on endothelial surrounding a tumor can also determine various cancer characteristics, such as tumor-induced neoangiogenesis, lymphangiogenesis, and immune resistance. Therefore, integrins are relevant pharmacological targets in tumor biology.

Ligand Specificity of Integrins Each unique a-chain combines with a b-chain to form heterodimers with unique selectivity for ECM, cell surface molecules, and plasma proteins (Loftus et  al. 1994; Plow et  al. 2000; van der Flier and Sonnenberg 2001). Integrins binds to their ligand to a divalent-cation dependent fashion (Smith et al. 1994; Mould et al. 1995). While some integrins selectively recognize primarily a single ECM protein ligand (e.g., integrin a5b1 recognizes primarily fibronectin), others can bind several ligands (e.g., integrin avb3 binds vitronectin, fibronectin, collagen, and other matrix proteins).

14  Integrin Signaling in Lymphangiogenesis

297

Many integrins, including avb3, a5b1, aIIbb3, avb6, and a3b1, recognize the tripeptide Arg-Gly-Asp (RGD) in their ligands (Plow et al. 2000). Sequences flanking the RGD peptide are also important for integrin selectivity (Haas and Plow 1994). Other integrins recognize alternative short peptide sequences; for example, integrin a4b1 recognizes Glu-Ile-Leu-Asp-Val (EILDV) and Arg-Glu-Asp-Val (REDV) in the alternatively spliced fibronectin domain known as IIICS (Komoriya et al. 1991). Some integrins, such as a4b1, can also bind cell surface receptors, such as Vascular Cell Adhesion Molecule-1 (VCAM-1), to promote cell–cell adhesion (Jin and Varner 2004) in addition to alternatively spliced CS-1 fibronectin. Spurred by the recent success to generate pharmaceutical mimetics of RGDdependent integrins and by the integrin’s easy accessibility on the cell surface, the hope is rising that also RGD-independent integrins, such as the collagen- and laminin-binding integrins, can be pharmacologically manipulated to fight integrindependent functions of cancer cells, which are necessary and at least partially specific for their proliferation and progression.

Integrin Signaling Integrins are essential for cell migration and invasion. They mediate adhesion to extracellular matrix and regulate intracellular signaling pathways that control cytoskeletal organization. These pathways involve phosphorylation of focal adhesion kinase (FAK), recruitment of adaptator proteins, activation of small GTPases, and downstream effector molecules (Fig. 14.3). Integrin ligation promotes integrin clustering and subsequent integrin-mediated intracellular signal transduction. Unlike growth factor receptors, integrins have no intrinsic enzymatic or kinase activities, but activate complex signaling pathways by coclustering with kinases and adaptor proteins in focal adhesion complexes. A number of signaling pathways are activated by integrins and many of these are found within focal adhesion complexes. Focal adhesion complexes are comprised of integrins, protein kinases – such as FAK and Src – adaptor proteins such as Shc, signaling intermediates such as Rho family GTPases, actin-binding cytoskeletal proteins such as talin, a-actinin, paxillin, tensin, and vinculin and other signaling proteins. Integrin signaling promotes cell migration, proliferation, and survival. Loss of integrin ligation inhibits these events and unligated integrins can actively initiate apoptosis, even without loss of cell attachment. This form of death is stress response- and death-receptor-independent, but caspase 8-dependent, and has been called “integrin mediated death” (Stupack et al. 2006). Fak FAK is a cytoplasmic protein kinase that colocalizes with integrins at structures called focal adhesion. Integrins binding of ECM ligands induces integrin clustering

298

B. Garmy-Susini

Fig. 14.3  Each integrin subunit consists of an extracellular domain, a single transmembrane region, and a short (approximately 30–40 amino acids) cytoplasmic cytoplasmic tails and no intrinsic enzymatic or kinase activities. Upon ligand binding, a series of intracellular signaling events is initiated. These pathways are associated with enhanced cell proliferation, migration, and survival. To integrate signals and activate intracellular signaling pathways, integrins cocluster with serine, threonine and tyrosine kinases, phosphatases, and adaptor proteins in focal adhesions. Focal adhesion complexes are comprised of integrins, protein kinases such as focal adhesion kinase (FAK), Src and many other kinases, adaptor proteins such as Shc, signaling intermediates such as PI-3-kinase, Rho and Rac GTPases and actin-binding cytoskeletal proteins such as talin, a-actinin, paxillin, tensin, and vinculin. Integrin signaling promotes cell migration by providing traction along the extracellular matrix and by promoting actin remodeling through the Rho family small GTPases. This actin remodeling leads to cytoplasmic flow in the direction of cell migration and cell body retraction at the trailing end of the cells. Individual components of integrin-mediated signaling cascades, such as FAK, Shc and Raf, play essential roles in angiogenesis. For example, FAK is a mediator of signal transduction by integrins and growth factor receptors in endothelial cells. In addition, Shc, an important adaptor protein that potentiates MAP kinase pathway signaling is activated by both integrins and growth factor receptors and plays critical roles in early vascular development. Like Shc, Raf-1 is an integral component of the MAP kinase signaling pathway. This signaling intermediate is activated by integrins and is critical for vascular morphogenesis. Thus, integrin mediated signaling likely plays important roles in vascular development in the adulthood

and FAK activation (Schlaepfer et al. 1999). Numerous studies have linked FAKmediated signaling pathways to cancer and a variety of other biological and disease processes. Integrin-mediated cell adhesion is the major upstream activator of FAK, and increased activation and tyrosine phosphorylation of FAK have been observed in essentially all adherent cells analyzed so far. Once it is activated, FAK undergoes

14  Integrin Signaling in Lymphangiogenesis

299

autophosphorylated at Y397, which creates a binding site for Src via the SH2 domain (Schaller and Parsons 1994; Xing et al. 1994; Cary et al. 1999). FAK also binds p85 subunit of PI3K through autophosphorylated Y397 (Reiske et al. 2000). Phospholipid production stimulated by FAK association and activation of PI3K can activate Akt kinase, which inhibits apoptosis by regulating various cell death machinery proteins (Luo et al. 2003; Hennessy et al. 2005). Several FAK signaling pathways have been shown to play an important role in cell proliferation. The formation of FAK/Src complex allows Src to phosphorylate Y925 on FAK to mediate its interaction with Grb2 leading to the activation of RAS-Erk signaling pathway (Schlaepfer et al. 1994). FAK interaction with another adaptor molecule SHC also contributes to the activation of this pathway (Schlaepfer and Hunter 1996). Numerous studies have shown that autophosphorylation of FAK at Y397 and its association with Src at the site is essential for FAK’s ability to promote cell migration, as mutation of Y397 to F both disrupted FAK association with Src and its stimulation of cell migration in many cell types including FAK-/- cells (Cary et al. 1996; Cary et  al. 1998; Owen et  al. 1999). Inhibition of cell spreading by blocking FAK functions with FRNK (C-terminal FAK fragment functioning as a dominant negative FAK) could be partially rescued by overexpression of Src (Richardson and Parsons 1996). Taken together, these studies demonstrate the crucial role of FAK in cell migration, an essential process during lymphangiogenesis. Shc The adapter protein Shc is a prototype adapter protein that has been quite useful in the understanding of the function of adapter proteins in cellular signaling. The importance of Shc in vivo has been demonstrated by the knockout of the shcA gene, which results in embryonic lethality at day 11.5 (Lai and Pawson 2000) due to defects in blood vessels formation and cardiovascular development (Lai and Pawson 2000). Although a role for Shc in activation of the Ras/MAPK pathway and its role in mitogenic signaling has been better described (Bonfini et al. 1996), what other roles it may play in signaling and why this protein is so often recruited/targeted for phosphorylation by many different types of receptors is not clear. SHC is recruited to activated tyrosine kinases in response to ligation of integrins a1b1, a6b4, a5b1, and avb3 (Wary et al. 1996, 1998). Studies from various laboratories in different model systems have shown an important role for Shc in leading to Ras activation. Shc is tyrosine-phosphorylated by receptor activation and it subsequently interacts with Grb2. Grb2, binds to Ras guanine nucleotide exchange factor, Sos. The Shc/Grb2/Sos complex gets localized to the membrane through the interaction of Shc with the phosphorylated receptor. In the case of the integrin family of receptors and G-protein-coupled receptors, the precise mechanism by which the Shc/Grb2/Sos complex is localized to the membrane is not clear. Sos has been found preferentially in complexes that also contain Shc (Pronk et al. 1994). Shc may also influence the extent of Ras activation. A number of studies using

300

B. Garmy-Susini

dominant negative Shc proteins and mice lacking Shc expression have definitively established a role for Shc in MAPK activation (Salcini et  al. 1994; Gotoh et  al. 1997; Pratt et al. 1999; Lai and Pawson 2000). Together, studies identify Shc as a critical regulator of integrin signaling pathways. Rho Family of Small GTPases Integrin engagement and subsequent clustering of these receptors in focal adhesions leads to the generation of intracellular macromolecular complexes (Miyamoto et al. 1995). Numerous proteins present at the cytoplasmic face of focal adhesions are considered to be structural components of focal adhesions, including cytoskeletal proteins such as vinculin and talin (Jockusch and Rudiger 1996). In addition, numerous “signaling” proteins with enzymatic activity (e.g., kinases and GTPases) are also components of focal adhesions. The Rho family of GTPases appears poised to contribute to these integrin-mediated signals, in particular, signals that control cytoskeletal organization involved in changes in cell morphology. Rho family members such as Cdc42, Rac1, and RhoA are part of the Ras superfamily of proteins that cycle between active and an active state. Activated RhoA is capable of stimulating microfilament bundling in serum-starved cells that are already adherent (Ridley and Hall 1992), similar to the response of cells to plating on an ECM-coated surface. Rho is also essential for the formation of focal complexes (Hotchin and Hall 1995). The Rho family member Rac controls growth factor-stimulated membrane ruffling and formation of lamellipodia (Ridley and Hall 1992). Finally, Cdc42 activation triggers the extension of filopodia (Kozma et al. 1995; Nobes and Hall 1995). Studies such as these have defined how soluble extracellular factors induce the assembly of focal adhesions and stress fibers in serum-starved adherent Swiss 3T3 fibroblasts through activation of the Rho family of GTPases (Ridley 1996; Tapon and Hall 1997). Rho family GTPases are therefore necessary to induce the signaling and cellular responses that are required to reorganize the actin cytoskeleton into an invasive and migratory phenotype. Talin Talin is a major cytoskeletal protein that colocalizes with and binds to integrins, and to actin and actin-binding proteins such as vinculin (Calderwood et al. 1999, 2002; Tadokoro et  al. 2003; Rose et  al. 2007). Talin interacts directly with the b-chain cytoplasmic domain of integrins. Knockdown of talin expression in CHO (Chinesehamster ovary) cells inhibits the activation of both b1 and b3 integrins without altering integrin expression, and this cannot be compensated for by the expression of activating molecules such as activated R-Ras or the CD98 heavy chain (Tadokoro et al. 2003). Talin colocalizes with activated integrins, and overexpression of talin’s N-terminus activates integrins (Calderwood et al. 1999). Talin represents therefore a potential key regulator of integrins signaling during lymphangiogenesis.

14  Integrin Signaling in Lymphangiogenesis

301

Vinculin Vinculin is a cytoskeletal protein associated with the cytoplasmic part of matrix adherens-type junctions. Vinculin is one of several interacting proteins involved in anchoring actin to the membrane (Weller et al. 1990). Vinculin also plays a pivotal role in cell adhesion and migration by providing the link between the actin cytoskeleton and the transmembrane receptors, integrin and cadherin. Additionaly, modulation of b1-integrin association with vinculin in human coronary artery endothelial cells (HCAEC) alter endothelial wound closure under shear stress (Albuquerque and Flozak 2003) demonstrating its role in integrin-mediated endothelial migration.

Paxillin Paxillin connect the actin cytoskeleton to the extracellular matrix within focal adhesions. Studies have shown that mutant of paxillin formed focal adhesions and exhibited limited movement associated with cell migration and wound-healing assays (Huang et al. 2003). Paxillin has also been shown to bind specifically the cytoplasmic tail of alpha4 chain to induce cell migration (Liu et al. 2000; Liu and Schnellmann 2003). This a4 binding protein suggests a potential role of the integrin alpha chain in regulating activation and signaling pathways.

Intergrins and Lymphangiogenesis The role of integrins in angiogenesis has been previously indicated in tumor angiogenesis (Hynes et al. 2002), but their role in the remodeling of lymphatic vessels is at an early stage of understanding.

a 9b1 Recent studies have shown that integrin a9b1 is expressed on quiescent LEC. Integrin a9b1 binds tenascin and oxteopontin (Kanayama et  al. 2009) and is required for development of the fully functional lymphatic system because mice deficient in a9b1 integrin die 6–12 days after birth due to chylothorax, an accumulation of lymph in the pleural cavity (Huang et al. 2000). Integrin a9b1 participates to the lymphatic valves morphogenesis as it has been recently demonstrated an upregulation of integrin a9b1 on LEC and deposition of its ligand fibronectin in the extracellular matrix during embryonic development (Bazigou et al. 2009). Integrin a9b1 plays a role in growth factor mediated lymphangiogenesis as Prox-1, a LEC selective transcription factor coordinately upregulates integrin a9b1 and VEGFR-3

302

B. Garmy-Susini

expression and endothelial cell motility in vivo (Mishima et al. 2007). Furthermore, studies have shown that this integrin promotes VEGF-C- and -D-stimulated cell migration by directly binding these growth factors (Vlahakis et al. 2005). Importantly, antagonism of a9b1 suppresses VEGF-C induced motility. Taken together, these studies indicate that a9b1 plays unique yet critical roles in lymphangiogenesis.

a 1b1 and a2b1 Other studies have shown that integrins a1b1 and a2b1 are expressed on lymphatic endothelium in healing wounds in response to VEGF-A. Inhibition of these integrins blocked lymphangiogenesis in these wounds (Hong et al. 2004). These integrins bind laminin and collagen and also play roles in regulating angiogenesis. Studies in cultured LECs revealed that VEGF-A induced expression of the alpha1 and alpha2 integrins, which promoted their in vitro tube formation and their haptotactic migration toward type I collagen. It has also been shown that systemic blockade of the alpha1 and alpha2 integrins inhibits VEGF-A-driven lymphangiogenesis in vivo (Hong et al. 2004).

a  5b1 Integrin a5b1 is receptor for fibronectin. The role of integrins a5b1 in tumor angiogenesis has been clearly demonstrated (Yang et  al. 1993; Francis et  al. 2002) its their role in tumor lymphangiogenesis remains unclear. Integrins a5b1 and av are expressed by a subpopulation of lymphatic vessels in the inflamed cornea and small molecule antagonists of this integrin inhibited inflammatory lymphangiogenesis (Dietrich et  al. 2007). Based on these studies, the role of integrin a5b1 during lymphangiogenesis seems to be restricted to inflammatory diseases. Recently, integrin a5b1 has been shown to be upregulated on lymphatic sprouts in a mice model of airways inflammation (Okazaki et  al. 2009) suggesting a role in lymphatic growth. The av integrins appear to play little or no role in lymphangiogenesis and integrin a5b1 appears to play no role in tumor lymphangiogenesis due to the lack of expression on lymphangiogenic LEC during tumor progression (Garmy-Susini et al. 2007; Avraamides et al. 2008; Garmy-Susini and Varner 2008).

a 4b1 Integrin a4b1 not only allows cells to interact with the ECM after binding a spliced variant of fibronectin (CS-1 fibronectin), but also participates to cell–cell adhesion after binding a cell-surface molecule VCAM-1, a member of the immunoglobulin

14  Integrin Signaling in Lymphangiogenesis

303

superfamily. Integrin a4b1 promotes angiogenesis after binding its ligand VCAM-1 on mural cells during blood vessels formation (Garmy-Susini et al. 2005). Integrin a4b1 regulate embryonic development as loss of its gene causes embryonic lethality by E11.5-E12.5 due to vascular defects (Kwee et  al. 1995; Yang et  al. 1995; Sengbusch et al. 2002). Integrin a4b1 is also highly expressed on tumor lymphatic endothelium (Garmy-Susini et al. 2007) and antagonists of this integrin can block lymphangiogenesis and tumor metastasis.

Conclusion Most of blood and lymphatic vessels remain quiescent in the adulthood and are activated during pathological conditions such as wound healing, inflammatory disease, or tumor development. Several integrins appear to play important roles in regulating lymphangiogenesis. Their profile during lymphangiogenesis is distinct from those regulating angiogenesis. Integrin expressed during lymphangiogenesis promote LEC migration, proliferation and survival, providing new ways for tumor cells to escape and metastasize to distant loci. They represent potentially relevant targets for antilymphangiogenic therapy. Many studies exhibit the evidence that drug therapies have to be considered in connection with the cell microenvironment that controls lymphangiogenesis. Targeting integrins for antilymphangiogenic therapy remains crucial, as they are expressed on activated vessels but not on quiescent vessels and therefore accessible by drugs only during lymphangiogenesis. Antagonists of these integrins may be useful in preventing tumor metastasis by blocking lymphangiogenesis. Moreover, agents targeting integrins receptors are now in clinical development for treating solid tumors. Due to the lack of side effect associated with integrins inhibitors the challenge for the future will be to optimize their use in combination with other treatment to fully exploit their therapeutic potential in lymphangiogenesis inhibition. However, the molecular events underlying the regulatory function of integrin receptors are not well understood and targeting of integrin function and signaling may be an alternative strategy to extracellular integrin antagonists for the therapeutic inhibition of tumor lymphangiogenesis.

References Albuquerque ML, Flozak AS (2003) Lamellipodial motility in wounded endothelial cells exposed to physiologic flow is associated with different patterns of beta1-integrin and vinculin localization. J Cell Physiol 195, 50–60. Alitalo K, Tammela T, Petrova TV (2005) Lymphangiogenesis in development and human disease. Nature 438, 946–953. Altinoz MA, Korkmaz R (2004) NF-kappaB, macrophage migration inhibitory factor and cyclooxygenase-inhibitions as likely mechanisms behind the acetaminophen- and NSAIDprevention of the ovarian cancer. Neoplasma 51, 239–247.

304

B. Garmy-Susini

Angelo LS, Kurzrock R (2007) Vascular endothelial growth factor and its relationship to inflammatory mediators. Clin Cancer Res 13, 2825–2830. Ardies CM (2003) Inflammation as cause for scar cancers of the lung. Integr Cancer Ther 2, 238–246. Arias-Salgado EG, Lizano S, Sarkar S, Brugge JS, Ginsberg MH, Shattil SJ (2003) Src kinase activation by direct interaction with the integrin beta cytoplasmic domain. Proc Natl Acad Sci U S A 100, 13298–13302. Arias-Salgado EG, Lizano S, Shattil SJ, Ginsberg MH (2005) Specification of the direction of adhesive signaling by the integrin beta cytoplasmic domain. J Biol Chem 280, 29699–29707. Avraamides CJ, Garmy-Susini B, Varner JA (2008) Integrins in angiogenesis and lymphangiogenesis. Nat Rev Cancer 8, 604–617. Baluk P, Fuxe J, Hashizume H, Romano T, Lashnits E, Butz S, Vestweber D, Corada M, Molendini C, Dejana E, McDonald DM (2007) Functionally specialized junctions between endothelial cells of lymphatic vessels. J Exp Med 204, 2349–2362. Bartsch H, Nair J (2004) Oxidative stress and lipid peroxidation-derived DNA-lesions in inflammation driven carcinogenesis. Cancer Detect Prev 28, 385–391. Bazigou E, Xie S, Chen C, Weston A, Miura N, Sorokin L, Adams R, Muro AF, Sheppard D, Makinen T (2009) Integrin-alpha9 is required for fibronectin matrix assembly during lymphatic valve morphogenesis. Dev Cell 17, 175–186. Biarc J, Nguyen IS, Pini A, Gosse F, Richert S, Thierse D, Van Dorsselaer A, Leize-Wagner E, Raul F, Klein JP, Scholler-Guinard M (2004) Carcinogenic properties of proteins with proinflammatory activity from Streptococcus infantarius (formerly S.bovis). Carcinogenesis 25, 1477–1484. Bonfini L, Migliaccio E, Pelicci G, Lanfrancone L, Pelicci PG (1996) Not all Shc’s roads lead to Ras. Trends Biochem Sci 21, 257–261. Breslin JW, Gaudreault N, Watson KD, Reynoso R, Yuan SY, Wu MH (2007) Vascular endothelial growth factor-C stimulates the lymphatic pump by a VEGF receptor-3-dependent mechanism. Am J Physiol Heart Circ Physiol 293, H709–718. Bridenbaugh E (2005) Literature watch. Complete and specific inhibition of adult lymphatic regeneration by a novel VEGFR-3 neutralizing antibody. Lymphat Res Biol 3, 87–88. Brower V (2005) Researchers attempting to define role of cytokines in cancer risk. J Natl Cancer Inst 97, 1175–1177. Calderwood DA, Yan B, de Pereda JM, Alvarez BG, Fujioka Y, Liddington RC, Ginsberg MH (2002) The phosphotyrosine binding-like domain of talin activates integrins. J Biol Chem 277, 21749–21758. Calderwood DA, Zent R, Grant R, Rees DJ, Hynes RO, Ginsberg MH (1999) The Talin head domain binds to integrin beta subunit cytoplasmic tails and regulates integrin activation. J Biol Chem 274, 28071–28074. Cary LA, Chang JF, Guan JL (1996) Stimulation of cell migration by overexpression of focal adhesion kinase and its association with Src and Fyn. J Cell Sci 109 (Pt 7), 1787–1794. Cary LA, Han DC, Guan JL (1999) Integrin-mediated signal transduction pathways. Histol Histopathol 14, 1001–1009. Cary LA, Han DC, Polte TR, Hanks SK, Guan JL (1998) Identification of p130Cas as a mediator of focal adhesion kinase-promoted cell migration. J Cell Biol 140, 211–221. Chang LK, Garcia-Cardena G, Farnebo F, Fannon M, Chen EJ, Butterfield C, Moses MA, Mulligan RC, Folkman J, Kaipainen A (2004) Dose-dependent response of FGF-2 for lymphangiogenesis. Proc Natl Acad Sci U S A 101, 11658–11663. Choi WW, Lewis MM, Lawson D, Yin-Goen Q, Birdsong GG, Cotsonis GA, Cohen C, Young AN (2005) Angiogenic and lymphangiogenic microvessel density in breast carcinoma: correlation with clinicopathologic parameters and VEGF-family gene expression. Mod Pathol 18, 143–152. Dadras SS, Lange-Asschenfeldt B, Velasco P, Nguyen L, Vora A, Muzikansky A, Jahnke K, Hauschild A, Hirakawa S, Mihm MC, Detmar M (2005) Tumor lymphangiogenesis predicts melanoma metastasis to sentinel lymph nodes. Mod Pathol 18, 1232–1242.

14  Integrin Signaling in Lymphangiogenesis

305

Dietrich T, Onderka J, Bock F, Kruse FE, Vossmeyer D, Stragies R, Zahn G, Cursiefen C (2007) Inhibition of inflammatory lymphangiogenesis by integrin alpha5 blockade. Am J Pathol 171, 361–372. Francis SE, Goh KL, Hodivala-Dilke K, Bader BL, Stark M, Davidson D, Hynes RO (2002) Central roles of alpha5beta1 integrin and fibronectin in vascular development in mouse embryos and embryoid bodies. Arterioscler Thromb Vasc Biol 22, 927–933. Garmy-Susini B, Jin H, Zhu Y, Sung RJ, Hwang R, Varner J (2005) Integrin alpha4beta1-VCAM1-mediated adhesion between endothelial and mural cells is required for blood vessel maturation. J Clin Invest 115, 1542–1551. Garmy-Susini B, Makale M, Fuster M, Varner JA (2007) Methods to study lymphatic vessel integrins. Methods Enzymol 426, 415–438. Garmy-Susini B, Varner JA (2008) Roles of integrins in tumor angiogenesis and lymphangiogenesis. Lymphat Res Biol 6, 155–163. Gotoh N, Toyoda M, Shibuya M (1997) Tyrosine phosphorylation sites at amino acids 239 and 240 of Shc are involved in epidermal growth factor-induced mitogenic signaling that is distinct from Ras/mitogen-activated protein kinase activation. Mol Cell Biol 17, 1824–1831. Haas TA, Plow EF (1994) Integrin-ligand interactions: a year in review. Curr Opin Cell Biol 6, 656–662. Hennessy BT, Smith DL, Ram PT, Lu Y, Mills GB (2005) Exploiting the PI3K/AKT pathway for cancer drug discovery. Nat Rev Drug Discov 4, 988–1004. Hirakawa S, Brown LF, Kodama S, Paavonen K, Alitalo K, Detmar M (2007) VEGF-C-induced lymphangiogenesis in sentinel lymph nodes promotes tumor metastasis to distant sites. Blood 109, 1010–1017. Hirakawa S, Kodama S, Kunstfeld R, Kajiya K, Brown LF, Detmar M (2005) VEGF-A induces tumor and sentinel lymph node lymphangiogenesis and promotes lymphatic metastasis. J Exp Med 201, 1089–1099. Hong YK, Harvey N, Noh YH, Schacht V, Hirakawa S, Detmar M, Oliver G (2002) Prox1 is a master control gene in the program specifying lymphatic endothelial cell fate. Dev Dyn 225, 351–357. Hong YK, Lange-Asschenfeldt B, Velasco P, Hirakawa S, Kunstfeld R, Brown LF, Bohlen P, Senger DR, Detmar M (2004) VEGF-A promotes tissue repair-associated lymphatic vessel formation via VEGFR-2 and the alpha1beta1 and alpha2beta1 integrins. Faseb J 18, 1111–1113. Hood JD, Cheresh DA (2002) Role of integrins in cell invasion and migration. Nat Rev Cancer 2, 91–100. Hotchin NA, Hall A (1995) The assembly of integrin adhesion complexes requires both extracellular matrix and intracellular rho/rac GTPases. J Cell Biol 131, 1857–1865. Huang C, Rajfur Z, Borchers C, Schaller MD, Jacobson K (2003) JNK phosphorylates paxillin and regulates cell migration. Nature 424, 219–223. Huang XZ, Wu JF, Ferrando R, Lee JH, Wang YL, Farese RV, Jr., Sheppard D (2000) Fatal bilateral chylothorax in mice lacking the integrin alpha9beta1. Mol Cell Biol 20, 5208–5215. Hussein MR, Ahmed RA (2005) Analysis of the mononuclear inflammatory cell infiltrate in the non-tumorigenic, pre-tumorigenic and tumorigenic keratinocytic hyperproliferative lesions of the skin. Cancer Biol Ther 4, 819–821. Hynes RO (2002) Integrins: bidirectional, allosteric signaling machines. Cell 110, 673–687. Hynes RO, Lively JC, McCarty JH, Taverna D, Francis SE, Hodivala-Dilke K, Xiao Q (2002) The diverse roles of integrins and their ligands in angiogenesis. Cold Spring Harb Symp Quant Biol 67, 143–153. Jain RK (1994) Barriers to drug delivery in solid tumors. Sci Am 271, 58–65. Jaiswal M, LaRusso NF, Gores GJ (2001) Nitric oxide in gastrointestinal epithelial cell carcinogenesis: linking inflammation to oncogenesis. Am J Physiol Gastrointest Liver Physiol 281, G626–634. Jin H, Varner J (2004) Integrins: roles in cancer development and as treatment targets. Br J Cancer 90, 561–565.

306

B. Garmy-Susini

Jockusch BM, Rudiger M (1996) Crosstalk between cell adhesion molecules: vinculin as a paradigm for regulation by conformation. Trends Cell Biol 6, 311–315. Kajiya K, Hirakawa S, Ma B, Drinnenberg I, Detmar M (2005) Hepatocyte growth factor promotes lymphatic vessel formation and function. Embo J 24, 2885–2895. Kanayama M, Kurotaki D, Morimoto J, Asano T, Matsui Y, Nakayama Y, Saito Y, Ito K, Kimura C, Iwasaki N, et al. (2009) Alpha9 integrin and its ligands constitute critical joint microenvironments for development of autoimmune arthritis. J Immunol 182, 8015–8025. Kaplan RN, Riba RD, Zacharoulis S, Bramley AH, Vincent L, Costa C, MacDonald DD, Jin DK, Shido K, Kerns SA, et al. (2005) VEGFR1-positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche. Nature 438, 820–827. Karkkainen MJ, Ferrell RE, Lawrence EC, Kimak MA, Levinson KL, McTigue MA, Alitalo K, Finegold DN (2000) Missense mutations interfere with VEGFR-3 signalling in primary lymphoedema. Nat Genet 25, 153–159. Karkkainen MJ, Haiko P, Sainio K, Partanen J, Taipale J, Petrova TV, Jeltsch M, Jackson DG, Talikka M, Rauvala H, et al. (2004) Vascular endothelial growth factor C is required for sprouting of the first lymphatic vessels from embryonic veins. Nat Immunol 5, 74–80. Kilic N, Oliveira-Ferrer L, Neshat-Vahid S, Irmak S, Obst-Pernberg K, Wurmbach JH, Loges S, Kilic E, Weil J, Lauke H, et al. (2007) Lymphatic reprogramming of microvascular endothelial cells by CEA-related cell adhesion molecule-1 via interaction with VEGFR-3 and Prox1. Blood 110, 4223–4233. Komoriya A, Green LJ, Mervic M, Yamada SS, Yamada KM, Humphries MJ (1991) The minimal essential sequence for a major cell type-specific adhesion site (CS1) within the alternatively spliced type III connecting segment domain of fibronectin is leucine-aspartic acid-valine. J Biol Chem 266, 15075–15079. Kozma R, Ahmed S, Best A, Lim L (1995) The Ras-related protein Cdc42Hs and bradykinin promote formation of peripheral actin microspikes and filopodia in Swiss 3T3 fibroblasts. Mol Cell Biol 15, 1942–1952. Krishnan J, Kirkin V, Steffen A, Hegen M, Weih D, Tomarev S, Wilting J, Sleeman JP (2003) Differential in vivo and in vitro expression of vascular endothelial growth factor (VEGF)-C and VEGF-D in tumors and its relationship to lymphatic metastasis in immunocompetent rats. Cancer Res 63, 713–722. Kubo H, Cao R, Brakenhielm E, Makinen T, Cao Y, Alitalo K (2002) Blockade of vascular endothelial growth factor receptor-3 signaling inhibits fibroblast growth factor-2-induced lymphangiogenesis in mouse cornea. Proc Natl Acad Sci U S A 99, 8868–8873. Kwee L, Baldwin HS, Shen HM, Stewart CL, Buck C, Buck CA, Labow MA (1995) Defective development of the embryonic and extraembryonic circulatory systems in vascular cell adhesion molecule (VCAM-1) deficient mice. Development 121, 489–503. Lai KM, Pawson T (2000) The ShcA phosphotyrosine docking protein sensitizes cardiovascular signaling in the mouse embryo. Genes Dev 14, 1132–1145. Leak LV (1976) The structure of lymphatic capillaries in lymph formation. Fed Proc 35, 1863–1871. Liu S, Rose DM, Han J, Ginsberg MH (2000) Alpha4 integrins in cardiovascular development and diseases. Trends Cardiovasc Med 10, 253–257. Liu X, Schnellmann RG (2003) Calpain mediates progressive plasma membrane permeability and proteolysis of cytoskeleton-associated paxillin, talin, and vinculin during renal cell death. J Pharmacol Exp Ther 304, 63–70. Loftus JC, Smith JW, Ginsberg MH (1994) Integrin-mediated cell adhesion: the extracellular face. J Biol Chem 269, 25235–25238. Luo J, Manning BD, Cantley LC (2003) Targeting the PI3K-Akt pathway in human cancer: rationale and promise. Cancer Cell 4, 257–262. Makinen T, Veikkola T, Mustjoki S, Karpanen T, Catimel B, Nice EC, Wise L, Mercer A, Kowalski H, Kerjaschki D, et  al. (2001) Isolated lymphatic endothelial cells transduce growth, survival and migratory signals via the VEGF-C/D receptor VEGFR-3. Embo J 20, 4762–4773.

14  Integrin Signaling in Lymphangiogenesis

307

Mandriota SJ, Jussila L, Jeltsch M, Compagni A, Baetens D, Prevo R, Banerji S, Huarte J, Montesano R, Jackson DG, et al. (2001) Vascular endothelial growth factor-C-mediated lymphangiogenesis promotes tumour metastasis. Embo J 20, 672–682. Maula SM, Luukkaa M, Grenman R, Jackson D, Jalkanen S, Ristamaki R (2003) Intratumoral lymphatics are essential for the metastatic spread and prognosis in squamous cell carcinomas of the head and neck region. Cancer Res 63, 1920–1926. Mishima K, Watabe T, Saito A, Yoshimatsu Y, Imaizumi N, Masui S, Hirashima M, Morisada T, Oike Y, Araie M, et al. (2007) Prox1 induces lymphatic endothelial differentiation via integrin alpha9 and other signaling cascades. Mol Biol Cell 18, 1421–1429. Miyamoto S, Teramoto H, Coso OA, Gutkind JS, Burbelo PD, Akiyama SK, Yamada KM (1995) Integrin function: molecular hierarchies of cytoskeletal and signaling molecules. J Cell Biol 131, 791–805. Mould AP, Akiyama SK, Humphries MJ (1995) Regulation of integrin alpha 5 beta 1-fibronectin interactions by divalent cations. Evidence for distinct classes of binding sites for Mn2+, Mg2+, and Ca2+. J Biol Chem 270, 26270–26277. Mumprecht V, Detmar M (2009) Lymphangiogenesis and cancer metastasis. J Cell Mol Med 13, 1405–1416. Nagy JA, Vasile E, Feng D, Sundberg C, Brown LF, Manseau EJ, Dvorak AM, Dvorak HF (2002) VEGF-A induces angiogenesis, arteriogenesis, lymphangiogenesis, and vascular malformations. Cold Spring Harb Symp Quant Biol 67, 227–237. Nobes CD, Hall A (1995) Rho, rac, and cdc42 GTPases regulate the assembly of multimolecular focal complexes associated with actin stress fibers, lamellipodia, and filopodia. Cell 81, 53–62. Okazaki T, Ni A, Ayeni OA, Baluk P, Yao LC, Vossmeyer D, Zischinsky G, Zahn G, Knolle J, Christner C, McDonald DM (2009) alpha5beta1 Integrin blockade inhibits lymphangiogenesis in airway inflammation. Am J Pathol 174, 2378–2387. Owen JD, Ruest PJ, Fry DW, Hanks SK (1999) Induced focal adhesion kinase (FAK) expression in FAK-null cells enhances cell spreading and migration requiring both auto- and activation loop phosphorylation sites and inhibits adhesion-dependent tyrosine phosphorylation of Pyk2. Mol Cell Biol 19, 4806–4818. Padera TP, Kadambi A, di Tomaso E, Carreira CM, Brown EB, Boucher Y, Choi NC, Mathisen D, Wain J, Mark EJ, et al. (2002) Lymphatic metastasis in the absence of functional intratumor lymphatics. Science 296, 1883–1886. Petrova TV, Bono P, Holnthoner W, Chesnes J, Pytowski B, Sihto H, Laakkonen P, Heikkila P, Joensuu H, Alitalo K (2008) VEGFR-3 expression is restricted to blood and lymphatic vessels in solid tumors. Cancer Cell 13, 554–556. Petrova TV, Makinen T, Makela TP, Saarela J, Virtanen I, Ferrell RE, Finegold DN, Kerjaschki D, Yla-Herttuala S, Alitalo K (2002) Lymphatic endothelial reprogramming of vascular endothelial cells by the Prox-1 homeobox transcription factor. Embo J 21, 4593–4599. Plow EF, Haas TA, Zhang L, Loftus J, Smith JW (2000) Ligand binding to integrins. J Biol Chem 275, 21785–21788. Pratt JC, van den Brink MR, Igras VE, Walk SF, Ravichandran KS, Burakoff SJ (1999) Requirement for Shc in TCR-mediated activation of a T cell hybridoma. J Immunol 163, 2586–2591. Prevo R, Banerji S, Ferguson DJ, Clasper S, Jackson DG (2001) Mouse LYVE-1 is an endocytic receptor for hyaluronan in lymphatic endothelium. J Biol Chem 276, 19420–19430. Pronk GJ, de Vries-Smits AM, Buday L, Downward J, Maassen JA, Medema RH, Bos JL (1994) Involvement of Shc in insulin- and epidermal growth factor-induced activation of p21ras. Mol Cell Biol 14, 1575–1581. Reiske HR, Zhao J, Han DC, Cooper LA, Guan JL (2000) Analysis of FAK-associated signaling pathways in the regulation of cell cycle progression. FEBS Lett 486, 275–280. Renyi-Vamos F, Tovari J, Fillinger J, Timar J, Paku S, Kenessey I, Ostoros G, Agocs L, Soltesz I, Dome B (2005) Lymphangiogenesis correlates with lymph node metastasis, prognosis, and angiogenic phenotype in human non-small cell lung cancer. Clin Cancer Res 11, 7344–7353.

308

B. Garmy-Susini

Richardson A, Parsons T (1996) A mechanism for regulation of the adhesion-associated proteintyrosine kinase pp125FAK. Nature 380, 538–540. Ridley AJ (1996) Rho: theme and variations. Curr Biol 6, 1256–1264. Ridley AJ, Hall A (1992) The small GTP-binding protein rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors. Cell 70, 389–399. Roberts N, Kloos B, Cassella M, Podgrabinska S, Persaud K, Wu Y, Pytowski B, Skobe M (2006) Inhibition of VEGFR-3 activation with the antagonistic antibody more potently suppresses lymph node and distant metastases than inactivation of VEGFR-2. Cancer Res 66, 2650–2657. Rose DM, Alon R, Ginsberg MH (2007) Integrin modulation and signaling in leukocyte adhesion and migration. Immunol Rev 218, 126–134. Salcini AE, McGlade J, Pelicci G, Nicoletti I, Pawson T, Pelicci PG (1994) Formation of Shc-Grb2 complexes is necessary to induce neoplastic transformation by overexpression of Shc proteins. Oncogene 9, 2827–2836. Schaller MD, Parsons JT (1994) Focal adhesion kinase and associated proteins. Curr Opin Cell Biol 6, 705–710. Schlaepfer DD, Hanks SK, Hunter T, van der Geer P (1994) Integrin-mediated signal transduction linked to Ras pathway by GRB2 binding to focal adhesion kinase. Nature 372, 786–791. Schlaepfer DD, Hauck CR, Sieg DJ (1999) Signaling through focal adhesion kinase. Prog Biophys Mol Biol 71, 435–478. Schlaepfer DD, Hunter T (1996) Evidence for in vivo phosphorylation of the Grb2 SH2-domain binding site on focal adhesion kinase by Src-family protein-tyrosine kinases. Mol Cell Biol 16, 5623–5633. Schmid-Schonbein GW (2003) The second valve system in lymphatics. Lymphat Res Biol 1, 25–29; discussion 29–31. Sengbusch JK, He W, Pinco KA, Yang JT (2002) Dual functions of [alpha]4[beta]1 integrin in epicardial development: initial migration and long-term attachment. J Cell Biol 157, 873–882. Shin JW, Min M, Larrieu-Lahargue F, Canron X, Kunstfeld R, Nguyen L, Henderson JE, Bikfalvi A, Detmar M, Hong YK (2006) Prox1 promotes lineage-specific expression of fibroblast growth factor (FGF) receptor-3 in lymphatic endothelium: a role for FGF signaling in lymphangiogenesis. Mol Biol Cell 17, 576–584. Skobe M, Hawighorst T, Jackson DG, Prevo R, Janes L, Velasco P, Riccardi L, Alitalo K, Claffey K, Detmar M (2001) Induction of tumor lymphangiogenesis by VEGF-C promotes breast cancer metastasis. Nat Med 7, 192–198. Smith JW, Piotrowicz RS, Mathis D (1994) A mechanism for divalent cation regulation of beta 3-integrins. J Biol Chem 269, 960–967. Stacker SA, Achen MG (2008) From anti-angiogenesis to anti-lymphangiogenesis: emerging trends in cancer therapy. Lymphat Res Biol 6, 165–172. Stacker SA, Caesar C, Baldwin ME, Thornton GE, Williams RA, Prevo R, Jackson DG, Nishikawa S, Kubo H, Achen MG (2001) VEGF-D promotes the metastatic spread of tumor cells via the lymphatics. Nat Med 7, 186–191. Stacker SA, Williams RA, Achen MG (2004) Lymphangiogenic growth factors as markers of tumor metastasis. Apmis 112, 539–549. Stupack DG, Teitz T, Potter MD, Mikolon D, Houghton PJ, Kidd VJ, Lahti JM, Cheresh DA (2006) Potentiation of neuroblastoma metastasis by loss of caspase-8. Nature 439, 95–99. Sugiura T, Inoue Y, Matsuki R, Ishii K, Takahashi M, Abe M, Shirasuna K (2009) VEGF-C and VEGF-D expression is correlated with lymphatic vessel density and lymph node metastasis in oral squamous cell carcinoma: Implications for use as a prognostic marker. Int J Oncol 34, 673–680. Tadokoro S, Shattil SJ, Eto K, Tai V, Liddington RC, de Pereda JM, Ginsberg MH, Calderwood DA (2003) Talin binding to integrin beta tails: a final common step in integrin activation. Science 302, 103–106.

14  Integrin Signaling in Lymphangiogenesis

309

Tammela T, Zarkada G, Wallgard E, Murtomaki A, Suchting S, Wirzenius M, Waltari M, Hellstrom M, Schomber T, Peltonen R, et al. (2008) Blocking VEGFR-3 suppresses angiogenic sprouting and vascular network formation. Nature 454, 656–660. Tapon N, Hall A (1997) Rho, Rac and Cdc42 GTPases regulate the organization of the actin cytoskeleton. Curr Opin Cell Biol 9, 86–92. Trzewik J, Mallipattu SK, Artmann GM, Delano FA, Schmid-Schonbein GW (2001) Evidence for a second valve system in lymphatics: endothelial microvalves. Faseb J 15, 1711–1717. Van der Auwera I, Van Laere SJ, Van den Eynden GG, Benoy I, van Dam P, Colpaert CG, Fox SB, Turley H, Harris AL, Van Marck EA, et  al. (2004) Increased angiogenesis and lymphangiogenesis in inflammatory versus noninflammatory breast cancer by real-time reverse transcriptase-PCR gene expression quantification. Clin Cancer Res 10, 7965–7971. van der Flier A, Sonnenberg A (2001) Function and interactions of integrins. Cell Tissue Res 305, 285–298. Vlahakis NE, Young BA, Atakilit A, Sheppard D (2005) The lymphangiogenic vascular endothelial growth factors VEGF-C and -D are ligands for the integrin alpha9beta1. J Biol Chem 280, 4544–4552. Wang W, Bergh A, Damber JE (2004) Chronic inflammation in benign prostate hyperplasia is associated with focal upregulation of cyclooxygenase-2, Bcl-2, and cell proliferation in the glandular epithelium. Prostate 61, 60–72. Wary KK, Mainiero F, Isakoff SJ, Marcantonio EE, Giancotti FG (1996) The adaptor protein Shc couples a class of integrins to the control of cell cycle progression. Cell 87, 733–743. Wary KK, Mariotti A, Zurzolo C, Giancotti FG (1998) A requirement for caveolin-1 and associated kinase Fyn in integrin signaling and anchorage-dependent cell growth. Cell 94, 625–634. Weller PA, Ogryzko EP, Corben EB, Zhidkova NI, Patel B, Price GJ, Spurr NK, Koteliansky VE, Critchley DR (1990) Complete sequence of human vinculin and assignment of the gene to chromosome 10. Proc Natl Acad Sci U S A 87, 5667–5671. Wicki A, Lehembre F, Wick N, Hantusch B, Kerjaschki D, Christofori G (2006) Tumor invasion in the absence of epithelial-mesenchymal transition: podoplanin-mediated remodeling of the actin cytoskeleton. Cancer Cell 9, 261–272. Wigle JT, Oliver G (1999) Prox1 function is required for the development of the murine lymphatic system. Cell 98, 769–778. Williams CS, Leek RD, Robson AM, Banerji S, Prevo R, Harris AL, Jackson DG (2003) Absence of lymphangiogenesis and intratumoural lymph vessels in human metastatic breast cancer. J Pathol 200, 195–206. Xing Z, Chen HC, Nowlen JK, Taylor SJ, Shalloway D, Guan JL (1994) Direct interaction of v-Src with the focal adhesion kinase mediated by the Src SH2 domain. Mol Biol Cell 5, 413–421. Yang JT, Rayburn H, Hynes RO (1993) Embryonic mesodermal defects in alpha 5 integrin-deficient mice. Development 119, 1093–1105. Yang JT, Rayburn H, Hynes RO (1995) Cell adhesion events mediated by alpha 4 integrins are essential in placental and cardiac development. Development 121, 549–560. Zamir E, Geiger B (2001) Molecular complexity and dynamics of cell-matrix adhesions. J Cell Sci 114, 3583–3590. Zeng Y, Opeskin K, Horvath LG, Sutherland RL, Williams ED (2005) Lymphatic vessel density and lymph node metastasis in prostate cancer. Prostate 65, 222–230.

Chapter 15

Role of Pericytes in Resistance to Antiangiogenic Therapy Koji Matsuo, Chunhua Lu, Mian M.K. Shazad, Robert L. Coleman, and Anil K. Sood

Abstract  Angiogenesis plays an important role in the progressive growth of primary tumor and metastasis, and targeting tumor angiogenesis as a therapeutic strategy is showing promise. While these approaches have shown improved survival for some cancer patients, most eventually develop progressive disease due to resistance to antiangiogenic therapy. Recent evidence suggests a functional role for pericytes in acquired resistance to antiangiogenesis agents. Pericytes play an important role in stabilizing blood vessels in the microvasculature regulated by the PDGF ligand (PDGF-BB) and receptor (PDGFR-b) homeostasis, and serve as a local source of survival factors for endothelial cells. Therefore, dual targeting of pericytes (PDGF axis blockers) and endothelial cells (VEGF pathway blockers) may be more efficacious than targeting either cell type alone.

Introduction Oxygen is vitally important to the survival of most organisms. Due to limitations in oxygen diffusion, mammalian cells require blood supply within 100–200 µm of its cell lining (Carmeliet and Jain 2000). As a result, it has been estimated that for a tumor to grow beyond 1mm in size, angiogenesis is required (Naumov et al. 2006). Recruitment of new blood vessels can occur by many pathways including vasculogenesis and angiogenesis (Carmeliet and Jain 2000; Semenza 2003). Pericytes and endothelial cells are

A.K. Sood (*) Department of Gynecologic Oncology, University of Texas M. D. Anderson Cancer Center, Houston, TX, USA and Center for RNA Interference and Non-Coding RNA, University of Texas M. D. Anderson Cancer Center, Houston, TX, USA and Department of Cancer Biology, University of Texas M. D. Anderson Cancer Center, Houston, TX, USA e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_15, © Springer Science+Business Media, LLC 2010

311

312

K. Matsuo et al.

two critical components of blood vessel formation, and the physical and chemical interactions between the two cells are indispensable for normal endothelial function (Allt and Lawrenson 2001; Cuevas et al. 1984; Gerhardt and Betsholtz 2003). For cancer, angiogenesis plays an important role in progressive growth of primary tumor and metastasis. Therefore, targeting tumor angiogenesis may provide an efficient strategy to block tumor growth. Previous studies targeting angiogenesis have widely focused on tumor endothelial cells and their progenitors. While these approaches have shown improved survival for some cancer patients, most eventually develop progressive disease due to resistance to antiangiogenic therapy (Kerbel 2008). Several mechanisms for this acquired antiangiogenic resistance have been hypothesized, one of which is the functional role of pericytes in the tumor microenvironment (Bergers and Hanahan 2008; Kerbel 2008). In this chapter, we will focus on physiology of pericytes in tumor vasculature, targeting pericytes as a therapeutic modality, role of pericytes in resistance to antiangiogenesis therapy, and strategies to overcome pericyte-associated antiangiogenic resistance.

Biology, Physiology, and Pathology of Pericytes Despite decades of research, the origin of pericytes is still not fully understood. It is currently believed that pericytes develop from various types of progenitors depending on their anatomic location in the body. For example, epicardial, mesenchymal, and neural crest cells are believed to be a source for pericytes in the cardiac coronary vasculature, dorsal aorta, and cardiac outflow tract, respectively (Bergwerff et  al. 1998; Drake et al. 1998; Vrancken Peeters et al. 1999). Pericytes play an important role in stabilizing blood vessels in the microvasculature (Nehls and Drenckhahn 1993; Sims 1986). A hallmark of pericyte function is their ability to provide vascular stability by depositing matrix or releasing factors that can promote endothelial cell differentiation or quiescence (Armulik et al. 2005). For example, it was demonstrated that pericytes release angiopoietin-1 (Ang-1) that binds to epidermal growth factor homology domains-2 (Tie2) receptor (von Tell et al. 2006). Tie2 receptor expression recently has been identified in mesenchymal cells that are present in the stroma; implicating a repository for tumor vessel pericytes (De Palma et al. 2005). Pericyte homeostasis in normal biology is regulated in significant part by signaling through the PDGF ligand and receptor system (Fig. 15.1) (Lindahl et al. 1999; Pietras and Hanahan 2005). PDGF is a potent mitogen for pericytes and fibroblasts and is composed of A, B, C, and D polypeptide chains and forms homodimers PDGF-AA, BB, CC, and DD and heterodimer PDGF-AB (Andrae et  al. 2008). Its biological activities are linked to two tyrosine kinase receptors, PDGFR-a and PDGFR-b (Heldin et al. 1998; Kelly et al. 1991). PDGFR-a binds to PDGF-AA, BB, AB, and CC, whereas PDGFR-b interacts with BB and DD (Betsholtz et al. 2001). Previous studies have shown that up to 90% reduction in pericyte coverage in mice is compatible with postnatal survival (Enge et al. 2002), whereas loss of >95% of pericytes is lethal (Enge et al. 2002; Lindahl et al. 1997), suggesting that a rather

15  Role of Pericytes in Resistance to Antiangiogenic Therapy

313

Fig. 15.1  PDGF pathways in tumor biology. PDGF ligands and receptor pathway is suggested to be involved in the recruitment of pericytes and stabilization of blood vessels

low threshold density of pericytes is required for basal function of microvasculature. Pericyte deficiency, seen in knockout mice lacking platelet-derived growth factor-B (PDGF-B) and its receptor, PDGF receptor-beta (PDGFR-b), results in various changes in microvasculature, including endothelial hyperplasia, vessel dilation, tortuosity, leakage, and rupture, leading to widespread and lethal microhemorrhage and edema in late gestation (Hellstrom et  al. 1996b, 2001; Lindahl et  al. 1997). Studies of implanted tumors have shown that pericytes initially accumulate at the interface of tumor and host tissue and later around new blood vessels exhibiting close contacts with endothelial cells. Maturation of the tumor-associated vasculature is accompanied quantitatively by a reduced pericyte volume and qualitatively by morphological changes wherein pericytes become flattened and elongated (Verhoeven and Buyssens 1988).

Pericytes and Tumor Angiogenesis Pericytes are tightly attached to the endothelial cells in normal vasculature; however, it is becoming apparent that pericyte coverage in tumor blood vessels is quite distinct. Studies with insulinoma, breast, and lung carcinoma models have demonstrated that pericytes were present on most blood vessels, but 30–50% of the endothelial surface had no pericyte coverage (Morikawa et al. 2002). The extent of pericyte coverage in different tumors ranges from expansive (Schlingemann et al. 1990; Wesseling et al. 1995) to little or none (Benjamin et al. 1999; Eberhard et al. 2000; Johnson and Bruce 1997). Some of these differences may be explained by

314

K. Matsuo et al.

Fig. 15.2  Pericyte coverage on the microvasculature. Dual immunofluorescence staining for CD31 (red) and desmin (green) was performed to visualize endothelial cells and pericytes, respectively. (a) In the normal vasculature, pericytes cover vessels extensively and tightly. (b) In the tumor vasculature, pericytes are morphologically abnormal and tortuous in shape, attach to endothelial cells loosely, and paradoxically extend cytoplasmic processes away from the vessel wall

differences in pericyte marker expression among tumors. Pericytes in tumor vasculature are morphologically abnormal and tortuous in shape, attach to endothelial cells loosely, and paradoxically extend cytoplasmic processes away from the vessel wall (Fig. 15.2) (Lu et al. 2008). Several markers, including a-SMA, NG2, and desmin, have been found to be useful for detecting pericytes (Schlingemann et al. 1990, 1991). Pericytes also play an important role in regulating local blood flow, phagocytosis, modulation of new blood vessel growth and protecting endothelial cells from apoptosis (Hirschi and D’Amore 1996; Reinmuth et al. 2001). It has been demonstrated that pericytes can serve as a local source of vascular endothelial growth factor (VEGF), which is a survival factor for endothelial cells (Brown et al. 2001; Fukumura et  al. 1998; Lu et  al. 2008; Reinmuth et  al. 2001) (Fig. 15.2). Tumor vessels lacking pericytes appear to be more dependent on VEGF for their survival than are vessels invested by pericytes (Benjamin et  al. 1999). It is possible that decreased pericyte coverage may make the vascular endothelium vulnerable to VEGF blockage. Tyrosine kinase inhibitors affecting multiple receptor tyrosine kinases may exert their antitumor activity in part by reducing pericyte density in tumor vessels, thereby sensitizing them to inhibition of endothelial receptor tyrosine kinases (Lindblom et al. 1999, 2003; Shaheen et al. 2001). Variable recruitment of pericytes into tumor-associated vasculature may be explained by differences in expression of the PDGF ligands. Recent studies suggest that PDGF-B expression stimulates pericyte recruitment in gliomas (Guo et  al. 2003). Both PDGF-AA and BB ligands are expressed in most ovarian cancer samples (Apte et al. 2004). PDGF-B produced by tumor endothelial cells has been shown to be required for the recruitment of adequate number of pericytes as well as proper integration of pericytes in the vascular wall (Abramsson et al. 2003). This may be further enhanced by PDGF-B production directly from tumor cells (Abramsson et al. 2003). Genetic ablation of PDGF-B in endothelial cells leads to

15  Role of Pericytes in Resistance to Antiangiogenic Therapy

315

impaired recruitment of pericytes, resulting in glomerular, cardiac and placental abnormalities using Cre-lox techniques (Bjarnegard et al. 2004). The close interrelationship between these control mechanisms suggests a short-range paracrine system. Using tyrosine kinase inhibitors, a functional role for PDGFR-b has also been implicated in pericyte recruitment in mouse insulinomas (Bergers et al. 2003). Interestingly, the tyrosine kinase inhibitors had no effect on normal tissues with regard to pericyte detachment, but they disrupted the association of pericytes from endothelial cells in tumors (Bergers et al. 2003). These results suggest that there may be differential sensitivity of pericytes in developing immature blood vessels compared to the mature vasculature with regard to dependence on PDGF.

Pericytes and Resistance to Antiangiogenic Therapy One of the first lines of evidence regarding the role of pericytes in protecting the tumor endothelium came from observations following treatment with anti-VEGF therapy. It was noted that anti-VEGF treatment with receptor tyrosine kinase inhibitors in RIP-Tag mice resulted in significant reductions in tumor vasculature, however, the antiangiogenic therapy failed to affect the basement membrane and pericytes (Mancuso et al. 2006).

VEGF Pathway and Pericytes in Tumor Angiogenesis VEGF affects not only endothelial cell proliferation but also cell survival (Carmeliet et al. 1996; Senger et al. 1983). Among several strategies for platinum refractory ovarian cancer, it was concluded that antiangiogenesis approaches targeting VEGF pathway were among the most efficacious (Burger et al. 2007). However, it is possible that anti-VEGF therapy selectively “prunes” blood vessels that have little or no pericyte coverage (Bergers and Hanahan 2008). Upon stimulation of pericytes (e.g., PDGF-BB), local production of factors such as VEGF may serve as a survival factor for the remaining endothelial cells (Lu et al. 2008; Reinmuth et al. 2001). For example, antiangiogenesis therapy with a VEGF receptor inhibitor (AEE788) significantly reduced microvessel density in orthotopic mouse models of ovarian cancer, but the percentage of pericyte-coated vessels was significantly increased, suggesting that pericytes may protect these vessels (Lu et al. 2007).

Resistance to Antiangiogenic Therapy The mechanisms of resistance to antiangiogenic compounds are not well understood, but may overlap to some degree with those factors responsible for resistance to cytotoxic agents (Broxterman and Georgopapadakou 2007; Shahzad et al. 2009).

316

K. Matsuo et al.

For instance, similar to cytotoxic drugs, antiangiogenic agents may face resistance due to decreased bioavailability in the tumors or by upregulation of factors that are involved in apoptosis inhibition. Antiangiogenic agents that affect endothelial cells may not be effectively targeting other components of the tumor vasculature. There are currently many antiangiogenic agents in clinical development for ovarian cancer. While antiangiogenesis treatment strategies are showing promise in clinical trials for patients with ovarian and other cancers (prolonging progressionfree and/or overall survival), most patients eventually develop progression resulting in eventual death (Burger et  al. 2007; Jain et  al. 2006; Spannuth et  al. 2008). Emerging evidence from preclinical and clinical data suggest at least two modes of resistance to antiangiogenic therapy (Bergers and Hanahan 2008). The first mode is adaptive (evasive) resistance and the second is intrinsic (preexisting) nonresponsiveness (Bergers and Hanahan 2008). Possible mechanisms associated with adaptive resistance to anti-VEGF therapy include: (1) activation and/or upregulation of alternative proangiogenic pathways in the tumor; (2) activation and enhancement of invasion and metastasis to provide access to normal tissue vasculature without neovascularization; (3) recruitment of bone marrow-derived proangiogenic cells that differentiate into pericytes; and (4) increased pericyte coverage of tumor microvasculature serving to support survival functions for tumor endothelial cells (Bergers and Hanahan 2008). Antiangiogenic approaches are highly effective in tumor vessels lacking adequate pericyte coverage (Benjamin et al. 1999). It was found that antiangiogenic therapy can effectively reduce the overall vessel density and increase endothelial cell apoptosis in pericyte-negative vessels, but the fraction of pericyte-positive vessels significantly increases after such therapy (McCarty et al. 2004). For example, a VEGFR inhibitor was effective against endothelial cells, affecting early stage angiogenic vasculature, but not against the more mature vasculature in larger tumors (Bergers et  al. 2003). The vessels in the tumors intrinsically resistant to antiangiogenic therapy showed slim and tightly covered blood vessels with pericytes (Jain and Booth 2003; Mancuso et al. 2006). These blood vessels are quite distinct from the untreated tumor vasculature, which is often characterized by pericytes that are loosely attached to endothelial cells (Jain 2005; Jain and Booth 2003). On the basis of such data, it is thought that endothelial cells may induce pericyte recruitment to protect themselves from the effects of antiangiogenic therapy. While disruption of the connections between pericytes and endothelial cells may be an attractive approach to increase the antiangiogenic effects on endothelial cells, there is an emerging concern of increased risk of metastasis associated with facilitation of tumor cell entry into the circulation through the disrupted vasculature (Xian et al. 2006). Recruitment of pericytes for external coating and reinforcement of immature blood vessels is essential in the process of normal vascular development (Allt and Lawrenson 2001; Cuevas et al. 1984; Gerhardt and Betsholtz 2003). Antiangiogenic agents may inhibit other kinases as well, and whether their effects on pericytes are mediated by kinases other than PDGFR-b cannot be ruled out. In addition, PDGF-B and PDGFR-b have other roles beside pericyte recruitment,

15  Role of Pericytes in Resistance to Antiangiogenic Therapy

317

including autocrine growth stimulation, formation of tumor stroma, and control of interstitial tumor pressure that may also affect tumor growth (Forsberg et  al. 1993; Heldin and Westermark 1999; Pietras et  al. 2001; Westermark and Heldin 1991). Other growth factors that may affect pericyte function include endothelin-1 (produced by endothelial cells) (Egginton et al. 1996), interleukin-2 (may increase localization of pericytes to endothelial cell junctions) (Sims et al. 1994), and transforming growth factor-b (produced by pericytes and may inhibit endothelial cell proliferation) (Wakui et  al. 1997). However, the precise mechanism underlying recruitment of pericytes to endothelial cells is not well understood. PDGF signaling is suggested as the key factor for pericyte recruitment due to its critical role in pericyte development and function (Hellstrom et al. 1999a, 2001; Lindahl et al. 1997). Tumor hypoxia due to the vascular regression following antiangiogenic treatment appears to induce recruitment of various bone marrow-derived cells to the tumor microenvironment (Du et  al. 2008). In the presence of inducible HIF1-a, populations of CD45+ myeloid cells containing Tie2, VEGFR1, CD11b, and F4/80 subpopulations, as well as endothelial and pericyte progenitor cells are released into the circulation where they promote neovascularization in glioblastoma (Du et  al. 2008). In absence of HIF1-a, fewer bone marrow-derived cells are recruited to the tumors, which severely impair tumor growth. These data suggest paradoxical induction of tumor angiogenesis via bone marrow-derived vessel progenitor cells after antiangiogenic therapy.

Targeting Pericytes for Antivascular Strategies Signaling through PDGF ligand and receptor system significantly regulates pericyte homeostasis (Lindblom et al. 2003; Pietras and Hanahan 2005). On the basis of the functional and biological roles of the PDGD-BB/PDGFR-b axis on pericyte function, inhibitors of this pathway may be effective in targeting pericytes. Several available clinical or preclinical approaches to inhibiting PDGF-B and PDGFR-b pathway include: (1) tyrosine kinase inhibitors of PDGF receptor, such as imatinib mesylate (Lu et al. 2007 2008), SU6668 (Erber et al. 2004); (2) PDGF-B aptamer, a modified DNA-based aptamer to PDGF-B chain that blocks binding of PDGF-B to its cell-surface receptor (Fredriksson et  al. 2002); (3) PDGF Trap targeting PDGF-BB (Lu et  al. 2008); and (4) RNA interference. Treatment of RIP-Tag tumors with AX102, an aptamer blocking PDGF-BB, showed loss of pericytes with subsequent loss of endothelial cells; thereby emphasizing the importance of pericytes in tumor associated vasculature (Sennino et al. 2007). However, not surprisingly, monotherapy with these agents has been ineffective and may actually accelerate tumor growth by making the tumor vasculature more immature (Jayson et  al. 2005; Spannuth et  al. 2008). In contrast, combination of PDGF-BB/ PDGFR-b inhibitors with anti-VEGF or cytotoxic agents may enhance efficacy of

318

K. Matsuo et al.

antiangiogenic therapy. For example, several studies with various tumor models recently have shown that dual targeting of endothelial cells with agents such as AEE788 or SU5416, and pericytes with STI571 or SU6668, is more efficacious than targeting either cell type alone, even in established or drug-resistant tumors (Bergers et al. 2003; Lu et al. 2007). Furthermore, triple combination therapy with a cytotoxic agent, AEE788, and STI571 was effective in regressing bulky tumors in orthotopic mouse models of ovarian cancer using in  vivo bioluminescence imaging (Lu et al. 2007). Other dual targeting approaches that showed efficacy in inhibiting in  vitro tumor growth include combination of VEGF Trap and PDGF Trap targeting PDGF-BB (Lu et al. 2008), and bevacizumab with AX102. We and others have shown that the PDGF-BB aptamer AX102 can significantly reduce pericyte coverage, and also enhance efficacy of anti-VEGF therapy. It is possible that the efficacy of some of the new tyrosine kinase inhibitors such as SU11248 and SU14813 may be related to targeting both endothelial cells (VEGFR) and pericytes (PDGFR) (Hartmann et  al. 2009; Hopkins et  al. 2008; Patyna et  al. 2006). Similarly, dual targeting of endothelial cells and pericytes has been reported to cause regression of pancreatic tumors in RIP1 Tag 2 mice (Bergers et al. 2003). Thus, for bulky tumors, which are often seen in recurrent disease, these approaches may have clinical relevance. In addition to PDGF-B/PDGFR-b pathway, other pathways such as Ang1 and a receptor tyrosine kinase with immunoglobulin and Tie2 pathway, sphingosine 1-phosphate (S1P) and endothelial differentiation gene-1 (Edg-1), TGF-b1 and activin-like kinase receptor (Alk5) may also offer additional opportunities for therapeutic targeting of pericytes (Erber et al. 2004). Neuropilin is a receptor of semaphorin that was initially implicated in the development of nervous system and in axon guidance (Capparuccia and Tamagnone 2009). However, recent studies have found that neuropilin is expressed in pericytes associated with angiogenesis. It is also suggested that blocking neuropilin-1 function inhibits vascular remodeling, rendering vessels more susceptible to anti-VEGF therapy (Pan et al. 2007). Neuropilin-2 is a high-affinity kinase-deficient receptor for VEGF and semaphorin, and expressed in certain types cancer (Gray et al. 2008). Dual targeting of neuropilin in tumor and pericytes may offer novel approaches for antivascular therapy.

Conclusions In summary, pericytes appear to be emerging as an important target for antivascular approaches. Pericytes not only function to stabilize the vasculature, but may play an important role in conferring resistance to anti-VEGF therapy. Therefore, dual endothelial and pericyte targeting holds promise for more efficacious cancer treatment strategies.

15  Role of Pericytes in Resistance to Antiangiogenic Therapy

319

Fig. 15.3  Interactions between tumor, endothelial cells, and pericytes. Cancer and endothelial cells produce PDGF-BB. PDGF-BB promotes VEGF production by pericytes. VEGF provides local survival signals for mature vasculature (adapted and modified, Lu et al. 2008)

Acknowledgments  The authors declare that there is no conflict of interest. KM is supported by the GCF/OCRF Ann Schreiber Ovarian Cancer Research grant and an award from the Meyer and Ida Gordon Foundation #2. MMS is supported by the GCF-Molly Cade ovarian cancer research grant and the NIH/NICHD Baylor WRHR scholarship grant (HD050128). Portions of this work were supported by NIH grants (CA 110793 and 109298), the Ovarian Cancer Research Fund, Inc. (Program Project Development Grant), U. T. M. D. Anderson Cancer Center SPORE (P50CA083639 and P50CA098258), the Marcus Foundation, the Entertainment Industry Foundation, the Blanton-Davis Ovarian Cancer Research Program, and the Betty Anne Asche Murray Distinguished Professorship. We thank Ms. Alison E. Schroeer for assistance with graphic design for Figs. 15.1 and 15.3.

320

K. Matsuo et al.

References Abramsson A, Lindblom P, Betsholtz C (2003) Endothelial and nonendothelial sources of PDGF-B regulate pericyte recruitment and influence vascular pattern formation in tumors. J Clin Invest 112:1142–1151. Allt G, Lawrenson JG (2001) Pericytes: cell biology and pathology. Cells Tissues Organs 169:1–11. Andrae J, Gallini R, Betsholtz C (2008) Role of platelet-derived growth factors in physiology and medicine. Genes Dev 22:1276–1312. Apte SM, Bucana CD, Killion JJ, Gershenson DM, Fidler IJ (2004) Expression of platelet-derived growth factor and activated receptor in clinical specimens of epithelial ovarian cancer and ovarian carcinoma cell lines. Gynecol Oncol 93:78–86. Armulik A, Abramsson A, Betsholtz C (2005) Endothelial/pericyte interactions. Circ Res 97:512–523. Benjamin LE, Golijanin D, Itin A, Pode D, Keshet E (1999) Selective ablation of immature blood vessels in established human tumors follows vascular endothelial growth factor withdrawal. J Clin Invest 103:159–165. Bergers G, Hanahan D (2008) Modes of resistance to anti-angiogenic therapy. Nat Rev Cancer 8:592–603. Bergers G, Song S, Meyer-Morse N, Bergsland E, Hanahan D (2003) Benefits of targeting both pericytes and endothelial cells in the tumor vasculature with kinase inhibitors. J Clin Invest 111:1287–1295. Bergwerff M, Verberne ME, DeRuiter MC, Poelmann RE, Gittenberger-de Groot AC (1998) Neural crest cell contribution to the developing circulatory system: implications for vascular morphology? Circ Res 82:221–231. Betsholtz C, Karlsson L, Lindahl P (2001) Developmental roles of platelet-derived growth factors. Bioessays 23:494–507. Bjarnegard M, Enge M, Norlin J, Gustafsdottir S, Fredriksson S, Abramsson A, Takemoto M, Gustafsson E, Fassler R, Betsholtz C (2004) Endothelium-specific ablation of PDGFB leads to pericyte loss and glomerular, cardiac and placental abnormalities. Development 131: 1847–1857. Brown EB, Campbell RB, Tsuzuki Y, Xu L, Carmeliet P, Fukumura D, Jain RK (2001) In vivo measurement of gene expression, angiogenesis and physiological function in tumors using multiphoton laser scanning microscopy. Nat Med 7:864–868. Broxterman HJ, Georgopapadakou NH (2007) Anticancer therapeutics: a surge of new developments increasingly target tumor and stroma. Drug Resist Updat 10:182–193. Burger RA, Sill MW, Monk BJ, Greer BE, Sorosky JI (2007) Phase II trial of bevacizumab in persistent or recurrent epithelial ovarian cancer or primary peritoneal cancer: a Gynecologic Oncology Group Study. J Clin Oncol 25:5165–5171. Capparuccia L, Tamagnone L (2009) Semaphorin signaling in cancer cells and in cells of the tumor microenvironment – two sides of a coin. J Cell Sci 122:1723–1736. Carmeliet P, Ferreira V, Breier G, Pollefeyt S, Kieckens L, Gertsenstein M, Fahrig M, Vandenhoeck A, Harpal K, Eberhardt C, Declercq C, Pawling J, Moons L, Collen D, Risau W, Nagy A (1996) Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature 380:435–439. Carmeliet P, Jain RK (2000) Angiogenesis in cancer and other diseases. Nature 407:249–257. Cuevas P, Gutierrez-Diaz JA, Reimers D, Dujovny M, Diaz FG, Ausman JI (1984) Pericyte endothelial gap junctions in human cerebral capillaries. Anat Embryol (Berl) 170:155–159. De Palma M, Venneri MA, Galli R, Sergi Sergi L, Politi LS, Sampaolesi M, Naldini L (2005) Tie2 identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitors. Cancer Cell 8:211–226. Drake CJ, Hungerford JE, Little CD (1998) Morphogenesis of the first blood vessels. Ann N Y Acad Sci 857:155–179.

15  Role of Pericytes in Resistance to Antiangiogenic Therapy

321

Du R, Lu KV, Petritsch C, Liu P, Ganss R, Passegue E, Song H, Vandenberg S, Johnson RS, Werb Z, Bergers G (2008) HIF1alpha induces the recruitment of bone marrow-derived vascular modulatory cells to regulate tumor angiogenesis and invasion. Cancer Cell 13:206–220. Eberhard A, Kahlert S, Goede V, Hemmerlein B, Plate KH, Augustin HG (2000) Heterogeneity of angiogenesis and blood vessel maturation in human tumors: implications for antiangiogenic tumor therapies. Cancer Res 60:1388–1393. Egginton S, Hudlicka O, Brown MD, Graciotti L, Granata AL (1996) In vivo pericyte-endothelial cell interaction during angiogenesis in adult cardiac and skeletal muscle. Microvasc Res 51:213–228. Enge M, Bjarnegard M, Gerhardt H, Gustafsson E, Kalen M, Asker N, Hammes HP, Shani M, Fassler R, Betsholtz C (2002) Endothelium-specific platelet-derived growth factor-B ablation mimics diabetic retinopathy. EMBO J 21:4307–4316. Erber R, Thurnher A, Katsen AD, Groth G, Kerger H, Hammes HP, Menger MD, Ullrich A, Vajkoczy P (2004) Combined inhibition of VEGF and PDGF signaling enforces tumor vessel regression by interfering with pericyte-mediated endothelial cell survival mechanisms. FASEB J 18:338–340. Forsberg K, Valyi-Nagy I, Heldin CH, Herlyn M, Westermark B (1993) Platelet-derived growth factor (PDGF) in oncogenesis: development of a vascular connective tissue stroma in xenotransplanted human melanoma producing PDGF-BB. Proc Natl Acad Sci U S A 90:393–397. Fredriksson S, Gullberg M, Jarvius J, Olsson C, Pietras K, Gustafsdottir SM, Ostman A Landegren U (2002) Protein detection using proximity-dependent DNA ligation assays. Nat Biotechnol 20:473–477. Fukumura D, Xavier R, Sugiura T, Chen Y, Park EC, Lu N, Selig M, Nielsen G, Taksir T, Jain RK, Seed B (1998) Tumor induction of VEGF promoter activity in stromal cells. Cell 94:715–725. Gerhardt H, Betsholtz C (2003) Endothelial-pericyte interactions in angiogenesis. Cell Tissue Res 314:15–23. Gray MJ, Van Buren G, Dallas NA, Xia L, Wang X, Yang AD, Somcio RJ, Lin YG, Lim S, Fan F, Mangala LS, Arumugam T, Logsdon CD, Lopez-Berestein G, Sood AK, Ellis LM (2008) Therapeutic targeting of neuropilin-2 on colorectal carcinoma cells implanted in the murine liver. J Natl Cancer Inst 100:109–120. Guo P, Hu B, Gu W, Xu L, Wang D, Huang HJ, Cavenee WK, Cheng SY (2003) Platelet-derived growth factor-B enhances glioma angiogenesis by stimulating vascular endothelial growth factor expression in tumor endothelia and by promoting pericyte recruitment. Am J Pathol 162:1083–1093. Hartmann JT, Haap M, Kopp HG, Lipp HP (2009) Tyrosine kinase inhibitors – A review on pharmacology, metabolism and side effects. Curr Drug Metab 10:470–481. Heldin CH, Ostman A, Ronnstrand L (1998) Signal transduction via platelet-derived growth factor receptors. Biochim Biophys Acta 1378:F79–F113. Heldin CH, Westermark B (1999) Mechanism of action and in vivo role of platelet-derived growth factor. Physiol Rev 79:1283–1316. Hellstrom AC, Tegerstedt G, Silfversward C, Pettersson F (1999a) Malignant mixed mullerian tumors of the ovary: histopathologic and clinical review of 36 cases. Int J Gynecol Cancer 9:312–316. Hellstrom M, Gerhardt H, Kalen M, Li X, Eriksson U, Wolburg H, Betsholtz C (2001) Lack of pericytes leads to endothelial hyperplasia and abnormal vascular morphogenesis. J Cell Biol 153:543–553. Hellstrom M, Kalen M, Lindahl P, Abramsson A, Betsholtz C (1999b) Role of PDGF-B and PDGFR-beta in recruitment of vascular smooth muscle cells and pericytes during embryonic blood vessel formation in the mouse. Development 3047–3055. Hirschi KK, D’Amore PA (1996) Pericytes in the microvasculature. Cardiovasc Res 32:687–698. Hopkins TG, Marples M, Stark D (2008) Sunitinib in the management of gastrointestinal stromal tumours (GISTs). Eur J Surg Oncol 34:844–850.

322

K. Matsuo et al.

Jain RK (2005) Normalization of tumor vasculature: an emerging concept in antiangiogenic therapy. Science 307:58–62. Jain RK, Booth MF (2003) What brings pericytes to tumor vessels? J Clin Invest 112:1134–1136. Jain RK, Duda DG, Clark JW, Loeffler JS (2006) Lessons from phase III clinical trials on antiVEGF therapy for cancer. Nat Clin Pract Oncol 3:24–40. Jayson GC, Parker GJ, Mullamitha S, Valle JW, Saunders M, Broughton L, Lawrance J, Carrington B, Roberts C, Issa B, Buckley DL, Cheung S, Davies K, Watson Y, Zinkewich-Peotti K, Rolfe L, Jackson A (2005) Blockade of platelet-derived growth factor receptor-beta by CDP860, a humanized, PEGylated di-Fab’, leads to fluid accumulation and is associated with increased tumor vascularized volume. J Clin Oncol 23:973–981. Johnson JP, Bruce JN (1997) Angiogenesis in human gliomas: prognostic and therapeutic implications. EXS 79:29–46. Kelly JD, Haldeman BA, Grant FJ, Murray MJ, Seifert RA, Bowen-Pope DF, Cooper JA, Kazlauskas A (1991) Platelet-derived growth factor (PDGF) stimulates PDGF receptor subunit dimerization and intersubunit trans-phosphorylation. J Biol Chem 266:8987–8992. Kerbel RS (2008) Tumor angiogenesis. N Engl J Med 358:2039–2049. Lindahl P, Bostrom H, Karlsson L, Hellstrom M, Kalen M, Betsholtz C (1999) Role of plateletderived growth factors in angiogenesis and alveogenesis. Curr Top Pathol 93:27–33. Lindahl P, Johansson BR, Leveen P, Betsholtz C (1997) Pericyte loss and microaneurysm formation in PDGF-B-deficient mice. Science 277:242–245. Lindblom P, Gerhardt H, Liebner S, Abramsson A, Enge M, Hellstrom M, Backstrom G, Fredriksson S, Landegren U, Nystrom HC, Bergstrom G, Dejana E, Ostman A, Lindahl P, Betsholtz C (2003) Endothelial PDGF-B retention is required for proper investment of pericytes in the microvessel wall. Genes Dev 17:1835–1840. Lindblom P, Valdemarsson S, Westerdahl J, Tennvall J, Bergenfelz A (1999) Hyperthyroidism after surgery for primary hyperparathyroidism. Langenbecks Arch Surg 384:568–575. Lu C, Kamat AA, Lin YG, Merritt WM, Landen CN, Kim TJ, Spannuth W, Arumugam T, Han LY, Jennings NB, Logsdon C, Jaffe RB, Coleman RL, Sood AK (2007) Dual targeting of endothelial cells and pericytes in antivascular therapy for ovarian carcinoma. Clin Cancer Res 13:4209–4217. Lu C, Thaker PH, Lin YG, Spannuth W, Landen CN, Merritt WM, Jennings NB, Langley RR, Gershenson DM, Yancopoulos GD, Ellis LM, Jaffe RB, Coleman RL, Sood AK (2008) Impact of vessel maturation on antiangiogenic therapy in ovarian cancer. Am J Obstet Gynecol 198:477 e471–e479; discussion 477 e479–e410. Mancuso MR, Davis R, Norberg SM, O’Brien S, Sennino B, Nakahara T, Yao VJ, Inai T, Brooks P, Freimark B, Shalinsky DR, Hu-Lowe DD, McDonald DM (2006) Rapid vascular regrowth in tumors after reversal of VEGF inhibition. J Clin Invest 116:2610–2621. McCarty MF, Wey J, Stoeltzing O, Liu W, Fan F, Bucana C, Mansfield PF, Ryan AJ, Ellis LM (2004) ZD6474, a vascular endothelial growth factor receptor tyrosine kinase inhibitor with additional activity against epidermal growth factor receptor tyrosine kinase, inhibits orthotopic growth and angiogenesis of gastric cancer. Mol Cancer Ther 3:1041–1048. Morikawa S, Baluk P, Kaidoh T, Haskell A, Jain RK, McDonald DM (2002) Abnormalities in pericytes on blood vessels and endothelial sprouts in tumors. Am J Pathol 160:985–1000. Naumov GN, Akslen LA, Folkman J (2006) Role of angiogenesis in human tumor dormancy: animal models of the angiogenic switch. Cell Cycle 5:1779–1787. Nehls V, Drenckhahn D (1993) The versatility of microvascular pericytes: from mesenchyme to smooth muscle? Histochemistry 99:1–12. Pan Q, Chanthery Y, Liang WC, Stawicki S, Mak J, Rathore N, Tong RK, Kowalski J, Yee SF, Pacheco G, Ross S, Cheng Z, Le Couter J, Plowman G, Peale F, Koch AW, Wu Y, Bagri A, Tessier-Lavigne M Watts RJ (2007) Blocking neuropilin-1 function has an additive effect with anti-VEGF to inhibit tumor growth. Cancer Cell 11:53–67. Patyna S, Laird AD, Mendel DB, O’Farrell M, Liang A,C, Guan H, Vojkovsky T, Vasile S, Wang X, Chen J, Grazzini M, Yang CY, Haznedar JO, Sukbuntherng J, Zhong WZ, Cherrington JM, Hu-Lowe D (2006) SU14813: a novel multiple receptor tyrosine kinase inhibitor with potent antiangiogenic and antitumor activity. Mol Cancer Ther 5:1774–1782.

15  Role of Pericytes in Resistance to Antiangiogenic Therapy

323

Pietras K, Hanahan D (2005) A multitargeted, metronomic, and maximum-tolerated dose “chemoswitch” regimen is antiangiogenic, producing objective responses and survival benefit in a mouse model of cancer. J Clin Oncol 23:939–952. Pietras K, Ostman A, Sjoquist M, Buchdunger E, Reed RK, Heldin CH, Rubin K (2001) Inhibition of platelet-derived growth factor receptors reduces interstitial hypertension and increases transcapillary transport in tumors. Cancer Res 61:2929–2934. Reinmuth N, Liu W, Jung YD, Ahmad SA, Shaheen RM, Fan F, Bucana CD, McMahon G, Gallick GE, Ellis LM (2001) Induction of VEGF in perivascular cells defines a potential paracrine mechanism for endothelial cell survival. FASEB J 15:1239–1241. Schlingemann RO, Rietveld FJ, de Waal RM, Ferrone S, Ruiter DJ (1990) Expression of the high molecular weight melanoma-associated antigen by pericytes during angiogenesis in tumors and in healing wounds. Am J Pathol 136:1393–1405. Schlingemann RO, Rietveld FJ, Kwaspen F, van de Kerkhof PC, de Waal RM, Ruiter DJ (1991) Differential expression of markers for endothelial cells, pericytes, and basal lamina in the microvasculature of tumors and granulation tissue. Am J Pathol 138:1335–1347. Semenza GL (2003) Angiogenesis in ischemic and neoplastic disorders. Annu Rev Med 54:17–28. Senger DR, Galli SJ, Dvorak AM, Perruzzi CA, Harvey VS, Dvorak HF (1983) Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites fluid. Science 219:983–985. Sennino B, Falcon BL, McCauley D, Le T, McCauley T, Kurz JC, Haskell A, Epstein DM, McDonald DM (2007) Sequential loss of tumor vessel pericytes and endothelial cells after inhibition of platelet-derived growth factor B by selective aptamer AX102. Cancer Res 67:7358–7367. Shaheen RM, Tseng WW, Davis DW, Liu W, Reinmuth N, Vellagas R, Wieczorek AA, Ogura Y, McConkey DJ, Drazan KE, Bucana CD, McMahon G, Ellis LM (2001) Tyrosine kinase inhibition of multiple angiogenic growth factor receptors improves survival in mice bearing colon cancer liver metastases by inhibition of endothelial cell survival mechanisms. Cancer Res 61:1464–1468. Shahzad MM, Lopez-Berestein G, Sood AK (2009) Novel strategies for reversing platinum resistance. Drug Resist Updat 12:148–152. Sims DE (1986) The pericyte – a review. Tissue Cell 18:153–174. Sims DE, Miller FN, Horne MM, Edwards MJ (1994) Interleukin-2 alters the positions of capillary and venule pericytes in rat cremaster muscle. J Submicrosc Cytol Pathol 26:507–513. Spannuth WA, Sood AK, Coleman RL (2008) Angiogenesis as a strategic target for ovarian cancer therapy. Nat Clin Pract Oncol 5:194–204. Verhoeven D, Buyssens N (1988) Desmin-positive stellate cells associated with angiogenesis in a tumour and non-tumour system. Virchows Arch B Cell Pathol Incl Mol Pathol 54:263–272. von Tell D, Armulik A, Betsholtz C (2006) Pericytes and vascular stability. Exp Cell Res 312:623–629. Vrancken Peeters MP, Gittenberger-de Groot AC, Mentink MM, Poelmann RE (1999) Smooth muscle cells and fibroblasts of the coronary arteries derive from epithelial-mesenchymal transformation of the epicardium. Anat Embryol (Berl) 199:367–378. Wakui S, Furusato M, Muto T, Ohshige H, Takahashi H, Ushigome S (1997) Transforming growth factor-beta and urokinase plasminogen activator presents at endothelial cell-pericyte interdigitation in human granulation tissue. Microvasc Res 54:262–269. Wesseling P, Schlingemann RO, Rietveld FJ, Link M, Burger PC, Ruiter DJ (1995) Early and extensive contribution of pericytes/vascular smooth muscle cells to microvascular proliferation in glioblastoma multiforme: an immuno-light and immuno-electron microscopic study. J Neuropathol Exp Neurol 54:304–310. Westermark B, Heldin CH (1991) Platelet-derived growth factor in autocrine transformation. Cancer Res 51:5087–5092. Xian X, Hakansson J, Stahlberg A, Lindblom P, Betsholtz C, Gerhardt H, Semb H (2006) Pericytes limit tumor cell metastasis. J Clin Invest 116:642–651.

Chapter 16

Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas Urszula M. Polanska, Kieran T. Mellody, and Akira Orimo

Abstract  Carcinomas are complex tissues comprised of neoplastic cells and a non-cancerous compartment referred to as the “stroma.” The stroma consists of an extracellular matrix (ECM) and a variety of mesenchymal cells, notably including fibroblasts, myofibroblasts, endothelial cells, pericytes and leukocytes. During tumourigenesis, the tumour-associated stroma is continuously exposed to substantial paracrine signals released by nearby carcinoma cells, and is often populated by considerable numbers of myofibroblasts. These cells are a hallmark of “activated fibroblasts” that are commonly observed in injured and fibrotic tissue. Importantly, their presence in large numbers within the stroma of human carcinomas is related to high-grade malignancies and poor prognoses in patients. Tumour-derived stromal myofibroblast-rich cell populations propagated in vitro stably maintain their myofibroblastic state, as well as an ability to significantly promote tumour growth in xenograft models. Differentiation of stromal cells into myofibroblasts within the tumour corroborates the evolution of the normal stroma towards a tumour-promoting stroma. However, the selective pressures responsible for instigating the generation of an altered stroma, and the molecular alterations that stably maintain the unique tumour-promoting myofibroblastic phenotype remain unclear. This chapter highlights the biological role of the tumour-associated stroma, with a particular focus on myofibroblasts and their ability to promote tumour progression through their interactions with carcinoma cells.

Introduction The epithelium is the tissue composed of highly organised cells that line the cavities and surfaces of structures in organs. The epithelium in the mammary gland, for example, consists of an outer layer of myoepithelial cells and an inner layer of A. Orimo () CR-UK Stromal-Tumour Interaction Group, Paterson Institute for Cancer Research, The University of Manchester, Wilmslow Road, Manchester M20 4BX, UK e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_16, © Springer Science+Business Media, LLC 2010

325

326

U.M. Polanska et al.

Fig. 16.1  Histology shows differences between stroma of the normal mammary gland and of a breast carcinoma. Paraffin sections prepared from an invasive breast cancer region (c, f ) and a non-tumour region (b, e) dissected from the breast tissue of a patient, and the normal breast region (a, d) prepared from a healthy donor, were stained with hematoxylin and eosin (H&E) (a–c) or immunostained with an anti-a-smooth muscle actin (a-SMA) antibody (d–f). Large numbers of myofibroblasts (indicated by arrows) in the cancer region (f) and myoepithelial cells (indicated by an arrow in d and e) surrounding luminal epithelial cells are shown. Scale bar, 75 mm [from Orimo et al. (2005)]

luminal epithelial cells attached to a basement membrane (Fig. 16.1a–e). The epithelium is surrounded by an adjacent stroma that consists of mesenchymal cells and a specialised ECM that helps maintain tissue integrity and homeostasis (Fig. 16.1a, b). Not only is the stroma important for cell differentiation and proliferation during tissue development, but it also plays a major role in wound healing, fibrosis and tumourigenesis. The malignant transformation of epithelial cells depends on genetic and/or epigenetic alterations that accumulate within certain oncogenes and/or tumour suppressor genes [reviewed in Hanahan and Weinberg (2000) and Vogelstein and Kinzler (2004)]. These perturbations alone, however, are not sufficient for the carcinoma to fully develop into a high-grade malignancy [for reviews, see DePinho (2000), Bissell and Radisky (2001), Coussens and Werb (2002) and Mueller and Fusenig (2004)]. Malignant epithelial cells require additional support from the surrounding stroma that evolves alongside and promotes tumourigenic progression [reviewed in Rowley (1998), Cunha et al. (2003), Bhowmick et al. (2004b), Lorusso and Ruegg (2008) and Polyak et al. (2009)]. The presence of large numbers of stromal cells and a dense ECM is characteristic of the stroma found in many types of carcinomas, including breast cancer, and is often referred to as a “desmoplastic” or “reactive” stroma (Fig. 16.1c, f). The desmoplastic stroma is believed to play a major role in promoting the growth and progression of the carcinoma [reviewed in Ronnov-Jessen et al. (1996), Bissell and Radisky (2001), Elenbaas and Weinberg (2001), Desmouliere et al. (2004), Kalluri and Zeisberg (2006), De Wever et al. (2008) and Shimoda et al. (2009)].

16  Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas

327

Fibroblasts and myofibroblasts, collectively designated carcinoma-associated fibroblasts (CAFs), are one of the most significant cell types found within the desmoplastic stroma. CAFs interact with cancer cells and collaborate with other components of the stroma through their production and secretion of various growth factors, cytokines and chemokines. During tumourigenesis, signalling molecules secreted by these cells effectively mediate neoangiogenesis, carcinoma cell proliferation, survival, motility and invasion [reviewed in Ronnov-Jessen et al. (1996), De Wever and Mareel (2003), Kalluri and Zeisberg (2006), Orimo and Weinberg (2006) and Ostman and Augsten (2009)].

Myofibroblasts Involved in Tissue Fibrosis Share Characteristics with Tumour-Associated Myofibroblasts Myofibroblasts exhibit a highly contractile phenotype through their expression of a-smooth muscle actin (a-SMA), which assembles into a robust network of stress fibers. These cells share characteristics with smooth muscle cells and pericytes [Desmouliere et al. 1993; reviewed in Serini and Gabbiani (1999)]. Myofibroblasts are commonly found in injured tissues (Gabbiani et  al. 1971), and play essential roles in regulating the healing response by facilitating contraction of the wound. They also secrete a number of soluble growth factors, pro-inflammatory cytokines, chemokines and matrix metalloproteinases (MMPs), which act together in concert to induce angiogenesis and further facilitate tissue repair [reviewed in Powell et al. (1999) and Ronnov-Jessen et al. (1996)]. Myofibroblasts are activated and proliferate in injured tissues to promote wound healing. These cells are cleared from the tissue once the healing process is complete. Myofibroblasts are also found in a number of pathological fibrotic diseases including rheumatoid arthritis, Dupuytren’s disease, idiopathic pulmonary fibrosis, scleroderma, pulmonary hypertension and arteriosclerosis [Kim et  al. 2009; reviewed in Varga (2002), Leask (2006), Varga and Abraham (2007) and Ihn (2008)]. In contrast to the normal healing response, myofibroblasts in fibrotic tissues continue to persist in an activated state. Their presence perpetuates the inflammatory process resulting in an improper tissue repair response that disrupts organ and tissue function. Several studies have identified crucial roles for the pleiotropic cytokine TGF-b in promoting fibrosis [for review, see Varga and Abraham (2007)]. TGF-b promotes mesenchymal cell proliferation, migration, differentiation, adhesion, survival, cytokine induction and ECM protein synthesis [reviewed in Derynck et al. (2001), Bierie and Moses (2006) and Massague (2008)]. It also induces and activates many profibrotic gene signalling pathways in stromal cells such as PDGF signalling that promotes the proliferation and survival of myofibroblasts during fibrogenesis [reviewed in Bonner (2004)]. This cytokine also induces expression level of endothelin-1 (ET-1) which signals through the ETA and ETB receptors. Binding of ET1 to the ETA and ETB receptors activates the JNK pathway that increases further expression of endogenous ET-1 (Shi-Wen et  al. 2006). The resulting autocrine

328

U.M. Polanska et al.

signalling loop contributes to the persistence of the myofibroblastic phenotype in myofibroblasts derived from scleroderma (Shi-Wen et al. 2006). In addition, TGF-b induces Jagged1-Notch autocrine signalling that is involved in myofibroblast differentiation during fibrosis (Liu et al. 2009). Importantly, myofibroblasts extracted from pathological fibrotic tissues, frequently show the establishment of a TGF-b-Smad autocrine signalling loop that is stably maintained during propagation of these cells in vitro [Kim et al. 2009; reviewed in Varga (2002)]. Myofibroblasts, derived from fibrotic tissues, secrete high levels of TGF-b and show upregulated expression levels of TGF-b receptors. This results in a constitutive activation of the TGF-b-Smad signalling pathway in an autocrine fashion that helps to maintain the myofibroblastic phenotype. Inhibition of this pathway not only substantially abates the myofibroblastic phenotype, as demonstrated using both in vitro and in vivo models of fibrosis, but is also associated with decreased ECM deposition and suppression of the inflammatory cytokine response [reviewed in Varga and Abraham (2007) and Varga and Pasche (2009)]. However, the mechanism(s) by which TGF-b autocrine signalling is initially triggered and programmed in these cells remain unclear. One possibility is that sustained exposure of stromal cells to chronic inflammation, and other selective pressures, results in altered gene expression and epigenetic modifications that generate the myofibroblastic phenotype. The tissue remodelling processes involved in wound healing and fibrosis physiologically resemble those occurring within the tumour-associated stroma [reviewed in Dvorak (1986) and Schafer and Werner (2008)]. An important study highlighted the ability of activated fibroblasts, derived from injured tissue, to actively promote tumourigenesis (Hu et al. 2009). Such fibroblasts, extracted from an inflamed rheumatoid arthritis patient, were co-injected with carcinoma cells into a recipient mouse. These activated fibroblasts increased levels of cyclooxygenase-2 (COX-2) expression in the carcinoma cells, an enzyme that is involved in inflammationassociated tumourigenesis and which boosts tumour progression. The carcinoma exploits the host’s wound healing response in order to promote tumourigenesis [reviewed in Bissell and Radisky (2001), Furlow (2005), Kalluri and Zeisberg (2006) and Orimo and Weinberg (2006)]. However, it is not known if molecular signalling, such as TGF-b autocrine signalling, which plays a central role in myofibrogenesis during fibrosis, also mediates the tumour-promoting myofibroblastic phenotype observed in tumour-associated myofibroblasts.

Carcinoma-Associated Fibroblast Characterised as Tumour-Promoting Myofibroblasts The presence of myofibroblasts in large numbers in the stroma of human tumours (Fig. 16.1c, f) is often associated with high-grade malignancies and poor prognoses in patients (Cardone et  al. 1997; Maeshima et  al. 2002; Tsujino et  al. 2007; Kellermann et al. 2008). In order to examine the functional role of tumour-associated myofibroblasts during tumourigenesis, CAFs have been isolated from various different

16  Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas

329

types of human carcinomas including carcinomas of the breast (Allinen et al. 2004; Orimo et al. 2005; Lebret et al. 2007; Casey et al. 2008; Mercier et al. 2008), prostate (Olumi et al. 1999; Ao et al. 2007), ovary (Yang et al. 2006), pancreas (Hwang et al. 2008), skin (Sneddon et al. 2006) and esophagus (Zhang et al. 2009). Our own observations indicate that CAFs extracted from a human breast tumour mass exhibit a different morphology in culture compared to “counterpart fibroblasts” extracted from a noncancerous region of the breast in the same individual or “normal fibroblasts” isolated from a healthy donor (Fig. 16.2a). CAFs appear more slender and spindly and upon reaching confluency, they often lose their contact inhibition of cell growth, continuing to proliferate and stack on top of one another (Fig. 16.2a). These observations are consistent with the findings of an earlier pioneer study that characterised CAFs isolated from human prostate carcinomas (Olumi et al. 1999). Primary cultured fibroblasts, extracted from human normal breast tissues, are a heterogeneous cell population that contains a small number of cells exhibiting a myofibroblastic characteristic, as determined by a-SMA expression. However, CAFs contain a noticeably larger proportion of myofibroblasts compared to those present within counterpart and normal fibroblasts (Fig. 16.2b) (Ronnov-Jessen et al. 1992; Orimo et al. 2005). Moreover, CAFs in comparison with control fibroblasts, when embedded within a collagen gel, display a more contractile phenotype that is indicative of the activated fibroblastic trait in myofibroblasts (Fig. 16.2c). They are also able to retain the activated phenotype during propagation in vitro for at least ten population doublings (PDs), without further ongoing interaction with carcinoma cells (Fig. 16.2c). Taken together, these observations demonstrate that CAFs include increased proportions of myofibroblasts that stably maintain an activated fibroblast phenotype. Genetic analyses of CAFs have previously been performed to determine if these cells are themselves neoplastic. Karyotype analysis to date has not detected any chromosomal abnormalities in CAFs (Olumi et  al. 1999; Orimo et  al. 2005). Consistently, genome-wide genetic analyses have also failed to detect any significant loss of heterozygosity (LOH) in these cells (Allinen et al. 2004; Qiu et al. 2008; Walter et al. 2008); (discussed in “Carcinoma-Associated Fibroblast Characterised as Tumour-Promoting Myofibroblasts” section). CAFs exhibit no anchorage-independent growth ability in vitro and no tumourigenicity in vivo when injected alone, without carcinoma cells, into a recipient mouse (Fig. 16.3a–c). They also undergo cellular senescence in vitro after 15 PDs (Orimo et al. 2005). Collectively, these findings strongly suggest that CAFs are intrinsically non-tumourigenic. CAFs, however, can promote tumour growth through their interactions with carcinoma cells. When injected along with human breast carcinoma MCF-7-ras cells subcutaneously into an immunodeficient mouse, CAFs show an ability to more significantly increase the rate and size of tumour growth compared to control counterpart and normal fibroblasts (Orimo et al. 2005). Importantly, CAFs retain such a tumour-promoting phenotype until, at least, ten PDs in the absence of continuous interaction with carcinoma cells (Orimo et al. 2005). Several other independent studies have also confirmed the ability of CAFs extracted from a number of different human carcinomas to promote tumourigenesis

330

a

U.M. Polanska et al. i

Normal fibroblasts

iii

CAFs

b

Normal fibroblasts

iii

Counterpart fibroblasts

iv

i

α-SMA Ab

ii

Counterpart fibroblasts

iv

ii

CAFs

c

α-SMA Ab

α-SMA Ab

Confluent

d

CAFs

CAFs

Control IgG

180 170 160

Normal fibroblasts-2

Counterpart fibroblasts

Area of collagen gels (mm2)

Normal fibroblasts-1

150 140 130 120 110 100 90 80

Normal fibroblast-1 Normal fibroblast-2 Counterpart fibroblasts CAFs (5 PDs) CAFs (10 PDs)

70 60

CAFs

50

0

10

20

30

40

hr

50

60

70

80

Fig. 16.2  CAFs include a larger proportion of myofibroblasts. (a) Normal fibroblasts (i) were extracted from a reduction mammoplasty performed on a healthy donor. Counterpart fibroblasts (ii) were also extracted from a noncancerous region of the breast at least 2 cm from the outer tumour margin. In addition, CAFs (iii, iv) were extracted from an invasive ductal breast carcinoma region in the same individual. CAFs (iv) show a more spindle-like morphology when they reach near confluency compared with normal (i) and counterpart fibroblasts (ii) (scale bar 50 mm). (b) An increased proportion of a-SMA-positive myofibroblasts are observed in CAFs. Immunofluorescence of normal fibroblasts (i), counterpart fibroblasts (ii) or CAFs (iii) was performed using an antia-SMA antibody. CAFs (iv) were stained with control IgG as a negative control (scale bar: 50 mm); a–d from Orimo et al. (2005). (c) Equal numbers of various fibroblasts were incubated in a collagen gel for 72 h. CAFs exhibit a greater ability to contract collagen gels compared to normal or counterpart fibroblasts. The graph shows the rate of contraction of the collagen gels for each group of fibroblasts tested. Error bars indicate standard deviation of the mean [from Orimo et al. (2005)]

(Olumi et  al. 1999; Yang et  al. 2006; Hwang et  al. 2008; Hu et  al. 2009). Immunohistochemical analysis using a human-specific anti-vimentin antibody, which detects human mesenchymal cells but not MCF-7-ras carcinoma cells or the mouse host cells, shows that either the injected CAFs or control fibroblasts are present in considerable numbers within the advanced tumour xenografts (Fig. 16.3d) (Orimo et  al. 2005). Taken together, these studies suggest that CAFs promote

16  Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas

331

Fig. 16.3  CAFs themselves are not neoplastic but are capable of promoting tumour growth through their interactions with apposed carcinoma cells. (a) Various human mammary fibroblasts, including CAFs, show no anchorage-independent growth activity when propagated in vitro. Tissue sections from mice inoculated with CAFs alone without cancer cells were stained with hematoxylin and eosin (b), and with a human-specific anti-vimentin antibody and hematoxylin (c). (d) CAFs survive alongside carcinoma cells within the advanced tumour xenografts. Immunohistochemical analysis of a paraffin section prepared from a 62-day-old tumour with antibodies against human vimentin (brown) and GFP (pink), allowing specific detection of human vimentin-expressing CAFs and GFP-labelled MCF-7-ras carcinoma cells, respectively. Nuclear staining was performed using hemotoxylin (blue) [from Orimo et al. (2005)]. Scale bars: 100 mm (a) and 50 mm (b–d)

tumour growth via their ongoing interactions with the apposed carcinoma cells throughout tumour progression. Moreover, these fibroblasts are able to retain their stable tumour-promoting properties in a cell-autonomous fashion.

Somatic Genetic and Epigenetic Alterations in Tumour-Associated Stroma The CAFs are non-neoplastic cells capable of promoting tumourigenesis through their interactions with carcinoma cells. However, there is much debate as to whether tumour-associated stromal cells harbour somatic genetic alterations [reviewed in

332

U.M. Polanska et al.

Weinberg (2008b), Haviv et al. (2009) and Polyak et al. (2009)]. Furthermore, the molecular mechanism(s) responsible for maintaining the stable tumour-promoting CAF phenotype also remain unclear. A number of studies support the presence of somatic genetic alterations, including LOH and mutations, in stromal regions microdissected from various human carcinomas, including those of the breast (Moinfar et  al. 2000; Wernert et  al. 2001; Kurose et al. 2002; Patocs et al. 2007), ovary (Tuhkanen et al. 2004), colon (Wernert et al. 2001), bladder (Paterson et al. 2003) and head and neck (Weber et al. 2007). Indeed, 25.6% in the hereditary group and 19.4% in the sporadic group of breast cancer patients show somatic p53 mutations in microdissected tumour-associated stromal regions (Patocs et al. 2007). In addition, there is a higher risk associated with regional lymph node metastases in these patients. In prostatic cell-specific oncogene-driven transgenic mice with a p53+/- background, the developing tumour puts selective pressure on the tumour-associated stroma. This results in frequent loss of the wild type p53 allele within the stroma (Hill et  al. 2005). Carcinoma cell-derived paracrine signals may act upon the adjacent stroma to give rise to p53-negative highly proliferative mesenchymal cells. However, whilst alterations in the p53 gene may confer upon stromal cells their tumour-promoting properties, it is unclear whether such alterations do give rise to the myofibroblast-rich desmoplastic stroma often seen in human carcinomas. Furthermore, patients with Li-Fraumeni syndrome are prone to developing soft tissue and bone sarcomas due to inherent p53 alterations, as do p53 knockout transgenic mice. However, it is not known why breast cancer patients, who may lose p53 function within the tumourassociated stroma, very rarely develop tumours of mesenchymal origin (Bolton and Sieunarine 1990; Tokudome et al. 2005). Most of the studies that detected genetic alterations in the tumour stromal compartments were performed using paraffin-fixed tumour tissues. Conversely, genome-wide genetic analyses, including single nucleotide polymorphism and comparative genomic hybridisation arrays, performed using freshly frozen tissues derived from human breast carcinomas, fail to detect any significant genetic alterations in cancer-associated stroma (Qiu et  al. 2008). This was highlighted in a recent study that reported only one chromosomal LOH event among 35 samples of cancer-associated stromal regions dissected from different breast and ovarian tumours [Qiu et  al. 2008; reviewed in Haviv et  al. (2009)]. Similar analyses performed using CAFs have also failed to detect any significant genetic alterations, as described earlier (Allinen et al. 2004; Qiu et al. 2008; Walter et al. 2008). Therefore, detection of genetic alterations in some studies, whilst not in others, may be in part due to the different methods used to acquire and prepare tissues for analysis, the use of paraffin-fixed tissues instead of frozen tissues or the different types of genomic analyses performed [commented in Weinberg (2008b) and reviewed in Haviv et  al. (2009)]. Future independent analyses are therefore required to clarify these issues. The presence of altered profiles of epigenetic modifications, such as DNA methylation within the genome of CAFs, has also been reported (Hu et al. 2005; Fiegl et al. 2006; Hanson et al. 2006; Jiang et al. 2008). In addition, CAFs, unlike

16  Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas

333

normal fibroblasts, are able to confer epigenetic modifications upon apposed human breast epithelial cells. This was demonstrated in an important study in which CAFs were co-cultured for three weeks with human breast epithelial cells in an assay that allowed cell-to-cell contacts to form (Lin et al. 2008). CAFs activated the AKT signalling pathway in the epithelial cells, which hypermethylated the promoter region for the tumour suppressor gene cystatin M. Future studies are required to determine if the CAF-induced epigenetic modifications are stably reprogrammed within the epithelial cells, and if this, in turn, promotes tumourigenesis in  vivo. Understanding what epigenetic modifications in CAFs mediate their phenotype, and how these cells confer epigenetic alterations upon neighbouring epithelial cells, will further highlight the molecular mechanisms that underlie their tumour-promoting properties.

Heterogeneous Cellular Origins of Carcinoma-Associated Myofibroblasts Human dermal fibroblasts are heterogeneous in nature and often display locationspecific gene expression signatures (Chang et al. 2002). Similarly, tumour-associated stromal fibroblasts are also considered to be a heterogeneous cell population (Sugimoto et  al. 2006; Anderberg and Pietras 2009). Although the expression of a-SMA defines the myofibroblastic phenotype, to date, no specific protein markers have been identified that determine the cellular origins of myofibroblasts found in the stroma of tumour. These cells are therefore characterised by expression of a number of mesenchymal protein markers that include tenascin-C, fibroblast specific protein-1, fibroblast activation protein, PDGF receptor-a/b, NG2 chondroitin sulfate proteoglycan, vimentin and prolyl 4-hydroxylase [Sugimoto et  al. 2006; reviewed in Kalluri and Zeisberg (2006), De Wever et al. (2008) and Anderberg and Pietras (2009)]. Various cell types are thought to act as a source for the emergence of tumourpromoting CAF myofibroblasts (Fig. 16.4). Resident fibroblasts are likely a major source of myofibroblasts [reviewed in Powell et al. (1999) and Hinz (2007)], whilst other mesenchymal cell types including endothelial cells, pericytes, smooth muscle cells and preadipocytes are also capable of converting into myofibroblasts [Zeisberg et al. 2007; reviewed in Kalluri and Zeisberg (2006) and De Wever et al. (2008)]. Bone marrow-derived progenitors, such as fibrocytes and mesenchymal stem cells (MSCs), are also reported to differentiate into myofibroblasts within the tumour. It has been shown, for example, that bone marrow-derived cells contribute to the generation of nearly 40% of myofibroblasts present in the tumour-associated stroma of 28-day-old pancreatic tumour xenografts (Ishii et  al. 2003), and of approximately 25% in spontaneous pancreatic tumours developing for 16–18 weeks in another transgenic mouse model (Direkze et al. 2004). It is also possible that a small number of myofibroblasts present within the normal fibroblast populations are clonally expanded in response to selective pressure imposed by the tumour

334

U.M. Polanska et al.

Endothelial cells

Myofibroblasts

Fibroblasts

Fibroblasts with genetic alterations (e.g., p53)

?

Pericytes

Differentiation Selection Differentiation/ Selection

Myofibroblasts BMDCs/MSCs

blood vessels

Tumour progression

Fibroblasts cancer cells

Carcinoma in situ Invasive carcinoma Fig. 16.4  Schematic representation of cellular origins of CAF myofibroblasts during tumour progression; CAFs are proposed to be derived from various cell types, such as pre-existing fibroblasts, pericytes or endothelial cells. Bone marrow-derived progenitor cells (BMDCs) including MSCs and fibrocytes also differentiate into myofibroblasts. A small population of residual myofibroblasts or mesenchymal cells that carry genetic alterations may be clonally selected during tumour progression. The genetically altered mesenchymal cells may or may not differentiate into myofibroblasts

microenvironment. In addition, the stromal cells that have acquired epigenetic or genetic alterations may differentiate into myofibroblasts and these cells are then also likely to undergo clonal expansion. The tumour stroma continues to remodel itself during tumour progression and actively recruits various cell types into the tumour mass where they act as different sources for myofibroblasts. It is not known, however, if different cells of origins in the myofibroblast populations in the stroma of tumour exhibit different tumourpromoting properties. Once generated, myofibroblasts maintain their ability to substantially promote tumour growth and progression in many aspects via their interactions with carcinoma cells and other host stromal cells.

16  Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas

335

Normal Stroma-Derived Tumour-Suppressive Signalling and Tumour Stroma-Derived Tumour-Promoting Signalling TGF-b signalling functions as a potent tumour suppressor in epithelial cells during the early stages of tumourigenesis. However, the role of this signalling pathway within the stroma is unclear. A study using a transgenic mouse model, in which TGF-b type II receptor (TbRII) gene is deleted specifically in stromal fibroblasts, has demonstrated the inherent tumour-suppressing effects of stromal TGF-b signalling on epithelial cells (Bhowmick et  al. 2004a). The TbRII-deficient fibroblasts secrete high levels of hepatocyte growth factor (HGF) that causes spontaneous malignant transformation of c-Met receptor-expressing forestomach epithelial cells and tumourigenesis within the mice. More recently, the tumour-suppressive role of Notch signalling within the stroma has been demonstrated in a transgenic mouse model. The chimeric deletion of the Notch 1 gene, within the epidermis of a conditional Notch 1flox/flox mouse, causes a wound-like microenvironment, as characterised by the infiltration of myofibroblasts and leukocytes into the affected tissue (Demehri et  al. 2009). These cells secrete elevated levels of TGF-b, keratinocyte growth factor and stromal cell-derived factor 1 (SDF-1, also called CXCL12). This altered microenvironment, rich in growth factors and cytokines, promotes chemical carcinogeninduced skin tumourigenesis of Notch 1-expressing keratinocytes in a non-cell autonomous manner. The inherent tumour-suppressive role of stromal Notch signalling is therefore important in maintaining tissue integrity and homeostasis within the skin. In addition to TGF-b and Notch signalling, phosphatase and tensin homologue (PTEN) signalling within the stroma plays an important function in inhibiting tumourigenesis. Myofibroblasts in lung tissues from idiopathic pulmonary fibrosis patients show decreased levels of PTEN expression (White et al. 2006). Inhibition of PTEN also increases numbers of myofibroblasts observed in bleomycin-induced fibrotic lung (White et al. 2006). Furthermore, suppression of PTEN expression and inhibition of its activity stimulate differentiation of cultured fibroblasts into myofibroblasts. Indeed, the loss of PTEN expression in mammary stromal fibroblasts within a transgenic mouse model accelerates the initiation and progression of ErbB2-driven breast epithelial cancers (Trimboli et al. 2009). The ErbB2 tumours grow more rigorously in the presence of PTEN-deficient stromal fibroblasts and show a highly desmoplastic stromal reaction, as determined by an increase in the numbers of infiltrating macrophages. However, it is not known if the PTENdeficient stromal fibroblasts per se are a source of tumour-associated myofibroblasts in the stroma. In addition, a gene expression profile obtained from PTEN-null mammary stromal fibroblasts indicates upregulation of genes predominantly involved in ECM remodelling, inflammation and fibrosis. These observations further highlight the inherent tumour-suppressive role of the normal stroma, and specifically stromal PTEN, TGF-b and Notch signalling pathways, involved in suppressing tumourigenesis.

336

U.M. Polanska et al.

Through its interactions with carcinoma cells, the initially tumour-suppressing stroma, however, overtime acquires an ability to promote tumourigenesis. SDF-1 is an important proangiogenic chemokine that is known to actively regulate tumourigenesis [reviewed in Burger and Kipps (2006), Nagasawa (2006), Ratajczak et al. (2006), Kryczek et  al. (2007) and Petit et  al. (2007)]. Unlike fibroblasts found within the non-cancerous stroma, myofibroblasts present within the tumour stroma actively secrete SDF-1 (Allinen et al. 2004; Orimo et al. 2005; Ao et al. 2007; Tait et al. 2007). This chemokine, upon binding to its cognate receptor CXCR4 constitutively expressed on the cell surface of carcinoma cells, stimulates carcinoma cell proliferation (Hall and Korach 2003; Allinen et al. 2004; Orimo et al. 2005). In the later stages of tumour progression, TGF-b signalling in epithelial cancer cells also promotes tumourigenesis. A recent study showed that elevated secretion of TGF-b by CAFs (San Francisco et  al. 2004; Rosenthal et  al. 2004; Ao et  al. 2007), upregulates the expression levels of CXCR4 in prostatic epithelial cells. This in turn boosts stromal SDF-1-stimulated tumourigenesis in vivo (Ao et al. 2007). TGF-b within the stroma therefore also plays a permissive role in promoting some aspects of tumourigenesis. The Hedgehog (Hh) signaling is a key regulator of embryonic development but has also been implicated in the development of many different type of cancers [Karhadkar et al. 2004; reviewed in Taipale and Beachy (2001)]. The activity of the transmembrane receptor Smoothened (Smo) is constitutively suppressed by the unbound Patched 1 (PTCH1) receptor, the cognate receptor for Hh ligands. However, subsequent binding of Hh ligands to PTCH1 reverses this effect, and thus activates the Smo signalling pathway. Recent studies have demonstrated that ligand-dependent activation of the Hh pathway in the tumour-associated stroma plays an important role in influencing tumourigenesis [reviewed in Theunissen and de Sauvage (2009)]. For example, the human pancreatic carcinoma cell-secreted Sonic hedgehog (Shh) ligand has been shown to induce a desmoplastic stromal reaction within the tumour (Bailey et al. 2008). Exposure of cultured pancreatic stellate cells to Shh also facilitated their differentiation into myofibroblasts (Bailey et al. 2008). Moreover, the expression levels of both PTCH1 and Gli1, a downstream target of the Smo signalling pathways, are elevated in the stromal region of human pancreatic carcinomas (Tian et  al. 2009). Hh ligands, however, are expressed within the epithelial carcinoma compartment. Importantly, inhibition of Smo protein activity on the tumour-associated stromal cells using a potent, small molecular antagonist of the Hh pathway attenuated growth of human colon adenocarcinoma cells in a tumour xenograft model (Yauch et  al. 2008). These findings demonstrate that the Hh ligands secreted by carcinoma cells instigate tumour-promoting Hh signalling in nearby stromal cells in a paracrine fashion, through binding of the PTCH1 receptor expressed by stromal cells. However, the contribution of soluble paracrine factors, released by the tumourpromoting stroma in response to the activation of Hh signalling, to tumourigenesis remains unknown. A recent study highlighted an ability of Hh signalling within the tumour-associated stroma to affect drug delivery in cancer (Olive et al. 2009). Systemic administration

16  Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas

337

of a Smo inhibitor into a murine model of pancreatic cancer, which expresses endogenous mutant K-ras and p53 alleles, inhibits Hh signalling. This alters the composition of the desmoplastic stroma and attenuates proliferation of stromal myofibroblasts. The authors also claim that the associated increase of the tumour vascular density facilitates efficient delivery of the chemotherapeutic agent into the tumour. As another possible mechanism, the increase in drug delivery may be due to decreased interstitial fluid pressure that is normally instigated by the tumourassociated stroma with high Hh signalling. This may prevent transport of the drug from the vasculature into the tumour tissue [for reviews, see Jain (2001) and Pietras et al. (2003)]. Stromal fibroblasts interact with carcinoma epithelial cells within the tumour, yet this relationship between the fibroblasts and carcinoma cells is reciprocal. It is likely that carcinoma cells initially secrete cell signalling molecules (e.g., TGF-b), which can induce the conversion of fibroblasts and/or their progenitors into myofibroblasts. In vitro experiments have demonstrated that stromal fibroblasts exposed to medium conditioned by carcinoma cells can undergo differentiation into myofibroblasts (Guo et al. 2008; Kellermann et al. 2008; Noma et al. 2008). Importantly, the ability of stromal fibroblasts to promote tumourigenesis and tumour progression is substantially associated with myofibroblast differentiation [De Wever et al. 2004; Orimo et  al. 2005; reviewed in Desmouliere et  al. (2004) and De Wever et  al. (2008)]. Carcinoma cells can induce myofibroblast differentiation of the apposed stromal cells and the resulting myofibroblasts can in turn reciprocate to promote growth and progression of carcinoma cells. The development of the malignant tumour and promotion towards advanced stages of carcinomas require the successful co-evolution of both cancer cells and stromal cells [reviewed in De Wever et al. (2008), Weinberg (2008b), Polyak et al. (2009) and Shimoda et al. (2009)].

Tumour-Associated Stroma Promotes Neoangiogenesis The CAF-promoted neoangiogenesis is mediated by a number of proangiogenic cytokines and chemokines including SDF-1, vascular endothelial growth factor-A (VEGF-A) and basic fibroblast growth factor 2 (FGF2) (Fig. 16.5a) (Orimo et al. 2005; Guo et al. 2008; Noma et al. 2008; Pietras et al. 2008). The elevated levels of VEGF promoter activity have been detected in stromal fibroblasts present in tumours developing within transgenic mice expressing VEGF promoter-driven GFP (Fukumura et al. 1998). Other studies have shown that inhibition of stroma cell-derived VEGF significantly decreases the growth rate of fibrosarcoma cells (Dong et  al. 2004). In addition, TGF-b-primed myofibroblasts secrete increased levels of VEGF that stimulates vascular formation by endothelial cells co-cultured in a 3D assay (Noma et al. 2008). These studies further support the pivotal roles of stromal VEGF in stimulating neoangiogenesis. A recent study has reported that tumour-associated stromal fibroblasts not only stimulate tumour angiogenesis but also mediate resistance to anti-angiogenic therapy.

338

U.M. Polanska et al.

Fibroblast-enriched cell fractions extracted from the mouse lymphomas, which were resistant to anti-VEGFA antibody treatment, upregulated PDGF-C expression by 200-fold compared to fibroblasts isolated from non-refractory tumours (Crawford et al. 2009). This increase in stromal PDGF-C expression in the therapy-resistant tumours was attributed to an increase in neoangiogenesis occurring during the course of anti-VEGF treatment. Similarly, in a cervical squamous cell carcinoma

a TGFβ PDGF Shh

HGF SDF-1 IGF-2 Gremlin-1 SFRP-1

proliferation

FGF2 VEGF-A SDF-1 Endothelial cells/EPCs TGFβ SDF-1 Tenascin C and W HGF MMPs

angiogenesis

SFRP1

invasion

survival

Fig. 16.5  Tumour-promoting effects exerted by CAFs. (a) Tumour-promoting paracrine signalling exerted by tumour-associated myofibroblasts: HGF (Tokunou et al. 2001; De Wever et al. 2004), SDF-1 (Orimo et  al. 2005; Daly et  al. 2008), IGF-2 (insulin-like growth factors 2) (Zhu et  al. 2007), Gremlin-1 (Sneddon et al. 2006), SFRP1 (secreted frizzled-related protein) (Joesting et al. 2005), FGF2 (fibroblasts growth factor 2) (Pietras et  al. 2008), VEGF-A (vascular endothelial growth factor A) (Fukumura et al. 1998), TGF-b (Weinberg 2007), MMPs, tenascin C (De Wever et al. 2004) and tenascin W (Degen et al. 2007). Carcinoma-derived signalling molecules causing myofibroblast differentiation and proliferation: TGF-b (Desmouliere et al. 1993; Ronnov-Jessen et al. 1996; Serini and Gabbiani 1999; Hinz et al. 2001), PDGF (Bostrom et al. 1996; Jester et al. 2002; Bonner 2004), Sonic hedgehog (Shh) (Bailey et  al. 2008; Olive et  al. 2009). (b) CAFs stimulate neoangiogenesis; Human breast MCF-7-ras tumours that developed in the presence of CAFs are highly angiogenic. Sections prepared from MCF-7-ras tumour xenografts containing various fibroblasts were stained with anti-CD31 antibody to identify endothelial cells (iv, v, vi) or by Masson’s trichrome (i, ii, iii), which stains collagen in blue, red blood cells in red and cancer cells in purple. Tumours, developed in the presence of CAFs, are rich in collagen (iii) and have a highly vascularised stroma (iii, vi). Scale bar: 100 mm; (c) Human breast MCF-7-ras tumours developed in the presence of CAFs contain increased numbers of Sca1+CD31+ endothelial progenitor cells (EPCs) compared to tumours derived from MCF-7-ras admixed with control fibroblasts. The advanced tumours were dissociated into a single cell suspension and stained with anti-Sca1 and anti-CD31 antibodies; p<0.05 [from Orimo et al. (2005)]

16  Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas

b

+Normal fibroblasts

i

339

+CAFs

+Counterpart fibroblasts

ii

iii

v

vi

capillary vessels

Masson’s trichrome

iv

CD31 Ab

c

0.9

n=8

Sca1+/CD31+ EPCs in tumour (%)

0.8 0.7 0.6 0.5 0.4 0.3 n=7

0.2 0.1 0

Fig. 16.5  (continued)

capillary vessels

*

n=6 n=6

340

U.M. Polanska et al.

transgenic mouse model, treatment with the PDGF receptor kinase inhibitor also resulted in suppression of FGF2 and FGF7 expression in tumour stromal fibroblasts, which attenuated neoangiogenesis and tumour growth (Pietras et al. 2008). The increased level of SDF-1 secreted by CAFs also substantially promotes neoangiogenesis as it actively recruits circulating endothelial progenitor cells (EPCs) into the primary tumour (Orimo et al. 2005) (Fig. 16.5c). These cells are able to differentiate into vascular endothelial cells that form tumour-associated capillaries. Also, varieties of leukocytes including neutrophils, macrophages, CD11b+Gr-1+myeloid-derived suppressor cells (MDSCs) and lymphocytes are often recruited into the tumour-associated stroma. Experimental data derived from both in vitro and in vivo studies demonstrate crucial roles for these tumourassociated leukocytes in boosting neoangiogenesis and tumour progression [DeNardo et al. 2009; reviewed in Condeelis and Pollard (2006), de Visser et al. (2006) and Mantovani et al. (2008)]. Importantly, the tumour microenvironment polarises tumour-associated macrophages towards either a pro-tumour or an anti-tumour phenotype (Sica et al. 2008). Recent literature also indicates that the immunosuppressive cytokine TGF-b, abundantly produced within the tumour microenvironment, plays an essential role in switching the polarity of tumourassociated neutrophils towards a pro-tumourigenic phenotype (Fridlender et  al. 2009). These studies highlight an important role for the tumour microenvironment in promoting tumour progression by regulating the polarity of tumour-associated leukocytes.

Roles of Tumour-Associated Stroma in Promoting Cancer Cell Invasion and Metastasis Invasion and metastasis are responsible for 90% of cancer-associated mortality, and the majority of cancer cells at the time of death are often found in metastases rather than in the primary tumour (Weinberg 2007). The tumour invasion-metastasis cascade is a complex multi-step process that includes localised invasion of carcinoma cells, entrance into the systemic circulation, survival during transportation, extravasation, the establishment of micrometastases in distal tissues and colonisation to form macroscopic metastases [reviewed by Fidler (2003)]. It has long been assumed that dissemination of metastatic carcinoma cells depends largely on their cell-autonomous effects due to epigenetic and/or genetic alterations that accumulate within these malignant cells. However, emerging evidence now proposes a different schema in which metastatic spread is determined relatively early on in tumour progression and is not totally dependant on the acquisition of additional genetic alterations within carcinoma cells [reviewed in Weinberg (2008a) and Klein (2009)]. Two distinct populations of normal mammary epithelial cells, originating from breast tissue within the same individual, upon receiving the same set of genetic aberrations, give rise to different types of carcinoma cells with distinct metastatic traits (Ince et  al. 2007). This suggests that the differentiation

16  Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas

341

program in the parental cell of origin, from which carcinoma cells are derived, is also actively present in the derivative malignant cells. This differentiation program appears to strongly influence the development of their metastatic propensity (Gupta et al. 2005; Ince et al. 2007). The tumour microenvironment likely serves as another important determinant that encourages carcinoma cells in the primary tumour to become motile and invasive, and to disseminate into distant organs. Interaction of carcinoma cells with the tumour-associated stroma facilitates the invasion-metastasis cascade (Karnoub et al. 2007). Tumour-associated stromal fibroblasts play a significant role in regulating such migratory and invasive behaviours in carcinoma cells. This is supported by evidence indicating that stromal myofibroblasts are frequently present at the invasive front of human carcinomas [Nakayama et  al. 1998; Sivridis et  al. 2005; reviewed in De Wever and Mareel (2003)]. In addition, it has been shown that CAF myofibroblasts, extracted from human carcinomas, increase the migratory and invasive propensity of the cancer cells co-cultured alongside them in collagen gels [De Wever et al. 2004; Gaggioli et al. 2007; Jodele et al. 2006; Daly et al. 2008; reviewed in Desmouliere et  al. (2004)]. Tenascin C, HGF and SDF-1, which are secreted by CAFs, also play a role in mediating CAF-stimulated invasion of carcinoma cells (De Wever et al. 2004; Daly et al. 2008) (Fig. 16.5a). A number of additional signalling molecules are involved in initiating and regulating the metastatic spread of cancer cells to distant organs [reviewed in Siegel and Massague (2003) and Nguyen et al. (2009)] (Fig. 16.5a). TGF-b signalling is a key molecular pathway involved in this process. Human cancer cells, interacting with the stroma at the leading edge of the invasive front, often undergo epithelial to mesenchymal transition (EMT), as determined by the downregulation of epithelial markers and upregulation of mesenchymal markers [reviewed in Thiery and Sleeman (2006), Weinberg (2007) and Thiery et  al. (2009)]. Therefore EMT, induced in cancer cells by signals released from the tumour-associated stroma (e.g., TGF-b), may be an early cellular event that promotes the invasion and metastatic dissemination of cancer cells. Cancer cells that undergo EMT lose their cell-to-cell junctions and adopt a more migratory and invasive phenotype. The reverse process, mesenchymal to epithelial transition, allows carcinoma cells that have undergone metastasis to anchor themselves to basement membranes, form cell-to-cell contacts and colonise distal organs and tissues [reviewed in Brabletz et al. (2005), Lee et al. (2006), Thiery and Sleeman (2006), Hugo et al. (2007) and Kalluri and Weinberg (2009)]. A recent study, using in  vivo imaging to monitor activation of TGF-b-Smad signalling in a rat mammary carcinoma xenograft model, found that TGF-b signalling is activated in carcinoma cells rapidly moving alone, but not slowly moving as an aggregate (Giampieri et al. 2009). The tumour-associated stroma, which is rich in TGF-b, may induce activation of the TGF-b signalling in carcinoma cells. Activation of this signalling pathway is likely to increase their motility along with their ability to invade basement membranes and intravasate into blood vessels. In contrast, carcinoma cells that metastasise into distal organs, where the stroma is not rich in TGF-b, lose activation of the TGF-b signalling pathway. These cells therefore

342

U.M. Polanska et al.

lose their migratory phenotype and are able to proliferate and colonise these tissues. Furthermore, primary tumour microenvironment-derived TGF-b primes breast carcinoma cells to elevate expression levels of angiopoietin-like 4, which disassembles the vascular endothelial cell-to-cell junctions and facilitates extravasation into the lung (Padua et al. 2008). Taken together, these observations reinforce the view that the stromal microenvironment in the primary tumour has the ability to influence tumour invasion and metastasis. The concept of a pre-metastatic niche in which non-cancer cells initiate metastasis of carcinoma cells into distal specific organs has recently been proposed (Kaplan et al. 2005). Bone marrow-derived hematopoietic progenitor cells, positive for vascular endothelial growth factor receptor-1 create pre-metastatic niches in a distal organ prior to the arrival of carcinoma cells. The pre-metastatic niche supports recruitment of disseminated carcinoma cells and facilitates the further formation of micro- and macro-metastases [Hiratsuka et  al. 2002; reviewed in Psaila and Lyden (2009)]. Tumour cell-derived paracrine soluble factors such as VEGF-A, placental growth factor, inflammatory S100 chemokines and the extracellular enzyme lysyl oxidase, play key roles in generating the pre-metastatic niche (Kaplan et al. 2005; Hiratsuka et al. 2008; Erler et al. 2009). Some of these factors are also substantially produced by the stroma within the primary tumour. Future studies are therefore likely to help clarify the role of the tumour-associated stroma in generating the pre-metastatic niche. During colonisation of tissues and organs distal to the site of the primary tumour, disseminated metastatic cells establish micro- and macro-metastases. In order to do so, cancer cells actively remodel the surrounding stroma, changing its inherent tumour-suppressing properties. The malignant cells thereby create a hospitable tissue microenvironment that supports their proliferation and the manifestation of metastatic lesions. Interestingly, the actively remodelled stroma often shares a histopathological resemblance with stroma present in the primary tumour.

Conclusions/Perspectives The importance of the tumour-associated stroma in regulating tumour growth and progression is becoming evermore apparent. During tumourigenesis, the stroma evolves in response to signals released by carcinoma cells. Myofibroblast differentiation during tumourigenesis corroborates the evolution of the normal stroma towards a tumour-promoting stroma. CAF myofibroblasts, frequently found within the stroma of many different types of carcinomas, contribute enormously to tumourigenesis. However, many aspects of the mesenchymal–epithelial interactions in the tumour have still to be elucidated. Delineating these complex paracrine and autocrine signalling pathways within the tumour microenvironment will be necessary to fully understand the role of the stroma in instigating and assisting the progression of cancer. This research will

16  Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas

343

also hopefully allow us to understand the mechanisms that initiate the differentiation of cells into myofibroblasts, and the factors involved in the maintenance of their tumour-promoting phenotype. Importantly, recent research highlights noticeable roles for the tumour-associated stroma in promoting invasion and metastasis. Understanding better the function of myofibroblasts during wound healing and tissue fibrosis may provide additional insights into the broad spectrum of stromal signalling that progresses cancer cell growth, invasion and metastasis. Drug resistance often occurs in carcinoma cells that are genetically unstable and thus accumulate adaptive mutations during the course of chemoprevention study. Novel therapies are being developed that target the tumour-promoting signalling in stromal cells that possess a more stable genome compared to carcinoma cells. Whilst the debate continues as to whether the tumour-associated stroma is genetically altered and may develop resistance to chemotherapies, it still remains an attractive therapeutic target in the fight against cancer. Disrupting the interactions that occur between the stroma and carcinoma cells using novel therapies, alone or in combination with conventional treatments that target the carcinoma cells directly, may result in greater efficacy. Drug design may focus on preventing progenitor cells, fibroblasts or other precursory cell types from differentiating into myofibroblasts. Understanding the complex mesenchymal–epithelial interactions will further our understanding of the nature of human cancer. Such research is challenging but holds great promise, particularly with the advent of new pharmacological agents designed to specifically target these stromal–tumour interactions. Acknowledgements  We would like to extend our gratitude to Dr. Robert A. Weinberg, Whitehead Institute for Biomedical Research, for his ongoing guidance and excellent supervision. We also thank Dr. Radoslaw Polanski for the critical reading of this chapter and Cancer research UK for the funding.

References Allinen M, Beroukhim R, Cai L et al (2004) Molecular characterization of the tumor microenvironment in breast cancer. Cancer Cell 6:17–32. Anderberg C, Pietras K (2009) On the origin of cancer-associated fibroblasts. Cell Cycle 8:1461–1462. Ao M, Franco OE, Park D et  al (2007) Cross-talk between paracrine-acting cytokine and chemokine pathways promotes malignancy in benign human prostatic epithelium. Cancer Res 67:4244–4253. Bailey JM, Swanson BJ, Hamada T et al (2008) Sonic hedgehog promotes desmoplasia in pancreatic cancer. Clin Cancer Res 14:5995–6004. Bhowmick NA, Chytil A, Plieth D et al (2004a) TGF-beta signaling in fibroblasts modulates the oncogenic potential of adjacent epithelia. Science 303:848–851. Bhowmick NA, Neilson EG, Moses HL (2004b) Stromal fibroblasts in cancer initiation and progression. Nature 432:332–337. Bierie B, Moses HL (2006) Tumour microenvironment: TGFbeta: the molecular Jekyll and Hyde of cancer. Nat Rev Cancer 6:506–520.

344

U.M. Polanska et al.

Bissell MJ, Radisky D (2001) Putting tumours in context. Nat Rev Cancer 1:46–54. Bolton B, Sieunarine K (1990) Carcinosarcoma: a rare tumour of the breast. Aust N Z J Surg 60:917–919. Bonner JC (2004) Regulation of PDGF and its receptors in fibrotic diseases. Cytokine Growth Factor Rev 15:255–273. Bostrom H, Willetts K, Pekny M et al (1996) PDGF-A signaling is a critical event in lung alveolar myofibroblast development and alveogenesis. Cell 85:863–873. Brabletz T, Hlubek F, Spaderna S et  al (2005) Invasion and metastasis in colorectal cancer: epithelial-mesenchymal transition, mesenchymal-epithelial transition, stem cells and betacatenin. Cells Tissues Organs 179:56–65. Burger JA, Kipps TJ (2006) CXCR4: a key receptor in the crosstalk between tumor cells and their microenvironment. Blood 107:1761–1767. Cardone A, Tolino A, Zarcone R et  al (1997) Prognostic value of desmoplastic reaction and lymphocytic infiltration in the management of breast cancer. Panminerva Med 39:174–177. Casey TM, Eneman J, Crocker A et al (2008) Cancer associated fibroblasts stimulated by transforming growth factor beta1 (TGF-beta 1) increase invasion rate of tumor cells: a population study. Breast Cancer Res Treat 110:39–49. Chang HY, Chi JT, Dudoit S et  al (2002) Diversity, topographic differentiation, and positional memory in human fibroblasts. Proc Natl Acad Sci USA 99:12877–12882. Condeelis J, Pollard JW (2006) Macrophages: obligate partners for tumor cell migration, invasion, and metastasis. Cell 124:263–266. Coussens LM, Werb Z (2002) Inflammation and cancer. Nature 420:860–867. Crawford Y, Kasman I, Yu L et  al (2009) PDGF-C mediates the angiogenic and tumorigenic properties of fibroblasts associated with tumors refractory to anti-VEGF treatment. Cancer Cell 15:21–34. Cunha GR, Hayward SW, Wang YZ et al (2003) Role of the stromal microenvironment in carcinogenesis of the prostate. Int J Cancer 107:1–10. Daly AJ, McIlreavey L, Irwin CR (2008) Regulation of HGF and SDF-1 expression by oral fibroblasts—implications for invasion of oral cancer. Oral Oncol 44:646–651. Degen M, Brellier F, Kain R et al (2007) Tenascin-W is a novel marker for activated tumor stroma in low-grade human breast cancer and influences cell behavior. Cancer Res 67:9169–9179. Demehri S, Turkoz A, Kopan R (2009) Epidermal Notch1 loss promotes skin tumorigenesis by impacting the stromal microenvironment. Cancer Cell 16:55–66. DeNardo DG, Barreto JB, Andreu P et al (2009) CD4(+) T cells regulate pulmonary metastasis of mammary carcinomas by enhancing protumor properties of macrophages. Cancer Cell 16: 91–102. DePinho RA (2000) The age of cancer. Nature 408:248–254. Derynck R, Akhurst RJ, Balmain A (2001) TGF-beta signaling in tumor suppression and cancer progression. Nat Genet 29:117–129. Desmouliere A, Geinoz A, Gabbiani F et  al (1993) Transforming growth factor-beta 1 induces alpha-smooth muscle actin expression in granulation tissue myofibroblasts and in quiescent and growing cultured fibroblasts. J Cell Biol 122:103–111. Desmouliere A, Guyot C, Gabbiani G (2004) The stroma reaction myofibroblast: a key player in the control of tumor cell behavior. Int J Dev Biol 48:509–517. de Visser KE, Eichten A, Coussens LM (2006) Paradoxical roles of the immune system during cancer development. Nat Rev Cancer 6:24–37. De Wever O, Mareel M (2003) Role of tissue stroma in cancer cell invasion. J Pathol 200:429–447. De Wever O, Nguyen QD, Van Hoorde L et  al (2004) Tenascin-C and SF/HGF produced by myofibroblasts in vitro provide convergent proinvasive signals to human colon cancer cells through RhoA and Rac. FASEB J 18:1016–1018. De Wever O, Demetter P, Mareel M et al (2008) Stromal myofibroblasts are drivers of invasive cancer growth. Int J Cancer 123:2229–2238. Direkze NC, Hodivala-Dilke K, Jeffery R et al (2004) Bone marrow contribution to tumor-associated myofibroblasts and fibroblasts. Cancer Res 64:8492–8495.

16  Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas

345

Dong J, Grunstein J, Tejada M et  al (2004) VEGF-null cells require PDGFR alpha signalingmediated stromal fibroblast recruitment for tumorigenesis. EMBO J 23:2800–2810. Dvorak HF (1986) Tumors: wounds that do not heal. Similarities between tumor stroma generation and wound healing. N Engl J Med 315:1650–1659. Elenbaas B, Weinberg RA (2001) Heterotypic signaling between epithelial tumor cells and fibroblasts in carcinoma formation. Exp Cell Res 264:169–184. Erler JT, Bennewith KL, Cox TR et al (2009) Hypoxia-induced lysyl oxidase is a critical mediator of bone marrow cell recruitment to form the premetastatic niche. Cancer Cell 15:35–44. Fidler IJ (2003) The pathogenesis of cancer metastasis: the ‘seed and soil’ hypothesis revisited. Nat Rev Cancer 3:453–458. Fiegl H, Millinger S, Goebel G et  al (2006) Breast cancer DNA methylation profiles in cancer cells and tumor stroma: association with HER-2/neu status in primary breast cancer. Cancer Res 66:29–33. Fridlender ZG, Sun J, Kim S et al (2009) Polarization of tumor-associated neutrophil phenotype by TGF-beta: “N1” versus “N2” TAN. Cancer Cell 16:183–194. Fukumura D, Xavier R, Sugiura T et al (1998) Tumor induction of VEGF promoter activity in stromal cells. Cell 94:715–725. Furlow B (2005) Wound healing processes and tumour growth. Lancet Oncol 6:363. Gabbiani G, Ryan GB, Majne G (1971) Presence of modified fibroblasts in granulation tissue and their possible role in wound contraction. Experientia 27:549–550. Gaggioli C, Hooper S, Hidalgo-Carcedo C et al (2007) Fibroblast-led collective invasion of carcinoma cells with differing roles for RhoGTPases in leading and following cells. Nat Cell Biol 9:1392–1400. Giampieri S, Manning C, Hooper S et  al (2009) Localized and reversible TGFbeta signalling switches breast cancer cells from cohesive to single cell motility. Nat Cell Biol 11:1287–1296. Guo X, Oshima H, Kitmura T et al (2008) Stromal fibroblasts activated by tumor cells promote angiogenesis in mouse gastric cancer. J Biol Chem 283:19864–19871. Gupta PB, Kuperwasser C, Brunet JP et al (2005) The melanocyte differentiation program predisposes to metastasis after neoplastic transformation. Nat Genet 37:1047–1054. Hall JM, Korach KS (2003) Stromal cell-derived factor 1, a novel target of estrogen receptor action, mediates the mitogenic effects of estradiol in ovarian and breast cancer cells. Mol Endocrinol 17:792–803. Hanahan D, Weinberg RA (2000) The hallmarks of cancer. Cell 100:57–70. Hanson JA, Gillespie JW, Grover A et al (2006) Gene promoter methylation in prostate tumorassociated stromal cells. J Natl Cancer Inst 98:255–261. Haviv I, Polyak K, Qiu W et  al (2009) Origin of carcinoma associated fibroblasts. Cell Cycle 8:589–595. Hill R, Song Y, Cardiff RD et al (2005) Selective evolution of stromal mesenchyme with p53 loss in response to epithelial tumorigenesis. Cell 123:1001–1011. Hinz B (2007) Formation and function of the myofibroblast during tissue repair. J Invest Dermatol 127:526–537. Hinz B, Mastrangelo D, Iselin CE et  al (2001) Mechanical tension controls granulation tissue contractile activity and myofibroblast differentiation. Am J Pathol 159:1009–1020. Hiratsuka S, Nakamura K, Iwai S et al (2002) MMP9 induction by vascular endothelial growth factor receptor-1 is involved in lung-specific metastasis. Cancer Cell 2:289–300. Hiratsuka S, Watanabe A, Sakurai Y et al (2008) The S100A8-serum amyloid A3-TLR4 paracrine cascade establishes a pre-metastatic phase. Nat Cell Biol 10:1349–1355. Hu M, Yao J, Cai L et al (2005) Distinct epigenetic changes in the stromal cells of breast cancers. Nat Genet 37:899–905. Hu M, Peluffo G, Chen H et al (2009) Role of COX-2 in epithelial-stromal cell interactions and progression of ductal carcinoma in situ of the breast. Proc Natl Acad Sci USA 106:3372–3377. Hugo H, Ackland ML, Blick T et al (2007) Epithelial-mesenchymal and mesenchymal-epithelial transitions in carcinoma progression. J Cell Physiol 213:374–383.

346

U.M. Polanska et al.

Hwang RF, Moore T, Arumugam T et  al (2008) Cancer-associated stromal fibroblasts promote pancreatic tumor progression. Cancer Res 68:918–926. Ihn H (2008) Autocrine TGF-beta signaling in the pathogenesis of systemic sclerosis. J Dermatol Sci 49:103–113. Ince TA, Richardson AL, Bell GW et al (2007) Transformation of different human breast epithelial cell types leads to distinct tumor phenotypes. Cancer Cell 12:160–170. Ishii G, Sangai T, Oda T et  al (2003) Bone-marrow-derived myofibroblasts contribute to the cancer-induced stromal reaction. Biochem Biophys Res Commun 309:232–240. Jain RK (2001) Delivery of molecular and cellular medicine to solid tumors. Adv Drug Deliv Rev 46:149–168. Jester JV, Huang J, Petroll WM et  al (2002) TGFbeta induced myofibroblast differentiation of rabbit keratocytes requires synergistic TGFbeta, PDGF and integrin signaling. Exp Eye Res 75:645–657. Jiang L, Gonda TA, Gamble MV et al (2008) Global hypomethylation of genomic DNA in cancerassociated myofibroblasts. Cancer Res 68:9900–9908. Jodele S, Blavier L, Yoon JM et al (2006) Modifying the soil to affect the seed: role of stromalderived matrix metalloproteinases in cancer progression. Cancer Metastasis Rev 25:35–43. Joesting MS, Perrin S, Elenbaas B et al (2005) Identification of SFRP1 as a candidate mediator of stromal-to-epithelial signaling in prostate cancer. Cancer Res 65:10423–10430. Kalluri R, Weinberg RA (2009) The basics of epithelial-mesenchymal transition. J Clin Invest 119:1420–1428. Kalluri R, Zeisberg M (2006) Fibroblasts in cancer. Nat Rev Cancer 6:392–401. Kaplan RN, Riba RD, Zacharoulis S et al (2005) VEGFR1-positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche. Nature 438:820–827. Karhadkar SS, Bova GS, Abdallah N et al (2004) Hedgehog signalling in prostate regeneration, neoplasia and metastasis. Nature 431:707–712. Karnoub AE, Dash AB, Vo AP et  al (2007) Mesenchymal stem cells within tumour stroma promote breast cancer metastasis. Nature 449:557–563. Kellermann MG, Sobral LM, da Silva SD et al (2008) Mutual paracrine effects of oral squamous cell carcinoma cells and normal oral fibroblasts: induction of fibroblast to myofibroblast transdifferentiation and modulation of tumor cell proliferation. Oral Oncol 44:509–517. Kim KK, Wei Y, Szekeres C et al (2009) Epithelial cell alpha3beta1 integrin links beta-catenin and Smad signaling to promote myofibroblast formation and pulmonary fibrosis. J Clin Invest 119:213–224. Klein CA (2009) Parallel progression of primary tumours and metastases. Nat Rev Cancer 9:302–312. Kryczek I, Wei S, Keller E et al (2007) Stroma-derived factor (SDF-1/CXCL12) and human tumor pathogenesis. Am J Physiol Cell Physiol 292:C987–C995. Kurose K, Gilley K, Matsumoto S et al (2002) Frequent somatic mutations in PTEN and TP53 are mutually exclusive in the stroma of breast carcinomas. Nat Genet 32:355–357. Leask A (2006) Scar wars: is TGFbeta the phantom menace in scleroderma? Arthritis Res Ther 8:213–219. Lebret SC, Newgreen DF, Thompson EW et  al (2007) Induction of epithelial to mesenchymal transition in PMC42-LA human breast carcinoma cells by carcinoma-associated fibroblast secreted factors. Breast Cancer Res 9:R19. Lee JM, Dedhar S, Kalluri R et al (2006) The epithelial-mesenchymal transition: new insights in signaling, development, and disease. J Cell Biol 172:973–981. Lin HJ, Zuo T, Lin CH et al (2008) Breast cancer-associated fibroblasts confer AKT1-mediated epigenetic silencing of Cystatin M in epithelial cells. Cancer Res 68:10257–10266. Liu T, Hu B, Choi YY et al (2009) Notch1 signaling in FIZZ1 induction of myofibroblast differentiation. Am J Pathol 174:1745–1755. Lorusso G, Ruegg C (2008) The tumor microenvironment and its contribution to tumor evolution toward metastasis. Histochem Cell Biol 130:1091–1103.

16  Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas

347

Maeshima AM, Niki T, Maeshima A et al (2002) Modified scar grade: a prognostic indicator in small peripheral lung adenocarcinoma. Cancer 95:2546–2554. Mantovani A, Allavena P, Sica A et  al (2008) Cancer-related inflammation. Nature 454: 436–444. Massague J (2008) TGFbeta in Cancer. Cell 134:215–230. Mercier I, Casimiro MC, Wang C et al (2008) Human breast cancer-associated fibroblasts (CAFs) show caveolin-1 downregulation and RB tumor suppressor functional inactivation: Implications for the response to hormonal therapy. Cancer Biol Ther 7:1212–1225. Moinfar F, Man YG, Arnould L et al (2000) Concurrent and independent genetic alterations in the stromal and epithelial cells of mammary carcinoma: implications for tumorigenesis. Cancer Res 60:2562–2566. Mueller MM, Fusenig NE (2004) Friends or foes —bipolar effects of the tumour stroma in cancer. Nat Rev Cancer 4:839–849. Nagasawa T (2006) Microenvironmental niches in the bone marrow required for B-cell development. Nat Rev Immunol 6:107–116. Nakayama H, Enzan H, Miyazaki E et al (1998) The role of myofibroblasts at the tumor border of invasive colorectal adenocarcinomas. Jpn J Clin Oncol 28:615–620. Nguyen DX, Bos PD, Massague J (2009) Metastasis: from dissemination to organ-specific colonization. Nat Rev Cancer 9:274–284. Noma K, Smalley KS, Lioni M et  al (2008) The essential role of fibroblasts in esophageal squamous cell carcinoma-induced angiogenesis. Gastroenterology 134:1981–1993. Olive KP, Jacobetz MA, Davidson CJ et al (2009) Inhibition of hedgehog signaling enhances delivery of chemotherapy in a mouse model of pancreatic cancer. Science 324: 1457–1461. Olumi AF, Grossfeld GD, Hayward SW et  al (1999) Carcinoma-associated fibroblasts direct tumor progression of initiated human prostatic epithelium. Cancer Res 59:5002–5011. Orimo A, Weinberg RA (2006) Stromal fibroblasts in cancer: a novel tumor-promoting cell type. Cell Cycle 5:1597–1601. Orimo A, Gupta PB, Sgroi DC et al (2005) Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion. Cell 121:335–348. Ostman A, Augsten M (2009) Cancer-associated fibroblasts and tumor growth—bystanders turning into key players. Curr Opin Genet Dev 19:67–73. Padua D, Zhang XH, Wang Q et al (2008) TGFbeta primes breast tumors for lung metastasis seeding through angiopoietin-like 4. Cell 133:66–77. Paterson RF, Ulbright TM, MacLennan GT et al (2003) Molecular genetic alterations in the lasercapture-microdissected stroma adjacent to bladder carcinoma. Cancer 98:1830–1836. Patocs A, Zhang L, Xu Y et al (2007) Breast-cancer stromal cells with TP53 mutations and nodal metastases. N Engl J Med 357:2543–2551. Petit I, Jin D, Rafii S (2007) The SDF-1-CXCR4 signaling pathway: a molecular hub modulating neo-angiogenesis. Trends Immunol 28:299–307. Pietras K, Sjöblom T, Rubin K et al (2003) PDGF receptors as cancer drug targets. Cancer Cell 3:439–443. Pietras K, Pahler J, Bergers G et al (2008) Functions of paracrine PDGF signaling in the proangiogenic tumor stroma revealed by pharmacological targeting. PLoS Med 5:e19. Polyak K, Haviv I, Campbell IG (2009) Co-evolution of tumor cells and their microenvironment. Trends Genet 25:30–38. Powell DW, Mifflin RC, Valentich JD et al (1999) Myofibroblasts. I. Paracrine cells important in health and disease. Am J Physiol 277:C1–C9. Psaila B, Lyden D (2009) The metastatic niche: adapting the foreign soil. Nat Rev Cancer 9:285–293. Qiu W, Hu M, Sridhar A et al (2008) No evidence of clonal somatic genetic alterations in cancerassociated fibroblasts from human breast and ovarian carcinomas. Nat Genet 40:650–655.

348

U.M. Polanska et al.

Ratajczak MZ, Zuba-Surma E, Kucia M et al (2006) The pleiotropic effects of the SDF-1-CXCR4 axis in organogenesis, regeneration and tumorigenesis. Leukemia 20:1915–1924. Ronnov-Jessen L, Van Deurs B, Nielsen M et al (1992) Identification, paracrine generation, and possible function of human breast carcinoma myofibroblasts in culture. In Vitro Cell Dev Biol 28A:273–283. Ronnov-Jessen L, Petersen OW, Bissell MJ (1996) Cellular changes involved in conversion of normal to malignant breast: importance of the stromal reaction. Physiol Rev 76:69–125. Rosenthal E, McCrory A, Talbert M et al (2004) Elevated expression of TGF-beta1 in head and neck cancer-associated fibroblasts. Mol Carcinog 40:116–121. Rowley DR (1998) What might a stromal response mean to prostate cancer progression? Cancer Metastasis Rev 17:411–419. San Francisco IF, DeWolf WC, Peehl DM et al (2004) Expression of transforming growth factorbeta 1 and growth in soft agar differentiate prostate carcinoma-associated fibroblasts from normal prostate fibroblasts. Int J Cancer 112:213–218. Schafer M, Werner S (2008) Cancer as an overhealing wound: an old hypothesis revisited. Nat Rev Mol Cell Biol 9:628–638. Serini G, Gabbiani G (1999) Mechanisms of myofibroblast activity and phenotypic modulation. Exp Cell Res 250:273–283. Shimoda M, Mellody KT, Orimo A (2010) Carcinoma-associated fibroblasts are a rate-limiting determinant for tumour progression. Semin Cell Dev Biol 21:19–25. Shi-Wen X, Rodriguez-Pascual F, Lamas S et  al (2006) Constitutive ALK5-independent c-Jun N-terminal kinase activation contributes to endothelin-1 overexpression in pulmonary fibrosis: evidence of an autocrine endothelin loop operating through the endothelin A and B receptors. Mol Cell Biol 26:5518–5527. Sica A, Larghi P, Mancino A et al (2008) Macrophage polarization in tumour progression. Semin Cancer Biol 18:349–355. Siegel PM, Massague J (2003) Cytostatic and apoptotic actions of TGF-beta in homeostasis and cancer. Nat Rev Cancer 3:807–821. Sivridis E, Giatromanolaki A, Koukourakis MI (2005) Proliferating fibroblasts at the invading tumour edge of colorectal adenocarcinomas are associated with endogenous markers of hypoxia, acidity, and oxidative stress. J Clin Pathol 58:1033–1038. Sneddon JB, Zhen HH, Montgomery K et al (2006) Bone morphogenetic protein antagonist gremlin 1 is widely expressed by cancer-associated stromal cells and can promote tumor cell proliferation. Proc Natl Acad Sci USA 103:14842–14847. Sugimoto H, Mundel TM, Kieran MW et al (2006) Identification of fibroblast heterogeneity in the tumor microenvironment. Cancer Biol Ther 5:1640–1646. Taipale J, Beachy PA (2001) The Hedgehog and Wnt signalling pathways in cancer. Nature 411:349–354. Tait LR, Pauley RJ, Santner SJ et  al (2007) Dynamic stromal-epithelial interactions during progression of MCF10DCIS.com xenografts. Int J Cancer 120:2127–2134. Theunissen JW, de Sauvage FJ (2009) Paracrine Hedgehog signaling in cancer. Cancer Res 69:6007–6010. Thiery JP, Sleeman JP (2006) Complex networks orchestrate epithelial-mesenchymal transitions. Nat Rev Mol Cell Biol 7:131–142. Thiery JP, Acloque H, Huang RJ et al (2009) Epithelial-mesenchymal transitions in development and disease. Cell 139:871–890. Tian H, Callahan CA, DuPree KJ et  al (2009) Hedgehog signaling is restricted to the stromal compartment during pancreatic carcinogenesis. Proc Natl Acad Sci USA 106:4254–4259. Tokudome N, Sakamoto G, Sakai T et al (2005) A case of carcinosarcoma of the breast. Breast Cancer 12:149–153. Tokunou M, Niki T, Eguchi K et al (2001) c-MET expression in myofibroblasts: role in autocrine activation and prognostic significance in lung adenocarcinoma. Am J Pathol 158:1451–1463. Trimboli AJ, Cantemir-Stone CZ, Li F et  al (2009) Pten in stromal fibroblasts suppresses mammary epithelial tumours. Nature 461:1084–1091.

16  Tumour-Promoting Stromal Myofibroblasts in Human Carcinomas

349

Tsujino T, Seshimo I, Yamamoto H et al (2007) Stromal myofibroblasts predict disease recurrence for colorectal cancer. Clin Cancer Res 13:2082–2090. Tuhkanen H, Anttila M, Kosma VM et al (2004) Genetic alterations in the peritumoral stromal cells of malignant and borderline epithelial ovarian tumors as indicated by allelic imbalance on chromosome 3p. Int J Cancer 109:247–252. Varga J (2002) Scleroderma and Smads: dysfunctional Smad family dynamics culminating in fibrosis. Arthritis Rheum 46:1703–1713. Varga J, Abraham D (2007) Systemic sclerosis: a prototypic multisystem fibrotic disorder. J Clin Invest 117:557–567. Varga J, Pasche B (2009) Transforming growth factor beta as a therapeutic target in systemic sclerosis. Nat Rev Rheumatol 5:200–206. Vogelstein B, Kinzler KW (2004) Cancer genes and the pathways they control. Nat Med 10:789–799. Walter K, Omura N, Hong SM et al (2008) Pancreatic cancer associated fibroblasts display normal allelotypes. Cancer Biol Ther 7:882–888. Weber F, Xu Y, Zhang L et al (2007) Microenvironmental genomic alterations and clinicopathological behavior in head and neck squamous cell carcinoma. JAMA 297:187–195. Weinberg RA (2007) The biology of cancer. Garland Science, New York, pp 34, 584–585, 605–614, 652. Weinberg RA (2008a) Mechanisms of malignant progression. Carcinogenesis 29:1092–1095. Weinberg RA (2008b) Coevolution in the tumor microenvironment. Nat Genet 40:494–495. Wernert N, Locherbach C, Wellmann A et al (2001) Presence of genetic alterations in microdissected stroma of human colon and breast cancers. Anticancer Res 21:2259–2264. White ES, Atrasz RG, Hu B et al (2006) Negative regulation of myofibroblast differentiation by PTEN (phosphatase and tensin homolog deleted on chromosome 10). Am J Respir Crit Care Med 173:112–121. Yang G, Rosen DG, Zhang Z et al (2006) The chemokine growth-regulated oncogene 1 (Gro-1) links RAS signaling to the senescence of stromal fibroblasts and ovarian tumorigenesis. Proc Natl Acad Sci USA 103:16472–16477. Yauch RL, Gould SE, Scales SJ et al (2008) A paracrine requirement for hedgehog signalling in cancer. Nature 455:406–410. Zeisberg EM, Potenta S, Xie L et al (2007) Discovery of endothelial to mesenchymal transition as a source for carcinoma-associated fibroblasts. Cancer Res 67:10123–10128. Zhang C, Fu L, Fu J et al (2009) Fibroblast growth factor receptor 2-positive fibroblasts provide a suitable microenvironment for tumor development and progression in esophageal carcinoma. Clin Cancer Res 15:4017–4027. Zhu CQ, Popova SN, Brown ER et  al (2007) Integrin alpha 11 regulates IGF2 expression in fibroblasts to enhance tumorigenicity of human non-small-cell lung cancer cells. Proc Natl Acad Sci USA 104:11754–11759.

Part IV

Immune-Mediated Cells

Chapter 17

Mast Cells and Tumor Microenvironment Theoharis C. Theoharides, Konstantinos-Dionysios Alysandratos, Asimenia Angelidou, and Bodi Zhang

Abstract  Increasing evidence indicates that a unique immune cell, the mast cell, accumulates in the stroma surrounding certain tumors, especially mammary and pancreatic adenocarcinoma, as well as melanoma. Many molecules secreted by mast cells could benefit the tumor in at least four ways: (1) angiogenin, heparin and vascular endothelial growth factor (VEGF), which induce neovascularization; (2) proteases that disrupt the surrounding matrix and facilitate metastases; (3) growth factors such as, epidermal growth factor (EGF), nerve growth factor (NGF), platelet derived growth factor (PDGF) and stem cell factor (SCF); (4) histamine, IL-10 and transforming growth factor-b (TGF-b), which are immunosuppressant, along with activation of certain dendritic cells that induce immunologic anergy. These actions could only occur through the unique ability of mast cells to release certain mediators selectively without degranulation. Blocking such release of pro-tumor mediators may constitute a novel therapeutic approach. Abbreviations BBB CRH CT CTMC

blood-brain-barrier cortocotropin-releasing hormone tryptase and chymase mast cells connective tissue mast cells

T.C. Theoharides (*) Molecular Immunopharmacology and Drug Development Laboratory, Department of Pharmacology and Experimental Therapeutics, Tufts University School of Medicine and Tufts Medical Center, 136 Harrison Avenue, Boston, MA 02111, USA and Department of Biochemistry, Tufts University School of Medicine and Tufts Medical Center, Boston, MA, USA and Department of Internal Medicine, Tufts University School of Medicine and Tufts Medical Center, Boston, MA, USA e-mail: [email protected]

R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_17, © Springer Science+Business Media, LLC 2010

353

354

DMBA EGF hCBMCs HDC HMC-1 IFN-a IFN-g MDSCs MMC MMP-9 NGF NMU NO NSCLC NT PAR-1 and -2 PDAC PDGF RBL-1 SCF SCLC SP T mast cells TAMs TGF-b TNF TRAIL TSLP VEGF VPF

T.C. Theoharides et al.

7, 12-dimethylbenz(a)anthracene epidermal growth factor human umbilical cord blood-derived cultured mast cells histidine decarboxylase human leukemic mast cells interferon-a interferon-g myeloid-derived stem cells mucosal mast cells metalloproteinase-9 nerve growth factor nitrosomethylurea nitric oxide non-small cell lung carcinomas neurotensin protease-activated receptors pancreatic ductal adenocarcinoma platelet-derived growth factor rat basophil leukemia cells stem cell factor small cell lung carcinomas substance P tryptase mast cells tumor-associated macrophages transforming growth factor-b tumor necrosis factor TNF-related apoptosis-related ligand thymic stromal lymphopoietin vascular endothelial growth factor vascular permeability factor

Introduction Despite substantial resources invested in basic cancer research, mortality rates for the most frequent forms of cancer have not decreased significantly. Metastasis facilitated by stromal proteolytic enzymes (Almholt and Johnsen 2003) and chemokines (Murphy 2001) remains the chief cause of morbidity and mortality. The stroma surrounding the tumor is increasingly acquiring importance for its growth and dissemination, with infiltrating inflammatory cells actually contributing to cancer proliferation (Mantovani et  al. 2002). For instance, tumor-associated macrophages (TAMs) and tumor-associated fibroblasts can be beneficial to tumor angiogenesis and growth (Silzle et al. 2003; Yu and Rak 2003) through secretion of vascular endothelial growth factor (VEGF) (Barbera-Guillem et  al. 2002) and platelet-derived growth factor (PDGF) (Kataki et  al. 2002). In one model of

17  Mast Cells and Tumor Microenvironment

355

subcutaneous melanoma, both angiogenesis and growth rate correlated with tumor infiltration by macrophages that expressed angiotensin II type 1 receptor and VEGF (Egami et al. 2003). TAMs were also significantly correlated with squamous cell carcinoma invasion (Li et al. 2002). Mast cells are important in allergic and late phase reactions, but also in inflammation and T-cell mediated immune responses (Mekori and Metcalfe 2000; Redegeld and Nijkamp 2003; Pedotti et  al. 2003). Yet, mast cells appear to be recruited by tumors and accumulate in the stroma (Theoharides 1988) (Fig. 17.1). Mast cells could be helpful to the tumor, but only if secretion of beneficial molecules could occur selectively without degranulation (Theoharides et al. 2007). In fact, the tumor stroma microenvironment could alter the phenotypic behavior of mast cells. For instance, acidity created by rapid cancer cell proliferation inhibits mast cell degranulation, but enhances IL-4 production (Frossi et al. 2003). Nitric oxide (NO) generated by new vessel growth inhibits mast cell degranulation (Coleman 2002), as do oxidized polyamines secreted by the tumor (Vliagoftis et al. 1992). Mast cells can promote tumor development by: (a) disturbing the normal stroma-epithelium

Fig. 17.1  Schematic representation of the possible role of increased number of mast cells in the stroma of certain tumors. Mast cells could be recruited by tumor-derived chemoattractants such as adrenomedullin, MCP-1, RANTES and SCF, to selectively secrete molecules beneficial to the tumor; these could include growth factors, histamine which is mitogenic (H1) and immunosuppressant (H2), neovascularization agents such as heparin, VEGF and IL-8, as well as proteases that could permit new blood vessel formation and metastases

356

T.C. Theoharides et al.

communication as was shown for matrix degradation at sites of tumor invasion in rat mammary adenocarcinoma, (b) facilitating tumor angiogenesis, (c) releasing growth factors (Conti et al. 2007), and (d) inducing a state of immunosuppression. The tumor enhancing effect of mast cells has been shown repeatedly with the use of W/Wv mast cell deficient mice, which developed fewer lung metastases to subcutaneous B16-BL6 tumors (Starkey et al. 1988), and in which mice premalignant angiogenesis of squamous epithelial carcinogenesis was blocked (Coussens et al. 1999). The development of 1, 2-dimethylhydrazine-induced intestinal tumors was slowed by 60% in W/Wv mice (Wedemeyer and Galli 2005). There was also reduced microvessel formation and tumor size in W/Wv mice injected with MB49 murine bladder carcinoma (Dethelfsen et al. 1994).

Mast Cell Biology Mast cells derive from a specific bone marrow progenitor cell, they migrate into tissues where they mature depending upon microenvironmental conditions, and they participate in allergic reactions, as well as innate and acquired immunity (Mekori and Metcalfe 2000; Galli et al. 2005a, b). Mast cells are located perivascularly close to neurons and could have a critical role in neuroinflammatory diseases (Theoharides and Cochrane 2004), as well as in stress-induced brain metastases (Theoharides et al. 2008). Mast cells vary considerably in their cytokine and proteolytic enzyme content: connective tissue mast cells (CTMC) contain tryptase and chymase (CT mast cells), while mucosal mast cells (MMC) contain only tryptase (T mast cells). However, the phenotypic expression of mast cells is not fixed, since MMC can develop into CTMC given the appropriate microenvironmental conditions (Galli et al. 2005a). Moreover, addition of IL-5 to human umbilical cord blood-derived cultured mast cells (hCBMCs) augmented IgE-induced production of distinct cytokines, such as tumor necrosis factor (TNF) and MIP-1a, but without histamine (Ochi et al. 2000). In addition to IgE and antigen, the main trigger in allergic reactions, anaphylatoxins (C3a, C5a), cytokines (IL-1, IL-33), hormones (CRH) and neuropeptides can stimulate mast cell activation; the latter include endorphins, substance P (SP), neurotensin (NT), and nerve growth factor (NGF) leading to secretion of numerous biologically active mediators, but through different pathways (Theoharides and Kalogeromitros 2006). In addition, thymic stromal lymphopoietin (TSLP) released from epithelial cells in response to infection, trauma, and inflammation activates mast cells in the absence of IgE, but in the presence of IL-1, to release IL-5 and IL-13 (Allakhverdi et al. 2007; Al-Shami et al. 2005). Mast cells can secrete either the content of individual granules (Theoharides and Douglas 1978) or distinct mediators selectively (Theoharides et  al. 1982), possibly through regulation by specific phosphoproteins (Sieghart et  al. 1978; Theoharides et  al. 1980). Vascular permeability/vascular endothelial cell growth factor (VPF/VEGF) can be secreted from bone marrow-derived mouse mast cells

17  Mast Cells and Tumor Microenvironment

357

(Boesiger et al. 1998). In view of the fact that acute stress increased tumor size and decreased survival (Sklar and Anisman 1979; Antoni et al. 2006), we investigated if cortocotropin-releasing hormone (CRH), secreted under stress, could induce VEGF release from hCBMCs. We reported that CRH induced selective VEGF release without histamine (Cao et  al. 2005). We further showed that IL-1 could induce selective secretion of IL-6 from hCBMCs without degranulation through a unique vesicular shuttle (Kandere-Grzybowska et  al. 2003). IL-1 can further stimulate secretion of VEGF (Salven et  al. 2002), thus promoting angiogenesis (Salven et al. 2002) and lung carcinoma growth (Saijo et al. 2002). Stem cell factor (SCF) can also induce selective release of IL-6 without histamine and without degranulation (Gagari et  al. 1997). This process has been termed “differential release”, “intragranular activation” or “piecemeal degranulation” (Letourneau et  al. 1996). Moreover, in certain diseases such as scleroderma and interstitial cystitis (Theoharides et al. 1995), mast cells could be almost totally depleted of their granule content, without classic degranulation, rendering them undetectable by light microscopy (“phantom mast cells”) (Claman et al. 1986).

Mast Cells Could Be Beneficial to the Tumor Mast cells could accumulate at sites of tumor growth in response to numerous chemoattractants (Table 17.1) such as RANTES or MCP-1 (Conti et al. 1997), and are associated with poor prognosis (Molin et  al. 2002). In addition, SCF at low doses mediates chemotaxis of mast cells, while a higher dose is necessary for release of mediators (Huang et  al. 2008), such as metalloproteinase-9 (MMP-9) (Huang et al. 2008). Adrenomedullin can stimulate histamine release from rat peritoneal mast cells (Yoshida et al. 2001), but can also be released from human cultured A549 lung carcinoma cells and stimulate human leukemic mast cells (HMC-1) (Zudaire et  al. 2006). Moreover, adrenomedullin is also produced by HMC-1 cells, it augments growth of lung cancer cells (Zudaire et al. 2006), and adrenomedullin-producing mast cells were shown to infiltrate human lung cancers (Zudaire et al. 2006). Mast cells have been consistently implicated in tumor angiogenesis (Crivellato et al. 2008), along with other myeloid cells (Murdoch et al. 2008). Mast cell-deficient W/Wv mice exhibited decreased rate of tumor angiogenesis (Starkey et  al. 1988). Mast cells could facilitate tumor angiogenesis through heparin-like molecules that would also permit metastases through their anti-clotting effects (Fig. 17.1). Mast cells

Table 17.1  Tumor-derived mast cell chemoattractants/ triggers

• Adrenomedullin • MCP-1 • RANTES • SCF

358

T.C. Theoharides et al.

also generate and secrete IL-8 which has been shown to be an angiogenesis factor, as well as a tumor cell chemotactic factor and tumor mitogen (Waugh and Wilson 2008). Mast cells secrete a number of growth factors, such as EGF, PDGF, NGF, and SCF (Galli et  al. 2005b). Moreover, VPF/VEGF is secreted from mouse bone marrowderived and human cultured mast cells (Boesiger et al. 1998), as well as from HMC-1 cells (Grutzkau et al. 1998) (Table 17.2). Mast cells are rich in metalloproteinases, such as MMP-9, that can facilitate tumor invasiveness (Almholt and Johnsen 2003). Such enzymes can disturb the normal stroma-epithelium communication, as was shown for matrix degradation at sites of tumor invasion in rat mammary adenocarcinoma (Dabbous et  al. 1986). Mast cells and stress could also disrupt the blood-brain-barrier (BBB) and promote brain metastases (Theoharides et al. 2008). Acute stress can activate mast cells and increase BBB permeability that is mast cell dependent (Esposito et  al. 2002a). These findings are important in view of the fact that acute stress increases metastases in breast and other tumors (Sklar and Anisman 1979; Antoni et al. 2006), and over 30% of breast cancer patients develop brain metastases with poor associated prognosis (Schouten et al. 2002). In fact, a number of cancers express CRH receptors (Reubi et al. 2003), prompting the suggestion that CRH may affect tumor cell behavior. For instance, a human breast cancer line MCF7 expresses CRH mRNA and secrets immunoreactive CRH (Graziani et al. 2006b), prompting the possibility of autocrine or paracrine effects. Another aspect of tumor microenvironment is immunosuppression. Histamine induces tumor cell proliferation through H1 receptors, while suppressing the immune system through H2 and possibly H4 receptors (Tiligada et  al. 2009; Gutzmer et al. 2005). It was also shown that the histamine content of human breast Table 17.2  Mast cell mediators relevant to tumor microenvironment Mediators Main pathophysiologic effects Histamine Vasodilation, mitogenesis, immunosuppression Enzymes   Chymase Tissue damage   Metalloproteinases Tissue disruption, metastases   Tryptase Tissue damage, metastases Growth factors Tumor growth   EGF, GM-CSF, NGF,   PDGF, SCF Angiogenic factors Angiogenesis, neovascularization   Angiogenin   Heparin   VEGF Cytokines   IL-8 Chemoattractant, tumor mitogen EGF epidermal growth factor, GM-CSF granulocyte monocyte-colony stimulating factor, NGF nerve growth factor, PDGF platelet-derived growth factor, SCF stem cell factor, VEGF vascular endothelial cell growth factor

17  Mast Cells and Tumor Microenvironment

359

cancer tissue was much higher than adjacent normal tissue and sufficient to act as a local immunosuppressant (Reynolds et  al. 1998). The H2 receptor antagonist, famotidine, given preoperatively enhanced tumor infiltrating lymphocytes and increased metastatic lymph node reactive changes in breast cancer in humans (Parshad et al. 2002). Mast cells can promote immunosuppression by also secreting the immunosuppressants transforming growth factor-b (TGF-b) and IL-10 (Conti et al. 2003). SCF-mediated mast cell infiltration of tumors further enhances immunosuppression (Huang et  al. 2008). Mast cells may further contribute to tumor anergy by promoting the development/recruitment of “tolerogenic host antigenpresenting cells” (Wasiuk et  al. 2009) and Treg cells (Wasiuk et  al. 2009). Mast cells may also influence migration and function of dendritic cells through distinct prostaglandins (PGE2 and PGD2) (Wasiuk et al. 2009). In addition, mast cells could indirectly down-regulate anti-tumor immunity by influencing the functions of immune suppressive myeloid-derived stem cells (MDSCs). MDSCs also secrete VEGF (Marx 2008), but this action requires MMP-9 (Yang et al. 2004), which is produced by mast cells. Mast cell-derived IL-1b induces MDSCs in mice with transplanted mammary carcinoma or fibrosarcoma (Bunt et al. 2006).

Breast Cancer Disruption of the normal flow of information between stroma and parenchyma could permit neoplastic progression (Barcellos-Hoff 1998). Stromal matrix metalloproteinases, rather than the target cell, were shown to promote mammary tumorigenesis (Sternlicht et al. 1999), while irradiated mammary gland stroma promoted carcinogenesis of unirradiated epithelial cells (Barcellos-Hoff and Ravani 2000). Mammary carcinogenesis in Wistar/Furth rats occurs when only the stroma of the mammary gland (fat pad) is exposed to the carcinogen nitrosomethylurea (NMU) (Maffini et al. 2004). The earliest effects of carcinogen administration in mammary gland carcinogenesis are manifested in the stroma with infiltration of inflammatory cells including mast cells (Maffini et  al. 2004). The number of mast cells was significantly increased in malignant as compared to benign, lesions in human breast biopsies (Kashiwase et al. 2004). Moreover, the number of mast cells was greater in scirrhous than papillotubular carcinoma (Kashiwase et al. 2004). The histamine content of human breast cancer tissue, an index of mast cell presence, was much higher than adjacent normal tissue (Reynolds et al. 1998). Recent papers also confirmed high number of mast cells in human mammary adenocarcinoma (Rajput et al. 2007; Ribatti et al. 2007). Some of these papers suggested that the presence of mast cells may indicate a favorable prognosis (Rajput et al. 2007; Dabiri et al. 2004); however, in these instances mast cells were “identified” by staining for c-kit (Rajput et al. 2007; Dabiri et al. 2004), which is not specific, as cancer cells also express c-kit (Charpin et  al. 2009). Moreover, reduction of c-kit expression was associated with malignant transformation of breast epithelium in human breast cancer (Polat 2007), and in carcinogen-induced rat mammary carcinoma (Maffini

360

T.C. Theoharides et al.

et  al. 2008). Consequently, tumor c-kit expression and not the presence of c-kit positive mast cells, appears to be associated with favorable outcomes. An increased number of mast cells was reported in cis-hydroxyproline-induced mammary tumors in Buffalo rats (Strum et al. 1981). Similar findings were obtained in 7, 12-dimethylbenz(a)anthracene (DMBA)-induced mammary adenocarcinoma in which mast cells accumulated, but appeared to be intact and resistant to the mast cell degranulator compound 48/80 (Andersson et al. 1976). The mast cell inhibitor disodium cromoglycate increased blood clotting and hypoxia in invasive murine breast cancer (Samoszuk and Corwin 2003). The location of mast cells in relation to tumor cells may also be important. Lymph nodes may behave differently (Munn and Mellor 2006) upon cancer cell entrapment. Tryptase-positive mast cells correlated with angiogenesis and presence of micrometastases in sentinel lymph nodes from 80 patients with breast cancer (Ribatti et al. 2007). In contrast, intermammary lymph node enlargement with mast cell infiltration was considered to be a positive prognostic sign (Quan et al. 2002).

Melanoma and Basal Cell Carcinoma Mast cells accumulate especially around invasive melanoma (Reed et  al. 1996; Dvorak et al. 1980), and their numbers correlate with increased neovascularization, mast cell overexpression of VEGF, tumor aggressiveness, and poor prognosis (Ch’ng et al. 2006). Tumor vascularity and tryptase-positive mast cells correlated with poor melanoma prognosis (Ribatti et al. 2003). Moreover, SCF splice variants were detected in melanoma (Welker et al. 2000), and could present new forms of mast cell growth factors related to melanoma growth. Mast cells have also been repeatedly noted to accumulate around basal cell carcinoma lesions and are thought to contribute to cancer growth by inducing immunosuppression (Grimbaldeston et al. 2000). Increased dermal mast cell numbers are associated with higher risk of developing basal cell carcinoma in humans possibly through UVB-induced immunosuppression (Grimbaldeston et  al. 2002). Mast cells can also mediate TNF-a dependent dendritic cell migration and consequently increase skin tumor antigen presentation, but in a manner that does not elicit an immune response (Munn and Mellor 2006; Hart et al. 2002).

Pancreatic Cancer Pancreatic ductal adenocarcinoma (PDAC) is the 4th cause of cancer-related deaths in the USA, with a prognosis of less than 6.0 months and 5-year survival of less than 5% (Welsch et al. 2007; Hezel et al. 2006). PDAC escapes early detection and resists treatment (Tuveson and Hingorani 2005). Even though PDAC had been

17  Mast Cells and Tumor Microenvironment

361

Fig. 17.2  Light photomicrograph of ductal pancreatic adenocarcinoma with a number of infiltrating intact mast cells (solid arrows) stained with acidified toluidine blue, magnification = 20×

thought not to be particularly vascular, it was shown to secrete VEGF (Luo et al. 2001). Adrenomedullin, discussed earlier as a mast cell trigger, is expressed by pancreatic cancer cells and increases their growth and survival (Nakamura et al. 2006; Keleg et al. 2007). Increased numbers of mast cells have also been noted in pancreatic cancer (Esposito et al. 2002b, 2004). Moreover, a recent paper reported that mast cells were an absolute requirement for angiogenesis and cancer development in Myc-activated mouse pancreatic tumors (Soucek et al. 2007). Mast cells are actually localized among ductal adenocarcinoma cells, and they appear morphologically intact (Fig. 17.2), suggesting that pro-tumor molecules may be released without degranulation.

Lung Cancer The relevance of lung mast cells to lung cancer is intriguing, given that mast cells are well known to be increased in asthma and other inflammatory lung diseases (Krishnaswamy et  al. 2007). In fact, lung mast cells are also increased in the epithelium of smokers (Lamb and Lumsden 1982), especially in the small airways (Battaglia et al. 2007). Gene profiling of disaggregated human lung mast cells from patients with interstitial diseases showed increased expression of matrix metalloproteinases (Edwards et  al. 2005). The number of mast cells was increased in bronchoalveolar lavage of patients with bronchial carcinoma (Walls et  al. 2007). Also, increased mast cell density using tryptase and surface CD34 immunocytochemistry was significantly correlated with tumor progression, angiogenesis and poor prognosis in human adenocarcinomas (Takanami et al. 2000). High counts of

362

T.C. Theoharides et al.

chymase-positive mast cells also correlated with worse prognosis in bronchoalveolar carcinoma (Nagata et al. 2003) and in lung adenocarcinoma (Ibaraki et al. 2005). This direct correlation between increased numbers of mast cells and lung cancer was apparently independent of tumor angiogenesis, as measured by the presence of endothelial cells stained with anti-human factor VIII antibody (Tomita et al. 2000). Nevertheless, increased mast cell density correlated with increased VEGF expression and poor prognosis in 33/53 cases of non-small cell lung carcinomas (NSCLC) (Imada et al. 2000). Histidine decarboxylase (HDC) immunoreactivity, an index of mast cell presence/activation, could distinguish 18/23 cases of small cell lung carcinomas (SCLC), but only 6/12 cases of NSCLC (Matsuki et al. 2003), suggesting that higher number of mast cells were infiltrating the SCLC, known to be more aggressive.

Mast Cells Could be Detrimental to the Tumor Even though mast cells could be detrimental to tumor growth, they apparently cannot secrete such mediators as they may be inhibited from degranulation by tumorderived blockers, such as oxidized polyamines (Vliagoftis et al. 1992), or chondroitin sulfate, which inhibits mast cell activation (Theoharides et al. 2000). It would be fascinating if one could inhibit mast cells from secreting pro-tumor mediators, but promote secretion of anti-tumor molecules. For instance, tryptase stimulates protease-activated receptors (PAR-1 and -2), also activated by thrombin and trypsin respectively, and induces widespread inflammation (D’Andrea et  al. 2001). IL-4 binds to IL-4 receptors expressed by human breast carcinoma cells, and leads to apoptosis (Gooch et al. 1998). TNF-a could also induce tumor cell death (Gordon and Galli 1990). Histamine inhibited human primary melanoma cell proliferation presumably by acting through H1 receptors, an action enhanced by IL-6 (LazarMolnar et al. 2002). Heparan sulfate proteoglycans could block binding of heparin to the cell surface and prevent neovascularization (Fannon et al. 2003). For instance protamine, which binds to heparin and neutralizes its anticoagulant properties, induced selective thrombosis of blood vessels within mammary adenocarcinoma (Su et al. 2001). On the other hand, cancer cell-associated chondroitin sulfate accumulates in mammary gland tumors and in metastatic lesions (Hinrichs et al. 1999); in fact, tumor cells metastasize through binding of their surface chondroitin sulfate to the interstitial matrix (Kokenyesi 2001). Interferon-a (IFN-a) or mast cell-derived interferon-g (IFN-g) may enhance TNF-related apoptosis-related ligand (TRAIL) gene expression and translation, leading to apoptosis of tumor cells in an autocrine and paracrine manner (Abadie et  al. 2004; Wang et  al. 2004). Moreover, treatment of melanoma patients with IFN-a increased TRAIL levels in serum (Tecchio et al. 2004). Receptor binding of TRAIL, in turn, activates an number of down steam events leading to regulation of the inflammatory response (Collison et al. 2009). In addition to its cytotoxic role,

17  Mast Cells and Tumor Microenvironment

363

TRAIL can inhibit angiogenesis indirectly by facilitating apoptosis of endothelial cells (Li et al. 2003). Certain studies indicate that CRH may inhibit growth of endometrial (Graziani et al. 2002) and breast cancer cells (Graziani et al. 2006b) in culture. Other studies suggest that endometrial cancer not expressing CRHR-1 may be associated with a more aggressive phenotype in humans (Graziani et al. 2006a). Mast cells (Kempuraj et al. 2004) and other inflammatory cells (Karalis et al. 1997) can synthesize and release CRH. It would, therefore, be interesting if mast cells could release CRH, but at the same time be prevented from CRH inducing VEGF release in an autocrine fashion.

Conclusion Mounting evidence indicates that mast cells accumulate in tumor stroma and can promote tumor growth and metastases. Mast cells may, therefore, serve as a new target for the adjuvant treatment of tumors (Groot et al. 2009), such as mammary adenocarcinoma or pancreatic cancer (Theoharides 2008), through the selective inhibition of tumor-promoting molecules. A possible therapeutic approach could involve the use of select flavonoids. Flavonoids are naturally occurring polyphenolic compounds present in green plants and seeds with anti-oxidant, anti-inflammatory and cancer-inhibiting properties (Middleton et al. 2000). A multiethnic epidemiological study identified an inverse relationship between flavonoid intake and pancreatic cancer (Nothlings et  al. 2007). An inverse relationship has also been reported between intake of flavonoids, such as quercetin, and risk of lung cancer in general (Le Marchand et al. 2000), as well as in male smokers in particular (Hirvonen et al. 2001). The flavonoids quercetin, luteolin and epigallocatechin also decrease proliferation of pancreatic carcinoma cells in culture (Lee et  al. 2002; Shankar et  al. 2008; Takada et  al. 2002), but the mechanism of this action is not known. We showed that these flavonoids inhibit proliferation and secretion from rat basophil leukemia (RBL-1) cells (Alexandrakis et al. 1999) and HMC-1 cells (Alexandrakis et  al. 2003), as well as pro-inflammatory cytokine release from normal hCBMCs (Kempuraj et al. 2005). In fact, recent reviews have re-emphasized the potential use of select flavonoids, such as epigallocatechin, quercetin and curcumin in cancer treatment (Saif et al. 2009; Sogno et al. 2009; Jagtap et al. 2009). Future studies should investigate any unique ability of select flavonoids with anti-cancer properties to also inhibit secretion of pro-cancer molecules, while permitting secretion of anti-cancer mediators from mast cells. Acknowledgments  Work discussed was supported in part by a “Concept Award” No. BC024430 from the United States Department of Defense to (TCT), and funds from Theta Biomedical Consulting and Development Co., Inc. (Brookline, MA). The possible therapeutic role of inhibiting mast cell-derived molecules beneficial to tumor growth is covered by patent application US 10/811,838 submitted by TCT and assigned to Theta, Inc.

364

T.C. Theoharides et al.

Many thanks are due to Dr. Eva Karamitropoulou (Attikon General Hospital, Athens Medical School, Athens, Greece) for the pancreatic cancer photo, and to Ms. Jessica Christian for her word processing skills.

References Abadie A, Besancon F, Wietzerbin J (2004) Type I interferon and TNFalpha cooperate with type II interferon for TRAIL induction and triggering of apoptosis in SK-N-MC EWING tumor cells. Oncogene 23:4911–4920 Alexandrakis M, Singh L, Boucher W, Letourneau R, Theofilopoulos P et al (1999) Differential effect of flavonoids on inhibition of secretion and accumulation of secretory granules in rat basophilic leukemia cells. Int J Immunopharmacol 21:379–390 Alexandrakis MG, Letourneau R, Kempuraj D, Kandere K, Huang M et al (2003) Flavones inhibit proliferation and increase mediator content in human leukemic mast cells (HMC-1). Eur J Haematol 71:448–454 Allakhverdi Z, Comeau MR, Jessup HK, yoon BP, Breuer A et al (2007) Thymic stromal lymphopoietin is released by human epithelial cell in response to microbes, trauma, or inflammation and potently activates mast cells. J Exp Med 19:253–258 Almholt K, Johnsen M (2003) Stromal cell involvement in cancer. Cancer Res 162:31–42 Al-Shami A, Spolski R, Kelly J, Keane-Myers A, Leonard WJ (2005) A role for TSLP in the development of inflammation in an asthma model. J Exp Med 202:829–839 Andersson A-C, Henningsson S, Lundell L, Rosengren E, Sundler F (1976) Diamines and polyamines in DMBA-induced breast carcinoma containing mast cells resistant to compound 48/80. Agents Actions 6/5:577–583 Antoni MH, Lutgendorf SK, Cole SW, Dhabhar FS, Sephton SE et  al (2006) The influence of bio-behavioural factors on tumour biology: pathways and mechanisms. Nat Rev Cancer 6:240–248 Barbera-Guillem E, Nyhus JK, Wolford CC, Friece CR, Sampsel JW (2002) Vascular endothelial growth factor secretion by tumor-infiltrating macrophages essentially supports tumor angiogenesis, and IgG immune complexes potentiate the process. Cancer Res 62:7042–7049 Barcellos-Hoff MH (1998) The potential influence of radiation-induced microenvironments in neoplastic progression. J Mammary Gland Biol Neoplasia 3:165–175 Barcellos-Hoff MH, Ravani SA (2000) Irradiated mammary gland stroma promotes the expression of tumorigenic potential by unirradiated epithelial cells. Cancer Res 60:1254–1260 Battaglia S, Mauad T, van Schadewijik AM, Vignola AM, Rabe KF et  al (2007) Differential distribution of inflammatory cells in large and small airways in smokers. J Clin Pathol 60: 907–911 Boesiger J, Tsai M, Maurer M, Yamaguchi M, Brown LF et al (1998) Mast cells can secrete vascular permeability factor/vascular endothelial cell growth factor and exhibit enhanced release after immunoglobulin E-dependent upregulation of Fce receptor I expression. J Exp Med 188:1135–1145 Bunt SK, Sinha P, Clements VK, Leips J, Ostrand-Rosenberg S (2006) Inflammation induces myeloid-derived suppressor cells that facilitate tumor progression. J Immunol 176:284–290 Cao J, Papadopoulou N, Kempuraj D, Boucher WS, Sugimoto K et al (2005) Human mast cells express corticotropin-releasing hormone (CRH) receptors and CRH leads to selective secretion of vascular endothelial growth factor. J Immunol 174:7665–7675 Charpin C, Giusiano S, Charfi S, Secq V, Carpentier S et  al (2009) Quantitative immunohistochemical expression of c Kit in breast carcinomas is predictive of patients’ outcome. Br J Cancer 101:48–54

17  Mast Cells and Tumor Microenvironment

365

Ch’ng S, Waallis RA, Yuan L, Davis PF, Tan ST (2006) Mast cells and cutaneous malignancies. Mod Pathol 19:149–159 Claman HN, Choi KL, Sujansky W, Vatter AE (1986) Mast cell “disappearance” in chronic murine graft-vs-host disease (GVHD)-ultrastructural demonstration of “phantom mast cells”. J Immunol 137:2009–2013 Coleman JW (2002) Nitric oxide: a regulator of mast cell activation and mast cell-mediated inflammation. Clin Exp Immunol 129:4–10 Collison A, Foster PS, Mattes J (2009) The emerging role of TRAIL as key regulator of inflammatory responses. Clin Exp Pharmacol Physiol 36:1049–1053 Conti P, Pang X, Boucher W, Letourneau R, Reale M et al (1997) Impact of Rantes and MCP-1 chemokines on in vivo basophilic mast cell recruitment in rat skin injection model and their role in modifying the protein and mRNA levels for histidine decarboxylase. Blood 89: 4120–4127 Conti P, Kempuraj D, Kandere K, Gioacchino MD, Barbacane RC et al (2003) IL-10, an inflammatory/inhibitory cytokine, but not always. Immunol Lett 86:123–129 Conti P, Castellani ML, Kempuraj D, Salini V, Vecchiet J et al (2007) Role of mast cells in tumor growth. Ann Clin Lab Sci 37:315–322 Coussens LM, Raymond WW, Nergers G, Laig-Webster M, Behrendtsen O et  al (1999) Inflammatory mast cells up-regulate angiogenesis during squamous epithelial carcinogenesis. Genes Dev 13:1382–1397 Crivellato E, Nico B, Ribatti D (2008) Mast cells and tumour angiogenesis: new insight from experimental carcinogenesis. Cancer Lett 269:1–6 D’Andrea MR, Derian CK, Santulli RJ, Andrade-Gordon P (2001) Differential expression of protease-activated receptors-1 and -2 in stromal fibroblasts of normal, benign, and malignant human tissues. Am J Pathol 158:2031–2041 Dabbous MK, Walker R, Haney L, Carter LM, Nicolson GL et al (1986) Mast cells and matrix degradation at sites of tumor invasion in rat mammary adenocarinoma. Br J Cancer 54:459–465 Dabiri S, Huntsman D, Makretsov N, Cheang M, Gilks B et al (2004) The presence of stromal mast cells identifies a subset of invasive breast cancers with a favorable prognosis. Mod Pathol 17:690–695 Dethelfsen SM, Matsumura N, Zetter BR (1994) Mast cell accumulation at sites of murine tumor implantation: implications for angiogenesis and tumor metastasis. Invasion Metastasis 14:395–408 Dvorak AM, Mihm MCJ, Osage JE, Dvorak HF (1980) Melanoma. An ultrastructural study of the host inflammatory and vascular responses. J Invest Dermatol 75:388–393 Edwards ST, Cruz AC, Donnelly S, Dazin PF, Schulman ES et al (2005) c-Kit immunophenotyping and metalloproteinase expression profiles of mast cells in intestinal lung diseases. J Pathol 206:279–290 Egami K, Murohara T, Shimada T, Sasaki K, Shintani S et al (2003) Role of host angiotensin II type 1 receptor in tumor angiogenesis and growth. J Clin Invest 112:67–75 Esposito P, Chandler N, Kandere-Grzybowska K, Basu S, Jacobson S et al (2002a) Corticotropinreleasing hormone (CRH) and brain mast cells regulate blood-brain-barrier permeability induced by acute stress. J Pharmacol Exp Ther 303:1061–1066 Esposito I, Kleeff J, Bischoff SC, Fischer L, Collecchi P et al (2002b) The stem cell factor-c-kit system and mast cells in human pancreatic cancer. Lab Invest 82:1481–1492 Esposito I, Menicagli M, Funel N, Bergmann F, Boggi U et al (2004) Inflammatory cells contribute to the generation of an angiogenic phenotype in pancreatic ductal adenocarcinoma. J Clin Pathol 57:630–636 Fannon M, Forsten-Williams K, Dowd CJ, Freedman DA, Folkman J et al (2003) Binding inhibition of angiogenic factors by heparan sulfate proteoglycans in aqueous humor: potential mechanism for maintenance of an avascular environment. FASEB J 17:902–904

366

T.C. Theoharides et al.

Frossi B, De Carli M, Daniel KC, Rivera J, Pucillo C (2003) Oxidative stress stimulates IL-4 and IL-6 production in mast cells by an APE/Ref-1-dependent pathway. Eur J Immunol 33:2168–2177 Gagari E, Tsai M, Lantz CS, Fox LG, Galli SJ (1997) Differential release of mast cell interleukin-6 via c-kit. Blood 89:2654–2663 Galli SJ, Kalesnikoff J, Grimbaldeston MA, Piliponsky AM, Williams CM et al (2005a) Mast cells as “tunable” effector and immunoregulatory cells: recent advances. Annu Rev Immunol 23:749–786 Galli SJ, Nakae S, Tsai M (2005b) Mast cells in the development of adaptive immune responses. Nat Immunol 6:135–142 Gooch JL, Lee AV, Yee D (1998) Interleukin 4 inhibits growth and induces apoptosis in human breast cancer cells. Cancer Res 58:4199–4205 Gordon JR, Galli SJ (1990) Mast cells as a source of both preformed and immunologically inducible TNF-a/cachectin. Nature 346:274–276 Graziani G, Tentori L, Portarena I, Barbarino M, Tringali G et al (2002) CRH inhibits cell growth of human endometrial adenocarcinoma cells via CRH-receptor 1-mediated activation of cAMP-PKA pathway. Endocrinology 143:807–813 Graziani G, Ferrandina G, Pozzoli G, Vergati M, Muzi A et  al (2006a) Corticotropin-releasing hormone receptor-1 in human endometrial cancer. Oncol Rep 15:375–379 Graziani G, Tentori L, Muzi A, Vergati M, Tringali G et al (2006b) Evidence that corticotropinreleasing hormone inhibits cell growth of human breast cancer cells via the activation of CRHR1 receptor subtype. Mol Cell Endocrinol 264:44–49 Grimbaldeston MA, Skov L, Baadsgaard O, Skov BG, Marshman G et al (2000) Communications: high dermal mast cell prevalence is a predisposing factor for basal cell carcinoma in humans. J Invest Dermatol 115:317–320 Grimbaldeston MA, Skov L, Finlay-Jones JJ, Hart PH (2002) Increased dermal mast cell prevalence and susceptibility to development of basal cell carcinoma in humans. Methods 28:90–96 Groot KT, Abudukelimu A, Redegeld FA (2009) Mast cells as target in cancer therapy. Curr Pharm Des 15:1868–1878 Grutzkau A, Kruger-Krasagakes S, Baumeister H, Schwarz C, Kogel H et  al (1998) Synthesis, storage and release of vascular endothelial growth factor/vascular permeability factor (VEGF/ VPF) by human mast cells: Implications for the biological significance of VEGF206. Mol Biol Cell 9:875–884 Gutzmer R, Diestel C, Mommert S, Kother B, Stark H et al (2005) Histamine H4 receptor stimulation suppresses IL-12p70 production and mediates chemotaxis in human monocyte-derived dendritic cells. J Immunol 174:5224–5232 Hart PH, Townley SL, Grimbaldeston MA, Khalil Z, Finlay-Jones JJ (2002) Mast cells, neuropeptides, histamine, and prostaglandins in UV-induced systemic immunosuppression. Methods 28:79–89 Hezel AF, Kimmelman AC, Stanger BZ, Bardeesy N, Depinho RA (2006) Genetics and biology of pancreatic ductal adenocarcinoma. Genes Dev 20:1218–1249 Hinrichs U, Rutteman GR, Nederbragt H (1999) Stromal accumulation of chondroitin sulphate in mammary tumours of dogs. Br J Cancer 80:1359–1365 Hirvonen T, Tornwali ME, Pietinen P, Albanes D, Vitamo J (2001) Flavonol and flavone intake and the risk of cancer in male smokers (Finland). Cancer Causes Control 12:789–796 Huang B, Lei Z, Zhang GM, Li D, Song C et al (2008) SCF-mediated mast cell infiltration and activation exacerbate the inflammation and immunosuppression in tumor microenvironment. Blood 112:1269–1279 Ibaraki T, Muramatsu M, Takai S, Jin D, Maruyama H et al (2005) The relationship of tryptase and chymase positive mast cells to angiogenesis in stage 1 non-small cell lung cancer. Eur J Cardiothorac Surg 28:621 Imada A, Shijubo N, Kojima H, Abe S (2000) Mast cells correlate with angiogenesis and poor outcome in stage I lung adenocarcinoma. Eur Respir J 15:1087–1093

17  Mast Cells and Tumor Microenvironment

367

Jagtap S, Meganathan K, Wagh V, Winkler J, Hescheler J et al (2009) Chemoprotective mechanism of the natural compounds, epigallocatechin-3-O-gallate, quercetin and curcumin against cancer and cardiovascular diseases. Curr Med Chem 16:1451–1462 Kandere-Grzybowska K, Letourneau R, Kempuraj D, Donelan J, Poplawski S et al (2003) IL-1 induces vesicular secretion of IL-6 without degranulation from human mast cells. J Immunol 171:4830–4836 Karalis K, Louis JM, Bae D, Hilderbrand H, Majzoub JA (1997) CRH and the immune system. J Neuroimmunol 72:131–136 Kashiwase Y, Morioka J, Inamura H, Yoshizawa Y, Usui R et al (2004) Quantitative analysis of mast cells in benign and malignant breast lesions. Immunohistochemical study on formalinfixed, paraffin-embedded tissues. Int Arch Allergy Immunol 134:199–205 Kataki A, Scheid P, Piet M, Marie B, Martinet N et al (2002) Tumor infiltrating lymphocytes and macrophages have a potential dual role in lung cancer by supporting both host-defense and tumor progression. J Lab Clin Med 140:320–328 Keleg S, Kayed H, Jiang X, Penzel R, Giese T et al (2007) Adrenomedullin is induced by hypoxia and enhances pancreatic cancer cell invasion. Int J Cancer 121:21–32 Kempuraj D, Papadopoulou NG, Lytinas M, Huang M, Kandere-Grzybowska K et  al (2004) Corticotropin-releasing hormone and its structurally related urocortin are synthesized and secreted by human mast cells. Endocrinology 145:43–48 Kempuraj D, Madhappan B, Christodoulou S, Boucher W, Cao J et al (2005) Flavonols inhibit proinflammatory mediator release, intracellular calcium ion levels and protein kinase C theta phosphorylation in human mast cells. Br J Pharmacol 145:934–944 Kokenyesi R (2001) Ovarian carcinoma cells synthesize both chondroitin sulfate and heparan sulfate cell surface proteoglycans that mediate cell adhesion to interstitial matrix. J Cell Biochem 83:259–270 Krishnaswamy G, Ajitawi O, Chi DS (2007) The human mast cells. Methods Mol Biol 315:13–34 Lamb D, Lumsden A (1982) Intra-epithelial mast cells in human airway epithelium: evidence for smoking-induced changes in their frequency. Thorax 37:334–342 Lazar-Molnar E, Hegyesi H, Pallinger E, Kovacs P, Toth S et  al (2002) Inhibition of human primary melanoma cell proliferation by histamine is enhanced by interleukin-6. Eur J Clin Invest 32:743–749 Le Marchand L, Murphy SP, Hankin JH, Wilkens LR, Kolonel LN (2000) Intake of flavonoids and lung cancer. J Natl Cancer Inst 92:154–160 Lee LT, Huang YT, Hwang JJ, Lee PP, Ke FC et al (2002) Blockade of the epidermal growth factor receptor tyrosine kinase activity by quercetin and luteolin leads to growth inhibition and apoptosis of pancreatic tumor cells. Anticancer Res 22:1615–1627 Letourneau R, Pang X, Sant GR, Theoharides TC (1996) Intragranular activation of bladder mast cells and their association with nerve processes in interstitial cystitis. Br J Urol 77:41–54 Li C, Shintani S, Terakado N, Nakashiro K, Hamakawa H (2002) Infiltration of tumor-associated macrophages in human oral squamous cell carcinoma. Oncol Rep 9:1219–1223 Li JH, Kirkiles-Smith NC, McNiff JM, Pober JS (2003) TRAIL induces apoptosis and inflammatory gene expression in human endothelial cells. J Immunol 171:1526–1533 Luo J, Guo P, Matsuda K, Truong N, Lee A et al (2001) Pancreatic cancer cell-derived vascular endothelial growth factor is biologically active in vitro and enhances tumorigenicity in vivo. Int J Cancer 92:361–369 Maffini MV, Soto AM, Calabro JM, Ucol AA, Sonnenschein C (2004) The stroma as a crucial target in rat mammary gland carcinogenesis. J Cell Sci 117:1495–1502 Maffini MV, Soto AM, Sonnenschein C, Papadopoulos N, Theoharides TC (2008) Lack of c-kit receptor promotes mammary tumors in N-nitrosomethylurea-treated Ws/Ws rats. Cancer Cell Int 8:5 Mantovani A, Sozzani S, Locati M, Allavena P, Sica A (2002) Macrophage polarization: tumorassociated macrophages as a paradigm for polarized M2 mononuclear phagocytes. Trends Immunol 23:549–555

368

T.C. Theoharides et al.

Marx J (2008) Cancer immunology. Cancer’s bulwark against immune attack: MDS cells. Science 319:154–156 Matsuki Y, Tanimoto A, Hamada T, Sasaguri Y (2003) Histidine decarboxylase expression as a new sensitive and specific marker for small cell lung carcinoma. Mod Pathol 16:72–78 Mekori YA, Metcalfe DD (2000) Mast cells in innate immunity. Immunol Rev 173:131–140 Middleton E Jr, Kandaswami C, Theoharides TC (2000) The effects of plant flavonoids on mammalian cells: implications for inflammation, heart disease and cancer. Pharmacol Rev 52:673–751 Molin D, Edstrom A, Glimelius I, Glimelius B, Nilsson G et  al (2002) Mast cell infiltration correlates with poor prognosis in Hodgkin’s lymphoma. Br J Haematol 119:122–124 Munn DH, Mellor AL (2006) The tumor-draining lymph node as an immune-privileged site. Immunol Rev 213:146–158 Murdoch C, Muthana M, Coffelt SB, Lewis CE (2008) The role of myeloid cells in the promotion of tumour angiogenesis. Nat Rev Cancer 8:618–631 Murphy PM (2001) Chemokines and the molecular basis of cancer metastasis. N Engl J Med 345:833–835 Nagata M, Shijubo N, Waalls AF, Ichimiya s, Abe S et al (2003) Chymase positive mast cells in small sized adenocarcinoma of the lung. Virchows Arch 443:565–573 Nakamura M, Han B, Nunobiki O, Kakudo K (2006) Adrenomedullin: a tumor progression factor via angiogenic control. Curr Cancer Drug Targets 6:635–643 Nothlings U, Murphy SP, Wilkens LR, Henderson BE, Kolonel LN (2007) Flavonols and pancreatic cancer risk: the multiethnic cohort study. Am J Epidemiol 166:924–931 Ochi H, DeJesus NH, Hsieh FH, Austen KF, Boyce JA (2000) IL-4 and -5 prime human mast cells for different profiles of IgE-dependent cytokine production. Proc Natl Acad Sci U S A 97:10509–10513 Parshad R, Kapoor S, Gupta SD, Kumar A, Chattopadhyaya TK (2002) Does famotidine enhance tumor infiltrating lymphocytes in breast cancer? Results of a randomized prospective pilot study. Acta Oncol 41:362–365 Pedotti R, De Voss JJ, Steinman L, Galli SJ (2003) Involvement of both ‘allergic’ and ‘autoimmune’ mechanisms in EAE, MS and other autoimmune diseases. Trends Immunol 24:479–484 Polat A (2007) c-KIT expression in columnar cell lesions of the breast accompanied by benign and malignant breast diseases. Pathol Res Pract 203:765–769 Quan C, Wang HL, Lu SJ (2002) Resistance to mammary carcinogenesis in Copenhagen rats: potential roles of vascular endothelial growth factor and mast cells. Cancer Lett 186:165–175 Rajput AB, Turbin DA, Cheang MCU, Voduc DK, Leung S et  al (2007) Stromal mast cells in invasive breast cancer are a marker of favourable prognosis: a study of 4,444 cases. Breast Cancer Res 107:249–257 Redegeld FA, Nijkamp FP (2003) Immunoglobulin free light chains and mast cells: pivotal role in T-cell-mediated immune reactions? Trends Immunol 24:181–185 Reed JA, McNutt NS, Bogdany JK, Albino AP (1996) Expression of the mast cell growth factor interleukin-3 in melanocytic lesions correlates with an increased number of mast cells in the perilesional stroma: implications for melanoma progression. J Cutan Pathol 23:495–505 Reubi JC, Waser B, Vale W, Rivier J (2003) Expression of CRF1 and CRF2 receptors in human cancers. J Clin Endocrinol Metab 88:3312–3320 Reynolds JL, Akhter JA, Magarey CJ, Schwartz P, Adams WJ et al (1998) Histamine in human breast cancer. Br J Surg 85:538–541 Ribatti D, Ennas MG, Vacca A, Ferreli F, Nico B et al (2003) Tumor vascularity and tryptasepositive mast cells correlate with a poor prognosis in melanoma. Eur J Clin Invest 33: 420–425 Ribatti D, Finato N, Crivellato E, Guidolin D, Longo V et al (2007) Angiogenesis and mast cells in human breast cancer sentinel lymph nodes with and without micrometastases. Histopathology 51:837–842

17  Mast Cells and Tumor Microenvironment

369

Saif MW, Tytler E, Lansigan F, Brown DM, Husband AJ (2009) Flavonoids, phenoxodiol, and a novel agent, triphendiol, for the treatment of pancreaticobiliary cancers. Expert Opin Investig Drugs 18:469–479 Saijo Y, Tanaka M, Miki M, Usui K, Suzuki T et al (2002) Proinflammatory cytokine IL-1 beta promotes tumor growth of Lewis lung carcinoma by induction of angiogenic factors: in vivo analysis of tumor-stromal interaction. J Immunol 169:469–475 Salven P, Hattori K, Heissig B, Rafii S (2002) Interleukin-1alpha promotes angiogenesis in vivo via VEGFR-2 pathway by inducing inflammatory cell VEGF synthesis and secretion. FASEB J 16:1471–1473 Samoszuk M, Corwin MA (2003) Mast cell inhibitor cromolyn increases blood clotting and hypoxia in murine breast cancer. Int J Cancer 107:159–163 Schouten LJ, Rutten J, Huveneers HA, Twijnstra A (2002) Incidence of brain metastases in a cohort of patients with carcinoma of the breast, colon, kidney, and lung and melanoma. Cancer 94:2698–2705 Shankar S, Ganapathy S, Hingorani SR, Srivastava RK (2008) EGCG inhibits growth, invasion, angiogenesis and metastasis of pancreatic cancer. Front Biosci 13:440–452 Sieghart W, Theoharides TC, Alper SL, Douglas WW, Greengard P (1978) Calcium-dependent protein phosphorylation during secretion by exocytosis in the mast cell. Nature 275:329–331 Silzle T, Kreutz M, Dobler MA, Brockhoff G, Knuechel R et al (2003) Tumor-associated fibroblasts recruit blood monocytes into tumor tissue. Eur J Immunol 33:1311–1320 Sklar LS, Anisman H (1979) Stress and coping factors influence tumor growth. Science 205:513–515 Sogno I, Vannini N, Lorusso G, Cammarota R, Noonan DM et al (2009) Anti-angiogenic activity of a novel class of chemopreventive compounds: oleanic acid terpenoids. Recent Results Cancer Res 181:209–212 Soucek L, Lawlor ER, Soto D, Shchors K, Swigart LB et al (2007) Mast cells are required for angiogenesis and macroscopic expansion of Myc-induced pancreatic islet tumors. Nat Med 13:1211–1218 Starkey JR, Crowle PK, Taubenberger S (1988) Mast-cell-deficient W/Wv mice exhibit a decreased rate of tumor angiogenesis. Int J Cancer 42:48–52 Sternlicht MD, Lochter A, Sympson CJ, Huey B, Rougier JP et al (1999) The stromal proteinase MMP3/Stromelysin-1 promotes mammary carcinogenesis. Cell 98:134–146 Strum JM, Lewko WM, Kidwell WR (1981) Structural alterations within N-nitrosomethylureainduced mammary tumors after in  vivo treatment with cis-hydroxyproline. Lab Invest 45:347–354 Su MY, Najafi A, Nalcioglu O (2001) Selective thrombosis of tumor blood vessels in mammary adenocarcinoma implants in rats. Am J Pathol 159:245–251 Takada M, Nakamura Y, Koizumi T, Toyama H, Kamigaki T et  al (2002) Suppression of human pancreatic carcinoma cell growth and invasion by epigallocatechin-3-gallate. Pancreas 25:45–48 Takanami I, Takeuchi K, Naruke M (2000) Mast cell density is associated with angiogenesis and poor prognosis in pulmonary adenocarcinoma. Cancer 88:2686–2692 Tecchio C, Huber V, Scapini P, Calzetti F, Margotto D et al (2004) IFNalpha-stimulated neutrophils and monocytes release a soluble form of TNF-related apoptosis-inducing ligand (TRAIL/ Apo-2 ligand) displaying apoptotic activity on leukemic cells. Blood 103:3837–3844 Theoharides TC (1988) Neuroimmunology of tumor growth: the role of mast cells. Int J Immunopath Pharmacol 1:89–98 Theoharides TC (2008) Mast cells and pancreatic cancer. N Engl J Med 358:1860–1861 Theoharides TC, Cochrane DE (2004) Critical role of mast cells in inflammatory diseases and the effect of acute stress. J Neuroimmunol 146:1–12 Theoharides TC, Douglas WW (1978) Secretion in mast cells induced by calcium entrapped within phospholipid vesicles. Science 201:1143–1145 Theoharides TC, Kalogeromitros D (2006) The critical role of mast cell in allergy and inflammation. Ann N Y Acad Sci 1088:78–99

370

T.C. Theoharides et al.

Theoharides TC, Sieghart W, Greengard P, Douglas WW (1980) Antiallergic drug cromolyn may inhibit histamine secretion by regulating phosphorylation of a mast cell protein. Science 207:80–82 Theoharides TC, Bondy PK, Tsakalos ND, Askenase PW (1982) Differential release of serotonin and histamine from mast cells. Nature 297:229–231 Theoharides TC, Sant GR, El-Mansoury M, Letourneau RJ, Ucci AA Jr et al (1995) Activation of bladder mast cells in interstitial cystitis: a light and electron microscopic study. J Urol 153:629–636 Theoharides TC, Patra P, Boucher W, Letourneau R, Kempuraj D et al (2000) Chondroitin sulfate inhibits connective tissue mast cells. Br J Pharmacol 131:1039–1049 Theoharides TC, Kempuraj D, Tagen M, Conti P, Kalogeromitros D (2007) Differential release of mast cell mediators and the pathogenesis of inflammation. Immunol Rev 217:65–78 Theoharides TC, Rozniecki JJ, Sahagian G, Kempuraj D, Kalogeromitros D et al (2008) Impact of stress and mast cells on brain metastases. J Neuroimmunol 205:1–7 Tiligada E, Zampeli E, Sander K, Stark H (2009) Histamine H3 and H4 receptors as novel drug targets. Expert Opin Investig Drugs 18:1519–1531 Tomita M, Matsuzaki Y, Onitsuka T (2000) Effect of mast cells on tumor angiogenesis in lung cancer. Ann Thorac Surg 69:1686–1690 Tuveson DA, Hingorani SR (2005) Ductal pancreatic cancer in humans and mice. Cold Spring Harb Symp Quant Biol 70:65–72 Vliagoftis H, Boucher WS, Mak LL, Theoharides TC (1992) Inhibition of mast cell secretion by oxidation products of natural polyamines. Biochem Pharmacol 43:2237–2245 Walls AF, Roberts JA, Godfrey RC, Holgate ST (2007) Histochemical heterogeneity of human mast cells: disease-related differences in mast cell subsets recovered by bronchoalveolar lavaage. Int Arch Allergy Appl Immunol 92:233–241 Wang SH, Mezosi E, Wolf JM, Cao Z, Utsugi S et  al (2004) IFNgamma sensitization to TRAIL-induced apoptosis in human thyroid carcinoma cells by upregulating Bak expression. Oncogene 23:928–935 Wasiuk A, de V V, Hartmann K, Roers A, Noelle RJ (2009) Mast cells as regulators of adaptive immunity to tumours. Clin Exp Immunol 155:140–146 Waugh DJ, Wilson C (2008) The interleukin-8 pathway in cancer. Clin Cancer Res 14:6735–6741 Wedemeyer J, Galli SJ (2005) Decreased susceptibility of mast cell-deficient kit(W)Kit(W-v) mice to the development of 1,2-dimethyldrazine-induced intestinal tumors. Lab Invest 85:388–396 Welker P, Schadendorf D, Artuc M, Grabbe J, Henz BM (2000) Expression of SCF splice variants in human melanocytes and melanoma cell lines: potential prognostic implications. Br J Cancer 82:1453–1458 Welsch T, Kleeff J, Friess H (2007) Molecular pathogenesis of pancreatic cancer: advances and challenges. Curr Mol Med 7:504–521 Yang L, DeBusk LM, Fukuda K, Fingleton B, Green-Jarvis B et  al (2004) Expansion of myeloid immune suppressor Gr+CD11b+ cells in tumor-bearing host directly promotes tumor angiogenesis. Cancer Cell 6:409–421 Yoshida M, Yoshida H, Kitaichi K, Hiramatsu K, Kimura T et  al (2001) Adrenomedullin and proadrenomedullin N-terminal 20 peptide induce histamine release from rat peritoneal mast cells. Regul Pept 101:163–168 Yu JL, Rak JW (2003) Host microenvironment in breast cancer development: inflammatory and immune cells in tumour angiogenesis and arteriogenesis. Breast Cancer Res 5:83–88 Zudaire E, Martinez A, Garayoa M, Pio R, Kaur G et al (2006) Adrenomedullin is a cross-talk molecule that regulates tumor and mast cell function during human carcinogenesis. Am J Pathol 168:280–291

Chapter 18

Macrophages in the Tumor Microenvironment Monica Escorcio-Correia and Thorsten Hagemann

Abstract  Solid tumors consist of neoplastic cells, non-malignant stromal cells and migratory haematopoietic cells. Complex interactions between the cell types in this microenvironment regulate tumor growth, progression, metastasis and angiogenesis. There is also strong evidence that this microenvironment is inflammatory and that activation of the innate immune system plays a role in the progression of cancer. One such inflammatory cell that has the potential to promote cancer progression is the macrophage. There is a growing body of pre-clinical and clinical evidence associating abundance of tumor-associated macrophages (TAM) with poor prognosis. According to Condeelis and Pollard, TAM are obligate partners for malignant cell migration, invasion and metastases in many different cancers. These conclusions are based not only on association studies, but also on experiments that show ablation of macrophage function, or their infiltration into experimental tumors, inhibits growth and metastases. The monocyte/macrophage lineage is composed of innate immune cells that can have a variety of functions. Monocytes are produced in the bone marrow and circulate in the blood before being recruited into various tissues and differentiating into tissue-resident macrophages, prototypical phagocytes that play an important role in tissue homeostasis and pathogen clearance. Their ability to produce immunostimulatory cytokines such as interleukin-12 (IL-12) means that they can also play an important role in the activation of other immune cells, for example natural killer (NK) cells and T lymphocytes. In addition, they are also professional antigen presenting cells and can therefore participate in the mounting of the adaptive immune response (Gordon and Taylor 2005). Similarly to what is observed in the context of T cell responses, which can be polarized to a particular phenotype such as Th1 or Th2, macrophages can also be polarized depending on the context in which they are activated. In fact, an M1 and T. Hagemann (*) Barts and The London School of Medicine and Dentistry, Institute of Cancer, Charterhouse Square, London, EC1M 6BQ, UK e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_18, © Springer Science+Business Media, LLC 2010

371

372

M. Escorcio-Correia and T. Hagemann

M2 nomenclature that parallels, to a certain extent, with Th1 and Th2 has also been used for macrophages. The M1 phenotype is observed when macrophages are classically activated by stimuli such as lipolysaccharide (LPS) and interferon gamma (IFN-g). M1 cells are characterized by an enhanced capacity for antigen presentation and cell cytotoxicity, thanks to the increased expression of surface molecules such as MHCII, CD80, CD86, as well as reactive oxygen species and pro-inflammatory cytokines. On the other hand, M2 macrophages are typically activated in the presence of cytokines such as IL-4 and IL-13, usually produced during immune responses to parasites and allergens as part of a Th2 response. M2 macrophages can be characterized by their enhanced expression of mannose and scavenger receptors and arginase (Solinas et al. 2009; Gordon 2003). However, macrophages are plastic cells and similarly to what has been observed for T cells, it can be difficult to assign them to an absolute category as any cell population is likely to be heterogeneous and have characteristics of more than one subtype (Stout et  al. 2009). For example, M1 macrophages are usually characterized by the expression inducible nitric oxide synthase (iNOS) which is required for the metabolism of arginine into nitric oxide (NO), an important mediator in macrophage cytotoxic function (Sica and Bronte 2007). On the other hand, M2 macrophages have higher expression of arginase, which allows them to metabolize arginine into urea and l-ornithine (Sica and Bronte 2007). However, some tumor-associated macrophages (TAMs) simultaneously express high levels of both iNOS and arginase (Tsai et al. 2007; Kusmartsev and Gabrilovich 2005). Tumors are composed not only of malignant cells and can include a variety of cell types, with macrophages sometimes representing a significant proportion of these cells, up to 40% of non-malignant cells in certain types of cancer (Pollard 2004). The macrophage population in tumors is composed not only of resident tissue macrophages but also includes differentiated monocytes that are recruited to the tumor from the blood, attracted by chemokines such as CCL2 (MCP-1), which can be highly expressed in tumors. Other chemoattractants involved in macrophage recruitment that are found to be expressed in the tumor microenvironment include CCL3 (MIP-1a), CCL4 (MIP-1b), CCL5 (RANTES), CXCL12 (SDF-1) and vascular endothelial growth factor (VEGF) (Pollard 2004; Bingle et  al. 2002). In human patients, the abundance of TAMs has been correlated with poor prognosis in certain types of malignancy and several studies have shown that macrophages can contribute to different aspects of tumor growth and progression (Pollard 2004; Bingle et al. 2002). In accordance, the high expression of macrophage growth factors and chemokines such as macrophage-colony stimulating factor (M-CSF) and CCL2, have also been correlated with poor prognosis (Pollard 2004).

Pro-tumor Aspects The role of macrophages in tumor growth has been studied in both experimental and clinical studies. We will discuss evidence that supports the involvement of macrophages in tumor growth and progression, including immune suppression and promotion of angiogenesis and metastasis.

18  Macrophages in the Tumor Microenvironment

373

Immunosuppressive Phenotype Malignant cells have been shown to express unique antigens that are capable of triggering immune responses and potent cytotoxicity from T cells (Dunn et al. 2002). In addition, studies with transgenic animals that are deficient for specific immune cell subsets, including NK cells, NKT and both ab and gd T cells, have shown that the ablation of these cells leads to increased susceptibility to tumors (Dunn et al. 2002). However, despite this evidence supporting the potential of the immune system to participate in tumor clearance, anti-tumor immune responses often seem to be insufficient in preventing tumor growth. One of the reasons for this is the presence of several tumor-derived factors such as IL-10, VEGF and transforming growth factor beta (TGF-b) that render the tumor microenvironment immunosuppressive (Kim et al. 2006; Lewis and Pollard 2006). In fact, it has been shown that there are functional differences between tumor specific cytotoxic T cells depending on whether they are circulating through peripheral lymphoid organs or have infiltrated into tumor sites. Specifically, it has been shown that in melanoma CD8+ T cells collected from unaffected lymph nodes are capable of potent cytotoxicity in vitro towards their cognate antigen (Zippelius et  al. 2004). In contrast, CD8+ T cells with the same antigen specificity and from the same individual, but directly associated with tumor sites are functionally tolerant in vitro (Zippelius et al. 2004). In many tumors, in addition to the malignant cells themselves, two other types of immunosuppressive cells are commonly found: Foxp3+ T regulatory cells (Tregs) and TAMs. The abundance of both these cells types in tumor lesions has been correlated with poor prognosis in human patients and their suppression of anti-tumor immunity has also been demonstrated (Pollard 2004; Khazaie and von Boehmer 2006; Zou 2005). One of the main roles of Tregs in the promotion of tumor growth seems to be the suppression of cytolytic activity by antigen-specific CD8+ T cells, with TGF-b having been identified as an important mediator of suppression (Khazaie and von Boehmer 2006). In addition, Tregs have also been shown to interact with innate immune cells and to play a role in the modulation of the tumor cytokine microenvironment. One study by Taams and colleagues showed that Tregs are capable of skewing macrophages into alternative activation (M2), which has been correlated with a tumor-promoting phenotype (Tiemessen et al. 2007). Following co-culture of human macrophages with Tregs (CD4+CD25+CD127low) or effector T cells (CD4+CD25CD127+), LPS was added to the cultures to induce macrophage activation. It was observed that in the presence of Tregs there was significant upregulation of alternative activation surface markers such as CD206 and CD163 on the macrophages when compared to cultures with effector T cells or macrophages alone (Tiemessen et al. 2007). Moreover, Tregs also prevented the production of pro-inflammatory cytokines, including tumor necrosis factor alpha (TNF-a), IL-6, IL-1, IL-8, macrophage inhibitory protein-1 alpha (MIP-1a) and macrophage chemoattractant protein-1 (MCP-1), likely to result from a defect in NF-kB activation, which the authors also observed (Tiemessen et al. 2007). In tumors, this interaction between TAMs and Tregs is perpetuated as the macrophages are able to recruit CD4+CD25+ T regulatory cells, for example via the production of IL-10 (Zhou et al. 2009).

374

M. Escorcio-Correia and T. Hagemann

In the tumor microenvironment, malignant cells are also capable of polarizing resident macrophages and recruited monocytes to a tumor-promoting M2-skewed phenotype. This polarization prevents the classical (M1) macrophage activation that would enable these cells to mediate tumor cytotoxicity and effectively present antigens to promote an anti-tumor adaptive immune response. This means that, in addition to other effector immune cells such as cytotoxic T cells, macrophages themselves are also suppressed by the tumor microenvironment. Moreover, their acquired TAM phenotype means that these macrophages are not only unable to participate in an anti-tumor immune response, but become suppressor cells themselves. TAMs have been shown to express a gene profile that shares characteristics with that of (alternatively activated) M2 macrophages (Biswas et  al. 2006; Saccani et al. 2006; Lepique et al. 2009; Mantovani et al. 2002). Common patterns of gene expression between these two macrophage subsets include high levels of IL-10, scavenger receptors and mannose receptor, as well as low IL-12. One marked difference between TAMs and M2 macrophages is however, the high expression in TAMs of the interferon-inducible chemokines CCL5, CXCL9, CXCL10 and CXCL16 (Biswas et al. 2006). In any case, and true to their plastic nature, macrophages might have a different phenotype depending on the tumor region in which they are found (Stout et  al. 2009; Biswas et  al. 2008). One example is the modulation of TAM function by hypoxia in tumors, which can induce a particular macrophage phenotype that promotes angiogenesis (Murdoch et al. 2005). There are also studies that have correlated an increase in infiltrating macrophages in certain human tumors with improved prognosis, depending on their location within the tumor, which indicates that these macrophages might maintain an anti-tumor phenotype (Forssell et al. 2007; Kawai et al. 2008; Ohno et al. 2004). Interestingly, another study showed that following chemotherapy a high TAM infiltrate also correlated with increased patient survival (Taskinen et al. 2007). In support of their immunosuppressive phenotype, IL-10 and TGF-b are highly expressed by TAMs and these cytokines, usually produced by Th2 cells and Tregs, can act as potent inhibitors of Th1 cytokines such as IFN-g. The production of prostaglandins and arginase-1 by TAMs has also been implicated in the suppression of T cell anti-tumor immunity (Balkwill and Mantovani 2001). Several studies further indicate that TAMs can thwart immune responses to tumors by promoting the polarization of T cells towards a regulatory rather than a cytotoxic phenotype characterized by increased expression of IL-10 and defective induction of IFN-g (Lepique et al. 2009; Loercher et al. 1999). Other aspects of macrophage-mediated immune suppression in tumors include their poor antigen presenting capacity and their contribution to the differentiation of suppressive dendritic cells (DCs) (Mantovani et al. 2002). A study by Curiel et al. (2003) identified a population of dendritic cells that express B7-H1 on their surface and act to downregulate T cell anti-tumor responses. The authors showed that TAMs, but not normal macrophages, were able to induce B7-H1 expression on DCs via IL-10 and these cells could no longer effectively stimulate T cell proliferation (Curiel et al. 2003).

18  Macrophages in the Tumor Microenvironment

375

Macrophages and Inflammation Another important aspect of the pro-tumor effect of TAMs is their contribution to persistent inflammation in the tumor microenvironment. There are several lines of evidence that indicate a potential link between cancer and inflammation. Malignancies often arise at sites of chronic inflammation and epidemiological studies have shown that chronic infection and inflammation are important cancer risk factors (Karin et al. 2006; Mantovani et al. 2008). In addition, the long-term use of non-steroidal anti-inflammatory drugs (NSAIDs) can protect from certain cancers and in some animal models inhibition of inflammatory responses results in reduced tumor growth. The link between malignancy and inflammation can occur through either extrinsic or intrinsic pathways (Mantovani et al. 2008). The extrinsic pathway refers to cases in which an inflammatory pathology or infection can lead to neoplastic transformation (Mantovani et al. 2008). On the other hand, the intrinsic pathway is characterized by an oncogenic mutation that manifests itself in the tumor microenvironment with features of inflammation that contribute to tumor growth and progression (Mantovani et al. 2008). Macrophages play an important role in innate immune responses but the same molecules that can contribute to efficient pathogen clearance can also cause significant tissue damage during chronic infection/inflammation. These mediators include reactive oxygen and nitrogen intermediates (ROI and RNI) and macrophage migration inhibitory factor (MIF), which in the setting of chronic inflammation can not only cause tissue damage but can also be mutagenic and lead to defective p53 activity. These effects can promote the appearance of pre-malignant cells, further supported by classic pro-inflammatory cytokines that can act as tumor growth factors, for instance TNF-a, IL-6 and IL-1b, all of which are produced by activated macrophages (Pollard 2004; Biswas et al. 2008). In order to study the role of inflammatory cytokines in tumor initiation and progression, one approach has been to target components of signaling pathways required for the transcription of these soluble mediators. For example, the nuclear factor k-B (NF-kB) family of transcription factors not only plays an important role in the activation of macrophages in response to infection and inflammation, but has also been implicated in cancer (Hagemann et  al. 2009). There are five different NF-kB proteins in mammals: RelA (p65), RelB, c-Rel, p50 (NF-kB1) and p52 (NF-kB2), and they can form different homo- or heterodimeric complexes (Perkins 2007). In the absence of the appropriate stimulus, NF-kB dimers are retained in the cytoplasm by IkB inhibitory proteins (IkBa, IkBb and IkBe) that mask the conserved nuclear localization sequence present in NF-kB proteins, thereby preventing their nuclear translocation (Perkins 2007). However, in response to inflammatory cytokines such as TNF-a and IL-1, or bacterial products like LPS that can bind to Toll-like receptors (TLR), IkB proteins are phosphorylated and targeted by ubiquitination for degradation in the proteasome. These phosphorylation events are mediated by IkB kinases, IKKa and IKKb, coupled to the regulatory subunit IKKg (NEMO). Once the IkB protein has been degraded, the NF-kB complex can then

376

M. Escorcio-Correia and T. Hagemann

translocate to the nucleus and bind its target genes to activate transcription. Importantly, positive feedback loops are often established as many NF-kB target genes, for example Tnf (TNF-a), encode proteins that will themselves lead to further NF-kB activation (Perkins 2007). The link between NF-kB and cancer was evident in studies of colitis-associated cancer in which this transcription factor was inhibited in myeloid cells, compromising their ability to induce, amongst other genes, TNF-a, IL-6 and IL-1b (Greten et al. 2004). In these experiments, ablation of IKKb resulted in a significant reduction in tumor incidence as well as tumor size, implicating myeloid cells in both tumor initiation and progression (Greten et al. 2004). In another model of chemically induced carcinogenesis, diethylnitrosamine (DEN)-induced liver cancer, deletion of IKKb and consequent ablation of NF-kB activation also inhibited tumor development (Maeda et al. 2005). Interestingly, this study showed that deletion of IKKb in hepatocytes actually resulted in increased tumor load, which was linked to increased hepatocyte death. DEN is a carcinogen that induces not only DNA damage but also cell death and the observed increase in apoptosis is explained by the lack of NF-kB signaling, which can protect cells against apoptotic cell death. In this case, instead of impeding the expansion of transformed hepatocytes, the increase in cell apoptosis actually induced compensatory cell proliferation, which in the presence of DEN resulted in increased malignancy (Maeda et al. 2005). However, ablation of NF-kB signaling in both hepatocytes and Kupffer cells (resident liver macrophages) had the opposite effect and a decrease in carcinogenesis was observed (Maeda et al. 2005). This decrease in tumor load correlated with decreased hepatocyte cell proliferation as these cells require cytokine growth factors such as TNF-a and IL-6, which are produced by Kupffer cells in an IKKb-dependent manner and are probably induced by factors released by dying hepatocytes (Maeda et al. 2005). Additional studies in a murine model of ovarian cancer also demonstrated that disruption of NF-kB signaling in TAMs resulted in several changes in their gene expression profile, including a reduction in arginase-1, IL-10 and TNF-a and an increase in IL12 and NO production (Hagemann et al. 2008). These changes in TAM phenotype correlated with overall reduced tumor growth, accompanied by an increase in the number of infiltrating NK cells which was dependent on the presence of IL-12 (Hagemann et al. 2008). Interestingly, the role of NF-kB activation in macrophages might be dependent on the stage of tumor development as there are also studies indicating that there is defective NF-kB activation in TAMs from established tumors due to the high expression of nuclear p50 homodimers (Biswas et al. 2006; Saccani et al. 2006).

Role in Angiogenesis and Metastasis An important aspect of tumor growth is the formation of new blood vessels, or neoangiogenesis, which is required to support the continuous expansion of the tumor mass beyond a certain size. TAMs can contribute to tumor angiogenesis

18  Macrophages in the Tumor Microenvironment

377

through the production of pro-angiogenic factors such as VEGF and platelet-derived endothelial cell growth factor. In addition, macrophage-derived cytokines can also promote angiogenesis, for example TNF-a induces matrix metalloproteinase-9 (MMP-9), a protease that cleaves latent VEGF from the extracellular matrix releasing it in its bioactive form (Pollard 2004). In fact, different studies have correlated high macrophage numbers with increased tumor vascularity in humans (Leek et al. 1996; Murdoch et al. 2008) and depletion of macrophages in some murine models of cancer has been shown to result in a significant decrease in the number of tumor blood vessels (Lin et al. 2006; Kimura et al. 2007). The role of macrophages in tumor angiogenesis provides yet another example of how the tumor microenvironment can influence the phenotype of infiltrating macrophages and polarize them to become tumor-promoting cells. Macrophages have been shown to preferentially accumulate in necrotic and hypoxic regions of solid tumors, in which the low oxygen microenvironment induces the upregulation of hypoxia-inducible transcription factors (HIF), HIF1 and HIF2 (Murdoch et  al. 2008). In turn, the HIF transcription factors control the expression of several genes involved in the regulation of not only cell growth and metabolism but also angiogenesis, including Vegfa (VEGF) and Mmp7 (MMP-7), and can polarize macrophages to a pro-angiogenic phenotype (Murdoch et  al. 2005; Burke et  al. 2002; Talks et  al. 2000). The cellular response to hypoxia and the expression of HIF transcription factors is also linked to the NF-kB signaling pathway. It has been shown that in the absence of IKKb, there is a defect in the upregulation of HIF-1a target genes, for example Vegfa, both in vitro and in vivo (Rius et al. 2008). The authors show that nuclear translocation of NF-kB complexes precedes the upregulation of HIF-1a, and the existence of a classical kB site on the Hif1a promoter further supports the hypothesis that Hif1a is an NF-kB target gene (Rius et al. 2008). A particular subpopulation of circulating monocytes has been implicated in the promotion of neoangiogenesis in the tumor microenvironment. These monocytes are identified by their expression of the angiopoietin receptor TIE2 and can be found in the peripheral blood of both humans and mice (Venneri et  al. 2007; De Palma et al. 2005). One study demonstrated that the selective depletion of TIE2expressing monocytes (TEMs) in murine cancer models, despite not affecting recruitment of other monocyte subpopulations into the tumor, resulted in a significant decrease in tumor vascularization (De Palma et al. 2005). In addition, in the presence of TIE2+ human monocytes there was an increase in angiogenesis in a xenograft tumor model when compared to TIE2- cells (Venneri et al. 2007). TEMs do not express CCR2, the surface receptor for CCL2, and might be recruited to tumors by other factors. One such chemoattractant might be angiopoietin 2 itself, found to be highly expressed by tumor cells in hypoxic regions and endothelial cells in tumor blood vessels, both regions in which TEM accumulate in human and mouse tumors (Murdoch et al. 2008). Metastasis refers to the spreading of malignant cell colonies from the primary tumor where they originally developed to other tissues in the body. This process involves different steps in tumor progression including angiogenesis, breakdown of extracellular matrix proteins, and an increase in tumor cell motility, driven by the

378

M. Escorcio-Correia and T. Hagemann

expression of chemokine ligands and their receptors. Macrophages have also been implicated in tumor metastasis. Pollard and colleagues showed that when MMTV-PyMT transgenic mice, susceptible to mammary adenocarcinoma, were crossed to M-CSF deficient mice (Csf1op) there was a significant decrease in tumor metastasis, although there were no differences in the number and size of primary tumors (Lin et  al. 2001). Conversely, the overexpression of M-CSF in the mammary epithelium of these mice led to an increase in lung metastasis, which correlated with an increase in the number of infiltrated macrophages in the primary tumor site (Lin et  al. 2001). These observations support the role of TAMs in metastatic spread in this breast cancer model, as the expression of the M-CSF receptor in these tumors was restricted to the macrophage population (Lin et al. 2001). The same murine tumor model was used to demonstrate that in the absence of M-CSF there is a delay in the transition from premalignant lesions to malignant tumors, which was in turn accelerated when M-CSF overexpression was induced (Lin et al. 2006). It has also been demonstrated that both M-CSF and epidermal growth factor (EGF) can induce malignant cell and macrophage migration in vitro, despite the fact that the receptors for these molecules are not expressed on both cell types (Wyckoff et al. 2004). In fact, the EGF receptor was only found on tumor cells whereas the M-CSF receptor was present in macrophages but not tumor cells. However, the absence of either EGF or M-CSF resulted in the inhibition of migration from both cell types, revealing a collaborating interaction between them (Wyckoff et al. 2004). The communication between malignant cells and TAMs was also illustrated in a multiphoton microscopy study (Wyckoff et  al. 2007). These experiments confirmed that, as previously observed in tissue sections, TAMs accumulate in the periphery of mammary tumors and the smaller numbers of TAMs present in the center of the tumor associate with blood vessels (perivascular). Time-lapse imaging showed that malignant cells have increased motility when they are in proximity to TAMs, towards which they were seen to migrate before binding to blood vessels and entering the circulation (Wyckoff et al. 2007). A recent study by DeNardo et al. (2009) further elucidates the role of TAMs in the promotion of tumor metastasis. The authors also used the MMTV-PyMT breast cancer transgenic mouse model, which they coupled to different transgenic mice deficient for both B and T cells (CD4+ and CD8+ subsets) and compared to mice that were deficient for each lymphocyte subset separately. It was observed that there were no differences in primary tumor incidence or phenotype, but there was a significant decrease in the lung tumor burden resulting from metastatic spread in the absence of CD4+ T cells (DeNardo et al. 2009). This effect was observed in MMTV-PyMT mice that had been crossed to RAG-/- mice (deficient for all B and T cells subsets), CD4-/-CD8-/- and CD4-/-, as well as MMTVPyMT mice treated with a depleting CD4 antibody, but not those selectively depleted of CD4+CD25+ T cells, a subpopulation enriched for Tregs. In the primary tumor, the numbers of infiltrating leukocytes remained the same but there were important differences in the TAM expression profile, which in the CD4-/- mice seemed to indicate a polarization towards an M1 phenotype. This included significantly increased mRNA expression of TNF-a, IL-6, IL-12p40, IL-1b and iNOS and

18  Macrophages in the Tumor Microenvironment

379

decreased expression of M2-associated mRNA products, for instance arginase-1 and TGF-b (DeNardo et  al. 2009). The cytokine profile of CD4+ T cells in this tumor model was also indicative of Th2 polarization (induction of IL-4, IL-13 and IL-10) and in the presence of these CD4+ T cells the induction of M1 macrophages by LPS and IFN-g was inhibited. In fact, the authors showed that the neutralization of IL-4 resulted in a similar phenotype to that of CD4+ T cell deficient mice, including reduced metastatic tumor spread and increased expression of M1 markers in TAMs (DeNardo et al. 2009). Further studies have implicated macrophages in the promotion of tumor metastasis, for example in models of lung cancer that use the Lewis Lung Carcinoma (LLC) cell line. Culture of macrophages with conditioned media from this cell line was shown to induce high levels of IL-6 and TNF-a production by macrophages and versican, an extracellular matrix proteoglycan, was identified as being an important tumor-secreted factor for this response (Kim et al. 2009). Macrophages were shown to respond to versican through TLR2:TLR6 complexes by producing high amounts of TNF-a, a cytokine that promoted LLC tumor growth and metastatic spread. Silencing of versican in the LLC cell line resulted in reduced tumor multiplicity and reduced metastasis to lung, liver and adrenal glands when the tumor cells were implanted subcutaneously (Kim et al. 2009). Another study demonstrated that alveolar macrophages expressing high levels of VEGFR-1 responded to VEGF produced by tumor cells in the primary tumor site, which promoted the upregulation of MMP-9 by both macrophages and lung endothelial cells (Hiratsuka et al. 2002). This induction of MMP-9 only occurred in the presence of a primary tumor and correlated with a significant increase in the metastatic potential of systemically injected tumor cells.

Anti-tumor Potential/ Therapeutic Implications There is ample evidence to support that the abundance of TAMs in human tumors can correlate with poor prognosis, particularly when these macrophages are located in hypoxic/necrotic areas of the tumor (Lewis and Pollard 2006). On the other hand, large numbers of TAMs in the proximity of large, well vascularized areas of tumor has been found to correlate with a good prognosis (Forssell et al. 2007; Kawai et al. 2008; Ohno et al. 2004). In addition, despite the fact that the majority of experimental and clinical data points to macrophages as promoters of tumor growth, these cells also have the potential to lyse tumor cells (Koestler et al. 1987; Chen et al. 2002; Buhtoiarov et al. 2006; Lum et al. 2006; Bracher et al. 2007; Wu et al. 2009). William Coley, a late nineteenth century surgeon from New York, was probably one of the first to induce innate immune activation for the treatment of solid tumors, primarily sarcomas. He began by infecting cancer patients with isolates of bacteria (Coley 1893), and in later studies he used a preparation of “mixed toxins”, slightly less dangerous filtrates from cultures of Streptococcus pyogenes and gram-negative endotoxin-producing Serratia marcasens (Coley 1906). This mixture of bacterial

380

M. Escorcio-Correia and T. Hagemann

toxins probably resulted in the activation of the innate immune system and led to release of TNF, IL-1 and IL-12 by activated macrophages. Coley’s work was controversial and few were able to reproduce the beneficial effects that he had reported but, if the published case histories are to be believed, Coley was able to obtain rapid and sustained responses in patients. Nearly a century later, the use of autologous macrophages activated in vitro with either IFN-g alone or in combination with LPS was tested as a cancer treatment in clinical trials but this approach had only limited success (Hennemann et al. 1998; Andreesen et al. 1990). Much evidence for an anti-tumor potential of macrophages has come from the use of elegant transgenic murine tumor models in which modification of single activation components or inhibition of macrophage influx has been shown to alter the pro-tumor phenotype of TAMs (Hagemann et  al. 2009; Hallam et  al. 2009). Many cancers have a significant macrophage population, but despite their intrinsic anti-tumor potential, epidemiological and experimental data suggests that these cells are more likely to promote tumor growth rather than inhibit it. The concept of targeting TAMs and harnessing their anti-tumoral properties has clear potential for the treatment of cancer however, the experimental approaches used so far are not immediately applicable to the clinic and the systemic impact of a general inhibition of macrophage function is also not fully understood.

References Andreesen R, Scheibenbogen C, Brugger W, Krause S, Meerpohl HG, Leser HG, Engler H, Lohr GW (1990) Adoptive transfer of tumor cytotoxic macrophages generated in  vitro from circulating blood monocytes: a new approach to cancer immunotherapy. Cancer Res 50:7450 Balkwill F, Mantovani A (2001) Inflammation and cancer: back to Virchow? Lancet 357:539 Bingle L, Brown NJ, Lewis CE (2002) The role of tumour-associated macrophages in tumour progression: implications for new anticancer therapies. J Pathol 196:254 Biswas SK, Gangi L, Paul S, Schioppa T, Saccani A, Sironi M, Bottazzi B, Doni A, Vincenzo B, Pasqualini F, Vago L, Nebuloni M, Mantovani A, Sica A (2006) A distinct and unique transcriptional program expressed by tumor-associated macrophages (defective NF-{kappa}B and enhanced IRF-3/STAT1 activation). Blood 107:2112 Biswas SK, Sica A, Lewis CE (2008) Plasticity of macrophage function during tumor progression: regulation by distinct molecular mechanisms. J Immunol 180:2011 Bracher M, Gould HJ, Sutton BJ, Dombrowicz D, Karagiannis SN (2007) Three-colour flow cytometric method to measure antibody-dependent tumour cell killing by cytotoxicity and phagocytosis. J Immunol Methods 323:160 Buhtoiarov IN, Lum HD, Berke G, Sondel PM, Rakhmilevich AL (2006) Synergistic activation of macrophages via CD40 and TLR9 results in T cell independent antitumor effects. J Immunol 176:309 Burke B, Tang N, Corke KP, Tazzyman D, Ameri K, Wells M, Lewis CE (2002) Expression of HIF-1alpha by human macrophages: implications for the use of macrophages in hypoxiaregulated cancer gene therapy. J Pathol 196:204 Chen GG, Chu YS, Chak EC, Leung BC, Poon WS (2002) Induction of apoptosis in glioma cells by molecules released from activated macrophages. J Neurooncol 57:179 Coley WB (1893) The treatment of malignant tumors by repeated inoculations of erysipelas: with a report of ten original cases. Am J Med Sci 105:487

18  Macrophages in the Tumor Microenvironment

381

Coley WB (1906) Late results of the treatment of inoperable sarcoma by the mixed toxins of erysipelas and bacillus prodigiosus. Am J Med Sci 131:375 Curiel TJ, Wei S, Dong H, Alvarez X, Cheng P, Mottram P, Krzysiek R, Knutson KL, Daniel B, Zimmermann MC, David O, Burow M, Gordon A, Dhurandhar N, Myers L, Berggren R, Hemminki A, Alvarez RD, Emilie D, Curiel DT, Chen L, Zou W (2003) Blockade of B7-H1 improves myeloid dendritic cell-mediated antitumor immunity. Nat Med 9:562 De Palma M, Venneri MA, Galli R, Sergi Sergi L, Politi LS, Sampaolesi M, Naldini L (2005) Tie2 identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitors. Cancer Cell 8:211 DeNardo DG, Barreto JB, Andreu P, Vasquez L, Tawfik D, Kolhatkar N, Coussens LM (2009) CD4+ T cells regulate pulmonary metastasis of mammary carcinomas by enhancing protumor properties of macrophages. Cancer Cell 16:91 Dunn GP, Bruce AT, Ikeda H, Old LJ, Schreiber RD (2002) Cancer immunoediting: from immunosurveillance to tumor escape. Nat Immunol 3:991 Forssell J, Oberg A, Henriksson ML, Stenling R, Jung A, Palmqvist R (2007) High macrophage infiltration along the tumor front correlates with improved survival in colon cancer. Clin Cancer Res 13:1472 Gordon S (2003) Alternative activation of macrophages. Nat Rev Immunol 3:23 Gordon S, Taylor PR (2005) Monocyte and macrophage heterogeneity. Nat Rev Immunol 5:953 Greten FR, Eckmann L, Greten TF, Park JM, Li ZW, Egan LJ, Kagnoff MF, Karin M (2004) IKKbeta links inflammation and tumorigenesis in a mouse model of colitis-associated cancer. Cell 118:285 Hallam S, Escorcio-Correia M, Soper R, Schultheiss A, Hagemann T (2009) Activated macrophages in the tumour microenvironment-dancing to the tune of TLR and NF-kappaB. J Pathol 219:143 Hagemann T, Lawrence T, McNeish I, Charles KA, Kulbe H, Thompson RG, Robinson SC, Balkwill FR (2008) “Re-educating” tumor-associated macrophages by targeting NF-{kappa} B. J Exp Med 205:1261 Hagemann T, Biswas SK, Lawrence T, Sica A, Lewis CE (2009) Regulation of macrophage function in tumors: the multifaceted role of NF-{kappa}B. Blood 113:3139 Hennemann B, Beckmann G, Eichelmann A, Rehm A, Andreesen R (1998) Phase I trial of adoptive immunotherapy of cancer patients using monocyte-derived macrophages activated with interferon gamma and lipopolysaccharide. Cancer Immunol Immunother 45:250 Hiratsuka S, Nakamura K, Iwai S, Murakami M, Itoh T, Kijima H, Shipley JM, Senior RM, Shibuya M (2002) MMP9 induction by vascular endothelial growth factor receptor-1 is involved in lung-specific metastasis. Cancer Cell 2:289 Karin M, Lawrence T, Nizet V (2006) Innate immunity gone awry: linking microbial infections to chronic inflammation and cancer. Cell 124:823 Kawai O, Ishii G, Kubota K, Murata Y, Naito Y, Mizuno T, Aokage K, Saijo N, Nishiwaki Y, Gemma A, Kudoh S, Ochiai A (2008) Predominant infiltration of macrophages and CD8(+) T cells in cancer nests is a significant predictor of survival in stage IV nonsmall cell lung cancer. Cancer 113:1387 Khazaie K, von Boehmer H (2006) The impact of CD4+CD25+ Treg on tumor specific CD8+ T cell cytotoxicity and cancer. Semin Cancer Biol 16:124 Kim R, Emi M, Tanabe K, Arihiro K (2006) Tumor-driven evolution of immunosuppressive networks during malignant progression. Cancer Res 66:5527 Kim S, Takahashi H, Lin W-W, Descargues P, Grivennikov S, Kim Y, Luo J-L, Karin M (2009) Carcinoma-produced factors activate myeloid cells through TLR2 to stimulate metastasis. Nature 457:102 Kimura YN, Watari K, Fotovati A, Hosoi F, Yasumoto K, Izumi H, Kohno K, Umezawa K, Iguchi H, Shirouzu K, Takamori S, Kuwano M, Ono M (2007) Inflammatory stimuli from macrophages and cancer cells synergistically promote tumor growth and angiogenesis. Cancer Sci 98:2009 Koestler TP, Johnson WJ, Rieman D, Dalton BJ, Greig RG, Poste G (1987) Differential expression of murine macrophage-mediated tumor cytotoxicity induced by interferons. Cancer Res 47:2804

382

M. Escorcio-Correia and T. Hagemann

Kusmartsev S, Gabrilovich DI (2005) STAT1 signaling regulates tumor-associated macrophage-mediated T cell deletion. J Immunol 174:4880 Leek RD, Lewis CE, Whitehouse R, Greenall M, Clarke J, Harris AL (1996) Association of macrophage infiltration with angiogenesis and prognosis in invasive breast carcinoma. Cancer Res 56:4625 Lepique AP, Daghastanli KRP, Cuccovia IM, Villa LL (2009) HPV16 tumor associated macrophages suppress antitumor T cell responses. Clin Cancer Res 15:4391 Lewis CE, Pollard JW (2006) Distinct role of macrophages in different tumor microenvironments. Cancer Res 66:605 Lin EY, Nguyen AV, Russell RG, Pollard JW (2001) Colony-stimulating factor 1 promotes progression of mammary tumors to malignancy. J Exp Med 193:727 Lin EY, Li J-F, Gnatovskiy L, Deng Y, Zhu L, Grzesik DA, Qian H, Xue X-N, Pollard JW (2006) Macrophages regulate the angiogenic switch in a mouse model of breast cancer. Cancer Res 66:11238 Loercher AE, Nash MA, Kavanagh JJ, Platsoucas CD, Freedman RS (1999) Identification of an IL-10-producing HLA-DR-negative monocyte subset in the malignant ascites of patients with ovarian carcinoma that inhibits cytokine protein expression and proliferation of autologous T cells. J Immunol 163:6251 Lum HD, Buhtoiarov IN, Schmidt BE, Berke G, Paulnock DM, Sondel PM, Rakhmilevich AL (2006) In vivo CD40 ligation can induce T cell-independent antitumor effects that involve macrophages. J Leukoc Biol 79:1181 Maeda S, Kamata H, Luo J-L, Leffert H, Karin M (2005) IKK[beta] couples hepatocyte death to cytokine-driven compensatory proliferation that promotes chemical hepatocarcinogenesis. Cell 121:977 Mantovani A, Sozzani S, Locati M, Allavena P, Sica A (2002) Macrophage polarization: tumorassociated macrophages as a paradigm for polarized M2 mononuclear phagocytes. Trends Immunol 23:549 Mantovani A, Allavena P, Sica A, Balkwill F (2008) Cancer-related inflammation. Nature 454:436 Murdoch C, Muthana M, Lewis CE (2005) Hypoxia regulates macrophage functions in inflammation. J Immunol 175:6257 Murdoch C, Muthana M, Coffelt SB, Lewis CE (2008) The role of myeloid cells in the promotion of tumour angiogenesis. Nat Rev Cancer 8:618 Ohno S, Ohno Y, Suzuki N, Kamei T, Koike K, Inagawa H, Kohchi C, Soma G-I, Inoue M (2004) Correlation of histological localization of tumor-associated macrophages with clinicopathological features in endometrial cancer. Anticancer Res 24:3335 Perkins ND (2007) Integrating cell-signalling pathways with NF-[kappa]B and IKK function. Nat Rev Mol Cell Biol 8:49 Pollard JW (2004) Tumour-educated macrophages promote tumour progression and metastasis. Nat Rev Cancer 4:71 Rius J, Guma M, Schachtrup C, Akassoglou K, Zinkernagel AS, Nizet V, Johnson RS, Haddad GG, Karin M (2008) NF-[kgr]B links innate immunity to the hypoxic response through transcriptional regulation of HIF-1[agr]. Nature 453:807 Saccani A, Schioppa T, Porta C, Biswas SK, Nebuloni M, Vago L, Bottazzi B, Colombo MP, Mantovani A, Sica A (2006) p50 nuclear factor-{kappa}B overexpression in tumor-associated macrophages inhibits M1 inflammatory responses and antitumor resistance. Cancer Res 66:11432 Sica A, Bronte V (2007) Altered macrophage differentiation and immune dysfunction in tumor development. J Clin Invest 117:1155 Solinas G, Germano G, Mantovani A, Allavena P (2009) Tumor-associated macrophages (TAM) as major players of the cancer-related inflammation. J Leukoc Biol jlb.0609385 Stout RD, Watkins SK, Suttles J (2009) Functional plasticity of macrophages: in situ reprogramming of tumor-associated macrophages. J Leukoc Biol jlb.0209073

18  Macrophages in the Tumor Microenvironment

383

Talks KL, Turley H, Gatter KC, Maxwell PH, Pugh CW, Ratcliffe PJ, Harris AL (2000) The expression and distribution of the hypoxia-inducible factors HIF-1{alpha} and HIF-2{alpha} in normal human tissues, cancers, and tumor-associated macrophages. Am J Pathol 157:411 Taskinen M, Karjalainen-Lindsberg ML, Nyman H, Eerola LM, Leppa S (2007) A high tumorassociated macrophage content predicts favorable outcome in follicular lymphoma patients treated with rituximab and cyclophosphamide-doxorubicin-vincristine-prednisone. Clin Cancer Res 13:5784 Tiemessen MM, Jagger AL, Evans HG, van Herwijnen MJC, John S, Taams LS (2007) CD4+CD25+Foxp3+ regulatory T cells induce alternative activation of human monocytes/ macrophages. Proc Natl Acad Sci U S A 104:19446 Tsai C-S, Chen F-H, Wang C-C, Huang H-L, Jung S-M, Wu C-J, Lee C-C, McBride WH, Chiang C-S, Hong J-H (2007) Macrophages from irradiated tumors express higher levels of iNOS, arginase-I and COX-2, and promote tumor growth. Int J Radiat Oncol Biol Phy 68:499 Venneri MA, Palma MD, Ponzoni M, Pucci F, Scielzo C, Zonari E, Mazzieri R, Doglioni C, Naldini L (2007) Identification of proangiogenic TIE2-expressing monocytes (TEMs) in human peripheral blood and cancer. Blood 109:5276 Wu Q-L, Buhtoiarov IN, Sondel PM, Rakhmilevich AL, Ranheim EA (2009) Tumoricidal effects of activated macrophages in a mouse model of chronic lymphocytic leukemia. J Immunol 182:6771 Wyckoff J, Wang W, Lin EY, Wang Y, Pixley F, Stanley ER, Graf T, Pollard JW, Segall J, Condeelis J (2004) A paracrine loop between tumor cells and macrophages is required for tumor cell migration in mammary tumors. Cancer Res 64:7022 Wyckoff JB, Wang Y, Lin EY, Li J-f, Goswami S, Stanley ER, Segall JE, Pollard JW, Condeelis J (2007) Direct Visualization of macrophage-assisted tumor cell intravasation in mammary tumors. Cancer Res 67:2649 Zhou J, Ding T, Pan W, Zhu LY, Li L, Zheng L (2009) Increased intratumoral regulatory T cells are related to intratumoral macrophages and poor prognosis in hepatocellular carcinoma patients. Int J Cancer 125:1640 Zippelius A, Batard P, Rubio-Godoy V, Bioley G, Lienard D, Lejeune F, Rimoldi D, Guillaume P, Meidenbauer N, Mackensen A, Rufer N, Lubenow N, Speiser D, Cerottini J-C, Romero P, Pittet MJ (2004) Effector function of human tumor-specific CD8 T cells in melanoma lesions: a state of local functional tolerance. Cancer Res 64:2865 Zou W (2005) Immunosuppressive networks in the tumour environment and their therapeutic relevance. Nat Rev Cancer 5:263

Chapter 19

The Prognostic Significance of Tumor-Infiltrating Lymphocytes Ping Yu and Yang-Xin Fu

Abstract  Tumors have been known to contain variable numbers of lymphocytes, referred to as tumor infiltrating lymphocytes (TILs). The degree of the lymphocytic infiltrate has been shown to correlate with positive outcome in some types of cancers. However, there are conflicting reports regarding the prognosis value of TIL even with the same types of cancers. Much research has gone into classifying TILs with respect to antigen receptor structure and the antigen to which the tumor-specific T cells react. However, these studies for the most part did not immunophenotype TILs, and recent data has revealed that the composition of TIL is not homogenous, but rather represents varying contributions from many lymphocytic subsets. Furthermore, the function of TILs is often compromised as a result of the accumulation of immunoregulatory cells and various tumor escape mechanisms. These recent findings stress the need to collect more data on the composition and function of TIL infiltrates before definitive conclusions about the prognostic significance of TILs can be drawn. This chapter summarizes the functional significance of different subsets of lymphocytes infiltrating the tumors.

Introduction The malignant tumors were noticed to contain variable numbers of lymphocytes since more than 100 years ago (Balkwill and Mantovani 2001). These lymphocytes have since been known as tumor-infiltrating lymphocytes (TIL). At the beginning, the TIL were thought to reflect the origin of cancer at the sites

P. Yu (*) The Committee on Immunology and Department of Pathology and Section of Dermatology/ Department of Medicine, University of Chicago, 5841 S. Maryland Ave, Chicago, IL 60637, USA e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_19, © Springer Science+Business Media, LLC 2010

385

386

P. Yu and Y.-x. Fu

of chronic inflammation (Balkwill and Mantovani 2001), and later it was debated whether TIL provided a favorable environment for cancer growth or were evidence of the host’s attempt to eliminate cancer (Rollins 2006). This debate continues until today. The relationship between the extent of immune cell infiltration and prognosis might give us some hint what these cells are doing at the site of cancers. This relationship was first identified in 1949 in cases of breast cancer (Moore and Foote 1949). Since then, it is the significance of such inflammatory cells within or at the periphery of solid tumors that has long been the subject of conflicting reports (Clark et  al. 1989; Clemente et al. 1996; Curiel et al. 2004; Zhang et al. 2003). Several studies have demonstrated that the increase of TIL is associated with better prognosis (Clark et  al. 1989; Clemente et  al. 1996; Zhang et  al. 2003). However a sweeping assumption that the influx of lymphocytes to the tumor site is invariably beneficial to the patient may be a bit presumptuous. Recent studies suggest that the type, not the quantity, of tumor-infiltrating cells may be more of a critical determinant for the prognosis (Curiel et al. 2004). For example, in the case of infiltrating regulatory CD4+ cells to the tumor site, it has been shown to be more deleterious than favorable for the patient (Curiel et  al. 2004). This subpopulation of TILs may behave in a more sinister manner to the detriment of the host’s defense against the malignant cells (Yu et al. 2005; Turk et  al. 2004; Tawara et  al. 2002; Li et  al. 2003; Golgher et  al. 2002; Casares et al. 2003; Machiels et al. 2001; Sutmuller et al. 2001; Steitz et al. 2001; Jones et  al. 2002; Shimizu et  al. 1999; Onizuka et  al. 1999). Another example is the tumor-induced Th17 cells that are producing cytokine IL17. A recent report showed that IL-17 production by CD4+ cells led to myeloid cell recruitment into the tumor microenvironment and enhanced tumor growth in the ovarian cancers (Charles et  al. 2009). Conversely, the antitumor lymphocytes migrating to the tumor site may become compromised once within the tumor milieu or become adversely adapted to the suppressive environment to promote growth instead of regression. In this chapter, we will strive to illustrate the type of T lymphocytes that behave either as friends or as foes in antitumor immunity. By understanding the functional peculiarities of the lymphocytic subtypes, it may serve to better explain the observable paradox that the evidence of TIL does not always correlate with better prognosis for the patient (Zhang et al. 2003; Marrogi et al. 1997; Naito et al. 1998; Schumacher et al. 2001; Saiki et al. 1996; Nakano et al. 2001; Banner et al. 1990; Kolbeck et al. 1992; Kowalczyk et al. 1997; Igarashi et al. 1992).

Antitumor Functions of T Lymphocytes The mature T cell population is composed of ab T cells expressing CD4 or CD8 and the CD4-/CD8- gd T cell receptor (TCR)-expressing cells. The unique function of CD4+ and CD8+ T cells is dictated by the expression of these coreceptors, CD4

19  The Prognostic Significance of Tumor-Infiltrating Lymphocytes

387

or CD8, for which the ligand is b2 domain of the major histocompatibility complex (MHC) class II molecule (Cammarota et al. 1992) and the a3 domain of MHC class I molecules (Potter et al. 1989), respectively. Due to these specificities, the ab TCR of CD8+ T cells is restricted to the recognition of antigens presented by MHC class I molecules and the ab TCR of CD4+ T cells to antigens presented by MHC class II molecules. However, relatively little is known about the function of gdT in antitumor immunity so far, thus this review will mainly focus on the ab T cells. The important role of T cells as effectors in antitumor immunity was first shown in numerous murine models. For instance, UV light-induced regressor tumors are rejected regularly by normal mice but grow progressively in the absence of T cells (Ward et al. 1990; Kripke 1974; Spellman and Daynes 1981). It has also been demonstrated convincingly that T cell-mediated immunity is essential for the rejection of virally and chemically induced tumors (Leclerc et al. 1972; Tevethia et al. 1974; Cheever et al. 1986; Cheever et al. 1986; Klein et al. 1960; Boyse et al. 1962; Rouse et  al. 1972). For example, in the model of murine chemical carcinogen methylcholanthrene-induced tumors, it was demonstrated that intravenous injection of immune cells, but not of immune serum, could transfer systemic tumor-specific immunity into sublethally irradiated mice (Boyse et  al. 1962). These results are consistent with studies showing that the protective immunity against a plasma-cell tumor was abolished by prior depletion of T-lymphocytes via anti-T cell antibodies and complement (Rouse et al. 1972). The importance of T cells in tumor immunity has also been implicated in human studies, albeit with discernable limitations. While most murine tumor models have the advantage of utilizing antigen-specific T cells generated from tumor-free syngeneic mice, generating a human homologue is not feasible. Thus, in human studies, T-lymphocytes are isolated from peripheral blood (Mukherji and MacAlister 1983; Knuth et al. 1984) or from the tumor (Klein et al. 1980) of cancer patients. Such T cells can react in vitro with autologous cancer cells (Fossati et al. 1987). Utilizing adoptive transfer of in vitro expanded TIL in combination with chemotherapy, recent clinical trials have shown up to 50% positive response rate in selected patients with late stage aggressive cancers (Rosenberg 2001a, b; Rosenberg et al. 2004; Dudley et al. 2002; Dudley and Rosenberg 2003).

CD8+ T Cells While the significance of lymphocytes in tumor immunity is rarely disputed, the relative importance of various T cell subsets in tumor rejection is the subject of great controversies (Robins 1986). Broadly speaking, since cancer is a disease caused by an array of mutations in various types of cells, differences in the T cell subsets required for mediating its rejection is not altogether surprising. One such subset is the CD8+ cytotoxic T lymphocytes (CTLs). Most tumors are positive for MHC class I but negative for MHC class II, and CTLs are able to induce tumor killing upon direct recognition of peptide antigens, presented by the tumor’s MHC

388

P. Yu and Y.-x. Fu

class I molecules. Thus, the initial attention to antitumor immune responses was preferentially given to CD8+ T cells. That the CD8+ T cells are critical effectors against tumor cells is further supported by numerous studies in murine models. For instance, for certain tumors, such as UV light-induced tumors, the CD8+ cytolytic T cell subset appears to be regularly required for rejection (Ward et  al. 1990). Murine CTLs that kill tumor targets in vitro can be freshly isolated from mice after repeated intraperitoneal injection of antigenic tumor cells or can be generated ex vivo in a 7-day mixed lymphocyte-tumor cell culture. Elimination of CD8+ T cells from mice either via depleting antibodies or using genetic knockout can, at least partially, abrogate the antitumor immunity induced by most cancer vaccines (Fearon et al. 1990; Golumbek et al. 1991; Dranoff et al. 1993; Lin et al. 1996). Consistent with the murine models, clinical data from the cancer patients uphold the importance of CD8+ T cells in bringing forth an antitumor response. Adoptive transfer studies of in vitro-stimulated CD8+ T cell lines and CD8+ clones specific for tumor antigens effectively mediate antitumor immunity when transferred back into tumor-bearing hosts (Rosenberg 1999, 2001a, b). Furthermore, recent reports suggest that immunization, using either adjuvant or dendritic cells (DCs) with pure tumor peptides, can result in productive antitumor immunity that is restricted by MHC class I (Noguchi et al. 1995; Feltkamp et al. 1993). Taken together, CD8+ T cells in tumor immunity can be unquestionably heralded as one of the principal subsets of T cells that constructively mediate an effective antitumor response.

CD4+ T Cells Undeniably, CD4+ T cells are an integral part of adaptive immunity, but the specific role they play in mounting an antitumor response remains dubious and somewhat conflicting. The critical function of CD4+ T cells in promoting immunity has been consistently demonstrated by vaccine and challenge experiments employing antibody-mediated depletion of CD4+ T cells or by using CD4-knockout mice (Fearon et  al. 1990; Golumbek et  al. 1991; Dranoff et  al. 1993; Lin et  al. 1996; Levitsky et  al. 1994; Ostrand-Rosenberg 1994; Pulaski et  al. 1993; Cavallo et  al. 1992; Hock et al. 1991; Pan et al. 1995). More importantly, similar to CD8+ T cells, tumor-specific CD4+ T cell that can recognize tumor antigens do exist and evidence of migration to the tumor site has been shown in both murine models and human cancer tissues (Pardoll and Topalian 1998; Beck-Engeser et al. 2001; Monach et al. 1995). However, complications arise when the accumulation of CD4+ T cells inside the tumor microenvironment during tumor progression seemingly hinder the effector function of CD8+ T cells (Yu et al. 2005; Awwad and North 1988; Berendt and North 1980; Bursuker and North 1984; Wang et al. 2004). The hallmark phenotype of the CD4+ T cell that impedes the antitumor response seems to harbor characteristics akin to regulatory cells (Curiel et al. 2004; Dieckmann et al. 2001; Liyanage et al. 2002; Woo et al. 2001, 2002). This paradoxical dualism of CD4+ T cells obligates further differentiation of this subtype into helper and regulatory CD4+ T cells.

19  The Prognostic Significance of Tumor-Infiltrating Lymphocytes

389

Although CD4+ T cells have been shown to be sufficient to eliminate tumor cells in the absence of CD8+ T cells in some tumor models (Monach et al. 1995; Beatty and Paterson 2001; Fujiwara et  al. 1984; Greenberg et  al. 1981; Mumberg et  al. 1999), it is more often the case that both CD4+ and CD8+ are required for effective tumor rejection to occur (Beatty and Paterson 2000). In part, this is due to a substantial portion of tumor cells expressing only the class I MHC molecules, not the class II molecules, thereby limiting direct recognition by the CD4+ T cells. Moreover, the predominant effector mechanism in tumor immunity is by direct lysis of tumor cells by the MHC class I recognizing CD8+ CTL, the role of CD4+ T cells in antitumor responses is often to aid in the activation of CD8+ T cells leading to the destruction of the tumor by CD8+ CTL. The CD4+ T cell help of the CD8+ CTLs in tumor immunity can be divided into three phases: early activation, effector, and memory. CD4+ T Cells Help for Cytotoxic T Lymphocytes Induction It has been debated for decades if CD4+ T cells are required for the priming of CTL. Accumulating evidence has indicated that for the induction of tumor-­ specific CD8+ T cell responses, cross-presentation of antigens that have been captured by professional antigen presenting cells (APCs) such as DCs plays a dominant role (Huang et al. 1994, 1996; Toes et al. 1996; Yu et al. 2003; Spiotto et  al. 2002). CD4+ T cell help has been hinted to be essential for such crosspriming in the induction of CTL immunity. This requirement for CD4 help is believed at least in part to activate APC (Ridge et  al. 1998), which in turn expresses costimulatory molecules such as ICAM-1 (Shinde et al. 1996), CD80, and CD86 (Cella et al. 1996), or to secrete cytokines including IL-12 (Cella et al. 1996; Koch et al. 1996). These factors are essential for better CD8+ T cell activation. Most T cell help for CTL priming is dependent on the interaction between CD40L expressed by CD4+ T cells and CD40 on APC (Lu et al. 2000; Schoenberger et  al. 1998; Bennett et  al. 1997, 1998). The CD40–CD40L interaction has also been proved important in the generation of protective T cell-­mediated tumor immunity (Mackey et  al. 1997, 1998). The requirement for CD4+ T cells to “license” APC for the priming of CLT helps explaining some scenarios in which the induction of CTL can be achieved in the absence of CD4+ T cells. The typical CTL priming that is independent of CD4+ T cells is via the direct activation of DCs by virus that provides the optimal inflammatory signal (Stevenson et  al. 1998) to activate DCs, which can subsequently prime antigen-specific CTL responses in the absence of CD4+ T cells (MacDonald and Johnston 2000). Evidence supporting CD4 independent CTL induction via activated DC is further demonstrated by the data showing that CD40-mediated activation with soluble ligand or activating antibodies licensed DC for cross-priming in the absence of CD4+ T cells (Mackey et al. 1997, 1998). However, these observations have not been confirmed in all experimental ­models. One study has shown that injection of MHC class I-deficient tumor cells

390

P. Yu and Y.-x. Fu

into MHC class II knockout mice resulted in the induction of tumor cell-specific CTL responses (Wolkers et al. 2001), demonstrating that cross-priming can occur even in the absence of CD4+ T cells, albeit with decreased efficiency. It cannot be excluded that the CD4 help required for the cross-priming of CLT was bypassed in this model because DCs were activated by cellular components released during tumor cell apoptosis, which can serve as adjuvant as shown in one study (Shi et al. 2000). Other evidence seems to suggest that high level of antigen can bypass CD4 help for CTL induction. For example, RMA-S cells loaded with a MHC class I-restricted peptide can induce CTL responses in vitro. The observed in vitro CTL priming was independent of CD4+ T cells or MHC class II-expressing cells, but was dependent upon the level of MHC class I expression on the RMA-S cells (De Bruijn et  al. 1991). A high expression level of MHC class I peptide complexes on the peptide presenting cells was the decisive requirement for the induction of CD4+ T cellindependent CTL response in this model. This hypothesis would explain other observations in the vaccination experiment using MHC class I-restricted peptides emulsified in noninflammatory incomplete Freund’s adjuvant, robust CTL function was induced in mice depleted of CD4+ T cells (Fayolle et al. 1996). Even with high level of antigen provided for CLT priming, other conditions may be required for CTL induction in the absence of CD4 help. Our own study has showed that CD4 help is dispensable for cross-priming of CTL when intact draining lymph nodes are available (Yu et al. 2003). In the absence of draining lymphoid tissue, this additional “help” becomes essential for the proliferation of naïve CD8+ T cells (Yu et al. 2003). Other observation appropriately illustrates that the CD4+ help even in the presence of activated DCs cannot be altogether excluded or ignored, suggesting that CD4+ T cells may play other roles in CTL induction than “licensing” DCs. For example, in experiments utilizing vaccinations with ovalbumin (OVA)-transduced, CD40-activated DCs, only in the presence of CD4+ T cells was there protection against formation of OVA-expressing tumors (Schnell et al. 2000). Furthermore, in some studies, vaccination with transduced DCs or DCs pulsed with peptides, tumor lysates, or tumor cell-derived exosomes has been shown to be at least partially CD4+ T cell-dependent, although, it is possible that DCs were not properly activated prior to injection (Wolfers et al. 2001; Tuting et al. 1999; Fields et al. 1998; Porgador and Gilboa 1995). In summary, the available data suggest that CD4+ T cells are generally required for CTL priming when DCs are not activated through another mechanism. However, other conditions, such as the level of antigen provided or the integrity of draining lymphoid tissues, may also impact on the requirement for CD4 help. Nevertheless, CD4+ T cell help has been considered essential for the induction of CTL responses against tumors in the most cases. Under the noninflammatory condition of the majority of the cancers in addition to the unavailability of tumor antigens for cross-presentation (Spiotto et al. 2002), DCs require activation by CD4+ T cells before they can induce the full activation and differentiation of naïive CD8+ T cells into CTLs.

19  The Prognostic Significance of Tumor-Infiltrating Lymphocytes

391

CD4+ T Cells for Maintenance of a Cytotoxic T Lymphocytes Response While the necessity of helper CD4+ T cells in the induction of CTL remains conflicting, it is generally accepted that CD4+ T cells are critical for the maintenance of CTL in both virally directed and tumor-specific immune response (Battegay et al. 1994; Matloubian et al. 1994; Cardin et al. 1996; Zitvogel et al. 1996). As clearly demonstrated in a human cytomegalovirus (CMV) infection, the survival of adoptively transferred anti-CMV CTLs were dependent on the presence of CMV-specific CD4+ T cells persisting in the host (Schomig et al. 1996). For the host to perpetually sustain an effector CTL function to counter a persisting virus infection may require help from CD4+ T cell through either cytokine secretions or stimulations that is independent of DC help. In certain reports, the need for helper CD4+ T cell can be replaced by an exogenously provided IL-2 (Kast et al. 1986, 1989; Bear et al. 1988), suggesting that the requirement for CD4+ T cells may simply be to supplement CD8+ T cells with IL-2, a cytokine traditionally thought to be essential for promoting growth and proliferation of T cells. Whether IL-2 is the primary mechanism or a byproduct utilized by CD4+ to provide help for the maintenance of CTL remains to be determined. However, the contribution of CD4+ help in sustaining tumor-specific CTL response is unequivocally recognized as an important means to an end. There is no shortage of corroborating data from therapeutic vaccination to adoptive transfer studies that necessitates CD4+ help in sustaining viable CTL function. In some murine models, while tumor growth was prevented by vaccinations with peptide-pulsed DC-mediated CTL induction, the same therapeutic immunization against established tumor, a scenario mimicking chronic persistence of antigen, required CD4+ T cells help (Zitvogel et al. 1996). Similarly, while adoptive transfer of tumor-specific CTLs prevented tumor formation in mice with CD4+ T cells depleted or MHC class II knockout mice, systemic metastases, a condition may require persistence of effector CTLs, could not be cured unless the hosts bear CD4+ T cells (Hu et al. 2000). Thus, as the tumor matures, CD4+ help becomes vital to the persistence of the effector response. Published reports on clinical trials of adoptive immunotherapy further support this concept (Dudley and Rosenberg 2003). Co-transfer of CD4+ T cells with CD8+ T cells expanded from autologous tumor infiltrating T cells prolonged the survival of the adoptively transferred T cells (Dudley and Rosenberg 2003). Taken together, it is evident that the longevity of a tumor-specific CTL response is stimulated by the presence of CD4+ T cells. CD4+ T Cells for the Induction and Maintenance of CD8+ T Cell Memory Responses After the clearance of the antigen, the greater part of the effector CTLs apoptosis while a marginal portion convert into lymphocytes with a memory phenotype; thus upon a second antigen encounter, CTL response propagates with alacrity and potency to provide the host with protection. It is generally accepted that memory cells persist in circulation subsequent to effector response, however, the role of

392

P. Yu and Y.-x. Fu

CD4+ T cells in the induction and maintenance of memory CTLs has been subjected to great controversy. While recent publications seem to have reached a consensus that CD4+ T cells are indispensable for an intact CTL memory response (Janssen et al. 2003; Shedlock and Shen 2003; Sun and Bevan 2003; Sun et al. 2004; Bevan 2004), several studies have suggested that CD4+ T cells are required in the primary responses to “program” the CD8+ T cells to differentiate into long-lived functional memory cells (Janssen et al. 2003; Shedlock and Shen 2003); others have argued that CD4+ T cells are required after antigen is eliminated to maintain the number or normal functions of CD8+ memory T cells (Sun et al. 2004; Bevan 2004). One earlier study reported a CD4+ T cell and B cell-independent persistence of memory H-Y-specific CTLs (Di Rosa and Matzinger 1996). Two other groups also found that virus-specific peptide-tetramer positive memory cells persisted in CD4 or MHC class II knockout mice using viral infection models (Stevenson et al. 1998; Zajac et  al. 1998). However, despite the clear presence of virus-specific CD8+ memory T cells, both groups were unable to revert memory T cells into effector T cells in the absence of CD4+ T cells (Stevenson et al. 1998; Zajac et al. 1998). In other words, the memory phenotype CD8+ T cells were nonfunctional. Similar results were obtained in the previously described mouse H-Y model, where exogenous cytokine addition was required to revert memory cells back into CTLs in the absence of CD4+ T cells (Di Rosa and Matzinger 1996). These earlier findings suggest that CD4+ T cells were essential for the conversion of memory CTLs into effectors upon secondary antigen encounter. Other groups have expanded on the aforementioned studies by utilizing adoptive transfer experiments to exemplify the necessity of CD4+ in sustaining the memory cells in circulation. They complemented earlier studies with MHC class II knockout mice in adoptive transfer of memory T cells to illustrate that in the absence of CD4+ help, reactivation into effectors could not occur (Hu et al. 2000). While these earlier studies did not address if CD4+ T cells were required in the priming phase of CLT to “program” them to later develop to long-lived functional memory cells, recent studies by Bevan’s group utilizing techniques that allowed tracking of antigen specific memory T cells concluded that CD4+ T cells are not required during the priming of CTL, but only required at the later stage for the maintenance of memory CD8+ T cells (Sun et al. 2004). However, contradicting debate ensued with two recent publications by Shedlock and Shen (2003) and Janssen et al. (2003), respectively, claiming the opposite of previous studies. They were able to show that adoptively transferred CTLs survive and function properly in an environment without CD4+ T cells after being primed in its presence. In addition, they show that they were not required for the secondary activation of CTLs. Taken together, the data available so far suggested a general requirement for CD4 help to generate a healthy CD8 memory. However, at which stage of CD8 T cell responses CD4 help is essential, or the nature of such help still await to be determined. T helper 1 Versus T helper 2 Responses for Antitumor Immunity The CD4+ T cell responses can be divided into different types depending on their cytokine profile (Mosmann and Sad 1996). The defined T helper 1 (Th1) cells are

19  The Prognostic Significance of Tumor-Infiltrating Lymphocytes

393

characterized by the production of IFN-g among others, whereas T helper 2 (Th2) cells produce IL-4, IL-5, and others. The balance between Th1 and Th2 cytokines has definite influence on the outcome of various immune responses as Th1 preferentially induces cellular immunity and Th2 tends to elicit humoral immunity (Romagnani 1997). The cytokine IFN-g positively impact on antigen processing and presentation, because MHC class I and II and the expression of several other molecules such as TAP and proteasome components are under the control of this cytokine (York et al. 1999). Therefore, the Th1 response is generally correlated with a better cellular and CTL response. Because a cellular immune response is preferable for tumor destruction, a Th1 response has been proposed to be beneficial for antitumor immunity. Several reports have supported this idea and demonstrated the parallelism between the generation of a Th1 response and a stronger antitumor immunity (Tsung et  al. 1997; Aruga et  al. 1997; Lowes et  al. 1997; Fallarino and Gajewski 1999; Kacha et al. 2000). A Th1 response has even been shown to be essential for antitumor immunity and Th2 cytokines downregulate antitumor immunity in some reports (Kacha et  al. 2000; Kobayashi et  al. 1998; Ostrand-Rosenberg et  al. 2000). The concept of immune deviation, a shift from Th1 to Th2 cytokine profile, has been hinted to be one of the major contributors to the failure of T cell-mediated immunity against tumors. Indeed, the immune deviation to Th2 cytokine production has been reported in progressive cancer patients (Pellegrini et  al. 1996). In contrast, an immunization-evoked Th2 to Th1 change was shown to induce tumor rejection in a murine tumor model (Hu et al. 1998). Additionally, Th2 cytokines have been shown to promote tumor growth in several experimental models (Hu et al. 1998). However, as in most immunological situations, one can never irrevocably conclude as one way or the other. There are plenty of data supporting the opposite. Th2 cytokines have been shown to be helpful for cancer gene therapy (Tepper et  al. 1989; Allione et al. 1994), and tumor-specific Th2 clones have been demonstrated to exhibit strong antitumor activity in vivo (Shen and Fujimoto 1996; Nishimura et al. 1999; Fallarino et al. 2000; Klugewitz et al. 2000). The mechanisms of how Th2 helper T cells destruct tumors is not yet clearly illustrated, but there is some evidence to suggest that the antitumor effect is mediated through the activation of innate immune cells such as eosinophils and macrophages by Th2 CD4+ T cells, which in turn secretes superoxide and nitric oxide (Hung et al. 1998). Although an effective antitumor immunity would be preferentially mediated through a Th1 response resulting in a direct killing by CTL, the results published up to this point seems to be directed towards a cooperative balance between the Th1 and Th2.

Regulatory CD4+ Cells To put it succinctly, the immune system is constantly working to find a balance between Th1 and Th2, activation and apoptosis, and pro-inflammatory and antiinflammatory to name a few. It has recently come to light that during tumor progression the tumor microenvironment becomes host to such an event, specifically between the effector and the regulatory response. Although not much is known regarding the

394

P. Yu and Y.-x. Fu

regulatory function pertaining to the tumor immunity, there is some evidence to suggest that the subtype of T cells responsible for “regulating” the effector immune response within the tumor site is similar to the now well-characterized regulatory CD4+ T cells involved in autoimmunity (Curiel et  al. 2004; Yu et  al. 2005; Wang et  al. 2004; Dieckmann et al. 2001; Liyanage et al. 2002; Woo et al. 2001, 2002).

Existence of Different Types of CD4+ Regulatory T Cells In Vivo Although during the past decades there had been much speculation and some evidence for suppressor T cells, such cells had not been phenotypically identified until recent years. A major advance resulted from the discovery by Sakaguchi and colleagues, and later confirmed by others that the small subset of CD4+ T cell, separate from CD4+ helper T cells by the expression of high levels of CD25 in the naïve mice, could induce organ-specific autoimmunity by its depletion followed by adoptive transfer or prevent its development (Sakaguchi et  al. 1985, 1995; Sugihara et  al. 1988). These data formed the basis into the regulatory nature of these cells, now defined as naturally occurring CD4+CD25+ T cells. This subset of cells has been established to be a powerful regulator of T cell responses in organ-specific autoimmunity and chronic infections (Bach and Francois Bach 2003; Belkaid et al. 2002; Maloy and Powrie 2001; Sakaguchi 2004; Shevach 2002). However, it is becoming increasingly clear that in many situations, CD4+CD25- T cells are as effective as CD4+CD25+ T cells in controlling T cell-mediated disease (Apostolou et al. 2002; Curotto de Lafaille and Lafaille 2002; Furtado et al. 2001). There are other subsets of CD4+ T cells that have been defined to manifest regulatory phenotypes. For example, CD4+CD45Rblow suppressor cells that secrete large quantities of either IL-10 and IL-4 (termed Tr1 cells) and other CD4+ suppressor T cells that secrete large quantities of TGF-b (termed Th3 cells) (Levings and Roncaralo 2000; O’Garra et al. 2004). The immune regulatory potential and functional significance of these cytokine-secreting CD4+ T cells is supported by the findings that TGF-b-deficient mice develop autoimmune disease (Gorelik and Flavell 2000) and that administration of neutralizing antibodies to IL-4 or TGF-b abrogates the in  vivo prevention of autoimmunity or tolerance-inducing activity of CD4+ T cells in the tumor model and some autoimmunity models (Yu et al. 2005; Seddon and Mason 1999; Zhai and Kupiec-Weglinski 1999). It remains to be determined, however, to what extent do these subtypes of CD4+ cells diverge from the naturally occurring CD4+CD25+ T cells in function and significance.

Markers to Identify CD4+ Regulatory T Cells In Vivo There are no known cell surface molecules that uniquely distinguish the CD4+ regulatory T cells from conventional activated CD4+ cells. For example, the CD25 molecule, which is the a-chain of the IL-2 receptor, is expressed on all peripheral

19  The Prognostic Significance of Tumor-Infiltrating Lymphocytes

395

antigen-reactive CD4+ T cells from one to several days following antigen activation. Moreover, many of the other cell-surface molecules in addition to CD25, including the TNF-family member GITR and cytotoxic T-lymphocyte antigen-4 (CTLA4) (Sakaguchi et  al. 2001; Wood and Sakaguchi 2003), that seem to distinguish CD4+CD25+ from CD4+CD25- effector cells are upregulated on CD4+CD25- T cells following antigen activation. In this regard, it is of great interest that a recently cloned transcription factor, termed Foxp3, a member of the forkhead family of DNA binding transcription factors, is not expressed in naïve CD4+CD25- cells but is highly expressed in the naturally occurring CD4+CD25+ regulatory cells. More importantly, mutational defects in the Foxp3 gene result in the fatal autoimmune and inflammatory disorder of the scurfy mouse and in the clinical and molecular features of the immunodysregulation, polyendocrinopathy, enteropathy, and X-linked syndrome (IPEX syndrome) in humans (Bennett et al. 2001; Brunkow et al. 2001; Wildin et al. 2001). In Foxp3-overexpressing mice, both CD4+CD25- and CD4-CD8+ T cells show suppressive activity, which suggests that expression of Foxp3 is linked to suppressor functions (Khattri et al. 2003; Fontenot et al. 2003; Hori et al. 2003). Taken together, these data strongly support the idea that Foxp3 may uniquely define the subset of CD4+ regulatory T cells in mice (Khattri et al. 2003; Fontenot et al. 2003; Hori et al. 2003) and humans (Walker et al. 2003). However, the recent findings that Foxp3 can be expressed in CD4+CD25- cells following activation and are also expressed in activated CD8+ T cells suggest that Foxp3 is linked to functional suppression, but not necessarily as a specific lineage marker (Walker et al. 2003; Chen et al. 2003; Cosmi et al. 2003; Manavalan et al. 2004). Whether a specific lineage marker even exists for these CD4+ regulatory cells awaits to be determined, but what can be concluded with little doubt is that a subset of CD4+ T cells indeed exist that regulate an inflammatory immune response. It can be speculated that such characterization of an inflammatory immune response may be ascribed to cancer.

Tumor-Induced CD4+ Regulatory T Cells Cancers generally develop over a long period of time. In addition, the major pathophysiologic characteristics of malignant cancer, invasion across natural tissue barriers and metastasis, are often associated with the disruption of normal tissue architecture leading to the initiation of inflammatory responses. In this regard, the cancers can be very much resembled to a chronic inflammation response. In view of this, it is possible to speculate that the anti-inflammatory mechanisms that are turned on at the beginning to protect normal tissues during a chronic inflammation coupled with the internal mechanisms controlled by malignant cells to produce cytokine like TGF-b start a process within the tumor microenvironment to set off a regulatory reaction that may adversely inhibit antitumor immunity. Given such correlative properties, the loss of regulatory function by the depletion of this CD4+ T cell subset may prolong the effector response resulting in tumor rejection.

396

P. Yu and Y.-x. Fu

Recent findings attest to possible negative regulatory roles by CD4+ T cells within the tumor environment. In some spontaneous tumor models, the presence of CD4+ T cells seems to promote cancer development instead of inhibiting (Daniel et al. 2003). One study by Schreiber’s group has shown that active immunization with antigen-specific CD4+ T cells in cancer-prone mice carrying a germline mutant ras oncogene resulted in immune responses that fail to eradicate mutant oncogene-expressing tumor cells, instead induced a remarkable enhancement of tumor growth (Siegel et al. 2000). Similarly, the studies by Robert North and his colleagues have shown that suppression by CD4+ T cells led to the progressive growth of an immunogenic tumor (Berendt and North 1980; Bursuker and North 1984) and intravenous depletion of these cells reversed the suppression to elicit CD8+ T cell-mediated antitumor immunity (Awwad and North 1988). While the presence of CD4+ T cells in some models have shown to be deleterious to the onset of tumor immunity, the necessity of CD4+ T cell help in mounting an effector immune response cannot be negated. In some studies, the depleting antibody given during the early stages of tumor growth was damaging to the generation of an immune response against the tumor, especially if T cell help was obligatory (Turk et al. 2004; Sutmuller et al. 2001), strongly suggesting the existence of functionally distinct CD4+ T cells subsets.

CD4+CD25+ Regulatory T Cells in Mice and Human Recent studies have suggested that CD4+CD25+ suppressor cells are relevant in to tumor immunology, albeit more deleterious than favorable for the host. Much of the study, up to recent times, have utilized a depletion of the entire CD4+ population, resulting in enhancement of immunity against tumors via the depletion of the suppressive CD4+CD25+ T cell population. In vivo experiments in several murine tumor models demonstrate this case as depletion of CD4+CD25+ T cells by anti-CD25 antibody treatment prior to tumor challenge significantly enhances the efficacy of vaccineinduced antitumor immunity (Tawara et al. 2002; Li et al. 2002; Golgher et al. 2002; Casares et al. 2003; Machiels et al. 2001; Sutmuller et al. 2001; Steitz et al. 2001; Jones et al. 2002; Shimizu et al. 1999; Onizuka et al. 1999). Another study revealed that splenic cells depleted of CD4+CD25+ T cells can mediate tumor regression presumably through promoting autoreactivity because autoimmune diseases were also induced (Shimizu et  al. 1999). These studies suggest that regulatory T cells may inhibit initial priming of CD8+ T cells, some of which recognize tumor antigens. One data demonstrated more clearly that the CD4+CD25+ T cells prevented priming of CTL, as depletion of CD25+ T cells had to be performed within the first 2 days after tumor inoculation, suggesting that suppression was ineffective once the priming was initiated (Onizuka et al. 1999). It was also shown that elimination of these regulatory T cells, despite causing increased autoreactivity in some cases, could increase immune responses to tumors such as melanomas over-expressing self-antigens (Sutmuller et al. 2001; Steitz et al. 2001; Jones et al. 2002; Shimizu et al. 1999). Even when the

19  The Prognostic Significance of Tumor-Infiltrating Lymphocytes

397

host bears a poorly immunogenic cancer, concomitant immunity can be rescued by systemic depletion of the CD4+CD25+ regulatory T cell subset (Turk et  al. 2004). These observations in the tumor models are consistent with the features that have been defined for CD4+CD25+ regulatory T cells in other disease models. It has been demonstrated that the equivalent of CD4+CD25+ regulatory T cells identified in mice also exist in humans. These cells, CD4+CD25+CD45RO+ T cells, occupy mean 6% of CD4+ T cells and are present in the blood of healthy adults (Dieckmann et al. 2001). Prevalence of regulatory T cells is increased in peripheral blood and tumor microenvironment of patients with pancreas or breast adenocarcinoma (Liyanage et al. 2002). The CD4+CD25+ T cells possessing regulatory properties have also been reported to be among the tumor-infiltrating T cells in different types of human cancers, such as lung, ovarian, pancreas, breast and gastrointestinal cancers, and lymphoma (Woo et al. 2001, 2002; Sasada et al. 2003; Marshall et al. 2004). Antigen-specific activation and cell–cell contact were required for these clones of Treg cells to exert suppressive activity on CD4+ effector cells. The finding that the presence CD4+ Treg cells at tumor sites suggests that they could have a profound effect on the inhibition of T cell responses against some human cancers (Curiel et al. 2004; Yu et al. 2005; Wang et al. 2004).

Suppression Occurred Inside Tumor Tissues It is possible that different subsets of CD4+ T cells, either providing help or regulating, dominate during various stages of tumor progression. A recent study by our group (Yu et al. 2005) has shown that suppression of immunity against tumor mainly occurs in the effector phase at tumor site and depletion of the regulatory T cells at the late stage of tumor progression did not mitigate the possible T helper function. We utilized a highly antigenic tumor that expresses a strong antigen on the surface, yet fails to regress in the host induced an accumulation of CD4+ regulatory T cells within the tumor microenvironment as the tumor persists and establishes, subsequently inhibiting CD8+ T cell function. In this model, local intratumoral depletion of these regulatory T cells unmasked the immunogenicity of tumor and reversed the CTL toleration leading to the rapid rejection of well-established tumors. From this study, we propose that CD4+ cells predominately play an enhancing helper role during the initial stages, but once tumors become chronically persistent and established the increased accumulation of CD4+ regulatory T cells inhibits CD8+ cell function and masks the immunogenicity of tumor. In fact, depletion of regulatory T cells unveils the immunogenicity of tumor cells and provides long-term protection against re-challenge of even the parental tumor cells lacking the strong antigen. This result suggests that the depletion of regulatory T cells promoted immunity against previously poorly immunogenic tumor antigens and expanded the tumor-reactive CD8+ T cell repertoire. Our study revealed that the population of tumor-infiltrating cells is skewed to favor regulatory CD4+ T cells over the helper CD4+ T cells within the tumor tissue, especially as tumor progresses and gets established in the host.

398

P. Yu and Y.-x. Fu

Origins of Tumor-Induced CD4+ Regulatory T Cells It is not very clear thus far if these CD4+CD25+ T cells inhibiting antitumor immunity in mice and human are naturally occurring regulatory T cells or others that generated in the periphery. It is possible that tumor environment preferentially recruits naturally occurring CD4+CD25+ T cells. One study published recently suggested that chemokine CCL22 abundantly expressed in the ovarian cancer tissues and tumor ascites cells preferentially attract the CD4+CD25+ T cells they identified in the ovarian cancers (Curiel et al. 2004). These cells were positive for the transcriptional factor Foxp3 and exhibited regulatory function similarly as the naturally occurring CD4+CD25+ T cells. It is not very clear so far if these tumor-infiltrating regulatory T cells can be defined as the counterpart of the naturally occurring CD4+CD25+ T cells in the mouse. What the specific trafficking clues are for the naturally occurring CD4+CD25+ T cells in the murine models still waits exploring. Another possibility is that tumor environment converts CD4+ T cells to CD4+CD25+ regulatory T cells or expands naturally occurring CD4+CD25+ T cells. There is evidence to indicate changes in the function of tumor-specific CD4+ T cells, from effectors to suppressors, during cancer progression (Zhou et al. 2004). Conversion from effector cells coincided with a substantial reduction in the antigen expression level, resulting in tumor persistence that ultimately led to T cell tolerance. They have evidence to further suggest that these antigen-specific T cells became CD4+CD25+ regulatory T cells (Zhou et al. 2004). An interesting model was suggested by this study that the processes of immunosurveillance and tumor editing coexist with a process in which the functional tumor-specific T cell repertoire is also edited by the tumor environment to benefit the tumor progression. Besides the anti-inflammatory cytokines, the CD4+CD25+ regulatory T cells inside the tumor may suppress antitumor immunity via other mechanisms. For example, these Tregs may inhibit immune response through their ability to control T cell numbers because they have been shown to regulate T cell proliferation in vitro (Sakaguchi 2004; Shevach 2002). Whether the regulatory cells that accumulate in the tumor site are ones that naturally exist in the host, or they initially migrate as helper CD4+ T cells, but become converted to regulatory cells by encountering the suppressive tumor environment is not altogether clear. It would be beneficial to better characterize the nature of CD4+ T cells isolated from or present in the tumor tissues by surface markers and cytokine profiles to utilize the CD4+ helper T cells instead of CD4+ regulatory T cells for adoptive transfer immunotherapy and association with better prognosis of cancer patients.

Th17 Cells CD4+ T cells differentiate into different lineages of helper T (TH) cells that are defined by distinct developmental regulation and biological functions following activation. TH17 cells producing IL17 have recently been identified as a new lineage of effector TH cells (Dong 2008) whose induction in mice is dependent on the

19  The Prognostic Significance of Tumor-Infiltrating Lymphocytes

399

synergistic activation by TGF-b and IL-6 (Bettelli et al. 2006; Mangan et al. 2006; Veldhoen et al. 2006). TGF-b is often present in various tissues and tumors, possibly contributing to the induction of either regulatory FOXP3+ or Th17 phenotype (Li et al. 2007; Nam et al. 2008). Indeed, Th17 cells have been found in the tumor microenvironment in animal models and in patients, although their specificity and function remain unclear (Kryczek et al. 2007; Tartour et al. 1999). Th17 cells have been shown to be important in immune responses to infectious agents and inflammation leading to autoimmune diseases (Dong 2008). The importance of an inflammatory milieu in the initiation of cancer has been well described (Lin and Karin 2007), consistent with the facts that tumors often arise in pro-inflammatory conditions induced by infections, chemical irritants, or autoimmunity. The IL-6/ IL-23/IL-17 axis of inflammation has been shown to be associated with carcinogenesis (Langowski et  al. 2006). IL-6 and IL-23 via STAT3 signaling might deliver prosurvival stimuli to some malignancies in an autocrine or paracrine mechanism, supported by the finding that the animals deficient in IL-23 were protected from cancer progression (Foulds et al. 2008; Kortylewski et al. 2005). IL-17 might promote tumor-associated neovascularization (Numasaki et al. 2003; Overwijk et al. 2006) and possibly recruit myeloid suppressor cells into the inflammatory environment, potentiating tumor growth (Charles et al. 2009). However, it has also been shown that systemic overexpression of IL-23 results in increased antitumor activity of T cells, involving both IFN-g and IL-17 secretion (Overwijk et al. 2006; Kaiga et al. 2007). In an adoptive transfer model, Th17-polarized cells were more effective in eradicating tumors than unpolarized T helper cells or Th1 cells. However, the treatment with tumor-specific Th17 was critically dependent on IFN-g (Muranski et  al. 2008). The authors interpreted the data as that the potential duality of the adoptively transferred effector Th17 cells later turned on T-bet to express a Th1 phenotype to contribute to the antitumor effects (Muranski and Restifo 2009). Since T-bet-expressing Th1 cells might represent a terminally differentiated phenotype that is more prone to senescence and/or apoptosis than their Th17 counterparts (Intlekofer et al. 2007), and Th17-defining transcription factor ROR-gt might have anti-apoptotic and antiproliferative activity, analogous to its well-described role during the thymocyte development (He et al. 1998; Kurebayashi et al. 2000; Sun et al. 2000; Xi et al. 2006), the early acquisition of t-bet might impair the ability of T cells to persist resulting in the disadvantages over Th17 cells in adoptive transfer therapy. Thus, the ability of the transferred Th17 cells to evolve in  vivo to Th1 phenotype is crucial for the therapeutic outcome.

Other Aspects That Complicate the Relationship Between the Tumor-Infiltrating Lymphocytes and Prognosis The association of TIL with better prognosis seems to be influenced by other factors besides the immunophenotypes of TIL. Melanomas are among the tumors that are often associated with TIL. Many correlation studies have been done since Clark et  al. (1969) first described the lymphocytic infiltration of PCM.

400

P. Yu and Y.-x. Fu

Patients with a moderate-to-marked lymphocytic infiltrate within their primary melanoma had a significantly better prognosis and a three-time higher 5-year survival rate than patients with a sparse or absent lymphocytic infiltrate (Day et  al. 1981). Elder et  al. (1985) differentiated the lymphocytic infiltrate into brisk, nonbrisk, or absent, according to its intensity, and demonstrated that TIL were of prognostic significance only in vertical growth phase (VGP) melanoma. In contrast, the extent of lymphocytic infiltration had no prognostic influence in radial growth phase (RGP) melanomas, regardless of whether the melanoma was in situ or invasive. These findings were later confirmed by another study (Clemente et al. 1996). The 5- and 10-year survival rates were 77 and 55% in melanomas with brisk VGP infiltrates; 53 and 45% with nonbrisk VGP infiltrates; and 37 and 27% without VGP infiltrates (Clemente et  al. 1996). Also, the number of TILs in the primary tumor has been found to be inversely correlated with the probability for lymph node metastases (Clemente et al. 1996). Patients with brisk TIL infiltrates in their primary tumors showed a 3.9% probability of a positive sentinel lymph node (SLN), compared to a 26.2% probability in patients with TILs absent from their primary melanoma (Taylor et  al. 2007). Furthermore, of those patients with regional lymph node metastases, patients with more marked lymphocytic responses in their metastatic melanoma showed a significantly higher 30-month disease-free survival rate (81.3% for patients with a brisk TIL infiltrate; 46.8% for patients with a nonbrisk infiltrate; and, 29.3% for patients with TILs absent from their lymph node metastases) (Day et al. 1981; Mihm et al. 1996). However, other studies could not convincingly demonstrate that brisk TIL infiltrates were associated with improved survival in melanoma patients (Barnhill et al. 1996; Thorn et al. 1994). These discrepant results may in part be explained by differences in patient populations investigated, with particular reference to the thickness of patients’ melanomas (Taylor et al. 2007). The study by Clemente et al. (1996) found the impact of TILs most pronounced in patients with high-risk lesions, thicker than 1.7mm but less than 6mm in Breslow depth (Taylor et al. 2007). This suggested that the briskness of the TIL infiltrate was prognostic for T2–T4 (TMN system) primary cutaneous melanoma (PCM), though the prognostic significance of TILs was lost in very thick lesions (advanced T4). In contrast, Barnhill et al. (1996) did not find any survival advantage to be associated with brisk TIL infiltrates; however, patients with both RGP and VGP were included in this study (Barnhill et al. 1996), even though other studies did not demonstrate a prognostic significance of TILs in RGP PCM (Clemente et  al. 1996; Elder et  al. 1985). Furthermore, only 25.6% of patients in Barnhill’s study had lesions thicker than 1.7mm (Barnhill et al. 1996), while 82 and 71% of patients had lesions thicker than 1.7mm in the studies by Clemente et al. (1996) and Tuthill et al. (2002). Taylor et al.(2007) did not find an impact of TILs on survival (44% of patients had lesions thicker than 2mm); however, they did show that TILs are an independent predictor of SLN positivity, which by itself is the most important independent predictor of recurrence and survival in malignant melanoma patients (Balch et  al. 2001). It is possible that there are distinct mechanisms for the

19  The Prognostic Significance of Tumor-Infiltrating Lymphocytes

401

induction of TIL at RGP and VGP, for which might be unique in each cancer type and wait for further investigation.

Targeted Tumor Tissues to Recruit and Train T Cells Tumor often forms barrier to limit T cell infiltration and reduce sufficient tumor antigens to draining lymph nodes. Since the positive roles played by CD8+ T cells and CD4+ helper T cells in antitumor immunity, it is reasonable to speculate that increase of infiltration of T lymphocytes and initiation of antitumor immune responses inside the tumor tissue at early phase could increase T cell repertoire and more antigens to stimulate T cells inside tumor would be a strategy for cancer immunotherapy. Several cytokines or chemokines have been used to attract and activate T cells at tumor sites. In this review, we focus on our recent finding that stimulation of lymphotoxin b receptor (LTbR) inside tumor tissues will promote strong infiltration of immune cells leading to tumor rejection. The LTbR plays an important role in the formation of lymphoid structures (Ettinger 2000; Fu and Chaplin 1999; Kim et al. 2000). LTbR is activated by two members of the TNF family, membrane lymphotoxin ab and LIGHT (Mauri et al. 1998). Signaling via LTbR regulates the expression of chemokines and adhesion molecules within secondary lymphoid organs. Chemokines and adhesion molecules control the migration and positioning of DCs and lymphocytes in the spleen (Wu et al. 1999; Ngo et al. 1999). Therefore, it is possible that enhanced LTbR signaling inside tumor tissues may promote the formation of lymphoid-like structure for direct T cell sequestration. TNFR signaling may also play similar but less effective role in regulating chemokine expression (Ngo et al. 1999). However, TNFR signaling may have a more toxic effect as seen in other systemic TNF treatments without extra costimulation (Spriggs et al. 1988). Soluble LTa can signal through the TNFR resulting in the upregulation of chemokines. To avoid toxic effect, recombinant lymphotoxin a (LTa) has been conjugated with antibody targeting specifically to the tumor tissues resulted in an effective antitumor immune response associated with the induction of peripheral lymphoid-like tissue (Schrama et al. 2001). However, LTa lacks of costimulation function, which would result in a less effective activation of recruited naïve T cells in the lymphoid-like structure inside tumor tissues. LIGHT is a ligand for LTbR and herpes virus entry mediator (Mauri et al. 1998; Rooney et al. 2000). LIGHT is predominantly expressed on lymphoid tissues, especially on the surface of activated DCs and T cells. LIGHT is a strong costimulatory molecule (Mauri et al. 1998; Tamada et al. 2000; Zhai et al. 1998). Our data indicates that the interactions between LIGHT and LTbR restore lymphoid structures in the spleen of LTa-/- mice. In addition, upregulation of LIGHT causes T cell activation and migration into nonlymphoid tissues and formation of lymphoid-like structures (Wang et  al. 2001, 2002). Therefore, LIGHT-mediated microenvironment inside tumor could be effective in both recruiting and activating naïve T cells into tumor tissues. The expression of LIGHT in the tumor environment induces a

402

P. Yu and Y.-x. Fu

massive infiltration of naïve T lymphocytes that correlates with an upregulation of both chemokine production and expression of adhesion molecules. Activation and expansion of these infiltrating T cells leads to the rejection of established, highly progressive tumors at local and distal sites (Yu et al. 2004). Our study indicates that indeed induction of infiltration of T lymphocytes and initiation of antitumor immune responses inside the tumor tissue may be an effective strategy for cancer immunotherapy.

Concluding Remarks Understanding the balance of antitumor effector T cells versus regulatory T cells may be more important to determine the outcome of immune responses inside tumors. Our recent study has demonstrated that rapid recruitment of naïve lymphocytes and expansion of CD8+ T cells inside the tumor may be a way of creating a dominant pro-inflammatory environment, leading to the rejection of tumor at local and distal sites (Yu et al. 2004). We later further demonstrated that the depletion of regulatory T cells inside the tumor is another efficient way of converting the antiinflammatory environment inside tumor to pro-inflammatory one (Yu et al. 2005). From clinical therapeutic point of the view, the local treatment to eliminate regulatory T cells has certain critical advantages over systemic treatment: First, local treatment may avoid side effects induced by systemic depletion of all CD4+ T cells, which may abrogate T helper-mediated protective immunity against pathogens; Second, it would not hinder effective priming of CD8+ T cell in the lymphoid tissues by the helper CD4+ T cells because depletion remains local. Third, the local treatment would be even expected to be more effective if the key suppression of CD4+CD25+ regulatory T cells resides inside tumor. Also intratumoral treatment would reduce the dose of antibody to be applied to the patients as well as be more affordable even for the developing countries. CD4+CD25+ T cells have been shown to be present in a variety of human cancer tissues (Woo et al. 2001, 2002; Sasada et al. 2003; Marshall et al. 2004) and these cells are negatively associated with the prognosis of the ovarian cancer patients in addition (Curiel et al. 2004). Therefore, regulatory T cells within the tumor environment represent an attractive target, and their depletion may lead to improvements in the current immunotherapy protocol in the future clinical trials. It is likely that a combination treatment which would rapidly expand the effector cells at the tumor site, while locally depleting the regulatory cells, could potentially provide a potent strategy for enhancing antitumor immunity and permitting a clinically desirable outcome for cancer patients. It is increasingly clear that some of TIL may be friend while others may be foes. Increase of TIL may not always be associated with better prognosis. Accumulation of regulatory T cells inside tumor tissue in both animal model and human patients suggest that aberrant immune response occurs inside growing tumor causing immune invasion. How to effectively reverse the immunologically suppressive environment or depleting regulatory T cells inside tumors may be a new avenue to

19  The Prognostic Significance of Tumor-Infiltrating Lymphocytes

403

generate effective immune responses against tumors at local and distal sites. Careful laboratory investigation is critical for us to better understand the role of TIL. It is possible that future immunotherapy and prognosis depend on better defining of TIL. When cancer tissue contains fewer TIL, targeting tumor tissues for the increase of TIL, especially CTL may result in positive responses. On the other hand, depletion of T regulatory cells may be important for patients with cancer tissues containing T regulatory cells. The pathologists may play critical diagnostic role in determining the type of TIL, which provide not only prognosis but also the guide for immunotherapy.

References Allione A et al (1994) Cancer Res 54(23):6022 Apostolou I, Sarukhan A, Klein L, von Boehmer H (2002) Nat Immunol 3(8):756 Aruga A et al (1997) J Immunol 159(2):664 Awwad M, North RJ (1988) J Exp Med 168(6):2193 Bach JF, Francois Bach J (2003) Nat Rev Immunol 3(3):189 Balch CM et al (2001) J Clin Oncol 19(16):3622 Balkwill F, Mantovani, A (2001) Lancet 357(9255):539 Banner BF et al (1990) Mod Pathol 3(2):129 Barnhill RL, Fine JA, Roush GC, Berwick M (1996) Cancer 78(3):427 Battegay M et al (1994) J Virol 68(7):4700 Bear HD, Susskind BM, Close KA, Barrett SK (1988) Cancer Res 48(6):1422 Beatty GL, Paterson Y (2000) J Immunol 165(10):5502 Beatty G, Paterson Y (2001) J Immunol 166(4):2276 Beck-Engeser GB et al (2001) J Exp Med 194(3):285 Belkaid Y et al (2002) Nature 420(6915):502 Bennett SR et al (1997) J Exp Med 186(1):65 Bennett SR et al (1998) Nature 393(6684):478 Bennett CL et al (2001) Nat Genet 27(1):20 Berendt MJ, North RJ (1980) J Exp Med 151(1):69 Bettelli E et al (2006) Nature 441(7090):235 Bevan MJ (2004) Nat Rev Immunol 4(8):595 Boyse EA, Old LJ, Stockert E (1962) Ann NY Acad Sci 99:574 Brunkow ME et al (2001) Nat Genet 27(1):68 Bursuker I, North RJ (1984) J Exp Med 159(5):1312 Cammarota G et al (1992) Nature 356(6372):799 Cardin RD, Brooks JW, Sarawar SR, Doherty PC (1996) J Exp Med 184(3):863 Casares N et al (2003) J Immunol 171(11):5931 Cavallo F et al (1992) J Immunol 149(11):3627 Cella M et al (1996) J Exp Med 184(2):747 Charles KA et al (2009) J Clin Invest Cheever MA, Thompson DB, Klarnet JP, Greenberg PD (1986) J Exp Med 163(5):1100 Chen W et al (2003) J Exp Med 198(12):1875 Clark WH Jr, From L, Bernardino EA, Mihm MC (1969) Cancer Res 29(3):705 Clark W H, Jr et al (1989) J Natl Cancer Inst 81(24):1893 Clemente C G et al (1996) Cancer 77(7):1303 Cosmi L et al (2003) Blood 102(12):4107 Curiel TJ et al (2004) Nat Med 10(9):942

404

P. Yu and Y.-x. Fu

Curotto de Lafaille MA, Lafaille JJ (2002) Curr Opin Immunol 14(6):771 Daniel D et al (2003) J Exp Med 197(8):1017 Day CL, Jr. et al (1981) Ann Surg 193(4):436 De Bruijn ML et al (1991) Eur J Immunol 21(12):2963 Dieckmann D et al (2001) J Exp Med 193(11):1303 Di Rosa F, Matzinger P (1996) J Exp Med 183(5):2153 Dong C (2008) Nat Rev Immunol 8(5):337 Dranoff G et al (1993) Proc Natl Acad Sci USA 90(8):3539 Dudley ME et al (2002) Science 298(5594):850 Dudley ME, Rosenberg SA (2003) Nat Rev Cancer 3(9):666 Elder DE et al (1985) Cancer 56(2):413 Ettinger R (2000) Curr Top Microbiol Immunol 251:203 Fallarino F, Gajewski TF (1999) J Immunol 163(8):4109 Fallarino F et al (2000) J Immunol 165(10):5495 Fayolle C et al (1996) Immunology 89(1):41 Fearon ER et al (1990) Cell 60(3):397 Feltkamp MC et al (1993) Eur J Immunol 23(9):2242 Fields RC, Shimizu K, Mule JJ (1998) Proc Natl Acad Sci USA 95(16):9482 Fontenot JD, Gavin MA, Rudensky AY (2003) Nat Immunol 4(4):330 Fossati G, Anichini A, Parmiani G (1987) Int J Cancer 39(6):689 Foulds KE et al (2008) J Immunol 180(2):842 Fu YX, Chaplin DD (1999) Annu Rev Immunol 17:399 Fujiwara H et al (1984) J Immunol 133(3):1671 Furtado GC et al (2001) Immunol Rev 182:122 Golgher D et al (2002) Eur J Immunol 32(11):3267 Golumbek PT et al (1991) Science 254(5032):713 Gorelik L, Flavell RA (2000) Immunity 12(2):171 Greenberg PD, Cheever MA, Fefer A (1981) J Exp Med 154(3):952 He YW, Deftos ML, Ojala EW, Bevan MJ (1998) Immunity 9(6):797 Hock H, Dorsch M, Diamantstein T, Blankenstein T (1991) J Exp Med 174(6):1291 Hori S, Nomura T, Sakaguchi S (2003) Science 299(5609):1057 Hu HM, Urba WJ, Fox BA (1998) J Immunol 161(6):3033 Hu HM, Winter H, Urba WJ, Fox BA (2000) J Immunol 165(8):4246 Huang AY et al (1994) Science 264(5161):961 Huang AY, Bruce AT, Pardoll DM, Levitsky HI (1996) Immunity 4(4):349 Hung K et al (1998) J Exp Med 188(12):2357 Igarashi T et al (1992) Eur Urol 22(4):323 Intlekofer AM et al (2007) J Exp Med 204(9):2015 Janssen EM et al (2003) Nature 421(6925):852 Jones E et al (2002) Cancer Immun 2:1 Kacha AK, Fallarino F, Markiewicz MA, Gajewski TF (2000) J Immunol 165(11):6024 Kaiga T et al (2007) J Immunol 178(12):7571 Kast WM, Bronkhorst AM, de Waal LP, Melief CJ (1986) J Exp Med 164(3):723 Kast WM et al (1989) Cell 59(4):603 Khattri R, Cox T, Yasayko SA, Ramsdell F (2003) Nat Immunol 4(4):337 Kim D et al (2000) J Exp Med 192(10):1467 Klein G, Sjogren HO, Klein E, Hellstrom KE (1960) Cancer Res 20:1561 Klein E et al (1980) Contemp Top Immunobiol 10:79 Klugewitz K, Scheffold A, Radbruch A, Hamann A (2000) Int J Cancer 87(5):673 Knuth A, Danowski B, Oettgen HF, Old LJ (1984) Proc Natl Acad Sci U S A 81(11):3511 Kobayashi M, Kobayashi H, Pollard RB, Suzuki F (1998) J Immunol 160(12):5869 Koch F et al (1996) J Exp Med 184(2):741 Kolbeck PC et al (1992) Mod Pathol 5(4):420

19  The Prognostic Significance of Tumor-Infiltrating Lymphocytes Kortylewski M et al (2005) Nat Med 11(12):1314 Kowalczyk D, Skorupski W, Kwias Z, Nowak J (1997) Br J Urol 80(4):543 Kripke ML (1974) J Natl Cancer Inst 53(5):1333 Kryczek I et al (2007) J Immunol 178(11):6730 Kurebayashi S et al (2000) Proc Natl Acad Sci U S A 97(18):10132 Langowski JL et al (2006) Nature 442(7101):461 Leclerc JC, Gomard E, Levy JP (1972) Int J Cancer 10(3):589 Levings MK, Roncarolo MG (2000) J Allergy Clin Immunol 106(1 Pt 2):S109 Levitsky HI, Lazenby A, Hayashi RJ, Pardoll DM (1994) J Exp Med 179(4):1215 Li J, Hu P, Khawli LA, Epstein AL (2003) Cancer Res 63(23):8384 Li MO, Wan YY, Flavell RA (2007) Immunity 26(5):579 Lin WW, Karin M (2007) J Clin Invest 117(5):1175 Lin KY et al (1996) Cancer Res 56(1):21 Liyanage UK et al (2002) J Immunol 169(5):2756 Lowes MA et al (1997) J Invest Dermatol 108(6):914 Lu Z et al (2000) J Exp Med 191(3):541 MacDonald GH, Johnston RE (2000) J Virol 74(2):914 Machiels JP et al (2001) Cancer Res 61(9):3689 Mackey MF et al (1997) Cancer Res 57(13):2569 Mackey MF et al (1998) J Immunol 161(5):2094 Maloy KJ, Powrie F (2001) Nat Immunol 2(9):816 Manavalan JS et al (2004) Int Immunol 16(8):1055 Mangan PR et al (2006) Nature 441(7090):231 Marrogi AJ et al (1997) Int J Cancer 74(5):492 Marshall NA et al (2004) Blood 103(5):1755 Matloubian M, Concepcion RJ, Ahmed R (1994) J Virol 68(12):8056 Mauri DN et al (1998) Immunity 8(1):21 Mihm MC, Jr., Clemente CG, Cascinelli N (1996) Lab Invest 74(1):43 Monach PA, Meredith SC, Siegel CT, Schreiber H (1995) Immunity 2(1):45 Moore OS Jr, Foote FW, Jr (1949) Cancer 2(4):635 Mukherji B, MacAlister TJ (1983) J Exp Med 158(1):240 Mumberg D et al (1999) Proc Natl Acad Sci U S A 96(15):8633 Muranski P, Restifo NP (2009) Curr Opin Immunol 21(2):200 Muranski P et al (2008) Blood 112(2):362 Mosmann TR, Sad S (1996) Immunol Today 17(3):138 Naito Y et al (1998) Cancer Res 58(16):3491 Nakano O et al (2001) Cancer Res 61(13):5132 Nam JS et al (2008) Cancer Res 68(10):3915 Ngo VN et al (1999) J Exp Med 189(2):403 Nishimura T et al (1999) J Exp Med 190(5):617 Noguchi Y, Richards EC, Chen YT, Old LJ (1995) Proc Natl Acad Sci U S A 92(6):2219 Numasaki M et al (2003) Blood 101(7):2620 O’Garra A, Vieira PL, Vieira P, Goldfeld AE (2004) J Clin Invest 114(10):1372 Onizuka S et al (1999) Cancer Res 59(13):3128 Ostrand-Rosenberg S (1994) Curr Opin Immunol 6(5):722 Ostrand-Rosenberg S, Grusby MJ, Clements VK (2000) J Immunol 165(11):6015 Overwijk WW et al (2006) J Immunol 176(9):5213 Pan ZK, Ikonomidis G, Pardoll D, Paterson Y (1995) Cancer Res 55(21):4776 Pardoll DM, Topalian SL (1998) Curr Opin Immunol 10(5):588 Pellegrini P et al (1996) Cancer Immunol Immunother 42(1):1 Porgador A, Gilboa E (1995) J Exp Med 182(1):255 Potter TA, Rajan TV, Dick RF, 2nd, Bluestone JA (1989) Nature 337(6202):73 Pulaski BA et al (1993) Cancer Res 53(9):2112

405

406

P. Yu and Y.-x. Fu

Ridge JP, Di Rosa F, Matzinger P (1998) Nature 393(6684):474 Robins RA (1986) Biochim Biophys Acta 865(3):289 Rollins BJ (2006) Eur J Cancer 42(6):760 Romagnani S (1997) Immunol Today 18(6):263 Rooney IA et al (2000) J Biol Chem 275(19):14307 Rosenberg SA (1999) Immunity 10(3):281 Rosenberg SA (2001a) Nature 411(6835):380 Rosenberg SA (2001b) J Intern Med 250(6):462 Rosenberg SA, Yang JC, Restifo NP (2004) Nat Med 10(9):909 Rouse BT, Rollinghoff M, Warner NL (1972) Nat New Biol 238(82):116 Saiki Y et al (1996) Lab Invest 75(1):67 Sakaguchi S (2004) Annu Rev Immunol 22:531 Sakaguchi S, Fukuma K, Kuribayashi K, Masuda T (1985) J Exp Med 161(1):72 Sakaguchi S et al (1995) J Immunol 155(3):1151 Sakaguchi S et al (2001) Immunol Rev 182:18 Sasada T et al (2003) Cancer 98(5):1089 Schnell S, Young JW, Houghton AN, Sadelain M (2000) J Immunol 164(3):1243 Schoenberger SP et al (1998) Nature 393(6684):480 Schomig A et al (1996) N Engl J Med 334(17):1084 Schrama D et al (2001) Immunity 14(2):111 Schumacher K, Haensch W, Roefzaad C, Schlag PM (2001) Cancer Res 61(10):3932 Seddon B, Mason D (1999) J Exp Med 189(2):279 Shedlock DJ, Shen H (2003) Science 300(5617):337 Shen Y, Fujimoto S (1996) Cancer Res 56(21):5005 Shevach EM (2002) Nat Rev Immunol 2(6):389 Shi Y, Zheng W, Rock KL (2000) Proc Natl Acad Sci U S A 97(26):14590 Shimizu J, Yamazaki S, Sakaguchi S (1999) J Immunol 163(10):5211 Shinde S et al (1996) J Immunol 157(7):2764 Siegel CT et al (2000) J Exp Med 191(11):1945 Spellman CW, Daynes RA (1981) Hum Pathol 12(4):299 Spiotto MT et al (2002) Immunity 17(6):737 Spriggs DR et al (1988) J Natl Cancer Inst 80(13):1039 Steitz J et al (2001) Cancer Res 61(24):8643 Stevenson PG, Belz GT, Altman JD, Doherty PC (1998) Proc Natl Acad Sci U S A 95(26):15565 Sugihara S et al (1988) J Immunol 141(1):105 Sun JC, Bevan MJ (2003) Science 300(5617):339 Sun Z et al (2000) Science 288(5475):2369 Sun JC, Williams MA, Bevan MJ (2004) Nat Immunol 5(9):927 Sutmuller RP et al (2001) J Exp Med 194(6):823 Tamada K et al (2000) Nat Med 6(3):283 Tartour E et al (1999) Cancer Res 59(15):3698 Tawara I et al (2002) Jpn J Cancer Res 93(8):911 Taylor RC et al (2007) J Clin Oncol 25(7):869 Tepper RI, Pattengale PK, Leder P (1989) Cell 57(3):503 Tevethia SS, Blasecki JW, Vaneck G, Goldstein AL (1974) J Immunol 113(5):1417 Thorn M et al (1994) J Natl Cancer Inst 86(10):761 Toes RE et al (1996) Cancer Res 56(16):3782 Tsung K et al (1997) J Immunol 158(7):3359 Turk MJ et al (2004) J Exp Med 200(6):771 Tuthill RJ et al (2002) Am J Clin Pathol 118(4):504 Tuting T et al (1999) J Gene Med 1(6):400 Veldhoen M et al (2006) Immunity 24(2):179 Walker MR et al (2003) J Clin Invest 112(9):1437

19  The Prognostic Significance of Tumor-Infiltrating Lymphocytes Wang J et al (2001) J Immunol 167(9):5099 Wang J et al (2002) Eur J Immunol 32(7):1969 Wang HY et al (2004) Immunity 20(1):107 Ward PL et al (1990) Cancer Res 50(13):3851 Wildin RS et al (2001) Nat Genet 27(1):18 Wolfers J et al (2001) Nat Med 7(3):297 Wolkers MC, Stoetter G, Vyth-Dreese FA, Schumacher TN (2001) J Immunol 167(7):3577 Woo EY et al (2001) Cancer Res 61(12):4766 Woo EY et al (2002) J Immunol 168(9):4272 Wood KJ, Sakaguchi S (2003) Nat Rev Immunol 3(3):199 Wu Q et al (1999) J Exp Med 190(5):629 Xi H et al (2006) Immunity 24(6):813 York IA, Goldberg AL, Mo XY, Rock KL (1999) Immunol Rev 172:49 Yu P et al (2003) J Exp Med 197(8):985 Yu P et al (2004) Nat Immunol 5(2):141 Yu P et al (2005) J Exp Med 201(5):779 Zajac AJ et al (1998) J Exp Med 188(12):2205 Zhai Y, Kupiec-Weglinski JW (1999) Curr Opin Immunol 11(5):497 Zhai Y et al (1998) J Clin Invest 102(6):1142 Zhang L et al (2003) N Engl J Med 348(3):203 Zhou G et al (2004) J Exp Med 200(12):1581 Zitvogel L et al (1996) J Exp Med 183(1):87

407

Chapter 20

The Pro-inflammatory Milieu and Its Role in Malignant Epithelial Initiation Adam Yagui-Beltrán, Qizhi Tang, and David M. Jablons

Abstract  Increasing research from preclinical and clinical studies is demonstrating that inflammation with its myriad of heterogeneous mediators and cellular effectors plays a role in the malignant initiation of epithelial cells. Pre-existing inflammation may exist prior to the development of cancer, or on the other hand it may be that oncogenic changes may lead to a certain inflammatory microenvironment that in turn will promote the development of a tumor. Examples of how inflammation may lead to cancer formation include enhanced proliferation of initiated cells, through resistance to apoptosis, induction of genomic instability, alterations in epigenetic events and subsequent inappropriate gene expression, abnormal tumor neovascularization or angiogenesis, and the promotion of metastasis among others. At the time of malignant transformation of epithelial cells, the many components of the tumor microenvironment (TME), such as the tumor cells themselves, the stromal cells in surrounding tissue and the infiltrating leukocytes cells generate an intratumoral inflammatory state by the aberrant expression and secretion of a florid array of pro-inflammatory molecules that include chemokines, cytokines, cyclooxygenase-2, prostaglandins (PGs), iNOS, nitric oxide (NO), and a vast network of intracellular signaling molecules including upstream kinases and transcription factors that facilitate tumor promotion and progression. In this chapter, we will explore current knowledge of the various components of the pro-inflammatory milieu within the TME; we will, for example, highlight the major molecular studies that support how pro-inflammatory cytokines, chemokines, and PGs are able to regulate some of the angiogenic switches controlled by vascular endothelial growth factors inducing inflammatory angiogenesis. We will highlight some of the principal tumor cell–stroma dynamic relationships and how often these interplays are incrementally manipulated in ways that favor malignant epithelial initiation, resulting in tumor

A. Yagui-Beltrán (*) Department of Surgery, Division of Adult Thoracic Surgery, The Helen Diller Family Comprehensive Cancer Center Thoracic Surgery, University of California San Francisco, 1600 Divisadero, Room A-743, San Francisco, CA 94143-1724, USA e-mail: [email protected] R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_20, © Springer Science+Business Media, LLC 2010

409

410

A. Yagui-Beltrán et al.

immune evasion, tumor growth, and metastasis. We hope that an increasing understanding of the specific molecular mechanisms associating inflammation, TME, and cancer will subsequently result in a better understanding of how these cancers to subvert the immune system and overcome response our wide array of chemoradiotherapeutic agents.

Introduction The relationship between chronic inflammation and cancer is not new and can be traced back to the days of Rudolf Virchow (Coussens and Werb 2002; Balkwill and Mantovani 2001). However, it is only in recent years that this link has seen a dynamic renaissance, fueled by increasing animal and human studies and strengthened by epidemiology (Balkwill et al. 2005; Mager et al. 2005; Mager 2006). As part of this strengthening, various pathogenic causes for chronic inflammation have been proposed as players in multiple tumor systems for malignant initiation, progression, and metastatic spread (Kuper et al. 2000; Fidler 1995; Karin and Greten 2005; Hold and El-Omar 2008); despite these clear advances, the specific mechanisms regulating their functional association are not fully understood. Examples of specific chronic inflammatory agents of infectious nature leading to cancer include Helicobacter pylori colonization in the stomach, which increases the risk of gastric cancer and mucosal lymphoma. Other examples of infectious agents leading to chronic inflammation and eventually cancer are Hepatitis B and C, Clonorchis sinensis, and Schistosoma mansoni among others. In certain pathologies, inflammatory such as pelvic inflammatory disease, Barretts’ esophagus, and chronic pancreatitis, the etiology of inflammation is noninfectious or idiopathic (Quante and Wang 2008), and this inflammation is responsible for increasing the risk of developing cancer. It is also well established that chronic inflammation associated with autoimmune inflammatory bowel disease may ultimately lead to the development of carcinoma of the colon. Interestingly, while defining the mechanisms regulating inflammation in malignancy it became evident that various oxidizing enzymes involved in DNA damage, such as nicotinamide adenine dinucleotide phosphate-oxidase (NADPH) and inducible nitric oxide synthase (iNOS), are also responsible for the malignant initiation of epithelial cells within the TME. In chronic inflammation, specifically where a direct effect of the pathogen on neoplastic transformation of epithelial cells has been shown but overwhelming evidence points to a more generalized role of the resultant inflammatory process as the primary mediator of the carcinogenic process, both eradication of infection and anti-inflammatory therapies (e.g., COX-2 inhibitors) have been successful in preventing progression and even curing some of these cancers (Koehne and Dubois 2004; Flossmann and Rothwell 2007; Chan et al. 2007). Recent research has demonstrated that not all inflammation results in cancer; for example, the adaptive immune response elicited by some cancers exerts a potentially effective role in immune surveillance (Raulet 2004). A fundamental question in the field is the role that specific immune and inflammatory cell populations play in the neoplastic

20  The Pro-inflammatory Milieu and Its Role in Malignant Epithelial Initiation

411

i­nitiation of the epithelium. In general terms, it can be suggested that either the tumor and its microenvironment actively affects the immune response so that it becomes pro-tumorigenic [where, for example, cytokines or reactive oxygen species (ROS) maximize and propagate tumor growth] or it remains as sustained chronic inflammation exerting an active and primary role in initiating and transforming epithelial stem cells into tumor cells; there is increasing evidence that tumor-derived signals appear to play a role in recruiting bone marrow derived cells (BMDC) where they can give rise to stroma and perhaps tumor cells that can contribute to tumor progression within the context of chronic inflammation.

Acute Versus Chronic Inflammation in the Context of the Tumor Microenvironment Acute inflammation often becomes chronic in nature especially when the causing inflammatory agent or agents persist; however, there are circumstances where chronic inflammation takes place right from the start as in the case of gastric cancer and infection with H. pylori. In contrast to the faster onset and the mostly vascular changes associated with acute inflammation, chronic inflammation is characterized by significant leukocyte infiltration in the affected tissues. In contrast to chronic inflammation, both the innate and the adaptive immune systems play a role in acute inflammation. They are responsible for two functionally reversible and opposing functions of the acute inflammatory response: (1) a pro-inflammatory response (mainly consisting of organized damaged cell death) and (2) a post-inflammatory or anti-inflammatory process (consisting of tissue repair and ultimately a resolution of inflammation). Subsequent to exposure to inflammatory agents, a series of complex inflammatory paraphernalia explodes through precise highly regulated crosstalk mechanisms between the innate and adaptive immune systems in an organ specific manner followed by expression of further death signals. Furthermore, such a crosstalk ensures the availability of the required amounts of inflammatory mediators such as cytokines/chemokines and their receptors, and antibodies responsible for recruiting, attracting, and activating other inflammatory cells and mediator cells to the site of inflammation via the activation and engagement of vascular cell components. The principal physiological purposes of the acute inflammatory response are dual (1) Obliterate any exogenous cells (organisms or harmful materials) or any endogenous harmful cells or cell components (e.g., cancerous cells, cells with aberrant nuclei, protein machinery, etc.) that are created during the inflammatory process. (2) Neutralize and eliminate the various apoptotic signals generated in the inflamed tissue after their signal has become redundant as well as eliminate the causative toxins. This is particularly important so that cellular and tissue repair can commence upon successful resolution of the initial acute inflammatory response. It is important to note that not all acute inflammatory conditions (e.g., minor exposure to chemical irritation) that can elicit instant response from innate immune cells result in extensive involvement of adaptive immunity. On the other hand, chronic

412

A. Yagui-Beltrán et al.

exposure to allergens, other environmental factors, or infectious agents often may result in the precise activation of innate and adaptive immunity pathways that produce suitable responses for defending the host tissue (Lodoen and Lanier 2006; Valent et al. 2001; Bonasio and von Andrian 2006; Drayton et al. 2006; Kabelitz and Medzhitov 2007). Perhaps of significance to human disease is the fact that the various dynamic changes in immune cell function associated to unresolved inflammation can result in the development of a myriad of inflammatory pathologies affecting the majority of human organs, and those can vary from local allergic reactions to major life threatening vasculitis, syndromes, and from pre-neoplastic syndromes to full blown cancers (Balkwill and Mantovani 2001; Dvorak 1986; Ferrantini et al. 2008; Dalgleish and O’Byrne 2006; de Visser et al. 2006; Swann et al. 2008; Wang et al. 2008). Although the precise molecular mechanisms governing the often complex multifaceted processes initiating, perpetuating, and linking the innate and adaptive immune systems during an acute inflammatory process are yet to be fully elucidated, an outline of the most important known molecular and cellular phenomena of acute inflammation will serve useful in subsequent understanding of the pleiotropic functional significance (tumorigenic versus tumoricidal) of the various innate and adaptive immune cell subpopulations in subsequent persistent or chronic inflammation specifically in the context of carcinogenesis. 1. The pro-inflammatory response phase:  The initial phase of acute inflammation is also known as the programmed cell death or the “apoptosis” phase of the acute inflammatory phase. Apoptotic cellular functions are specifically engaged subsequent to recognition and encountering of antigens, mitogens, or infective agents and infected host cells, which result in presentation of processed antigen/pathogen to effector T or B cells for additional cell-mediated and/or humoral responses. The characteristics of the injurious stimuli – pathogens, neoplastic cells with aberrant DNA and/or proteins – as well as the target organs and associated tissues (immune privileged, epithelial, lymphoid, or mucosal tissues) each with the ability to initiate specialized inflammatory cells, such as, for example, natural killer cells (NKs), macrophages (MFs), polymorphonuclear neutrophils (PMNs), dendritic cells (DCs), mast cells (MCs), basophils or eosinophils as well as T or B cells become stimulated and specifically differentiated in order to produce the required amounts of proinflammatory mediators. Acute pro-inflammatory apoptotic mediators may be produced and secreted from the various cytosolic, endoplasmic reticulum, nuclear or mitochondrial compartments of the activated cells. An example of a pro-inflammatory response phase process is demonstrated in the respiratory tract, subsequent to exposure of the lungs to certain pathogens or certain pathogenically secreted compounds such as bacterial lipopolysaccharides (LPS). A cascade of signals in the membranes of DCs are induced, which in turn instigate adequate pro-inflammatory responses such as subsequent expression of pathogen (LPS)-specific receptor molecules, [toll-like-receptors (TLRs) TLR-1–11 to additionally promote the production of mediators that are also termed alarmins (danger molecules) or death factors (DFs)]. Pro-inflammatory signals or DFs that are activated in response to TLRs include TNF-a and TNF receptor protein (TNF-R), cell surface antigen molecules and their respective receptors and/or recognition molecules [e.g., TNF-related

20  The Pro-inflammatory Milieu and Its Role in Malignant Epithelial Initiation

413

apoptosis inducing ligand (TRAIL), TCR, TNF-R/TNF-SF5, IgE-FceR, TGF-R, CCL13/monocyte chemoattractant protein 4 (mcp-4), phosphatase and tensin homolog (PTEN), B cell leukemia/lymphoma associated protein 2-associated X protein (Bax), bcl2-associated death promoter (BAD), Fas-associated via death domain (FADD), or CD2, CD11, CD18, CD25, CD50, CD54, CD63, CD69, CD88]. Apoptotic responses of the pro-inflammatory response phase also include expression of a myriad of preformed cytokines, chemokines, and associated factors such as macrophage inflammatory protein 1 (MIP-1), eosinophil-derived neurotoxin (EDN), eosinophil chemotactic factor of anaphylaxis (ECFA), CCL11/eotaxin-1, MPIF-2/eotaxin-2, TGF-a, b, ILs (e.g., IL-1, IL-3, IL-4, IL-6, IL-7), (capsases 1–10; extracellular matrix proteases or membrane metalloproteases (MMPs)). Importantly, many newly synthesized vasoactive mediators, [such as histamine (his)], heparin (hep)], and various enzymes (like elastase, tryptase, chymase, and perforin) MHC class I/II molecules and their co-stimulatory molecules (e.g., DC28, DC86), and chemokine receptor family CCRs (e.g., CX3CR1, ILBrA, XCR1/CCXCR1), are also synthesized. Moreover, during the apoptotic pro-inflammatory phase a number of oxidants [e.g., ROS, reactive nitrogen species (RNS), superoxide (SO), hydroxyl ions (OH–), hydrogen peroxide (H2O2), or hydroclorous acid (HOCl)] are generated from the myriad activated inflammatory MCs, DCs, MFs, NKs, basophils, PMNs, eosinophils, T or B cells as well as vascular cells (Bonasio and von Andrian 2006; Kabelitz and Medzhitov 2007; Wang et al. 2008; Banchereau and Steinman 1998; Fukata and Abreu 2008; Conroy et al. 2008; Wilson and Villadangos 2004). It is important to note that the programmed cell death process of this part of the acute inflammatory response is also associated with the mobilization of various cell membrane enzymes and/or enzyme components, such as the case of phospholipases A, C, or D, of the metabolism of arachidonic acid, activation of the COX and lipooxygenase pathways and the synthesis of a multitude of PGs and leukotrienes, of activation of sphingosine kinase binding proteins and calcium mobilization resulting in increased cellular energy NADP+, of and increased mitochondrial oxidative metabolism and the production of oxidants as already indicated. Mast cell-derived programmed cell death often requires antibody-induced activation of blood platelets and complement cascades to assist cell lysis as well as the expression of receptor molecules essential for B cell-derived antigen-specific antibody synthesis, e.g. IgG, IgE, IgM, or IgA, and antibody-receptor aggregation (Lodoen and Lanier 2006; Valent et al. 2001; Bonasio and von Andrian 2006; Drayton et al. 2006; Kabelitz and Medzhitov 2007; Banchereau and Steinman 1998; Conroy et al. 2008; Wilson and Villadangos 2004; Yano et al. 1997; Juliet et al. 2003; Khatami et al. 1984; Fukata et al. 2005; Selsing 2006; Caro and Cederbaum 2006; Ibrahim et al. 2004; Peerschke et  al. 2006; Monteseirin et  al. 2004; Nyakern et  al. 2006; Fischetti and Tedesco 2006). In conclusion, during the programmed cell death or the “apoptosis” part of the acute inflammatory phase, DFs are orchestrated in a timely spatial-temporal manner that will ensure that subsequent to the recognition of the causing pathogen, its correct processing will occur, and that through the deployment of the adequate molecular and cellular mechanisms its presentation to the relevant innate immune cells or their counterparts adaptive immune subpopulations T or B lymphocytes for additional cellular responses will ensue.

414

A. Yagui-Beltrán et al.

2. The resolution phase of acute inflammation.  The termination or resolution part of the acute inflammatory response ensures that any redundant toxicity remaining in target organs after the relevant cascade of apoptotic danger signals has exerted their function (to resolve and terminate the initial inflammatory response, while allowing normalization of functional relationship between the host and target organ so that tissue repair can also commence) is efficiently nullified by postinflammatory processes facilitated by a myriad of recruited anti-inflammatory and anti-apoptotic mediators. The mediators are specific to the type of stimuli, degree of tissue injury, and the quantity of death signals produced in the host during apoptosis. Examples of mediators involved in wound healing include growth factors, anti-apoptotic mediators including lymphocyte NF-kB and mitogen-activated kinases (MAPKs). Other anti-inflammatory mediators that are generated in this phase include IFNs: IFN-a, IFN-g: ILs (IL-5, IL-8, IL-10, IL-13) B cell leukemia/lymphoma associate protein 2 (BCL-2), PKC, Jun kinase (JNK), IkB kinase (IKK), membrane phosphotidylinositide protein kinases (PI3K)/Akt pathways; ataxia telangiectasia mutated homolog (ATM); fibroblast growth factor (FGF); platelet-derived growth factor (PDGF), nuclear expression of Ataxia Telangiectasia, and a variety of inhibitors of toxins, antioxidants, and reducing enzymes [e.g., superoxide dismutase (SOD), NADP+ reductase, catalase, thioredoxin reductase (Trx)]; NO; adaptor molecules and protein complexes [e.g., toll-interleukin 1 receptor domain containing adaptor protein (TRAP), translocation-associated membrane protein (TRAM), MyD88, interleukin-1 receptor-associated kinases (IRAKs), TNF receptor-associated factors (TRAFs)] (Lodoen and Lanier 2006; Ferrantini et  al. 2008; de Visser et  al. 2006; Wang et al. 2008; Banchereau and Steinman 1998; Yano et al. 1997; Juliet et al. 2003; Khatami et al. 1984; Selsing 2006; Caro and Cederbaum 2006; Ibrahim et  al. 2004; Peerschke et  al. 2006; Monteseirin et  al. 2004; Nyakern et  al. 2006; Fischetti and Tedesco 2006; Perez et  al. 2006; Mantovani et  al. 2008; Snook et al. 2006; Ledeen and Wu 2006; Hanukoglu 2006; Berger 2006; Ii et al. 2006; Stenmark et al. 2006).

Malignant Epithelial Initiation Within a Pro-inflammatory Milieu The process of tumor development is associated with the growth and expansion of not only tumor cells but also stroma, new blood vessels, and a myriad of infiltrating inflammatory cells, and it is the interaction between these different cell populations that helps disseminate tumor development and spread. Traditionally for many decades, studies supported and focused around the dogma that acute inflammation through its deployment of various aspects of the immune system exerted a major role to mainly eliminate cancer (Nauts and McLaren 1990). Nevertheless, acute inflammatory processes are by definition self-limiting, and often develop into chronic inflammation, with links to cancer initiation as explained above (Kuper

20  The Pro-inflammatory Milieu and Its Role in Malignant Epithelial Initiation

415

et al. 2000; Fidler 1995; Karin and Greten 2005; Hold and El-Omar 2008). Immune responses elicited against cancerous cells, despite some similitude to those deployed against pathogenic organisms are generally weak and not adequate to eradicate malignant initiated epithelial cells (Whiteside 2008). A plausible explanation is that the initial signals originated by the malignant epithelial cells are not recognized as equivalently strong as those danger signals emanating from exogenous pathogens (Gallucci and Matzinger 2001). Other explanation would be that the host harboring the cancerous cells subsequently downregulate the perceived immune self– responses, or alternatively that certain components of the TME act as barriers blocking access of the immune system that would normally function to suppress or eliminate the cancer cells (Zhang 2008). More recently, the NF-kB and STAT3 pathways have gained significant importance as major protagonists playing important roles ensuring the persistence of chronic inflammation through the regulation of pro-inflammatory cytokines and as pivotal mediators of the pro-inflammatory milieu that facilitates the growth and malignant transformation of initiated epithelial cells (Karin and Greten 2005; Karin 2006; Rius et al. 2008).

Tumor Immune Evasion and Progression Within Sustained Chronic Inflammation Studies have shown that in order for cancer development to occur successfully, suppression of T cell as well as humoral immune responses must occur. Various factors, including the often heterogeneous nature of tumors, as well as their still poorly understood ability to evade and suppress the attack of the host’s immunity, have been postulated as likely contributors to the inadequacy of an antitumor response. Autologous effector T cells that normally function to suppress or eliminate cancer cells are downregulated by regulatory T cells (Tregs) that infiltrate and accumulate the TME. These regulatory T cells are often activated or recruited by DCs infiltrating the TME (Bacchetta et al. 2007). The other population of cells that have been suggested to play an important role in suppression of tumor infiltrating lymphocytes are myeloid-derived suppressor cells or MDSCs (in the case of mouse models, the CD11b+/Gr-1+ myeloid progenitor cell populations) that are present in increased number both in the cancerous lesions and peripherally in lymph nodes and the spleens of the tumor bearing animals (Talmadge et al. 2007; Murdoch et al. 2008). Despite the abundant ongoing MDSC research across Europe and the United States; it is important to emphasise that the nature and functional significance of MDSCs in the field of tumor immunology that remains rather controversial and undefined. For example, MDSCs were originally believed to exert an immune suppressive function; however, newer research in the field proposes a pro-inflammatory as well as a pro-angiogenic function for these immune cell populations (Murdoch et al. 2008). It seems therefore that tumor-associated inflammation exerts both a pro-inflammatory and an anti-inflammatory set of signals; and when a polarization of the balance favors a pro-inflammatory milieu malignant initiated epithelial cells are then able to

416

A. Yagui-Beltrán et al.

evade the various mechanisms of immune surveillance, and are then able to grow and mature into cancers while developing their own blood vessels, eventually breaking the basement membrane and spreading into distal sites (metastasis).

Tumor Progression, Metastatic Potential, and Inflammation Recent research by various groups has indicated that inflammation also exerts an imperative function in the ability of cancer cells to spread and metastasize; this is thought to be due to the capacity of initiated epithelial cancer cells, with the acquired ability to spread, to express chemokine receptors (Muller et  al. 2001). The expression of specific chemokine receptors is an essential part of the machinery necessary to support the signals elicited by paracrine secretion of pro-inflammatory cytokines, such as IL1-b, IL-6, and TNF-a as well as the signals elicited through the production of other cytokines produced in an autocrine manner (Kulbe et  al. 2007). An example of how the pro-inflammatory milieu can influence epithelial malignant initiation and metastatic spread is demonstrated by a knockdown study of TNF-a by Burger et al. in 2006 (Burger and Kipps 2006); in their experiments, they were able to corroborate that mesenchymal or marrow-derived stromal cells, constituting a large proportion of the noncancer cell population within the TME, and secreting the chemokine stromal cell-derived factor-1 (SDF-1/CXCL12), attract cancer cells (through its receptor CXCR4, expressed by both hematopoietic and nonhematopoietic tumor cells). CXCR4 promotes tumor progression by direct and indirect mechanisms. CXCR4 is essential for metastatic spread to organs where CXCL12 is expressed, allowing tumor cells to access cellular niches, such as the marrow that favor tumor-cell survival and growth. Stromal-derived CXCL12 itself can stimulate survival and growth of neoplastic cells in a paracrine fashion. Moreover, CXCL12 can promote tumor angiogenesis by attracting endothelial cells to the TME. Burger and colleagues’ knockdown of TNF-a resulted in decreased cellular CXCR4 and CXCL12 expression and decreased ability of the tumor to spread, highlighting that at least in preclinical tumor models CXCR4 antagonists may have antitumor activity in patients with various malignancies. This may be potentially exploited to target the TME therapeutically in the treatment of neoplastic disease (Burger and Kipps 2006). Apart the specific cytokines released by the tumor cells from the various other elements of the microenvironment, from the stroma, endothelial cells, and from noncancer cells, studies have importantly shown that for cancer development to occur certain infiltrating leukocytes and particular tumor-associated macrophages or TAMs must exert their function. TAMs have been the focus of research of many groups across the globe in the last two decades with a multitude of data on their functional significance in cancer development. TAMs have been traditionally subdivided into pro-inflammatory M1 and anti-inflammatory M2, and their

20  The Pro-inflammatory Milieu and Its Role in Malignant Epithelial Initiation

417

polarization into either in the tumor dictates the behavior of the neoplasm, its ability to metastasize and its degree of malignancy and this has been corroborated in an organ specific manner (Andreu et al. 2010; DeNardo et al. 2009; Lin et al. 2001; Mantovani et al. 2008).

Soluble Mediators of the Immune Response in the Pro-inflammatory Milieu Responsible for Cancer Development as well as Maintenance of Chronic Inflammation Various animal and human studies have showed a dynamic interplay between initiated epithelial cells and the myriad components of the TME necessary for the development of cancer. Undoubtedly, the fields of tumor immunology and the TME have been rapidly expanding scientifically while gaining popularity as an attractive integral model that serves to explain the process of carcinogenesis and more importantly as a model system to design novel therapies targeting the various components of the TME in a specific manner, affecting the various hallmarks of cancer progression. Inevitably, such an expansion in the field means that many of the findings are not devoid of controversy and require further ongoing corroboration before scientific robustness is achieved. The following are the best-known inflammatory mediators exerting important roles within the TME initiating epithelial cells promoting their carcinogenesis.

Oxidative Stress Species and Their Functional Significance in Cancer Development A major source of endogenous reactive oxygen (RO) and RNS: O2-, NO, H2O2, OH, ONOO-, HOCl are innate immune cells. As explained above, during the acute phase of inflammation, the release of these compounds plays a beneficial role in the elimination of pathogens (Hussain and Harris 2007). However, a persistence of inflammation, as is the case in chronic inflammation, results in sustained exposure of epithelial cells to oxidative stress – an exposure that is eventually harmful, as it can lead to abnormal genomic alterations (DNA strand breaks and base modifications), lipid and protein peroxidation and aberrant activation of signal transduction by posttranslational modification (RNS is a mediator of MAPK and AP-1 signal transduction. NO can induce posttranslation modification of p53 via ATM or ATR, or specific induction of anti-apoptosis genes (AIF) (Wu 2006; Azad et  al. 2008; Hussain et al. 2000). These mechanisms link ROS and RNS production, and hence the innate immune system, to the ability of tumors to induce angiogenesis, to proliferate, and to promote distant metastasis. Additionally, oxidative stress has also been

418

A. Yagui-Beltrán et al.

associated with impaired DNA methyltransferase activity (DNMT-1) or MBP binding, and therefore provides a link to a myriad of epigenetic alterations (Shahrzad et al. 2007; Lim et al. 2008).

The Role of Matrix Remodeling Proteases in the Tumor Microenvironment Matrix metalloproteinases are crucial in establishing crosstalk and co-regulation among the various cellular elements of the TME. In a nutshell, the majority of MMPs are expressed by stromal and immune cells in the pro-inflammatory milieu in the TME (MMP7 is an exception and is predominantly expressed by epithelial cells). For the majority of MMPs, it logically follows that in cancer, a higher MMP expression tends to correlate with poorer prognosis and worse patient outcome. Research in various study models has demonstrated that extracellular proteases in the TME, as well as having a function of remodeling of certain extracellular matrix (ECM) components, can also modulate a range of growth factors, cytokines and chemokines. In this respect, MMP9 has been shown to be important in the regulation of malignant epithelial proliferation, specifically in angiogenesis through the processing of pro-growth factors such as vascular endothelial growth factor (VEGF). Interestingly, studies have shown that MMP9 expression correlates with a more invasive and aggressive phenotype in breast and skin cancers (van ’t Veer et al. 2002), which may perhaps be related to its ability to promote vasculogenesis. Anh et  al. in a study in 2008 demonstrated that MMP9 may be necessary for the recruitment of bone marrow derived progenitor cells (MDSCs), such as endothelial progenitor cells (Ahn and Brown 2008); they showed that intra-tumoral expression of MMP9 by MDSCs is sufficient to restore angiogenesis in irradiated mice (Ahn and Brown 2008). MMP7 is able to process RANKL, positively influencing metastasis through tumorinduced osteolysis (Ahn and Brown 2008). McCaig et al. showed in a study in 2006 that infection with H. pylori directly induced MMP7 in gastric cancer epithelial cells; they also showed that MMP-7 could act as an epithelial-derived signal increasing the bioavailability of IGF-II released from myofibroblasts and because IGF-II could function on both stromal and epithelial cells, their findings suggested that increased MMP-7 expression contributes to redefining the niche occupied by dividing cells and leading to hyperproliferation occurring in H. pylori infection (McCaig et al. 2006). The field of MMPs is regaining momentum despite the fact that their targeting has been often complicated because of their parallel functions in both normal tissue homeostasis and cancer development and their varying functions depending on tumor biology and stage; this is particularly true as the gap in the understanding of the molecular mechanisms governing the function of myriad components of the TME and their dynamic interplay decrease.

20  The Pro-inflammatory Milieu and Its Role in Malignant Epithelial Initiation

419

Functional Significance of Specific Transcription Factors and Primary Inflammatory Cytokines In the wide array of molecular factors in the pro-inflammatory milieu promoting cancer development, a number of principal or players have emerged through the consistent work of various research groups across the world. These include transcription factors such as nuclear factor-kappaB (NF-kB), signal transducer activator of transcription-3 (Stat3), and primary inflammatory cytokines such as IL-1b, IL-6, IL-23, and TNF-a (Karin 2006; Yu et  al. 2007; Voronov et  al. 2003; Langowski et al. 2006). NF-kB is a pivotal orchestrator of innate immunity/inflammation, and aberrant NF-kB regulation has been observed in a number of malignancies (Karin 2006). In both tumor and infiltrating inflammatory cells, NF-kB is activated downstream of the toll-like receptor (TLR)-MyD88 pathway (sensing pathogenic microbes and tissue damage) and of the inflammatory cytokines TNF-a and IL-1b. In addition, NF-kB activation can be the result of cell-autonomous genetic alterations (amplification, mutations or deletions) in cancer cells. Interestingly, NF-kB can be activated in response to hypoxia though to a lesser extent than hypoxia inducible factor (HIF)-1a (Balkwill 2009; Carbia-Nagashima et al. 2007; Taylor 2008). Accumulating evidence suggests that intersections and compensatory pathways may exist between the NF-kB and HIF-1a systems linking innate immunity to the hypoxic response. NF-kB induces the expression of inflammatory cytokines, adhesion molecules, key enzymes in the prostaglandin synthase pathway (COX-2), NO synthase, and angiogenic factors. Furthermore, by inducing anti-apoptotic genes (e.g., Bcl2), it promotes survival in tumor cells and in epithelial cells targeted by carcinogens. Research has unequivocally demonstrated that NF-kB is involved in malignant epithelial initiation and progression in tissues where sustained chronic inflammation has been classically described, for example, the gastrointestinal tract and the hepatobiliary tract (Karin 2006; Greten et al. 2004; Pikarsky et al. 2004). Similarly, as when discussing the ability of IL-6 to polarize T cell differentiation toward a pro-inflammatory phenotype, precise inactivation of NF-kB in tumor-infiltrating leukocytes, by a strategy targeting Ik-Bkinase beta, inhibited colitis-associated cancer, thus providing genetic proof for the specific function of NF-kB and infiltrating leukocytes in intestinal cancer development (Greten et al. 2004). Various inhibitors functioning at different levels of the NF-kB pathway ensure its adequate control. For example, Tir8 is an orphan member of the IL-1R family that is highly expressed in the mucosa of the gastrointestinal tract. In a murine model of intestinal cancer, Tir8-deficient mice, in response to dextran sulfate sodium salt and azoxymethane administration, demonstrated exaggerated inflammation with an associated increase for colon cancer risk compared to normal animals. Furthermore, this was associated with local production of PG E2, proinflammatory cytokines (IL-1, IL-6), and chemokines (KC/CXC, JE/CCL2 and CCL3) (Garlanda et  al. 2007; Xiao et  al. 2007). Thus, the lack of a checkpoint (Tir8) of NF-kB activation leads to increased carcinogenesis in the gastrointestinal tract, highlighting once more the association between chronic inflammation and

420

A. Yagui-Beltrán et al.

malignant initiation of epithelial cells. As indicated already, there is a role for STAT3 as a point of convergence and also quite importantly for perpetuation of several oncogenic signaling pathways (Yu et al. 2007; Lee et al. 2009). This transcription factor is constitutively activated both in malignant cells and also in immune cells and plays a role in carcinogenesis as well as in tumor immune evasion by halting DCs maturation (Yu et al. 2007; Kortylewski et al. 2005; Becker et al. 2004). Research on colon cancer revealed that STAT3 is a major controller of cell proliferation and survival, regulating the expression of c-Myc, Mcl-1, Cyclin D, and Bcl-2 (Becker et al. 2004). Genetic studies in animal and human experiments have demonstrated that cytokines play a key role in linking chronic inflammation and cancer. They are activated by both inflammatory and tumor cells and their effect is similarly important for both sustaining chronic inflammation, promoting the progression of malignant epithelial cells, and also in inhibiting immune-mediated tumor surveillance. Although, conveniently, cytokines have been divided into either pro-inflammatory – IL-1, 6, 8, 11, 12, 18, 23, IFN-g, TNF-a, and MIF (e.g., IL-1b overexpression alone, in the absence of Helicobacter infection, is sufficient to cause gastric cancer (Tu et al. 2008) – or anti-inflammatory (IL-4, 10, IFN-a and b, TGF-b), compounding research is showing that an increasing number of these molecules have dichotomous roles. TGF-b is an example of a pleiotropic cytokine that can function to suppress cancer in certain circumstances, while having the capacity to promote cancer progression through the manipulation of definite microenvironment pathways during others. For example, complete deletion of TGF-b has been linked with increased tumor incidence; nevertheless most cancers are also able to progress and spread in the presence of high levels of TGF-b (Gonda et al. 2009). TGF-b has also been associated with the recruitment of MDSCs into the TME, suppression of cytotoxic T lymphocytes (CTLS) CD8+ T cell and infiltration of NK, i.e., facilitating an inflammatory milieu that is more favorable to tumor growth and development. IL-6 has been linked to chronic inflammation and tumor development in the colon and stomach mainly through an aberrant activation of STAT 1 and 3 (Bromberg and Wang 2009). Like IL-6, TGF-b polarizes T cell differentiation toward a pro-inflammatory phenotype. This year, two separate research groups have shown that IL-6 deficient mice were not able to develop colitis-associated carcinoma in a DSS/AOM mouse model and that the IL-6 effect was regulated by STAT-3 in intestinal epithelial cells (Bollrath et al. 2009; Grivennikov et al. 2009). Interestingly, in this murine model bone marrow derived myeloid cells were shown to be the source of IL-6. Moreover, IL-6 in the cancers was demonstrated to be effective not only on intestinal epithelial cells, but also its effect was important in other components of the microenvironmental milieu, leading for example to activation of DCs and T helper cells. The overall result of this dynamic interplay between IL-6, malignant epithelial cells, and other components of the microenvironment is thought to be the perseverance of cytokine production and secretion and further propagation of the chronic inflammatory state exacerbated by a pro-tumorigenic cytokine soup (Bromberg and Wang 2009). This is an example of various integrative models advocating that a proinflammatory cytokine, IL-1b, exerts an essential role in the recruitment of bone

20  The Pro-inflammatory Milieu and Its Role in Malignant Epithelial Initiation

421

marrow-derived myeloid cells, which in turn function as a source for IL-6 production. Moreover, IL-6 induction of STAT3 in epithelial cells may result in further malignant epithelial transformation. It is important to highlight the potential importance that inhibitors of these cytokines and their signal transducers (STAT-3/JAK) may have as novel anticancer targeted therapies. We have also indicated that IL-6, a multifunctional cytokine with growthpromoting and anti-apoptotic activity, is considered to be a pivotal effector molecule of the NF-kB pathway (Lin and Karin 2007; Naugler and Karin 2008). We also highlighted some of the recent research on the NF-kB–IL-6–STAT3 cascade, where IL-6 was found to be produced by myeloid cells and to be a key to the development of gastrointestinal cancer. IL-6 protects normal and premalignant intestinal epithelial cells from apoptosis and favors the proliferation of tumor-initiating cells (Bollrath et al. 2009; Grivennikov et al. 2009). Interestingly, STAT3 also regulates the balance between IL-12 and IL-23 in the TME, and consequently the polarization of T-helper subsets (Kortylewski et  al. 2009). New mechanistic insights into links between IL-6 and cancer have been shown in a mouse model hepatocellular carcinoma, where in male animals IL-6 was shown to promote liver inflammation, injury, compensatory cell proliferation, and cancer development. In females, estrogen steroid hormones had a cancer protective effect through the inhibition of IL-6 production (Naugler et al. 2007; Rakoff-Nahoum and Medzhitov 2007). IL-1 is a pro-inflammatory cytokine with a well-described ability to promote cancer development and spread by influencing important pro-cancer mediators in the pro-inflammatory milieu (Mantovani et al. 2008; Giavazzi et al. 1990; Luo et al. 2007). The IL-1R1 gene-targeted mouse model system have provided good evidence for the pro-tumor potential of IL-1 (Dinarello 2006). Particularly, in models of chemical carcinogen-induced tumors, IL-1B secreted by malignant cells or infiltrating leukocytes contributes to properties of malignancy such as angiogenesis, immune suppression increased tumor adhesiveness and invasion, whereas IL-1ra negatively controls these processes (Krelin et  al. 2007). In diethyl-nitrosamineinduced hepatocarcinoma, the unique membrane-associated form of IL-1a acts as protumorigenic mediator; diethyl-nitrosamine-induced hepatocyte death results in the release of IL-1a and activation of IL-1R signaling, leading to IL-6 induction and compensatory proliferation, critical for hepatocarcinogenesis (Sakurai et al. 2008). The importance of IL-1a is also worth mentioning in 3-methylcholantrene-induced fibrosarcoma for its efficiency in activating antitumor innate and specific immune responses by acting as a focused adjuvant through binding to IL-1RI on cells deputed to immune surveillance (Marhaba et al. 2008; Elkabets et al. 2009). This “danger signal” has been successfully collected artificially for mounting antitumor immunity from dying cells (Chen et  al. 2007). Stomach-specific expression of human IL-1b in transgenic mice lead to spontaneous gastric inflammation and cancer that correlated with early recruitment of MDSCs (Tu et al. 2008). Last, but not least, recent studies have demonstrated a relationship between steroid hormones, IL-1, and cancer; in carcinoma of the prostate, androgen-dependent tumor sensitivity to hormonal stimulation is regulated by selective androgen receptor modulators. The

422

A. Yagui-Beltrán et al.

inflammatory cytokine IL-1 produced by macrophages in the TME converts selective androgen receptor modulator from inhibitors to stimulators, thus inducing resistance to hormonal therapies (Zhu et al. 2006) and highlighting the complexity and potential for novel cancer therapy that a good understanding of the functional significance of the pro-inflammatory milieu and the microenvironment entails. TNF is a major cytokine and player in the pro-inflammatory milieu exerting a cancer-inducing role on malignant initiated epithelial cells. TNF, which originally was identified as a cytokine inducing hemorrhagic necrosis in tumors, was soon after discovered to also have pro-tumoral functions. The finding that TNF-deficient mice are protected from skin cancers offered genetic proof associating TNFmediated inflammation and cancer (Balkwill 2009; Moore et al. 1999). TNF has been shown to be versatile in how it is able to enhance tumor development through a number of mechanisms: it can successfully facilitate the infiltration of tumor promoting leukocytes; it promotes angiogenesis, metastatic spread, and also assists epithelial-to-mesenchymal transition or EMT (e.g., through the activation of the Wnt/beta-catenin pathway subsequent to TNF signaling from activated gastric cancer TAMs), processes that have the overall result of tumor enhancement and growth (Kulbe et al. 2007; Balkwill 2009; Popivanova et al. 2008; Oguma et al. 2008). Not to ignore is the immune suppressive functions of some members of the TNF family. An example is that of the decoy receptor-3 that has been linked to the downregulation of major histocompatibility complex class II in TAM (Chang et al. 2008). Like in the case of other components of the pro-inflammatory milieu in the TME, a better understanding of the molecular mechanisms underlying this cytokine and its dynamic interplay with others is serving as a platform for the design of novel anticancer therapies. Indeed, there have already been some phase I and II clinical cancer studies with TNF antagonists to assess their efficacy and safety in a variety of neoplasms (Harrison et al. 2007; Brown et al. 2008).

Functional Significance of Myeloid Cell Recruitment Within Tumors It is estimated that in any single tumor, half of the component cells are tumor epithelial cells and the other half is composed of noncancerous epithelial cells and a stroma containing fibroblasts, vessels, and leukocytes, also often referred to as the TME (Allavena et al. 2008). In the majority of published studies, TAMs seem to be the main subset of leukocytes driving an amplification of the inflammatory response in the tumor milieu; although recent studies are showing that the predominating infiltrating immune cell subpopulations may be other than TAMs and their presence may be organ specific (Jablons and Coussens, unpublished). In fact, part from TAMs, also mast cells, neutrophils, and even effectors of the adaptive immunity (especially in the form of antibodies) may activate inflammatory reactions that promote cancer progression (de Visser et al. 2006; Galli et al. 2008). Chemokines have long been associated with the recruitment of TAM in tumors (e.g., CCL2 and

20  The Pro-inflammatory Milieu and Its Role in Malignant Epithelial Initiation

423

CCL5) (Mantovani et al. 2008; Balkwill 2004). For their phenotypic and functional properties, TAM resembles M2-polarized macrophages, though there are some specific characteristics (Mantovani et al. 2002; Sica et al. 2006). In the majority of published animal and human studies, accumulation of TAM correlates with the angiogenic switch, poorer patient outcome, and reduced survival (Balkwill et  al. 2005; Mantovani et al. 2008; Bingle et al. 2002; Pollard 2004). TAMs have been shown to be pro-tumorigenic because as they preferentially accumulate in hypoxic regions of tumors they facilitate the release of pro-inflammatory growth factors, cytokines, and matrix-degrading enzymes, such as MMPs (described above) (Wyckoff et  al. 2007; Mantovani et  al. 2006; Condeelis and Pollard 2006; Yang et al. 2008) while enriching the pro-tumorigenic niche with a florid array of angiogenic factors: VEGF, PDGF, FGF, and CXCL8 (Murdoch et  al. 2008; Ahn and Brown 2008; Du et al. 2008; Seandel et al. 2008; Dineen et al. 2008; Kusmartsev et al. 2008; Duluc et al. 2007; Morandi et al. 2007; Robinson-Smith et al. 2007; Stearman et  al. 2008; Lewis and Pollard 2006) facilitating the growth of further tumor cells while perpetuating an environment of sustained chronic inflammation with everything that entails in terms of malignancy and metastasis. Proving some of the mechanisms behind secondary localization and seeding of cancer are some recent studies showing that VEGF1R+ hematopoietic cells home to tumor-specific pre-metastatic sites that favors secondary localization of cancer (monocytes express VEGF receptors and VEGF being a recognized chemoattractant of myeloid cells in cancers) (Fischer et al. 2007; Kaplan et al. 2005; Witz 2008). Chemokines (e.g., CXCL5 and CXCL12) are also involved in the attraction of MDSCs (Sawanobori et al. 2008; Sica and Bronte 2007). MDSCs, like TAM, are important effectors in tumor angiogenesis (Gonda et al. 2009; Bromberg and Wang 2009) and Gr+Mac1+ cells, presumably MDSC, have been shown to mediate resistance to anti-angiogenic therapy (Shojaei et al. 2007). Subsequent to the malignant initiation of epithelial cells within the rich proinflammatory pro-cancer milieu, further sustainment of chronic inflammation and cancer development and spread is principally mediated by the host’s failure to mount a protective antitumor immune response. TAM and MDSCs express an of immunosuppressive armamentum responsible for this. As well as inhibiting the CTL response, which in physiological circumstances is mediated by CD8 T cell activation through the expression of NOS2 and Arg1, MDSCs seem to promote the expansion of CD4+FOXP3+ T-regulatory cells and an M2 polarization of TAM (Sica and Bronte 2007; Huang et  al. 2006; Sinha et  al. 2007; Nagaraj and Gabrilovich 2008). The immunoregulatory activity of TAM is mostly influenced by tissue specific local cues. In the tumor milieu, a number of immunosuppressive factors (e.g., IL-10 and transforming growth factor-b) have been postulated by several studies to be important in the differentiation of infiltrating monocytes toward an M2 macrophage phenotype (Mantovani et  al. 2002; Pollard 2004; Mantovani et  al. 2006; Allavena et  al. 2008). NF-kB, an important tumorigenic mediator of the pro-inflammatory milieu and a pivotal player connecting inflammation to many of the essential and known processes required to drive malignant transformation and already highlighted earlier, has also been recently associated to

424

A. Yagui-Beltrán et al.

the M2 polarization of TAM (Hagemann et al. 2008). Increasing evidence supports that metabolic changes in the tumor pro-inflammatory milieu, apart from providing the advantageous environment through the availability of a multitude of growth factors, can also have a more profound effect on infiltrating leukocytes (Tennant et al. 2009), a premise not difficult to comprehend. For example, several studies have demonstrated that tumor cells are able to secrete lactic acid that in turn stimulates the IL-23/IL-17 axis in TAM (Shime et al. 2008) at the expense of the immunoprotective IL-12-inducible Th1 pathway.

Relationship of Bone Marrow-Derived Cells and the Tumor Microenvironment This topic will be extensively discussed in another chapter. Cytokines and other tumor derived soluble factors not only recruit to tumors and well-defined inflammatory cells but also, as increasingly recognized, bone marrow-derived progenitor cell populations. BMDC that are recruited to the site of chronic inflammation have demonstrated remarkable plasticity and have been shown to differentiate into cells of diverse lineages. In vitro, this differentiation is governed by secreted factors. Similarly, in vivo their differentiation seems to be governed by the tissue (tumor) microenvironment (Gonda et al. 2009).

Conclusion Signals from the microenvironment have a profound influence on the maintenance and/or progression of epithelial cancers. Increasing research evidence from animal and human studies demonstrates that the pro-inflammatory milieu with its various components and soluble mediators plays a fundamental role in the initiation of malignant epithelial cells while sustaining and perpetuating chronic inflammation that in turn results in further malignant proliferation and distant malignant spread and seeding. As the fields of tumor immunology and tumor biology progress, the focus of investigation is starting to shift from merely identifying and describing the components of the protagonists of infiltrating immune cell populations in the proinflammatory milieu to a more careful definition of the functional significance and mechanisms behind their complex dynamic interplays. This is slowly helping the scientific and medical community to understand how chronic inflammation is exactly contributing to the genetic and epigenetic aberrant alterations that are observed in the pre-cancerous and cancerous cells dwelling within the pro-tumor milieu and cancer microenvironment. Of importance in this dynamic understanding of the TME, inflammation, and the neoplastic process is the recognition of the increasing evidence supporting that there may be small populations of pluripotent cells with the ability of self-renewal that may be recruited and forced by pathways

20  The Pro-inflammatory Milieu and Its Role in Malignant Epithelial Initiation

425

not yet fully understood into malignant initiation contributing to the development and spread of cancer. It is hoped that with better knowledge of the molecular mechanisms that link the stem-cell like populations and niches, with the proinflammatory microenvironment, we will be able to design novel targeted therapies aimed at specific components of these exciting pathways. Acknowledgment  The authors are supported by grants from the National Institutes of Health (CA132566 and CA130980), Department of Defense Award for mesothelioma research (W81XWH-09-1-0342), an Era of Hope Scholar Award (W81XWH-06-1-0416), the Bonnie J. Addario Lung Cancer Research Foundation, the Kazan, McClain, Abrams, Fernandez, Lyons, Greenwood, Harley & Oberman Research Foundation Inc., and the Eileen D. Ludwig Endowed Chair Fund for Thoracic Oncology Research.

References Ahn GO, Brown JM (2008) Matrix metalloproteinase-9 is required for tumor vasculogenesis but not for angiogenesis: role of bone marrow-derived myelomonocytic cells. Cancer Cell 13(3):193–205. Allavena P et  al (2008) Pathways connecting inflammation and cancer. Curr Opin Genet Dev 18(1):3–10. Allavena P et al (2008) The Yin-Yang of tumor-associated macrophages in neoplastic progression and immune surveillance. Immunol Rev 222:155–161. Andreu P et al (2010) FcRgamma activation regulates inflammation-associated squamous carcinogenesis. Cancer Cell 17(2):121–134. Azad N, Rojanasakul Y, Vallyathan V (2008) Inflammation and lung cancer: roles of reactive oxygen/nitrogen species. J Toxicol Environ Health B Crit Rev 11(1):1–15. Bacchetta R, Gambineri E, Roncarolo MG (2007) Role of regulatory T cells and FOXP3 in human diseases. J Allergy Clin Immunol 120(2):227–235; quiz 236–237. Balkwill, F (2004) Cancer and the chemokine network. Nat Rev Cancer 4(7):540–550. Balkwill, F (2009) Tumour necrosis factor and cancer. Nat Rev Cancer 9(5):361–371. Balkwill F, Mantovani A (2001) Inflammation and cancer: back to Virchow? Lancet 357(9255):539–545. Balkwill F, Charles KA, Mantovani A (2005) Smoldering and polarized inflammation in the initiation and promotion of malignant disease. Cancer Cell 7(3):211–217. Banchereau J, Steinman RM (1998) Dendritic cells and the control of immunity. Nature 392(6673):245–252. Becker C et al (2004) TGF-beta suppresses tumor progression in colon cancer by inhibition of IL-6 trans-signaling. Immunity 21(4):491–501. Berger SA (2006) Signaling pathways influencing SLF and c-kit-mediated survival and proliferation. Immunol Res 35(1–2):1–12. Bingle L, Brown NJ, Lewis CE (2002) The role of tumour-associated macrophages in tumour progression: implications for new anticancer therapies. J Pathol 196(3):254–265. Bollrath J et al (2009) gp130-mediated Stat3 activation in enterocytes regulates cell survival and cell-cycle progression during colitis-associated tumorigenesis. Cancer Cell 15(2):91–102. Bonasio R, von Andrian UH (2006) Generation, migration and function of circulating dendritic cells. Curr Opin Immunol 18(4):503–511. Bromberg J, Wang TC (2009) Inflammation and cancer: IL-6 and STAT3 complete the link. Cancer Cell 15(2):79–80. Brown ER et al (2008) A clinical study assessing the tolerability and biological effects of infliximab, a TNF-alpha inhibitor, in patients with advanced cancer. Ann Oncol 19(7):1340–1346.

426

A. Yagui-Beltrán et al.

Burger JA, Kipps TJ (2006) CXCR4: a key receptor in the crosstalk between tumor cells and their microenvironment. Blood 107(5):1761–1767. Carbia-Nagashima A et  al (2007) RSUME, a small RWD-containing protein, enhances SUMO conjugation and stabilizes HIF-1alpha during hypoxia. Cell 131(2):309–323. Caro AA, Cederbaum AI (2006) Role of cytochrome P450 in phospholipase A2- and arachidonic acid-mediated cytotoxicity. Free Radic Biol Med 40(3):364–375. Chan AT, Ogino S, Fuchs CS (2007) Aspirin and the risk of colorectal cancer in relation to the expression of COX-2. N Engl J Med 356(21):2131–2142. Chang YC et al (2008) Epigenetic control of MHC class II expression in tumor-associated macrophages by decoy receptor 3. Blood 111(10):5054–5063. Chen CJ et al (2007) Identification of a key pathway required for the sterile inflammatory response triggered by dying cells. Nat Med 13(7):851–856. Condeelis J, Pollard JW (2006) Macrophages: obligate partners for tumor cell migration, invasion, and metastasis. Cell 124(2):263–266. Conroy H, Marshall NA, Mills KH (2008) TLR ligand suppression or enhancement of Treg cells? A double-edged sword in immunity to tumours. Oncogene 27(2):168–180. Coussens LM, Werb Z (2002) Inflammation and cancer. Nature 420(6917):860–867. Dalgleish AG, O’Byrne K (2006) Inflammation and cancer: the role of the immune response and angiogenesis. Cancer Treat Res 130:1–38. DeNardo DG et al (2009) CD4(+) T cells regulate pulmonary metastasis of mammary carcinomas by enhancing protumor properties of macrophages. Cancer Cell 16(2):91–102. de Visser KE, Eichten A, Coussens LM (2006) Paradoxical roles of the immune system during cancer development. Nat Rev Cancer 6(1):24–37. Dinarello CA (2006) The paradox of pro-inflammatory cytokines in cancer. Cancer Metastasis Rev 25(3):307–313. Dineen SP et al (2008) Vascular endothelial growth factor receptor 2 mediates macrophage infiltration into orthotopic pancreatic tumors in mice. Cancer Res 68(11):4340–4346. Drayton DL et  al (2006) Lymphoid organ development: from ontogeny to neogenesis. Nat Immunol 7(4):344–353. Duluc D et al (2007) Tumor-associated leukemia inhibitory factor and IL-6 skew monocyte differentiation into tumor-associated macrophage-like cells. Blood 110(13):4319–4330. Du R et al (2008) HIF1alpha induces the recruitment of bone marrow-derived vascular modulatory cells to regulate tumor angiogenesis and invasion. Cancer Cell 13(3):206–220. Dvorak HF (1986) Tumors: wounds that do not heal. Similarities between tumor stroma generation and wound healing. N Engl J Med 315(26):1650–1659. Elkabets M et al (2009) Host-derived interleukin-1alpha is important in determining the immunogenicity of 3-methylcholantrene tumor cells. J Immunol 182(8):4874–4881. Ferrantini M, Capone I, Belardelli F (2008) Dendritic cells and cytokines in immune rejection of cancer. Cytokine Growth Factor Rev 19(1):93–107. Fidler IJ (1995) Modulation of the organ microenvironment for treatment of cancer metastasis. J Natl Cancer Inst 87(21):1588–1592. Fischer C et al (2007) Anti-PlGF inhibits growth of VEGF(R)-inhibitor-resistant tumors without affecting healthy vessels. Cell 131(3):463–475. Fischetti F, Tedesco F (2006) Cross-talk between the complement system and endothelial cells in physiologic conditions and in vascular diseases. Autoimmunity 39(5):417–428. Flossmann E, Rothwell PM (2007) Effect of aspirin on long-term risk of colorectal cancer: consistent evidence from randomised and observational studies. Lancet 369(9573):1603–1613. Fukata M, Abreu MT (2008) Role of toll-like receptors in gastrointestinal malignancies. Oncogene 27(2):234–243. Fukata S et  al (2005) Levels of angiogenesis and expression of angiogenesis-related genes are prognostic for organ-specific metastasis of renal cell carcinoma. Cancer 103(5):931–942. Galli SJ, Grimbaldeston M, Tsai M (2008) Immunomodulatory mast cells: negative, as well as positive, regulators of immunity. Nat Rev Immunol 8(6):478–486.

20  The Pro-inflammatory Milieu and Its Role in Malignant Epithelial Initiation

427

Gallucci S, Matzinger P (2001) Danger signals: SOS to the immune system. Curr Opin Immunol 13(1):114–119. Garlanda C et al (2007) Increased susceptibility to colitis-associated cancer of mice lacking TIR8, an inhibitory member of the interleukin-1 receptor family. Cancer Res 67(13):6017–6021. Giavazzi R et al (1990) Interleukin 1-induced augmentation of experimental metastases from a human melanoma in nude mice. Cancer Res 50(15):4771–4775. Gonda TA, Tu S, Wang TC (2009) Chronic inflammation, the tumor microenvironment and carcinogenesis. Cell Cycle 8(13):2005–2013. Greten FR et al (2004) IKKbeta links inflammation and tumorigenesis in a mouse model of colitisassociated cancer. Cell 118(3):285–296. Grivennikov S et al (2009) IL-6 and Stat3 are required for survival of intestinal epithelial cells and development of colitis-associated cancer. Cancer Cell 15(2):103–113. Hagemann T et al (2008) “Re-educating” tumor-associated macrophages by targeting NF-kappaB. J Exp Med 205(6):1261–1268. Hanukoglu I (2006) Antioxidant protective mechanisms against reactive oxygen species (ROS) generated by mitochondrial P450 systems in steroidogenic cells. Drug Metab Rev 38(1–2):171–196. Harrison ML et al (2007) Tumor necrosis factor alpha as a new target for renal cell carcinoma: two sequential phase II trials of infliximab at standard and high dose. J Clin Oncol 25(29):4542–4549. Hold GL, El-Omar EM (2008) Genetic aspects of inflammation and cancer. Biochem J 410(2):225–235. Huang B et al (2006) Gr-1+CD115+ immature myeloid suppressor cells mediate the development of tumor-induced T regulatory cells and T-cell anergy in tumor-bearing host. Cancer Res 66(2):1123–1131. Hussain SP, Harris CC (2007) Inflammation and cancer: an ancient link with novel potentials. Int J Cancer 121(11):2373–2380. Hussain SP et al (2000) Increased p53 mutation load in noncancerous colon tissue from ulcerative colitis: a cancer-prone chronic inflammatory disease. Cancer Res 60(13):3333–3337. Ibrahim FB, Pang SJ, Melendez AJ, (2004) Anaphylatoxin signaling in human neutrophils. A key role for sphingosine kinase. J Biol Chem 279(43):44802–44811. Ii M et al (2006) Role of matrix metalloproteinase-7 (matrilysin) in human cancer invasion, apoptosis, growth, and angiogenesis. Exp Biol Med (Maywood) 231(1):20–27. Juliet PA et  al (2003) Concomitant production of nitric oxide and superoxide in human macrophages. Biochem Biophys Res Commun 310(2):367–370. Kabelitz D, Medzhitov R (2007) Innate immunity – cross-talk with adaptive immunity through pattern recognition receptors and cytokines. Curr Opin Immunol 19(1):1–3. Kaplan RN et al (2005) VEGFR1-positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche. Nature 438(7069):820–827. Karin M (2006) Nuclear factor-kappaB in cancer development and progression. Nature 441(7092):431–436. Karin M, Greten FR (2005) NF-kappaB: linking inflammation and immunity to cancer development and progression. Nat Rev Immunol 5(10):749–759. Khatami M et al (1984) Vernal conjunctivitis. Model studies in guinea pigs immunized topically with fluoresceinyl ovalbumin. Arch Ophthalmol 102(11):1683–1688. Koehne CH, Dubois RN (2004) COX-2 inhibition and colorectal cancer. Semin Oncol 31(2 Suppl 7):12–21. Kortylewski M et  al (2005) Inhibiting Stat3 signaling in the hematopoietic system elicits multicomponent antitumor immunity. Nat Med 11(12):1314–1321. Kortylewski M et al (2009) Regulation of the IL-23 and IL-12 balance by Stat3 signaling in the tumor microenvironment. Cancer Cell 15(2):114–123. Krelin Y et al (2007) Interleukin-1beta-driven inflammation promotes the development and invasiveness of chemical carcinogen-induced tumors. Cancer Res 67(3):1062–1071.

428

A. Yagui-Beltrán et al.

Kulbe H et al (2007) The inflammatory cytokine tumor necrosis factor-alpha generates an autocrine tumor-promoting network in epithelial ovarian cancer cells. Cancer Res 67(2):585–592. Kuper H, Adami HO, Trichopoulos D (2000) Infections as a major preventable cause of human cancer. J Intern Med 248(3):171–183. Kusmartsev S et al (2008) Oxidative stress regulates expression of VEGFR1 in myeloid cells: link to tumor-induced immune suppression in renal cell carcinoma. J Immunol 181(1):346–353. Langowski JL et al (2006) IL-23 promotes tumour incidence and growth. Nature 442(7101):461–465. Ledeen RW, Wu G (2006) GM1 ganglioside: another nuclear lipid that modulates nuclear calcium. GM1 potentiates the nuclear sodium-calcium exchanger. Can J Physiol Pharmacol 84(3–4):393–402. Lee H et  al (2009) Persistently activated Stat3 maintains constitutive NF-kappaB activity in tumors. Cancer Cell 15(4):283–293. Lewis CE, Pollard JW (2006) Distinct role of macrophages in different tumor microenvironments. Cancer Res 66(2):605–612. Lim SO et  al (2008) Epigenetic changes induced by reactive oxygen species in hepatocellular carcinoma: methylation of the E-cadherin promoter. Gastroenterology 135(6):2128–2140, 2140, e1–e8. Lin EY et  al (2001) Colony-stimulating factor 1 promotes progression of mammary tumors to malignancy. J Exp Med 193(6):727–740. Lin WW, Karin M (2007) A cytokine-mediated link between innate immunity, inflammation, and cancer. J Clin Invest 117(5):1175–1183. Lodoen MB, Lanier LL (2006) Natural killer cells as an initial defense against pathogens. Curr Opin Immunol 18(4):391–398. Luo JL et al (2007) Nuclear cytokine-activated IKKalpha controls prostate cancer metastasis by repressing Maspin. Nature 446(7136):690–694. Mager DL (2006) Bacteria and cancer: cause, coincidence or cure? A review. J Transl Med 4:14. Mager DL et al (2005) The salivary microbiota as a diagnostic indicator of oral cancer: a descriptive, non-randomized study of cancer-free and oral squamous cell carcinoma subjects. J Transl Med 3:27. Mantovani A et al (2002) Macrophage polarization: tumor-associated macrophages as a paradigm for polarized M2 mononuclear phagocytes. Trends Immunol 23(11):549–555. Mantovani A et al (2006) Role of tumor-associated macrophages in tumor progression and invasion. Cancer Metastasis Rev 25(3):315–322. Mantovani A et al (2008) Cancer-related inflammation. Nature 454(7203):436–444. Mantovani A et al (2008) Tumour immunity: effector response to tumour and role of the microenvironment. Lancet 371(9614):771–783. Marhaba R et al (2008) Opposing effects of fibrosarcoma cell-derived IL-1 alpha and IL-1 beta on immune response induction. Int J Cancer 123(1):134–145. McCaig C et al (2006) The role of matrix metalloproteinase-7 in redefining the gastric microenvironment in response to Helicobacter pylori. Gastroenterology 130(6):1754–1763. Monteseirin J et  al (2004) Human neutrophils synthesize IL-8 in an IgE-mediated activation. J Leukoc Biol 76(3):692–700. Moore RJ et al (1999) Mice deficient in tumor necrosis factor-alpha are resistant to skin carcinogenesis. Nat Med 5(7):828–831. Morandi F et  al (2007) Human neuroblastoma cells trigger an immunosuppressive program in monocytes by stimulating soluble HLA-G release. Cancer Res 67(13):6433–6441. Muller A et  al (2001) Involvement of chemokine receptors in breast cancer metastasis. Nature 410(6824):50–56. Murdoch C et al (2008) The role of myeloid cells in the promotion of tumour angiogenesis. Nat Rev Cancer 8(8):618–631. Nagaraj S, Gabrilovich DI (2008) Tumor escape mechanism governed by myeloid-derived suppressor cells. Cancer Res 68(8):2561–2563. Naugler WE, Karin M (2008) The wolf in sheep’s clothing: the role of interleukin-6 in immunity, inflammation and cancer. Trends Mol Med 14(3):109–119.

20  The Pro-inflammatory Milieu and Its Role in Malignant Epithelial Initiation

429

Naugler WE et  al (2007) Gender disparity in liver cancer due to sex differences in MyD88dependent IL-6 production. Science 317(5834):121–124. Nauts HC, McLaren JR (1990) Coley toxins – the first century. Adv Exp Med Biol 267:483–500. Nyakern M et al (2006) Synergistic induction of apoptosis in human leukemia T cells by the Akt inhibitor perifosine and etoposide through activation of intrinsic and Fas-mediated extrinsic cell death pathways. Mol Cancer Ther 5(6):1559–1570. Oguma K et al (2008) Activated macrophages promote Wnt signalling through tumour necrosis factor-alpha in gastric tumour cells. Embo J 27(12):1671–1681. Peerschke EI et al (2006) Blood platelets activate the classical pathway of human complement. J Thromb Haemost 4(9):2035–2042. Perez R et  al (2006) Blockade of arachidonic acid incorporation into phospholipids induces apoptosis in U937 promonocytic cells. J Lipid Res 47(3):484–491. Pikarsky E et al (2004) NF-kappaB functions as a tumour promoter in inflammation-associated cancer. Nature 431(7007):461–466. Pollard JW (2004) Tumour-educated macrophages promote tumour progression and metastasis. Nat Rev Cancer 4(1):71–78. Popivanova BK et al (2008) Blocking TNF-alpha in mice reduces colorectal carcinogenesis associated with chronic colitis. J Clin Invest 118(2):560–570. Quante M, Wang TC (2008) Inflammation and stem cells in gastrointestinal carcinogenesis. Physiology (Bethesda) 23:350–359. Rakoff-Nahoum S, Medzhitov R (2007) Regulation of spontaneous intestinal tumorigenesis through the adaptor protein MyD88. Science 317(5834):124–127. Raulet DH (2004) Interplay of natural killer cells and their receptors with the adaptive immune response. Nat Immunol 5(10):996–1002. Rius J et al (2008) NF-kappaB links innate immunity to the hypoxic response through transcriptional regulation of HIF-1alpha. Nature 453(7196):807–811. Robinson-Smith TM et  al (2007) Macrophages mediate inflammation-enhanced metastasis of ovarian tumors in mice. Cancer Res 67(12):5708–5716. Sakurai T et  al (2008) Hepatocyte necrosis induced by oxidative stress and IL-1 alpha release mediate carcinogen-induced compensatory proliferation and liver tumorigenesis. Cancer Cell 14(2):156–165. Sawanobori Y et  al (2008) Chemokine-mediated rapid turnover of myeloid-derived suppressor cells in tumor-bearing mice. Blood 111(12):5457–5466. Seandel M et al (2008) A catalytic role for proangiogenic marrow-derived cells in tumor neovascularization. Cancer Cell 13(3):181–183. Selsing E (2006) Ig class switching: targeting the recombinational mechanism. Curr Opin Immunol 18(3):249–254. Shahrzad S et  al (2007) Induction of DNA hypomethylation by tumor hypoxia. Epigenetics 2(2):119–125. Shime H et al (2008) Tumor-secreted lactic acid promotes IL-23/IL-17 proinflammatory pathway. J Immunol 180(11):7175–7183. Shojaei F et al (2007) Tumor refractoriness to anti-VEGF treatment is mediated by CD11b+Gr1+ myeloid cells. Nat Biotechnol 25(8):911–920. Sica A, Bronte V (2007) Altered macrophage differentiation and immune dysfunction in tumor development. J Clin Invest 117(5):1155–1166. Sica A et  al (2006) Tumour-associated macrophages are a distinct M2 polarised population promoting tumour progression: potential targets of anti-cancer therapy. Eur J Cancer 42(6):717–727. Sinha P et al (2007) Cross-talk between myeloid-derived suppressor cells and macrophages subverts tumor immunity toward a type 2 response. J Immunol 179(2):977–983. Snook CF, Jones JA, Hannun YA (2006) Sphingolipid-binding proteins. Biochim Biophys Acta 1761(8):927–946. Stearman RS et al (2008) A macrophage gene expression signature defines a field effect in the lung tumor microenvironment. Cancer Res 68(1):34–43.

430

A. Yagui-Beltrán et al.

Stenmark KR, Fagan KA, Frid MG (2006) Hypoxia-induced pulmonary vascular remodeling: cellular and molecular mechanisms. Circ Res 99(7):675–691. Swann JB et al (2008) Demonstration of inflammation-induced cancer and cancer immunoediting during primary tumorigenesis. Proc Natl Acad Sci U S A 105(2):652–656. Talmadge JE, Donkor M, Scholar E (2007) Inflammatory cell infiltration of tumors: Jekyll or Hyde. Cancer Metastasis Rev 26(3–4):373–400. Taylor CT (2008) Interdependent roles for hypoxia inducible factor and nuclear factor-kappaB in hypoxic inflammation. J Physiol 586(Pt 17):4055–4059. Tennant DA et al (2009) Metabolic transformation in cancer. Carcinogenesis 30(8):1269–1280. Tu S et al (2008) Overexpression of interleukin-1beta induces gastric inflammation and cancer and mobilizes myeloid-derived suppressor cells in mice. Cancer Cell 14(5):408–419. Valent P et al (2001) Variable expression of activation-linked surface antigens on human mast cells in health and disease. Immunol Rev 179:74–81. van ’t Veer LJ et al (2002) Gene expression profiling predicts clinical outcome of breast cancer. Nature 415(6871):530–536. Voronov E et al (2003) IL-1 is required for tumor invasiveness and angiogenesis. Proc Natl Acad Sci U S A 100(5):2645–2650. Wang RF, Miyahara Y, Wang HY (2008) Toll-like receptors and immune regulation: implications for cancer therapy. Oncogene 27(2):181–189. Whiteside TL (2008) The tumor microenvironment and its role in promoting tumor growth. Oncogene 27(45):5904–5912. Wilson NS, Villadangos JA (2004) Lymphoid organ dendritic cells: beyond the Langerhans cells paradigm. Immunol Cell Biol 82(1):91–98. Witz IP (2008) Tumor-microenvironment interactions: dangerous liaisons. Adv Cancer Res 100:203–229. Wu WS (2006) The signaling mechanism of ROS in tumor progression. Cancer Metastasis Rev 25(4):695–705. Wyckoff JB et al (2007) Direct visualization of macrophage-assisted tumor cell intravasation in mammary tumors. Cancer Res 67(6):2649–2656. Xiao H et al (2007) The Toll-interleukin-1 receptor member SIGIRR regulates colonic epithelial homeostasis, inflammation, and tumorigenesis. Immunity 26(4):461–475. Yang L et  al (2008) Abrogation of TGF beta signaling in mammary carcinomas recruits Gr-1+CD11b+ myeloid cells that promote metastasis. Cancer Cell 13(1):23–35. Yano K et al (1997) Production of macrophage inflammatory protein-1alpha by human mast cells: increased anti-IgE-dependent secretion after IgE-dependent enhancement of mast cell IgEbinding ability. Lab Invest 77(2):185–193. Yu H, Kortylewski M, Pardoll D (2007) Crosstalk between cancer and immune cells: role of STAT3 in the tumour microenvironment. Nat Rev Immunol 7(1):41–51. Zhang B (2008) Targeting the stroma by T cells to limit tumor growth. Cancer Res 68(23):9570–9573. Zhu P et al (2006) Macrophage/cancer cell interactions mediate hormone resistance by a nuclear receptor derepression pathway. Cell 124(3):615–629.

Chapter 21

Natural Killer Cells for Adoptive Immunotherapy Jonathan E. Benjamin and Sally Arai

Abstract  Natural killer (NK) cell-based immunotherapy in cancer is gaining prominence yet is still early in its development as we learn more about NK cell biology and the molecular mechanisms that drive NK cell expansion, direct trafficking, and determine target cell recognition. Clinical investigation of NK cell therapy has ranged from in vivo modulation of NK cell activity to ex vivo purification/expansion and adoptive transfer with modest results. Future trials will likely move toward refining the current techniques of NK expansion and overcoming the barriers of NK cell engraftment and host factors.

Introduction Natural Killer (NK) cells have attracted the attention of cancer immunologists for their ability to lyse a broad range of cancer cell types in vitro, to eradicate transplanted tumors, and to participate in the surveillance of spontaneously arising tumors in susceptible mouse strains. Their name reflects functional activity that was originally observed in vitro, namely, rapid, contact-dependent killing that did not require prior sensitization, was not genotypically restricted, and did not require the addition of antibodies (Herberman et al. 1975; Kiessling et al. 1975). Unlike T and B cells, NK cells do not rearrange germline encoded antigen receptor genes, and NK target recognition is not, strictly speaking, MHC restricted. Rather, NK cell function is governed by the integration of opposing signals from inhibitory and activating cell surface with diverse ligand-binding specificities. Moreover, cytotoxicity represents only one facet of NK cell function. Although previously considered to be a homogenous population, NK cells display phenotypic diversity that is matched by a broad repertoire of immune regulatory functions. It is likely that some of these functions will be relevant to the antitumor capabilities of NK cells. The J.E. Benjamin (*) Division of Blood and Marrow Transplantation, Stanford University Medical Center, Stanford, CA, USA e-mail: [email protected]

R.G. Bagley (ed.), The Tumor Microenvironment, Cancer Drug Discovery and Development, DOI 10.1007/978-1-4419-6615-5_21, © Springer Science+Business Media, LLC 2010

431

432

J.E. Benjamin and S. Arai

increasing appreciation of the molecular mechanisms that drive NK cell expansion, trafficking, and target cell recognition improve the likelihood that NK cell-based therapies may be successfully employed to treat cancer. However, NK cell-based therapies must be considered to be at an early stage. In this chapter, we attempt to review the current understanding of NK cell biology as pertains to recent approaches toward their therapeutic utilization.

Immunophenotype The study of NK cell biology has lagged behind that of B and T cells because no surface marker has yet been identified that is expressed specifically and exclusively by all NK cells. Moreover, mouse and human NK cells do not necessarily express orthologous markers, making comparisons of NK cell subsets between these species difficult. In humans, NK cells are characterized by the expression of the adhesion glycoprotein CD56 (NCAM) and the absence of CD3. The function of CD56 on NK cells remains unknown, but its surface density further subdivides NK cells into two populations with distinct functional properties (Baume et  al. 1992; Carson et  al. 1997; Lanier et al. 1991). CD56dim cells account for 90% of circulating peripheral blood NK cells, are potently cytotoxic, and express high amounts of intracellular perforin (Jacobs et al. 2001). This population expresses FcgRIIIa (CD16), a receptor for IgG implicated in antibody-dependent cellular cytotoxicity (ADCC). In contrast, CD56bright cells account for approximately 10% of circulating CD3-CD56+ cells but comprise the vast majority of NK cells in the lymph nodes (Fehniger et al. 2003). There, they are found primarily in the para-follicular T-cell rich regions, colocalizing with dendritic cells. The CD56bright cells lack CD16, do not express high levels of perforin, and are minimally cytotoxic (Jacobs et al. 2001). Instead, they function as modulators of the immune system by virtue of their abundant cytokine secretion (Cooper et  al. 2001). Depending on the specific activating signal, CD56bright cells have the capacity to secrete IFN-g, TNF-a, TNF-b, GM-CSF, IL-10, and IL-13. Recently, multiple groups have described a novel subset of CD56bright NK cells in tonsils and Peyer’s patches that secrete high levels of IL-22 (Cella et al. 2009; Sanos et  al. 2009; Satoh-Takayama et  al. 2008). It is likely that additional, functionally distinct populations of cells with NK-like phenotypes will be discovered.

Ontogeny NK cells are derived from Lin- CD34+ CD38+ common lymphoid progenitors (Galy et al. 1995; Miller et al. 1994), and bone marrow appears to be the primary, but not exclusive, site of NK cell development. Extramedullary NK cell development occurs in the lymph nodes and thymus (Ferlazzo et al. 2004; Freud et al. 2005). In the lymph node, a unique population of CD34dim NK cell precursors may develop into CD56bright NK cells under the influence of T cell-derived IL-2 and possibly

21  Natural Killer Cells for Adoptive Immunotherapy

433

myeloid-derived IL-15 (Freud et al. 2005). In the mouse, the thymus-derived NK cell population has functional properties similar to those of CD56bright cells, and is dependent upon IL-7R signaling (Vosshenrich et al. 2006). The developmental relationship between CD56dim and CD56bright NK cells remains controversial (Colucci et al. 2003). There is some evidence to suggest that the CD56bright subset represents a precursor population of the CD56dim (Chan et al. 2007; Romagnani et  al. 2007). When cultured with IL-2, CD56brightCD16- lymph node NK cells upregulate perforin, CD16, and demonstrate potent cytoxicity (Ferlazzo et  al. 2004). Conversely, in  vitro cytokine stimulation of CD56dim cells leads to upregulation of CD56 expression, although this does not necessarily ­signify a switch to an immunoregulatory phenotype (Takahashi et al. 2007).

Localization and Trafficking NK cells are broadly distributed in lymphoid and non-lymphoid organs, consistent with their role as early responders to infection. In the peripheral blood, NK cells account for 10–25% of lymphocytes, yielding a concentration of 100–600 ml-1, and are predominantly CD56dim. In the spleen, NK cells constitute 25% of lymphocytes, are predominantly CD56dim, and are concentrated in the red pulp adjacent to antigen-presenting macrophages and dendritic cells (Witte et al. 1990). In the lymph node, NK cells represent 1–5% of cells (Fehniger et al. 2003). They represent the predominant lymphocyte subset in the liver, with a variable distribution of CD56dim versus CD56bright (Burt et al. 2009). In the lung, NK cells make up 10% of all lymphocytes, representing a large reservoir that is second only to the spleen in absolute numbers (Culley 2009). Interestingly, a large number of CD56bright NK cells are also found in the placenta and are essential to the formation of the decidua and may play a role in the development of maternal tolerance to the semiallogeneic fetus (Croy et al. 1997). The anatomic distribution of NK cell subsets is determined in part by soluble chemotactic factors and membrane-bound adhesion molecules. NK cell maturation occurs with the acquisition of S1P5, a G-protein coupled receptor for the lysophospholipid sphingosine-1-phosphate. Genetic ablation of this receptor in mice results in the accumulation of NK cells in bone marrow and lymph nodes, and their absence in the blood, spleen, and lung (Jenne et al. 2009; Walzer et al. 2007). It is unknown whether S1P5 agonists might function to mobilize NK cells into peripheral blood. Migration of CD56bright cells to lymph nodes is mediated by CCR7, the receptor for the CCL19 and CCL21 chemokines. CD62L, or L-selectin, interacts with ligands on high endothelial venules and mediates the entry of CD56bright cells into lymph nodes (Frey et al. 1998). CD56dim cells express CX3CR1 and ChemR23, allowing their homing to inflamed tissues (Nishimura et  al. 2002; Parolini et  al. 2007). Notably, tumor cell expression of CX3CL1, a ligand for CX3CR1, has been associated with NK cell infiltration and better prognosis (Hyakudomi et al. 2008; Ohta et al. 2005).

434

J.E. Benjamin and S. Arai

Activation of NK Cells by Cytokines and Accessory Cells Peripheral blood NK effector function can be augmented by treatment with pharmacologic doses of cytokines and by co-culture with APC, leading to a distinction between resting and activated NK cells (Fernandez et  al. 1999; Trinchieri et al. 1984). IL-2 was the first cytokine demonstrated to enhance NK cytotoxicity and remains the most frequently used cytokine for NK cell activation in clinical trials (Caligiuri et  al. 1993; Miller et  al. 2005; Shi et  al. 2008). However, IL-2 preferentially expands the CD56 bright subset (Baume et al. 1992; Soiffer et al. 1996). IL-15 is singularly important for NK cell development and survival (Cooper et al. 2002; Ranson et al. 2003). Deficiencies in IL-15 or IL-15R α result in profoundly reduced numbers of NK cells (Kennedy et al. 2000; Lodolce et al. 1998). Furthermore, adoptive transfer of NK cells into either IL-15 or IL-15R α recipients results in the rapid disappearance of NK cells, suggestive of an important role in NK cell survival (Cooper et al. 2002; Ranson et al. 2003). IL-15 is a member of the four alpha-helix family of cytokines. Its receptor includes a beta subunit shared with the IL-2R and a g-chain subunit shared by IL-2, IL-4, IL-7, IL-9, and IL-21. A feature apparently unique to IL-15 signaling is the requisite expression in cis of IL-15 and the IL-15R a subunit. As currently formulated, the transpresentation model posits that IL-15 binds to IL-15R a on the cell surface where it can stimulate the IL-15 signaling pathway of neighboring cells that express the IL-2/15R b and common g subunits (Burkett et al. 2004; Dubois et al. 2002; Koka et al. 2003; Lodolce et al. 2001). Both plasmacytoid and myeloid dendritic cells are important physiologic inducers of NK cell cytotoxicity and cytokine production, albeit by different mechanisms (Ferlazzo et  al. 2002; Fernandez et  al. 1999; Gerosa et  al. 2002). Dendritic cell priming of NK cells is dependent on cell contact and IL-15Ra expression on the dendritic cell (Koka et al. 2004). Dendritic cell-derived type I IFN potentiates NK cell cytotoxicity. In contrast, NK cell IFN-g production is enhanced by direct cellular contact with IL-12 producing DC, usually in concert with additional cytokines such as IL-1, IL-2, IL-15, or IL-18 (Cooper et al. 2001).

Cytokine Secretion Interferon g is perhaps the most important of the cytokines secreted by NK cells, with higher levels produced by the CD56bright subset. Its effects are protean, including polarization of T cells to the Th1 phenotype, upregulation of MHC I molecules on APC, maturation and activation of dendritic cells, antiviral effects against herpes viruses, and antiproliferative effects on transformed cells (Boehm et al. 1997; Schroder et al. 2004). NK cells also secrete tumor necrosis factor-a and GM-CSF, which can promote dendritic cell maturation and phagocytosis (Gerosa et al. 2002; Pistoia et al. 1989).

21  Natural Killer Cells for Adoptive Immunotherapy

435

Cytotoxicity As has been observed with cytotoxic CD8 T cells, NK cells may kill targets via multiple mechanisms. Lytic granule exocytosis is the result of a multistep pathway resulting in cytoskeletal reorganization leading to localization of cytoplasmic perforin and granzyme-containing vesicles to the point of contact with the target cell (Bryceson et  al. 2005). Perforin creates pores in the target cell membrane, allowing granzyme to enter and cleave substrates such as caspases. Downstream of caspase activation is DNAse that degrades target cell DNA, resulting in apoptosis. Fusion of the lytic granule membrane with the plasma membrane enables the lysosome-associated membrane protein LAMP-1 (CD107a) to appear on the cell surface, providing a marker of NK cell activity that can be detected by flow cytometry (Alter et al. 2004). NK cells can also kill targets in a perforin-independent mechanism that involves TNF family of ligands and receptors. Both FasL and TNF-related apoptosisinducing ligand (TRAIL) are expressed on the NK membrane. They bind to Fas (CD95) and TRAIL-R1 or TRAIL-R2, leading to trimerization and the recruitment of Fas-Associated Death Domain (FADD) proteins in the target, which triggers the caspase cascade (Arase et  al. 1995; Oshimi et  al. 1996). Studies with blocking antibodies and genetically modified mice have provided ample evidence that these pathways are important for tumor cell recognition and clearance mediated by NK cells (Kashii et al. 1999; Smyth et al. 2001).

Activating and Inhibitory Signals NK cells employ a panoply of transmembrane receptors to assess potential targets for evidence of infection, stress, or transformation. Of particular interest to the field of cellular therapy are the receptors, both activating and inhibitory, that recognize Major Histocompatibility Class (MHC) I molecules. These include the killer immunoglobulin-like receptors (KIR) in humans, Ly49 receptors in mouse, and CD94/NKG2A in both species. KIR expression is stochastic, generating an NK cell compartment with individual members expressing one or more inhibitory receptors. The Class I specificity of an NK cell is determined by the repertoire of receptors it expresses. Failure to engage the inhibitory receptors, known as missing-self recognition (Karre et al. 1986), is thought to contribute to the graft-versus-malignancy effect observed in Human Leukocyte Antigen (HLA) haploidentical hematopoietic cell transplants (Ruggeri et al. 2002). KIR expression is noted predominantly on the CD56dim subset and thus appears to be more relevant to the cytotoxic pathways. In humans, the KIR gene cluster is located on chromosome 19 and thus segregates independently of HLA genes on chromosome 6. KIR genotype, rather than HLA, is the major determinant of KIR expression pattern (Shilling et  al. 2003). Fourteen KIR genes and two pseudogenes have been identified. KIR genes are

436

J.E. Benjamin and S. Arai

highly polymorphic, and growing evidence suggests that allelic diversity may correlate with functional differences in both ligand recognition and signaling strength (Bari et al. 2009; Hsu et al. 2002; Single et al. 2007). KIR gene nomenclature is determined by the number of extracellular immunoglobulin domains and the length of the cytoplasmic tails. For instance, a KIR gene with the 2D extracellular domain may have a long (2DL2) or short (2DS2) cytoplasmic tail. Long cytoplasmic tails contain Immunoreceptor Tyrosine-Based Inhibitory Motifs (ITIMs) that facilitate the docking of phosphatase molecules. In KIR-S molecules, the short cytoplasmic tails generally contain transmembrane charged residues that enable pairing to the DAP12 adapter molecule. DAP12 contains an Immunoreceptor Tyrosine-Based Activating Motif (ITAM). Crosslinking of these receptors results in tyrosine phosphorylation of the ITAM and recruitment of the Syk and Zap70 tyrosine kinases. The extracellular domains of activating KIR share sequence similarity with the corresponding inhibitory receptors and may share HLA ligand binding specificities, although perhaps with reduced affinity. Four KIR specificities have been determined. Based on motifs in the carboxy terminal of the Class I alpha helix, HLA-C molecules can be divided into two groups by an allelic dimorphism at positions 77 and 80. Group 1 has a serine at position 77 and asparagine at position 80 (HLA-Cw3 and related alleles), whereas group 2 has asparagine and lysine (HLA-Cw4). KIR2DL2/2DL3 binds C1 and KIR2DL1 binds C2. HLA-B allotypes with the Bw4 motif bind to KIR3DL1. KIR3DL2 recognizes HLA-A3 and A11. In the context of allogeneic transplantation, KIR ligand mismatching in the GVH direction refers to a situation in which the recipient lacks one or more KIR ligands expressed by the donor.

Activating Receptors Absence of self-MHC antigens is insufficient to trigger an NK attack. Rather, NK cells require stimulation via cell surface receptors that pair with ITAM-containing signaling molecules. A number of activating NK cell receptors have now been characterized, and most, if not all NK cells, express many of these receptors. Consequently, unlike T and B cells, NK cells do not have clonally distributed antigen specificity. With the exception of CD16, crosslinking of a single activating receptor does not trigger NK effector function. Rather, cytotoxicity or cytokine secretion requires crosslinking of pairs or multiple receptors (Bryceson et al. 2009). The repertoire of activating NK receptors is large and has been reviewed elsewhere (Bottino et al. 2005; Lanier 2009). Certain receptors merit discussion for the purpose of this review. CD16, the intermediate-affinity receptor for IgG, mediates ADCC and is essential to the therapeutic benefit of depleting monoclonal antibodies such as rituximab or trastuzumab. CD16 levels on NK cells can be augmented by pharmacologic doses of IL-2, and multiple clinical trials employing monoclonal antibodies with IL-2 have been published (Eisenbeis et al. 2004; Fleming et al. 2002;

21  Natural Killer Cells for Adoptive Immunotherapy

437

Repka et  al. 2003). Another well-studied receptor is NKG2D, a C-type lectin expressed on all NK cells as well as certain T cell subsets. Ligands for NKG2D, including the human MIC-A/B, and ULBP proteins, are structurally similar to MHC-I molecules (Bauer et  al. 1999; Cosman et  al. 2001). As described below, target cells may be sensitized to NK cell attack by the upregulation of NKG2D ligands by drugs such as 5-azacytidine, bortezomib, and arsenic trioxide. Finally, the natural cytotoxicity receptors (NCR) NKp46, NKp30, and NKp44 have been implicated in the recognition and targeting of tumor cells, but information regarding their ligand specificities remains elusive.

Licensing The mechanism of self-tolerance by NK cells is a fascinating question, with significant implications for clinical trial design. It is known that subsets of NK cells express only inhibitory KIRs for absent Class I antigens. Rather than being autoreactive, as would be predicted by the missing-self model, these cells are hyporesponsive to activating stimuli (Wu and Raulet 1997). The licensing theory of NK cell maturation suggests that engagement of an inhibitory KIR by the cognate Class I molecule is necessary to acquire effector function (Anfossi et al. 2006; Kim et al. 2005). The requirement for NK licensing would appear to preclude the efficacy of KIR-ligand mismatched NK cells. However, it seems that licensing can be bypassed by inflammatory stimuli, such as might be anticipated after the administration of chemotherapy (Cooley et al. 2007; Kim et al. 2005).

NK Cells and Anti-tumor Response Many studies have demonstrated NK cell killing of a variety of human and mouse cancer cell lines in  vitro, as well as NK-dependent eradication of transplanted tumor cell lines in vivo (Carlsten et al. 2009). To date, NKG2D is the best-studied activating NK cell receptor linked to tumor recognition and lysis. In the mouse, blockade, downregulation, or targeted deletion of NKG2D results in an increased incidence of experimentally induced sarcomas or spontaneous lymphoma and prostate cancer in susceptible mouse strains (Guerra et al. 2008; Oppenheim et al. 2005; Smyth et al. 2005). NKG2D ligand (NKG2D-L) expression is minimal on healthy cells but upregulated in cancerous cells, likely by multiple mechanisms including those involving the DNA-damage repair pathway (Gasser et al. 2005). However, chemotherapies with agents such as bortezomib, 5-azacytidine, and arsenic trioxide, i.e., whose primary effect is not DNA damage, also enhance NKG2D-L expression (Diermayr et  al. 2008; Poggi et  al. 2009; Soriani et  al. 2009). Clinical trials employing these agents as a component of NK cell therapy are underway.

438

J.E. Benjamin and S. Arai

Tumor Infiltrating Lymphocytes Infiltration of NK cells into numerous different tumor types has correlated with better outcome (Coca et  al. 1997; Ishigami et  al. 2000; Takanami et  al. 2001; Villegas et al. 2002). Preclinical studies with in vitro activated NK cells infused into tumor-bearing mice have suggested that the infused cells can accumulate preferentially into the tumor tissue (Daldrup-Link et al. 2005; Meier et al. 2008). The ability of NK cells to penetrate into tumor tissue likely depends on the location of the tumor (e.g., epithelial tissue vs. lymph node) as well as tumor-specific factors such as extracellular matrix components, tumor-derived chemokines, and adhesion molecules.

NK Cells and Haploidentical Transplantation Landmark studies by the Perugia group have generated greater interest in the therapeutic use of NK cells. Allogeneic HCT is a potentially curative therapy for hematolymphoid neoplasms including myeloid and lymphoid leukemias, lymphoproliferative disorders, and marrow failure syndromes including myelodsyplasia and aplastic anemia. The benefits of allogeneic HCT are derived from both the conditioning regimen of chemoradiotherapy and the donor alloimmune response, primarily mediated by donor T lymphocytes recognizing minor histocompatibility antigens. The curative potential of allogeneic HCT is offset by the toxicity of the therapy, the most important being graft-versus-host disease (GVHD), also mediated by donor T cells. GVHD risk increases with the degree of HLA mismatching between donor and recipient. HLA identical siblings are preferred donors. If a sibling donor is not available, then an HLA-matched unrelated donor is sought via registries of healthy volunteers. Many patients, particularly those from minority populations, lack a suitably matched donor. Haploidentical (i.e., sharing only one HLA haplotype) related donors, including parents and offspring, are available for most patients. Transplantation with T-cell containing grafts from haploidentical donors carries an unacceptable risk of GVHD. The Perugia group has pioneered the use of haploidentical transplantation for patients with high-risk leukemia lacking an HLA-matched donor. In order to avoid GVHD, the donor grafts are T cell depleted, and “megadose” CD34 infusions are required to allow for engraftment (Aversa et  al. 1994; Aversa et  al. 1998). Superior outcomes were observed when KIR ligand mismatching was predicted in the GVH direction, but only in the setting of myeloid and not lymphoid malignancies, possibly reflecting the lack of NKG2D-L expression on lymphoid leukemias (Romanski et al. 2005; Ruggeri et al. 2002). KIR ligand incompatibility was also protective against GVHD and graft rejection. Protection against GVHD has been explored in mouse models, where it was found that Ly49 ligand mismatching

21  Natural Killer Cells for Adoptive Immunotherapy

439

in the GVH direction resulted in the depletion of recipient antigen-presenting cells (Lundqvist et al. 2007; Ruggeri et al. 2002). As has been shown after myeloablative conditioning, NK cells are the first lymphocyte population to repopulate the peripheral blood. In the Perugia studies and in others, most NK cells isolated from the recipient after the transplant lacked KIR but expressed CD94/NKG2A, more consistent with the CD56bright immunoregulatory phenotype, although this may represent a developmental intermediate to the CD56dim cytotoxic population (Nguyen et al. 2005; Ruggeri et al. 2007; Schulze et al. 2008; Vago et al. 2008). A mature donor-type KIR repertoire emerges within 3 months to 3 years (Shilling et al. 2003). The potential benefits of NK cell alloreactivity have been explored in cohorts of patients receiving mismatched unrelated donor transplants, often with discordant conclusions. In an early review of the experience at the University of Minnesota, Davies et  al. found no benefit of KIR ligand incompatibilities (Davies et  al. 2002). The authors noted that a variety of transplant protocols were employed at that center over the 10 years of accrual, most differing from those used in Perugia by virtue of the presence of T cells in the graft and by the use of post-grafting immunosuppression, both of which may skew developing NK cells to the CD56bright lineage (Cooley et al. 2005; Lowe et al. 2003; Wang et al. 2007). In contrast, Beelen et  al. reviewed transplant outcomes at Essen and found a marked reduction of AML relapse in patients receiving T-cell replete, KIR ligand incompatible grafts (Beelen et  al. 2005). This benefit was offset by an increased risk of chronic GVHD; therefore, no survival advantage was observed with KIR ligand incompatibility. Not surprisingly, multi-institutional studies have not clarified matters. Giebel et  al. analyzed the outcomes of 130 patients receiving URD transplants at three European centers (Giebel et al. 2003). A total of 61 patients received HLA-matched transplants, and of the remaining patients who received mismatched transplants, 20 had KIR ligand mismatching in the GVH direction. Approximately two-thirds of the patients had myeloid malignancies. Among patients without KIR ligand mismatches, survival at 4.5 years was not significantly different regardless of whether the patients were HLA matched or mismatched (48% vs. 58%). In contrast, patients receiving KIR-ligand mismatched transplants experienced 87% survival. The incidence of Grade III–IV acute GVHD was 0 and 15% in patients receiving KIRligand mismatched and matched transplants, respectively. The authors, in seeking to reconcile their results with those of Perugia and Minnesota, noted that although T cell-replete grafts were used, all patients received pre-transplant ATG, which, because of its persistence post-transplant, may have negatively influenced donor T cell recovery. Farag et al. reviewed international registry data of 1,571 unrelated bone marrow donor transplants for myeloid malignancies (Farag et al. 2006). Two-thirds of the patients were HLA matched. In the HLA-mismatched donor recipient pairs, 30% had KIR ligand incompatibility in the GVH direction, 24% in the

440

J.E. Benjamin and S. Arai

HVG direction, and the remainder had no predicted NK alloreactivity. T cell depletion was employed in approximately 20% of transplants in all groups. Overall survival and TRM were superior in patients receiving HLA-matched transplants. Among those receiving mismatched transplants, there were no significant differences in outcomes according to the presence or vector of KIR ligand mismatching.

Umbilical Cord Transplantation As with haploidentical transplantation, umbilical cord blood (UCB) transplants are frequently characterized by a high degree of HLA mismatching, allowing an evaluation of the role of KIR ligand mismatching. In a review of 218 transplants, Willemze et  al. found that the recipients of KIR-ligand incompatible grafts had improved overall survival due to decreased relapse when compared with those without these incompatibilities (Willemze et al. 2009). As in the Perugia studies, benefits of KIR ligand incompatibility were most evident among patients with AML, although UCB recipients with ALL also had a trend toward improved leukemia-free survival. Brunstein et al., however, failed to observe any benefit of KIR-L mismatch after myeloablative or non-myeloablative conditioning regimens (Brunstein et al. 2009).

Non-myeloablative Transplantation More than 40 years ago, the demonstration of hybrid (a × b F1) resistance to parental (a or b) hematopoietic grafts contradicted the prevailing understanding of allograft rejection (Cudkowicz and Stimpfling 1964). In skin and solid organ transplants, F1 progeny tolerates parental grafts. In the subsequent decades, recipient NK cells were identified as the mediator of hybrid resistance. In the hybrid resistance model, the conditioning regimen employed is total body irradiation, which may allow the persistence of significant numbers of recipient NK cells. Likewise, non-myeloablative conditioning is marked by a period of mixed chimerism in which recipient NK cells may exert anti-donor effects. Sobecks et  al. analyzed the impact of recipient anti-donor NK-cell alloreactivity in 31 patients conditioned with fludarabine and 2Gy TBI (Sobecks et al. 2008). Donor–recipient pairs were stratified by the number of recipient KIR genes for which the donor lacked the appropriate HLA Class I ligand. Those with the greatest number of unmatched KIR-L and KIR exhibited the highest incidence of graft rejection and the lowest incidence of complete donor chimerism.

21  Natural Killer Cells for Adoptive Immunotherapy

441

Killer Cell Immunotherapy Rosenberg and colleagues pioneered the field of adoptive transfer of killer cells with a series of clinical trials performed in the 1980s (Rosenberg et al. 1985). Autologous peripheral blood mononuclear cells (PBMC) from patients with advanced renal cell carcinoma or melanoma were treated in vitro with high-dose IL-2, yielding a population of lymphokine activated killer cells (LAK) that demonstrated broad cytotoxicity toward a variety of tumor cell lines. Following reinfusion of LAK cells, patients were treated with high-dose IL-2 with the hope of maintaining the cytotoxic activity of the cells. Objective responses were observed in approximately 20% of patients, although subsequent studies found similar response rates with high-dose IL-2 alone (Yang et al. 2003). From the current perspective of NK cell biology, the LAK method has inherent shortcomings. First, the LAK product, derived from unsorted mononuclear cells, consisted predominantly of T cells. Second was the use of autologous cells, which with current knowledge of KIR biology would be predicted to be inhibited by tumors that maintained Class I expression.

NK Cell Adoptive Therapy Following the observations in haploidentical transplantation, along with abundant preclinical data, multiple groups have attempted to infuse purified NK cells in patients with refractory malignancies or poor graft function after transplant. Pure NK cell products can be obtained from leukapheresis specimens by immunomagnetic selection using commercially available products under GMP conditions. The CliniMACS system can achieve purities of 40–99%, a 3–5 log reduction in T cell content, and yields of 20–80% (Iyengar et al. 2003; Koehl et al. 2005; McKenna et  al. 2007; Meyer-Monard et  al. 2009). It is expected that NK isolation from peripheral blood yields a predominantly CD56dim population. Haploidentical NK Cells A pilot study of NK donor lymphocyte infusion (DLI) was attempted in five patients as treatment for poor chimerism function or early relapse after haploidentical HCT. PBMC were collected from the original donor by 10-L leukapheresis, and then purified by immunomagnetic T cell depletion followed by positive selection with anti-CD56. NK cells were