Tumour-Associated Macrophages

Tumour-Associated Macrophages

Toby Lawrence • Thorsten Hagemann Editors

Tumour-Associated Macrophages

Editors Toby Lawrence Centre d’Immunologie Marseille Luminy Inserm-CNRS-Universitie de Mediteranee Inflammation Biology Group Parc Scientifique de Luminy, Case 906 Marseille Cedex 9, France [email protected]

Thorsten Hagemann Centre for Cancer and Inflammation Barts Cancer Institute, Barts and the London School of Medicine and Dentistry Queen Mary University of London London, EC1M 6BQ, UK [email protected]

ISBN 978-1-4614-0661-7 e-ISBN 978-1-4614-0662-4 DOI 10.1007/978-1-4614-0662-4 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2011937444 © Springer Science+Business Media, LLC 2012 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 software, 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

Macrophages are tissue resident phagocytes derived from blood monocytes; they have diverse functions in development and immunity and display enormous phenotypic heterogeneity. Macrophages in different tissues have specialized and specific functions that support organ development and physiology, for example, kupffer cells in the liver filter debris from the blood and aid liver regeneration after injury, Langerhans cells in the skin are important immune sentinel cells and mediate immune surveillance, osteoclasts mediate bone morphogenesis, and microglia in the brain support the development and maintenance of neuronal networks. In response to inflammation or injury, monocytes are recruited into tissue and differentiate locally into macrophages and depending on the nature of the insult or injury these macrophages may acquire distinct phenotypes. Tumours are frequently infiltrated by large number of macrophages and in most cases this is linked with tumour progression and poor prognosis. Macrophage polarization is a poorly defined phenomenon; the mediators and mechanisms that maintain the phenotype of distinct macrophage subsets in both physiology and disease remain to be described. Based primarily on in vitro studies, two particular macrophage phenotypes have been described: “classically” activated or M1 macrophages are characterized by the production of pro-inflammatory cytokines and increased microbicidal or even tumouricidal activity. The second, “alternatively” activated or M2 macrophages, in contrast produce anti-inflammatory cytokines and are linked with angiogenesis, tissue repair, and remodeling. These polarized phenotypes have been described based on in vitro stimulation of macrophages with either interferon (IFN) g, in the case of M1 macrophages, or interleukin (IL)-4 for M2 macrophages. It still is not clear what correlates these populations have in vivo and their physiological relevance remains ambiguous. While these classifications have been useful in that they allow the functional grouping of different macrophage phenotypes, M1 macrophages being proinflammatory cells and M2 macrophages linked with trophic functions and wound healing, there are undoubtedly several intermediates between these polarized phenotypes. However, this classification is too restrictive and it is clear that the functional diversity macrophages in vivo may not be associated with these distinct phenotypic subsets. In fact, the question remains in the context of inflammation and v

vi

Preface

tumours if “the macrophage” merely displays functional plasticity within tissue responding to environmental cues, or distinct stable subsets of macrophages exist with specialized functions. This issue is particularly pertinent in the case of TAM; these cells often display an M2-like phenotype associated with trophic functions promoting tumour angiogenesis, invasion, and metastasis. However, TAM also often produce pro-inflammatory cytokines and have been associated with the promotion of inflammation-associated cancer. This volume provides an overview of current research on the form and function of TAM, highlighting both the mechanistic roles they play in carcinogenesis and tumour progression as well as the molecular mechanisms that control their phenotype and function. Marseille, France London, UK

Toby Lawrence Thorsten Hagemann

Contents

Part I

Form and Function

1

Macrophage Phenotype in Tumours ..................................................... Hsi-Hsien Lin and Siamon Gordon

2

Role of Tumour-Associated Macrophages in the Regulation of Angiogenesis ........................................................................................ Russell Hughes, Hsin-Yu Fang, Munitta Muthana, and Claire E. Lewis

3

The Role of Tumour-Associated Macrophages in Malignant Invasion ............................................................................. Claudia Binder

3

17

31

4

Tumour-Induced Immune Suppression by Myeloid Cells................... Serena Zilio, Giacomo Desantis, Mariacristina Chioda, and Vincenzo Bronte

49

5

TAM: A Moving Clinical Target ............................................................ Simon Hallam and Thorsten Hagemann

63

Part II

Mechanisms of Action

6

Arginine Metabolism and Tumour-Associated Macrophages............. Melissa Phillips and Peter W. Szlosarek

7

Indoleamine 2,3-Dioxygenase Amino Acid Metabolism and Tumour-Associated Macrophages: Regulation in Cancer-Associated Inflammation and Immune Escape .................. George C. Prendergast, Richard Metz, Mee Young Chang, Courtney Smith, Alexander J. Muller, and Suzanne Ostrand-Rosenberg

77

91

vii

viii

Contents

8

Vascular Endothelial Growth Factor and Tumour-Associated Macrophages............................................................................................ 105 Christian Stockmann and Randall S. Johnson

Part III

Molecular Regulation

9

TLR Signaling and Tumour-Associated Macrophages........................ 119 Oscar R. Colegio and Ruslan Medzhitov

10

SHIP and Tumour-Associated Macrophages ....................................... 135 Victor W. Ho, Melisa J. Hamilton, Etsushi Kuroda, Jens Ruschmann, Frann Antignano, Vivian Lam, and Gerald Krystal

11

NF-KappaB-Mediated Regulation of Tumour-Associated Macrophages: Mechanisms and Significance ....................................... 153 Antonio Sica and Alberto Mantovani

12

Role of Hypoxia-Inducible Transcription Factors in TAM Function ..................................................................................... 167 Nadine Rohwer and Thorsten Cramer

Index ................................................................................................................. 183

Contributors

Frann Antignano The Terry Fox Laboratory, B.C. Cancer Agency, Vancouver, BC V5Z 1L3, Canada Claudia Binder Department of Hematology/Oncology, University of Göttingen, Göttingen 37099, Germany Vincenzo Bronte Verona University Hospital and Department of Pathology, Immunology Section, Piazzale L.A. Scuro 10, 37134 Verona, Italy Mee Young Chang Lankenau Institute for Medical Research, Wynnewood, PA, USA Mariacristina Chioda Istituto Oncologico Veneto (IOV), IRCCS, Via Gattamelata 64, 35128 Padova, Italy Oscar R. Colegio Department of Immunobiology, Howard Hughes Medical Institute, Yale University, New Haven, CT, USA Thorsten Cramer Medizinische Klinik mit Schwerpunkt Hepatologie und Gastroenterologie and Molekulares Krebsforschungszentrum, Charité – Universitätsmedizin Berlin, Campus Virchow-Klinikum, Augustenburger Platz 1, 13353 Berlin, Germany Giacomo Desantis Department of Oncology and Surgical Sciences, University of Padova, Via Gattamelata 64, 35128 Padova, Italy Hsin-Yu Fang Academic Unit of Inflammation & Tumour Targeting, University of Sheffield Medical School, Beech Hill Road, Sheffield S10 2RX, UK Siamon Gordon Sir William Dunn School of Pathology, University of Oxford, South Parks Road, Oxford OX1 3RE, UK

ix

x

Contributors

Thorsten Hagemann Centre for Cancer and Inflammation, Barts Cancer Institute, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, London EC1M 6BQ, UK Simon Hallam Centre for Cancer and Inflammation, Barts Cancer Institute, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, London EC1M 6BQ, UK Melisa J. Hamilton The Terry Fox Laboratory, B.C. Cancer Agency, Vancouver, BC V5Z 1L3, Canada Victor W. Ho The Terry Fox Laboratory, B.C. Cancer Agency, Vancouver, BC V5Z 1L3, Canada Russell Hughes Academic Unit of Inflammation & Tumour Targeting, University of Sheffield Medical School, Beech Hill Road, Sheffield S10 2RX, UK Randall S. Johnson Molecular Biology Section, Division of Biology, University of California, San Diego, CA 92093, USA Gerald Krystal The Terry Fox Laboratory, B.C. Cancer Agency, Vancouver, BC V5Z 1L3, Canada Etsushi Kuroda The Terry Fox Laboratory, B.C. Cancer Agency, Vancouver, BC V5Z 1L3, Canada Vivian Lam The Terry Fox Laboratory, B.C. Cancer Agency, Vancouver, BC V5Z 1L3, Canada Claire E. Lewis Academic Unit of Inflammation & Tumour Targeting, University of Sheffield Medical School, Beech Hill Road, Sheffield S10 2RX, UK Hsi-Hsien Lin Department of Microbiology and Immunology, College of Medicine, Chang Gung University, 259 Wen-Hwa 1st Road, Kwei-Shan, Tao-Yuan, Taiwan Alberto Mantovani Istituto Clinico Humanitas IRCCS, via Manzoni 56, 20089 Rozzano, Italy Department of Translational Medicine, University of Milan, Milan, Italy Ruslan Medzhitov Department of Immunobiology, Howard Hughes Medical Institute, Yale University, New Haven, CT, USA Richard Metz New Link Genetics Corporation, Ames, IA, USA Alexander J. Muller Lankenau Institute for Medical Research, Wynnewood, PA, USA Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA, USA Munitta Muthana Academic Unit of Inflammation & Tumour Targeting, University of Sheffield Medical School, Beech Hill Road, Sheffield S10 2RX, UK

Contributors

xi

Suzanne Ostrand-Rosenberg Department of Biological Sciences, University of Maryland at Baltimore, Baltimore, MD, USA Melissa Phillips Centre for Molecular Oncology and Imaging, Institute of Cancer, Barts and The London School of Medicine, Queen Mary College, John Vane Science Centre, Charterhouse Square, London EC1M 6BQ, UK George C. Prendergast Lankenau Institute for Medical Research, Wynnewood, PA, USA Department of Pathology, Anatomy & Cell Biology, Jefferson Medical School, Philadelphia, PA, USA Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA, USA Nadine Rohwer Medizinische Klinik mit Schwerpunkt Hepatologie und Gastroenterologie and Molekulares Krebsforschungszentrum, Charité – Universitätsmedizin Berlin, Campus Virchow-Klinikum, Augustenburger Platz 1, 13353 Berlin, Germany Jens Ruschmann The Terry Fox Laboratory, B.C. Cancer Agency, Vancouver, BC V5Z 1L3, Canada Antonio Sica Istituto Clinico Humanitas IRCCS, via Manzoni 56, 20089 Rozzano, Italy DiSCAFF, University of Piemonte Orientale A. Avogadro, 28100 Novara, Italy Courtney Smith Lankenau Institute for Medical Research, Wynnewood, PA, USA Christian Stockmann Molecular Biology Section, Division of Biology, University of California, San Diego, CA 92093, USA Institut für Physiologie, University of Duisburg-Essen, Duisburg-Essen, Germany Peter W. Szlosarek Centre for Molecular Oncology and Imaging, Institute of Cancer, Barts and The London School of Medicine, Queen Mary College, John Vane Science Centre, Charterhouse Square, London EC1M 6BQ, UK Serena Zilio Department of Oncology and Surgical Sciences, University of Padova, Via Gattamelata 64, 35128 Padova, Italy Istituto Oncologico Veneto (IOV), IRCCS, Via Gattamelata 64, 35128 Padova, Italy

Part I

Form and Function

Chapter 1

Macrophage Phenotype in Tumours Hsi-Hsien Lin and Siamon Gordon

Introduction Monocytes and macrophages are a prominent component of the host response to, and manipulation by, tumour cells (Gordon and Martinez 2010; Mantovani et al. 2008). Together with other myeloid and lymphoid cells, they influence tumour development, both positively and negatively. Although the factors that determine outcome of the host–tumour relationship are not well understood, many tumours recruit immature myelomonocytic cells, block their differentiation, subvert their cytotoxicity, suppress lymphoid effector cells, and induce peripheral tolerance. In addition, they mimic and utilise macrophage functions to enhance growth, produce a stroma and promote angiogenesis, local invasion of their micro-environment and metastasis (Qian and Pollard 2010). In particular, the uptake of apoptotic tumour cells can suppress anti-tumour inflammatory responses by TGF-beta and prostaglandins. The macrophage growth factor CSF-1 stimulates macrophage recruitment and modulates its phenotype, limiting the activation of cytotoxic effector functions; Interleukin-4 and -13, acting through common and specific receptors, induce a trophic, alternative M2 activation phenotype, distinct from cytotoxic M1, classically activated (Interferon-gamma-dependent) macrophages (Reviewed by Gordon and Martinez 2010). Interleukin-10 is a potent deactivator of macrophage inflammatory properties, whereas TGF-beta, another deactivator, promotes fibrosis and vascular remodelling.

H.-H. Lin Department of Microbiology and Immunology, College of Medicine, Chang Gung University, 259 Wen-Hwa 1st Road, Kwei-Shan, Tao-Yuan, Taiwan S. Gordon (*) Sir William Dunn School of Pathology, University of Oxford, South Parks Road, Oxford OX1 3RE, UK e-mail: [email protected] T. Lawrence and T. Hagemann (eds.), Tumour-Associated Macrophages, DOI 10.1007/978-1-4614-0662-4_1, © Springer Science+Business Media, LLC 2012

3

4

H.-H. Lin and S. Gordon

A wide range of chemokines such as MCP-1, often produced by tumour cells, attract mononuclear and myeloid cells. TNF-alpha has also been implicated in tumourigenesis (Mantovani et al. 2008). Monocyte-macrophages express a wide range of plasma membrane receptors which govern their response to chemokines, cytokines, growth factors and other tumour- and host-derived ligands (Taylor et al. 2005). Other membrane molecules regulate cellular responses to diverse agonists, inhibiting or enhancing macrophage effector mechanisms. These molecules provide useful markers for the presence, characterisation and possible functions of tumour-associated monocyte/macrophages, and targets for therapeutic intervention. In this review, we present a range of possible molecular markers for in situ characterisation, with special reference to the EGF-TM7 family of myeloid G protein-coupled receptors (GPCRs) with large extracellular domains. Their potential is reviewed in the context of macrophage heterogeneity and plasticity (Auffray et al. 2009; Gordon and Taylor 2005) and the experimental analysis of macrophage phenotype in tumours.

Macrophage Heterogeneity in Tumours Some of the earliest studies on the presence and possible role of macrophages in tumours were undertaken by Evans and Alexander, Mantovani, Pollard and their collaborators (Mantovani et al. 2008; Qian and Pollard 2010). The topic received renewed impetus in recent years with the work of Bronte (Peranzoni et al. 2010) and Gabrilovitch (Gabrilovich and Nagaraj 2009) and their groups. Important contributions came from Balkwill (Mantovani et al. 2008), Lewis (Coffelt et al. 2009), Karin (Grivennikov et al. 2010) and Coussens (Coussens and Werb 2002), Rosenberg (Domachowske et al. 2000) and Joyce (Joyce and Pollard 2009). A great deal of confusion has resulted from myeloid cell heterogeneity and terms such as TAMs (tumour-associated macrophages) and MDSC (myeloid-derived suppessor cells) are currently in wide use. The former embraces cells with macrophagerestricted markers such as F4/80 and alternative activation markers such as Arginase-1 (Gordon and Martinez 2010); the latter term includes cells with immature monocytic phenotype (Gr-1 low) and granulocyte characteristics (Gr-1 high). Mononuclear phagocyte heterogeneity associated with stages of differentiation and activation status gives rise to considerable plasticity within and among cell populations. Studies by Geissmann (Geissmann et al. 2010) and Jung (Varol et al. 2009) have utilised the fractalkine receptor, in combination with other chemokine receptors, to define precursors of tissue macrophages during development, adult life, physiologically and in various inflammatory and pathologic states. Studies by Nussenzweig (Dudziak et al. 2007), Merad (Merad 2010; Merad and Manz 2009) and their colleagues have helped to clarify the origins and population dynamics of myeloid dendritic cells, vis-à-vis monocyte/macrophages. Their fluorescence and transgenic methods will be useful to trace precursors of myelomonocytic cells in mouse tumours.

1

Macrophage Phenotype in Tumours

5

Tumours are obviously heterogeneous themselves, not only in their ability to invade (benign or malignant), but also in their micro-environment (lung, liver, bone and lymph nodes), origin (epithelial, mesenchymal and haemopoietic), vascularisation, within individual tumours as well as among different primary or secondary tumour populations. Other differences pertain as tumours induce matrix synthesis and catabolism, undergo hypoxia, apoptosis and necrosis. The concomitant presence of CD4+, CD8+ lymphocytes, FoxP3 positive suppressor cells, as well as innate lymphoid cells (NKT and NK cells) modulates myeloid cells, reciprocally. Tumour cells themselves often express characteristic properties of leukocytes that can contribute to their migration and invasion. Macrophages can also be tolerogenic and contribute to lymphocyte suppression by cell contact or secretory products. Dendritic cell maturation and antigen presentation can also be subverted by tumour- or other myeloid-derived products. Apart from the above considerations, many difficulties hinder experimental analysis of macrophage phenotype in tumours. Ideally, one should study naturally occurring tumours in situ, rather than transplantable models. Isolation of myeloid cells, especially macrophages, is difficult and prone to artefact, particularly if FACS analysis is not combined with immunocytochemistry in situ. The use of oncogenic transgenes, e.g. by Hanahan and colleagues (Hanahan 1989) made it possible to synchronise defined stages of experimental tumours. Mouse models do not necessarily replicate human tumours, often studied at late stages, or after chemotherapy and irradiation. Finally, macrophage markers used in the mouse and human may differ markedly between species. The interactions between macrophages and tumour cells result in novel gene expression profiles in both cell types, only partially reproduced during co-cultivation in vitro. Microarray and proteomic analyses, while powerful indicators of signatures, e.g. of type 1 interferon activation pathways, need refinement. The traditional methods of morphologic, diagnostic pathology are undergoing rapid advances, but have not yet progressed to interpret function at the single-cell level sufficiently.

Membrane Markers for Macrophages in Tumours Given the above caveats, we present a list of validated and candidate antigen markers to define macrophage heterogeneity in tumours (Table 1.1). We feel that the present focus reported in the literature is too narrow, that FACS analysis of isolated macrophages is insufficient and that whilst antigens are reasonably well-defined in the mouse, markers for human antigens are limited and not sufficiently characterised. Monoclonal and polyclonal antibodies for FACS and western blotting are not necessarily suitable for immunocytochemistry. Tissue preservation, antigen stability and antibody staining need to be optimised for each epitope. Table 1.1 includes members of a range of molecular families, varying in cell specificity. Markers include opsonic and non-opsonic phagocytic receptors, lectins and scavenger receptors, as well as cytokine receptors and other differentiation antigens, with some functional correlates.

6

H.-H. Lin and S. Gordon

Table 1.1 Selected membrane markers for macrophages in tumours Molecule Property Comment F4/80 EGF-TM7/adhesion-GPCR Peripheral tolerance, MI subpopulation CD97 EGF-TM7/adhesion-GPCR Myeloid, other cells EMR2 EGF-TM7/adhesion-GPCR Human, not mouse, aberrant in breast cancers CD68 LAMP family Pan-MI and DC, some tumours Gr-1 Ly-6 family PMN, immature monocytes Polymorphic, PMN, immature monocytes 7/4 Ly-6 family (Rosas et al. 2010) Siglec-1 IgSF Sialyl-ligand, e.g. Muc-1 CD163 SRCR family Glucocorticoid, IL-10 induced CD200/CD200R IgSF Receptor/ligand pair, negative regulator FcR IgSF Activatory/inhibitory CR3 Beta-2 integrin Opsonic and non-opsonic phagocytosis Adhesion SR-A SRCR family Clearance apoptotic cells, CSF-1 upregn MARCO SRCR family Adhesion, induced via TLRs CD36 Bispanner SR-B Ox-LDL, Apoptotic, Thrombospondin R MR C-type lectin Alternative activation marker Dectin-1 C-type lectin-like Beta-glucan R, ITAM-like domain Dectin-2 C-type lectin-like Subset macrophages, Mannose-ligand TLRs Leucine-rich repeat Sensor exogenous, host ligands IL-4/13 R Cytokine R Common, specific R, alternative activation CSF-1R Receptor tyrosine kinase fms GM-CSF R Haemopoietic R Fc-GMCSF chimeric ligand (Rosas et al. 2007) CX3CR1 GPCR Membrane bound fractalkine R CCR2 GPCR MCP-1 ligand

Curiously, in some cases, e.g. CD68, non-haemopoietic tumour cells are able to express leukocyte markers ectopically. Giant cells and hybrids arising from fusion of tumour cells and macrophages provide another mechanism for aberrant marker expression. Some of these markers have been utilised in inflammatory and infectious models in the mouse but not in tumours. The need for co-localisation and double/multiple labelling, so useful in FACS, is more difficult to achieve in immunohistochemistry, which also lacks quantitation. Laser capture microscopy and tissue arrays may overcome some of these difficulties.

EGF-TM7 Receptors and Tumour-Associated Macrophages The mouse differentiation antigen F4/80 is a well-characterised marker for mouse macrophages and has been implicated in peripheral tolerance in a non-tumour model (see below for references). The mouse and human antigen CD97 is not only associated

1

Macrophage Phenotype in Tumours

7

with myeloid cell differentiation and activation, but has also been studied in a variety of tumour settings in vivo. The closely related antigen EMR2 provides a sensitive marker for macrophage identification in human tissues. We review the common and selective characteristics of these molecules in detail, in relation to tissue specificity and as potential markers of macrophage heterogeneity and function in tumour–host interactions.

Common Characteristics of the EGF-TM7 Receptors F4/80, EMR2 and CD97 all belong to the group of EGF-TM7 molecules that make up the second largest GPCR sub-family in man, the adhesion-GPCRs (Fig. 1.1) (McKnight and Gordon 1996; McKnight and Gordon 1998; Stacey et al. 2000; Yona et al. 2008a; Bjarnadottir et al. 2004; Bjarnadottir et al. 2007). The EGF-TM7 receptors share many common characteristics in protein structure, cellular function

E E E E E

E

E

E E

E E

E

E

E

E

E E

F4/80

Expression

EGF-like domains GPS proteolysis Ligand(s)

Function(s)

EMR2

Monocytes Resident tissue macrophages (Kupffer cells, microglia, etc.) Immature dendritic cells Eosinophils TAMs 7

Monocytes Macrophages Dendritic cells Neutrophils

5

No Unknown

Yes Chondroitin sulfate

Induction of Peripheral tolerance

Priming of neutrophils

CD97

Lymphocytes Monocytes Macrophages Dendritic cells Neutrophils Smooth muscle cells Tumour cells 5 Yes CD55, Chondroitin sulfate Integrins (A5B1/AvB3 ) Migration and homeostasis of PMNs, HSC mobilization, CD4+ T cell co-stimulation, Angiogenesis, Tumour invasion

Fig. 1.1 Characteristics of F4/80, EMR2 and CD97. The three receptors are represented schematically. The EGF-like (E) motifs are shown as triangles, the GPS motif as a triangle with two disulfide bonds and the 7TM domain is represented by seven cylinders

8

H.-H. Lin and S. Gordon

and expression patterns (Stacey et al. 2000; Yona et al. 2008a). As suggested by the nomenclature, EGF-TM7 molecules are chimeric proteins composed of two major protein modules, namely the epidermal growth factor (EGF)-like motifs and the seven-span transmembrane (7TM) domains. A typical EGF-TM7 molecule contains tandem repeats of EGF-like motifs at the N terminus followed by a stalk region, which connects to a 7TM domain (McKnight and Gordon 1996; McKnight and Gordon 1998). Most of the EGF-like motifs of the EGF-TM7 molecules belong to the Ca2+binding subtype usually found in extracellular matrix proteins. Thus, Ca2+ binding is important for the cellular function of these receptors (Lin et al. 2001). The 7TM domains of the EGF-TM7 molecules share strong sequence similarity to the class B or secretin-like GPCRs. However, recent phylogenetic analyses have suggested an independent evolutionary lineage and group them with others to form the adhesionGPCR sub-family (Bjarnadottir et al. 2004; Bjarnadottir et al. 2007; Fredriksson et al. 2003). The stalk region between the EGF-like motifs and the 7TM domain usually contains numerous Ser and Thr residues and N-link glycosylation sites. Therefore, it is believed to be heavily decorated with O- and N-link glycans and is thought as a rigid structure. In addition, a highly conserved Cys-rich motif of ~50 residues located immediately upstream of the first TM region has been identified. A posttranslational auto-proteolytic reaction at this Cys-rich motif would cleave the receptor molecule into an extracellular- and 7TM-subunits (Lin et al. 2004). The specific proteolytic cleavage site is, therefore, named the GPCR proteolysis site (GPS) (Krasnoperov et al. 1999). It is thought that the EGF-TM7 receptors would bind to specific cellular ligands through the extracellular domain, especially by the EGF-like motifs, which in turn activate the 7TM domain to transmit intracellular signals. The majority of the EGF-TM7 receptors are restrictedly expressed in myeloid cells, including monocytes, macrophages, polymorphonuclear cells and dendritic cells (Stacey et al. 2000; Yona et al. 2008a). Thus, a role in innate as well as adaptive immune functions was predicted for these molecules.

F4/80 The F4/80 molecule is a ~160 kDa cell surface glycoprotein (Austyn and Gordon 1981). However, the full-length F4/80 cDNA predicted a mature protein of 904 amino acid residues with an estimated molecular weight of 99 kDa (Lin et al. 1997; McKnight et al. 1996). It was shown that the ~60 kDa difference in size was due to heavy glycosylation of the molecule as predicted by the multiple O- and N-linked glycosylation sites in the stalk region. In addition, a total of 7 EGF-like motifs, one glycosaminoglycan attachment site and an Arg-Gly-Asp (RGD) motif, were also identified in the extracellular domain of F4/80 (McKnight and Gordon 1996; McKnight and Gordon 1998; McKnight et al. 1996). Despite the presence of a typical

1

Macrophage Phenotype in Tumours

9

GPS motif, no GPS proteolytic cleavage was identified in F4/80 because of the lack of a consensus cleavage site. Hence, unlike typical adhesion-GPCRs, the F4/80 receptor is a single-chain polypeptide. The main cellular functions of F4/80 was found to be involved in the generation of Ag-specific efferent CD8+ regulatory T (Treg) cells responsible for peripheral immune tolerance (Lin et al. 2005). During the induction of peripheral immune tolerance, it is thought that F4/80 is required for the cellular interactions among Ag-presenting cells (APC) and other immune effector cells that lead to the generation of efferent CD8+ Treg cells. In addition, it was also suggested that F4/80 is involved in macrophage-NK interaction in a Listeria monocytogenes infection model (Warschkau and Kiderlen 1999). Overall, F4/80 is believed to play an immuneregulatory role through the interaction with an unidentified cellular ligand expressed on other immune effector cells (Lin et al. 2005; van den Berg and Kraal 2005). F4/80 is one of the best surface markers for the majority of mouse tissue macrophages (Taylor et al. 2005; Gordon and Taylor 2005; Austyn and Gordon 1981). The F4/80 Ag was detected strongly in many tissue macrophage populations, including Kupffer cells in liver, red pulp macrophages in spleen, microglia cells in brain as well as other resident macrophages in bone marrow stroma, gut lamina propria, testis, kidney, lymph nodes and peritoneum. On the other hand, macrophages within T-cell areas of lymph nodes (paracortex), spleen (white pulp) and Peyer’s patches are usually negative for F4/80. Very low levels of F4/80 were expressed in some other resident tissue macrophages such as alveolar macrophages, marginal zone and subcapsular sinus macrophages in the spleen and lymph nodes (Taylor et al. 2005; McKnight and Gordon 1998). Blood monocytes, the precursors of tissue macrophages, express less F4/80 than their tissue counterparts, suggesting that the expression of F4/80 is regulated during differentiation. Furthermore, it was also noted that F4/80 expression is modulated according to the activation status of macrophages. With regards to cells of other hematopoietic origins, no reactivity was observed in lymphoid cells, neutrophils and monocyte-derived osteoclasts. Nevertheless, eosinophils were shown to also express the F4/80 Ag. Likewise, Langerhans cells, a type of dendritic cells (DC) in the epidermis, were found to be F4/80 positive, but the Ag is down-regulated upon subsequent DC maturation and migration to the draining lymph nodes (Taylor et al. 2005; McKnight and Gordon 1998). In mice, F4/80-expressing TAMs have been documented for many types of tumours, either occurred spontaneously or induced experimentally (Qian & Pollard 2010; Mantovani et al. 2002; Rabinovich et al. 2007; Umemura et al. 2008). Therefore, the presence or absence of the F4/80 reactivity has been used to determine the efficiency of macrophage ablation experiments in tumour studies. In addition, F4/80 was also detected in tumour-infiltrating MDSC. The Ag has been useful to define the cellular origin of MDSC in tumours as MDSC was initially thought to represent immature myelomonocytic cells with common monocytic and granulocytic phenotypes. It is now generally accepted that TAMs and MDSC include cells with similar phenotypes to alternatively activated macrophages, which all express F4/80 (Gordon and Martinez 2010; Umemura et al. 2008; Martinez et al. 2009; Sinha et al. 2005).

10

H.-H. Lin and S. Gordon

CD97 The full-length CD97 molecule contains a total of five EGF-like motifs. However, as a result of RNA alternative splicing, three major isoforms containing different combinations of EGF-like motifs were predicted. These include CD97(1, 2, 5), CD97(1, 2, 3, 5) and CD97(1, 2, 3, 4, 5) (Gray et al. 1996; Hamann et al. 1995). Apart from multiple O- and N-linked glycosylation sites, one RGD motif and a complete GPS motif were present in the extracellular region. Hence, CD97 was cleaved at the predicted GPS site into two subunits (Gray et al. 1996). CD97 was found to interact through its extracellular region with several cellular ligands, including CD55 (DAF) (Hamann et al. 1998; Hamann et al. 1996), dermatan sulphate (Kwakkenbos et al. 2005; Stacey et al. 2003) and D5E1/DvE3 integrins (Wang et al. 2005). Interestingly, the CD97-ligand interaction is mostly isoform-specific such that CD55 binds better to CD97(1, 2, 5), while dermatan sulphate only interacts with CD97(1, 2, 3, 4, 5) (Lin et al. 2001; Stacey et al. 2003). CD97–CD55 interaction is species-specific so that human CD97 only reacts with human CD55 but not murine CD55 (Lin et al. 2001). Through the utilisation of specific mAbs, soluble CD97 proteins, and the generation and analysis of knock-out animals, CD97 has been implicated in the cellular migration and homeostasis of polymorphonuclear cells (Kop et al. 2006; Leemans et al. 2004; Wang et al. 2007), hematopoietic stem cell/progenitor cell mobilisation (van Pel et al. 2008a; van Pel et al. 2008b), co-stimulation of CD4+ T cells (Abbott et al. 2007; Capasso et al. 2006) and angiogenesis (Wang et al. 2005). CD97 was also shown to be potentially involved in the pathogenesis of arthritis (Kop et al. 2006). CD97 was first identified as an early activation marker for T and B lymphocytes, which express low levels of CD97 in resting conditions (Gray et al. 1996; Hamann et al. 1995). On the other hand, CD97 is constitutively expressed on granulocytes and monocytes/macrophages (Jaspars et al. 2001). Immunohistochemistry staining of normal human tissues in situ has shown that CD97 is abundantly expressed in resident macrophages of most tissues (Jaspars et al. 2001). These include liver (Kupffer cells and periportal histiocytes), lung (alveolar macrophages), skin (including Langerhans cells), brain (perivascular macrophages but not microglia), kidney (mesangial cells of the glomeruli) and secondary lymphoid organs such as lymph nodes, spleen (red pulp and white pulp macrophages), tonsil and mucosa-associated lymphoid tissues (MALT). Dendritic cells in most of the lymphoid tissues also express CD97, whereas CD97-expressing lymphocytes are mostly restricted in intraepithelial and sub-epithelial locations. Lymphocytes in paracortical areas, follicles and germinal centres are mostly CD97 weak or negative. Outside lymphoid tissues, smooth muscle cells also express CD97. In addition, CD97 was often detected in many types of tumours, including thyroid, gastric, pancreatic, esophageal and colorectal carcinomas (Aust et al. 1997; Aust et al. 2002; Steinert et al. 2002). Most interestingly, stronger CD97 expression levels were usually detected at the invasion fronts of tumours, suggesting a role in tumour cell migration/invasion (Steinert et al. 2002; Galle et al. 2006; Wobus et al. 2006).

1

Macrophage Phenotype in Tumours

11

Specifically, in colorectal and gastric carcinomas, the strongest CD97 staining was identified in disseminated or scattered tumour cells at the invasion front of the tumour. Furthermore, a poorer clinical stage and increased lymph vessel invasion were found to be correlated positively in tumour patients with more CD97-high scattered tumour cells. In the cell level, higher CD97 expression stimulated singlecell motility, enhanced the proteolytic activity of matrix metalloproteinases and secretion of chemokines. Tumour growth in scid mice was enhanced in cells overexpressing CD97 (Steinert et al. 2002; Galle et al. 2006; Wobus et al. 2006). Moreover, CD97 was shown recently to stimulate angiogenesis through binding to D5E1 integrin on endothelial cells (Wang et al. 2005). Briefly, soluble CD97 was tested in a quantitative, directed in vivo angiogenesis assay (DIVAA). It was found that CD97 at 25 and 100 ng/ml is as efficient as bFGF in inducing blood vessel development in vivo. Furthermore, developing tumours derived from CD97expressing cells display a significantly greater vessel density than those from control CD97-negative tumour cells. The angiogenic response of CD97 is most likely mediated by the RGD motif in the stalk as it interacts with the D5E1 and DvE3 integrins. Through these interactions, CD97 promotes the adhesion, migration and invasion of human umbilical vein endothelial cells (HUVECs). Nevertheless, another ligand of CD97, chondroitin sulphate, seems to also contribute to these effects (Wang et al. 2005). A more recent study using CD97 transgenic mice, however, indicates that the role of CD97 in tumourigenesis might be more complicated than we thought earlier (Becker et al. 2010). To investigate the involvement of CD97 in colorectal carcinogenesis, Becker et al. generated transgenic mice that overexpress CD97 specifically in enterocytes (Becker et al. 2010). These animals were then subjected to azoxymethane (AOM)/dextran sodium sulphate (DSS)-induced colitis-associated tumourigenesis. Interestingly, DSS-induced colitis was reduced in transgenic mice when compared with the wild-type control. This reduction was dependent on the copy number of the CD97 transgene. Through ultrastructural and biochemical analyses, it was concluded that CD97 over-expression can enhance the structural integrity of enterocytic adherens junctions, which in turn enforce intestinal epithelial strength leading to the attenuation of experimental colitis.

EMR2 Highly similar to CD97, the full-length EMR2 molecule also contains a total of five EGF-like motifs. A total of four major isoforms, namely EMR2(1, 2), EMR2 (1, 2, 5), EMR2(1, 2, 3, 5) and EMR2(1, 2, 3, 4, 5), were predicted from spliced mRNA sequences (Lin et al. 2000). EMR2 shares ~97% sequence identity in the EGF-like domains with CD97. Interestingly, however, a highly restricted ligand binding specificity was identified for these EGF-TM7 molecules. Thus, EMR2 (1, 2, 5) was shown to bind CD55 with a tenfold weaker affinity than CD97(1, 2, 5), even though the two molecules only differ in three residues (Lin et al. 2001).

12

H.-H. Lin and S. Gordon

Nevertheless, EMR2(1, 2, 3, 4, 5) and CD97(1, 2, 3, 4, 5) were found to bind dermatan sulphate equally well (Stacey et al. 2003). These results suggest potential overlapping and also unique functions for EMR2 and CD97. Phylogenetic analysis has revealed an interesting evolutionary relationship between EMR2 and CD97; although both genes are highly homologous in the human genome, the EMR2 gene is lost in rodents, while the CD97 gene is present in man and mouse. In fact, it seems that EMR2 evolved through gene conversion and duplication by combining part of CD97 and another EGF-TM7 receptor, EMR3. Thus, EMR2 is a chimeric gene evolved relatively recently, probably to meet the need of the immune system in primates (Kwakkenbos et al. 2004; Kwakkenbos et al. 2006). A role for EMR2 in regulating neutrophil activation was found recently (Yona et al. 2008b). Binding of EMR2 receptor by a specific mAb was shown to strongly enhance the inflammatory responses of neutrophils to a panel of stimuli. Interestingly, mAb treatment alone did not activate neutrophils. Hence, EMR2 receptor activation seems to have a priming effect on neutrophil activation. Expression of EMR2 is more restricted than CD97 to myeloid cells (Lin et al. 2000; Chang et al. 2007; Kwakkenbos et al. 2002). Monocytes, macrophages, neutrophils and dendritic cells have all been shown to express EMR2. Specifically, the strongest EMR2 signal was detected on CD16+ blood monocytes, macrophages and BDCA-3+ myeloid dendritic cells. No expression was ever detected in resting or activated lymphocytes. Interestingly, it was found that EMR2 is up-regulated during the differentiation and maturation of macrophages in vitro, but is downregulated during dendritic cell maturation (Chang et al. 2007). EMR2 expression in monocytes and macrophages can be up-regulated by LPS and IL-10 in an IL-10dependent manner. In some limited tissues surveyed, EMR2 was detected in certain macrophage sub-populations of skin, spleen, lung, placenta, and tonsil. Macrophages of liver and kidney do not seem to express EMR2. In inflamed tissues, EMR2 was found in sub-populations of macrophages and neutrophils. Recently, it was shown that foamy macrophages in atherosclerotic vessels and Gaucher cells in spleen express high levels of EMR2. In contrast, multiple sclerosis brain foam cells expressed little if any EMR2, but strong CD97 (van Eijk et al. 2010). Unlike CD97, EMR2 expression in tumours has not been observed much. In fact, in thyroid, gastric, pancreatic, and esophageal carcinomas, many tumour cells are CD97 positive but EMR2 negative (Aust et al. 2002). In some colorectal tumour cell lines and adenocarcinomas, alternatively spliced EMR2 mRNA transcripts were identified, but EMR2 protein expression remained low in the tumour cells (Aust et al. 2003). Within the carcinomas, certain tumour-infiltrating macrophages expressed strong surface EMR2. Interestingly, by staining breast cancer tissue sections, we recently identify strong EMR2 reactivity within tumour cells (Davies et al. unpublished results). The significance of this finding awaits further investigation, but suggests that aberrantly expressed EMR2 could be linked to certain tumour types. In conclusion, members of the myeloid-restricted adhesion-GPCRs, the EGF-TM7 receptors, could be considered as important new molecules associated with tumourigenesis. Based on the structural, functional and expressional characteristics, the

1

Macrophage Phenotype in Tumours

13

EGF-TM7 proteins could serve as the cell surface marker of tumour-infiltrating or tumour-associated myeloid cells (macrophages and neutrophils). These receptors can also be expressed by tumour cells to facilitate the interaction with the tumour microenvironment including cells and extracellular matrix. By doing so, the EGF-TM7 receptors might promote angiogenesis and the migration and invasion of tumour cells. Finally, a potential role of the EGF-TM7 receptors in immunosuppression might provide tumour cells an opportunity to evade immune surveillance.

Conclusion A larger range of molecular markers is available than currently in use, to characterise macrophages within and isolated from tumours. The leap from macrophage marker to function, taking into account the relevant tumour subpopulation actually responsible for tumour progression, is considerable. A great deal of further characterisation of promising markers is needed, especially in humans, who provide a large pool of material of natural history and diversity. Studies on individual macrophage markers, membrane and otherwise, combined with biomarkers in whole blood and tissues, should aid diagnosis and possible targeting of macrophage functions in the future.

References Abbott RJ, Spendlove I, Roversi P, Fitzgibbon H, Knott V, Teriete P, McDonnell JM, Handford PA, Lea SM (2007) Structural and functional characterization of a novel T cell receptor co-regulatory protein complex, CD97–CD55. J Biol Chem 282(30):22023–22032 Auffray C, Sieweke MH, Geissmann F (2009) Blood monocytes: development, heterogeneity, and relationship with dendritic cells. Annu Rev Immunol 27:669–692 Aust G, Eichler W, Laue S, Lehmann I, Heldin NE, Lotz O, Scherbaum WA, Dralle H, Hoang-Vu C (1997) CD97: a dedifferentiation marker in human thyroid carcinomas. Cancer Res 57(9): 1798–1806 Aust G, Steinert M, Schutz A, Boltze C, Wahlbuhl M, Hamann J, Wobus M (2002) CD97, but not its closely related EGF-TM7 family member EMR2, is expressed on gastric, pancreatic, and esophageal carcinomas. Am J Clin Pathol 118:699–707 Aust G, Hamann J, Schilling N, Wobus M (2003) Detection of alternatively spliced EMR2 mRNAs in colorectal tumor cell lines but rare expression of the molecule in colorectal adenocarcinomas. Virchows Arch 443(1):32–37 Austyn JM, Gordon S (1981) F4/80, a monoclonal antibody directed specifically against the mouse macrophage. Eur J Immunol 11(10):805–85 Becker S, Wandel E, Wobus M, Schneider R, Amasheh S, Sittig D, Kerner C, Naumann R, Hamann J, Aust G (2010) Overexpression of CD97 in intestinal epithelial cells of transgenic mice attenuates colitis by strengthening adherens junctions. PLoS One 5(1):e8507 Bjarnadottir TK, Fredriksson R, Hoglund PJ, Gloriam DE, Lagerstrom MC, Schioth HB (2004) The human and mouse repertoire of the adhesion family of G-protein-coupled receptors. Genomics 84(1):23–33 Bjarnadottir TK, Fredriksson R, Schioth HB (2007) The adhesion GPCRs: a unique family of G protein-coupled receptors with important roles in both central and peripheral tissues. Cell Mol Life Sci 64(16):2104–2119

14

H.-H. Lin and S. Gordon

Capasso M, Durrant LG, Stacey M, Gordon S, Ramage J, Spendlove I (2006) Costimulation via CD55 on human CD4+ T cells mediated by CD97. J Immunol 177(2):1070–1077 Chang GW, Davies JQ, Stacey M, Yona S, Bowdish DM, Hamann J, Chen TC, Lin CY, Gordon S, Lin HH (2007) CD312, the human adhesion-GPCR EMR2, is differentially expressed during differentiation, maturation, and activation of myeloid cells. Biochem Biophys Res Commun 353(1):133–138 Coffelt SB, Hughes R, Lewis CE (2009) Tumor-associated macrophages: effectors of angiogenesis and tumor progression. Biochim Biophys Acta 1796(1):11–18 Coussens LM, Werb Z (2002) Inflammation and cancer. Nature 420(6917):860–867 Domachowske JB, Bonville CA, Gao JL, Murphy PM, Easton AJ, Rosenberg HF (2000) The chemokine macrophage-inflammatory protein-1 alpha and its receptor CCR1 control pulmonary inflammation and antiviral host defense in paramyxovirus infection. J Immunol 165(5):2677–2682 Dudziak D, Kamphorst AO, Heidkamp GF, Buchholz VR, Trumpfheller C, Yamazaki S, Cheong C, Liu K, Lee HW, Park CG, Steinman RM, Nussenzweig MC (2007) Differential antigen processing by dendritic cell subsets in vivo. Science (New York, NY) 315(5808): 107–111 Fredriksson R, Lagerstrom MC, Lundin LG, Schioth HB (2003) The G-protein-coupled receptors in the human genome form five main families. Phylogenetic analysis, paralogon groups, and fingerprints. Mol Pharmacol 63(6):1256–1272 Gabrilovich DI, Nagaraj S (2009) Myeloid-derived suppressor cells as regulators of the immune system. Nat Rev Immunol 9(3):162–174 Galle J, Sittig D, Hanisch I, Wobus M, Wandel E, Loeffler M, Aust G (2006) Individual cell-based models of tumor-environment interactions: Multiple effects of CD97 on tumor invasion. Am J Pathol 169(5):1802–1811 Geissmann F, Manz MG, Jung S, Sieweke MH, Merad M, Ley K (2010) Development of monocytes, macrophages, and dendritic cells. Science (New York, NY) 327(5966):656–661 Gordon S, Martinez FO (2010) Alternative activation of macrophages: mechanism and functions. Immunity 32(5):593–604 Gordon S, Taylor PR (2005) Monocyte and macrophage heterogeneity. Nat Rev Immunol 5(12): 953–964 Gray JX, Haino M, Roth MJ, Maguire JE, Jensen PN, Yarme A, Stetler-Stevenson MA, Siebenlist U, Kelly K (1996) CD97 is a processed, seven-transmembrane, heterodimeric receptor associated with inflammation. J Immunol 157(12):5438–5447 Grivennikov SI, Greten FR, Karin M (2010) Immunity, inflammation, and cancer. Cell 140(6):883–899 Hamann J, Eichler W, Hamann D, Kerstens HM, Poddighe PJ, Hoovers JM, Hartmann E, Strauss M, van Lier RA (1995) Expression cloning and chromosomal mapping of the leukocyte activation antigen CD97, a new seven-span transmembrane molecule of the secretion receptor superfamily with an unusual extracellular domain. J Immunol 155(4):1942–1950 Hamann J, Vogel B, van Schijndel GM, van Lier R (1996) The seven-span transmembrane receptor CD97 has a cellular ligand (CD55, DAF). J Exp Med 184(3):1185–1189 Hamann J, Stortelers C, Kiss-Toth E, Vogel B, Eichler W, van Lier RA (1998) Characterization of the CD55 (DAF)-binding site on the seven-span transmembrane receptor CD97. Eur J Immunol 28(5):1701–1707 Hanahan D (1989) Transgenic mice as probes into complex systems. Science (New York, NY) 246(4935):1265–1275 Jaspars LH, Vos W, Aust G, Van Lier RA, Hamann J (2001) Tissue distribution of the human CD97 EGF-TM7 receptor. Tissue Antigens 57(4):325–331 Joyce JA, Pollard JW (2009) Microenvironmental regulation of metastasis. Nat Rev Cancer 9(4):239–252 Kop EN, Adriaansen J, Smeets TJ, Vervoordeldonk MJ, van Lier RA, Hamann J, Tak PP (2006) CD97 neutralisation increases resistance to collagen-induced arthritis in mice. Arthritis Res Ther 8(5):R155

1

Macrophage Phenotype in Tumours

15

Krasnoperov V, Bittner MA, Holz RW, Chepurny O, Petrenko AG (1999) Structural requirements for alpha-latrotoxin binding and alpha-latrotoxin-stimulated secretion. A study with calciumindependent receptor of alpha-latrotoxin (CIRL) deletion mutants. J Biol Chem 274(6): 3590–3596 Kwakkenbos MJ, Chang GW, Lin HH, Pouwels W, de Jong EC, van Lier RA, Gordon S, Hamann J (2002) The human EGF-TM7 family member EMR2 is a heterodimeric receptor expressed on myeloid cells. J Leukoc Biol 71(5):854–862 Kwakkenbos MJ, Kop EN, Stacey M, Matmati M, Gordon S, Lin HH, Hamann J (2004) The EGF-TM7 family: a postgenomic view. Immunogenetics 55(10):655–666 Kwakkenbos MJ, Pouwels W, Matmati M, Stacey M, Lin HH, Gordon S, van Lier RA, Hamann J (2005) Expression of the largest CD97 and EMR2 isoforms on leukocytes facilitates a specific interaction with chondroitin sulfate on B cells. J Leukoc Biol 77(1):112–119 Kwakkenbos MJ, Matmati M, Madsen O, Pouwels W, Wang Y, Bontrop RE, Heidt PJ, Hoek RM, Hamann J (2006) An unusual mode of concerted evolution of the EGF-TM7 receptor chimera EMR2. FASEB J 20(14):2582–2584 Leemans JC, te Velde AA, Florquin S, Bennink RJ, de Bruin K, van Lier RA, van der Poll T, Hamann J (2004) The epidermal growth factor-seven transmembrane (EGF-TM7) receptor CD97 is required for neutrophil migration and host defense. J Immunol 172(2):1125–1131 Lin HH, Stubbs LJ, Mucenski ML (1997) Identification and characterization of a seven transmembrane hormone receptor using differential display. Genomics 41(3):301–308 Lin HH, Stacey M, Hamann J, Gordon S, McKnight AJ (2000) Human EMR2, a novel EGF-TM7 molecule on chromosome 19p13.1, is closely related to CD97. Genomics 67(2):188–200 Lin HH, Stacey M, Saxby C, Knott V, Chaudhry Y, Evans D, Gordon S, McKnight AJ, Handford P, Lea S (2001) Molecular analysis of the epidermal growth factor-like short consensus repeat domain-mediated protein-protein interactions: dissection of the CD97–CD55 complex. J Biol Chem 276(26):24160–24169 Lin HH, Chang GW, Davies JQ, Stacey M, Harris J, Gordon S (2004) Autocatalytic cleavage of the EMR2 receptor occurs at a conserved G protein-coupled receptor proteolytic site motif. J Biol Chem 279(30):31823–31832 Lin HH, Faunce DE, Stacey M, Terajewicz A, Nakamura T, Zhang-Hoover J, Kerley M, Mucenski ML, Gordon S, Stein-Streilein J (2005) The macrophage F4/80 receptor is required for the induction of antigen-specific efferent regulatory T cells in peripheral tolerance. J Exp Med 201(10):1615–1625 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(11):549–555 Mantovani A, Allavena P, Sica A, Balkwill F (2008) Cancer-related inflammation. Nature 454(7203):436–444 Martinez FO, Helming L, Gordon S (2009) Alternative activation of macrophages: an immunologic functional perspective. Annu Rev Immunol 27:451–483 McKnight AJ, Gordon S (1996) EGF-TM7: a novel subfamily of seven- transmembrane-region leukocyte cell-surface molecules. Immunol Today 17(6):283–287 McKnight AJ, Gordon S (1998) The EGF-TM7 family: unusual structures at the leukocyte surface. J Leukoc Biol 63(3):271–280 McKnight AJ, Macfarlane AJ, Dri P, Turley L, Willis AC, Gordon S (1996) Molecular cloning of F4/80, a murine macrophage-restricted cell surface glycoprotein with homology to the G-protein-linked transmembrane 7 hormone receptor family. J Biol Chem 271(1):486–489 Merad M (2010) PU.1 takes control of the dendritic cell lineage. Immunity 32(5):583–585 Merad M, Manz MG (2009) Dendritic cell homeostasis. Blood 113(15):3418–3427 Peranzoni E, Zilio S, Marigo I, Dolcetti L, Zanovello P, Mandruzzato S, Bronte V (2010) Myeloid-derived suppressor cell heterogeneity and subset definition. Curr Opin Immunol 22(2):238–244 Qian BZ, Pollard JW (2010) Macrophage diversity enhances tumor progression and metastasis. Cell 141(1):39–51

16

H.-H. Lin and S. Gordon

Rabinovich GA, Gabrilovich D, Sotomayor EM (2007) Immunosuppressive strategies that are mediated by tumor cells. Annu Rev Immunol 25:267–296 Rosas M, Gordon S, Taylor PR (2007) Characterisation of the expression and function of the GM-CSF receptor alpha-chain in mice. Eur J Immunol 37(9):2518–2528 Rosas M, Thomas B, Stacey M, Gordon S, Taylor PR (2010) The myeloid 7/4-antigen defines recently generated inflammatory macrophages and is synonymous with Ly-6B. J Leukoc Biol 88(1):169–180 Sinha P, Clements VK, Ostrand-Rosenberg S (2005) Reduction of myeloid-derived suppressor cells and induction of M1 macrophages facilitate the rejection of established metastatic disease. J Immunol 174(2):636–645 Stacey M, Lin HH, Gordon S, McKnight AJ (2000) LNB-TM7, a group of seven-transmembrane proteins related to family-B G-protein-coupled receptors. Trends Biochem Sci 25(6):284–289 Stacey M, Chang GW, Davies JQ, Kwakkenbos MJ, Sanderson RD, Hamann J, Gordon S, Lin HH (2003) The epidermal growth factor-like domains of the human EMR2 receptor mediate cell attachment through chondroitin sulfate glycosaminoglycans. Blood 102(8):2916–2924 Steinert M, Wobus M, Boltze C, Schutz A, Wahlbuhl M, Hamann J, Aust G (2002) Expression and regulation of CD97 in colorectal carcinoma cell lines and tumor tissues. Am J Pathol 161(5):1657–1667 Taylor PR, Martinez-Pomares L, Stacey M, Lin HH, Brown GD, Gordon S (2005) Macrophage receptors and immune recognition. Annu Rev Immunol 23:901–944 Umemura N, Saio M, Suwa T, Kitoh Y, Bai J, Nonaka K, Ouyang GF, Okada M, Balazs M, Adany R, Shibata T, Takami T (2008) Tumor-infiltrating myeloid-derived suppressor cells are pleiotropic-inflamed monocytes/ macrophages that bear M1- and M2-type characteristics. J Leukoc Biol 83(5):1136–1144 van den Berg TK, Kraal G (2005) A function for the macrophage F4/80 molecule in tolerance induction. Trends Immunol 26(10):506-509 van Eijk M, Aust G, Brouwer MS, van Meurs M, Voerman JS, Dijke IE, Pouwels W, Sandig I, Wandel E, Aerts JM, Boot RG, Laman JD, Hamann J (2010) Differential expression of the EGF-TM7 family members CD97 and EMR2 in lipid-laden macrophages in atherosclerosis, multiple sclerosis and Gaucher disease. Immunol Lett 129(2):64–71 van Pel M, Hagoort H, Hamann J, Fibbe WE (2008a) CD97 is differentially expressed on murine hematopoietic stem-and progenitor-cells. Haematologica 93(8):1137–1144 van Pel M, Hagoort H, Kwakkenbos MJ, Hamann J, Fibbe WE (2008b) Differential role of CD97 in interleukin-8-induced and granulocyte-colony stimulating factor-induced hematopoietic stem and progenitor cell mobilization. Haematologica 93(4):601–604 Varol C, Yona S, Jung S (2009) Origins and tissue-context-dependent fates of blood monocytes. Immunol Cell Biol 87(1):30–38 Wang T, Ward Y, Tian L, Lake R, Guedez L, Stetler-Stevenson WG, Kelly K (2005) CD97, an adhesion receptor on inflammatory cells, stimulates angiogenesis through binding integrin counterreceptors on endothelial cells. Blood 105(7):2836–2844 Wang T, Tian L, Haino M, Gao JL, Lake R, Ward Y, Wang H, Siebenlist U, Murphy PM, Kelly K (2007) Improved antibacterial host defense and altered peripheral granulocyte homeostasis in mice lacking the adhesion class G protein receptor CD97. Infect Immun 75(3):1144–1153 Warschkau H, Kiderlen AF (1999) A monoclonal antibody directed against the murine macrophage surface molecule F4/80 modulates natural immune response to Listeria monocytogenes. J Immunol 163(6):3409–3416 Wobus M, Huber O, Hamann J, Aust G (2006) CD97 overexpression in tumor cells at the invasion front in colorectal cancer (CC) is independently regulated of the canonical Wnt pathway. Mol Carcinog 45(11):881–886 Yona S, Lin HH, Siu WO, Gordon S, Stacey M (2008a) Adhesion-GPCRs: emerging roles for novel receptors. Trends Biochem Sci 33(10):491–500 Yona S, Lin HH, Dri P, Davies JQ, Hayhoe RP, Lewis SM, Heinsbroek SE, Brown KA, Perretti M, Hamann J, Treacher DF, Gordon S, Stacey M (2008b) Ligation of the adhesion-GPCR EMR2 regulates human neutrophil function. FASEB J 22(3):741–751

Chapter 2

Role of Tumour-Associated Macrophages in the Regulation of Angiogenesis Russell Hughes, Hsin-Yu Fang, Munitta Muthana, and Claire E. Lewis

Introduction As mentioned in previous chapters, tumours consist not only of malignant cells but also of various stromal cell types including tumour-associated macrophages (TAMs) (Sica et al. 2008). One early sign that TAMs might influence tumour angiogenesis was the finding that TAM numbers positively correlate with tumour angiogenesis in breast carcinomas (Leek et al. 1996). Several subsequent studies have confirmed such a link in a wide array of tumour types (Aharinejad et al. 2004; Bailey et al. 2007; Koide et al. 2004; Ohta et al. 2003; Saji et al. 2001) and showed that high-TAM numbers are also often linked to poor prognosis (Bingle et al. 2002; Lewis and Pollard 2006). However, definitive evidence for the pro-angiogenic effect of TAMs in tumours was provided using various murine tumour models. First, we demonstrated that macrophage infiltration into small breast tumour nodules (tumour spheroids) markedly enhanced their ability to induce neovascularisation following implantation into window chambers on the flanks of mice (Bingle et al. 2006). Then Lin and colleagues showed that, when MMTV-PyMT mice (which develop mammary tumours) were crossed with transgenic mice carrying a colony stimulating factor-1 (Csf1) null mutation (Csf1op/op), the absence of CSF-1 markedly decreased macrophage infiltration in pre-malignant tumours, which, in turn, resulted in the inhibition of tumour angiogenesis and delayed tumour progression (Lin et al. 2006; Lin et al. 2001). Indeed, they showed that TAMs control the ‘angiogenic switch’ associated with the malignant transition of the spontaneous mammary tumours that form in MMTV-PyMT mice (Lin et al. 2006). Other strategies employing an antibody to CSF-1 (Paulus et al. 2006) or clodronate liposomes (Zeisberger et al. 2006) to deplete TAMs have also resulted in a marked reduction in tumour angiogenesis.

2(UGHESs( 9&ANGs--UTHANAs#%,EWIS*) Academic Unit of Inflammation and Tumour Targeting, University of Sheffield Medical School, Beech Hill Road, Sheffield S10 2RX, UK e-mail: claire.lewis@sheffield.ac.uk T. Lawrence and T. Hagemann (eds.), Tumour-Associated Macrophages, DOI 10.1007/978-1-4614-0662-4_2, © Springer Science+Business Media, LLC 2012

17

18

R. Hughes et al.

%LEVATEDEXPRESSIONOFSUCHMACROPHAGECHEMOATTRACTANTSASTHE## CHEMOKINE CCL2 (MCP-1) in solid tumours correlates both with increased TAM recruitment (Hoshino et al. 1995; Leung et al. 1997) and with increased tumour neovascularisation (Barleon et al. 1996; Zhu et al. 2004). Additionally, the accumulation of TAMs INSOLIDTUMOURSISASSOCIATEDWITHTHEINCREASEDEXPRESSIONOFSUCHPRO ANGIOGENIC MOLECULESAS6%'&! ANDB&'& ANDCORRELATESWITHINCREASEDFORMATIONOFANGIOgenic blood vessels (Bolat et al. 2006; Tsutsui et al. 2005). Functional plasticity is a hallmark feature of macrophages and they can respond to diverse environmental signals with a wide array of functional phenotypes. Consistent with this, a variety of overlapping macrophage activation states have been reported with classically activated, ‘M1’ and alternatively activated, ‘M2’ MACROPHAGESREPRESENTINGOPPOSINGEXTREMESOFTHISCONTINUUM%XPOSURETOPRO inflammatory cytokines and microbial products such as interferon gamma (IFNJ) and lipopolysaccharide (LPS) polarizes macrophages into the pro-inflammatory, -PHENOTYPE WHICHISCHARACTERIZEDBYTHEINCREASEDEXPRESSIONOFTHEIROWNPRO INmAMMATORYMEDIATORSINCLUDINGINDUCIBLENITRICOXIDESYNTHASEI./3 REACTIVE NITROGEN INTERMEDIATES 2.) REACTIVE OXYGEN INTERMEDIATES 2/) AND THEIR increased microbicidal/tumoricidal activity. Anti-inflammatory molecules such as the Th2 cytokines, IL-4, IL-13 and IL-10, as well as apoptotic cells and immune COMPLEXESINCOMBINATIONWITH4OLL LIKERECEPTOR4,2 STIMULATIONINDUCEMACROPHAGES TO EXPRESS AN - ACTIVATION PHENOTYPE 3UCH @- MACROPHAGES ARE thought to be involved in promoting angiogenesis and tissue remodelling/repair, dampening inflammation by producing high levels of anti-inflammatory cytokines (e.g. IL-10 and transforming growth factor-E 4'&E) and up-regulating products of the arginase pathway (arginase II, ornithine and polyamines). They also display a HIGHLEVELOFPHAGOCYTICACTIVITYANDEXPRESSHIGHLEVELSOFBOTHMANNOSEANDSCAVENGERRECEPTORS'ORDON2003; Mantovani et al. 2004; Mantovani et al. 2002; Mills et al. 2000; Siciliano et al. 2006). However, it should be noted that variations within the M2 phenotype of macrophages have recently been described leading to the subdivisions, M2a, -b and -c. Interleukins 4 and -13 stimulate the onset of the M2a phenotype in macrophages (Mantovani et al. 2002; Martinez et al. 2009), whereas EXPOSURETOIMMUNECOMPLEXESINCOMBINATIONWITH4,2ACTIVATIONINDUCES-B TYPEMACROPHAGES&INALLY EXPOSURETOGLUCOCORTICOIDSOR), RESULTSINTHEPRODUCtion of M2c-type macrophages, with greater immunosuppressive, pro-angiogenic ANDTISSUEREMODELLINGFUNCTIONS'OERDTAND/RFANOS1999). Murine TAMs show many characteristics of an M2-polarized state (Biswas et al. 2008; Mantovani et al. 2002 &OREXAMPLE 4!-SOFTENUP REGULATEPRO ANGIOGENIC MEDIATORS SUCH AS 6%'&! "ALKWILL AND -ANTOVANI 2001 CYCLO OXYGENASE (COX)-2 (Nakao et al. 2005) and MMP9 (Coussens et al. 2000). They also have impaired EXPRESSION OF SUCH PRO INmAMMAOTORY MEDIATORS AS EG ),  AND 4.&D), RNI AND MAJOR HISTOCOMPATIBILITY COMPLEX -(# )) WHICH FUNCTIONALLY CORRELATE TO REDUCEDCYTOTOXICITYANDANTIGENPRESENTINGCAPACITY#ONVERSELY 4!-SUP REGULATE ANTI INmAMMATORY MOLECULES SUCH AS ),  AND 4'&E. Other hallmark markers of M2 activation, such as up-regulated levels of the M2-specific genes, arginase,

2

Role of Tumour-Associated Macrophages in the Regulation of Angiogenesis

19

macrophage galactose-type C-type lectin-2 (Mgl2), found in inflammatory zone 1 &IZZ AND 9M -ANTOVANI ET AL 2003 6AN 'INDERACHTER ET AL 2006) are also EXPRESSEDBYMURINE4!-S(OWEVER ITSHOULDBENOTEDTHATSOME4!-SINMOUSE tumour models, as well as human tumours (e.g. breast, ovarian and renal cell carciNOMA EXHIBIT TYPICAL - CYTOKINES SUCH AS 4.&D, IL-1E, IL-6, IL-23p19 and CXCL8, depending on tumour type and stage (Biswas et al. 2006; Dinapoli et al. 1996; Langowski et al. 2006). This has lead to the suggestion that the phenotype of TAMs may vary between different stages of tumour development and possibly different regions of the same tumours (Murdoch et al. 2008; Ohno et al. 2004).

Pro-angiogenic Functions of Macrophages in Tumours 4!-SEXPRESSABROADNUMBEROFCYTOKINESTHATARECAPABLEOFDIRECTLYINDUCINGTHE FORMATIONOFBLOODVESSELSWITHINTUMOURSINCLUDING6%'&!ANDB&'&"OLATETAL 2006; Tsutsui et al. 2005). However, TAMs also produce a wide range of pleiotrophic factors including IL1E (Voronov et al. 2003 4'& E1, and TNFD (Luo et al. 2006; 9OSHIDA ET AL 1997), which promote new blood vessel formation by promoting THERELEASEOFSUCHFACTORSAS6%'&!FROMOTHERCELLTYPESCOMMONLYFOUNDWITHIN the tumour micro-environment (Pertovaara et al. 1994). TAMs also indirectly promote the formation of blood vessels in tumours using a variety of TAM-derived enzymes SUCHASTHEMATRIXMETALLOPROTEINASES--0S   AND 'IRAUDOETAL2004; Hildenbrand et al. 1995; Houghton et al. 2006; Huang et al. 2002; Nozawa et al. 2006  4HESE ARE EXTRACELLULAR MATRIX %#- REMODELLING ENZYMES THAT ALTER THE STRUCTUREOFTHE%#-TOSUPPORTTHEATTACHMENT MIGRATIONANDINVASIONOFACTIVATED%#S)NADDITION THEINTERACTIONOF--0SWITHTHE%#-CANALSOINCREASETHE BIOAVAILABILITYOFSUCHPRO ANGIOGENICFACTORSAS6%'&!ANDB&'&!CCORDINGLY THERECRUITMENTOFMYELOIDCELLSEXPRESSING--0ISIMPORTANTFORTHELIBERATIONOF 6%'&! FROM THE %#- )T WAS THEREFORE HARDLY SURPRISING TO lND THAT --0 EXPRESSINGMYELOIDCELLSWEREESSENTIALFORTUMOURANGIOGENESISINAMURINEMODEL of glioblastoma (Du et al. 2008). In addition, mice lacking MMPs 2 and 7 show a reduced rate of tumour progression when implanted with B16 melanoma and Lewis Lung carcinoma cells or in a murine colorectal carcinoma model (Itoh et al. 1998; Wilson et al. 1997  4!-S ALSO EXPRESS THE IMPORTANT PRO ANGIOGENIC ENZYME thymidine phosphorylase in tumours (TP) and this correlates with tumour neovascularisation and poor prognostic outcome in patients with gastric cancer (Kawahara et al. 2010). This enzyme is capable of modifying DNA released by dead or dying cells by converting the nucleoside, thymidine, into thymine and the potent proANGIOGENIC MOLECULE DEOXYRIBOSE  PHOSPHATE 4AKEN TOGETHER THESE lNDINGS highlight the importance of TAM-derived enzymes in the regulation of tumour neovascularisation. TAMs might also promote the formation of a tumour vasculature in a number of additional ways. These can include the recruitment of CD34+ AC133+ endothelial

20

R. Hughes et al.

NORMOXIA

RECRUITMENT

HYPOXIA ACTIVATION

MONO E/MPC TEM ANGPT2 CXCL12

ANGPT2 CCL2 CSF1 EMAPII

TAM

MIF IL1B TP PDGFB VEGFA CXCL8 bFGF

BIOAVAILABILITY

ECM MODULATION

ECM

bFGF VEGFA MMP-2 MMP-7 MMP-9 MMP-12

ECM

Fig. 2.1 Mechanisms by which TAM promote the formation of the tumour vasculature. TAMs promote the formation of the tumour vasculature by a variety of mechanisms including the recruit OFSUCHPRO ANGIOGENICCELLTYPESASMONOCYTES-/./ AND4)% EXPRESSINGMONOCYTES4%-  4!-S ALSO ACTIVELY RECRUIT ENDOTHELIAL AND MYELOID PROGENITORS %-0#S CAPABLE OF DIRECTLY INCORPORATINGINTOTHETUMOURVASCULATURE&URTHERMORE 4!-SRELEASEAVARIETYOF%#- MODIFYING ENZYMES 4HESE ENHANCE THE ABILITY OF ACTIVATED %# TO MIGRATE AND INVADE THE TUMOUR MICRO ENVIRONMENTBYRESTRUCTURINGTHE%#-!DDITIONALLY THESESAMEMATRIX REMODELLINGENZYMESALSO INCREASETHEBIOAVAILABILITYOFSUCHPRO ANGIOGENICCYTOKINESANDGROWTHFACTORSAS6%'&!AND B&'&VIATHEIRLIBERATIONFROMTHE%#-&INALLY 4!-SALSODIRECTLYPROMOTEANGIOGENESISTHROUGH THE RELEASE OF SUCH PRO ANGIOGENIC CYTOKINES AND GROWTH FACTORS THAT ACT DIRECTLY ON %# THIS induces their activation, migration, proliferation and differentiation into tumour neo-vessels

PROGENITORSOR#$–#$HUMANPERIPHERALBLOODMONOCYTES'ILLETAL2001; Peichev et al. 2000). Indeed, CD14+ CD34− peripheral blood mononuclear cells HAVE BEEN FOUND EXPRESSING THE CLASSIC %# MARKERS 6%'&2 AND V7& (Von Willebrand Factor) (Schmeisser et al. 2001 4HEPRESENCEOF6% CADHERIN6% '&!2'2#$B--0CELLSHASBEENFOUNDINSOLIDTUMOURSWHERETHEY contributed to tumour neovascularisation by becoming incorporated into the tumour ENDOTHELIUM9ANGETAL2004) (Fig. 2.1).

2

Role of Tumour-Associated Macrophages in the Regulation of Angiogenesis

21

Re-education of TAMs by Tumour-Derived Signals The tumour micro-environment is now thought to stimulate many of the proANGIOGENIC FUNCTIONS OF 4!-S &OR EXAMPLE 4.&D secreted by ovarian tumour CELLS ENHANCES 4!- EXPRESSION OF 6%'&! --0  AND OTHER IMPORTANT PRO angiogenic factors (Hagemann et al. 2006). Antibodies released from tumourinfiltrating B cells can also stimulate macrophages to promote tumour angiogenesis (Andreu et al. 2010). As mentioned in the chapter by Cramer and colleagues in this book, tumour HYPOXIA VERY LOW OXYGEN LEVELS ALSO PLAYS AN IMPORTANT PART IN RE EDUCATING MACROPHAGESINTUMOURS4HEOXYGENTENSIONINNORMALTISSUESRANGESFROMTO MM(G(YPOXIA ORLOWOXYGENTENSION DElNEDASANOXYGENTENSIONLESSTHAN 10 mmHg, arises when an imbalance occurs between the supply of O2 and its consumption by local cells. This usually results from the abnormally high proliferation of tumour cells or the presence of a defective vasculature (Shannon et al. 2003; Thews et al. 1998; Vaupel et al. 1989). This leads to cellular dysfunction and ultimately CELLDEATH-OSTTUMOURSWITHADIAMETERGREATERTHANMMCONTAINAREASOFHYPOXIA (Brown, 1979; Hill et al. 1996; Shannon et al. 2003 4HEOXYGENTENSIONSINHEAD and neck tumour metastases were found to be less than 30 mmHg with the median BEINGMM(GINDICATINGTHEPRESENCEOFSEVEREHYPOXIA"ROWN1999). Additional studies have reported similar findings in a variety of other tumour types including prostate (Movsas et al. 1999), cervical (Hockel et al. 1991), breast, brain, head/neck and soft tissue sarcoma (Vaupel et al. 1989; Vaupel et al. 1991). )NTERESTINGLY 4!-S HAVE BEEN SHOWN TO ACCUMULATE IN SUCH HYPOXIC ANDOR necrotic areas of breast, endometrial, ovarian, colorectal and buccal tumours (Lewis and Pollard 2006; Murdoch et al. 2008). Indeed, a positive correlation between the ACCUMULATIONOF4!-SANDTHEPRESENCEOFHYPOXIAHASALSOBEENREPORTEDINTHE liver metastases of breast and colorectal tumours (Stessels et al. 2004). It is thought THATTHESEAREASATTRACT4!-SBYRELEASINGSUCHHYPOXIA INDUCEDCHEMOATTRACTANTS AS6%'&! ENDOTHELINS %-!0))ALSOKNOWNAS3#9% AND#8#,3$&  (Kioi et al. 2010; Murdoch et al. 2004; Murdoch et al. 2008). It might also involve THERELEASEOFFACTORSSUCHASHIGH MOBILITYGROUPBOX(-'" BYNECROTICCELLS INSITESOFCHRONICTUMOURHYPOXIAASTHISHASBEENSHOWNTOALTERTHEPHENOTYPE of macrophages (Andersson et al. 2000; Stros 2010). However, it remains to be seen whether necrotic debris can act as a chemoattractant to recruit monocytes into tumours. )T HAS BEEN SUGGESTED THAT ONCE DRAWN INTO SUCH HYPOXIC SITES 4!-S BECOME TRAPPEDANDUNABLETOLEAVEDUETOTHEIRREDUCEDEXPRESSIONOFCHEMOKINERECEPTORSANDTHEREDUCEDEXPRESSIONOFCHEMOKINESBYTUMOURCELLS.EGUSETAL1998; Sica et al. 2000 %XPOSURETOHYPOXIACANSTIMULATETHEEXPRESSIONOFMACROPHAGE migration inhibitory factor (MIF) by both macrophages and tumour cells (Koong et al. 2000; White et al. 2004). This may inhibit the ability of TAMs to migrate out of HYPOXICSITESINTOOTHERAREASOFTHETUMOUR2ECENTSTUDIESHAVEALSOSHOWNTHE UP REGULATEDEXPRESSIONOF#8#2 THERECEPTORFOR#8#,3$& BYHYPOXIC

22

R. Hughes et al.

macrophages (Fang et al. 2009; Schioppa et al. 2003 ANDTHATHYPOXIA INDUCIBLE transcription factor-1 (HIF-1) influences the positioning and function of TAMs by REGULATINGTHEIREXPRESSIONOF#8#23CHIOPPAETAL2003). Moreover, HIF-1 actiVATIONALSOUP REGULATES#8#,INHYPOXICENDOTHELIALCELLS#ERADINIETAL2004).

Hypoxic Signalling and Regulation of Transcription in TAMs !NUMBEROFRECENTSTUDIESHAVESUGGESTEDTHAT INHYPOXICAREAS 4!-SEXPRESSA highly pro-angiogenic phenotype mainly due to their up-regulation of HIFs-1 and -2 (Burke et al. 2003; Fang et al. 2009). This is known to trigger increased tranSCRIPTION OF MANY ()& TARGET GENES THAT PROMOTE ANGIOGENESIS SUCH AS 6%'&! -)& !.'04AND#8#,&ANGETAL2009). Moreover, we showed that TAMs UP REGULATE THE PRO ANGIOGENIC CHEMOKINE @6%'&! IN AVASCULAR PERI NECROTIC tumours (Lewis et al. 2000). However, there is considerable debate regarding the relative contributions of ()&S AND TOHYPOXIA INDUCEDPRO ANGIOGENICGENEEXPRESSIONINMACROPHAGES Some studies have suggested that HIF-2 is the dominant form of HIF up-regulated BYHYPOXIAINSUCHCELLS'RIFlTHSETAL2000; Talks et al. 2000). Additionally, White ANDCOLLEAGUESUSEDADENOVIRALLYTRANSFECTEDHUMANMACROPHAGESTOOVER EXPRESS HIFs-1 and -2 and showed that HIF-2 rather than HIF-1 appeared to mediate the HYPOXICINDUCTIONOFVARIOUSPRO ANGIOGENICGENESSUCHAS6%'&! ), #/8AND B&'&7HITEETAL2004). However, their data is difficult to interpret as transfected MACROPHAGESEXPRESSEDSUPER PHYSIOLOGICALLEVELSOFTHESETWOTRANSCRIPTIONFACTORS ANDNORMOXICRATHERTHANHYPOXICCULTURESWEREUSEDTHROUGHOUT Other studies have shown that human macrophages accumulate high levels of BOTH()& THAN()& WHENEXPOSEDTOTUMOUR SPECIlCLEVELSOFHYPOXIAINVITRO (Burke et al. 2002). When siRNA was used to knock down the alpha subunit of each ()& THEMAJORITYOFHYPOXIA INDUCEDPRO ANGIOGENICGENESWERESEENTOHAVEEQUAL dependence on both HIFs, suggesting that each HIF was capable of compensating for the loss of the other (Fang et al. 2009). An intriguing recent study by Johnson and colleagues has shown that IL-4 up-regulates HIF-2 but not HIF-1 levels in macrophages and in doing so, skews them towards a more M2-like phenotype. This SUGGESTSTHAT()& MAYPLAYAPARTINMEDIATINGTHEINDUCTIONBYHYPOXIAANDOR ), OFTHE-PHENOTYPEEXPRESSEDBYSOME4!-S4AKEDAETAL2010) and accords WELLWITHTHElNDINGTHATHIGHNUMBERSOF()& EXPRESSING4!-SCORRELATEWITH increased tumour vascularity in breast carcinomas (Leek et al. 2002). Furthermore, THE MACROPHAGE EXPRESSION OF ()&  HAS RECENTLY BEEN FOUND TO IMPACT ON THE progression of murine hepatocelullar carcinomas (HCC) – as evidenced by their reduced growth in myeloid-specific HIF-2D conditional knock-out mice. Moreover, TUMOURCELLSEXHIBITEDALOWERMITOTICINDEXINTHESETUMOURS)MTIYAZETAL2010). &INALLY ITSHOULDBENOTEDTHAT%LBARGHATIANDCOLLEAGUESSHOWEDTHAT()&S AND  MAYBEDIFFERENTIALLYREGULATEDDURINGRE OXYGENATIONOFMACROPHAGESAFTERTHECESSATIONOFHYPOXIAFOREXAMPLE INHYPOXICTUMOURAREASUNDERGOINGREVASCULARIZATION 

2

Role of Tumour-Associated Macrophages in the Regulation of Angiogenesis

23

()&  UP REGULATION WAS RAPIDLY REVERSED DURING RE OXYGENATION WHEREAS ()&  LEVELSTOOKLONGERTODROPTONORMOXICLEVELS%LBARGHATIETAL2008). Recently, the activation of another transcription factor, NF-NB, has also been IMPLICATED IN THE TRANSCRIPTIONAL REGULATION OF HYPOXIC MACROPHAGES 3HORT TERM EXPOSUREnH OFMURINEMACROPHAGESTOHYPOXIAACTIVATES.& NB, which in turn up-regulates HIF-1D levels (Rius et al. 2008). This study demonstrated the impaired ability of IKKE−/− macrophages to up-regulate HIF-1DINTHEPRESENCEOFHYPOXIA 2ECENTDATAFROMOURGROUPHASNOWSHOWNTHATALTHOUGHMACROPHAGESEXPOSEDTO MOREPROLONGEDPERIODSOFHYPOXIAEGH UP REGULATETHEEXPRESSIONOFSUCH NF-NB members as p65, p50 and IKKE, inhibition of canonical NF-NB signalling HADNOEFFECTONTHEIRUP REGULATIONOFVARIOUSGENESINHYPOXIA&ANGETAL2009). Together these findings suggest that NF-NB may be more important in the initial, EARLYRESPONSESOFMACROPHAGESTOTUMOURHYPOXIATHANWHENTHEYEXPERIENCEITFOR longer periods in tumours.

TIE2-Expressing Monocyte/Macrophages A subset of human and murine monocyte/macrophages have been identified that DISPLAY HIGHLY PRO ANGIOGENIC FUNCTIONS n THOSE THAT EXPRESS THE TYROSINE KINASE WITH)GAND%'&HOMOLOGYDOMAIN 4)% RECEPTOR WHICHISTYPICALLYEXPRESSED on endothelial cells, (De Palma et al. 2007; De Palma et al. 2005; De Palma et al. 2003; Murdoch et al. 2007; Nowak et al. 2004 ANDAREKNOWNAS4)% EXPRESSING MONOCYTEMACROPHAGES 4%-S  4%-S HAVE BEEN FOUND IN A VARIETY OF HUMAN tumours including kidney, colon, pancreas, lung and in soft tissue sarcomas (Venneri et al. 2007). 4HERECRUITMENTOF4%-SINTOSOLIDTUMOURSISTHOUGHTTOBEVIAMECHANISMS THATAREDISTINCTFROMTHOSEUSEDBY4)%n4!-S!SMENTIONEDPREVIOUSLY -#0  (CCL2) plays a prominent role in the recruitment of monocytes from the peripheral BLOODINTOSOLIDTUMOURS APROCESSMEDIATEDBYTHEEXPRESSIONOF##2 THERECEPTOR FOR##, ONTHEMAJORITYOFCIRCULATINGMONOCYTES4%-SDONOTEXPRESS##2 (Venneri et al. 2007) and therefore CCL2 appears not to play a significant role in THEIR RECRUITMENT )NTERESTINGLY WE AND OTHERS HAVE SHOWN THAT 4%-S COULD BE RECRUITED INTO TUMOURS BY THE 4)% LIGAND ANGIOPOIETIN  !.'04 'U ET AL 2006; Murdoch et al. 2007; Stratmann et al. 1998; Venneri et al. 2007), a cytokine UP REGULATED BY ENDOTHELIAL CELLS IN TUMOUR BLOOD VESSELS !.'04 IS TYPICALLY stored in Weibel-Palade bodies in these cells, secretory granules that rapidly fuse with the cell membrane upon endothelial cell activation, and release stored cytokines and GROWTHFACTORS INCLUDING!.'04 INTOTHEEXTRACELLULARSPACE&IEDLERETAL2004). 4%-SAPPEARTOBEINHERENTLYBETTERATPROMOTINGTHEFORMATIONOFTUMOURBLOOD VESSELSTHAN4!-S!CCORDINGLY CIRCULATING4%-SWEREFOUNDTOEXPRESSHIGHER LEVELS OF IMPORTANT PRO ANGIOGENIC MEDIATORS SUCH AS 6%'&! --0 #/8 40 AND #ATHEPSIN " #ONSISTENT WITH THE ELEVATED EXPRESSION OF PRO ANGIOGENIC EFFECTORMOLECULES 4%-SPROMOTEDANGIOGENESISTOAGREATEREXTENTINBOTHINVITRO endothelial cell and murine in vivo tumour assays (Coffelt et al. 2010).

24

R. Hughes et al.

When human U87 glioma cells were co-injected subcutaneously into nude mice WITH4%-S ASIGNIlCANTLYHIGHERLEVELOFTUMOURVASCULARISATIONWASSEENCOMPARED WITHMICERECEIVINGNON 4%-S$E0ALMAETAL2005). Moreover, De Palma and COLLEAGUESUSEDACONDITIONALGENETIC4%-DEPLETIONMODELTOSHOWTHEESSENTIAL pro-angiogenic effect of these cells in murine tumours. To do this, sub-lethally irradiated mice were implanted with orthotopic human glioma cells and their bone MARROWRECONSTITUTEDFROMAMURINETRANSGENICSTRAININWHICHTHEEXPRESSIONOFTHE thymidine kinase gene was under the control of the Tie2 promoter. When the pro-drug, ganciclovir, was then administered to tumour-bearing mice to specifically ablate the 4%-POPULATION ITRESULTEDINAMARKEDREDUCTIONINBOTHTUMOURANGIOGENESISAND growth (De Palma et al. 2005). Taken together, these data clearly demonstrate the ROLETHAT4%-SPLAYINREGULATINGTUMOURANGIOGENESIS

Concluding Remarks The findings discussed in this chapter highlight the multifaceted way in which TAMs stimulate tumour angiogenesis. These highly versatile cells up-regulate a plethora of cytokines and growth factors in response to such micro-environmental SIGNALSASCYTOKINESSECRETEDBYTUMOURCELLSANDHYPOXIA4HESEFACTORSEITHERACTIvate endothelial cells directly in the tumour vasculature or indirectly via a variety OF%#- REMODELLINGENZYMES SUCHASVARIOUS--0STHATRE SHAPETHE%#-TO FACILITATE%#INVASIONANDMIGRATIONANDINCREASETHEBIOAVAILABILITYOFVARIOUSPRO angiogenic mediators. Taken together, these findings indicate that the development of therapeutic strategies aimed at re-directing the phenotype of macrophages back towards more M1-like, anti-tumour functions in vivo might prove efficacious in the treatment of cancer. Acknowledgements The authors gratefully acknowledge the support of Cancer research UK, the "REAST#ANCER#AMPAIGNAND9ORKSHIRE#ANCER2ESEARCHFORTHEIRWORKINTHISAREAOFRESEARCH

References !HARINEJAD3 0AULUS0 3IOUD- (OFMANN- :INS+ 3CHAFER2 3TANLEY%2 !BRAHAM$ Colony-stimulating factor-1 blockade by antisense oligonucleotides and small interfering 2.!S SUPPRESSES GROWTH OF HUMAN MAMMARY TUMOR XENOGRAFTS IN MICE #ANCER 2ES  5378–5384 !NDERSSON5 7ANG( 0ALMBLAD+ !VEBERGER!# "LOOM/ %RLANDSSON (ARRIS( *ANSON! +OKKOLA2 :HANG- 9ANG(ETAL (IGHMOBILITYGROUPPROTEIN(-'  STIMULATES PROINmAMMATORYCYTOKINESYNTHESISINHUMANMONOCYTES*%XP-EDn Andreu P, Johansson M, Affara NI, Pucci F, Tan T, Junankar S, Korets L, Lam J, Tawfik D, $E.ARDO$'ETAL &C2GAMMAACTIVATIONREGULATESINmAMMATION ASSOCIATEDSQUAMOUS carcinogenesis. Cancer Cell 17:121–134

2

Role of Tumour-Associated Macrophages in the Regulation of Angiogenesis

25

"AILEY# .EGUS2 -ORRIS! :IPRIN0 'OLDIN2 !LLAVENA0 0ECK$ $ARZI! #HEMOKINE EXPRESSION IS ASSOCIATED WITH THE ACCUMULATION OF TUMOUR ASSOCIATED MACROPHAGES 4!-S ANDPROGRESSIONINHUMANCOLORECTALCANCER#LIN%XP-ETASTASISn Balkwill F, Mantovani A (2001) Inflammation and cancer: back to Virchow? Lancet 357:539–545 Barleon B, Sozzani S, Zhou D, Weich HA, Mantovani A, Marme D (1996) Migration of human MONOCYTESINRESPONSETOVASCULARENDOTHELIALGROWTHFACTOR6%'& ISMEDIATEDVIATHE6%'& receptor flt-1. Blood 87:3336–3343 "INGLE , "ROWN .* ,EWIS #%  4HE ROLE OF TUMOUR ASSOCIATED MACROPHAGES IN TUMOUR progression: implications for new anticancer therapies. J Pathol 196:254–265 "INGLE, ,EWIS#% #ORKE+0 2EED-7 "ROWN.* -ACROPHAGESPROMOTEANGIOGENESIS in human breast tumour spheroids in vivo. Br J Cancer 94:101–107 "ISWAS3+ 'ANGI, 0AUL3 3CHIOPPA4 3ACCANI! 3IRONI- "OTTAZZI" $ONI! 6INCENZO" 0ASQUALINI&ETAL !DISTINCTANDUNIQUETRANSCRIPTIONALPROGRAMEXPRESSEDBYTUMOR associated macrophages (defective NF-kappaB and enhanced IRF-3/STAT1 activation). Blood 107:2112–2122 "ISWAS3+ 3ICA! ,EWIS#% 0LASTICITYOFMACROPHAGEFUNCTIONDURINGTUMORPROGRESSION regulation by distinct molecular mechanisms. J Immunol 180:2011–2017 "OLAT & +AYASELCUK & .URSAL 4: 9AGMURDUR -# "AL . $EMIRHAN "  -ICROVESSEL DENSITY 6%'&EXPRESSION ANDTUMOR ASSOCIATEDMACROPHAGESINBREASTTUMORSCORRELATIONS WITHPROGNOSTICPARAMETERS*%XP#LIN#ANCER2ESn "ROWN*- %VIDENCEFORACUTELYHYPOXICCELLSINMOUSETUMOURS ANDAPOSSIBLEMECHANISM OFREOXYGENATION"R*2ADIOLn "ROWN*- 4HEHYPOXICCELLATARGETFORSELECTIVECANCERTHERAPYnEIGHTEENTH"RUCE& Cain Memorial Award lecture. Cancer Res 59:5863–5870 "URKE " 4ANG . #ORKE +0 4AZZYMAN $ !MERI + 7ELLS - ,EWIS #%  %XPRESSION OF()& ALPHABYHUMANMACROPHAGESIMPLICATIONSFORTHEUSEOFMACROPHAGESINHYPOXIA regulated cancer gene therapy. J Pathol 196:204–212 "URKE " 'IANNOUDIS ! #ORKE +0 'ILL $ 7ELLS - :IEGLER (EITBROCK , ,EWIS #%  (YPOXIA INDUCEDGENEEXPRESSIONINHUMANMACROPHAGESIMPLICATIONSFORISCHEMICTISSUES ANDHYPOXIA REGULATEDGENETHERAPY!M*0ATHOLn #ERADINI $* +ULKARNI !2 #ALLAGHAN -* 4EPPER /- "ASTIDAS . +LEINMAN -% #APLA *- 'ALIANO2$ ,EVINE*0 'URTNER'# 0ROGENITORCELLTRAFlCKINGISREGULATEDBYHYPOXIC gradients through HIF-1 induction of SDF-1. Nat Med 10:858–864 Coffelt SB, Tal AO, Scholz A, De Palma M, Patel S, Urbich C, Biswas SK, Murdoch C, Plate KH, 2EISS9ETAL !NGIOPOIETIN REGULATESGENEEXPRESSIONIN4)% EXPRESSINGMONOCYTES and augments their inherent proangiogenic functions. Cancer Res 70:5270–5280 Coussens LM, Tinkle CL, Hanahan D, Werb Z (2000) MMP-9 supplied by bone marrow-derived cells contributes to skin carcinogenesis. Cell 103:481–490 $E0ALMA- 6ENNERI-! 2OCA# .ALDINI, 4ARGETINGEXOGENOUSGENESTOTUMORANGIOgenesis by transplantation of genetically modified hematopoietic stem cells. Nat Med 9: 789–795 $E0ALMA- 6ENNERI-! 'ALLI2 3ERGI3ERGI, 0OLITI,3 3AMPAOLESI- .ALDINI, 4IE identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitors. Cancer Cell 8:211–226 $E0ALMA- -URDOCH# 6ENNERI-! .ALDINI, ,EWIS#% 4IE EXPRESSINGMONOCYTES regulation of tumor angiogenesis and therapeutic implications. Trends Immunol 28:519–524 Dinapoli MR, Calderon CL, Lopez DM (1996) The altered tumoricidal capacity of macrophages ISOLATEDFROMTUMOR BEARINGMICEISRELATEDTOREDUCEEXPRESSIONOFTHEINDUCIBLENITRICOXIDE SYNTHASEGENE*%XP-EDn $U2 ,U+6 0ETRITSCH# ,IU0 'ANSS2 0ASSEGUE% 3ONG( 6ANDENBERG3 *OHNSON23 7ERB: et al (2008) HIF1alpha induces the recruitment of bone marrow-derived vascular modulatory cells to regulate tumor angiogenesis and invasion. Cancer Cell 13:206–220 %LBARGHATI, -URDOCH# ,EWIS#% %FFECTSOFHYPOXIAONTRANSCRIPTIONFACTOREXPRESSION in human monocytes and macrophages. Immunobiology 213:899–908

26

R. Hughes et al.

&ANG(9 (UGHES2 -URDOCH# #OFFELT3" "ISWAS3+ (ARRIS!, *OHNSON23 )MITYAZ(: 3IMON-# &REDLUND%ETAL (YPOXIA INDUCIBLEFACTORSANDAREIMPORTANTTRANSCRIPTIONALEFFECTORSINPRIMARYMACROPHAGESEXPERIENCINGHYPOXIA"LOODn &IEDLER5 3CHARPFENECKER- +OIDL3 (EGEN! 'RUNOW6 3CHMIDT*- +RIZ7 4HURSTON' !UGUSTIN(' 4HE4IE LIGANDANGIOPOIETIN ISSTOREDINANDRAPIDLYRELEASEDUPON stimulation from endothelial cell Weibel-Palade bodies. Blood 103:4150–4156 'ILL- $IAS3 (ATTORI+ 2IVERA-, (ICKLIN$ 7ITTE, 'IRARDI, 9URT2 (IMEL( 2AlI3  6ASCULAR TRAUMA INDUCES RAPID BUT TRANSIENT MOBILIZATION OF 6%'&2 !# endothelial precursor cells. Circ Res 88:167–174 'IRAUDO % )NOUE - (ANAHAN $  !N AMINO BISPHOSPHONATE TARGETS --0  EXPRESSING macrophages and angiogenesis to impair cervical carcinogenesis. J Clin Invest 114:623–633 'OERDT3 /RFANOS#% /THERFUNCTIONS OTHERGENESALTERNATIVEACTIVATIONOFANTIGEN PREsenting cells. Immunity 10:137–142 'ORDON3 !LTERNATIVEACTIVATIONOFMACROPHAGES.AT2EV)MMUNOLn 'RIFlTHS, "INLEY+ )QBALL3 +AN/ -AXWELL0 2ATCLIFFE0 ,EWIS# (ARRIS! +INGSMAN3 Naylor S (2000) The macrophage – a novel system to deliver gene therapy to pathological HYPOXIA'ENE4HERn 'U * 9AMAMOTO ( /GAWA - .GAN #9 $ANNO + (EMMI ( +YO . 4AKEMASA ) )KEDA - 3EKIMOTO-ETAL (YPOXIA INDUCEDUP REGULATIONOFANGIOPOIETIN INCOLORECTALCANCER Oncol Rep 15:779–783 (AGEMANN4 7ILSON* "URKE& +ULBE( ,I.& 0LUDDEMANN! #HARLES+ 'ORDON3 "ALKWILL&2 (2006) Ovarian cancer cells polarize macrophages toward a tumor-associated phenotype. J Immunol 176:5023–5032 Hildenbrand R, Dilger I, Horlin A, Stutte HJ (1995) Urokinase and macrophages in tumour angiogenesis. Br J Cancer 72:818–823 (ILL3! 0IGOTT+( 3AUNDERS-) 0OWELL-% !RNOLD3 /BEID! 7ARD' ,EAHY- (OSKIN0* Chaplin DJ (1996) Microregional blood flow in murine and human tumours assessed using laser Doppler microprobes. Br J Cancer Suppl 27:S260–263 (OCKEL - 3CHLENGER + +NOOP # 6AUPEL 0  /XYGENATION OF CARCINOMAS OF THE UTERINE CERVIXEVALUATIONBYCOMPUTERIZED/TENSIONMEASUREMENTS#ANCER2ESn (OSHINO 9 (ATAKE + +ASAHARA 4 4AKAHASHI 9 )KEDA - 4OMIZUKA ( /HTSUKI 4 5WAI - Mukaida N, Matsushima K et al (1995) Monocyte chemoattractant protein-1 stimulates tumor necrosis and recruitment of macrophages into tumors in tumor-bearing nude mice: increased GRANULOCYTE AND MACROPHAGE PROGENITORS IN MURINE BONE MARROW %XP (EMATOL  1035–1039 (OUGHTON !- 'RISOLANO *, "AUMANN -, +OBAYASHI $+ (AUTAMAKI 2$ .EHRING ,# #ORNELIUS ,! 3HAPIRO 3$  -ACROPHAGE ELASTASE MATRIX METALLOPROTEINASE  suppresses growth of lung metastases. Cancer Res 66:6149–6155 Huang S, Van Arsdall M, Tedjarati S, McCarty M, Wu W, Langley R, Fidler IJ (2002) Contributions of stromal metalloproteinase-9 to angiogenesis and growth of human ovarian carcinoma in mice. J Natl Cancer Inst 94:1134–1142 )MTIYAZ(: 7ILLIAMS%0 (ICKEY-- 0ATEL3! $URHAM!# 9UAN,* (AMMOND2 'IMOTTY0! +EITH" 3IMON-# (YPOXIA INDUCIBLEFACTORALPHAREGULATESMACROPHAGEFUNCTION in mouse models of acute and tumor inflammation. J Clin Invest 120:2699–2714 )TOH4 4ANIOKA- 9OSHIDA( 9OSHIOKA4 .ISHIMOTO( )TOHARA3 2EDUCEDANGIOGENESIS and tumor progression in gelatinase A-deficient mice. Cancer Res 58:1048–1051 +AWAHARA! (ATTORI3 !KIBA* .AKASHIMA+ 4AIRA4 7ATARI+ (OSOI& 5BA- "ASAKI9 Koufuji K et al (2010) Infiltration of thymidine phosphorylase-positive macrophages is closely associated with tumor angiogenesis and survival in intestinal type gastric cancer. Oncol Rep 24:405–415 +IOI- 6OGEL( 3CHULTZ' (OFFMAN2- (ARSH'2 "ROWN*- )NHIBITIONOFVASCULOgenesis, but not angiogenesis, prevents the recurrence of glioblastoma after irradiation in mice. J Clin Invest 120:694–705

2

Role of Tumour-Associated Macrophages in the Regulation of Angiogenesis

27

Koide N, Nishio A, Sato T, Sugiyama A, Miyagawa S (2004) Significance of macrophage CHEMOATTRACTANTPROTEIN EXPRESSIONANDMACROPHAGEINlLTRATIONINSQUAMOUSCELLCARCINOMA OFTHEESOPHAGUS!M*'ASTROENTEROLn +OONG!# $ENKO.# (UDSON+- 3CHINDLER# 3WIERSZ, +OCH# %VANS3 )BRAHIM( ,E14 4ERRIS $* ET AL  #ANDIDATE GENES FOR THE HYPOXIC TUMOR PHENOTYPE #ANCER 2ES  883–887 Langowski JL, Zhang X, Wu L, Mattson JD, Chen T, Smith K, Basham B, McClanahan T, Kastelein RA, Oft M (2006) IL-23 promotes tumour incidence and growth. Nature 442: 461–465 ,EEK 2$ ,EWIS #% 7HITEHOUSE 2 'REENALL - #LARKE * (ARRIS !,  !SSOCIATION OF macrophage infiltration with angiogenesis and prognosis in invasive breast carcinoma. Cancer Res 56:4625–4629 ,EEK2$ 4ALKS+, 0EZZELLA& 4URLEY( #AMPO, "ROWN.3 "ICKNELL2 4AYLOR- 'ATTER+# (ARRIS !,  2ELATION OF HYPOXIA INDUCIBLE FACTOR  ALPHA ()&  ALPHA EXPRESSION IN TUMOR INlLTRATIVEMACROPHAGESTOTUMORANGIOGENESISANDTHEOXIDATIVETHYMIDINEPHOSPHOrylase pathway in Human breast cancer. Cancer Res 62:1326–1329 ,EUNG39 7ONG-0 #HUNG,0 #HAN!3 9UEN34 -ONOCYTECHEMOATTRACTANTPROTEIN  EXPRESSIONANDMACROPHAGEINlLTRATIONINGLIOMAS!CTA.EUROPATHOLn ,EWIS#% 0OLLARD*7 $ISTINCTROLEOFMACROPHAGESINDIFFERENTTUMORMICROENVIRONMENTS Cancer Res 66:605–612 ,EWIS *3 ,ANDERS 2* 5NDERWOOD *# (ARRIS !, ,EWIS #%  %XPRESSION OF VASCULAR endothelial growth factor by macrophages is up-regulated in poorly vascularized areas of breast carcinomas. J Pathol 192:150–158 ,IN%9 .GUYEN!6 2USSELL2' 0OLLARD*7 #OLONY STIMULATINGFACTORPROMOTESPROGRESSIONOFMAMMARYTUMORSTOMALIGNANCY*%XP-EDn ,IN%9 ,I*& 'NATOVSKIY, $ENG9 :HU, 'RZESIK$! 1IAN( 8UE8. 0OLLARD*7 Macrophages regulate the angiogenic switch in a mouse model of breast cancer. Cancer Res 66: 11238–11246 ,UO9 :HOU( +RUEGER* +APLAN# ,EE3( $OLMAN# -ARKOWITZ$ 7U7 ,IU# 2EISFELD RA et al (2006) Targeting tumor-associated macrophages as a novel strategy against breast cancer. J Clin Invest 116:2132–2141 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 Mantovani A, Schioppa T, Biswas SK, Marchesi F, Allavena P, Sica A (2003) Tumor-associated macrophages and dendritic cells as prototypic type II polarized myeloid populations. Tumori 89:459–468 Mantovani A, Sica A, Sozzani S, Allavena P, Vecchi A, Locati M (2004) The chemokine system in diverse forms of macrophage activation and polarization. Trends Immunol 25:677–686 -ARTINEZ&/ (ELMING, 'ORDON3 !LTERNATIVEACTIVATIONOFMACROPHAGESANIMMUNOLOGIC functional perspective. Annu Rev Immunol 27:451–483 Mills CD, Kincaid K, Alt JM, Heilman MJ, Hill AM (2000) M-1/M-2 macrophages and the Th1/Th2 paradigm. J Immunol 164:6166–6173 -OVSAS" #HAPMAN*$ (ORWITZ%- 0INOVER7( 'REENBERG2% (ANLON!, )YER2 (ANKS'%  (YPOXICREGIONSEXISTINHUMANPROSTATECARCINOMA5ROLOGYn -URDOCH# 'IANNOUDIS! ,EWIS#% -ECHANISMSREGULATINGTHERECRUITMENTOFMACROPHAGES INTOHYPOXICAREASOFTUMORSANDOTHERISCHEMICTISSUES"LOODn -URDOCH# 4AZZYMAN3 7EBSTER3 ,EWIS#% %XPRESSIONOF4IE BYHUMANMONOCYTES and their responses to angiopoietin-2. J Immunol 178:7405–7411 -URDOCH# -UTHANA- #OFFELT3" ,EWIS#% 4HEROLEOFMYELOIDCELLSINTHEPROMOTION of tumour angiogenesis. Nat Rev Cancer 8:618–631 .AKAO3 +UWANO4 4SUTSUMI -IYAHARA# 5EDA3 +IMURA9. (AMANO3 3ONODA+( 3AIJO9 .UKIWA 4 3TRIETER 2- ET AL  )NlLTRATION OF #/8  EXPRESSING MACROPHAGES IS A

28

R. Hughes et al.

prerequisite for IL-1 beta-induced neovascularization and tumor growth. J Clin Invest 115: 2979–2991 .EGUS20 4URNER, "URKE& "ALKWILL&2 (YPOXIADOWN REGULATES-#0 EXPRESSION implications for macrophage distribution in tumors. J Leukoc Biol 63:758–765 .OWAK' +ARRAR! (OLMEN# .AVA3 5ZUNEL- (ULTENBY+ 3UMITRAN (OLGERSSON3 %XPRESSIONOFVASCULARENDOTHELIALGROWTHFACTORRECEPTOR OR4IE ONPERIPHERALBLOODCELLS defines functionally competent cell populations capable of reendothelialization. Circulation 110:3699–3707 Nozawa H, Chiu C, Hanahan D (2006) Infiltrating neutrophils mediate the initial angiogenic switch in a mouse model of multistage carcinogenesis. Proc Natl Acad Sci USA 103: 12493–12498 /HNO3 /HNO9 3UZUKI. +AMEI4 +OIKE+ )NAGAWA( +OHCHI# 3OMA' )NOUE- Correlation of histological localization of tumor-associated macrophages with clinicopathological features in endometrial cancer. Anticancer Res 24:3335–3342 /HTA- +ITADAI9 4ANAKA3 9OSHIHARA- 9ASUI7 -UKAIDA. (ARUMA+ #HAYAMA+ -ONOCYTE CHEMOATTRACTANT PROTEIN  EXPRESSION CORRELATES WITH MACROPHAGE INlLTRATION AND tumor vascularity in human gastric carcinomas. Int J Oncol 22:773–778 0AULUS0 3TANLEY%2 3CHAFER2 !BRAHAM$ !HARINEJAD3 #OLONY STIMULATINGFACTOR  ANTIBODYREVERSESCHEMORESISTANCEINHUMAN-#& BREASTCANCERXENOGRAFTS#ANCER2ES 4349–4356 Peichev M, Naiyer AJ, Pereira D, Zhu Z, Lane WJ, Williams M, Oz MC, Hicklin DJ, Witte L, -OORE-!ETAL %XPRESSIONOF6%'&2 AND!#BYCIRCULATINGHUMAN#$ cells identifies a population of functional endothelial precursors. Blood 95:952–958 Pertovaara L, Kaipainen A, Mustonen T, Orpana A, Ferrara N, Saksela O, Alitalo K (1994) Vascular endothelial growth factor is induced in response to transforming growth factor-beta in fibroblastic and epithelial cells. J Biol Chem 269:6271–6274 2IUS* 'UMA- 3CHACHTRUP# !KASSOGLOU+ :INKERNAGEL!3 .IZET6 *OHNSON23 (ADDAD '' +ARIN- .& KAPPA"LINKSINNATEIMMUNITYTOTHEHYPOXICRESPONSETHROUGHTRANscriptional regulation of HIF-1alpha. Nature 453:807–811 3AJI( +OIKE- 9AMORI4 3AJI3 3EIKI- -ATSUSHIMA+ 4OI- 3IGNIlCANTCORRELATION OFMONOCYTECHEMOATTRACTANTPROTEIN EXPRESSIONWITHNEOVASCULARIZATIONANDPROGRESSIONOF breast carcinoma. Cancer 92:1085–1091 Schioppa T, Uranchimeg B, Saccani A, Biswas SK, Doni A, Rapisarda A, Bernasconi S, Saccani S, .EBULONI- 6AGO,ETAL 2EGULATIONOFTHECHEMOKINERECEPTOR#8#2BYHYPOXIA *%XP-EDn 3CHMEISSER! 'ARLICHS#$ :HANG( %SKAl3 'RAFFY# ,UDWIG* 3TRASSER2( $ANIEL7'  -ONOCYTES COEXPRESS ENDOTHELIAL AND MACROPHAGOCYTIC LINEAGE MARKERS AND FORM cord-like structures in Matrigel under angiogenic conditions. Cardiovasc Res 49:671–680 3HANNON !- "OUCHIER (AYES $* #ONDRON #- 4OOMEY $  4UMOUR HYPOXIA CHEMOTHERAPEUTICRESISTANCEANDHYPOXIA RELATEDTHERAPIES#ANCER4REAT2EVn 3ICA! 3ACCANI! "OTTAZZI" "ERNASCONI3 !LLAVENA0 'AETANO" &EI& ,A2OSA' 3COTTON# "ALKWILL&ETAL $EFECTIVEEXPRESSIONOFTHEMONOCYTECHEMOTACTICPROTEIN RECEPTOR CCR2 in macrophages associated with human ovarian carcinoma. J Immunol 164:733–738 Sica A, Allavena P, Mantovani A (2008) Cancer related inflammation: The macrophage connection. Cancer Lett 267:204–215 3ICILIANO .! 3KINNER *! 9UK -(  "ORDETELLA BRONCHISEPTICA MODULATES MACROPHAGE phenotype leading to the inhibition of CD4+ T cell proliferation and the initiation of a Th17 immune response. J Immunol 177:7131–7138 3TESSELS & 6AN DEN %YNDEN ' 6AN DER !UWERA ) 3ALGADO 2 6AN DEN (EUVEL % (ARRIS !, *ACKSON$' #OLPAERT#' VAN-ARCK%! $IRIX,9ETAL "REASTADENOCARCINOMALIVER metastases, in contrast to colorectal cancer liver metastases, display a non-angiogenic growth PATTERNTHATPRESERVESTHESTROMAANDLACKSHYPOXIA"R*#ANCERn 3TRATMANN ! 2ISAU 7 0LATE +(  #ELL TYPE SPECIlC EXPRESSION OF ANGIOPOIETIN  AND angiopoietin-2 suggests a role in glioblastoma angiogenesis. Am J Pathol 153:1459–1466

2

Role of Tumour-Associated Macrophages in the Regulation of Angiogenesis

29

3TROS- (-'"PROTEINSINTERACTIONSWITH$.!ANDCHROMATIN"IOCHIM"IOPHYS!CTA 1799:101–113 4AKEDA. /$EA%, $OEDENS! +IM*7 7EIDEMANN! 3TOCKMANN# !SAGIRI- 3IMON-# Hoffmann A, Johnson RS (2010) Differential activation and antagonistic function of HIF-{alpha} ISOFORMSINMACROPHAGESAREESSENTIALFOR./HOMEOSTASIS'ENES$EVn 4ALKS +, 4URLEY ( 'ATTER +# -AXWELL 0( 0UGH #7 2ATCLIFFE 0* (ARRIS !,  4HE EXPRESSION AND DISTRIBUTION OF THE HYPOXIA INDUCIBLE FACTORS ()& ALPHA AND ()& ALPHA IN normal human tissues, cancers, and tumor-associated macrophages. Am J Pathol 157: 411–421 4HEWS / +ELLEHER $+ ,ECHER " 6AUPEL 0  "LOOD mOW OXYGENATION METABOLIC AND energetic status in different clonal subpopulations of a rat rhabdomyosarcoma. Int J Oncol 13: 205–211 4SUTSUI3 9ASUDA+ 3UZUKI+ 4AHARA+ (IGASHI( %RA3 -ACROPHAGEINlLTRATIONANDITS PROGNOSTICIMPLICATIONSINBREASTCANCERTHERELATIONSHIPWITH6%'&EXPRESSIONANDMICROVESSEL density. Oncol Rep 14:425–431 6AN'INDERACHTER*! -OVAHEDI+ (ASSANZADEH'HASSABEH' -EERSCHAUT3 "ESCHIN! 2AES' De Baetselier P (2006) Classical and alternative activation of mononuclear phagocytes: picking the best of both worlds for tumor promotion. Immunobiology 211:487–501 6AUPEL0 +ALLINOWSKI& /KUNIEFF0 "LOODmOW OXYGENANDNUTRIENTSUPPLY ANDMETABOLIC microenvironment of human tumors: a review. Cancer Res 49:6449–6465 6AUPEL0 3CHLENGER+ +NOOP# (OCKEL- /XYGENATIONOFHUMANTUMORSEVALUATIONOF TISSUEOXYGENDISTRIBUTIONINBREASTCANCERSBYCOMPUTERIZED/TENSIONMEASUREMENTS#ANCER Res 51:3316–3322 6ENNERI-! $E0ALMA- 0ONZONI- 0UCCI& 3CIELZO# :ONARI% -AZZIERI2 $OGLIONI# .ALDINI ,  )DENTIlCATION OF PROANGIOGENIC 4)% EXPRESSING MONOCYTES 4%-S IN human peripheral blood and cancer. Blood 109:5276–5285 6ORONOV% 3HOUVAL$3 +RELIN9 #AGNANO% "ENHARROCH$ )WAKURA9 $INARELLO#! !PTE2. (2003) IL-1 is required for tumor invasiveness and angiogenesis. Proc Natl Acad Sci USA 100:2645–2650 7HITE*2 (ARRIS2! ,EE32 #RAIGON-( "INLEY+ 0RICE4 "EARD', -UNDY#2 .AYLOR3  'ENETICAMPLIlCATIONOFTHETRANSCRIPTIONALRESPONSETOHYPOXIAASANOVELMEANSOF IDENTIFYINGREGULATORSOFANGIOGENESIS'ENOMICSn Wilson CL, Heppner KJ, Labosky PA, Hogan BL, Matrisian LM (1997) Intestinal tumorigenesis is suppressed in mice lacking the metalloproteinase matrilysin. Proc Natl Acad Sci USA 94: 1402–1407 9ANG, $E"USK,- &UKUDA+ &INGLETON" 'REEN *ARVIS" 3HYR9 -ATRISIAN,- #ARBONE$0 ,IN0# %XPANSIONOFMYELOIDIMMUNESUPPRESSOR'R#$BCELLSINTUMOR BEARING host directly promotes tumor angiogenesis. Cancer Cell 6:409–421 9OSHIDA3 /NO- 3HONO4 )ZUMI( )SHIBASHI4 3UZUKI( +UWANO- )NVOLVEMENTOF interleukin-8, vascular endothelial growth factor, and basic fibroblast growth factor in tumor necrosis factor alpha-dependent angiogenesis. Mol Cell Biol 17:4015–4023 Zeisberger SM, Odermatt B, Marty C, Zehnder-Fjallman AH, Ballmer-Hofer K, Schwendener RA (2006) Clodronate-liposome-mediated depletion of tumour-associated macrophages: a new and highly effective antiangiogenic therapy approach. Br J Cancer 95:272–281 :HU9- 7EBSTER3* &LOWER$ 7OLL0* )NTERLEUKIN #8#,ISAGROWTHFACTORFORHUMAN lung cancer cells. Br J Cancer 91:1970–1976

Chapter 3

The Role of Tumour-Associated Macrophages in Malignant Invasion Claudia Binder

Introduction Infiltrates of leukocytes in various cancers were first observed more than 100 years ago and were interpreted as a defense mechanism of the host’s immune system. Attempts of clinicians such as Sir William Coley to eliminate malignant tumours by stimulating inflammation, however, were not reproducibly successful, not least because of uncontrollable side effects (Nauts et al. 1946). That there is also another side of the coin has long been known among pathologists, who observed high amounts of infiltrating leukocytes, particularly, in clinically unfavorable entities, such as inflammatory breast cancer. Nevertheless, it has only gradually become realized that these cells, in contrast to their expected cytotoxic function, can foster tumour promotion and progression (Steele et al. 1984; Leek et al. 1996). During the last two decades, evidence has accumulated that especially the tumour-associated macrophages (TAM), representing the main population of these infiltrates (Mantovani et al. 1992), are key actors in this dubious play. Among the various ways, TAM can promote malignant progression, we will here focus on their effect on tumour cell invasion. Movement of cells through 3D environments during invasion usually requires enhanced motile capacity as well as reorganization of the invaded structures through pericellular proteolysis. To acquire single cell motility, cells run through a program that involves cell polarization, loosening of cell–cell contacts, establishment of focal adhesions to the substrate as well as exertion of contractile force via actin-binding proteins, such as myosin, at the trailing edge (Friedl 2004; Kedrin et al. 2007) This is associated with extensive actin remodeling and development of various kinds of cellular protrusions at the leading edge, such as finger-like filopodia with

C. Binder (*) Department of Hematology/Oncology, University of Göttingen, Göttingen 37099, Germany e-mail: [email protected] T. Lawrence and T. Hagemann (eds.), Tumour-Associated Macrophages, DOI 10.1007/978-1-4614-0662-4_3, © Springer Science+Business Media, LLC 2012

31

32

C. Binder

thin actin bundles, important for chemotaxis, and sheet-like lamellipodia, containing broad actin meshworks. A further type of protrusions is found in malignant cells, which use invadopodia to penetrate the surrounding extracellular matrix (ECM). Invadopodia extend vertically into the basement membrane and are capable of locally restricted proteolysis via the expression of the membrane-bound matrix metalloproteinase (MMP)-14, which activates the major ECM-degrading MMPs 2 and -9 (Yamaguchi et al. 2005). It has been reported that malignant cells can also move as amoebas and squeeze through the ECM without proteolysis (Friedl 2004). However, this applies predominantly to hematopoetic tumours and seems to play a rather secondary role in vivo (Sabeh et al. 2009). The above described type of single cell movement corresponds to a well-known process in development, called epithelial-mesenchymal transition (EMT), characterized by the downregulation of epithelial markers, such as E-cadherin, and the upregulation of mesenchymal ones (Yang and Weinberg 2008). Poorly differentiated cancer cells are often constitutively highly motile and strongly express the mesenchymal markers Vimentin, Snail, and Twist. Well-differentiated cells, in contrast, use a different kind of locomotion and frequently move in cohorts with largely preserved epithelial characteristics (Friedl and Gilmour 2009). Only the leading cells show certain features of EMT, and, to acquire these, need strong stimuli from the surrounding tissue. Macrophages express and secrete a huge arsenal of factors with well-described activity in this context. Thus, they are attractive candidates to stimulate tumour cell motility either directly through the activation of migrationassociated signaling cascades or indirectly through the supply of downstream effectors which enable tumour cells to penetrate tissue structures.

Experimental Evidence for Macrophage-Induced Invasion Tumour Cell–Macrophage Interactions Already in 1987, it was reported that the invasive capacity of AH 130 rat ascites hepatoma cells was strongly increased upon cultivation on a macrophage feeder layer (Mukai et al. 1987). Subsequently, an elegant proof that macrophages in fact are crucial for cancer cell invasion was provided by Lin et al. (2001) in the transgenic MMTV-PyMT mouse model. Due to targeted overexpression of the oncogenic polyoma middle T-antigen, these mice spontaneously develop hormone receptornegative, her2-positive breast cancer which rapidly metastasizes to the lungs (Lin et al. 2003). When crossed with mice deficient of colony-stimulating factor-1 (CSF-1), thus lacking significant amounts of macrophages, metastasis formation was greatly reduced, while the oncogene-driven development of the primary tumours remained unaffected. Upon reconstitution of CSF-1, either exogenously or through the transgene, macrophage counts were restored, and metastatic dissemination was reestablished. This indicates that macrophages in this model are not involved in tumour initiation, but

3

The Role of Tumour-Associated Macrophages in Malignant Invasion

33

are critical for tumour progression. The fact that they foster the escape from the primary tumour strongly suggests that they can modulate tumour cell invasiveness. To better understand how TAM influence tumour cell motility, Hagemann et al. (2004) co-cultivated peripheral blood-derived human macrophages with various breast cancer cell lines in a modified Boyden chamber assay. Coculture enhanced invasiveness of the otherwise only weakly invasive cancer cell lines, while a benign mammary epithelial cell line remained noninvasive. Induced invasion relied on the induction of MMPs as well as on the upregulation of the cytokine TNFD, both in the macrophages. Interestingly, treatment with the Toll-like receptor (TLR) 4-agonist bacterial lipopolysacharide (LPS) abolished this effect, presumably by shifting the tumour-educated macrophages into a classical inflammatory phenotype. These in vitro findings were reproduced also for other cancer entities (Lin et al. 2006; Zins et al. 2007; Wu et al. 2009), however, they cannot be generalized. Invasion of MDA-MB 231, a highly motile breast cancer cell line with mesenchymal features, is much less dependent on external macrophage signals, since motility is already constitutively upregulated through an autocrine CSF-1 loop (Patsialou et al. 2009). Direct evidence of what happens between macrophages and tumour cells in vivo came from Wyckoff et al. (2004). In the above-mentioned PyMT mouse model as well as in SCID mice orthotopically injected with the rat mammary adenocarcinoma cell line MTLn3, microneedles were placed into the tumours for collection and further analysis of the invading cells. Needles contained either epidermal growth factor (EGF) or CSF-1 to provide chemoattraction. Two cell types were found in the collected population. Macrophages comigrated with cancer cells in an approximate ratio of 1:4, which was much higher than expected from the primary tumour and indicated enrichment of macrophages in the motile population. That macrophages were indeed critical for malignant invasion was shown in CSF-1-deficient PyMT variants, where collection yielded not only fewer macrophages but also much lower amounts of tumour cells. The same group (Wyckoff et al. 2007) showed later that macrophages not only contribute to the first of the four steps of tumour progression, namely invasion from the primary tumour into the immediate surroundings (DeNardo et al. 2008). Obviously, they are equally involved in the second step, the intravasation of tumour cells into blood vessels as a prerequisite for further dissemination. Using intravital multiphoton imaging and the PyMT mouse model, Wyckoff and colleagues (2007) visualized the direct interaction of breast cancer cells with macrophages. Motile tumour cells were predominantly found at sites of macrophage accumulation at the tumour margin and in the perivascular space. There, tumour cells entered the vasculature together with the macrophages, again, in a CSF-1 and EGF-dependent way.

34

C. Binder

Fig. 3.1 Tumour cells on their way into the brain. (a) MCF-7 breast cancer cells (black asterisks) are contacted by a resident brain macrophage (white arrow) protruding from living brain tissue (S). They are then pulled to the brain slice border (b) and use the microglial cell as a guiding rail for directional locomotion (c). Time lapse sequence of an organotypic brain slice coculture (obtained in collaboration with Chuang, Pukrop, and Hanisch, University of Göttingen)

The Role of Resident Macrophages To successfully accomplish colonization of distant tissues, circulating tumour cells have to extravasate from blood vessels and become invasive again. Since breast cancers in the PyMT mouse metastasize only on condition that the animals are not macrophage-depleted (Lin et al. 2001), it stands to reason that macrophages play a role also in this last step of tumour progression. However, apart from the peripheral blood-derived macrophages, most of the major target organs of metastasis harbor large populations of specialized macrophages with structural functions, also termed “trophic macrophages” (Qian and Pollard 2010). Until now, only few researchers have tried to answer the question of whether these resident macrophages are equally relevant for cancer cell motility and invasion. Liver-specific macrophages, the Kupffer cells, have been reported to exert tumoricidal as well as prometastatic effects (Gjoen et al. 1989; Heuff et al. 1995; Sturm et al. 2003; Gorden et al. 2007). In the lung, cells of the monocytic lineage establish the niche for successful colonization (Hiratsuka et al. 2002; Kaplan et al. 2005; Hiratsuka et al. 2008; Kim et al. 2009). Microglia, the resident macrophages of the central nervous system (CNS), representing a particularly secluded population, have been demonstrated to trigger invasion and secrete a variety of cytokines and growth factors which may contribute to successful immune evasion of CNSderived tumours (Markovic et al. 2005; Wesolowska et al. 2008). Recently, it has been shown that microglia are also essential for the invasion of solid tumours (Pukrop et al. 2010). Apart from enhancing the invasive capacity of breast cancer cells in Boyden chamber assays, microglia, as visualized in an organotypic slice coculture model, served as active transporters and guiding rails for the colonization of living brain tissues (Fig. 3.1). As a proof of principle, the bisphosphonate

3

The Role of Tumour-Associated Macrophages in Malignant Invasion

35

clodronate, a macrophage inactivator, abolished these effects. Unlike TAM, microglia did not undergo the expected cytokine profile switch to M2-like patterns. However, they were still able to react to exposure to LPS with an M1-like profile and the upregulation of TNFD, IL-6, CCL3 (MIP-1D), CCL5 (RANTES), CXCL1 (KC), and CXCL2 (MIP-2), which coincided with the loss of their pro-invasive function.

Signaling in Macrophage-Induced Invasion The EGF/CSF-1 Loop Establishment of cellular protrusions during invasion is based on enhanced actin turnover and extensive remodeling of the cytoskeleton. One of the key players in these changes is cofilin which was found upregulated in the invasive cancer cell population in a rat model (Wang et al. 2004). Local activation of cofilin defines the site of actin polymerization, protrusion and cell direction. Signaling through the EGF receptor was identified as the critical event (Yamaguchi et al. 2005). In particular, stimulation of the EGF receptor pathway was responsible for the formation of invadopodia, which was dependent on neural Wiskott-Aldrich syndrome protein (N-WASP) and the actin-related protein (Arp) 2/3 complex, while cofilin was required for the stabilization and maturation of these protrusions. The relevance of EGF signaling for invasion was further emphasized by the observation that the overexpression of the EGF receptor in the rat mammary adenocarcinoma cell line MTLn3 fosters in vitro motility and enhances in vivo invasion and metastasis (Xue et al. 2006). Although constitutive overexpression of the EGF receptor as well as production of the respective ligand has been shown in many cancer cells, Wyckoff and colleagues (2004) were the first to identify TAM as the major source of EGF in this context. As blockade of the EGF receptor as well as of CSF-1 signaling inhibited comigration of macrophages and tumour cells, the authors postulated the existence of a paracrine loop, where CSF-1 was secreted by the tumour cells to attract macrophages which, in turn, responded by the release of EGF and stimulated invasion. Consistently, cocultivation of macrophages with MTLn3 cells resulted in the formation of protrusions and enhanced in vitro invasion dependent on EGF and CSF-1 (Goswami et al. 2005). EGF-induced actin remodeling was associated with the activation of phosphoinositide 3 kinase and phospholipase J, followed by the generation of phosphoinositide 2 and 3 phosphate. These signaling molecules can interact with a variety of actin-binding proteins, and also with other regulators of actin polymerization, such as the GTPases Rho and Rac. Interestingly, the latter two can additionally be activated by Wnt-signaling, which will be discussed in detail below.

36

C. Binder

The impact of EGF for TAM-induced invasion was confirmed also in other cancer entities. Comparable effects were shown when hepG2 cells were cocultivated with conditioned media of various types of macrophages (Lin et al. 2006). This resulted in the reduction of E-cadherin in adherens junctions, acquisition of an EMT phenotype as well as enhanced invasive capacity. Again, the effect was dependent on EGF receptor signaling. The second partner in the pro-invasive loop, CSF-1 or M-CSF in humans, represents the major growth and differentiation factor for macrophages (Lin et al. 2002). It has long been known that CSF-1 promotes proliferation and, in particular, invasiveness of cancer cells expressing the cognate receptor CSF-1R, encoded by the c-fms protooncogene (Filderman et al. 1992). As mentioned above, MDA-MB 231 cells can upregulate invasion through their autocrine CSF-1 loop alone without any contribution of EGF (Patsialou et al. 2009). However, c-fms negative cells such as MCF-7, which have been shown to be unresponsive to the pro-invasive action of exogenous CSF-1 (Filderman et al. 1992), can still react with invasion to macrophage exposure (Hagemann et al. 2004). An elegant explanation to reconcile these divergent observations was recently provided by Abraham et al. (2010). They demonstrated that CSF-1 is also secreted by stromal cells, in particular, TAM and tumour-associated fibroblasts. In the same way as tumour-derived CSF-1, stromal CSF-1 serves as a chemoattractant to recruit additional macrophages, which then enhance the progression of c-fms positive as well as negative tumours, in the latter case via the upregulation of stromal target genes with pro-invasive function, such as MMP-2, VEG-F, and TNFD (Zins et al. 2007). Obviously, CSF-1 acts at the junction of several major tumour-promoting pathways, influencing not only cancer cell motility but also proteolytic tissue remodeling and neoangiogenesis.

Tumour Necrosis Factor a As the name implies, the inflammatory cytokine TNFD was originally discovered for its ability to elicit hemorrhagic necrosis in tumours. As a putative antitumour agent, it was even administered clinically in phase I/II studies. During these trials, however, it became increasingly clear that most of the TNFD effects point into the opposite direction (Balkwill 2009). In a murine model of chemically induced skin cancer, TNFD, while not interfering with tumour initiation, clearly promoted tumour progression (Moore et al. 1999). Although TNFD was produced by the tumour cells themselves, the fact that it was detected in whole epidermal homogenates of the tumour region (Scott et al. 2003) implies that stromal production may also have contributed to this effect. Consistently, macrophages were later identified as significant suppliers of TNFD, while enhancing invasion of breast cancer cells (Hagemann et al. 2004). Neutralization of TNFD inhibited the induction of downstream targets, such as MMPs, together with macrophageinduced invasion, indicating that TNFD is a critical modulator of the pro-invasive macrophage–tumour cell interactions. This was further underlined by the observation

3

The Role of Tumour-Associated Macrophages in Malignant Invasion

37

that TNFD was crucial for the macrophage switch to an alternatively activated M2-like phenotype (Hagemann et al. 2006). Of the two receptors for TNFD, the ubiquitously expressed 55 kDa TNFR I represents the key mediator of TNFD effects at the macrophage–tumour interface (Balkwill 2009). TNFR I governs several downstream signaling cascades which result either in the induction of cell death or in (chronic) inflammation and tumour promotion. Signaling via the latter branches involves the adaptor proteins TRADD and TRAF2 and is further propagated by the activation of Jun-N-terminal kinase (JNK) and the transcription factor AP-1, mitogen-activated protein kinases (MAPK) and the nuclear factor (NF)NB. The functional outcome of TNFD-signaling is determined not only by differential channeling of the signals, but also by the amount of available TNFD. High concentrations in classical M1-macrophages, especially in combination with other cytotoxic cytokines, clearly induce cell death, whereas lower levels in alternative activation do not. Instead, they induce mechanisms of enhanced survival and stimulate pro-invasive pathways (Sica et al. 2006). Among these pathways are the JNK as well as the NFNB cascade which have been described to mediate invasion in ovarian and breast cancer cell lines triggered by macrophage-derived TNFD (Hagemann et al. 2005). One of the downstream targets was the MMP-inducer EMMPRIN, whose inhibition by siRNA led to significant reduction of MMP secretion and invasion. Apart from increased proteolysis, TNFD can also directly influence cell motility. In pancreatic and colon cancer cell lines, macrophage-derived TNFD induced typical EMT features and enhanced invasion, acting cooperatively with another well-known EMT inducer, transforming growth factor (TGF)E (Bates and Mercurio 2003; Baran et al. 2009). Accumulation of the mesenchymal marker Snail due to TNFD-triggered inhibition of its degradation was responsible for this phenomenon (Wu et al. 2009). TNFD and the NFNB pathway also modulate cancer cell invasion via the activation of additional pro-migratory factors, such as VEGF and various chemokines. This has been demonstrated in ovarian cancer cell lines, where TNFD leads to the upregulation of VEGF, CCL2, CXCL12, and its receptor CXCR4, with corresponding immunohistochemical findings in primary ovarian cancers (Kulbe et al. 2005; Kulbe et al. 2007).

The Chemokine Connection Chemokines are small proteins which govern directional cell movement along chemotactic gradients. Although originally identified during leukocyte recruitment in tissue damage, it is now known that many cells can express chemokines and their receptors, in particular, malignant ones (Balkwill 2004). In cancer tissues, these cytokines are involved in various functions, most of them more or less based on their ability to induce directional migration (Barbieri et al. 2010). Of the many pairs of chemokines and cognate receptors, the CXCR4/CXCL12 axis is paradigmatic for the interplay between tumour cells and microenvironment.

38

C. Binder

The receptor is mostly, but not exclusively, expressed on cancer cells (Schioppa et al. 2003), while the ligand CXCL12, as implied by its alternative name stromal-derived factor (SDF)-1, is predominantly produced by cells of the stromal compartment, not least of which, by macrophages (Xu et al. 2009; Zlotnik 2006). Activation of CXCR4 stimulates directional movement of cancer cells and increases their invasion through artificial basement membranes as well as monolayers of various cell types (Scotton et al. 2002). Muller et al. (2001) demonstrated that CXCL12, either recombinant or in conditioned media from bone-marrow cells, induced actin polymerization, pseudopodia formation and directed migration of breast cancer cells. In SCID mice, these cells preferentially metastasized into lung, liver, and lymph nodes, containing high amounts of CXCL12. This is in line with the observation that collectively migrating cells, as often seen in the colonization of distant tissues, rely on the activation of CXCR4 in the leading cells by signals from the stroma (Friedl and Gilmour 2009). The recently identified second receptor for CXCL12, CXCR7, which is expressed at the rear end of migrating cohorts, can exert inhibitory functions and specifically interfere with CXCL12/CXCR4-induced transendothelial migration of cancer cells (Zabel et al. 2009). There are many cross-links to other tumour-promoting signaling cascades. CXCR4 as well as its ligand are modulated by hypoxia through hypoxia-inducible factor (HIF) 1D, which upregulates both molecules in tumour cells and macrophages (Schioppa et al. 2003). Hypoxic necrosis is frequently observed in growing tumours, which additionally attracts macrophages, thus, amplifying the pro-invasive feedback loop. Moreover, CXCL12 has been shown to upregulate TNFD in ovarian cancer cells (Scotton et al. 2002), which, in turn, can trigger invasion. In breast cancer cells, motility and CXCR4 expression are regulated via the NFNB pathway (Helbig et al. 2003). There is evidence also for cross-talk with the EGF/c-erbB receptor family (Hernandez et al. 2009). Overexpression of her-2/c-erbB-2 in breast cancer leads to enhanced invasiveness through the upregulation of CXCR4 and the inhibition of its ligand-induced degradation (Li et al. 2004). Similarly, transactivation of the EGF/c-erbB-1 receptor via CXCR4 and CXCL12 has been demonstrated in ovarian carcinoma cells (Porcile et al. 2005). Comparable to the CXCR4/CXCL12 axis, which elicits a complex to and fro of para- as well as autocrine actions and reactions between tumour and stromal cells to foster invasion, there are many other chemokines and receptors (Robinson and Coussens 2005) following more or less the same functional patterns. For example, expression of CCL2 or monocyte chemoattractant protein (MCP)-1, present in tumour cells as well as macrophages, is upregulated in the latter through tumour cell coculture and results in enhanced macrophage chemotaxis (Fujimoto et al. 2009). Blockade of CCL2 production in non-small cell lung cancers inhibits tumour growth and metastasis not only by altered recruitment of TAM, but also by shifting their phenotype to more M1-like characteristics which then stimulate antitumour activity of the innate immune system via recruitment of CD8-positive T-cells (Fridlender et al. 2011). In conclusion, the chemokine network, apart from direct modulation of tumour cell motility, can potentiate this effect by controlling the amount and com-

3

The Role of Tumour-Associated Macrophages in Malignant Invasion

39

position of the leukocyte infiltrate, its action at the tumour–host boundary being crucial for tumour progression.

Wnt Signaling Wnt ligands are secreted glycoproteins, involved in cell proliferation and directional migration in embryonic development as well as in the adult organism (Clevers 2006). For example, so-called canonical Wnt signaling via E-catenin regulates cohort movement of lateral line cells in the zebrafish embryo by spatial distribution of the two CXCL 12 receptors CXCR4 and seven to either the leading or the trailing end (Aman and Piotrowski 2008). Given these functions, it is not surprising that Wnt signals contribute to cancer initiation and progression when deregulated. Wnts and their receptors are overexpressed in a variety of malignancies, in particular, in colon cancer (Fodde and Brabletz 2007). Pathologic activation of the Wnt/E-catenin pathway is associated with aggressive tumour biology and has been found especially at the invasive frontier between tumour and surrounding tissue (Brabletz et al. 2005). Invasion can also be mediated by another group of Wnts, referred to as noncanonical, whose signals are channeled either into the Wnt/Ca2+ or the planar cell polarity (PCP) pathway (Wallingford and Habas 2005). PCP-signaling results in the activation of the small Rho-GTPases RhoA and Rac, leading to reorganization of the cytoskeleton during establishment of cellular protrusions as well as phosphorylation of JNK and AP-1 (Montcouquiol et al. 2006). While most of the investigations focused on Wnt signaling in tumour cells, the first notion that stroma-derived Wnts might play a role in tumour progression came from Smith et al. (1999) who observed the expression of the noncanonical ligand Wnt 5a in TAM in colon cancer. It was not until 2006 that macrophage-induced invasion of breast cancer cells was demonstrated to depend in fact on the upregulation of macrophage Wnt 5a (Pukrop et al. 2006). Although a functional canonical pathway in the tumour cells was indispensable, noncanonical signaling via JNK was the critical event. The biological significance of these findings was underlined by the detection of Wnt 5a in TAM in primary breast cancers and at the invasive front of lymph node metastases. Upregulation of Wnt 5a has been observed in various types of inflammatory reactions, thus linking Wnt signaling to other inflammatory cytokines, such as TNFD, and the TLR pathway (Pereira et al. 2009). It was expressed in macrophages in granulomatous lesions in tuberculosis (Blumenthal et al. 2006) as well as in septic conditions, where signaling was conferred via Ca2+-influx (Pereira et al. 2008). It stands to reason that these macrophages correspond to the classical M1 phenotype with potential antitumour activity. Obviously, Wnt 5a in macrophages exerts a dual role with opposite functional outcome (Pukrop and Binder 2008). When part of an M1-like reaction and signaling through the Wnt/Ca2+ cascade, it may act as a tumour suppressor, whereas Wnt 5a in tumour-educated macrophages facilitates malignant invasion.

40

C. Binder

A recent study of the transcriptome of TAM underlines the significance of Wnt signaling in this setting. Gene expression of murine macrophages comigrating with tumour cells, thus representing the actual pro-invasive subpopulation, was compared with that of the remaining nonmotile TAM (Ojalvo et al. 2010). The Wnt pathway was found enriched in this fraction with the upregulation of Wnt 5b and Wnt 7b, among other Wnt-associated molecules. The positive correlation between Wnt 7b expression and lymph node metastasis, identified by comparative analysis of previously published gene sets of human breast cancer, emphasizes the in vivo relevance of these findings. Interestingly, the pro-invasive interaction of tumour cells with the resident macrophages of the brain was also shown to be Wnt-dependent (Pukrop et al. 2010). Microglial cells constitutively express a series of Wnt ligands, among them Wnt 5a and b, and microglia-induced invasion was inhibited by the physiological Wnt-antagonists Dickkopf-1 and -2.

The Role of Proteases and the Extracellular Matrix Mostly as downstream effectors, but also as modulators of signaling by proteolytic modification of both cell surface proteins and ECM components, proteases play a key role in invasion. MMPs are the most well-known candidates in this context. They belong to a large family of zinc-dependent proteases with ECM-degrading activity and specificity for either collagens or other matrix molecules. During the 1990s, there were many reports that MMPs, especially the collagenases MMP-2 and -9, are crucial for tumour cell invasion (Westermarck and Kahari 1999). They are required, in particular, for the penetration of the collagen-rich basement membrane (Coussens and Werb 1996), their expression correlating with advanced tumour stage and unfavorable clinical outcome. Although many cancer cells express MMPs and are capable of local proteolysis through their invadopodia (Buccione et al. 2009), it has gradually become clear that the stroma significantly contributes to progression-associated MMP production, and a multitude of MMPs are supplied in high concentrations by TAM (Wiesen and Werb 1996; Swallow et al. 1996). MMP inhibitors can completely abolish macrophageinduced invasion in various settings (Hagemann et al. 2004; Grimshaw et al. 2004). The fact that these inhibitors simultaneously block coculture-induced TNFD production, presumably by antagonizing TNFD-shedding through MMP-7 (Haro et al. 2000), suggests that MMPs do not only act as terminal downstream effectors of invasion but also as signaling modulators. At the same time, MMPs are targets of several important regulators of macrophage-induced invasion, such as TNFD itself, SDF-1, EGF, and Wnt 5a (Jodele et al. 2006; Pukrop et al. 2006; Prieve and Moon 2003), indicating that they are involved in para- as well as autocrine loops within the complex signaling network between TAM and tumour cells. Cathepsins, a family of lysosomal cysteine proteases, also contribute to the proinvasive functions of TAM. They have been found upregulated in various cancers,

3

The Role of Tumour-Associated Macrophages in Malignant Invasion

41

where they are mislocated to the cellular surface and secreted into the ECM (Gocheva et al. 2006). Several members of this family, in particular cathepsin B, enhance tumour cell invasiveness via direct cleavage of E-cadherin, leading to loosening of cell–cell contacts and acquisition of an EMT phenotype. Interestingly, apart from the tumour cells themselves, cathepsins can also be produced by TAM (Vasiljeva et al. 2006). In the PyMT mouse as well as in the RIP1-Tag2 model of pancreatic cancer, cathepsin B is increasingly upregulated in macrophages, while they are recruited to the tumour and switch to the full TAM phenotype. This is dependent on IL-4 secretion by the cancer cells (Gocheva et al. 2010). TAMderived cathepsin B then enhances malignant invasion through matrigel and collagen I in vitro, as well as tumour growth and lung colonization in vivo. Obviously, proteases are crucial for the extensive ECM remodeling during tumour cell invasion. However, the ECM, mainly composed of collagens, fibronectin, laminins, elastin, and glycosaminoglycans, is not merely an inert structural scaffold for the embedded epithelial cells, where tumour cells just hack their way through. It rather represents a dynamic signaling network, which can either give rise to or modulate the activity of signaling molecules (Schenk and Quaranta 2003). Wnt molecules, due to their lipid modification (Mikels and Nusse 2006), are typical candidates for such interactions. Much of the secreted Wnts bind to heparan sulfate proteoglycans (Ai et al. 2003; Kurayoshi et al. 2007). Chemical linkage interferes with their signaling activity as well as with the establishment of gradients for directional cell movement. Similarly, matrix components can interact with chemokines and modulate the secretion of cytokines as well as MMP-9 by macrophages (Chung and Kao 2009). Versican, another ECM proteoglycan, found in interstitial tissues at the invasive margins of various cancers, interacts with EGF receptor signaling and promotes migration (Wu et al. 2005; Zheng et al. 2004). When produced by Lewis lung carcinoma cells, it activates TLR 2 and its coreceptors TLR 6 and CD14 and induces IL-6 and TNFD in macrophages, resulting in strongly enhanced metastasis to the lungs (Kim et al. 2009).

Conclusion Taken together, whether a malignant cell invades is not only determined by its intrinsic invasive capacity, but also is regulated by TAM. Invasiveness of poorly differentiated cancer cells with mesenchymal features is constitutively high, and often only marginally influenced by TAM, in contrast to their more epithelial-like counterparts, found in the majority of solid cancers. Differentiated cells require support from TAM, which either directly induce EMT in the tumour cells, or serve as mesenchymal guides at the tip of invading tumour cell cohorts with largely preserved epithelial characteristics. Both settings are driven by the activation of pro-invasive signaling pathways which either end or start in tumour or stromal cells and vice versa, thus, modulating and amplifying the respective effects (Fig. 3.2).

42

C. Binder

Fig. 3.2 Hypothetical model of macrophage-induced invasion. Tumour cells secrete cytokines and chemoattractants to recruit macrophages. This switches macrophages to an alternatively activated phenotype and leads to the induction of stimulators and effectors of motility and invasion, among them EGF, which is part of the pro-invasive EGF/CSF-1 loop, TNFD, a major inductor of the NFkB pathway, chemokines as well as Wnt ligands and matrix metalloproteinases. Several of these can also be provided by other stromal cells, in particular CXCL12/SDF-1, which fosters migration and invasion by interaction with the CXCR4 chemokine receptor on cancer cells. Their effects on malignant invasion are additionally regulated by interaction with components of the extracellular matrix as well as by autocrine loops. In some cases, these loops can substitute for the assistance of macrophages, e.g., the autocrine stimulation by CSF-1 without EGF in cells with constitutive mesenchymal features. Macrophages contribute to invasion by directly inducing EMT and enhanced invasive capacity in the tumour cells via the provided stimuli. Additionally and/or alternatively, they can act as mesenchymal assistants at the tip of invading tumour cell cohorts with largely preserved epithelial characteristics. (T = tumour cells, M = macrophages, F = fibroblasts, V = vasculature, L = lymphocytes, ECM = extracellular matrix, EMT = epithelial–mesenchymal transition)

Given the complexity of the underlying signaling and effector networks, it is clear that therapeutic intervention is not trivial. These networks are cross-linked at multiple levels, and activation patterns in the cytotoxic and the pro-invasive setting are at least partially overlapping. Thus, attempts to modulate activity of single factors may result in unexpected upregulation of substitute pathways and other escape mechanisms and have, until now, not proven satisfactorily successful. Targeting the TAM to shift them back from a tumour-promoting phenotype to their

3

The Role of Tumour-Associated Macrophages in Malignant Invasion

43

genuine defense functions, seems a more promising strategy, as this would interfere not only with their pro-invasive but also with their angiogenic and growth-supporting actions. Coming back to Coley, in principle, he seems to have been on the right track. However, there is still a lot more to know to reprogram inflammation and disrupt the involved cascades at the appropriate point to eventually achieve anti-invasive features without undesired side effects.

References Abraham D, Zins K, Sioud M, Lucas T, Schafer R, Stanley ER, Aharinejad S (2010) Stromal cell-derived CSF-1 blockade prolongs xenograft survival of CSF-1-negative neuroblastoma. Int J Cancer 126:1339–1352 Ai X, Do AT, Lozynska O, Kusche-Gullberg M, Lindahl U, Emerson CP Jr (2003) QSulf1 remodels the 6-O sulfation states of cell surface heparan sulfate proteoglycans to promote Wnt signaling. J Cell Biol 162:341–351 Aman A, Piotrowski T (2008) Wnt/beta-catenin and Fgf signaling control collective cell migration by restricting chemokine receptor expression. Dev Cell 15:749–761 Balkwill F (2004) Cancer and the chemokine network. Nat Rev Cancer 4:540–550 Balkwill F (2009) Tumour necrosis factor and cancer. Nat Rev Cancer 9:361–371 Baran B, Bechyne I, Siedlar M, Szpak K, Mytar B, Sroka J, Laczna E, Madeja Z, Zembala M, Czyz J (2009) Blood monocytes stimulate migration of human pancreatic carcinoma cells in vitro: the role of tumour necrosis factor - alpha. Eur J Cell Biol 88:743–752 Barbieri F, Bajetto A, Florio T (2010) Role of chemokine network in the development and progression of ovarian cancer: a potential novel pharmacological target. J Oncol 2010:426956 Bates RC, Mercurio AM (2003) Tumor necrosis factor-alpha stimulates the epithelial-to-mesenchymal transition of human colonic organoids. Mol Biol Cell 14:1790–1800 Blumenthal A, Ehlers S, Lauber J, Buer J, Lange C, Goldmann T, Heine H, Brandt E, Reiling N (2006) The Wingless homolog WNT5A and its receptor Frizzled-5 regulate inflammatory responses of human mononuclear cells induced by microbial stimulation. Blood 108:965–973 Brabletz T, Hlubek F, Spaderna S, Schmalhofer O, Hiendlmeyer E, Jung A, Kirchner T (2005) Invasion and metastasis in colorectal cancer: epithelial-mesenchymal transition, mesenchymalepithelial transition, stem cells and beta-catenin. Cells Tissues Organs 179:56–65 Buccione R, Caldieri G, Ayala I (2009) Invadopodia: specialized tumor cell structures for the focal degradation of the extracellular matrix. Cancer Metastasis Rev 28:137–149 Chung AS, Kao WJ (2009) Fibroblasts regulate monocyte response to ECM-derived matrix: the effects on monocyte adhesion and the production of inflammatory, matrix remodeling, and growth factor proteins. J Biomed Mater Res A 89:841–853 Clevers H (2006) Wnt/beta-catenin signaling in development and disease. Cell 127:469–480 Coussens LM, Werb Z (1996) Matrix metalloproteinases and the development of cancer. Chem Biol 3:895–904 DeNardo DG, Johansson M, Coussens LM (2008) Immune cells as mediators of solid tumor metastasis. Cancer Metastasis Rev 27:11–18 Filderman AE, Bruckner A, Kacinski BM, Deng N, Remold HG (1992) Macrophage colonystimulating factor (CSF-1) enhances invasiveness in CSF-1 receptor-positive carcinoma cell lines. Cancer Res 52:3661–3666 Fodde R, Brabletz T (2007) Wnt/beta-catenin signaling in cancer stemness and malignant behavior. Curr Opin Cell Biol 19:150–158 Fridlender ZG, Kapoor V, Buchlis G, Cheng G, Sun J, Wang LC, Singhal S, Snyder LA, Albelda SM (2011) CCL2 blockade inhibits lung cancer tumor growth by altering macrophage phenotype and activating CD8+ Cells. Am J Respir Cell Mol Biol 44:230–237

44

C. Binder

Friedl P (2004) Prespecification and plasticity: shifting mechanisms of cell migration. Curr Opin Cell Biol 16:14–23 Friedl P, Gilmour D (2009) Collective cell migration in morphogenesis, regeneration and cancer. Nat Rev Mol Cell Biol 10:445–457 Fujimoto H, Sangai T, Ishii G, Ikehara A, Nagashima T, Miyazaki M, Ochiai A (2009) Stromal MCP-1 in mammary tumors induces tumor-associated macrophage infiltration and contributes to tumor progression. Int J Cancer 125:1276–1284 Gjoen T, Seljelid R, Kolset SO (1989) Binding of metastatic colon carcinoma cells to liver macrophages. J Leukoc Biol 45:362–369 Gocheva V, Zeng W, Ke D, Klimstra D, Reinheckel T, Peters C, Hanahan D, Joyce JA (2006) Distinct roles for cysteine cathepsin genes in multistage tumorigenesis. Genes Dev 20:543–556 Gocheva V, Wang HW, Gadea BB, Shree T, Hunter KE, Garfall AL, Berman T, Joyce JA (2010) IL-4 induces cathepsin protease activity in tumor-associated macrophages to promote cancer growth and invasion. Genes Dev 24:241–255 Gorden DL, Fingleton B, Crawford HC, Jansen DE, Lepage M, Matrisian LM (2007) Resident stromal cell-derived MMP-9 promotes the growth of colorectal metastases in the liver microenvironment. Int J Cancer 121:495–500 Goswami S, Sahai E, Wyckoff JB, Cammer M, Cox D, Pixley FJ, Stanley ER, Segall JE, Condeelis JS (2005) Macrophages promote the invasion of breast carcinoma cells via a colony-stimulating factor-1/epidermal growth factor paracrine loop. Cancer Res 65:5278–5283 Grimshaw MJ, Hagemann T, Ayhan A, Gillett CE, Binder C, Balkwill FR (2004) A role for endothelin-2 and its receptors in breast tumor cell invasion. Cancer Res 64:2461–2468 Hagemann T, Robinson SC, Schulz M, Trumper L, Balkwill FR, Binder C (2004) Enhanced invasiveness of breast cancer cell lines upon co-cultivation with macrophages is due to TNF-alpha dependent up-regulation of matrix metalloproteases. Carcinogenesis 25:1543–1549 Hagemann T, Wilson J, Kulbe H, Li NF, Leinster DA, Charles K, Klemm F, Pukrop T, Binder C, Balkwill FR (2005) Macrophages induce invasiveness of epithelial cancer cells via NF-kappa B and JNK. J Immunol 175:1197–1205 Hagemann T, Wilson J, Burke F, Kulbe H, Li NF, Pluddemann A, Charles K, Gordon S, Balkwill FR (2006) Ovarian cancer cells polarize macrophages toward a tumor-associated phenotype. J Immunol 176:5023–5032 Haro H, Crawford HC, Fingleton B, Shinomiya K, Spengler DM, Matrisian LM (2000) Matrix metalloproteinase-7-dependent release of tumor necrosis factor-alpha in a model of herniated disc resorption. J Clin Invest 105:143–150 Helbig G, Christopherson KW, Bhat-Nakshatri P, Kumar S, Kishimoto H, Miller KD, Broxmeyer HE, Nakshatri H (2003) NF-kappaB promotes breast cancer cell migration and metastasis by inducing the expression of the chemokine receptor CXCR4. J Biol Chem 278:21631–21638 Hernandez L, Smirnova T, Kedrin D, Wyckoff J, Zhu L, Stanley ER, Cox D, Muller WJ, Pollard JW, Van RN, Segall JE (2009) The EGF/CSF-1 paracrine invasion loop can be triggered by heregulin beta1 and CXCL12. Cancer Res 69:3221–3227 Heuff G, van de Loosdrecht AA, Betjes MG, Beelen RH, Meijer S (1995) Isolation and purification of large quantities of fresh human Kupffer cells, which are cytotoxic against colon carcinoma. Hepatology 21:740–745 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 2:289–300 Hiratsuka S, Watanabe A, Sakurai Y, Akashi-Takamura S, Ishibashi S, Miyake K, Shibuya M, Akira S, Aburatani H, Maru Y (2008) The S100A8-serum amyloid A3-TLR4 paracrine cascade establishes a pre-metastatic phase. Nat Cell Biol 10:1349–1355 Jodele S, Blavier L, Yoon JM, DeClerck YA (2006) Modifying the soil to affect the seed: role of stromal-derived matrix metalloproteinases in cancer progression. Cancer Metastasis Rev 25:35–43 Kaplan RN, Riba RD, Zacharoulis S, Bramley AH, Vincent L, Costa C, MacDonald DD, Jin DK, Shido K, Kerns SA, Zhu Z, Hicklin D, Wu Y, Port JL, Altorki N, Port ER, Ruggero D, Shmelkov SV, Jensen KK, Rafii S, Lyden D (2005) VEGFR1-positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche. Nature 438:820–827

3

The Role of Tumour-Associated Macrophages in Malignant Invasion

45

Kedrin D, van Rheenen J, Hernandez L, Condeelis J, Segall JE (2007) Cell motility and cytoskeletal regulation in invasion and metastasis. J Mammary Gland Biol Neoplasia 12:143–152 Kim S, Takahashi H, Lin WW, Descargues P, Grivennikov S, Kim Y, Luo JL, Karin M (2009) Carcinoma-produced factors activate myeloid cells through TLR2 to stimulate metastasis. Nature 457:102–106 Kulbe H, Hagemann T, Szlosarek PW, Balkwill FR, Wilson JL (2005) The inflammatory cytokine tumor necrosis factor-alpha regulates chemokine receptor expression on ovarian cancer cells. Cancer Res 65:10355–10362 Kulbe H, Thompson R, Wilson JL, Robinson S, Hagemann T, Fatah R, Gould D, Ayhan A, Balkwill F (2007) The inflammatory cytokine tumor necrosis factor-alpha generates an autocrine tumor-promoting network in epithelial ovarian cancer cells. Cancer Res 67:585–592 Kurayoshi M, Yamamoto H, Izumi S, Kikuchi A (2007) Post-translational palmitoylation and glycosylation of Wnt-5a are necessary for its signalling. Biochem J 402:515–523 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 56:4625–4629 Li YM, Pan Y, Wei Y, Cheng X, Zhou BP, Tan M, Zhou X, Xia W, Hortobagyi GN, Yu D, Hung MC (2004) Upregulation of CXCR4 is essential for HER2-mediated tumor metastasis. Cancer Cell 6:459–469 Lin EY, Nguyen AV, Russell RG, Pollard JW (2001) Colony-stimulating factor 1 promotes progression of mammary tumors to malignancy. J Exp Med 193:727–740 Lin EY, Gouon-Evans V, Nguyen AV, Pollard JW (2002) The macrophage growth factor CSF-1 in mammary gland development and tumor progression. J Mammary Gland Biol Neoplasia 7:147–162 Lin EY, Jones JG, Li P, Zhu L, Whitney KD, Muller WJ, Pollard JW (2003) Progression to malignancy in the polyoma middle T oncoprotein mouse breast cancer model provides a reliable model for human diseases. Am J Pathol 163:2113–2126 Lin CY, Lin CJ, Chen KH, Wu JC, Huang SH, Wang SM (2006) Macrophage activation increases the invasive properties of hepatoma cells by destabilization of the adherens junction. FEBS Lett 580:3042–3050 Mantovani A, Bottazzi B, Colotta F, Sozzani S, Ruco L (1992) The origin and function of tumorassociated macrophages. Immunol Today 13:265–270 Markovic DS, Glass R, Synowitz M, Rooijen N, Kettenmann H (2005) Microglia stimulate the invasiveness of glioma cells by increasing the activity of metalloprotease-2. J Neuropathol Exp Neurol 64:754–762 Mikels AJ, Nusse R (2006) Wnts as ligands: processing, secretion and reception. Oncogene 25: 7461–7468 Montcouquiol M, Crenshaw EB III, Kelley MW (2006) Noncanonical Wnt signaling and neural polarity. Annu Rev Neurosci 29:363–386 Moore RJ, Owens DM, Stamp G, Arnott C, Burke F, East N, Holdsworth H, Turner L, Rollins B, Pasparakis M, Kollias G, Balkwill F (1999) Mice deficient in tumor necrosis factor-alpha are resistant to skin carcinogenesis. Nat Med 5:828–831 Mukai M, Shinkai K, Tateishi R, Mori Y, Akedo H (1987) Macrophage potentiation of invasive capacity of rat ascites hepatoma cells. Cancer Res 47:2167–2171 Muller A, Homey B, Soto H, Ge N, Catron D, Buchanan ME, McClanahan T, Murphy E, Yuan W, Wagner SN, Barrera JL, Mohar A, Verastegui E, Zlotnik A (2001) Involvement of chemokine receptors in breast cancer metastasis. Nature 410:50–56 Nauts HC, Swift WE, Coley BL (1946) The Treatment of Malignant Tumors by Bacterial Toxins As Developed by the Late Coley, William, B., Reviewed in the Light of Modern Research. Cancer Res 6:205–216 Ojalvo LS, Whittaker CA, Condeelis JS, Pollard JW (2010) Gene expression analysis of macrophages that facilitate tumor invasion supports a role for Wnt-signaling in mediating their activity in primary mammary tumors. J Immunol 184:702–712

46

C. Binder

Patsialou A, Wyckoff J, Wang Y, Goswami S, Stanley ER, Condeelis JS (2009) Invasion of human breast cancer cells in vivo requires both paracrine and autocrine loops involving the colonystimulating factor-1 receptor. Cancer Res 69:9498–9506 Pereira C, Schaer DJ, Bachli EB, Kurrer MO, Schoedon G (2008) Wnt5A/CaMKII signaling contributes to the inflammatory response of macrophages and is a target for the antiinflammatory action of activated protein C and interleukin-10. Arterioscler Thromb Vasc Biol 28:504–510 Pereira CP, Bachli EB, Schoedon G (2009) The wnt pathway: a macrophage effector molecule that triggers inflammation. Curr Atheroscler Rep 11:236–242 Porcile C, Bajetto A, Barbieri F, Barbero S, Bonavia R, Biglieri M, Pirani P, Florio T, Schettini G (2005) Stromal cell-derived factor-1alpha (SDF-1alpha/CXCL12) stimulates ovarian cancer cell growth through the EGF receptor transactivation. Exp Cell Res 308:241–253 Prieve MG, Moon RT (2003) Stromelysin-1 and mesothelin are differentially regulated by Wnt-5a and Wnt-1 in C57mg mouse mammary epithelial cells. BMC Dev Biol 3:2 Pukrop T, Binder C (2008) The complex pathways of Wnt 5a in cancer progression. J Mol Med 86:259–266 Pukrop T, Klemm F, Hagemann T, Gradl D, Schulz M, Siemes S, Trumper L, Binder C (2006) Wnt 5a signaling is critical for macrophage-induced invasion of breast cancer cell lines. Proc Natl Acad Sci USA 103:5454–5459 Pukrop T, Dehghani F, Chuang H-N, Lohaus R, Bayanga K, Heermann S, Regen T, van Rossum D, Klemm F, Schulz M, Siam L, Hoffmann A, Trümper L, Stadelmann Chr, Bechmann I, Hanisch U-K, Binder C (2010) Microglia promote colonization of brain tissue by breast cancer cells in a Wnt-dependent way. Glia 58:1477–1489 Qian BZ, Pollard JW (2010) Macrophage diversity enhances tumor progression and metastasis. Cell 141:39–51 Robinson SC, Coussens LM (2005) Soluble mediators of inflammation during tumor development. Adv Cancer Res 93:159–187 Sabeh F, Shimizu-Hirota R, Weiss SJ (2009) Protease-dependent versus -independent cancer cell invasion programs: three-dimensional amoeboid movement revisited. J Cell Biol 185:11–19 Schenk S, Quaranta V (2003) Tales from the crypt[ic] sites of the extracellular matrix. Trends Cell Biol 13:366–375 Schioppa T, Uranchimeg B, Saccani A, Biswas SK, Doni A, Rapisarda A, Bernasconi S, Saccani S, Nebuloni M, Vago L, Mantovani A, Melillo G, Sica A (2003) Regulation of the chemokine receptor CXCR4 by hypoxia. J Exp Med 198:1391–1402 Scott KA, Moore RJ, Arnott CH, East N, Thompson RG, Scallon BJ, Shealy DJ, Balkwill FR (2003) An anti-tumor necrosis factor-alpha antibody inhibits the development of experimental skin tumors. Mol Cancer Ther 2:445–451 Scotton CJ, Wilson JL, Scott K, Stamp G, Wilbanks GD, Fricker S, Bridger G, Balkwill FR (2002) Multiple actions of the chemokine CXCL12 on epithelial tumor cells in human ovarian cancer. Cancer Res 62:5930–5938 Sica A, Schioppa T, Mantovani A, Allavena P (2006) Tumour-associated macrophages are a distinct M2 polarised population promoting tumour progression: potential targets of anti-cancer therapy. Eur J Cancer 42:717–727 Smith K, Bui TD, Poulsom R, Kaklamanis L, Williams G, Harris AL (1999) Up-regulation of macrophage wnt gene expression in adenoma-carcinoma progression of human colorectal cancer. Br J Cancer 81:496–502 Steele RJ, Eremin O, Brown M, Hawkins RA (1984) A high macrophage content in human breast cancer is not associated with favourable prognostic factors. Br J Surg 71:456–458 Sturm JW, Magdeburg R, Berger K, Petruch B, Samel S, Bonninghoff R, Keese M, Hafner M, Post S (2003) Influence of TNFA on the formation of liver metastases in a syngenic mouse model. Int J Cancer 107:11–21 Swallow CJ, Murray MP, Guillem JG (1996) Metastatic colorectal cancer cells induce matrix metalloproteinase release by human monocytes. Clin Exp Metastasis 14:3–11 Vasiljeva O, Papazoglou A, Kruger A, Brodoefel H, Korovin M, Deussing J, Augustin N, Nielsen BS, Almholt K, Bogyo M, Peters C, Reinheckel T (2006) Tumor cell-derived and macrophage-

3

The Role of Tumour-Associated Macrophages in Malignant Invasion

47

derived cathepsin B promotes progression and lung metastasis of mammary cancer. Cancer Res 66:5242–5250 Wallingford JB, Habas R (2005) The developmental biology of Dishevelled: an enigmatic protein governing cell fate and cell polarity. Development 132:4421–4436 Wang W, Goswami S, Lapidus K, Wells AL, Wyckoff JB, Sahai E, Singer RH, Segall JE, Condeelis JS (2004) Identification and testing of a gene expression signature of invasive carcinoma cells within primary mammary tumors. Cancer Res 64:8585–8594 Wesolowska A, Kwiatkowska A, Slomnicki L, Dembinski M, Master A, Sliwa M, Franciszkiewicz K, Chouaib S, Kaminska B (2008) Microglia-derived TGF-beta as an important regulator of glioblastoma invasion–an inhibition of TGF-beta-dependent effects by shRNA against human TGF-beta type II receptor. Oncogene 27:918–930 Westermarck J, Kahari VM (1999) Regulation of matrix metalloproteinase expression in tumor invasion. FASEB J 13:781–792 Wiesen JF, Werb Z (1996) The role of stromelysin-1 in stromal-epithelial interactions and cancer. Enzyme Protein 49:174–181 Wu YJ, La Pierre DP, Wu J, Yee AJ, Yang BB (2005) The interaction of versican with its binding partners. Cell Res 15:483–494 Wu Y, Deng J, Rychahou PG, Qiu S, Evers BM, Zhou BP (2009) Stabilization of snail by NF-kappaB is required for inflammation-induced cell migration and invasion. Cancer Cell 15:416–428 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 64:7022–7029 Wyckoff JB, Wang Y, Lin EY, Li JF, Goswami S, Stanley ER, Segall JE, Pollard JW, Condeelis J (2007) Direct visualization of macrophage-assisted tumor cell intravasation in mammary tumors. Cancer Res 67:2649–2656 Xu L, Duda DG, di Tomaso E, Ancukiewicz M, Chung DC, Lauwers GY, Samuel R, Shellito P, Czito BG, Lin PC, Poleski M, Bentley R, Clark JW, Willett CG, Jain RK (2009) Direct evidence that bevacizumab, an anti-VEGF antibody, up-regulates SDF1alpha, CXCR4, CXCL6, and neuropilin 1 in tumors from patients with rectal cancer. Cancer Res 69:7905–7910 Xue C, Wyckoff J, Liang F, Sidani M, Violini S, Tsai KL, Zhang ZY, Sahai E, Condeelis J, Segall JE (2006) Epidermal growth factor receptor overexpression results in increased tumor cell motility in vivo coordinately with enhanced intravasation and metastasis. Cancer Res 66:192–197 Yamaguchi H, Lorenz M, Kempiak S, Sarmiento C, Coniglio S, Symons M, Segall J, Eddy R, Miki H, Takenawa T, Condeelis J (2005) Molecular mechanisms of invadopodium formation: the role of the N-WASP-Arp2/3 complex pathway and cofilin. J Cell Biol 168:441–452 Yang J, Weinberg RA (2008) Epithelial-mesenchymal transition: at the crossroads of development and tumor metastasis. Dev Cell 14:818–829 Zabel BA, Wang Y, Lewen S, Berahovich RD, Penfold ME, Zhang P, Powers J, Summers BC, Miao Z, Zhao B, Jalili A, Janowska-Wieczorek A, Jaen JC, Schall TJ (2009) Elucidation of CXCR7-mediated signaling events and inhibition of CXCR4-mediated tumor cell transendothelial migration by CXCR7 ligands. J Immunol 183:3204–3211 Zheng B, Lavoie C, Tang TD, Ma P, Meerloo T, Beas A, Farquhar MG (2004) Regulation of epidermal growth factor receptor degradation by heterotrimeric Galphas protein. Mol Biol Cell 15:5538–5550 Zins K, Abraham D, Sioud M, Aharinejad S (2007) Colon cancer cell-derived tumor necrosis factor-alpha mediates the tumor growth-promoting response in macrophages by up-regulating the colony-stimulating factor-1 pathway. Cancer Res 67:1038–1045 Zlotnik A (2006) Chemokines and cancer. Int J Cancer 119:2026–2029

Chapter 4

Tumour-Induced Immune Suppression by Myeloid Cells Serena Zilio, Giacomo Desantis, Mariacristina Chioda, and Vincenzo Bronte

Tumour-Induced Immunosuppression Since the description of the immunosurveillance theory, tumour immunologists have faced one critical paradox: the host maintains a general good responsiveness to exogenous antigens in spite of the profound dysfunction of immune reactions against tumour-antigens, except under some chemotherapeutic and radiotherapy treatments or in advanced stages of the disease. Even with the identification of tumour-associated antigens, which could be used as selective targets for cancer immunotherapy, the inability of the immune system to provide a strong response against the tumour, and the consequent low efficiency in eliminating tumour cells represents one of the major limitations to clinical translation. Indeed, the process of immune selection can be exploited by tumours for evolving in cells escaping the host immune response. This allows the generation of clinically identifiable tumour variants growing in immunocompetent hosts. It is, therefore, clear that the immune system not only has the potential of protecting the host in early stages of tumourigenesis but can also select cancer cells to give rise to species with reduced immunogenicity (Dunn et al. 2004).

S. Zilio Department of Oncology and Surgical Sciences, University of Padova, Via Gattamelata 64, 35128 Padova, Italy Istituto Oncologico Veneto (IOV), IRCCS, Via Gattamelata 64, 35128 Padova, Italy G. Desantis Department of Oncology and Surgical Sciences, University of Padova, Via Gattamelata 64, 35128 Padova, Italy M. Chioda Istituto Oncologico Veneto (IOV), IRCCS, Via Gattamelata 64, 35128 Padova, Italy V. Bronte (*) Verona University Hospital and Department of Pathology, Immunology Section, Piazzale L.A. Scuro 10, 37134 Verona, Italy e-mail: [email protected] T. Lawrence and T. Hagemann (eds.), Tumour-Associated Macrophages, DOI 10.1007/978-1-4614-0662-4_4, © Springer Science+Business Media, LLC 2012

49

50

S. Zilio et al.

Tumours can elude immune response by three main mechanisms: (1) cancer cells can mask their identity to escape recognition by the immune effectors, (2) they can directly modulate immunity, or (3) recruit other immune regulatory cells whose normal function is to dampen immune reactions to avoid the detrimental effects of an excessive immune stimulation. Probably, the most robust and efficient mechanism of “tumour escape” relies on tumour ability to create a tolerant microenvironment through the production of tumour-derived soluble factors (TDSFs). TDSFs modify the differentiation of myeloid precursors into either dendritic cells (DCs) or myeloid-derived suppressor cells (MDSCs) (Smyth et al. 2001; Drake et al. 2006) in the bone marrow and other hematopoietic organs. The imbalance in the number and type of myeloid cells has also profound influences on myeloid cell recruitment and function at the tumour site and in secondary lymphoid organs. The first part of this chapter considers cancer-induced modification in myelopoiesis, whereas the second one analyzes how tumour microenvironment can alter the function of recruited cells to enforce a tolerogenic mileu.

Tumour-Dependent Alteration in Myelopoiesis Results in Defective Accumulation of DCs and Enhanced MDSC Generation The presence of tumours alters the distribution and function of cells of the innate immune system. One of the main cell population targeted by this mechanism is represented by DCs, potent antigen-presenting cells (APCs) able to initiate the activation of naïve T cells (Guermonprez et al. 2002). In the bone marrow, hematopoietic progenitors cells (HPCs) give rise to immature DCs (iDCs). This first step of lineage commitment of the HPC is due to both exposure to soluble growth factors such as granulocyte/macrophage colony-stimulating factor (GM-CSF), interleukin-3 (IL-3), and fms-related tyrosine kinase 3 ligand (FLT3L) (Gabrilovich 2004) and to cell–cell contacts with bone marrow stroma cells. Differentiation of HPCs in iDCs is tightly regulated by the cooperation between the signaling pathways of Notch-Dll1 and the canonical Wnt cascade (Cheng et al. 2010). To achieve complete maturation, iDCs require further stimuli because, although able to uptake, process, and present antigens, they express little or no costimulatory molecules such CD80, CD86, and CD40 necessary to exert their full set of functions (Rabinovich et al. 2007). Maturation occurs physiologically thanks to microbial stimuli, proinflammatory cytokines, or CD40L-expressing T cells. During inflammation iDCs are activated by the contact with both exogenous antigens such as bacterial and viral products (for example LPS, flagellin, and viral particles) and endogenous antigens released from dying cells, as double-stranded RNA and other small molecules, such as urea and ATP (Macagno et al. 2007). In theory, stimuli for promoting iDC terminal differentiation are not induced only by microbial antigens, given that an inflammatory context is also often associated to several pathologies including cancer.

4

Tumour-Induced Immune Suppression by Myeloid Cells

51

When iDCs become activated DCs, they upregulate costimulatory and MHC class II molecules, increase IL-12 production and lose the ability to process antigens (Mellman and Steinman 2001; Banchereau et al. 2000). At the tumour site, the large number of tumour proteins released by dying or necrotic cancer cells represent, in theory, potent activators of DCs (Sauter et al. 2000). Activated DCs engulf these cells, process and present their antigens. Once they have reached the draining lymph nodes, DCs loaded with tumour antigens have the potential to initiate an immune response against tumour. However, despite this potential, DCs do not contribute to trigger an efficient immune response aimed at eliminating tumours, as often these manage to progress until the death of the host. One of the first explanations of this DC “failure” could derive from the fact that some tumours can affect the generation of functionally active DCs. This hypothesis is substantiated by the observation that tumour-bearing hosts have fewer activated DCs and that a physiological number of DCs can be restored upon resection of the primary tumour (Ishida et al. 1998; Gabrilovich et al. 1996a). Reduction of DCs was also measured in lymph nodes, skin, and spleen (Ishida et al. 1998). Growing tumours affect the number of activated DCs by inhibiting the transition from iDCs to DCs. In fact, DCs found at tumour site had a phenotype similar to the iDCs that leave the bone marrow after the initial phase of their differentiation. Furthermore, the higher number of iDCs stems out from defects in myelopoiesis rather than simply from the lack of appropriate activation signals at the tumour site. In vitro treatment of tumour-infiltrating DCs with appropriate stimuli (GM-CSF and TNF-D or CD40L) was in fact not sufficient to induce the expression of markers proper of activated DCs (Chaux et al. 1997). Taken together, this evidence supports the concept that the reduced functionality of DCs against some tumours is most probably due to defects in differentiation from their iDC progenitors. iDC are not the only myeloid cell population altered in tumour pathologies. In the bone marrow, hematopoietic stem cells give rise to the lymphoid (LP) and the myeloid (MP) multipotent precursors cells. From the MP originate other pluripotent cell types: the common DC (CDP) and the immature myeloid cell precursors (IMC); the first gives rise to iDCs and plasmacytoid dendritic cells (PDC), the second is the common progenitor of macrophages, granulocytes, and monocyte-derived DCs such as the Tip-DCs (TNF and inducible iNOS-producing DCs). In healthy organisms, IMCs rapidly differentiate into their descendant lineages; consequently they represent a relative low percentage of myeloid cells. However, under pathological conditions, including cancer, there is a partial block of differentiation of the IMCs leading to the accumulation of MDSCs and a consequent reduction of mature myeloidderived cells (Geissmann et al. 2010; Gabrilovich and Nagaraj 2009). Key features of MDSCs, besides their myeloid origin, are their immature phenotype and the exceptional ability to suppress T-cell-mediated immune response. (Gabrilovich et al. 2007). MDSCs are not a uniform cell population rather they include several subgroups distinct on their morphology and expression of surface markers. MDSCs have been described in several mouse tumour models and in human cancer patients. In mice, these cells are defined by the co-expression of the myeloid-cell lineage differentiation

52

S. Zilio et al.

marker Gr-1 and of CD11b (Serafini et al. 2004; Serafini et al. 2006). Based on the expression levels of Gr-1 (low, intermediate, and high), MDSCs cells were more precisely defined as a macropopulation composed by at least three different groups. (Dolcetti et al. 2010). MDSCs were also distinct depending on their phenotype and morphology into monocytic MDSCs (CD11b+Ly6C+F4/80+Gr-1low/int) and granulocytic MDSCs (CD11b+ Ly6G+F4/80-Gr-1high). MDSCs also express markers associated at the early stage of myeloid differentiation (CD31 and ER-MP58), low levels of MHC class II and co-stimulatory molecules (e.g., CD80), according to their origin from immature myelo-monocytic precursors (Bronte et al. 2000; Rossner et al. 2005). Although the reciprocal interplay among these subpopulations requires further studies, the immunosuppressive role of the monocytic MDSC subset has been confirmed by several groups (Dolcetti et al. 2010; Movahedi et al. 2008; Huang et al. 2006). Fundamental for the suppressive activity of MDSCs are their expansion and activation. Their activation is regulated by factors released by activated T-cells and tumour stromal cells, while their expansion is due mainly to factors secreted by the tumour itself (TDSFs). TDSFs induce both stimulation of myelopoiesis and the block of differentiation of myeloid cells with the consequent accumulation of immature myeloid precursors. It appears, therefore, that such a cancer-induced massive imbalance of myelopoiesis has profound consequences on the ability of the immune system to react to tumours: a lower number of mature DCs (DC maturation is blocked from both the iDC and the IMC branches) of macrophages and granulocytes results in a reduced potential of tumour antigen presentation, T-cell activation, and elimination of cancer cells. The immune system is then further compromised by the suppressive activity of the MDSC population (see next section). Many are the factors known to play a role in tumour-mediated MDSC or iDC expansion and alteration of myelopoiesis: they include cytokines, chemokines, and other factors such as prostaglandins, stem-cell factor (SCF), M-CSF, IL-6, GM-CSF, and the vascular endothelial growth factor (VEGF) (Gabrilovich and Nagaraj 2009). We will illustrate some examples of how TDSFs can regulate the number of mature and immature myeloid cells. VEGF is responsible of tumour-dependent angiogenesis and contributes to both tumour growth and metastasis. Tumour-produced VEGF (tVEGF) also impairs the recruitment and differentiation of DCs and expands MDSCs by suppressing the nuclear factor-kB (NF-kB) in hematopoietic progenitor cells. In facts, NF-kB is a key transcription factor in myelopoiesis regulating differentiation and survival of many precursors (Ouaaz et al. 2002) thus impacting on the composition of the mature myeloid population. Further, there is a feedback loop between tVEGF and MDSC expansion as treatment with tVEGF increases recruitment of MDSCs in the spleen of tumour-free mice; MDSCs, in turn, can regulate the bioavailability of the VEGF through the upregulation of matrix metallo-protease 9 (MMP9) (Yang et al. 2004). MMP9 acts as the molecular switch in tumour-dependent angiogenesis mediating release of VEGF (Oyama et al. 1998; Fricke et al. 2007). Through a VEGF receptor (R)2-mediated mechanism, VEGF was also shown to

4

Tumour-Induced Immune Suppression by Myeloid Cells

53

cause a profound alteration in hematopoiesis, by impairing the bone marrow hematopoiesis and, although, a parallel activation of extramedullary hematopoiesis in liver and spleen was seen, it was not sufficient to compensate for the defects of the first one (Xue et al. 2009). Another key cytokine is IL-10 as it can induce the downregulation of MHC II, co-stimulatory and adhesion molecules on DC surface, converting competent DCs in tolerogenic APCs (Steinbrink et al. 1997). IL-10 functions also in preventing differentiation of monocyte precursors to monocytes-derived DCs and rather forces them toward macrophagic lineage (Allavena et al. 1998; Buelens et al. 1997). Furthermore, DCs matured in the presence of IL-10 showed a reduced production of inflammatory cytokines and a lack of IL-12 synthesis, resulting in the induction of anergic CD8+ and CD4+ T regulatory lymphocytes, named CD8+/CD4+ Tr cells, with an antigen-specific suppressive activity. These cells present high expression of the CTLA-4, which is a inhibitor of T-cell activation (Koch et al. 1996; Steinbrink et al. 2002). However, the immunosuppressive activity of IL-10 is limited to iDCs since during terminal differentiation, DCs downregulate the IL-10 receptor, becoming non-responsive to IL-10 (Mahnke et al. 2002). IL-13 is a fundamental cytokine for the activation of the suppressive function of MDSCs. IL-13 interacts with the D chain of IL-4 receptor and cooperates with IFN-J required for the stable expression of IL-4RD on cell surface. IL-13 and IFN-J are essential for the induction of the two enzymes (ARG1 and NOS2) involved in l-arginine metabolism, whose upregulation has been linked to the immunosuppressive activation of MDSCs (see next section) (Gallina et al. 2006). Key roles in regulating myelopoiesis are played by M-CSF and IL-6, which also prevent myeloid precursor cells to become DCs and promote differentiation toward the monocytic/IMC lineage (Gabrilovich et al. 1996b; Menetrier-Caux et al. 1998). Treatment of iDC with IL-6 induces inability of these cells to respond to stimulation with LPS or TNF-D and peptidoglycan and therefore to achieve full maturation as DCs (Park et al. 2004). Another important cytokine involved in tumour tolerance is the granulocyte– monocyte colony stimulating factor (GM-CSF). GM-CSF is produced by a wide range of mouse and human tumour cell lines (Bronte et al. 1999). Chronic administration of GM-CSF results in the accumulation of MDSCs in the bone marrow, spleen, and blood of healthy mice mirroring the pattern induced by tumours (Bronte et al. 1999). Recent reports in both mice and humans show that bone marrow cells treated in vitro with GM-CSF and IL-6, become phenotipically similar to MDSCs induced by cancer. These in vitro-derived MDSCs also retain the ability to suppress in vivo the activity of T lymphocytes in an allogenic transplant models (Marigo et al. 2010). The molecules described above influence the ability of the HPCs in the bone marrow to commit to a given cell lineage by either favoring or preventing one of several differentiation routes. This is achieved molecularly by a restricted number of signal transduction pathways where many of these factors converge. Recent evidence suggests that at least one of these common pathways could be controlled by

54

S. Zilio et al.

Janus-activated kinase 2 (JAK2) and signal transducer and activator of transcription 3 (STAT3). The JAK and STAT families are involved in the transcriptional regulation of cellular survival, apoptosis, proliferation, and differentiation (Rane and Reddy 2000). JAK2 has been associated to the immune system because of its correlations to various cytokines and growth factor receptors including IL-2, IL-3, IL-4, IL-6, IL-10, IL-12, and IFN-D,E,J (Ihle et al. 1997; Imada and Leonard 2000). Further, direct stimulation of STAT3 activity in HPCs was demonstrated for all the described factors involved in DC dysfunction (Nefedova et al. 2004). Cytokines activate the JAK2 pathway causing the recruitment and phosphorylation of STAT3. Phosporylated STAT3 omodimerizes and traslocates to the nucleus where it regulates the expression of target genes involved in the control of cell survival and proliferation. Indeed, the aberrant activation of JAK or its constitutive expression has been observed in many tumours (Nosaka and Kitamura 2000; Steelman et al. 2004). The JAK2– STAT3 pathway is physiologically required for the early steps of differentiation of myeloid cells (Duarte and Franf 2002; Smithgall et al. 2000). Studies carried on conditional STAT3 knockout mice described the abrogation of the signaling pathway of Flt3L responsible for the activation of many hematopoietic progenitors and the consequent impairment in the development of DCs (Laouar et al. 2003). However, full differentiation of DCs requires a time-depending downregulation of STAT3 activation (Nefedova et al. 2004), emphasizing that a tight control of this signaling route is required for regulating various stages of myeloid differentiation and survival. (Nefedova et al. 2004). Another molecular switch involved in the expansion of MDSC population is represented by the signaling pathway of CCAAT-enhancer-binding protein-beta (C/EBPE). Tumour-produced GM-CSF, IL-6, and G-CSF are potent inducer of C/ EBPE a known regulator of ‘emergency’ granulopoiesis (Hirai et al. 2006). This factor plays a fundamental role in MDSCs expansion since the ablation of C/EBPE in the myeloid lineage results in a reduced accumulation of MDSCs, preferentially of the monocytic subset, in the spleen of tumour bearing mice. The reduced expansion of this population of MDSCs correlates with the full restoration of CD8+ T lymphocyte function and in a less invasive tumour behavior measured by the number of secondary metastasis (Marigo et al. 2010). Arginase 1 (ARG1) (Gray et al. 2005) and nitric oxide synthase 2 (NOS2) (Gupta and Kone 1999) are also regulated by C/ EBPE; therefore, the induction of this transcriptional factor by tumour-derived GM-CSF, IL-6, or G-CSF is one of the direct links between tumour and myelopoiesis resulting in the expansion of a MDSC monocytic population permissive to tumour and metastasis (Marigo et al. 2010). We described TDSFs as the main mechanism by which tumour can alter homeostasis of myelopoiesis. It is important to mention that TDSFs normally act locally and they rapidly undergo degradation in the blood stream. Tumour can use an additional strategy, physiologically employed for trafficking molecules. Recent evidences point out that cancer cells could release microvescicles containing bioactive molecules, nucleic acid, and protein (Mytar et al. 2008; Ratajczak et al. 2006; Cocucci et al. 2009). Probably, by protecting TDSFs from degradation, this mechanism facilitates their delivery to distant tissues, such as the bone marrow.

4

Tumour-Induced Immune Suppression by Myeloid Cells

55

Generation of Tumour Permissive Microenvironment The ongoing discovery of the fine regulation of the adaptive immunity by myeloid cells (a sort of “historic revaluation” of the innate immunity importance in antigenic responses) is providing us some answers to the puzzle regarding the apparent loss of immunocompetence at tumour sites. To simplify the complexity of this regulation, we can say that tumour-established suppressive network aims at two main goals: the direct inhibition of CTL antitumoural responses and the creation and maintenance of an immunosuppressive environment where this is needed, with the consequent induction of regulatory phenotypes in cells of the immune system that reside in those sites. To achieve these two goals, tumour-induced myeloid populations must first migrate to strategic key points throughout the organism, i.e. secondary lymphoid organs and the tumour site. The spleen in tumour-bearing hosts is an evident example of the abnormal tumour-driven accumulation of myeloid cells with immunosuppressive activity, most of which are CD11b+ Gr-1+ (in case of the mouse host) MDSCs (Gabrilovich and Nagaraj 2009; Diaz-Montero et al. 2009; Almand et al. 2001). These cells are very powerful suppressors of CD8+ T-cell response; their ability requires direct cell-to-cell contact, antigen presentation through MHC class I (Gabrilovich et al. 2001) and acts through the generation of free radicals and metabolism of l-arginine in the microenvironment by ARG1 and NOS2 enzymes (Kusmartsev et al. 2004; Bronte and Zanovello 2005). Although matter of debate, it seems that MDSC immunosuppression may be both antigen- or non-antigen-specific, depending on the tissue where these cells reside. In peripheral lymphoid organs, MDSCs seem to suppress in an antigen-specific manner. Among the possible mechanisms, the post-trasductional modifications of the TCR through the generation of peroxynitrites at the site of the immunologic synapse has been advanced (Willimsky et al. 2008; Nagaraj et al. 2007). Peroxynitrites (ONOO−) are generated by the combination of reactive oxygen species (ROS) with nitric oxide (NO), mainly produced by NOS2 with minor contribution of other NOS isoforms. During presentation of antigens, MHC-I molecules expressed by MDSCs engage with the TCR of CD8+ T cells, and this strict interaction favors the nitration and nitrosylation of TCR amino acids by MDSC-released peroxynitrites, finally resulting in the modification of the TCR specificity against those antigens presented in the context of MHC-I on MDSC. Reduced TCR specificity causes an impairment of TCR ability to respond to antigenic stimulation (Nagaraj et al. 2007). MDSC generation has another important effect on lymphoid tissues. As we discussed earlier, an increased production of these cells will result in a lower number of mature competent professional APCs (i.e., DCs) in lymphoid tissues (Ishida et al. 1998; Almand et al. 2000; Della Bella et al. 2003); moreover, the few DCs present in lymph nodes and tumour site display an immature phenotype. The lack of stimuli due to the absence of mature DCs has an active detrimental effect on T-cell function, causing anergy, which is a state of non-responsiveness to antigenic stimulation (Bonifaz et al. 2002). In this way, all tumour antigens presented in draining lymph nodes by iDCs will induce tolerance instead of immune reaction. Interestingly,

56

S. Zilio et al.

a class of lymphoid DCs called plasmacytoid DCs is not affected by tumour-induced alteration in myelopoiesis (Della Bella et al. 2003; Hoffmann et al. 2002), and this particular DCs can induce IL-10+ CCR7+ suppressive CD8+ T cells, with inhibitory effects on priming of tumour-specific T-cell in draining lymph nodes (Wei et al. 2005). All these phenomena may explain how the tumour manages to suppress the immune response against its own antigens, while leaving substantially unaltered the immunocompetence against general non-self-antigens. At the tumour site, the aim of dampening the immune response is achieved by a quite different strategy. In this case, MDSC suppression is believed to be non-antigenspecific and to be mediated by the concomitant upregulation of ARG1 and NOS2 induced by local stimuli (Gabrilovich and Nagaraj 2009; Gallina et al. 2006). The effects of the combined activity of these two enzymes (Bronte and Zanovello 2005) bring to a general inability of T cells to respond to proliferative stimuli, leading them to apoptosis, eventually. ARG1 and NOS2 have as common substrate l-arginine. Upregulation of these two enymatic activities implies a greater consumption of this essential aminoacid. l-Arginine starvation limits protein synthesis and reduces the re-expression of the ]-chain subunit of the TCR, which is usually internalized after TCR-mediated stimulation (Baniyash 2004). This mechanism could be mediated by GCN2, a sensor of amino acid deprivation, whose activation leads to an impairment of protein synthesis (Zhang et al. 2002; Lee et al. 2003). A second similar strategy operated to reduce immunocompetence at tumour sites, consists in controlling the availability of other essential nutrients to limit the proliferation of particular cell types, as in the case of indoleamine 2,3 dioxygenase (IDO) enzyme, which catalyze tryptophane degradation, expressed in a particular subset of DCs (Munn et al. 2002; Uyttenhove et al. 2003). An alternative way through which MDSCs achieve non-antigen-specific suppression is the production of NO by NOS2. NO levels impair IL-2 signaling by destabilizing IL-2 mRNA (Fischer et al. 2001; Macphail et al. 2003) and by blocking downstream effectors of the IL-2 receptor signaling pathway such as JAK1, JAK3, STAT5, ERK, and AKT (Bingisser et al. 1998; Mazzoni et al. 2002). The expression of ARG1 and NOS2 in MDSCs is a peculiar feature of these cells, which makes them quite different from tumour-associated macrophages (TAMs). In fact, the mutually exclusive expression of these enzymes has been historically used to define the polarization of mouse macrophages driven by antithetical cytokines. NOS2 is generally considered a marker of M1 macrophage orientation, while ARG1 defines the M2 state. Induction of both NOS2 and ARG1 is not only a way through which MDSCs can exert suppression in tumour environment, but these enzymes are also necessary for the MDSCs in lymphoid compartment to acquire full suppressive potential. At least in some tumour models, splenic MDSCs need to be armed by antithetic cytokines (IFN-J and IL-4/IL-13) to sustain a long-lasting upregulation of both NOS2 and ARG1 (Gallina et al. 2006). In particular, IFN-J released by antigen-activated CD8+ T cells and a membrane-coupled signal are sensed by MDSCs, which react to this stimulus by autocrine secretion of both IFN-J and IL-13. The combination of the signal transduction pathways activated by these two

4

Tumour-Induced Immune Suppression by Myeloid Cells

57

cytokines (which requires the expression of the marker CD124, the D chain of the IL-4/IL-13 receptor), leads MDSCs to upregulate ARG1 and NOS2 (Gallina et al. 2006). Interestingly, CD124 is one of the first markers to be upregulated in bone marrow cells under cytokine regimes that generate human and mouse MDSCs (Marigo et al. 2010). The peculiarity of MDSCs to respond with an M2-oriented program to Th1-cytokines is common to TAMs (Sica and Bronte 2007), but in both cell types the molecular ground for this phenotypic behavior needs further investigation. Actually, the relationship between MDSCs and TAMs may be an “ontologic” one. Some data suggest that, once they reach the tumour site, circulating MDSCs are capable of differentiating into TAMs, as indicated by the downregulation of Gr-1 and upregulation of F4/80 surface markers (Kusmartsev and Gabrilovich 2005); given the strict connection in their molecular pathways, just discussed, this could indeed be the case (Sica and Bronte 2007). Although TAMs are capable of expressing either ARG1 or NOS2, their suppressive capacity has been mainly associated with their ability to secrete large amounts of IL-10, TGF-J, and prostaglandins (Mantovani et al. 2002). Due to their extraordinary ability to release a wide range of cytokines, TAMs are some of the main actors of the immunosoppressive environment established at the tumour site. For example, they attract Tregs through the release of CCL22 (Curiel et al. 2004), and Treg regulatory activity is enhanced by the secretion of IL-10 and TGF-E. The role of TAMs as “masters of puppets” in the induction of other regulatory cells is strictly connected with the relationship between cancer and inflammation illustrated by the model of the “balance needle.” For several years, a relationship between chronic inflammation and incidence of cancer has been observed, and tumour is often considered an example of smolder inflammation (Mantovani et al. 2008). Although many issues need to be clarified, neoplastic lesions are thought to begin with an initial phase of inflammation, which promotes the neoplastic transformation and evolve into a state of persistent mild inflammation. This latter favors proliferation and survival of tumour cells, promotes angiogenesis, metastasis and prevents an opportune immune reaction to be initiated. As a confirmation of this, the transcriptome of human tumour cells in which the oncogene RET has been activated (which is sufficient for the pathogenesis of papillary thyroid carcinoma), includes among others upregulation of mRNAs encoding for various inflammation-linked molecules, such as colony stimulating factors (CSFs), IL1-E, COX2 (which synthetize prostaglandins), chemokines attracting monocytes, and DCs (CCL2 and CCL20) (Borrello et al. 2005). TAMs could represent the most probable candidates for explaining the link between cancer and inflammation as they inflict tissue damage during chronic inflammation and favor tumourigenesis (through an M1 phenotype), while sustain smolder inflammation during tumour progression (through an M2 phenotype) (Sica and Bronte 2007). A molecular explanation for the polyedric behavior of macrophages during various steps of tumourigenesis can be related to the transcription factor NF-kB. Given the broad implications that NF-kB has on cell

58

S. Zilio et al.

survival, proliferation, and differentiation, its link to TAMs is described in depth in a separate chapter of this book. According to the model of the “balance needle,” the immunologic fate of an antigen (induction of response or tolerance) depends on the kind of signals provided in the context of its expression and/or presentation. Each signal (cytokines, co-stimulatory signals, activity of immunoregulatory enzymes, etc.) supports the preservation or the elimination of that antigen, and their combination determines its fate. At the tumour site, there is a clear imbalance toward immunological tolerance to or disregard of tumour antigens, due to the secretion of molecules that suppress DC maturation, as above described (Gabrilovich 2004; Zou 2005); these molecules are secreted by tumour cells, stromal cells, and TAMs. Further, low levels of cytokines necessary for DC differentiation (e.g., IL-4, IL-12, and IFN-J) are produced at the tumours site, where the presence of a smolder inflammation phenotype also implies that insufficient danger/activating signals are delivered. These conditions lead to the accumulation of iDCs responsible for CD8+ T cell anergy. Furthermore, the imbalance of cytokines induces a regulatory behavior of DCs together with the expression of co-inhibitory molecules such as B7-H1 and B7-H4 (Zou and Chen 2008), which determines both T-cell unresponsiveness and generation of Tregs (Krupnick et al. 2005). TAMs contribute to the establishment of a tolerogenic microenvironment also by the secretion of chemokines (in particular CCL2, CCL22, and CCL18) that attract other regulatory cells. CCL2 is the principal chemoattractant for TAMs (Mantovani et al. 2002), CCL22 for Tregs, while CCL18 acts on naïve T cells (Schutyser et al. 2002), which are caught in a signaling web where they are rendered either anergic, tolerant, or regulatory. MDSCs seem to contribute to the induction of Tregs, but this aspect of their biology has been poorly studied so far. Several works have shown, both in vitro and in vivo, that MDSCs are able to expand Tregs through stimulation of natural occurring Tregs, instead of converting CD4+ T effector cells into regulatory ones (Huang et al. 2006; Ghiringhelli et al. 2005; Serafini et al. 2008; Yang et al. 2006; Liu et al. 2008; MacDonald et al. 2005). Many mechanisms for the MDSC-mediated Treg expansion have been proposed: these involve the secretion of IFN-J, IL-10, and TGF-E, the expression of arginase and the co-stimulatory molecules CD80 and B7-H1. Although these data propose different models and require further validation, the majority of them clearly shows the need for MHC class II-restricted antigen presentation in MDSC-Treg direct interaction. Moreover, it has been shown that APCs acquire the ability of expanding Tregs after direct or indirect interaction with tumour cells, both in vivo and in vitro (Ghiringhelli et al. 2005; Serafini et al. 2008; Yang et al. 2006). However, this concept is still a matter of debate as other lines of evidence have shown no correlation between MDSC expansion and induction of Tregs during tumour progression (Movahedi et al. 2008; Dugast et al. 2008). In summary, the induction of immunoregulatory phenotypes at tumour site is a complex network where signal transduction, transcriptional regulation, cytokines secretion, surface molecules, nutrients metabolism, and reciprocal cell–cell interaction are tightly bound. Tumour-induced tolerance is not simply hierarchical or unidirectional and it requires fine modulation in the function of lymphoid cells involved, as they have the potential of either inducing tolerance or initiating an immune reaction:

4

Tumour-Induced Immune Suppression by Myeloid Cells

59

for example, Tregs can trigger alternative activation of macrophages or suppress the expression of IL-12 in DCs, by reverse signaling of B7-H1 (Tiemessen et al. 2007; Curiel et al. 2003).

References Allavena P et al (1998) IL-10 prevents the differentiation of monocytes to dendritic cells but promotes their maturation to macrophages. Eur J Immunol 28(1):359–369 Almand B et al (2000) Clinical significance of defective dendritic cell differentiation in cancer. Clin Cancer Res 6(5):1755–1766 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 Banchereau J et al (2000) Immunobiology of dendritic cells. Annu Rev Immunol 18:767–811 Baniyash M (2004) TCR zeta-chain downregulation: curtailing an excessive inflammatory immune response. Nat Rev Immunol 4(9):675–687 Bingisser RM et al (1998) Macrophage-derived nitric oxide regulates T cell activation via reversible disruption of the Jak3/STAT5 signaling pathway. J Immunol 160(12):5729–5734 Bonifaz L et al (2002) Efficient targeting of protein antigen to the dendritic cell receptor DEC-205 in the steady state leads to antigen presentation on major histocompatibility complex class I products and peripheral CD8+ T cell tolerance. J Exp Med 196(12):1627–1638 Borrello MG et al (2005) Induction of a proinflammatory program in normal human thyrocytes by the RET/PTC1 oncogene. Proc Natl Acad Sci USA 102(41):14825–14830 Bronte V, Zanovello P (2005) Regulation of immune responses by L-arginine metabolism. Nat Rev Immunol 5(8):641–654 Bronte V et al (1999) Unopposed production of granulocyte-macrophage colony-stimulating factor by tumors inhibits CD8+ T cell responses by dysregulating antigen-presenting cell maturation. J Immunol 162(10):5728–5737 Bronte V et al (2000) Identification of a CD11b(+)/Gr-1(+)/CD31(+) myeloid progenitor capable of activating or suppressing CD8(+) T cells. Blood 96(12):3838–3846 Buelens C et al (1997) Human dendritic cell responses to lipopolysaccharide and CD40 ligation are differentially regulated by interleukin-10. Eur J Immunol 27(8):1848–1852 Chaux P et al (1997) Tumor-infiltrating dendritic cells are defective in their antigen-presenting function and inducible B7 expression in rats. Int J Cancer 72(4):619–624 Cheng P, Zhou J, Gabrilovich D (2010) Regulation of dendritic cell differentiation and function by Notch and Wnt pathways. Immunol Rev 234(1):105–119 Cocucci E, Racchetti G, Meldolesi J (2009) Shedding microvesicles: artefacts no more. Trends Cell Biol 19(2):43–51 Curiel TJ et al (2003) Blockade of B7-H1 improves myeloid dendritic cell-mediated antitumor immunity. Nat Med 9(5):562–567 Curiel TJ et al (2004) Specific recruitment of regulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survival. Nat Med 10(9):942–949 Della Bella S et al (2003) Altered maturation of peripheral blood dendritic cells in patients with breast cancer. Br J Cancer 89(8):1463–1472 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 Dolcetti L et al (2010) Hierarchy of immunosuppressive strength among myeloid-derived suppressor cell subsets is determined by GM-CSF. Eur J Immunol 40(1):22–35 Drake CG, Jaffee E, Pardoll DM (2006) Mechanisms of immune evasion by tumors. Adv Immunol 90:51–81

60

S. Zilio et al.

Duarte RF, Franf DA (2002) The synergy between stem cell factor (SCF) and granulocyte colonystimulating factor (G-CSF): molecular basis and clinical relevance. Leuk Lymphoma 43(6): 1179–1187 Dugast AS et al (2008) Myeloid-derived suppressor cells accumulate in kidney allograft tolerance and specifically suppress effector T cell expansion. J Immunol 180(12):7898–7906 Dunn GP, Old LJ, Schreiber RD (2004) The three Es of cancer immunoediting. Annu Rev Immunol 22:329–360 Fischer TA et al (2001) Activation of cGMP-dependent protein kinase Ibeta inhibits interleukin 2 release and proliferation of T cell receptor-stimulated human peripheral T cells. J Biol Chem 276(8):5967–5974 Fricke I et al (2007) Vascular endothelial growth factor-trap overcomes defects in dendritic cell differentiation but does not improve antigen-specific immune responses. Clin Cancer Res 13(16):4840–4848 Gabrilovich D (2004) Mechanisms and functional significance of tumour-induced dendritic-cell defects. Nat Rev Immunol 4(12):941–952 Gabrilovich DI, Nagaraj S (2009) Myeloid-derived suppressor cells as regulators of the immune system. Nat Rev Immunol 9(3):162–174 Gabrilovich DI, Ciernik IF, Carbone DP (1996a) Dendritic cells in antitumor immune responses. I. Defective antigen presentation in tumor-bearing hosts. Cell Immunol 170(1):101–110 Gabrilovich DI et al (1996b) Production of vascular endothelial growth factor by human tumors inhibits the functional maturation of dendritic cells. Nat Med 2(10):1096–1103 Gabrilovich DI et al (2001) Mechanism of immune dysfunction in cancer mediated by immature Gr-1+ myeloid cells. J Immunol 166(9):5398–5406 Gabrilovich DI et al (2007) The terminology issue for myeloid-derived suppressor cells. Cancer Res 67(1):425, author reply 426 Gallina G et al (2006) Tumors induce a subset of inflammatory monocytes with immunosuppressive activity on CD8+ T cells. J Clin Invest 116(10):2777–2790 Geissmann F et al (2010) Development of monocytes, macrophages, and dendritic cells. Science 327(5966):656–661 Ghiringhelli F et al (2005) Tumor cells convert immature myeloid dendritic cells into TGF-betasecreting cells inducing CD4 + CD25+ regulatory T cell proliferation. J Exp Med 202(7): 919–929 Gray MJ et al (2005) Induction of arginase I transcription by IL-4 requires a composite DNA response element for STAT6 and C/EBPbeta. Gene 353(1):98–106 Guermonprez P et al (2002) Antigen presentation and T cell stimulation by dendritic cells. Annu Rev Immunol 20:621–667 Gupta AK, Kone BC (1999) CCAAT/enhancer binding protein-beta trans-activates murine nitric oxide synthase 2 gene in an MTAL cell line. Am J Physiol 276(4 Pt 2):F599–F605 Hirai H et al (2006) C/EBPbeta is required for ‘emergency’ granulopoiesis. Nat Immunol 7(7): 732–739 Hoffmann TK et al (2002) Alterations in the frequency of dendritic cell subsets in the peripheral circulation of patients with squamous cell carcinomas of the head and neck. Clin Cancer Res 8(6):1787–1793 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 Ihle JN et al (1997) Jaks and Stats in cytokine signaling. Stem Cells 15(Suppl 1):105–111, discussion 112 Imada K, Leonard WJ (2000) The Jak-STAT pathway. Mol Immunol 37(1–2):1–11 Ishida T et al (1998) Defective function of Langerhans cells in tumor-bearing animals is the result of defective maturation from hemopoietic progenitors. J Immunol 161(9):4842–4851 Koch F et al (1996) High level IL-12 production by murine dendritic cells: upregulation via MHC class II and CD40 molecules and downregulation by IL-4 and IL-10. J Exp Med 184(2): 741–746

4

Tumour-Induced Immune Suppression by Myeloid Cells

61

Krupnick AS et al (2005) Murine vascular endothelium activates and induces the generation of allogeneic CD4 + 25 + Foxp3+ regulatory T cells. J Immunol 175(10):6265–6270 Kusmartsev S, Gabrilovich DI (2005) STAT1 signaling regulates tumor-associated macrophagemediated T cell deletion. J Immunol 174(8):4880–4891 Kusmartsev S et al (2004) Antigen-specific inhibition of CD8+ T cell response by immature myeloid cells in cancer is mediated by reactive oxygen species. J Immunol 172(2): 989–999 Laouar Y et al (2003) STAT3 is required for Flt3L-dependent dendritic cell differentiation. Immunity 19(6):903–912 Lee J et al (2003) Translational control of inducible nitric oxide synthase expression by arginine can explain the arginine paradox. Proc Natl Acad Sci USA 100(8):4843–4848 Liu Y et al (2008) B7-H1 on myeloid-derived suppressor cells in immune suppression by a mouse model of ovarian cancer. Clin Immunol 129(3):471–481 Macagno A et al (2007) Duration, combination and timing: the signal integration model of dendritic cell activation. Trends Immunol 28(5):227–233 MacDonald KP et al (2005) Cytokine expanded myeloid precursors function as regulatory antigenpresenting cells and promote tolerance through IL-10-producing regulatory T cells. J Immunol 174(4):1841–1850 Macphail SE et al (2003) Nitric oxide regulation of human peripheral blood mononuclear cells: critical time dependence and selectivity for cytokine versus chemokine expression. J Immunol 171(9):4809–4815 Mahnke K et al (2002) Immature, but not inactive: the tolerogenic function of immature dendritic cells. Immunol Cell Biol 80(5):477–483 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 (2008) Cancer-related inflammation. Nature 454(7203):436–444 Marigo I et al (2010) Tumor-Induced Tolerance and Immune Suppression Depend on the C/EBP[beta] Transcription Factor. Immunity 32(6):790–802 Mazzoni A et al (2002) Myeloid suppressor lines inhibit T cell responses by an NO-dependent mechanism. J Immunol 168(2):689–695 Mellman I, Steinman RM (2001) Dendritic cells: specialized and regulated antigen processing machines. Cell 106(3):255–258 Menetrier-Caux C et al (1998) Inhibition of the differentiation of dendritic cells from CD34(+) progenitors by tumor cells: role of interleukin-6 and macrophage colony-stimulating factor. Blood 92(12):4778–4791 Movahedi K et al (2008) Identification of discrete tumor-induced myeloid-derived suppressor cell subpopulations with distinct T cell-suppressive activity. Blood 111(8):4233–4244 Munn DH et al (2002) Potential regulatory function of human dendritic cells expressing indoleamine 2,3-dioxygenase. Science 297(5588):1867–1870 Mytar B et al (2008) Human monocytes both enhance and inhibit the growth of human pancreatic cancer in SCID mice. Anticancer Res 28(1A):187–192 Nagaraj S et al (2007) Altered recognition of antigen is a mechanism of CD8+ T cell tolerance in cancer. Nat Med 13(7):828–835 Nefedova Y et al (2004) Hyperactivation of STAT3 is involved in abnormal differentiation of dendritic cells in cancer. J Immunol 172(1):464–474 Nosaka T, Kitamura T (2000) Janus kinases (JAKs) and signal transducers and activators of transcription (STATs) in hematopoietic cells. Int J Hematol 71(4):309–319 Ouaaz F et al (2002) Dendritic cell development and survival require distinct NF-kappaB subunits. Immunity 16(2):257–270 Oyama T et al (1998) Vascular endothelial growth factor affects dendritic cell maturation through the inhibition of nuclear factor-kappa B activation in hemopoietic progenitor cells. J Immunol 160(3):1224–1232 Park SJ et al (2004) IL-6 regulates in vivo dendritic cell differentiation through STAT3 activation. J Immunol 173(6):3844–3854

62

S. Zilio et al.

Rabinovich GA, Gabrilovich D, Sotomayor EM (2007) Immunosuppressive strategies that are mediated by tumor cells. Annu Rev Immunol 25:267–296 Rane SG, Reddy EP (2000) Janus kinases: components of multiple signaling pathways. Oncogene 19(49):5662–5679 Ratajczak J et al (2006) Embryonic stem cell-derived microvesicles reprogram hematopoietic progenitors: evidence for horizontal transfer of mRNA and protein delivery. Leukemia 20(5): 847–856 Rossner S et al (2005) Myeloid dendritic cell precursors generated from bone marrow suppress T cell responses via cell contact and nitric oxide production in vitro. Eur J Immunol 35(12): 3533–3544 Sauter B et al (2000) Consequences of cell death: exposure to necrotic tumor cells, but not primary tissue cells or apoptotic cells, induces the maturation of immunostimulatory dendritic cells. J Exp Med 191(3):423–434 Schutyser E et al (2002) Identification of biologically active chemokine isoforms from ascitic fluid and elevated levels of CCL18/pulmonary and activation-regulated chemokine in ovarian carcinoma. J Biol Chem 277(27):24584–24593 Serafini P et al (2004) Derangement of immune responses by myeloid suppressor cells. Cancer Immunol Immunother 53(2):64–72 Serafini P et al (2006) Phosphodiesterase-5 inhibition augments endogenous antitumor immunity by reducing myeloid-derived suppressor cell function. J Exp Med 203(12):2691–2702 Serafini P et al (2008) Myeloid-derived suppressor cells promote cross-tolerance in B-cell lymphoma by expanding regulatory T cells. Cancer Res 68(13):5439–5449 Sica A, Bronte V (2007) Altered macrophage differentiation and immune dysfunction in tumor development. J Clin Invest 117(5):1155–1166 Smithgall TE et al (2000) Control of myeloid differentiation and survival by Stats. Oncogene 19(21):2612–2618 Smyth MJ, Godfrey DI, Trapani JA (2001) A fresh look at tumor immunosurveillance and immunotherapy. Nat Immunol 2(4):293–299 Steelman LS et al (2004) JAK/STAT, Raf/MEK/ERK, PI3K/Akt and BCR-ABL in cell cycle progression and leukemogenesis. Leukemia 18(2):189–218 Steinbrink K et al (1997) Induction of tolerance by IL-10-treated dendritic cells. J Immunol 159(10):4772–4780 Steinbrink K et al (2002) CD4(+) and CD8(+) anergic T cells induced by interleukin-10-treated human dendritic cells display antigen-specific suppressor activity. Blood 99(7):2468–2476 Tiemessen MM et al (2007) CD4 + CD25 + Foxp3+ regulatory T cells induce alternative activation of human monocytes/macrophages. Proc Natl Acad Sci USA 104(49):19446–19451 Uyttenhove C et al (2003) Evidence for a tumoral immune resistance mechanism based on tryptophan degradation by indoleamine 2,3-dioxygenase. Nat Med 9(10):1269–1274 Wei S et al (2005) Plasmacytoid dendritic cells induce CD8+ regulatory T cells in human ovarian carcinoma. Cancer Res 65(12):5020–5026 Willimsky G et al (2008) Immunogenicity of premalignant lesions is the primary cause of general cytotoxic T lymphocyte unresponsiveness. J Exp Med 205(7):1687–1700 Xue Y et al (2009) Tumor-derived VEGF modulates hematopoiesis. J Angiogenes Res 1:9 Yang L et al (2004) Expansion of myeloid immune suppressor Gr + CD11b + cells in tumor-bearing host directly promotes tumor angiogenesis. Cancer Cell 6(4):409–421 Yang R et al (2006) CD80 in immune suppression by mouse ovarian carcinoma-associated Gr-1 + CD11b + myeloid cells. Cancer Res 66(13):6807–6815 Zhang P et al (2002) The GCN2 eIF2alpha kinase is required for adaptation to amino acid deprivation in mice. Mol Cell Biol 22(19):6681–6688 Zou W (2005) Immunosuppressive networks in the tumour environment and their therapeutic relevance. Nat Rev Cancer 5(4):263–274 Zou W, Chen L (2008) Inhibitory B7-family molecules in the tumour microenvironment. Nat Rev Immunol 8(6):467–477

Chapter 5

TAM: A Moving Clinical Target Simon Hallam and Thorsten Hagemann

TAM and Potential Therapeutic Strategies As previously detailed, TAM are an important component of a tumour-supporting microenvironment and consist of newly recruited and resident tissue macrophages. TAM have roles in supporting angiogenesis, tumour growth, invasion and metastasis. TAM phenotype is variously described as “wound healing” and it has been postulated that this may indeed describe how they have evolved this phenotypic response to the presence of tumours. It is possible that TAM are, maladaptively, responding to tumour-inflicted tissue damage in a similar way that they would to sterile wounds, and so display a pro-angiogenic and immune-suppressive phenotype. Rational therapeutic approaches can be considered in terms of whether they involve macrophage ablation, inhibition of monocyte recruitment, re-education or adoptive transfer.

Macrophage Ablation An attractive strategy would be to selectively ablate pro-tumoural TAM, while preserving physiological macrophages in other tissues. In the 1980s, a “macrophage suicide” technique was developed, whereby liposome-encapsulated hydrophilic bisphosphonates were delivered into cells, exploiting the preferential engulfment of liposomes by cells of the mononuclear–phagocyte system. Intracellular accumulation of bisphosphonate led to apoptosis, primarily of monocytes and tissue macrophages. The most effective bisphosphonate for this purpose has been clodronate

3(ALLAMs4(AGEMANN*) Centre for Cancer and Inflammation, Barts Cancer Institute, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, London EC1M 6BQ, UK e-mail: [email protected]; [email protected] T. Lawrence and T. Hagemann (eds.), Tumour-Associated Macrophages, DOI 10.1007/978-1-4614-0662-4_5, © Springer Science+Business Media, LLC 2012

63

64

S. Hallam and T. Hagemann

(dichloromethylene bisphosphonate) (Van Rooijen & Sanders 1996). Liposomal clodronate has been extensively used to try and demonstrate proof-of-concept in mouse models of cancer (Zeisberger et al. 2006; Gazzaniga et al. 2007; Hiraoka et al. 2008; Miselis et al. 2008), to establish a functional relationship between the presence of TAM and tumour progression. Liposomal clodronate was recently evaluated as therapy in dogs with a spontaneous histiocytic neoplasm (Hafeman et al. 2010). Two of five dogs demonstrated tumour regression with an intravenous schedule of 0.5 ml/kg over 60 min, repeated once or twice at 2-week intervals. The authors believe that direct tumour cell killing was at least partly responsible for the clinical response observed in this situation, rather than only that mediated by macrophage depletion. Nonetheless, this work showed that such a short course of liposomal clodronate was feasible and well tolerated. Encouraging results have recently been published with a murine model whereby macrophage ablation with liposomal clodronate given prior to radiotherapy inhibits tumour relapse (Meng et al. 2010). Identification of such critical therapeutic windows for macrophage ablation, perhaps to augment responses to chemotherapy or radiotherapy, offers the prospect of optimising anti-tumour effects while minimising detrimental effects to physiological processes. Consequently, there have been calls for urgent clinical trials of liposomal clodronate in humans (Hafeman et al. 2010; Meng et al. 2010). However, this relatively unselective macrophage ablation technique destroys many physiological tissue macrophages, and in its unrefined form is likely to be limited in its human application. Systemic delivery of un-encapsulated bisphosphonates has been common clinical practise for many years, intended to inhibit osteoclast-mediated bone resorption in the setting of osteoporosis, multiple myeloma, and cancer metastasising to bone. Intriguingly, recent data strongly suggests that the effects of systemic bisphosphonates go beyond simply reducing bone resorption or metastasis, also appearing to modulate overall cancer progression. In a clinical trial of 1,803 pre-menopausal women with oestrogen-responsive early breast cancer, the addition of the intravenous bisphosphonate zoledronic acid, improved disease-free survival (Gnant et al. 2009). There were reductions in loco-regional recurrence, distant recurrence, bone metastasis and disease in the contralateral breast, implying anti-tumour effects extending beyond the bone microenvironment. Similarly, the Medical Research Council Myeloma IX Study of 1,960 patients is reporting a survival benefit from zoledronic acid, distinct from that associated with the prevention of skeletal-related events, again implying anti-cancer properties by mechanisms that remain unclear (Oral presentation, Prof G. Morgan ASCO 2010: Evaluating the Effects of Zoledronic Acid (ZOL) on Overall Survival (OS) in Patients (Pts) With Multiple Myeloma (MM): Results of the Medical Research Council (MRC) Myeloma IX Study. Abstract 8021). The contribution of macrophages to this effect remains unclear, as there may well also be direct anti-cancer cell effects. In vivo work in a mouse model of mesothelioma has indicated that systemic zoledronic acid might impair myeloid cell differentiation to tumour-associated macrophages (Veltman et al. 2010), and further work is needed to clarify and confirm this mechanism of action.

5 TAM: A Moving Clinical Target

65

Inhibition of Monocyte Recruitment A recent seminal study has exploded the dogma of TAM simply arising from circulating monocytes recruited to sites of cancer-related inflammation by chemokines including CSF-1 (Macdonald et al. 2010). This work reveals that monocyte production and inflammatory function are independent of maturation and replacement of resident tissue macrophages. It is the latter population of resident macrophages, whose development depends upon intact CSF1-Receptor signalling, that constitute TAM in the context of malignancy. In a tumour model, administration of a novel CSF1R blocking antibody reduced the numbers of TAM, but had no effect on monocyte recruitment using non-cancer inflammation models induced by lipo-polysaccharide, wound healing, peritonitis or acute graft-versus-host disease. Identification that TAM arise along separate maturation pathways from inflammatory monocyte/macrophages raises the prospect of reducing TAM numbers without severely impairing physiological monocyte/macrophage functions. This is conceptually more appealing than the far less selective macrophage ablation achieved with liposomal clodronate. It will be exciting to follow the development of human trials of CSF1R blocking antibodies or small molecules. Further insights into leukocyte trafficking have been gained by the study of a humanised monoclonal antibody against CD11b and CD18 cell adhesion proteins, expressed on macrophages and neutrophils. In a rabbit model of ischaemic brain injury, Rovelizumab (LeukArrest, ICOS) reduced neutrophil infiltration by 90%. Rovelizumab is well tolerated in humans and entered Phase III clinical trials in the late 1990s for use in ischaemic stroke (Jones 2000). Development of this drug for reducing inflammatory insults after ischaemic injury was halted in the light of disappointing clinical outcomes in this and other inflammatory conditions. Interest in Rovelizumab has been re-ignited by work investigating post-radiotherapy use of a CD11b antibody in a xenograft model of human hypopharyngeal carcinoma in mice (Ahn et al. 2010). Radiotherapy is known to induce tissue hypoxia, which in turn drives not only macrophage infiltration but also a pro-tumoural phenotype (Vujaskovic et al. 2001). In this way, TAM may promote tumour re-growth following radiotherapy. Having observed recruitment of CD11b + myeloid cells to tumours growing in previously irradiated tissues (Ahn and Brown 2008), it was found that systemic administration of a neutralising CD11b monoclonal antibody following irradiation reduced infiltration of myeloid cells, which correlated with the inhibition of tumour re-growth. Additionally, murine tumours transplanted to CD18 hypomorphic mice were more sensitive to irradiation than wild-type controls. CD11b and CD18 are expressed on such a large population of circulating and resident macrophages and neutrophils that timing of administration of neutralising antibodies is likely to be crucial when trying to influence TAM infiltration. CX3CL1, also known as Fractalkine, mediates several properties of inflammatory cells, including chemotaxis, adhesion, maturation and survival, and has a highly specific receptor, CX3CR1. This variety of functions as both a chemokine and adhesion molecule results from CX3CL1 existing in both membrane-tethered and

66

S. Hallam and T. Hagemann

soluble isoforms (Bazan et al. 1997). Murine monocytes have been classified as CX3CR1hi or CX3CR1lo, correlating with their migratory properties (Geissmann et al. 2003) such that CX3CR1lo/CCR2+/GR1+ monocytes are actively recruited to inflammatory tissues but short lived, and CX3CR1hi/CCR2-/GR1− monocytes are recruited in a CX3CR1-dependent manner to non-inflamed tissues and are longer lived. A murine model of ovarian cancer found that tumour progression was associated with massive accumulation in the peritoneum of CX3CR1hi TAM (Hart et al. 2009). A wealth of research has identified important roles for the CX3CL1–CX3CR1 axis in the pathogenesis of many inflammatory as well as malignant disorders. CX3CL1 antagonists as well as blocking antibodies to CX3CR1 (Furuichi et al. 2006), depletion of CX3CR1 positive cells and CX3CR1 knockout animals (Landsman et al. 2009; Jung et al. 2000; Yu et al. 2007) have developed this proof-ofconcept, but as yet there are no associated clinical agents undergoing human trials in cancer.

TAM Re-education Ablation of macrophage populations must be extremely selective to not only disable tumours, but also to enable normal healthy physiology. Phenotypic plasticity is a hallmark of macrophages, and one that, we believe, is hijacked by tumours to facilitate their growth and spread, such that TAM adopt an alternatively activated phenotype. An attractive approach, therefore, is to exploit this phenotypic plasticity and re-educate TAM towards a classically activated phenotype. In principal, this might not only result in the withdrawal of pro-tumoural phenomena, but also incorporate anti-tumoural activities through recruitment of Th1 immune responses. Previous chapters have outlined in great detail the nature of macrophage plasticity and phenotype determination. Animal models have clearly demonstrated the potential therapeutic benefits of re-educating macrophages in a number of different settings including cancer (Hagemann et al. 2008; Movahedi et al. 2010; Fong et al. 2008; Guiducci et al. 2005). Realising this potential with clinically feasible compounds is a goal that is being actively pursued. Small molecule inhibitors of enzymes at critical points in pathways of alternative activation of macrophages, such as IKKE, hold promise. Designing such drugs to be macrophage-specific is a huge challenge, yet in the case of the IKKE/NF-NB pathway, there may be advantages to inhibition in both cancer cells and macrophages. Using IKKE deletion in epithelial cells and separately myeloid cells, Greten et al. demonstrated attenuation of colitis-associated tumours with both approaches (Greten et al. 2004). Similar benefits of IKKE inhibition in both tumour cells and macrophages have been demonstrated in a chemically induced hepatocellular cancer model (Maeda et al. 2005). Inhibition of the NF-NB pathway appeared to become a clinical reality, with the development of Bortezomib, a highly effective anti-myeloma agent. Recent work has exposed the complexity of its true mechanism of action, such that it cannot be considered simply a pure NF-NB inhibitor of myeloma cells and highlighted the need for specificity regarding target

5 TAM: A Moving Clinical Target

67

cell type, and inhibition of inducible versus constitutive NF-NB activation in rational drug design (Hideshima et al. 2009). In this regard, it was found that Bortezomib did not actually induce NF-NB inhibition in patient myeloma cells, nor in peripheral blood mononuclear cells, but did inhibit NF-NB in bone marrow stromal cells. Novel small molecule inhibitors of IKKE, such as MLN120B (Nagashima et al. 2006), were found to inhibit myeloma cell NF-NB activation. It is interesting to speculate on the future clinical benefits that may be derived from combined specific inhibition of tumour cell and micro-environmental cell NF-NB, including in TAM. As we improve our understanding of existing therapies, it is becoming evident that certain anti-cancer drugs and radiotherapy act also as immuno-modulatory agents, and re-educate the tumour microenvironment, some exploiting TAM in their mechanism of action. One example of this exploitation is Rituximab, a chimaeric monoclonal antibody directed against CD20, as found on malignant B-lymphocytes. Rituximab is highly effective therapy against B-lymphomas, such as Follicular Lymphoma (FL), and is thought to exert its effects partly through direct lymphoma killing and partly through immune-mediated mechanisms, in which activatory FcJ receptor-expressing macrophages play a prominent role (Beers et al. 2010; Glennie et al. 2007). Although there is not yet a consensus on this matter, there is a growing body of evidence that a large number of infiltrating TAM predict a poor prognosis in FL when treated with conventional chemotherapy (Farinha et al. 2005). Moreover, TAM display alternatively activated features in this disease. However, when treated with Rituximab alone or in conjunction with chemotherapy, a large number of TAM predict a good response to treatment (Taskinen et al. 2007; Canioni et al. 2008). A fascinating insight has been provided by in vitro work with malignant B-lymphocytes, demonstrating that alternatively activated macrophages phagocytose Rituximab-opsonised malignant B-cells more efficiently than classically activate macrophages, leading the authors to suggest that TAM actually switch from a tumour-promoting to a tumour-inhibiting function with the addition of Rituximab (Leidi et al. 2009). Radiotherapy, at sufficient doses, certainly seems to exert some of its beneficial effects through re-education of the tumour microenvironment. Radiation induced tumour cell damage and death releases “danger signals,” provoking in situ autoimmunisation, augmenting tumour-specific T-cell and macrophage responses. The direction and consequences of macrophage activation following radiotherapy appear to depend upon the dose, the cancer being treated and the genetic background of the individual host. A better understanding of these variables is required to optimise this phenomenon (Formenti and Demaria 2009). Nanoparticles have been extensively investigated as vehicles capable of delivering anti-cancer drugs directly to tumour cells, aiming to avoid drug resistance mechanisms and collateral damage to normal tissue. However, as previously discussed in the example of liposomal clodronate, macrophages avidly take up simple liposomal nanoparticles, and prior to liposome PEGylation, was one of the major obstacles encountered in their use against cancer cells (Illum et al. 1984). This phenomenon does, of course, present the opportunity to target macrophages themselves with

68

S. Hallam and T. Hagemann

nanoparticles in such a way that they are activated toward a tumouricidal phenotype. In vitro studies with lung cancer cell lines and murine alveolar macrophages suggested that empty endotoxin-free poly(isobutylcyanoacrylate) nanoparticles are able to induce tumouricidal activity in macrophages with an associated increase in Th1 cytokines in the co-culture supernatant (Al-Hallak et al. 2010). Silica nanoparticles, under investigation as gene delivery vectors to the central nervous system, induce phenotypic changes in rat brain resident macrophages (microglia) (Choi et al. 2010). Increasing the production of reactive oxygen and nitrogen species, and downregulating TNF-D gene expression, macrophages adopt a potentially tumouricidal classically activated phenotype. Nanoparticles with additional functionality can be engineered with surface ligands to enhance binding to target cells, be they cancer cells or TAM. An alternative approach has been to explore combining the, often hypoxiadriven, tumour-homing behaviour of macrophages, with the low-toxicity and other benefits of nanoparticle drug-encapsulation, by immobilising nanoparticles on the cell surface of macrophages (Holden et al. 2010).

Adoptive Transfer Having considered ablation or re-education of pro-tumoural TAM, it is now logical to discuss the merits of adoptive transfer of monocytes or macrophages as a therapeutic strategy in cancer. Given that tumours appear to recruit monocytes from the circulation by a variety of mechanisms, it is reasonable to suppose that supplementary populations of injected monocyte/macrophages might preferentially accumulate within tumours. Without manipulation, these cells might contribution to the TAM population and so encourage tumour growth. Manipulation of the monocyte/ macrophages either in vitro or subsequently in vivo might provide the potential to deliver to the tumour, classically activated macrophages with anti-cancer properties. Alternatively, or in addition, these macrophages might be used as vectors to express genes within the tumour, such as inflammatory cytokines, or pro-drug conversion genes to optimise local delivery of active chemotherapeutics (Griffiths et al. 2000). In vivo studies in mice have demonstrated the therapeutic potential of macrophage adoptive transfer. In 1974, Isaiah Fidler and colleagues studied the ability of syngeneic macrophages to inhibit pulmonary metastases in C57BL/6 mice bearing a progressively growing B16 melanoma (Fidler 1974). Their work demonstrated that peritoneal macrophages, harvested after thioglycollate-induced inflammation, reduced the number of pulmonary metastases. This activity was dependent upon the macrophages being specifically treated in vitro and then injected intravenously. Whilst unable to describe the macrophage phenotype in language we are now familiar with, it was clear that the tumour-cytotoxic activity of the macrophages was influenced by their prior in vitro exposure to sensitising agents during maturation. Those exposed to tumour supernatants as well as rat lymphocytes sensitised to the tumour in vivo were highly active in preventing pulmonary metastases, compared with those exposed

5 TAM: A Moving Clinical Target

69

only to tumour supernatants, which had no effect. Later work described that the subsequent tissue distribution in a normal host mouse of transferred macrophages was determined by the method used to elicit and activate them from the donor peritoneum (Wiltrout et al. 1983). It is much clearer to us now that macrophage subtype and phenotype is likely to be crucial in successfully populating and influencing growing tumours. Similarly successful adoptive transfer of macrophages with tumour regression has been repeated and refined in other mouse models of cancer (Hagemann et al. 2008; Ben-Efraim et al. 1994). Human studies in the late 1980s demonstrated that macrophage adoptive transfer is feasible and safe in patients with cancer (Lacerna et al. 1988; Stevenson et al. 1987). Greater than 3×109 autologous monocyte-derived macrophages, sensitised in vitro with interferon-J or lipopolysaccharide, or in vivo with GM-CSF, can be safely infused by intravenous (i.v), intraperitoneal (i.p) or intrapleural routes. Systemically delivered macrophages first are retained in the lungs before pooling in the liver and spleen, concentrating at sites of metastases (Andreesen et al. 1998). Macrophages delivered i.p remain within this cavity for more than 7 days and accumulate at sites of major tumour growth. In one study of patients with ovarian or gastric cancer, i.p macrophages significantly reduced the volume of malignant ascites in three of seven patients, but without impacting on tumour mass or activity (Andreesen et al. 1990). Unfortunately, these results proved to be exceptional, with the majority of early human in vivo studies showing little or no measurable benefit in a variety of cancers (Lacerna et al. 1988; Faradji et al. 1991a; Faradji et al. 1991b; Lopez et al. 1991; Eymard et al. 1996; Hennemann et al. 1995; Hennemann et al. 1997). One reason for this failure of effect, despite successful transfer, might be an overwhelming ability of the tumour to manipulate the phenotype of transferred macrophages, so down-regulating their tumour-cytotoxic and pro-inflammatory functions. This might be overcome by transfecting macrophages with critical pro-inflammatory cytokines. In vitro and in vivo studies in mice support the feasibility of this approach with respect to IFN, IL4, IL6, TNF, also showing increased tumouricidal activity (Nishihara et al. 1995).

Concluding Remarks As the field of macrophage biology continues to expand, so the prospect of effective and highly specific TAM therapies comes closer to a reality. Studying the macrophagerelated mechanisms of action of existing therapies will continue to provide valuable insights, but it is likely to be the development of novel molecules, specific to certain macrophage subtypes and almost certainly adjuvant to other treatment modalities, that will deliver the greatest rewards. The specific aims in TAM therapy are likely to differ between cancer diagnoses and situations. The objectives, however, will remain constant: removing pro-tumoural influences, introducing, maximising and maintaining tumouricidal effects, whilst preserving physiological macrophage functions in normal tissues.

70

S. Hallam and T. Hagemann

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 13:193–205 Ahn GO, Tseng D, Liao CH, Dorie MJ, Czechowicz A, Brown JM (2010) Inhibition of Mac-1 (CD11b/CD18) enhances tumor response to radiation by reducing myeloid cell recruitment. Proc Natl Acad Sci USA 107:8363–8368 Al-Hallak KM, Azarmi S, Anwar-Mohamed A, Roa WH, Lobenberg R (2010) Secondary cytotoxicity mediated by alveolar macrophages: a contribution to the total efficacy of nanoparticles in lung cancer therapy? Eur J Pharm Biopharm 76:112–119 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 50:7450–7456 Andreesen R, Hennemann B, Krause SW (1998) Adoptive immunotherapy of cancer using monocyte-derived macrophages: rationale, current status, and perspectives. J Leukoc Biol 64: 419–426 Bazan JF, Bacon KB, Hardiman G, Wang W, Soo K, Rossi D, Greaves DR, Zlotnik A, Schall TJ (1997) A new class of membrane-bound chemokine with a CX3C motif. Nature 385:640–644 Beers SA, French RR, Chan HT, Lim SH, Jarrett TC, Vidal RM, Wijayaweera SS, Dixon SV, Kim H, Cox KL, Kerr JP, Johnston DA, Johnson PW, Verbeek JS, Glennie MJ, Cragg MS (2010) Antigenic modulation limits the efficacy of anti-CD20 antibodies: implications for antibody selection. Blood 115:5191–5201 Ben-Efraim S, Tak C, Romijn JC, Fieren MJ, Bonta IL (1994) Therapeutical effect of activated human macrophages on a human tumor line growing in nude mice. Med Oncol 11:7–12 Canioni D, Salles G, Mounier N, Brousse N, Keuppens M, Morchhauser F, Lamy T, Sonet A, Rousselet MC, Foussard C, Xerri L (2008) High numbers of tumor-associated macrophages have an adverse prognostic value that can be circumvented by rituximab in patients with follicular lymphoma enrolled onto the GELA-GOELAMS FL-2000 trial. J Clin Oncol 26: 440–446 Choi J, Zheng Q, Katz HE, Guilarte TR (2010) Silica-based nanoparticle uptake and cellular response by primary microglia. Environ Health Perspect 118:589–595 Eymard JC, Lopez M, Cattan A, Bouche O, Adjizian JC, Bernard J (1996) Phase I/II trial of autologous activated macrophages in advanced colorectal cancer. Eur J Cancer 32A:1905–1911 Faradji A, Bohbot A, Schmitt-Goguel M, Roeslin N, Dumont S, Wiesel ML, Lallot C, Eber M, Bartholeyns J, Poindron P et al (1991a) Phase I trial of intravenous infusion of ex-vivoactivated autologous blood-derived macrophages in patients with non-small-cell lung cancer: toxicity and immunomodulatory effects. Cancer Immunol Immunother 33:319–326 Faradji A, Bohbot A, Frost H, Schmitt-Goguel M, Siffert JC, Dufour P, Eber M, Lallot C, Wiesel ML, Bergerat JP et al (1991b) Phase I study of liposomal MTP-PE-activated autologous monocytes administered intraperitoneally to patients with peritoneal carcinomatosis. J Clin Oncol 9:1251–1260 Farinha P, Masoudi H, Skinnider BF, Shumansky K, Spinelli JJ, Gill K, Klasa R, Voss N, Connors JM, Gascoyne RD (2005) Analysis of multiple biomarkers shows that lymphoma-associated macrophage (LAM) content is an independent predictor of survival in follicular lymphoma (FL). Blood 106:2169–2174 Fidler IJ (1974) Inhibition of pulmonary metastasis by intravenous injection of specifically activated macrophages. Cancer Res 34:1074–1078 Fong CH, Bebien M, Didierlaurent A, Nebauer R, Hussell T, Broide D, Karin M, Lawrence T (2008) An antiinflammatory role for IKKbeta through the inhibition of “classical” macrophage activation. J Exp Med 205:1269–1276

5 TAM: A Moving Clinical Target

71

Formenti SC, Demaria S (2009) Systemic effects of local radiotherapy. Lancet Oncol 10: 718–726 Furuichi K, Gao JL, Murphy PM (2006) Chemokine receptor CX3CR1 regulates renal interstitial fibrosis after ischemia-reperfusion injury. Am J Pathol 169:372–387 Gazzaniga S, Bravo AI, Guglielmotti A, van Rooijen N, Maschi F, Vecchi A, Mantovani A, Mordoh J, Wainstok R (2007) Targeting tumor-associated macrophages and inhibition of MCP-1 reduce angiogenesis and tumor growth in a human melanoma xenograft. J Invest Dermatol 127: 2031–2041 Geissmann F, Jung S, Littman DR (2003) Blood monocytes consist of two principal subsets with distinct migratory properties. Immunity 19:71–82 Glennie MJ, French RR, Cragg MS, Taylor RP (2007) Mechanisms of killing by anti-CD20 monoclonal antibodies. Mol Immunol 44:3823–3837 Gnant M, Mlineritsch B, Schippinger W, Luschin-Ebengreuth G, Postlberger S, Menzel C, Jakesz R, Seifert M, Hubalek M, Bjelic-Radisic V, Samonigg H, Tausch C, Eidtmann H, Steger G, Kwasny W, Dubsky P, Fridrik M, Fitzal F, Stierer M, Rucklinger E, Greil R, Marth C (2009) Endocrine therapy plus zoledronic acid in premenopausal breast cancer. N Engl J Med 360:679–691 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 118:285–296 Griffiths L, Binley K, Iqball S, Kan O, Maxwell P, Ratcliffe P, Lewis C, Harris A, Kingsman S, Naylor S (2000) The macrophage - a novel system to deliver gene therapy to pathological hypoxia. Gene Ther 7:255–262 Guiducci C, Vicari AP, Sangaletti S, Trinchieri G, Colombo MP (2005) Redirecting in vivo elicited tumor infiltrating macrophages and dendritic cells towards tumor rejection. Cancer Res 65: 3437–3446 Hafeman S, London C, Elmslie R, Dow S (2010) Evaluation of liposomal clodronate for treatment of malignant histiocytosis in dogs. Cancer Immunol Immunother 59:441–452 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-kappaB. J Exp Med 205:1261–1268 Hart KM, Bak SP, Alonso A, Berwin B (2009) Phenotypic and functional delineation of murine CX(3)CR1 monocyte-derived cells in ovarian cancer. Neoplasia 11:564–573, 1 p following 573 Hennemann B, Scheibenbogen C, Schumichen C, Andreesen R (1995) Intrahepatic adoptive immunotherapy with autologous tumorcytotoxic macrophages in patients with cancer. J Immunother Emphasis Tumor Immunol 18:19–27 Hennemann B, Rehm A, Kottke A, Meidenbauer N, Andreesen R (1997) Adoptive immunotherapy with tumor-cytotoxic macrophages derived from recombinant human granulocyte-macrophage colony-stimulating factor (rhuGM-CSF) mobilized peripheral blood monocytes. J Immunother 20:365–371 Hideshima T, Ikeda H, Chauhan D, Okawa Y, Raje N, Podar K, Mitsiades C, Munshi NC, Richardson PG, Carrasco RD, Anderson KC (2009) Bortezomib induces canonical nuclear factor-kappaB activation in multiple myeloma cells. Blood 114:1046–1052 Hiraoka K, Zenmyo M, Watari K, Iguchi H, Fotovati A, Kimura YN, Hosoi F, Shoda T, Nagata K, Osada H, Ono M, Kuwano M (2008) Inhibition of bone and muscle metastases of lung cancer cells by a decrease in the number of monocytes/macrophages. Cancer Sci 99:1595–1602 Holden CA, Yuan Q, Yeudall WA, Lebman DA, Yang H (2010) Surface engineering of macrophages with nanoparticles to generate a cell-nanoparticle hybrid vehicle for hypoxia-targeted drug delivery. Int J Nanomedicine 5:25–36 Illum L, Jones PD, Baldwin RW, Davis SS (1984) Tissue distribution of poly(hexyl 2-cyanoacrylate) nanoparticles coated with monoclonal antibodies in mice bearing human tumor xenografts. J Pharmacol Exp Ther 230:733–736 Jones R (2000) Rovelizumab (ICOS Corp). IDrugs 3:442–446

72

S. Hallam and T. Hagemann

Jung S, Aliberti J, Graemmel P, Sunshine MJ, Kreutzberg GW, Sher A, Littman DR (2000) Analysis of fractalkine receptor CX(3)CR1 function by targeted deletion and green fluorescent protein reporter gene insertion. Mol Cell Biol 20:4106–4114 Lacerna LV Jr, Stevenson GW, Stevenson HC (1988) Adoptive cancer immunotherapy utilizing lymphokine activated killer cells and gamma interferon activated killer monocytes. Pharmacol Ther 38:453–465 Landsman L, Bar-On L, Zernecke A, Kim KW, Krauthgamer R, Shagdarsuren E, Lira SA, Weissman IL, Weber C, Jung S (2009) CX3CR1 is required for monocyte homeostasis and atherogenesis by promoting cell survival. Blood 113:963–972 Leidi M, Gotti E, Bologna L, Miranda E, Rimoldi M, Sica A, Roncalli M, Palumbo GA, Introna M, Golay J (2009) M2 macrophages phagocytose rituximab-opsonized leukemic targets more efficiently than m1 cells in vitro. J Immunol 182:4415–4422 Lopez M, Fechtenbaum J, David B, Martinache C, Chokri M, Canepa S, De Gramont A, Louvet C, Gorin I, Mortel O et al (1991) (1992) Adoptive immunotherapy with activated macrophages grown in vitro from blood monocytes in cancer patients: a pilot study. J Immunother 11: 209–217 Macdonald KP, Palmer JS, Cronau S, Seppanen E, Olver S, Raffelt NC, Kuns R, Pettit AR, Clouston A, Wainwright B, Branstetter D, Smith J, Paxton RJ, Cerretti DP, Bonham L, Hill GR, Hume DA (2010) An antibody against the colony-stimulating factor 1 receptor (CSF1R) depletes the resident subset of monocytes and tissue and tumor-associated macrophages but does not inhibit inflammation. Blood 116(19):3955–3963 Maeda S, Kamata H, Luo JL, Leffert H, Karin M (2005) IKKbeta couples hepatocyte death to cytokine-driven compensatory proliferation that promotes chemical hepatocarcinogenesis. Cell 121:977–990 Meng Y, Beckett MA, Liang H, Mauceri HJ, van Rooijen N, Cohen KS, Weichselbaum RR (2010) Blockade of tumor necrosis factor alpha signaling in tumor-associated macrophages as a radiosensitizing strategy. Cancer Res 70:1534–1543 Miselis NR, Wu ZJ, Van Rooijen N, Kane AB (2008) Targeting tumor-associated macrophages in an orthotopic murine model of diffuse malignant mesothelioma. Mol Cancer Ther 7:788–799 Movahedi K, Laoui D, Gysemans C, Baeten M, Stange G, Van den Bossche J, Mack M, Pipeleers D, In’t Veld P, De Baetselier P, Van Ginderachter JA (2010) Different tumor microenvironments contain functionally distinct subsets of macrophages derived from Ly6C(high) monocytes. Cancer Res 70:5728–5739 Nagashima K, Sasseville VG, Wen D, Bielecki A, Yang H, Simpson C, Grant E, Hepperle M, Harriman G, Jaffee B, Ocain T, Xu Y, Fraser CC (2006) Rapid TNFR1-dependent lymphocyte depletion in vivo with a selective chemical inhibitor of IKKbeta. Blood 107:4266–4273 Nishihara K, Barth RF, Wilkie N, Lang JC, Oda Y, Kikuchi H, Everson MP, Lotze MT (1995) Increased in vitro and in vivo tumoricidal activity of a macrophage cell line genetically engineered to express IFN-gamma, IL-4, IL-6, or TNF-alpha. Cancer Gene Ther 2:113–124 Stevenson HC, Keenan AM, Woodhouse C, Ottow RT, Miller P, Steller EP, Foon KA, Abrams PG, Beman J, Larson SM et al (1987) Fate of gamma-interferon-activated killer blood monocytes adoptively transferred into the abdominal cavity of patients with peritoneal carcinomatosis. Cancer Res 47:6100–6103 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 13:5784–5789 Van Rooijen N, Sanders A (1996) Kupffer cell depletion by liposome-delivered drugs: comparative activity of intracellular clodronate, propamidine, and ethylenediaminetetraacetic acid. Hepatology 23:1239–1243 Veltman JD, Lambers ME, van Nimwegen M, Hendriks RW, Hoogsteden HC, Hegmans JP, Aerts JG (2010) Zoledronic acid impairs myeloid differentiation to tumour-associated macrophages in mesothelioma. Br J Cancer 103:629–641

5 TAM: A Moving Clinical Target

73

Vujaskovic Z, Anscher MS, Feng QF, Rabbani ZN, Amin K, Samulski TS, Dewhirst MW, Haroon ZA (2001) Radiation-induced hypoxia may perpetuate late normal tissue injury. Int J Radiat Oncol Biol Phys 50:851–855 Wiltrout RH, Brunda MJ, Gorelik E, Peterson ES, Dunn JJ, Leonhardt J, Varesio L, Reynolds CW, Holden HT (1983) Distribution of peritoneal macrophage populations after intravenous injection in mice: differential effects of eliciting and activating agents. J Reticuloendothel Soc 34:253–269 Yu YR, Fong AM, Combadiere C, Gao JL, Murphy PM, Patel DD (2007) Defective antitumor responses in CX3CR1-deficient mice. Int J Cancer 121:316–322 Zeisberger SM, Odermatt B, Marty C, Zehnder-Fjallman AH, Ballmer-Hofer K, Schwendener RA (2006) Clodronate-liposome-mediated depletion of tumour-associated macrophages: a new and highly effective antiangiogenic therapy approach. Br J Cancer 95:272–281

Part II

Mechanisms of Action

Chapter 6

Arginine Metabolism and Tumour-Associated Macrophages Melissa Phillips and Peter W. Szlosarek

Introduction The metabolism of the amino acid l-arginine is critical in several key biological processes, including maintenance of normal immune cell function. Recent findings indicate that dysregulated metabolism of l-arginine by macrophages in the tumour microenvironment may promote tumour growth and development by impairing the anti-tumour immune response. Two enzymes that compete for l-arginine as a substrate – arginase and nitric oxide synthase (NOS) – are critical components of this immune suppression pathway. In contrast, argininosuccinate synthetase (ASS1)negative tumours, which depend on exogenous l-arginine for growth (‘arginine auxotrophs’), may instead succumb to an l-arginine deficient tumour microenvironment secondary to arginase released by alternatively activated macrophages or via the delivery of pharmacological doses of pegylated arginine-degrading enzymes. Since the rate-limiting enzyme for l-arginine biosynthesis, ASS1, may be elevated in macrophages, understanding its role in the context of l-arginine auxotrophic tumours and in channelling l-arginine to the enzymes arginase and NOS, remains a priority for further research. This chapter reviews the role of arginine and macrophages in tumour growth and development and examines the therapeutic potential of strategies that modulate the metabolism of this amino acid in the treatment of cancer.

-0HILLIPSs073ZLOSAREK*) Centre for Molecular Oncology and Imaging, Institute of Cancer, Barts and The London School of Medicine, Queen Mary College, John Vane Science Centre, Charterhouse Square, London EC1M 6BQ, UK e-mail: [email protected] T. Lawrence and T. Hagemann (eds.), Tumour-Associated Macrophages, DOI 10.1007/978-1-4614-0662-4_6, © Springer Science+Business Media, LLC 2012

77

78

-0HILLIPSAND073ZLOSAREK

l-Arginine Metabolism: An Overview l-Arginine (l-2-amino-5-guanidinovaleric acid) is a dibasic D amino acid first discovered over 100 years ago from lupin seedlings (Boger and Bode-Boger 2001). The L-form is one of the 20 most common natural amino acids. In humans, it is classified as a semi-essential or conditionally essential amino acid, as the body’s ability to synthesise sufficient quantities depends on the developmental stage and health status of the individual. Newborns, for example, are unable to produce l-arginine efficiently, but healthy adults can synthesise enough so that it is not nutritionally essential. It does, however, need to be supplemented in cases of physical stress and illness. Free l-arginine in the body is derived from the diet, endogenous synthesis, and protein turnover. Thus, in healthy adults, homeostasis of this amino acid is mainly achieved by the modulation of intake and modulation of its catabolism (reviewed by Morris 2007). Endogenous l-arginine synthesis occurs predominantly within the kidney, where it is formed from l-citrulline and aspartate by successive actions of the rate-limiting enzyme ASS1 and argininosuccinate lyase (ASL). The liver is also capable of synthesising considerable amounts of l-arginine, but this is compartmentalised within the urea cycle, rather than contributing to plasma l-arginine levels (Boger and Bode-Boger 2001). l-Arginine is regarded as a highly versatile amino acid, playing an important role in the synthesis of proteins and nitric oxide (NO). In addition, l-arginine is a precursor for at least 6 other biologically important compounds in mammals: polyamines, nucleotides, proline, glutamate, creatine and urea (Fig. 6.1). These processes are differentially expressed according TOCELLTYPE AGEANDDEVELOPMENTALSTAGE DIETANDSTATEOFHEALTHORDISEASE7UAND Morris 1998). The metabolism of l-arginine is complex and tightly regulated by several enzymes, including ASS1, arginase, and NOS, which are expressed as a number of different isoforms (reviewed by Husson et al. 2003; Fukumura et al. 2006; Morris 2009; Fig. 6.1). Two different isoforms of arginase have been identified in humans with 58% sequence identity at the amino acid level, similar enzyme activity, but differential tissue expression. Arginase I (argI) is constitutively present in the liver as part of the urea cycle and is induced in different myeloid cells by exposure to pro-inflammatory cytokines, as discussed below. On the other hand, arginase II (argII) is located in the mitochondria of various cell types, including renal cells, neurons and enterocytes and macrophages. Three distinct isoforms of NOS have also been identified with ~50% identity, but differing in intracellular localisation, regulation, catalytic properties and inhibitor sensitivity. NOS1 (also known as nNOS) is prevalent in neuronal tissue, NOS 2 (or iNOS) is the inducible isoform, present in cells of the immune system and various cancer cells, and NOS3 (or eNOS) is found in endothelial cells. Although there are 10–14 copies of the argininosuccinate synthetase gene in the human genome, all but one, residing on 9q34 (ASS1), are pseudogenes. Most of the evidence to date suggests that activities of iNOS, argI/II and ASS1 play major roles in determining the metabolic fates of l-arginine in health and disease (Peranzoni et al. 20077UETAL2009).

6

Arginine Metabolism and Tumour-Associated Macrophages

79

Fig. 6.1 Arginine utilisation in the tumour cell. Arginine is a substrate for multiple metabolic and inflammatory pathways in health and disease (see text). Arginine may be sourced via the cationic amino acid transporter, ASS1, or autophagy. The subcellular locations of key enzymes are shown and enumerated in the figure as follows: 1, arginase 1; 2, ornithine transcarbamylase (OTC); 3, argininosuccinate synthetase (ASS1); 4, argininosuccinate lyase; 5, nitric oxide synthase; 6, ornithine decarboxylase; 7, pyrroline-5-carboxylate reductase; 8, pyrroline-5-carboxylate dehydrogenase; 9, proline oxidase (dehydrogenase); 10, ornithine aminotransferase; 11, pyrroline-5carboxylate synthase; 12, arginine decarboxylase

l-Arginine and the Macrophage Interest in l-arginine metabolism has increased greatly in the last 20 years, triggered primarily by the discovery that in mammals l-arginine is the sole precursor for the multi-functional messenger molecule, NO. Initially NO production from l-arginine was identified in endothelial cells, followed by the key discovery that murine macrophages express iNOS upon stimulation by cytokines and other inflammatory stimuli (Hibbs et al. 1988). Macrophages are crucial to the regulation of an effective immune response, playing an important role in both innate and adaptive immunity. In addition to their role as resident tissue macrophages (i.e., Kupffer cells in the liver and microglial cells of the brain), activated macrophages are also recruited to sites of inflammation and infection, where they become involved in complex interactions with various cells present at the site. As a result, macrophages have a diverse range of biological functions essential for tissue remodelling, inflammation and immunity, depending

80

-0HILLIPSAND073ZLOSAREK

on their mode of activation. These include phagocytosis, cytotoxicity, and secretion of a wide array of cytokines, growth factors, lysozymes, proteases, complement components, coagulation factors and prostaglandins (Lin and Pollard 2004). To perform their functions, macrophages must be activated either by Th1-type cytokines IL-1, TNF-D, IFN-J and IL-2 and bacterial lipopolysaccharide (often referred to as M1 or ‘classically’ activated macrophages) or by Th2-type cytokines IL-4, IL-10, IL-13 and TGF-E, as well as GM-CSF, prostaglandin and catecholamines (M2 or ‘alternatively’ activated macrophages). Importantly, both pathways of macrophage activation employ l-arginine as a key substrate: Th1-type cytokines transcriptionally upregulate the enzyme iNOS, and Th2-type cytokines activate arginase expression (reviewed by Bronte and Zanovello 2005; Classen et al. 2009). The activities of these enzymes are tightly co-ordinated with several counter-regulatory feedback loops. For example, N-hydroxy-l-arginine (NOHA), an intermediate product in the iNOS pathway, is known to strongly inhibit arginase function. Conversely, arginase can reduce iNOS activity by competitively utilising the common substrate, l-arginine. The activities of iNOS and arginase are also dependent on l-arginine availability. Macrophages utilise both exogenous and endogenous sources of l-arginine to meet their metabolic requirements. l-arginine can enter a cell through several transporters, although an Na+-independent transport system, named system y+ is postulated to be the main route for l-arginine entry in most cells (Closs et al. 2004). Transporters in the y+ system include those of the high-affinity cationic amino acid transporter (CAT) family. To date, four members in the CAT family have been identified, CAT-1 to CAT-4, encoded by Slc7A1–4GENES RESPECTIVELY7HILE#!4   AND  are known to transport cationic amino acids – such as l-arginine, l-ornithine, and l-lysine – through a pathway that is independent of pH, sensitive to stimulation, and saturable, the function of CAT-4 is unclear. The most relevant CAT member involved in macrophage function is CAT-2, which can be expressed as two splice variants (Yeramian et al. 2006). The high-affinity transporter CAT-2B is inducible in many cell types including macrophages and is strongly upregulated by IFN-J (±LPS), IL-4, and GM-CSF, while CAT-2A is constitutively expressed in several cell types and is poorly sensitive to trans-stimulation. Several data implicate the CAT system in macrophage activation. First, genetic deletion of CAT-2 in mice revealed a reduction in l-arginine uptake by 95% and led to a 92% decrease in NO production, indicating that the uptake of l-arginine via this transporter is critical for iNOS activity in macrophages (Nicholson et al. 2001). Second, there is evidence that CAT-2 on macrophages supplies arginase with l-arginine in a rodent model of bleomycin-induced fibrosis, although inflammation per se was not modulated by the deletion of CAT-2 (Niese et al. 2010). Interestingly, high levels of extracellular l-arginine are catabolised preferentially by iNOS rather than arginase, generating proapoptotic levels of NO in the murine macrophage-like 2!7  CELL LINE 'OTOH AND -ORI 1999). Third, IFN-J-mediated growth enhancement of leishmania amazonensis amastigotes in murine macrophages is associated with l-arginine transport via CAT-2B, as parasite growth was found to be ABSENTINMACROPHAGESDERIVEDFROM#!4 "DElCIENTMICE7ANASENETAL2007). Since the activities of iNOS (i.e., classical pathway for parasite killing) and arginase

6

Arginine Metabolism and Tumour-Associated Macrophages

81

(i.e., alternative pathway for parasite growth) were unchanged in these IFN-J-treated, infected macrophages, this implicates a novel direct pathway of arginine supply via CAT-2B for parasitic proliferation. Alternatively, l-arginine can be obtained by endogenous synthesis, as first described with the conversion of l-citrulline into l-arginine in rat peritoneal macROPHAGES7UAND"ROSNAN1992). Furthermore, LPS-activated macrophages were shown to be threefold more active in the synthesis of l-arginine from l-citrulline than unstimulated macrophages. LPS and certain cytokines are able to induce ASS1, but not ASL, both of which are normally only present at low levels in resting macrophages. ASS1 induction occurs in parallel with upregulation of iNOS such that a large fraction of the l-citrulline, produced as a by-product of the iNOS reaction, is recycled back to l-arginine, termed the ‘citrulline-NO’ cycle (Nussler et al. 1994). In summary, l-arginine from exogenous and endogenous sources can be utilised by macrophages in a variety of metabolic pathways, with the differential catabolism via iNOS and arginase being the best characterised to date. Arginine utilisation, therefore, may directly affect the role of macrophages and the type of host immune response within the tumour microenvironment.

TAMs, Cancer Cells and the Tumour Microenvironment An inflammatory component consisting of cells and their secreted products is present in the microenvironment of most neoplastic tissues, and it is increasingly recognised that inflammation plays an indispensable role in cancer progression (reviewed by (Mantovani et al. 2008). Indeed, tumour-associated macrophages (TAMs) are key regulators of the link between inflammation and cancer, and although most are derived from peripheral blood monocytes recruited into the tumour mass from the circulation, there is also evidence of local macrophage proliferation. Thus, the fate of a developing tumour is determined by the properties of the malignant cells and by the phenotype of the tumour infiltrating myeloid cells, with the ‘arginine metabolome’ of both tumour and TAMs playing a potentially critical modulatory role.

TAMs Recent work has identified varying proportions of M1 (argI low and iNOS high) and M2 (argI high and iNOS low) TAMs depending upon the type of chemically or genetically induced epithelial tumour murine model. For example, macrophagesexhibited argI(high)iNOS(low) polarisation in early stage chemically induced lung tumours, whereas a mixed population of M1 and M2 TAMs was observed with late stage lung adenocarcinomas (Redente et al. 2010). Moreover, lung tumour regression secondary to the inactivation of the Kras or FGF10-driven transgene was associated with a switch from an argI(high)iNOS(low) TAM polarisation to an argI(low) iNOS(low) pulmonary macrophage phenotype (no polarisation). In contrast, studies

-0HILLIPSAND073ZLOSAREK

82

L-ARGININE

IFN-G IL-1 IL-2 TNF-A

CAT2B

TAM 1 ‘Classically’ activated macrophage

L-ARGININE

iNOS

NOHA ASS1 ASL L-CITRULLINE

extracellular

TAM 2 ‘Alternatively’ activated macrophage

IL4 IL10 IL13

Arginase

ORNITHINE L-Arginine Due to high ODC arginase activity PUTRESCINE

OAT

+ RNS ROS

PROLINE

NO SPERMIDINE + SPERMINE

L-Arginine

Pro-tumourigenic Tumour cytotoxicity (high NO) Cell proliferation T cell hypo-responsiveness effect (DNA damage) Pro-tumourigenic effect (low NO) Pro-tumourigenic Depletion by excess arginase effect may inhibit tumour growth

Fig. 6.2 Schematic representation of TAM 1 and TAM 2 metabolic pathways. The activities of the enzymes in both ‘classically’ activated (M1), and ‘alternatively’ activated (M2) macrophages in the tumour microenvironment are illustrated. Solid arrows indicate the main enzymatic activity, whereas dashed arrows indicate alternative metabolic activity. In particular, when l-arginine concentrations are low, iNOS activity changes from the prevalent production of NO to the generation of superoxide and highly reactive nitrogen species (RNS) (Ohshima et al. 2003, Bronte and Zanovello 2005). Low extracellular l-arginine concentration, overexpression of ARG 1, or reduction of the capacity for l-arginine uptake can all decrease intracellular l-arginine levels. The T helper 1 cytokines (IFN Ź) and T helper 2 cytokines (IL4, IL13) are the main inducers of iNOS and Arginase 1 (ARG 1), respectively. Pro-inflammatory signals (such as IL1 and TNF ŷ) and antiinflammatory signals (IL10) can contribute to regulate the final balance between iNOS and ARG 1 activity. Moreover, ARG 1 and iNOS directly activate several biochemical circuits that negatively regulate each other. The depletion of l-arginine by overexpression of ARG 1 reduces the activity of iNOS in the production of NO. Polyamines can also inhibit production of NO. Conversely, NOHA can inhibit ARG 1. l-arginine depletion causes T cell hyporesponsiveness, affecting the anti-tumour immune response. However, if enough ARG 1 is produced by increasing numbers of alternatively activated macrophages, the further l-arginine depletion that results can inhibit tumour growth in l-arginine auxotrophic tumours (Ellyard et al. 2010). For further details, see main text. NOHA, N hydroxy l-arginine; OAT, ornithine aminotransferase; ODC, ornithine decarboxylase; RNS, reactive nitrogen species; ROS, reactive oxygen species; ASS1 argininosuccinate synthetase; ASL, argininosuccinate lyase

of TAMs associated with different stages of melanoma progression, revealed dominant iNOS expression in in situ and thin melanomas which declined with the development of thicker melanomas (Massi et al. 2007). The shift in the iNOS/arginase balance may reflect a change in biology from tumour promotion secondary to iNOS activity (producing low prosurvival levels of NO) to proliferation secondary to comparatively greater argI activity (Fig. 6.2).

6

Arginine Metabolism and Tumour-Associated Macrophages

83

Arginase Arginase generates urea and ornithine, the latter a precursor for various polyamines produced via ornithine decarboxylase (ODC), and proline via ornithine aminotransferase (OAT), thereby playing an important role in tumour cell proliferation and collagen production, respectively (Gerner and Meyskens 2004; Morris 2007). Co-culture experiments using murine macrophages transfected with rat Arg I and the human breast cancer cell line ZR-75-1, have revealed enhanced growth of the latter via the upregulation of ornithine and the polyamine putrescine (Chang et al. 2001). This was accompanied by a reduction in tumour cytotoxic NO levels, illustrating the opposite actions of iNOS and argI in this model. TAMs expressing argI and argII may also be directly involved in tumour angiogenesis, by providing a source of polyamines for increased tumour vascularisation. A marked reduction in vessel diameter was observed with arginase compared with NOS inhibition in the context of peritoneal TAMs in the LMM3 xenograft model (Davel et al. 2002). Particularly relevant to immunogenic tumours, such as melanoma and renal cell carcinoma, is the finding that M2-polarised macrophages impair T-cell anti-tumour activity by depleting l-arginine (references by Rodriguez and Ochoa 2008; and Fig. 6.2). Macrophages stimulated with Th2 cytokines (IL-4 and IL-13) produce ArgI, rapidly reducing l-arginine concentration in the microenvironment and inducing T-cell dysfunction. In the absence of l-arginine, T cells failed to upregulate cyclin D3 and cyclindependent kinase 4 and remained in the G0–G1 phase of the cell cycle. There was a concomitant downregulation of the principal signal transduction element of the T-cell receptor (TCR) or the ] chain (CD3]), which is rate-limiting in TCR assembly and memBRANE EXPRESSION 7HILST !RG) INHIBITION OR ARGININE SUPPLEMENTATION RESTORED 4 CELL function, no effect on T-cell function was observed with ArgII or iNOS, confirming the importance of arginine and ArgI in this host-dependent immunosuppression pathway. Similarly, myeloid-derived suppressor cells (MDSCs), a heterogenous population of inflammatory cells that includes precursors of the macrophage, upregulate argI and CAT2B resulting in efficient depletion of extracellular l-arginine and T-cell suppression using a murine lung carcinoma cell line. This mechanism appears to be relevant for tumour escape in vivo, because injection of the arginase inhibitor NOHA reduced lung tumour growth in a dose-dependent manner. Although ArgI is absent in human lung epithelial carcinoma cells, a large infiltrate of strongly ArgI-expressing MDSCs cells has been identified, consistent with the impaired T-cell immunity described in patients with non-small cell lung cancer. In contrast, a recent study revealed that ArgI depletion of l-arginine by alternatively activated (M2 polarised) macrophages directly inhibited B16-F1 melanoma cell proliferation in vitro, a process that was enhanced with exogenous IL-4 or IL-13 treatment (Ellyard et al. 2010). Furthermore, the ratio of alternatively activated macrophages (effector, E) to tumour (T) cells impacted tumour cell survival in vitro: low E:T ratios (<4:1) increased tumour growth, whereas higher E:T ratios elicited anti-tumour immunity. It is thought that at lower E:T ratios, l-arginine depletion would be insufficient to have a negative effect on tumour growth, and instead, the growth promoting properties of the macrophages become dominant. However, above a certain E:T ratio,

84

-0HILLIPSAND073ZLOSAREK

macrophages release enough argI to deplete l-arginine from the microenvironment and inhibit tumour growth in l-arginine auxotrophic tumours, such as B16-F1 melanoma cells. This new evidence could open the door for potential immunotherapies aimed at recruiting large numbers of these macrophages into tumours that are dependent on exogenous l-arginine for survival (v. Targeting l-Arginine below).

iNOS Under conditions of low l-arginine availability – common to many patients with cancer – macrophage-derived iNOS appears to promote tumour development, consistent with the view that TAMs are largely detrimental to the host. Instead of generating L-citrulline and NO, under low levels of l-arginine, iNOS generates instead reactive oxygen species (ROS) and highly reactive nitrogen species (RNS), leading to enhanced carcinogenesis (Ohshima et al. 2003). These free radicals may combine to produce more reactive species, such as peroxynitrite and nitrogen oxides that, left unchecked, induce DNA damage with the activation of oncogenes and inhibition of tumour suppressor genes. Chemical modification of RNA, protein, and lipid molecules may also occur, resulting in functional alterations that promote tumourigenesis. Highly reactive peroxynitrite species also impair T-cell function, as another mechanism of macrophage-induced immunosuppression (Bronte and Zanovello 2005). Overall, elevated levels of various nitrated proteins have been described in inflamed tissues (e.g., gastric mucosa of patients with H. Pylori-induced gastritis) and the plasma of lung cancer patients and cigarette smokers. Macrophage iNOS may also generate low concentrations of NO, promoting tumour development via a number of mechanisms (reviewed by Fukumura et al. 2006). The protumourigenic effects of NO include the activation of other enzymes involved in inflammation, such as cyclo-oxygenase-2 (COX 2) and metalloproteinases, the upregulation of proinflammatory cytokines, telomerase (enhanced replicative potential), and DNA methyltransferase (inhibiting tumour-suppressor genes). NO may also promote tumour growth via the activation of DNA repair enzymes and inhibition of caspases, and indirectly via tumour-induced immunsuppression and angiogenesis. Indeed, it is thought that low levels of NO may stimulate tumour neo-vascularisation via the modulation of angiogenesis inducers (e.g. vascular endothelial growth factor and basic fibroblastic growth factor) and maintaining a vasodilator tone in blood vessels in and around the tumour.

Tumoural l-Arginine Metabolism ASS1 As discussed above, arginine auxotrophic tumours are highly sensitive to l-arginine depletion secondary to the loss of ASS1 and this may explain how arginase-producing alternatively activated macrophages induce the rejection of such tumours. A number

6

Arginine Metabolism and Tumour-Associated Macrophages

85

of ASS1-negative tumours have been identified in recent years, although further work is needed to clarify the relationship between the immune infiltrate, in particular the contribution of macrophages, and the l-arginine auxotrophism of the tumour. Thus, tumours commonly associated with chemoresistance and a poor clinical outcome, including melanoma, hepatocellular carcinoma (HCC), mesothelioma, renal cell carcinoma, and prostate cancer, exhibit the loss of ASS1 expression (reviewed by Delage et al. 2010). In part, ASS1 loss is due to epigenetic silencing of the ASS1 promoter in various human cancer cell lines and tumours, and it is silencing that confers arginine auxotrophy. The reason for the loss of ASS1 is unknown, however, as a rate-limiting enzyme involved in providing l-arginine for multiple metabolic pathways, it suggests an intrinsic biological advantage for these tumours. Exogenously sourced l-arginine may be used preferentially or diverted into an existing biosynthetic pathway, such as the generation of NO by iNOS. ASS1-negative tumours may depend on stromal cells including TAMs or fibroblasts for l-arginine or its precursor argininosuccinate for tumour survival, growth, and progression. Such a mechanism has been identified in patients with acute lymphoblastic leukaemia (ALL) deficient in the rate-limiting enzyme for l-asparagine synthesis known as asparagine synthetase (ASNS). ASNS negative ALL tumours resisted the cytotoxic agent, l-asparaginase due to the supply of l-asparagine by bone marrow-derived mesenchymal cells (Iwamoto et al. 2007). In contrast, several platinum sensitive tumours, including primary ovarian, stomach, and colorectal cancer, are characterised by ASS1 overexpression, which is regulated by proinflammatory cytokines. The differential expression of ASS1 between and within tumours requires further study both in the context of TAMs and alternatively activated macrophages, as well as other types of stromal cells.

iNOS and Arginase 7ITHREGARDTOENZYMES@DOWNSTREAMOF!33 INCREASEDARGINASEACTIVITYHASBEEN detected in patients with breast, colon, lung, and prostate cancer. Indeed, elevated serum arginase has been identified as a marker of disease progression in colorectal and breast cancer and has been proposed as a tool to monitor disease progression (reviewed by Morris 2009). Similarly, iNOS has been documented in a range of tumours (gynaecological, stomach, and lung tumours) with a positive correlation between expression, p53 mutation or loss, and tumour progression (Fukumura et al. 2006). 7HILSTBOTHARGINASEANDI./3PRODUCEDBYMACROPHAGESPLAYACRITICALROLEINSOME tumours, in others, such as prostate cancer, arginine-utilising enzymes sourced from the malignant cells themselves may modulate the level of immunosuppression. Bronte et al. revealed that tumour infiltrating lymphocytes (TILs) regained their ability to destroy prostate cancer cells abundant in argII and iNOS following incubation with the specific inhibitors NOHA and L-NMMA, respectively. Notably, inactivation of both pathways was necessary for the reversal of TIL immunosuppression, as shown using collagen-gel matrix-supported organ cultures of human prostate carcinomas and a transgenic animal model of prostate malignancy (Bronte et al. 2005).

-0HILLIPSAND073ZLOSAREK

86

Pharmacological Modulation of l-Arginine Metabolism in Malignancy Clearly, arginine utilisation by TAMs within the tumour microenvironment is complex and modulated by tumour phenotype. In general, TAMs are largely detrimental to the host, playing a significant role in tumour growth and progression. Targeting TAMs may, therefore, offer important opportunities for improving the efficacy of existing cancer vaccination strategies; in addition, pharmacological strategies that directly target l-arginine hold promise for the treatment of several ASS1-deficient cancers (Fig. 6.3).

CLIP

TAM 1

TAM 2

iNOS

ARG 1

Vaccines

CTL CTL CTL CTL CTL CTL CTL

Arginase inhibitors

Tumour

T cell Proliferation

Cytotoxicity Growth inhibition

ASS1-ve tumours Nitroaspirin L-arginine

PDE 5 inhibitor

ADI-PEG20 rhuArg-PEG

Fig. 6.3 Schematic representation of pharmacological manipulation of TAMs and l-arginine metabolism. In ASS1 negative tumours, the tumour relies on extracellular l-arginine (l-arginine auxotroph). Pharmacological depletion (PEG-ADI, Recombinant human arginase) of the amino acid inhibits tumour growth. l-arginine deficiency, in contrast, can also suppress anti-tumour immune responses by down-regulating key functions of T lymphocytes. Both pharmacological interference with l-arginine metabolism and targeting TAMs specifically, as demonstrated here, restores T cell function and the anti-tumour immune response, resulting in inhibition of tumour growth. PDE 5, phosphodiesterase 5, PEG-ADI, pegylated arginine deiminase, CTL, T lymphocyte, CLIP, liposome encapsulated clodronate

6

Arginine Metabolism and Tumour-Associated Macrophages

87

Targeting TAMs TAM depletion within the tumour microenvironment has been investigated using liposome-encapsulated clodronate (CLIP) in several animal cancer models. For instance, injection of mice with peritoneal malignant mesothelioma cells or tumour spheroids in a syngeneic orthotopic model, followed by repeated injections of CLIP, attenuated mesothelioma growth, invasion, and metastases via a reduction in the TAM population. Moreover, treatment of mice bearing established tumours also showed a significant reduction in tumour growth, although there was no inhibition of invasion or metastases. Several mechanisms may be involved, including the proposal that targeting TAMs may modulate the immunosuppressive nature of the M2-polarised tumour microenvironment by allowing naïve T cells to mature, thereby stimulating an adaptive anti-tumour immune reaction (Miselis et al. 2008). The CLIP strategy appears to be critical since zolendronic acid alone did not prolong survival in a similar model of murine mesothelioma (Veltman et al. 2010). Thus, despite a switch in macrophage phenotype from M2 to M1 that resulted in reduced levels of VEGF and CCL2, there was a concomitant increase in immature myeloid cells displaying an immunosuppressive phenotype (i.e., monocytic-MDSCs). Many studies have explored the role of iNOS and arginase inhibition in a variety of tumour models; however, few have addressed targeting specifically argininemetabolising enzymes expressed by TAMs (Bowlin et al. 1986; Morris 2009; Munder 2009). Blocking ODC with difluoromethylornithine (DFMO) suppresses polyamine synthesis and was shown by Bowlin et al. to attenuate B16 melanoma tumorigenesis at 6 days by increasing macrophage-mediated tumouricidal activity; however, this effect disappeared with prolonged DFMO treatment for 18 days (Bowlin et al. 1986). Clinically, encouraging activity of DFMO combined with nonsteroidal anti-inflammatory drugs has been documented in a ‘proof of principle’ study of colorectal cancer chemoprevention (Gerner and Meyskens 2009). Recent work in H. pylori-infected mice showed that DFMO treatment restored l-arginine uptake (via CAT2), iNOS expression, and NO production in gastric macrophages, significantly reducing H. pylori colonisation levels and gastritis (Chaturvedi et al. 2010). Several additional approaches have confirmed upregulation of macrophage iNOS expression and NO synthesis thereby suppressing tumourigenesis, including adenoviral-mediated IL-12 gene therapy in an orthotopic model of prostate cancer with pulmonary metastases, macrophage anti-CD40 ligation in a murine model of B16 melanoma, and chronic exercise promoting NO signalling in resident peritoneal macrophages in mice and cytolysis of P815 tumour cells (Nasu et al. 1999; Lu et al. 1999; Lum et al. 2006). However, in certain situations, iNOS and arginase pathways function co-operatively to inhibit tumour immunity, and this has led to the development of a novel drug class with the ability to block the activity of both enzymes, the nitro-aspirins (NO-aspirins). NO-donating aspirins consist of a conventional aspirin molecule bound covalently to one of several NO donors. Devoid of direct anti-tumour activity, NO-aspirin abrogated myeloid cell-dependent immunosuppression in vivo resulting in an increase in the number and function of tumour-antigen-specific T cells and an increase in the preventative and therapeutic effectiveness of cancer

88

-0HILLIPSAND073ZLOSAREK

vaccines (De Santo et al. 2005).The mechanism of action is partly dependent on NO-aspirin-mediated suppression of arginase and iNOS, and this suggests that these drugs may have a significant role as immune adjuvants by preventing the myeloid suppressive cells from inhibiting antigen-primed T cells in specific tumours. Lastly, phosphodiesterase 5 (PDE5) inhibitors (e.g., sildenafil or Viagra) have also been shown to inhibit iNOS and arginase activity in tumour infiltrating myeloid cells, reversing their suppressor activity and triggering T-cell infiltration with a significant delay in tumour growth (Serafini et al. 2006). Similarly, sildenafil triggered the in vitro proliferation of peripheral blood mononuclear cells, with differential activation of T cell subsets in patients with myeloma and head and neck cancer. Sildenafil replicated the in vitro activity of combined treatment with NOHA and L-NMMA, suggesting that PDE5 inhibition reverses clinical T-cell immunosuppression via an effect on MDSC-derived arginase and iNOS, respectively.

Targeting L-Arginine Alternatively activated TAMs may be recruited at high E:T ratios to elicit an anti-tumour innate immune response in l-arginine auxotrophic tumours as described earlier. l-arginine-degrading enzymes such as pegylated arginine deiminase derived from mycoplasma spp. (ADI-PEG20) and pegylated human recombinant arginase (rhuArg-peg) have been developed and are in various stages of preclinical and clinical development in ASS1-negative tumours. Phase I/II clinical trials of intramuscular ADI PEG 20 administered on a weekly schedule have reported 25 and 47% (complete and partial) response rates in small numbers of patients with advance melanoma and HCC, respectively (reviewed by Delage et al. 2010). Toxicity has been confined primarily to instances of hyperuricaemia, especially in patients with HCC, occasional mild neutropaenic episodes and self-limiting injection site reactions. Evidence to date reveals that l-arginine deprivation may also have a role in the treatment of several other ASS1 negative tumours, including malignant mesothelioma, pancreatic, prostate, renal cell, and certain haematological cancers, either as monotherapy or in combination with other cytotoxics.

Conclusion l-Arginine metabolism is now recognised as being of increasing importance in the study of tumour development and progression. Indeed, l-arginine is a precursor for a diverse range of molecules influencing tumour growth, proliferation, invasion, immunity, metastasis, and angiogenesis. It is becoming clear that amino acid metabolic pathways can also be chemotherapeutic targets, and several strategies are currently being investigated: depletion of TAMs from the tumour microenvironment, ‘re-education’ of TAM-derived l-arginine-metabolising enzymes, especially iNOS and arginase, and in contrast, l-arginine deprivation for l-arginine auxotrophic

6

Arginine Metabolism and Tumour-Associated Macrophages

89

(ASS1-deficient) tumours. Further exploration into the l-arginine metabolome of TAMs, together with a better understanding of the relationship between neoplastic and host-derived immune cells, may yield novel approaches for the treatment and prevention of malignant disease.

References Boger RH, Bode-Boger SM (2001) The clinical pharmacology of L-arginine. Annu Rev Pharmacol Toxicol 41:79–99 Bowlin TL, McKown BJ et al (1986) Effect of polyamine depletion in vivo by DL-alphadifluoromethylornithine on functionally distinct populations of tumoricidal effector cells in normal and tumor-bearing mice. Cancer Res 46(11):5494–5498 Bronte V, Zanovello P (2005) Regulation of immune responses by L-arginine metabolism. Nat Rev Immunol 5(8):641–654 Bronte V, Kasic T et al (2005) Boosting antitumor responses of T lymphocytes infiltrating human prostate cancers. J Exp Med 201(8):1257–1268 Chang CI, Liao JC et al (2001) Macrophage arginase promotes tumor cell growth and suppresses nitric oxide-mediated tumor cytotoxicity. Cancer Res 61(3):1100–1106 Chaturvedi R, Asim M et al (2010) Polyamines impair immunity to helicobacter pylori by inhibiting L-arginine uptake required for nitric oxide production. Gastroenterology 139(5):1686–1698 Classen A, Lloberas J et al (2009) Macrophage activation: classical versus alternative. Methods Mol Biol 531:29–43 Closs EI, Simon A et al (2004) Plasma membrane transporters for arginine. J Nutr 134(10 Suppl): 2752S–2759S, discussion 2765S–2767S Davel LE, Jasnis MA et al (2002) Arginine metabolic pathways involved in the modulation of tumor-induced angiogenesis by macrophages. FEBS Lett 532(1–2):216–220 De Santo C, Serafini P et al (2005) Nitroaspirin corrects immune dysfunction in tumor-bearing hosts and promotes tumor eradication by cancer vaccination. Proc Natl Acad Sci USA 102(11): 4185–4190 Delage B, Fennell DA et al (2010) Arginine deprivation and argininosuccinate synthetase expression in the treatment of cancer. Int J Cancer 126(12):2762–2772 Ellyard JI, Quah BJ et al (2010) Alternatively activated macrophage possess antitumor cytotoxicity that is induced by IL-4 and mediated by arginase-1. J Immunother 33(5):443–452 Fukumura D, Kashiwagi S et al (2006) The role of nitric oxide in tumour progression. Nat Rev Cancer 6(7):521–534 'ERNER%7 -EYSKENS&,*R 0OLYAMINESANDCANCEROLDMOLECULES NEWUNDERSTANDING Nat Rev Cancer 4(10):781–792 'ERNER %7 -EYSKENS &, *R  #OMBINATION CHEMOPREVENTION FOR COLON CANCER TARGETING polyamine synthesis and inflammation. Clin Cancer Res 15(3):758–761 Gotoh T, Mori M (1999) Arginase II downregulates nitric oxide (NO) production and prevents ./ MEDIATEDAPOPTOSISINMURINEMACROPHAGE DERIVED2!7CELLS*#ELL"IOL  427–434 Hibbs JB Jr, Taintor RR et al (1988) Nitric oxide: a cytotoxic activated macrophage effector molecule. Biochem Biophys Res Commun 157(1):87–94 Husson A, Brasse-Lagnel C et al (2003) Argininosuccinate synthetase from the urea cycle to the citrulline-NO cycle. Eur J Biochem 270(9):1887–1899 Iwamoto S, Mihara K et al (2007) Mesenchymal cells regulate the response of acute lymphoblastic leukemia cells to asparaginase. J Clin Invest 117(4):1049–1057 ,IN%9 0OLLARD*7 2OLEOFINlLTRATEDLEUCOCYTESINTUMOURGROWTHANDSPREAD"R*#ANCER 90(11):2053–2058 Lu Q, Ceddia MA et al (1999) Chronic exercise increases macrophage-mediated tumor cytolysis in young and old mice. Am J Physiol 276(2 Pt 2):R482–R489

90

-0HILLIPSAND073ZLOSAREK

Lum HD, Buhtoiarov IN et al (2006) Tumoristatic effects of anti-CD40 mAb-activated macrophages involve nitric oxide and tumour necrosis factor-alpha. Immunology 118(2):261–270 Mantovani A, Allavena P et al (2008) Cancer-related inflammation. Nature 454(7203):436–444 Massi D, Marconi C et al (2007) Arginine metabolism in tumor-associated macrophages in cutaneous malignant melanoma: evidence from human and experimental tumors. Hum Pathol 38(10): 1516–1525 -ISELIS.2 7U:*ETAL 4ARGETINGTUMOR ASSOCIATEDMACROPHAGESINANORTHOTOPICMURINE model of diffuse malignant mesothelioma. Mol Cancer Ther 7(4):788–799 Morris SM Jr (2007) Arginine metabolism: boundaries of our knowledge. J Nutr 137(6 Suppl 2): 1602S–1609S Morris SM Jr (2009) Recent advances in arginine metabolism: roles and regulation of the arginases. Br J Pharmacol 157(6):922–930 Munder M (2009) Arginase: an emerging key player in the mammalian immune system. Br J Pharmacol 158(3):638–651 Nasu Y, Bangma CH et al (1999) Adenovirus-mediated interleukin-12 gene therapy for prostate cancer: suppression of orthotopic tumor growth and pre-established lung metastases in an orthotopic model. Gene Ther 6(3):338–349 Nicholson B, Manner CK et al (2001) Sustained nitric oxide production in macrophages requires the arginine transporter CAT2. J Biol Chem 276(19):15881–15885 Niese KA, Chiaramonte MG et al (2010) The cationic amino acid transporter 2 is induced in inflammatory lung models and regulates lung fibrosis. Respir Res 11:87 Nussler AK, Billiar TR et al (1994) Coinduction of nitric oxide synthase and argininosuccinate synthetase in a murine macrophage cell line. Implications for regulation of nitric oxide production. J Biol Chem 269(2):1257–1261 Ohshima H, Tatemichi M et al (2003) Chemical basis of inflammation-induced carcinogenesis. Arch Biochem Biophys 417(1):3–11 Peranzoni E, Marigo I et al (2007) Role of arginine metabolism in immunity and immunopathology. Immunobiology 212(9–10):795–812 Redente EF, Dwyer-Nield LD et al (2010) Tumor progression stage and anatomical site regulate tumor-associated macrophage and bone marrow-derived monocyte polarization. Am J Pathol 176(6):2972–2985 Rodriguez PC, Ochoa AC (2008) Arginine regulation by myeloid derived suppressor cells and tolerance in cancer: mechanisms and therapeutic perspectives. Immunol Rev 222:180–191 Serafini P, Meckel K et al (2006) Phosphodiesterase-5 inhibition augments endogenous antitumor immunity by reducing myeloid-derived suppressor cell function. J Exp Med 203(12): 2691–2702 Veltman JD, Lambers ME et al (2010) Zoledronic acid impairs myeloid differentiation to tumourassociated macrophages in mesothelioma. Br J Cancer 103(5):629–641 7ANASEN. -AC,EOD#,ETAL , ARGININEANDCATIONICAMINOACIDTRANSPORTER"REGULATE growth and survival of Leishmania amazonensis amastigotes in macrophages. Infect Immun 75(6):2802–2810 7U'9 "ROSNAN*4 -ACROPHAGESCANCONVERTCITRULLINEINTOARGININE"IOCHEM*0T  45–48 7U' -ORRIS3-*R !RGININEMETABOLISMNITRICOXIDEANDBEYOND"IOCHEM*0T  1–17 7U' "AZER&7ETAL !RGININEMETABOLISMANDNUTRITIONINGROWTH HEALTHANDDISEASE Amino Acids 37(1):153–168 Yeramian A, Martin L et al (2006) Arginine transport via cationic amino acid transporter 2 plays a critical regulatory role in classical or alternative activation of macrophages. J Immunol 176(10): 5918–5924

Chapter 7

Indoleamine 2,3-Dioxygenase Amino Acid Metabolism and Tumour-Associated Macrophages: Regulation in Cancer-Associated Inflammation and Immune Escape George C. Prendergast, Richard Metz, Mee Young Chang, Courtney Smith, Alexander J. Muller, and Suzanne Ostrand-Rosenberg

Inflammatory and Immune Escape Mechanisms in the Tumour Microenvironment Cancer is initiated by a series of cumulative genetic and epigenetic changes in a normal cell, but it is also a disease of tissue microenvironment and immunity. While genetic and epigenetic alterations drive genomic plasticity and transformation, it is also clear that signals from the tumour microenvironment, modifier genes, stromal and endothelial cells, and immune cells are critical factors in determining progression versus dormancy or destruction of an initiated lesion, and also whether metastasis occurs (Witz 2008). Indeed, the loss of microenvironmental control that may be promoted by chronic inflammation may be a rate-limiting step in unleashing invasive and metastatic behaviors posing the main clinical challenge. Immune cells in the host contribute powerfully to tumour suppression (Shankaran et al. 2001). G.C. Prendergast (*) Lankenau Institute for Medical Research, Wynnewood, PA, USA Department of Pathology, Anatomy and Cell Biology, Jefferson Medical School, Philadelphia, PA, USA Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA, USA e-mail: [email protected] R. Metz New Link Genetics Corporation, Ames, IA, USA -9#HANGs#3MITH Lankenau Institute for Medical Research, Wynnewood, PA, USA A.J. Muller Lankenau Institute for Medical Research, Wynnewood, PA, USA Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA, USA S. Ostrand-Rosenberg Department of Biological Sciences, University of Maryland at Baltimore, Baltimore, MD, USA T. Lawrence and T. Hagemann (eds.), Tumour-Associated Macrophages, DOI 10.1007/978-1-4614-0662-4_7, © Springer Science+Business Media, LLC 2012

91

92

G.C. Prendergast et al.

Yet it is also clear that tumours can tilt immune balance from antagonistic to supportive roles (Ostrand-Rosenberg 2008). How tumours escape immune control is a central question that could challenge our fundamental understanding of the clinical nature of cancer and the most appropriate strategies to treat it. Indeed, an important trend in cancer research over the past decade has been a gradual shift in focus away from the malignant cell-centric fixation of the last 30 years toward a more integrated immuno-environmental perspective on the nature of cancer. Chronic inflammation is now accepted as a key component in the genesis and progression of many solid tumours, the earliest descriptions of which were made in the nineteenth century by Virchow and others who noted striking hallmarks of inflammation in early histopathological analysis of tumour tissues (Balkwill and Mantovani 2001). However, there remains a pressing need to define the specific factors that define ‘cancer-associated’ inflammation as it is distinguished from other contexts of chronic or classical inflammatory processes that are not associated with cancer (Peek et al. 2005). Explanative impact would be strengthened by shedding light on how ‘cancer-associated’ inflammation is integrated with the processes of oncogenesis and immunoediting (Prendergast et al. 2010). Emerging findings discussed below suggest the concept that pathogenic inflammation and immune escape processes are mechanistically as well as temporally connected processes during tumour development, as also discussed in detail elsewhere (Prendergast et al. 2010). There is an emerging appreciation that effective chemotherapy may rely in part on degradation of immune escape barriers erected by tumours, such that successful chemotherapy is immunotherapy. Lastly, there is an emerging realization that certain combined approaches of immunochemotherapy may be much more effective than one would have suspected given the traditional but probably mistaken view that chemotherapy and immunotherapy inherently antagonize each other. Overall, it is becoming increasing important to understand the cellular and molecular underpinnings of inflammation and immune escape as critical elements in cancer pathogenesis, progression, and treatment.

Immune Escape in Cancer Progression Immune escape is a fundamental trait of cancer (Prendergast 2008) that represents the end-stage of immunoediting, an influential model originally proposed by Schreiber, Smyth, and colleagues to explain how neoplastic cells evade immunity (Dunn et al. 2004). Immunoediting has three stages that represent a conceptual veneer atop the more widely accepted model of oncogenesis. Figure 7.1 integrates these models and recognizes the impact of genetic modifiers and microenvironment on tumour progression (Prendergast 2008). Briefly, the first stage of immunoediting is immune surveillance as enunciated originally by Ehrlich and later by Burnett and Lewis. Tumour neoantigens arising from oncogenesis stimulate immune surveillance against tumours, but this process also imposes a selection for immune escape that in the context of genomic plasticity can lead to the evolution of immune equilib-

7 Indoleamine 2,3-Dioxygenase Amino Acid Metabolism…

93

Fig. 7.1 An integrated model of immunoediting and oncogenesis. Oncogenesis produces transformed cells that are initially eradicated by immune surveillance as a result of neoantigen presentation. This process imposes a selection for cells that escape control. Genetic instability enabled by oncogenesis facilitates the evolution of escape tactics. The battle between tumour cells and the immune system ‘sculpts’ the immune system to equilibrium, which corresponds to a dynamic yet dormant disease. Further iteration leads to immune escape that may facilitate invasion and metastasis

rium, an incomplete state of tumour control (akin to a chronic low-grade infection). The clinical phenomenon of tumour dormancy (Curtis et al. 1997) may correspond to the state of immune equilibrium that can be observed in animals (Koebel et al. 2007). By iterating escape mechanisms, tumours may further ‘sculpt’ the immune response (Dunn et al. 2004) to achieve immune escape, an endpoint representing a defeat of immune tumour suppressor function. Whether immune escape is achieved by ‘sculpted’ evolution or early deformations of immunity is debated (Willimsky and Blankenstein 2007). Nevertheless, genetic evidence (Shankaran et al. 2001) supports the existence of some type of immune escape to ‘tilt’ the immune system toward tumour tolerance, either early in tumourigenesis or later in progression.

IDO in Immune Tolerance and Escape The tryptophan catabolic enzyme indoleamine 2,3-dioxygenase-1 (IDO) mediates immune tolerance in a variety of physiological and pathological settings, including cancer (Katz et al. 2008). This single chain oxidoreductase was initially discovered

94

G.C. Prendergast et al.

by its ability to catalyze tryptophan degradation to kynurenine, the first step in biosynthesis of nicotinamide adenine dinucleotide (NAD). IDO does not handle dietary tryptophan catabolism and NAD levels in mammals are maintained by salvage rather than synthesis. Thus, the role of IDO in mammalian cells was generally unclear before seminal studies of Munn, Mellor, and colleagues who correlated IDO activity with T-cell suppression (Munn et al. 1998; Munn et al. 1999). IDO can be expressed in a variety of cell types including antigen-presenting cells (APCs) in lymph nodes, where an immunoregulatory function has been suggested (Munn et al. 2004). Through local tryptophan starvation, IDO can stimulate a Gcn2-induced stress signaling pathway (Munn et al. 2005) that upregulates translation of the NF-IL6/CEBPE transcription factor isoform LIP, which alters immune-related gene expression including the key immune modulatory factors IL-6, TGF-E, and IL-10 (Metz et al. 2007). In a dramatic illustration of how IDO can mediate local immunosuppression to neoantigens, the bioactive IDO inhibitor 1-methyl-tryptophan (1MT) will trigger MHC-restricted T cell-mediated rejection of allogeneic pregnancies by revealing “foreign” paternal antigens to the maternal immune system (Munn et al. 1998).

IDO and Immune Escape in Cancer Recently, IDO deficient mice have been used to show how IDO is needed to create pathologic antigenic tolerance in many settings, including cancers where IDO promotes outgrowth through T-cell inhibitory mechanisms (Friberg et al. 2002). Importantly, IDO is expressed not only in tumour cells but also in tumour-draining lymph nodes (TDLNs). IDO acts to restrict antigen-induced T-cell activation (Munn et al. 2004; Fallarino et al. 2004) and recruit T regulatory cells (Tregs) (Munn and Mellor 2007). Effects on other immune cell types may be important as well, since IDO activity has been shown to interfere with NK cell activity (Della Chiesa et al. 2006). Further, IDO deficiency has been found to attenuate the T-cell suppressive activity of myeloid-derived suppressor cells which support metastasis by murine 4T1 breast cancer cells (C. Smith, K. Parker, D. Beury, J. DuHadaway, R. Metz, S. Ostrand-Rosenberg, G.C. Prendergast and A.J. Muller. IDO promotes pulmonary metastasis through IL6-based potentiation of myeloid-derived suppressor cells. Manuscript in preparation.). Strikingly, IDO is widely dysregulated in tumours and TDLNs in which it has been implicated in facilitating immune escape (Prendergast 2008). Figure 7.1 summarizes literature on dual roles of IDO in cancer as related to its dysregulation in tumours or TDLNs. Genetic studies in the mouse revealed that IDO is suppressed by the tumour suppressor gene Bin1, which is often attenuated in human cancers (Muller et al. 2005; Prendergast et al. 2009). Bin1 also limits the IFN-J-mediated activation of IDO in myeloid cells (Muller et al. 2005; Prendergast et al. 2009). Moreover, 1MT and other structural classes of IDO inhibitors identified in our laboratory can correct IDO-mediated immune escape, stanching tumour growth and

7 Indoleamine 2,3-Dioxygenase Amino Acid Metabolism…

95

synergizing with chemotherapy to eradicate established tumours in a variety of transgenic and allograft mouse models of cancer (in a T cell-dependent manner). Recent experiments in IDO-deficient mice confirm that IDO must be present for 1MT or other inhibitors to manifest their antitumour effects (Hou et al. 2007; Banerjee et al. 2008; Kumar et al. 2008; Muller et al. 2008). In cancer, IDO may be overexpressed in tumour cells, TDLNs, or both (Munn et al. 2004; Munn and Mellor 2007; Muller et al. 2005; Uyttenhove et al. 2003). In TDLNs, specific subsets of APCs appear to be preferentially capable of expressing IDO. In mice, these “IDO-competent” APCs include a subset of plasmacytoid dendritic cells (pDCs) distinguished by the cell surface markers B220 and CD19 (typically associated with the B-cell lineage), which have been implicated in supporting Treg development and malignant outgrowth (Munn 2006; Sharma et al. 2007). Interestingly, IDO induction in these pDCs appears to be sufficient to convert them from a T cell-activating to T cell-suppressing phenotype. Dermal application of a pro-inflammatory phorbol ester induces IDO + pDCs that potently suppress antigendependent T-cell activation (Muller et al. 2008). In addition to directly suppressing T-cell activation, IDO + pDCs have also been reported to inform the organization of a population of T regulatory cells (Tregs) by inducing naïve CD4+ T cells to differentiate to FoxP3+ Tregs, as well as by activating resting natural Tregs to become potently suppressive (Sharma et al. 2007; Fallarino et al. 2006). Interestingly, Tregs have been shown to induce IDO in DCs through CTLA-4-mediated reverse signaling (Grohmann et al. 2002; Munn et al. 2002; Fallarino et al. 2003). Therefore, it has been suggested that communication between IDO + pDCs and Tregs may inform a positive feedback loop of immunosuppression that helps drive cancer progression. As an initial step in efforts to therapeutically disable this pathway in cancer, the D isomer of 1MT (D-1MT) has progressed to Phase I human studies, with some early promising signs that its oral administration can yield biological activity at safely tolerated doses permitting further clinical evaluation (Soliman et al. 2009).

IDO is Essential for Inflammatory Skin Carcinogenesis In skin, where IDO connects chronic inflammation to the production of a potent immune suppressive class of IDO + pDCs, we hypothesized that IDO would also connect chronic inflammation to cancer progression, where immune escape may be pivotal (Shankaran et al. 2001; Prendergast 2008; Koebel et al. 2007). If IDO had a generalized immunosuppressive role in inflammation, its deficiency in mice might alter the course of normal wound healing or the epidermal thickening that occurs after application of phorbol esters. Alternately, if IDO had a more subtle role in contextualizing the inflammatory microenvironment in a manner that was apparent only in certain settings, such as cancer development, then IDO deficiency in mice would not be expected to affect general processes of inflammation. Indeed, no differences were noted in the histological response of IDO-deficient mice to abrasion wounds or topical treatment with a pro-inflammatory phorbol ester.

96

G.C. Prendergast et al.

However, within the same time frame, IDO was critical for the generation of immune suppressive IDO + pDCs appearing in inflamed skin-draining LNs after phorbol ester treatment (Muller et al. 2008). These observations suggested that IDO contributes to a specific or peculiar aspect of inflammation rather than overtly impacting its progression. Strikingly, this aspect of the inflammatory response to phorbol ester was associated with a deficiency in the formation and malignant progression of skin tumours (papillomas and carcinomas) (Muller et al. 2008). Briefly, the susceptibility of IDO-null mice to inflammatory skin carcinogenesis was compared with genetically identical wild-type mice using a classical multistage protocol of initiation by carcinogen 7,12-dimethylbenz[a]anthracene (DMBA) followed by promotion with twice weekly application of the phorbol ester 12-O-tetradecanylphorbol-13-acetate (TPA). Although we observed no difference in IDO-null mice with regard to the time of onset of skin papillomas, which arose at 8–10 weeks after initiation as expected, IDO-null mice exhibited a greatly reduced incidence of papilloma formation, with significantly fewer lesions arising over the 20 weeks course of TPA treatment (Muller et al. 2008). Continued monitoring beyond 20 weeks permitted efficient conversion to skin carcinoma in wildtype mice but not in IDO-deficient mice, with the wild-type mice exhibiting tumours that were also more advanced and obviously transformed than those of their IDO-null counterparts. Thus, IDO was critical in this classical model of carcinogenesis driven by chronic inflammation, both at the level of tumour formation and tumour progression. Since IDO-deficient mice can support the growth of primary tumours formed by engrafted cancer cell lines of various types, it is clearly not required for tumour growth per se but instead may facilitate cancer development as a mediator of cancer-associated inflammation. Thus, IDO may offer a defining element to understand the nature of this peculiar pathological condition.

IDO in Inflammation and Immune Escape in Cancer: Genetically Synonymous Pathways? As discussed above, studies in IDO-deficient mice suggest that IDO may define an immune escape mechanism and also a critical element of cancer-associated inflammation (Muller et al. 2008). Thus, IDO impacts a key question in cancer research (Peek et al. 2005) concerning the nature of the inflammatory state associated with cancer, a chronic condition that is distinct from classical inflammation and widely implicated in cancer susceptibility and progression. The molecular factors in this state have been unclear. Inflammation can be caused by infection, tissue damage, or both. After infection, inflammation activates innate immune functions and enhances immune cell infiltration into affected tissues and proximal lymph nodes. After tissue damage, “alarmins” or other “danger-associated” signals released by dying cells in the absence of infection can also provoke inflammation. In chronic infections or chronic tissue damage, inflammatory responses that stimulate effective T-cell immunity and memory appear to be blunted through the generation of active

7 Indoleamine 2,3-Dioxygenase Amino Acid Metabolism…

97

states of T-cell suppression. By analogy, chronic inflammatory responses associated with the development and progression of cancer may be similar to those associated with chronic infection, where functions such as IDO that can suppress T-cell immunity and memory are invoked. Neoantigens and cell death that arise during tumourigenesis can each provoke inflammation, but the molecular mechanisms that permit tumour cells to persist in immunocompetent individuals are unknown. What is clear is that local inflammation provides growth factors and cytokines that help sustain the growth deregulation, survival, and movement of pre-malignant cells (Coussens and Werb 2002; Balkwill et al. 2005). Additionally, inflammation can facilitate tumour progression by creating microenvironments at both the primary site of the tumour and the TDLNs that restrict immune surveillance (Dunn et al. 2004; Koebel et al. 2007; Gajewski 2007; Munn and Mellor 2006). Active suppression of adaptive immunity obviously occurs, because by the time tumours are overtly manifested both the tumour microenvironment and the TDLNs harbor potent T-cell suppressor functions (Sotomayor et al. 2001; Yu et al. 2006; Zhang et al. 2007). Thus, in considering how tumour cells acquire the ability to exploit inflammation and subvert immune surveillance, it is clear that they retain supportive features of the inflammatory microenvironment as tactics evolve to escape features hostile to the tumour. Thus, mechanisms that integrate immune escape with chronic inflammation would be predicted to promote carcinogenesis. IDO may provide a functional link between immune escape and chronic inflammatory states that drives the development and progression of many cancers. In offering a defining element of cancer-associated inflammation, the connection to immune escape provided by IDO has conceptual and practical implications. One implication is that the genetic mechanisms underpinning immune escape and cancer-associated inflammation may be overlapping rather than distinct. Models suggesting how the inflammatory microenvironment contributes to cancer tend to be based on the notion that the inflammatory conditions precede and engender the later evolution of immune escape mechanisms. However, studies in the IDO-deficient mouse argue that this concept may be invalid, since a single gene that supports immune escape is also needed at early times to support the generation of a cancerassociated inflammatory state needed for tumourigenesis. If immune escape pathways are understood merely as pathogenic elements of inflammation needed to support common cancers – a “molecular flavor” – then cancer-associated inflammation might be defined as a general chronic inflammation plus a genetic or epigenetic signature of immune escape, such as chronic IDO activation. A second implication is that immune escape may be initiated early in cancer, rather than later during tumour progression as the standard model of immunoediting proposes. This concept impacts other skepticisms raised to immunoediting, which at root question the idea that immune escape is the end result of an evolutionary process of immune “sculpting” by tumour-microenvironment interplay, rather than an aberration of the immune system that is fixed at early times in tumour formation (Willimsky and Blankenstein 2007; Willimsky and Blankenstein 2005). A third implication concerns the common use of mouse xenograft models to study cancer development and treatment. The preclinical development of most cancer drugs employs such models, which

98

G.C. Prendergast et al.

involve the engraftment of human tumour cells into immune compromised animals. In fact, these models remain an industry standard to evaluate traditional cytotoxics as well as modern molecular-targeted drugs, rooted in seminal work done in immunodeficient nude mice by Rygaard and Povlsen over 40 years ago (Rygaard and Povlsen 1969). Nude mice do not have higher rates of spontaneous cancer, which suggested to many at the time that immunity was nonessential in tumour suppression. However, while setting a tone that later helped push tumour immunology away from the mainstream of cancer research, tumour suppressive NK cells that are intact in nude mice were unknown until later (Kiessling et al. 1975), and a convincing genetic refutation of the skepticism about immunity as an essential tumour suppression function was not achieved until this century by Schreiber et al. (2001). With the more recent emergence of similar proofs of the great significance of immune-mediated inflammatory processes to oncogenesis and progression (Grivennikov and Karin 2010), such as processes supported by IDO (Muller et al. 2008), already existing concerns about the pathophysiological relevance of mouse xenograft models to clinical oncology are heightened still further. It is intriguing to consider that tactics being pursued clinically to defeat or correct immune escape in cancer might be understood as simply a reprogramming of the inflammatory state of the tumour microenvironment, altering its “molecular flavor” to antagonize rather than nurture the tumour. With regard to IDO inhibitors being explored for cancer therapy, this notion is intriguing in light of recent evidence that IDO blockade can promote the conversion of tumour-associated Tregs to Th17-like effector T cells which might stimulate an antitumour response (Baban et al. 2009; Sharma et al. 2009). As we first suggested some time ago (Prendergast and Jaffee 2007), investigations which delve further into how immune escape mechanisms and cancer-associated inflammation are connected will develop these concepts for prognostic or therapeutic applications. IDO activity clearly blunts chemotherapeutic efficacy (Banerjee et al. 2008; Kumar et al. 2008), and these findings align well with provocative evidence that chemotherapeutic efficacy may rely on resuscitating immune surveillance (Zitvogel et al. 2008; Apetoh et al. 2007). In summary, ongoing studies of the IDO pathway in cancer offer significance to several important questions in cancer research, including the nature of clinical cancer, which relate more to the failure of the immune system to control tumours rather than merely the presence of tumours; the mechanisms by which cancers escape immune control; the nature of cancer-associated inflammation; and how thwarting immune escape mechanisms may increase therapeutic efficacy.

Beyond IDO: Arginine and Cysteine Deprivation in Tumoural Immune Escape Immune suppression by amino acid deprivation is a common theme of myeloid cells and not limited to IDO, tryptophan, and macrophages. Myeloid-derived suppressor cells (MDSC), another key myeloid cell population that impedes antitumour immu-

7 Indoleamine 2,3-Dioxygenase Amino Acid Metabolism…

99

nity, also perturb protein synthesis to inhibit T-cell activation and promote tumour progression. One of the suppressive mechanisms employed by MDSC involves the degradation of arginine, an amino acid that is essential for T-cell activation and function. MDSC produce arginase which depletes arginine in the local environment, thereby inhibiting T-cell activation. This suppressive mechanism and its immune consequences are discussed in detail in Chap. 4. MDSC also inhibit T-cell activation and function by sequestering cystine and preventing T cells from obtaining cysteine, an amino acid that is also essential for T-cell activation and function. The following sections provide a brief description of MDSC with an overview of how MDSC exploit cystine/cysteine to tolerize T cells. A detailed introduction to MDSC is provided by Bronte in Chap. 4.

Most Cells Synthesize Cysteine or Import Cystine and Reduce it to Cysteine All eukaryotic cells require cysteine for protein synthesis. Most cells synthesize cysteine, or import the oxidized form of cysteine, cystine, from the extracellular oxidizing environment and convert it to cysteine in the cytosolic reducing environment (Arner & Holmgren 2000). Importation of cystine occurs via the cystine/ glutamate antiporter, also known as the xc− transporter. This plasma membrane structure consists of two polypeptide chains: the 4 F2 heavy chain (CD98) and the xCT light chain (Mansoor et al. 1992). Cells can also generate cysteine from intracellular methionine through the action of cystathionase (Gout et al. 2001; Ishii et al. 2004). In contrast to most cells, T cells lack the xCT chain of the cystine transporter and, therefore, are unable to import cystine to generate cysteine. T cells also lack cystathionase so they are unable to convert methionine to cysteine (Bannai 1984). As a result, cysteine is an “essential” amino acid for T cells because they are unable to generate it themselves.

During T-Cell Activation, T Cells Obtain Cysteine from APC Low levels of cystine are present in serum, and these levels are probably sufficient to maintain viability of resting T cells. However, once T cells are activated by antigen, their rate of protein synthesis increases dramatically and their need for cysteine increases correspondingly. During T-cell activation, APC such as dendritic cells (DC) and macrophages synapse with T cells to deliver the requisite antigen-specific and costimulatory signals to initiate T-cell activation and proliferation. This close contact between T cells and APC not only initiates T-cell activation, but also facilitates T cell uptake of cysteine released by the APC. DC and macrophages contain functional cystathionase and xc− transporters and they typically generate excess cysteine which is exported via their plasma membrane ASC neutral amino acid

100

G.C. Prendergast et al.

transporter (Sato et al. 1987). DC and macrophages also secrete thioredoxin which reduces extracellular cystine to cysteine. As a result, the local environment surrounding APC is enriched for cysteine which can be imported by T cells via their ASC transporters (Angelini et al. 2002; Castellani et al. 2008). The transfer of cysteine from APC to T cells is relatively efficient because the cells are in close proximity, minimizing the oxidation of cysteine to cystine.

MDSC Prevent APC from Providing Cysteine The first indication that MDSC might perturb cysteine availability came from experiments in which cultures of transgenic T cells, cognate peptide, APC, and MDSC were supplemented with the reducing agent E-mercaptoethanol (2-ME) or N-acetyl-cysteine (NAC), a form of cysteine that is stable in an oxidizing environment. Inclusion of 2-ME or NAC partially restored T-cell activation, indicating that MDSC suppression could be partially overcome by the provision of cysteine (Srivastava et al. 2010). Examination of MDSC by immunofluorescence and flow cytometry demonstrated that MDSC had normal levels of the two chains of the xc− cystine transporter. Measurements of cystine uptake confirmed the functionality of the cystine transporter in these cells and showed that MDSC, DC, and macrophages import cystine at approximately the same rate. However, DC and macrophages, but not MDSC, expressed ASC transporters as detected by immunofluorescence and flow cytometry, suggesting that MDSC cannot export cysteine. The inability of MDSC to export cysteine was confirmed by experiments measuring the release of cysteine (Srivastava et al. 2010). In addition to not releasing cysteine, MDSC also deplete their local environment of cystine. PCR studies demonstrated that similar to T cells, MDSC do not contain cystathionase and therefore must generate their cysteine exclusively through the uptake of extracellular cystine. Despite their lack of cystathionase, MDSC have higher intracellular pools of cysteine compared with DC and macrophages, suggesting that MDSC may sequester cystine/cysteine (Srivastava et al. 2010). These findings led to the hypothesis that MDSC may inhibit T-cell activation by limiting the availability of cystine and thereby reducing the amount of cysteine produced by APC during T-cell activation. This hypothesis was confirmed when it was observed that APC release of cysteine was inversely proportional to the quantity of MDSC and that at ratios of 2:1 MDSC to APC, cysteine release by APC was reduced by more than 75%. MDSC uptake of cystine also limits the amount of cysteine that can be generated extracellularly by thioredoxin, thereby further decreasing the amount of extracellular cysteine available for uptake by T cells. MDSC-mediated cystine/cysteine deprivation is also likely to impair the ability of activated T cells to eliminate malignant cells within solid tumours and metastases. Cysteine is an essential precursor for the biosynthesis of glutathione (GSH) (Sakakura et al. 2007), a major intracellular redox molecule that protects cells from oxidative stress. MDSC accumulate to high levels within solid tumours (Turovskaya

7 Indoleamine 2,3-Dioxygenase Amino Acid Metabolism…

101

et al. 2008), so tumour-infiltrating T cells will have low levels of GSH due to cysteine starvation, and therefore be unable to survive in the many hypoxic regions within solid tumours. Clinical observations and experimental findings indicate that cystine/cysteine deprivation favors tumour progression, although the connection with MDSC has only recently been appreciated (Srivastava et al. 2010). Women with high levels of cysteine in their serum have lower risk of developing breast cancer. Reduced risk for these women was attributed to the general protective qualities of GSH (Zhang et al. 2003); however, reduced risk could also be due to reversal of MDSC-mediated cystine/cysteine deprivation and increased activation of antitumour immunity. Similarly, the delay of tumour progression that occurs in mice fed the stabilized form of cysteine (NAC) (Gao et al. 2007) may be due, at least in part, to increased antitumour immunity. Recent unpublished studies showing that T-cell activation improved in tumour-bearing mice with high levels of MDSC following the administration of NAC support this concept (Sinha and Ostrand-Rosenberg unpublished data). MDSC-mediated cystine/cysteine deprivation and subsequent inhibition of T-cell activation will be most pronounced in local microenvironments such as primary and metastatic tumours where MDSC, T cells, and APC co-localize, and TAMS are also present. In addition to their localization to tumour sites where they both facilitate tumour progression, MDSC and TAMS share phenotypic characteristics and have a common progenitor cell origin. Despite these similarities, TAMS do not use cystine/ cysteine deprivation to inhibit antitumour immunity, demonstrating once again the diversity of mechanisms that enable malignant cells to escape the host’s immune response.

References Witz IP (2008) Tumor-microenvironment interactions: dangerous liaisons. Adv Cancer Res 100: 203–229 Shankaran V, Ikeda H, Bruce AT, White JM, Swanson PE, Old LJ et al (2001) IFNgamma and lymphocytes prevent primary tumour development and shape tumour immunogenicity. Nature 410(6832):1107–1111 Ostrand-Rosenberg S (2008) Immune surveillance: a balance between protumor and antitumor immunity. Curr Opin Genet Dev 18(1):11–18 Balkwill F, Mantovani A (2001) Inflammation and cancer: back to Virchow? Lancet 357(9255): 539–545 Peek RM, Mohla S, DuBois RN (2005) Inflammation in the genesis and perpetuation of cancer: summary and recommendations from a National Cancer Institute-sponsored meeting. Cancer Res 65:8583–8586 Prendergast GC, Metz R, Muller AJ (2010) Towards a genetic definition of cancer-associated inflammation: role of the IDO pathway. Am J Pathol 176(5):2082–2087 Prendergast GC (2008) Immune escape as a fundamental trait of cancer: focus on IDO. Oncogene 27(28):3889–3900 Dunn GP, Old LJ, Schreiber RD (2004) The immunobiology of cancer immunosurveillance and immunoediting. Immunity 21(2):137–148

102

G.C. Prendergast et al.

Curtis RE, Rowlings PA, Deeg HJ, Shriner DA, Socie G, Travis LB et al (1997) Solid cancers after bone marrow transplantation. N Engl J Med 336(13):897–904 Koebel CM, Vermi W, Swann JB, Zerafa N, Rodig SJ, Old LJ et al (2007) Adaptive immunity maintains occult cancer in an equilibrium state. Nature 450(7171):903–907 Willimsky G, Blankenstein T (2007 Dec) The adaptive immune response to sporadic cancer. Immunol Rev 220:102–112 Katz JB, Muller AJ, Metz R, Prendergast GC (2008) Indoleamine 2,3-dioxygenase in T-cell tolerance and tumoral immune escape. Immunol Rev 222:206–221 Munn DH, Zhou M, Attwood JT, Bondarev I, Conway SJ, Marshall B et al (1998) Prevention of allogeneic fetal rejection by tryptophan catabolism. Science 281:1191–1193 Munn DH, Shafizadeh E, Attwood JT, Bondarev I, Pashine A, Mellor AL (1999) Inhibition of T cell proliferation by macrophage tryptophan catabolism. J Exp Med 189:1363–1372 Munn DH, Sharma MD, Hou D, Baban B, Lee JR, Antonia SJ et al (2004) Expression of indoleamine 2,3-dioxygenase by plasmacytoid dendritic cells in tumor-draining lymph nodes. J Clin Invest 114(2):280–290 Munn DH, Sharma MD, Baban B, Harding HP, Zhang Y, Ron D et al (2005) GCN2 kinase in T cells mediates proliferative arrest and anergy induction in response to indoleamine 2,3-dioxygenase. Immunity 22(5):633–642 Metz R, DuHadaway JB, Kamasani U, Laury-Kleintop L, Muller AJ, Prendergast GC (2007) Novel tryptophan catabolic enzyme IDO2 is the preferred biochemical target of the antitumor IDO inhibitory compound D-1MT. Cancer Res 67:7082–7087 Friberg M, Jennings R, Alsarraj M, Dessureault S, Cantor A, Extermann M et al (2002) Indoleamine 2,3-dioxygenase contributes to tumor cell evasion of T cell-mediated rejection. Int J Cancer 101(2):151–155 Fallarino F, Asselin-Paturel C, Vacca C, Bianchi R, Gizzi S, Fioretti MC et al (2004) Murine plasmacytoid dendritic cells initiate the immunosuppressive pathway of tryptophan catabolism in response to CD200 receptor engagement. J Immunol 173(6):3748–3754 Munn DH, Mellor AL (2007) Indoleamine 2,3-dioxygenase and tumor-induced tolerance. J Clin Invest 117(5):1147–1154 Della Chiesa M, Carlomagno S, Frumento G, Balsamo M, Cantoni C, Conte R et al. (2006) The tryptophan catabolite L-kynurenine inhibits the surface expression of NKp46- and NKG2Dactivating receptors and regulates NK-cell function. Blood 108(13):4118–4125 Muller AJ, DuHadaway JB, Sutanto-Ward E, Donover PS, Prendergast GC (2005) Inhibition of indoleamine 2,3-dioxygenase, an immunomodulatory target of the tumor suppressor gene Bin1, potentiates cancer chemotherapy. Nat Med 11:312–319 Prendergast GC, Muller AJ, Ramalingam A, Chang MY (2009) BAR the door: cancer suppression by amphiphysin-like genes. Biochim Biophys Acta 1795(1):25–36 Hou DY, Muller AJ, Sharma MD, DuHadaway J, Banerjee T, Johnson M et al (2007) Inhibition of indoleamine 2,3-dioxygenase in dendritic cells by stereoisomers of 1-methyl-tryptophan correlates with antitumor responses. Cancer Res 67(2):792–801 Banerjee T, DuHadaway JB, Gaspari P, Sutanto-Ward E, Munn DH, Mellor AL et al (2008) Antitumor properties of chemopreventive natural product brassinin are based upon inhibition of indoleamine 2,3-dioxygenase (IDO). Oncogene 27(20):2851–2857 Kumar S, Malachowski WP, DuHadaway JB, LaLonde JM, Carroll PJ, Jaller D et al (2008) Indoleamine 2,3-dioxygenase is the anticancer target for a novel series of potent naphthoquinonebased inhibitors. J Med Chem 51(6):1706–1718 Muller AJ, Sharma MD, Chandler PR, Duhadaway JB, Everhart ME, Johnson BA 3rd et al (2008) Chronic inflammation that facilitates tumor progression creates local immune suppression by inducing indoleamine 2,3 dioxygenase. Proc Natl Acad Sci USA 105(44):17073–17078 Uyttenhove C, Pilotte L, Theate I, Stroobant V, Colau D, Parmentier N et al (2003) Evidence for a tumoral immune resistance mechanism based on tryptophan degradation by indoleamine 2,3-dioxygenase. Nat Med 9(10):1269–1274 Munn DH (2006) Indoleamine 2,3-dioxygenase, tumor-induced tolerance and counter-regulation. Curr Opin Immunol 18(2):220–225

7 Indoleamine 2,3-Dioxygenase Amino Acid Metabolism…

103

Sharma MD, Baban B, Chandler P, Hou DY, Singh N, Yagita H et al (2007) Plasmacytoid dendritic cells from mouse tumor-draining lymph nodes directly activate mature Tregs via indoleamine 2,3-dioxygenase. J Clin Invest 117(9):2570–2582 Fallarino F, Grohmann U, You S, McGrath BC, Cavener DR, Vacca C et al (2006) The combined effects of tryptophan starvation and tryptophan catabolites down-regulate T cell receptor zetachain and induce a regulatory phenotype in naive T cells. J Immunol 176(11):6752–6761 Grohmann U, Orabona C, Fallarino F, Vacca C, Calcinaro F, Falorni A et al (2002) CTLA-4-Ig regulates tryptophan catabolism in vivo. Nat Immunol 3(11):1097–1101 Munn DH, Sharma MD, Lee JR, Jhaver KG, Johnson TS, Keskin DB et al (2002) Potential regulatory function of human dendritic cells expressing indoleamine 2,3-dioxygenase. Science 297(5588): 1867–1870 Fallarino F, Grohmann U, Hwang KW, Orabona C, Vacca C, Bianchi R et al (2003) Modulation of tryptophan catabolism by regulatory T cells. Nat Immunol 4(12):1206–1212 Soliman HH, Antonia S, Sullivan D, Vahanian N, Link C (2009) Overcoming tumor antigen anergy in human malignancies using the novel indeolamine 2,3-dioxygenase (IDO) enzyme inhibitor, 1-methyl-D-tryptophan (1MT). J Clin Oncol 27:15s Coussens LM, Werb Z (2002) Inflammation and cancer. Nature 420(6917):860–867 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 Gajewski TF (2007) Failure at the effector phase: immune barriers at the level of the melanoma tumor microenvironment. Clin Cancer Res 13(18 Pt 1):5256–5261 Munn DH, Mellor AL (2006) The tumor-draining lymph node as an immune-privileged site. Immunol Rev 213:146–158 Sotomayor EM, Borrello I, Rattis FM, Cuenca AG, Abrams J, Staveley-O’Carroll K et al (2001) Cross-presentation of tumor antigens by bone marrow-derived antigen-presenting cells is the dominant mechanism in the induction of T-cell tolerance during B-cell lymphoma progression. Blood 98(4):1070–1077 Yu P, Rowley DA, Fu YX, Schreiber H (2006) The role of stroma in immune recognition and destruction of well-established solid tumors. Curr Opin Immunol 18(2):226–231 Zhang B, Bowerman NA, Salama JK, Schmidt H, Spiotto MT, Schietinger A et al (2007) Induced sensitization of tumor stroma leads to eradication of established cancer by T cells. J Exp Med 204(1):49–55 Willimsky G, Blankenstein T (2005) Sporadic immunogenic tumours avoid destruction by inducing T-cell tolerance. Nature 437(7055):141–146 Rygaard J, Povlsen CO (1969) Heterotransplantation of a human malignant tumour to “Nude” mice. Acta Pathol Microbiol Scand 77(4):758–760 Kiessling R, Klein E, Pross H, Wigzell H (1975) “Natural” killer cells in the mouse. II. Cytoxic cells with specificity for mouse Moloney leukemia cells. Characteristics of the killer cell. Eur J Immunol 5:117–121 Grivennikov SI, Karin M (2010) Inflammation and oncogenesis: a vicious connection. Curr Opin Genet Dev 20(1):65–71 Baban B, Chandler PR, Sharma MD, Pihkala J, Koni PA, Munn DH et al (2009) IDO activates regulatory T cells and blocks their conversion into Th17-like T cells. J Immunol 183(4):2475–2483 Sharma MD, Hou DY, Liu Y, Koni PA, Metz R, Chandler P et al (2009) Indoleamine 2,3-dioxygenase controls conversion of Foxp3+ Tregs to TH17-like cells in tumor-draining lymph nodes. Blood 113(24):6102–6111 Prendergast GC, Jaffee EM (2007) Cancer immunologists and cancer biologists: why we didn’t talk then but need to now. Cancer Res 67(8):3500–3504 Zitvogel L, Apetoh L, Ghiringhelli F, Andre F, Tesniere A, Kroemer G (2008) The anticancer immune response: indispensable for therapeutic success? J Clin Invest 118(6):1991–2001 Apetoh L, Ghiringhelli F, Tesniere A, Obeid M, Ortiz C, Criollo A et al (2007) Toll-like receptor 4-dependent contribution of the immune system to anticancer chemotherapy and radiotherapy. Nat Med 13(9):1050–1059 Arner ES, Holmgren A (2000) Physiological functions of thioredoxin and thioredoxin reductase. Eur J Biochem 267(20):6102–6109

104

G.C. Prendergast et al.

Mansoor MA, Svardal AM, Ueland PM (1992) Determination of the in vivo redox status of cysteine, cysteinylglycine, homocysteine, and glutathione in human plasma. Anal Biochem 200(2):218–229 Gout PW, Buckley AR, Simms CR, Bruchovsky N (2001) Sulfasalazine, a potent suppressor of lymphoma growth by inhibition of the x(c)- cystine transporter: a new action for an old drug. Leukemia 15(10):1633–1640 Ishii I, Akahoshi N, Yu XN, Kobayashi Y, Namekata K, Komaki G et al (2004) Murine cystathionine gamma-lyase: complete cDNA and genomic sequences, promoter activity, tissue distribution and developmental expression. Biochem J 381(Pt 1):113–123 Bannai S (1984) Transport of cystine and cysteine in mammalian cells. Biochim Biophys Acta 779(3):289–306 Sato H, Watanabe H, Ishii T, Bannai S (1987) Neutral amino acid transport in mouse peritoneal macrophages. J Biol Chem 262(27):13015–13019 Angelini G, Gardella S, Ardy M, Ciriolo MR, Filomeni G, Di Trapani G et al (2002) Antigenpresenting dendritic cells provide the reducing extracellular microenvironment required for T lymphocyte activation. Proc Natl Acad Sci USA 99(3):1491–1496 Castellani P, Angelini G, Delfino L, Matucci A, Rubartelli A (2008) The thiol redox state of lymphoid organs is modified by immunization: role of different immune cell populations. Eur J Immunol 38(9):2419–2425 Srivastava MK, Sinha P, Clements VK, Rodriguez P, Ostrand-Rosenberg S (2010) Myeloid-derived suppressor cells inhibit T-cell activation by depleting cystine and cysteine. Cancer Res 70(1): 68–77 Sakakura Y, Sato H, Shiiya A, Tamba M, Sagara J, Matsuda M et al (2007) Expression and function of cystine/glutamate transporter in neutrophils. J Leukoc Biol 81(4):974–982 Turovskaya O, Foell D, Sinha P, Vogl T, Newlin R, Nayak J et al (2008) RAGE, carboxylated glycans and S100A8/A9 play essential roles in colitis-associated carcinogenesis. Carcinogenesis 29(10):2035–2043 Zhang SM, Willett WC, Selhub J, Manson JE, Colditz GA, Hankinson SE (2003) A prospective study of plasma total cysteine and risk of breast cancer. Cancer Epidemiol Biomarkers Prev 12(11 Pt 1):1188–1193 Gao P, Zhang H, Dinavahi R, Li F, Xiang Y, Raman V et al (2007) HIF-dependent antitumorigenic effect of antioxidants in vivo. Cancer Cell 12(3):230–238

Chapter 8

Vascular Endothelial Growth Factor and Tumour-Associated Macrophages Christian Stockmann and Randall S. Johnson

Intensive research efforts over the past three decades have provided significant insight into the angiogenic process and its role in cancer biology. As mentioned above, the development of a vascular supply is a critical factor in the growth of malignant tumours. Tumours can absorb sufficient nutrients and oxygen up to a size of 1–2 mm at which point their further growth requires the elaboration of vascular supply. This process is thought to involve recruitment of the mature host vasculature to begin sprouting new blood vessel capillaries which grow toward and subsequently infiltrate the tumour mass (Gimbrone et al. 1973). Among the multitude of growth factors that regulate physiological and pathological angiogenesis, vascular endothelial growth factor (VEGF) is believed to be the most important and the cloning of VEGF in 1989 was a major milestone in our understanding of tumour angiogenesis (Keck et al. 1989; Leung et al. 1998). Therefore, strategies targeting tumour angiogenesis have been the focus of intense research over the past decade, with the notion that they would inhibit new blood vessel growth and thus starve tumours of necessary oxygen and nutrients. These efforts culminated in the development of the first VEGF-targeting agent bevacizumab (Avastin, Genentech), showing clinical benefit in cancer patients when combined with chemotherapy (Hurwitz et al. 2004). However, it has become increasingly apparent that the process of angiogenesis and the therapeutic benefit associated with anti-angiogenic therapy is complex and most likely involves multiple mechanisms (Ellis and Hicklin 2008).

C. Stockmann Molecular Biology Section, Division of Biology, University of California, San Diego, CA 92093, USA Institut für Physiologie, University of Duisburg-Essen, Duisburg-Essen, Germany R.S. Johnson (*) Molecular Biology Section, Division of Biology, University of California, San Diego, CA 92093, USA e-mail: [email protected] T. Lawrence and T. Hagemann (eds.), Tumour-Associated Macrophages, DOI 10.1007/978-1-4614-0662-4_8, © Springer Science+Business Media, LLC 2012

105

106

C. Stockmann and R.S. Johnson

VEGF promotes tumour angiogenesis through several mechanisms, including enhanced endothelial cell proliferation and survival, increased migration and invasion of endothelial cells, and increased permeability of existing vessels (Leung et al. 1998; Gimbrone et al. 1973). The mammalian family of VEGF genes consists of five glycoproteins: VEGFA, VEGFB, VEGFC, VEGFD, and placenta growth factor (PGF) (Dvorak 2002; Hicklin and Ellis 2005). The best characterized of the VEGF family members is VEGFA (commonly referred to as VEGF), which is expressed as various isoforms due to alternative splicing that leads to 121-, 165-, 189-, and 206-amino acid proteins. VEGF165 is the predominant isoform and is commonly overexpressed in various solid tumours. The VEGF ligands bind to and activate three structurally similar type III receptor tyrosine kinases, designated VEGFR1 (also known as FLT1), VEGFR2 (also known as KDR), and VEGFR3 (also known as FLT4). The VEGF ligands have distinctive binding specifities for each of these tyrosine kinase receptors, which contributes to their diversity of function. In response to ligand binding, the VEGFR tyrosine kinases activate a network of distinct downstream signaling pathways (Kowanetz and Ferrara 2006). VEGFR1 is expressed on the vasculature as well as on several other cell types, but the exact role of VEGFR1 on tumour endothelium remains to be elucidated. VEGFR2 is primarily expressed on the vasculature and is the key mediator of VEGF-induced angiogenesis (Waltenberger et al. 1994). A solid tumour forms an organ-like entity composed of neoplastic cells and nontransformed host stromal cells, including bone marrow-derived myeloid cells such as macrophages, neutrophils, eosinophils, mast cells, and dendritic cells. Although most of the studies focused on the tumour cell as the driving force of angiogenesis, considerable evidence has now emerged for the pivotal role of tumour-infiltrating macrophages in regulating the formation of new blood vessels in tumours (Murdoch et al. 2008). Macrophages are highly versatile, multifunctional cells that are characterized by their ability to engulf invading microbes or cell debris from injured sites, secrete a wide array of immunomodulatory cytokines, present antigens to T cells and act as accessory cells in lymphocyte activation. They display a high degree of plasticity, altering their phenotype to suit the microenvironment in which they reside. The conventional understanding is that macrophages can be subdivided into M1 (classically activated) or M2 (alternatively activated) phenotypes. M1 macrophages exhibit a pro-inflammatory phenotype and are activated by lipopolysaccharide and interferon J (IFN-J) to secrete bactericidal factors and promote T-helper-1 (TH1) responses. In contrast, M2 macrophages have an immunosuppressive phenotype and release cytokines that promote a TH2 response (Mantovani et al. 2002). Macrophages in tumours – usually termed tumour-associated macrophages (TAM) – often express many genes typical of the M2 phenotype (Biswas et al. 2006; Saccani et al. 2006) and have, therefore, been described as “M2-skewed.” However, recent evidence has suggested that the phenotype of TAM varies with the stage of tumour development, with M1-like cells often predominating at sites of chronic inflammation where tumours develop, then switching to an M2-like phenotype as the tumour begins to invade, vascularize, and progress (Biswas et al. 2008).

8

Vascular Endothelial Growth Factor and Tumour-Associated Macrophages

107

There are usually higher numbers of TAM in malignant tumours than in surrounding normal tissues (Bingle et al. 2002). These cells initially extravasate across the tumour vasculature as monocytes from the blood and differentiate into TAM (Yamashiro et al. 1994). Monocyte recruitment is driven by chemoattractants secreted by both malignant and stromal cells in tumours (Murdoch et al. 2004). The main one is the chemokine CCL2 (MCP-1) which is expressed in tumours (Bottazzi et al. 1992; Loberg et al. 2007; Gazzaniga et al. 2007). Increased levels of other monocyte-attracting chemokines such as placental growth factor (PlGF, PGF), CCL3 (MIP-1D), CCL4 (MIP-1E), and CCL5 (RANTES) have also been described in tumours (Fischer et al. 2007; Scotton et al. 2001). And also the potent pro-angiogenic factor VEGFA is upregulated in tumours and has been shown to be chemotactic for monocytes in vitro (Barleon et al. 1996), acting through VEGFR1 on these cells (Hiratsuka et al. 1998). Alternatively, VEGFR2 has been reported to mediate VEGF-induced monocyte recruitment (Dineen et al. 2008). TAM have a profound influence on the regulation of tumour angiogenesis. Several clinical studies have shown a correlation between a high number of TAM in human tumours and increased microvessel density, suggesting that these cells might promote tumour angiogenesis (Knowles et al. 2004). In a transgenic mouse mammary tumour virus model that expresses the Polyoma Middle T antigen (MMTV-PyMT), TAM were seen to increase in number in premalignant lesions just before the angiogenic switch that precedes the transition to malignancy. Depletion of macrophages in this model resulted in a 50% reduction in vascular density, causing delayed tumour progression and metastasis. Reintroduction of macrophages into these knock-out mice led to a significant increase in vascular density and enhanced tumour progression (Lin et al. 2006). Further evidence of the importance of TAM has come from studies in which monocytes were removed from the circulation using clodronate liposomes. This was shown to significantly reduce TAM numbers and angiogenesis in Lewis lung carcinoma xenografts (Kimura et al. 2007). TAM express many pro-angiogenic and angiogenesis-modulating factors in vitro, such as VEGFA (Dirkx et al. 2006). The tumour microenvironment is now known to stimulate the pro-angiogenic functions of macrophages. For example, TNFD secreted by ovarian tumour cells enhances the release of VEGF, MMP9, and other important pro-angiogenic factors by macrophages (Hagemann et al. 2006). Tumour hypoxia (low oxygen tension) also plays an important part in this phenomenon. This is extremely important since many TAM appear to accumulate in hypoxic and/or necrotic areas of such human tumours as those in the breast (Leek et al. 1999), endometrium (Ohno et al. 2004), ovary (Negus et al. 1997) bladder (Onita et al. 2002), colon (Bailey et al. 2007), and the oral cavity (Li et al. 2002). Hypoxia develops within some areas of tumours because of inadequate vascular perfusion due to the development of disorganized and immature blood vessels (Vaupel et al. 2001). It is thought that these areas attract TAM by releasing such hypoxia-induced chemoattractants such as VEGF, endothelins, and EMAPII (also known as SCYE1) (Murdoch et al. 2004). A number of recent findings have shown that in such hypoxic areas TAM have a particularly marked effect on tumour angiogenesis. Hypoxic macrophages are

108

C. Stockmann and R.S. Johnson

known to respond to hypoxia by upregulating hypoxia-inducible transcription factors (mainly hypoxia-inducible factors HIF1 and HIF2) (Burke et al. 2002; Talks et al. 2000), the activation of which leads to increased transcription of many genes that regulate cell proliferation, metabolism, and angiogenesis (Lewis and Murdoch 2005). Hypoxia induces a distinct pro-angiogenic phenotype in human macrophages in vitro including the upregulation of VEGF (White et al. 2004; Burke et al. 2002). Moreover, human TAM express VEGF preferentially in hypoxic and/or perinecrotic areas of breast carcinomas (Lewis et al. 2000) and upregulate VEGF when they migrate into the hypoxic, perinecrotic areas of human breast tumour spheroids. When implanted into window chambers on the flanks of nude mice these small, macrophage-infiltrated tumour masses stimulated significantly more angiogenesis than spheroids without macrophages (Bingle et al. 2006). The process of tumour angiogenesis was initially defined as classical sprouting angiogenesis, which is the outgrowth of post-capillary venules from preexisting vessels (Gimbrone et al. 1973). Accordingly, anti-angiogenic therapy should be the opposite: the inhibition of new blood vessel formation, with subsequent induction of tumour dormancy. However, angiogenesis in response to VEGF involves a complex series of events including classic vessel sprouting, the loss of pericyte– endothelial cell adhesion as well as increased vessel permeability (Dvorak 2002; Casanovas et al. 2005; Jain 2005) and the value of vascular density as a measure of anti-angiogenic activity has been questioned by Folkmann and colleagues (Hlatky et al. 2002). Therefore, changes in the function of the vascular bed are likely to be an important parameter of anti-angiogenic activity than the presence of vascular structures itself (Jain 2005). In fact, recent studies have shown that tumour blood flow and growth are decreased, whereas vessel count is increased (Hicklin 2007; Li et al. 2007; Noguera-Troise et al. 2006; Ridgway et al. 2006; Cohen et al. 2007), which further supports the notion that vascular function is more important than simple vessel counts. In most tumours, despite high vascular density, the vasculature is characterized by a relatively inefficient blood supply that differs from “normal” vascular networks. The abnormalities of the tumour vasculature include increased permeability and tortuosity as well as decreased pericyte coverage (Jain 2005; Morikawa et al. 2002). Due to the fact that many of the abnormalities of the tumour vasculature are secondary to VEGF, Jain and colleagues have hypothesized that VEGF-targeted therapy can temporarily “normalize” the vascular bed of tumours during the so-called window of normalization. The definition of “vascular normalization” includes the reversion of vascular abnormalities (that are, increased permeability, tortuosity, and loss of pericytes) and redistribution of the blood flow with increased delivery of cytotoxic agents and oxygen during the normalization window (Jain 2005). In fact, in a phase II study with glioblastoma patients, a VEGFR tyrosine kinase-inhibitor led to structural and functional to normalization of the tumour vasculature, as measured by MRI (Batchelor et al. 2007). Furthermore, the combination of chemotherapy and anti-angiogenic therapy exceeds the clinical outcome of VEGF-targeted therapy as a single agent (Hurwitz et al. 2004; Escudier et al. 2007; Llovet et al. 2007; Miller et al. 2007; Motzer et al. 2007; Sandler et al. 2006).

8

Vascular Endothelial Growth Factor and Tumour-Associated Macrophages

109

Several hypotheses have been proposed to explain tumour growth in the absence of VEGF. For example, other angiogenic factors may compensate for the lack of VEGF or selection for a subpopulation of tumour cell that are “hypoxia resistant” may lead to their preferential outgrowth (Yu et al. 2002). Alternatively, remodeling of tumour vasculature may result in the generation of mature, stabilized vessels that are less responsive to angiogenesis inhibitors. However, detailed mechanisms of resistance to anti-VEGF treatment in cancer patients remain to be determined. To gain insight into the role of myeloid cells in refractoriness to anti-VEGF, a recent study screened a series of murine cell lines to establish experimental tumour models that are either responsive or refractory to the treatment (Shojaei et al. 2007a). Interestingly, refractory tumours were associated with a significant increase in the frequency of tumour-infiltrating CD11b + Gr1+ myeloid cells, compared with sensitive ones. This was consistent with a greater frequency of these cells in the bone marrow and spleen of mice bearing refractory tumours. Experiments in which tumour cells were mixed with myeloid cells before host introduction elucidated the functional relevance of these cells. Myeloid cells isolated from refractory tumours, but not from sensitive tumours, were able to mediate refractoriness to anti-VEGF treatment. These studies suggested that myeloid cells could facilitate tumour angiogenesis. Indeed, mixing sensitive tumours with CD11b + Gr1+ myeloid cells from refractory tumours resulted in a more pronounced neovascular supply compared with that resulting from mixture with CD11b−Gr1− cells. Moreover, the combination of an anti-Gr1 antibody with anti-VEGF delayed the onset of refractoriness (Shojaei et al. 2007a). To further investigate the role of CD11b + Gr1+ cells in refractoriness to antiVEGF, the same tumours were treated with soluble VEGFR1 that not only neutralizes VEGFA but also PlGF and VEGFB (Davis-Smyth et al. 1996). No difference was observed between anti-VEGF and soluble VEGFR1-treated groups in the efficiency of tumour formation or in the accumulation of CD11b + Gr1+ cells, suggesting that other VEGFR1 ligands do not compensate for the lack of VEGFA. Contrary to these observations, Fischer and colleagues recently reported that targeting PlGF alone with neutralizing antibodies elicits significant antitumour effects in some antiVEGF refractory models (Fischer et al. 2007). In addition, to address whether a lack of response to anti-VEGF might reflect a general refractoriness to other antitumour agents, tumour-bearing mice were treated with several cytotoxic agents. No correlation was seen between response to anti-VEGF therapy or to cytotoxic chemotherapy. Therefore, refractoriness to anti-VEGF did not predict a lack of response to other antitumour agents (Shojaei et al. 2007a). Gene array analysis in anti-VEGF-treated tumours revealed a unique profile of gene expression in each tumour type. Analysis of differentially expressed genes in refractory versus sensitive tumours identified several cytokines, such as G-CSF and MCP-1, known to be involved in the mobilization of bone marrow-derived myeloid cells to the peripheral blood. In addition, proinflammatory factors such as MIP-2 and IL-1R were found to be highly expressed in refractory tumours (Shojaei et al. 2007a). Therefore, this gene array analysis identified factors that might enable tumours to instruct the bone marrow to mobilize of myeloid cells to the peripheral blood, and

110

C. Stockmann and R.S. Johnson

help them to home to the tumours. These investigations provided a potential cellular and molecular basis to explain refractoriness to anti-VEGF treatment. In evaluating the mechanism of angiogenesis mediated by myeloid cells, the ortholog of the secreted protein Bv8 as a critical mediator has been identified (Shojaei et al. 2007b). Bv8, also called prokineticin-2, belongs to a larger class of peptides that are defined by a five disulfide bridge motif called a colipase fold (Kaser et al. 2003). Both Bv8 and the related factor EG-VEGF bind two highly homologous G-protein-coupled receptors termed EG-VEGRF/PKR-1 and EG-VEGFR/PKR-2 (Masuda et al. 2002). Bv8 is upregulated in CD11b + Gr1+ cells after implantation of tumour cells (Shojaei et al. 2007b). Earlier studies showed that Bv8 may mobilize hematopoietic cells to the peripheral blood and also stimulate the production of granulocytic and monocytic colonies in vitro (LeCouter et al. 2004). Notably, anti-Bv8 treatment of mice implanted with a variety of human tumours resulted in a significant reduction in tumour growth, arguing that Bv8 had a critical role. Bv8 was found to be dramatically upregulated by G-CSF both in vitro and in vivo. This upregulation seemed to reflect the operation of a crosstalk between the tumour and CD11b + Gr1+ cells because neutralizing anti-G-CSF antibodies ablated the increase in Bv8 expression in CD11b + Gr1+ cells after tumour implantation (Shojaei et al. 2007b). More recently, Bv8 has been found to play a role in myeloid cell-dependent angiogenesis in tumour xenografts as well as in the RIP-Tag transgenic model of pancreatic tumourigenesis (Shojaei et al. 2008). A recent study created an in vivo myeloid cell-specific deletion of VEGF and tested its impact on vessel density and tumour progression in various murine tumour models to determine the role of myeloid cell-derived VEGF in this context (Stockmann et al. 2008). In the MMTV-PyMT model of mammary tumourigenesis increased vascular density was found as tumours progressed to malignancy, consistent with an “angiogenic switch.” However, in mutant mice with a deletion of VEGF restricted to myeloid cells, the malignancy-associated increase in vascularization, thus the “angiogenic switch,” did not occur. Along with impaired angiogenesis, a decrease in vessel length and reduced vessel tortuosity was observed in the absence of myeloid cell-derived VEGF. Although, VEGF protein levels did not vary in tumour lysates from wild-type and mutant animals, loss of myeloid-derived VEGF caused an approximately 50% reduction in VEGFR2 phosphorylation, suggesting that myeloid cell-derived VEGF plays a unique role in tumour vascularization that cannot be compensated for by VEGF from other sources within the tumour. Noteworthy, the onset of tumour growth was not affected by the lack VEGF in myeloid cells. However, surprisingly mutant mice had a significantly higher tumour burden at endpoint than their wild-type littermates and along with this a higher number of proliferating cells, indicating that tumours develop at a more rapid pace in the absence of myeloid cell-derived VEGF. These experiments have also been carried out in a more broadly used tumour model, with subcutaneously injected Lewis lung carcinoma (LLC) cells. Again, LLC tumours in myeloid VEGF null mutant mice had significantly greater tumour volumes. Noteworthy, no genotype-specific differences in tumour infiltration were detected. FACS analysis also showed that macrophages were the predominant

8

Vascular Endothelial Growth Factor and Tumour-Associated Macrophages

111

myeloid infiltrate and that levels of infiltrating macrophages, and less abundant neutrophils, were similar across genotypes. A significant reduction in vascular density and vessel tortuosity was again observed in mutant mice. An analysis of VEGF protein and mRNA showed no significant differences between genotypes. However, when phosphorylated VEGFR2 was quantified, the activation of VEGFR2 was once again reduced approximately 50% in mutant tumours relative to wild-type tumours. Furthermore, the loss of VEGF expression in myeloid cells resulted in a marked increase in the level of pericyte coverage, indicating vascular normalization and suggesting that VEGF expression from infiltrating myeloid cells is essential for intratumoural loss of vessel pericyte association. Interestingly, vessel permeability was also reduced in tumours from mutant animals, representing another indicator of vascular normalization. As previously mentioned, hypoxia is associated with solid tumour growth and with tumourigenic vasculature. It was found that the more slowly growing wild-type tumours had higher levels of hypoxia and that this occurred even in areas that were extensively vascularized. Interestingly, we observed that intense VEGF immunostaining in wild-type tumours occurs in regions with high levels of macrophage infiltration, whereas in mutant tumours VEGF staining is only found outside of infiltrated regions; this depicts how VEGF from myeloid cells could have an acute and localized effect on tumour vasculature. Consistent with the vascular changes and the concept of vascular normalization, the loss of myeloid-derived VEGF increased the efficacy of chemotherapeutic treatment. Cyclophosphamide treatment resulted in a significantly increased level of tumour cell death in myeloid VEGF null tumour-bearing mice. Similar experiments carried out with cis-platinum gave comparable results. To determine the differential contributions of tumour cell and myeloid cellderived VEGF, matching wild-type and VEGF null fibrosarcoma cell lines were injected subcutaneously into mice wild-type or null for myeloid VEGF expression. Fibrosarcomas that were wild-type for VEGF expression grew faster and significantly larger in mice lacking VEGF in myeloid cells, which is similar to what has been shown in the other models described above. However, consistent with previous findings (Grunstein et al. 1999), tumour cells lacking VEGF expression grew more slowly than those expressing VEGF when implanted into wild-type mice. Interestingly, when VEGF was absent from both the tumour and the myeloid compartment, tumour growth essentially collapsed, with greatly increased apoptosis and necrosis. Further, although there was a reduction in tumour vessel density ranging from the highest vessel count in tumours wild-type in all tissue compartments, to the lowest in those lacking VEGF in tumour cells and myeloid cells, these values did not directly correspond to changes in tumour growth. Rather, they demonstrate the unique role for myeloid-derived VEGF in facilitating changes in tumour vessel function and normalization. TAM have been shown to be crucial for malignant progression in a mouse model of breast cancer (Lin et al. 2006); however, these studies removed macrophages completely from tumours, and thus were removing all of the many aspects of

112

C. Stockmann and R.S. Johnson

macrophage involvement in tumour progression. In isolating one of those aspects experimentally, it could be shown that myeloid-derived VEGF is required for the formation of a high-density vessel network, that the loss of VEGF from this source greatly reduces the activation of VEGFR2 in spite of little or no loss in total tumour VEGF, and that the absence of this increased vascular density does not hamper, but rather increases tumour progression. When administered as single agents, anti-angiogenic drugs have produced modest objective responses in clinical trials, but overall, these single agent approaches have not yielded significant long-term survival benefits. In contrast, when given in combination with chemotherapy, antibodies targeted against VEGF result in an increase in survival in colorectal cancer patients (Hurwitz et al. 2004). Taken together this argues that a critical mediator of this sensitization to chemotherapy lies in the infiltrating innate immune cells, rather than in the malignant cells of the tumour. These findings open an important new avenue for the investigation of anti-angiogenic therapies targeted to tumour-associated cells of the immune system. Acknowledgements Acknowledgements and Competing Interests statement: We acknowledge the support of the Deutsche Forschungs Gemeinschaft to C.S. (STO 787/1-1 and STO 787/1-2) and NIH CA82515 to R.S.J.

References Bailey C et al (2007) Chemokine expression is associated with the accumulation of tumour associated macrophages (TAMs) and progression in human colorectal cancer. Clin Exp Metastasis 24: 121–130 Barleon B et al (1996) Migration of human monocytes in response to vascular endothelial growth factor (VEGF) is mediated via the VEGF receptor flt-1. Blood 87:3336–3343 Batchelor TT et al (2007) AZD2171, a pan-VEGF receptor tyrosine kinase inhibitor, normalizes tumor vasculature and alleviates edema in glioblastoma patients. Cancer Cell 11:83–95 Bingle L, Brown NJ, Lewis CE (2002) The role of tumour-associated macrophages in tumour progression: implications for new anticancer therapies. J Pathol 196:254–265 Bingle L, Lewis CE, Corke KP, Reed MW, Brown NJ (2006) Macrophages promote angiogenesis in human breast tumour spheroids in vivo. Br J Cancer 94:101–107 Biswas SK et al (2006) A distinct and unique transcriptional program expressed by tumor-associated macrophages (defective NF-NB and enhanced IRF-3/STAT1 activation). Blood 107:2112–2122 Biswas SK, Sica A, Lewis CE (2008) Plasticity of macrophage function during tumor progression: regulation by distinct molecular mechanisms. J Immunol 180:2011–2017 Bottazzi B, Walter S, Govoni D, Colotta F, Mantovani A (1992) Monocyte chemotactic cytokine gene transfer modulates macrophage infiltration, growth, and susceptibility to IL-2 therapy of a murine melanoma. J Immunol 148:1280–1285 Burke B et al (2002) Expression of HIF-1D by human macrophages: implications for the use of macrophages in hypoxia-regulated cancer gene therapy. J Pathol 196:204–212 Casanovas O, Hicklin DJ, Bergers G, Hanahan D (2005) Drug resistance by evasion of antiangiogenic targeting of VEGF signaling in late-stage pancreatic islet tumors. Cancer Cell 8: 299–309

8

Vascular Endothelial Growth Factor and Tumour-Associated Macrophages

113

Cohen MH, Gootenberg J, Keegan P, Pazdur R (2007) FDA drug approval summary: bevacizumab (Avastin) plus Carboplatin and Paclitaxel as first-line treatment of advanced/metastatic recurrent nonsquamous non-small cell lung cancer. Oncologist 12:713–718 Davis-Smyth T, Chen H, Park J, Presta LG, Ferrara N (1996) The second immunoglobulin-like domain of the VEGF tyrosine kinase receptor Flt-1 determines ligand binding and may initiate a signal transduction cascade. EMBO J 15:4919–4927 de Visser KE, Eichten A, Coussens LM (2006) Paradoxical roles of the immune system during cancer development. Nature Rev Cancer 6:24–37 Dineen SP et al (2008) Vascular endothelial growth factor receptor 2 mediates macrophage infiltration into orthotopic pancreatic tumors in mice. Cancer Res 68:4340–4346 Dirkx AE, Oude Egbrink MG, Wagstaff J, Griffioen AW (2006) Monocyte/macrophage infiltration in tumors: modulators of angiogenesis. J Leukoc Biol 80:1183–1196 Dvorak HF (2002) Vascular permeability factor/vascular endothelial growth factor: a critical cytokine in tumor angiogenesis and a potential target for diagnosis and therapy. J Clin Oncol 20:4368–4380 Ellis LM, Hicklin DJ (2008) VEGF-targeted therapy: mechanisms of anti-tumour activity. Nat Rev Cancer 8:579–591 Escudier B et al (2007) Sorafenib in advanced clear-cell renal- cell carcinoma. N Engl J Med 356:125–134 Fischer C et al (2007) Anti-PlGF inhibits growth of VEGF(R)- inhibitor-resistant tumors without affecting healthy vessels. Cell 131:463–475 Gazzaniga S et al (2007) Targeting tumor-associated macrophages and inhibition of MCP-1 reduce angiogenesis and tumor growth in a human melanoma xenograft. J Invest Dermatol 127: 2031–2041 Gimbrone MA Jr, Leapman SB, Cotran RS, Folkman J (1973) Tumor angiogenesis: iris neovascularization at a distance from experimental intraocular tumors. J Natl Cancer Inst 50:219–228 Grunstein J, Roberts WG, Mathieu-Costello O, Hanahan D, Johnson RS (1999) Tumor-derived expression of vascular endothelial growth factor is a critical factor in tumor expansion and vascular function. Cancer Res 59:1592–1608 Hagemann T et al (2006) Ovarian cancer cells polarize macrophages toward a tumor-associated phenotype. J Immunol 176:5023–5032 Hicklin DJ, Ellis LM (2005) Role of the vascular endothelial growth factor pathway in tumor growth and angiogenesis. J Clin Oncol 23:1011–1027 Hicklin DJ (2007) Promoting angiogenesis to a fault. Nat Biotech 25:300–302 Hiratsuka S, Minowa O, Kuno J, Noda T, Shibuya M (1998) Flt-1 lacking the tyrosine kinase domain is sufficient for normal development and angiogenesis in mice. Proc Natl Acad Sci USA 95:9349–9354 Hlatky L, Hahnfeldt P, Folkman J (2002) Clinical application of antiangiogenic therapy: microvessel density, what it does and doesn’t tell us. J Natl Cancer Inst 94:883–893 Hurwitz H et al (2004) Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 350:2335–2342 Jain RK (2005) Normalization of tumor vasculature: an emerging concept in antiangiogenic therapy. Science 307:58–62 Kaser A, Winklmayr M, Lepperdinger G, Kreil G (2003) The AVIT protein family. EMBO Rep 4:469–473 Keck PJ et al (1989) Vascular permeability factor, an endothelial cell mitogen related to PDGF. Science 246:1309–1312 Kimura YN et al (2007) Inflammatory stimuli from macrophages and cancer cells synergistically promote tumor growth and angiogenesis. Cancer Sci 98:2009–2018 Knowles H, Leek R, Harris AL (2004) Macrophage infiltration and angiogenesis in human malignancy. Novartis Found Symp 256:189–200 Kowanetz M, Ferrara N (2006) Vascular endothelial growth factor signaling pathways: therapeutic perspective. Clin Cancer Res 12:5018–5022

114

C. Stockmann and R.S. Johnson

LeCouter J, Zlot C, Tejada M, Peale F, Ferrara N (2004) Bv8 and endocrine gland-derived vascular endothelial growth factor stimulate hematopoiesis and hemato- poietic cell mobilization. Proc Natl Acad Sci USA 101:16813–16818 Leek RD, Landers RJ, Harris AL, Lewis CE (1999) Necrosis correlates with high vascular density and focal macrophage infiltration in invasive carcinoma of the breast. Br J Cancer 79:991–995 Leung DW, Cachianes G, Kuang WJ, Goeddel DV, Ferrara N (1998) Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 246:1306–1309 Lewis JS, Landers RJ, Underwood JC, Harris AL, Lewis CE (2000) Expression of vascular endothelial growth factor by macrophages is up-regulated in poorly vascularized areas of breast carcinomas. J Pathol 192:150–158 Lewis CE, Murdoch C (2005) Macrophage responses to hypoxia: implications for tumor progression and anti- cancer therapies. Am J Pathol 167:627–635 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 JL et al (2007) Delta-like 4 Notch ligand regulates tumor angiogenesis, improves tumor vascular function, and promotes tumor growth in vivo. Cancer Res 67:11244–11253 Lin EY et al (2006) Macrophages regulate the angiogenic switch in a mouse model of breast cancer. Cancer Res 66:11238–11246 Loberg RD et al (2007) CCL2 as an important mediator of prostate cancer growth in vivo through the regulation of macrophage infiltration. Neoplasia 9:556–562 Llovet J. et al. (2007) Sorafenib improves survival in advanced hepatocellular carcinoma (HCC): results of a phase III randomized placebo-controlled trial (SHARP trial). J Clin Oncol. Part I. ASCO Annual Meeting Proceedings LBA1, 25, 2007 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 Masuda Y, Takatsu Y, Terao Y et al (2002) Isolation and identification of EG-VEGF/prokineticins as cognate ligands for two orphan G-protein-coupled receptors. Biochem Biophys Res Commun 293:396–402 Miller K et al (2007) Paclitaxel plus bevacizumab versus paclitaxel alone for metastatic breast cancer. N Engl J Med 357:2666–2676 Morikawa S et al (2002) Abnormalities in pericytes on blood vessels and endothelial sprouts in tumors. Am J Pathol 160:985–1000 Motzer RJ et al (2007) Sunitinib versus interferon alfa in metastatic renal-cell carcinoma. N Engl J Med 356:115–124 Murdoch C, Giannoudis A, Lewis CE (2004) Mechanisms regulating the recruitment of macrophages into hypoxic areas of tumors and other ischemic tissues. Blood 104:2224–2234 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 Negus RP, Stamp GW, Hadley J, Balkwill FR (1997) Quantitative assessment of the leukocyte infiltrate in ovarian cancer and its relationship to the expression of C-C chemokines. Am J Pathol 150:1723–1734 Noguera-Troise I et al (2006) Blockade of Dll4 inhibits tumour growth by promoting non-productive angiogenesis. Nature 444:1032–1037 Ohno S et al (2004) Correlation of histological localization of tumor-associated macrophages with clinicopathological features in endometrial cancer. Anticancer Res 24:3335–3342 Onita T et al (2002) Hypoxia-induced, perinecrotic expression of endothelial Per-ARNT-Sim domain protein-1/hypoxia-inducible factor-2D correlates with tumor progression, vascularization, and focal macrophage infiltration in bladder cancer. Clin Cancer Res 8:471–480 Ridgway J et al (2006) Inhibition of Dll4 signalling inhibits tumour growth by deregulating angiogenesis. Nature 444:1083–1087 Saccani A et al (2006) p50 nuclear factor-NB overexpression in tumor-associated macrophages inhibits M1 inflammatory responses and antitumor resistance. Cancer Res 66:11432–11440

8

Vascular Endothelial Growth Factor and Tumour-Associated Macrophages

115

Sandler A et al (2006) Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer. N Engl J Med 355:2542–2550 Scotton C, Milliken D, Wilson J, Raju S, Balkwill F (2001) Analysis of CC chemokine and chemokine receptor expression in solid ovarian tumours. Br J Cancer 85:891–897 Shojaei F, Wu X, Malik AK et al (2007a) Tumor refrectoriness to anti-VEGF treatment is mediated by CD11b + Gr1+ myeloid cells. Nat Biotechnol 25:911–920 Shojaei F, Wu X, Zhong C et al (2007b) Bv8 regulates myeloid cell-dependent tumour angiogenesis. Nature 450:825–831 Shojaei F, Singh M, Thompson JD, Ferrara N (2008) Role of Bv8 in neutrophil-dependent angiogenesis in a transgenic model of cancer progression. Proc Natl Acad Sci USA 105: 2640–2645 Stockmann C et al (2008) Deletion of vascular endothelial growth factor in myeloid cells accelerates tumorigenesis. Nature 456:814–818 Talks KL et al (2000) The expression and distribution of the hypoxia-inducible factors HIF-1D andHIF-2D in normal human tissues, cancers, and tumor-associated macrophages. Am J Pathol 157:411–421 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 Waltenberger J, Claesson-Welsh L, Siegbahn A, Shibuya M, Heldin CH (1994) Different signal transduction properties of KDR and Flt1, two receptors for vascular endothelial growth factor. J Biol Chem 269:26988–26995 White JR et al (2004) Genetic amplification of the transcriptional response to hypoxia as a novel means of identifying regulators of angiogenesis. Genomics 83:1–8 Yamashiro S et al (1994) Tumor-derived monocyte chemoattractant protein-1 induces intratumoral infiltration of monocyte-derived macrophage subpopulation in transplanted rat tumors. Am J Pathol 145:856–867 Yu JL, Rak JW, Coomber BL, Hicklin DJ, Kerbel RS (2002) Effect of p53 status on tumor response to antiangio- genic therapy. Science 295:1526–1528

Part III

Molecular Regulation

Chapter 9

TLR Signaling and Tumour-Associated Macrophages Oscar R. Colegio and Ruslan Medzhitov

Introduction Macrophages are bone marrow-derived cells that initially circulate as monocytes and subsequently undergo differentiation into tissue resident macrophages. Macrophages support tissues by remodeling, secreting growth and angiogenic factors, and phagocytosing apoptotic cells. Virchow first described the presence of leukocytes in human tumours in 1863 and suggested that cancer originates in sites of chronic inflammation (Balkwill and Mantovani 2001). Wounding of normal tissues results in the production of numerous growth factors, cytokines, and chemokines, all of which recruit and differentiate circulating monocytes to tissues. Tumour cells have been found to produce similar growth factors and cytokines that recruit macrophages to tumours (Condeelis and Pollard 2006). Further, there is an emerging characterization of macrophage activation by tumour-derived ligands of Tolllike receptors (TLRs), a family of transmembrane proteins initially characterized as sensors of infection. Whereas the role of these tumour-associated macrophages (TAMs) has yet to be clearly characterized, a majority (>80%) of clinical studies correlate an increased number of TAMs with a worse prognosis (Pollard 2004). Several in vivo studies have supported these clinical observations: mice deficient in CSF-1 and thus deficient in macrophages (Wiktor-Jedrzejczak et al. 1990), when crossed with mice prone to develop metastatic breast cancer, develop the same numbers of breast tumours but nearly no metastases (Lin et al. 2001). The role that TLR signaling may have in stimulating macrophages to become tumour promoting is not clearly defined. However, given the role that macrophages play in restoring tissue homeostasis such as in the tissue repair response, it is likely that tumour-induced TLR signaling will

/2#OLEGIOs2-EDZHITOV*) Department of Immunobiology, Howard Hughes Medical Institute, Yale University, New Haven, CT, USA e-mail: [email protected] T. Lawrence and T. Hagemann (eds.), Tumour-Associated Macrophages, DOI 10.1007/978-1-4614-0662-4_9, © Springer Science+Business Media, LLC 2012

119

120

O.R. Colegio and R. Medzhitov

provide macrophages with signals of persistent injury in need of repair. In this chapter, we review the molecular determinants of TLR signaling, the role TLR signaling plays in infection, wound healing, and tumourigenesis and discuss how TLR signaling in TAMs may result in tumour progression.

Toll-Like Receptors The TLR family of transmembrane proteins is best characterized as pattern recognition receptors (PRR) for sensing conserved pathogen-associated molecular patterns (PAMPs) (Medzhitov et al. 1997). PAMPs are conserved molecular patterns that are universal to all microorganisms within a defined class. Further, PAMPs are present within microorganisms regardless of their pathogenicity. In addition to microbial products, TLRs can also recognize endogenous, host-derived ligands (Jiang et al. 2007; Sims et al. 2010). Ten TLRs have been identified in humans and 12 TLRs in mice (Takeuchi and Akira 2010). TLRs can be divided into two general groups based on the types of ligands they recognize and their subcellular localization. TLR1, TLR2, TLR4, TLR5, TLR6, and TLR11 are expressed on the plasma membrane and can recognize lipids, lipoproteins, and proteins from exogenous and endogenous sources. In contrast, TLR3, TLR7, TLR8, and TLR9 localize to intracellular compartments including the endoplasmic reticulum, endosomes, lysosomes, and endolysosomes, and they recognize nucleic acids (Kawai and Akira 2010). TLRs are type I single spanning transmembrane proteins. Their molecular structure is defined by an extracellular or endolysosomal N-terminal leucine-rich repeat (LRR) domain followed by a transmembrane domain and then by a C-terminal cytoplasmic Toll/IL-1 receptor (TIR) domain. Below is a concise description of the known TLRs and their ligands. TLR2 recognizes PAMPs from a variety of microorganisms including bacteria, fungi, Mycoplasma, and viruses by forming heterodimers with TLR1 and TLR6. The TLR1/TLR2 heterodimer recognizes triacyl lipoproteins, whereas the TLR6/ TLR2 heterodimer recognizes diacyl lipoproteins. TLR4 recognizes lipopolysaccharide (LPS), a component of the outer membranes of Gram-negative bacteria, by forming a homodimeric complex with LPS and myeloid differentiation factor-2 (MD2). In addition to bacterial products, TLR4 can recognize components of viral envelopes. TLR4 also recognizes endogenous ligands such as oxidized phospholipids produced during acute lung injury caused by avian influenza (Imai et al. 2008). TLR5 recognizes flagellin, a component of the bacterial flagellum. In the lamina propria of the small intestine, flagellin stimulates lamina propria dendritic cells (DCs) to induce T cells to differentiate into Th17 cells and Th1 cells and B cells to differentiate into immunoglobulin A-producing plasma cells (Uematsu et al. 2008). TLR11, which is found in mice but not in humans, is closely related to TLR5. It is thought to recognize uropathogenic bacterial components (Zhang et al. 2004). In addition, it can recognize a profilin-like molecule from Toxoplasma gondii.

9

TLR Signaling and Tumour-Associated Macrophages

121

Nucleic acids are recognized by TLR3, TLR7, TLR8, and TLR9. TLR3 recognizes double-stranded RNA (dsRNA) from viruses (Alexopoulou et al. 2001) that are presented within endolysosomes. An analog of dsRNA, polyinosinic:polycytidylic acid (poly I:C), is also recognized by TLR3 and is commonly used to simulate viral infections. Activation of TLR3, as all TLRs that recognize nucleic acids, leads to the production of type I interferons as well as other pro-inflammatory cytokines. TLR7 and TLR8 recognize single-stranded RNA (ssRNA) as well as the imidazoquinoline analogs of ssRNA within endolysosomes. TLR7 can also detect RNA from bacteria, also within the context of the endolysosome (Mancuso et al. 2009). Unmethylated DNA with CpG oligodeoxynucleotides (cytosine–guanosine repeats), commonly found in bacteria and viruses, is recognized by TLR9. In addition to DNA, TLR9 may recognize the crystalline metabolite of malaria, hemozoin (Coban et al. 2005). However, there is some evidence that hemozoin binds to the malaria DNA which can be recognized by TLR9 in the endolysosome (Parroche et al. 2007). Although the ligand for human TLR10 has not yet been identified, it shares sequence homology with TLR1 and TLR6. In mice, the gene for TLR10 is disrupted by a retrovirus.

TLR Signaling Pathways Upon recognition of their ligands, TLRs signal for the transcription of genes that allow the stimulated cell to appropriately respond to the stimulus. Specificity of this response is, in part, determined by the combination of adaptor molecules that associate with the C-terminal cytoplasmic domains of TLRs. These adaptors have in common a TIR domain. The five known TIR domain-containing adaptors of TLRs are MyD88 (myeloid differentiation factor 88), TRIF (TIR domain-containing adaptor inducing interferon beta; also known as TICAM-1), TIRAP (also known as MAL), TRAM (TRIF-related adaptor molecule), and SARM (Sterile-alpha and Armadillo motif-containing protein) (O’Neill and Bowie 2007). TLR signaling can be divided into two signaling pathways depending on whether they use either MyD88 or TRIF as adaptor molecules. The downstream targets vary in that MyD88 induces inflammatory cytokines, whereas TRIF induces type I interferons and inflammatory cytokines. TRAM and TIRAP are bridging adaptors that help to recruit and bind TRIF to TLR4 and MyD88 to TLR2, respectively (Takeuchi and Akira 2010). SARM functions as a negative regulator of NF-kB and IRF activation by blocking TRIF-dependent transcription factor activation (Carty et al. 2006).

MyD88 Signaling MyD88 was the first characterized TLR adaptor and can signal downstream of all TLRs except TLR3. MyD88 binds to IRAK4 (IL-1 receptor-associated kinase),

122

O.R. Colegio and R. Medzhitov

which can activate IRAK1 and IRAK2. The IRAKs then associate with the E3 ubiquitin ligase TRAF6 (TNFR-associated factor 6), which with E2-ubiquitin conjugating enzyme complex forms a polyubiquitin chain on TRAF6 (Xia et al. 2009). This polyubiquitin chain can activate kinases such as TAK1 (TGF-Eactivating kinase-1) that phosphorylate and thus inactivate the inhibitor of NF-kB, IkBa. Free of its inhibitor, NF-kB can induce the expression of a cascade of proinflammatory genes upon translocation to the nucleus. MAPKs (mitogen-activated protein kinases) can also be activated through MyD88 signaling, leading to the induction of pro-inflammatory cytokines.

TRIF Signaling TLR3 and TLR4 can both signal through the adaptor TRIF. Whereas TLR3 can bind directly to TRIF for signaling, TLR4 requires the adaptor TRAM to signal through TRIF. The TRIF-dependent pathway results in the activation of NF-kB and IRF3. Upon activation, TRIF binds to TRAF6, TRAF3 as well as RIP1 (C-terminal receptorinteracting protein-1) and RIP3. The TRAFs then associate with TRADD (TNFRassociated death domain protein) which, in combination with FADD (FAS-associated death domain-containing protein), can result in the ubiquitination of RIP1 and then the activation of NF-kB (Ermolaeva et al. 2008; Pobezinskaya et al. 2008).

TLRs in Host Defense An evolutionarily conserved role for TLRs is in the host defense against pathogens. The first barriers encountered by most pathogens are surface epithelial cells that line the skin, lungs, gastrointestinal tract, and genitourinary tract. Signaling of TLRs in these tissues can lead to the production and secretion of antimicrobial factors such as D- and E-defensins, cathelicidin, phospholipase A2, and lysozyme. Moreover, release of these effectors can further enhance the immune response to signal through TLRs. Further, TLR signaling on immune cells such as macrophages and neutrophils leads to the release of the microbicidal compounds, reactive oxygen species, and reactive nitrogen intermediates. Beyond inducing an acute innate immune response, signaling through TLRs is critical to establishing an adaptive immune response. Antigen uptake with concomitant TLR activation on DCs allows for the discrimination between self and non-self. One way in which appropriate antigens can be targeted by the adaptive immune system is by their proximity to the TLR ligand (Blander and Medzhitov 2004, 2006a, b). This has been demonstrated through cooccurrence of antigens and TLR ligands in DCs, allowing for efficient presentation of antigen on MHC Class II or through the ligation of an antigen to a TLR ligand enabling an efficient production of antibodies (Palm and Medzhitov 2009). Further, the appropriate immune response

9

TLR Signaling and Tumour-Associated Macrophages

123

is determined by the general class of pathogen as determined by the detected TLR ligands. For instance, TLR4 recognizes LPS from Gram-negative bacteria, while TLR2 recognizes peptidoglycan from Gram-positive bacteria. The adaptive immune response can be controlled by TLRs at multiple levels including the control of antigen uptake (West et al. 2004) and presentation (Blander and Medzhitov 2006b) by DCs, the maturation of DCs, cytokine production by antigen-presenting cells, and control of survival of activated T cells. Signaling through TLRs on B cells is crucial for activation, proliferation, immunoglobulin isotype class switching, and maturation during infection as well as immunization (Gerondakis et al. 2007). Further, TLRs can induce antibody production by memory B cells (Bernasconi et al. 2002).

Endogenous Ligands of TLRs Exogenous, non-self molecules were the first identified and characterized TLR ligands. However, more recently, an increasing number of endogenous, self ligands have been identified and characterized (Karin et al. 2006; Kawai and Akira 2010). These endogenous ligands have in common their association with either infection or tissue destruction and cell death. Therefore, the ligands are more accurately described as altered self rather than self. Most of these endogenous ligands are released during tissue damage, such as that observed in tumour progression. The emerging list of endogenous ligands is complicated by the potential contamination of reagents with exogenous, pathogen-associated products. Therefore, biochemical analysis demonstrating direct interaction between endogenous ligands and TLRs will be required to address the skepticism that presently exists. Identification and characterization of endogenous ligands may be especially relevant to TAMs, as many tumours exist in a sterile environment. Therefore, for many tumour types, the TLR ligands that activate TAMs will likely be derived from an endogenous source.

Intracellular-Derived Ligands Intracellular-derived endogenous ligands include nuclear proteins and heat shock proteins. HMGB1 (high-mobility group box 1) is a nuclear protein that can bind chromatin that is released during cell necrosis or inflammation (Scaffidi et al. 2002; Wang et al. 1999). HMGB1 has been reported to be recognized by a variety of TLRs including TLR2, TLR4, and TLR9 (Sims et al. 2010). In models of ischemia– reperfusion and septic shock, HMGB1 is proinflammatory. Neutralizing antibodies against HMGB1 in the TLR4−/− background are both protective in the ischemia– reperfusion model, suggesting that endogenous, non-pathogen-derived proinflammatory signaling is mediated through TLR4 (Tsung et al. 2005b). Further, other HMGB proteins – namely HMGB1, HMGB2, and HMGB3 have been found to function as universal sentinels for nucleic acids that can activate a variety of PRRs including TLRs, NLRs (NOD-like receptors), and RLRs (RIG-I-like receptors)

124

O.R. Colegio and R. Medzhitov

(Yanai et al. 2009). As a DNA-binding protein, HMGB1 can bind both host- and pathogen-derived DNA. When the HMGB1–nucleic acid complex binds the RAGE (receptor for advanced glycation end products) receptor, it is endocytosed and presented to TLR9, where it leads to the activation of DCs and B cells (Tian et al. 2007). How the release of HMGB1 is regulated, whether through a non-classical secretory lysosomal pathway or during necrosis, has not yet been determined. However, the release of HMGB1 was found to be dependent on the cytosolic NLR, NLRP3, and its adaptor ASC (apoptotic speck-like protein containing a caspase recruitment domain) but independent of caspase-1 in a lung infection model (Willingham et al. 2009). Whether there is a physiologic correlate to controlled HMGB1 release has yet to be determined. However, HMGB1 is highly expressed in a variety of solid tumours including colon cancer, breast cancer, melanoma, prostate cancer, and pancreatic cancer. Further, this elevated expression of HMGB1 is associated with tumour progression and metastasis (Ellerman et al. 2007). One consequence of TLR signaling in TAMs by endogenous ligands was suggested in a report of the induction of apoptosis in activated T cells by TAMs that were stimulated by unidentified tumour-derived TLR4 ligands (Liu et al. 2010). Heat-shock proteins including Hsp60, Hsp70, Hsp90, and gp96 have been reported to activate macrophages and DCs through TLR2 and TLR4. However, heat-shock proteins have also been shown to bind exogenous ligands for TLRs such as LPS and lipoproteins. Therefore, it is unclear whether the effects of these molecular chaperones is direct or through contamination (Tsan and Gao 2009).

TLR Ligands Induced by TLR Signaling TLR signaling by PAMPs leads to the production of antimicrobial products. In a possible feed-forward mechanism of signal amplification, two of these products have been reported to signal through TLRs. E-Defensin-2 is a pore-forming antimicrobial protein secreted by mucosal epithelium in response to TLR signaling. However, it was also suggested to directly activate immature DCs to express costimulatory molecules and undergo maturation through TLR4, leading to an adaptive immune response (Biragyn et al. 2002). Cathelicidin (LL37) is an antimicrobial peptide that is secreted by keratinocytes and neutrophils upon TLR stimulation. In addition to its antimicrobial properties, LL37 can bind with host DNA or RNA released from necrotic cells and, upon phagocytosis of this aggregate, signal through TLR9 and TLR7, respectively (Ganguly et al. 2009; Lande et al. 2007). A cellular byproduct of infection by the H5N1 avian influenza virus and from aspiration of acidic solutions that leads to TLR signaling is oxidized phospholipid. In a murine model of lung infection, both TLR4−/− and TRIF−/− mice were protected from the acute lung injury resulting from the avian flu infection, suggesting that this pathway is critical in the detection of this form of oxidative damage (Imai et al. 2008). Of note, all forms of acute lung injury examined including SARS (severe acute respiratory syndrome), pulmonary anthrax, monkey pox, and Yersinia pestis infection resulted in significantly elevated levels of oxidized phospholipids.

9

TLR Signaling and Tumour-Associated Macrophages

125

Extracellular Matrix Ligands Altered extracellular matrix glycoproteins are another class of endogenous ligands of TLRs. Biglycan is a proteoglycan that binds collagen and is expressed in the extracellular matrix. Biglycan was found to induce pro-inflammatory cytokines in a TLR2- and TLR4-dependent manner (Schaefer et al. 2005). Further, biglycandeficient mice were more resilient to shock induced by LPS or zymosan, and this was associated with lower levels of circulating TNFD. These findings suggest that release of biglycan from the extracellular matrix through stresses such as inflammation may result in a pro-inflammatory feed-forward mechanism. Similar to biglycan, hyaluronan, a glycosaminoglycan, is a major constituent of the extracellular matrix. Hyaluronan is also a cell wall component of the bacteria Streptococcus groups A and C as well as Pasturella multocida (Jiang et al. 2007). Upon tissue damage or inflammation, small fragments of hyaluronan can be recognized by TLR2 and TLR4 to produce pro-inflammatory cytokines. Signaling through these TLRs may be important in tissue repair as TLR2- and TLR4-deficient mice challenged with bleomycin-induced sterile lung injury had greater lung pathology and a higher mortality rate than wild-type mice. The converse corollary experiment in which lung-specific overexpression of hyaluronan is protective of sterile injury supports the hypothesis that TLR signaling via hyaluronan is protective of the lung epithelium (Jiang et al. 2005). These findings suggest that the physiologic correlate of these systems may be that with small deviations from homeostasis, signaling by endogenous ligands through TLRs may be used to restore tissue integrity, whereas in a more severe infection with increased tissue damage, this mechanism may lead to immunopathology. Versican, a proteoglycan of the extracellular matrix, was identified as the active component of media conditioned by Lewis lung carcinoma that led to the induction of pro-inflammatory cytokines by macrophages (Kim et al. 2009). The induction of these cytokines was determined to be through TLR2/TLR6 signaling. Further, mice deficient in TLR2 or its adaptor, MyD88, had a higher rate of survival and fewer metastases than the wild-type mice. In this study, direct binding of versican to TLR2 but not TLR4 was demonstrated. Whether the versican produced by Lewis lung carcinoma is altered in a way to make it detectable by TLRs is not known.

TLRs and Tissue Repair TLR signaling has been found to have significant roles in tissue regeneration and repair (Kluwe et al. 2009; Rakoff-Nahoum and Medzhitov 2009). The process of tissue repair requires the clearance of damaged tissues, repopulation of lost cell types, reestablishment of the tissue vasculature and innervation, and remodeling of the overall tissue architecture. TLR ligands can be categorized into exogenous and endogenous categories, based on from where they are derived. As such, upon damage of tissue, signaling

126

O.R. Colegio and R. Medzhitov

through TLRs can be mediated by exogenous and endogenous ligands. Damage of epithelial surfaces, including the skin and intestine, compromise the integrity of these barriers, allowing for exposure to TLR ligands from microorganisms whether commensals or pathogens. In contrast, damage to internal organs that do not have surfaces colonized by mircroorganisms results in the release of stress signals, which can relay the nature and degree of tissue damage through TLRs. Various models have demonstrated that TLR signaling can either prevent or promote tissue injury. In models of sterile lung injury using bleomycin, signaling through TLR2 and TLR4 was found to be protective of the lung epithelium. The endogenous ligand in this model was determined to be the extracellular matrix proteoglycan hyaluronan. When TLR signaling induced by hyaluronan was blocked with inhibitory peptides, the increased lung pathology was similar to that seen in the TLR2/TLR4 deficient mice (Jiang et al. 2005). In an analogous system, mice sterilized of their gut flora demonstrated increased intestinal injury after treatment with dextran sulfate sodium (DSS) compared with wild-type mice. Further, this increased DSS-induced intestinal damage was similar to that seen in MyD88 deficient mice, indicating that signals from the commensal bacteria is important in tissue preservation (Rakoff-Nahoum et al. 2004). One mechanism by which TLR signaling may result in tissue protection is the delivery of anti-apoptotic signals to cells, preventing cell loss. This may allow TLR signaling to establish the threshold of tolerable injury and cell death. Downstream products of TLR signaling such as cyclooxygenase-2 provide a signal for progenitor cells to proliferate, restoring the damaged tissue. In other models of tissue injury, signaling through TLRs has been associated with increased tissue pathology. Ischemia followed by reperfusion results in inflammation and oxidative stress of the affected tissue. In various models of ischemia–reperfusion including liver, kidney, heart, and brain, TLR4-deficient mice are relatively protected (Oyama et al. 2004; Tang et al. 2007; Tsung et al. 2005b; Wu et al. 2007). The cells in which TLR signaling is critical for this pathology vary. For instance, in the liver, the abrogation of TLR4 signaling in hematopoietic cells is protective (Tsung et al. 2005a), whereas in the kidney, the abrogation of TLR4 signaling in parenchymal cells is protective (Zhang et al. 2008). HMGB1 is an endogenous TLR ligand that is upregulated in both liver and kidney after ischemia–reperfusion. Antibodies that bind and deplete HMGB1 after liver ischemia–reperfusion are protective of subsequent tissue damage. Further, the depletion of HMGB1 in mice deficient in both TLR2 and TLR4 does not confer any additional protection from tissue pathology (Tsung et al. 2005b). This suggests that the protection conferred in the TLR2 and TLR4 deficient mouse model is due to the loss of signaling downstream of TLR ligand HMGB1. Two extracellular matrix proteins known to be endogenous TLR ligands, biglycan and hyaluronan, are induced by ischemia–reperfusion (Wu et al. 2007), but their role in mediating tissue injury has yet to be determined. TLR signaling also promotes tissue regeneration in several models. After resection of part of the liver, MyD88-deficient mice exhibit delayed regeneration of the liver (Seki et al. 2005). Although the ligands for TLRs leading to this regenerative effect of MyD88 signaling are not known, it is thought they may be from intestinal commensal bacteria as mice grown in germ-free conditions have similarly impaired

9

TLR Signaling and Tumour-Associated Macrophages

127

liver regeneration after resection. Models of skin wound healing have also demonstrated that MyD88 deficient mice heal at a slower rate with less granulation tissue and fewer vessels within granulation tissue of the wound (Macedo et al. 2007).

TLRs in Cancer and Tumour-Associated Macrophages Correlations between inflammation and cancer have been made over the past several hundred years. In some instances, it has been observed that infections correlate with the remission of tumours. Deidier observed that tumours resolved after patients developed venereal infections in the eighteenth century (Rakoff-Nahoum and Medzhitov 2009). In contrast, Virchow suggested in the nineteenth century that sites of chronic inflammation are the origin of some cancers (Balkwill and Mantovani 2001). Experimental models have demonstrated evidence supporting both of these observations, with TLR signaling leading to either the regression or the promotion of various tumour types. Further, genetic association studies have linked polymorphisms within various TLR loci (TLR1, TLR2, TLR3, TLR4, TLR6, and TLR10) to an increased risk of prostate, breast, colorectal, and nasopharyngeal cancers (El-Omar et al. 2008; Rakoff-Nahoum and Medzhitov 2009). Whether these TLR variants display a gain or loss of function remains to be determined. Although the cell types in which TLR signaling is relevant to disease outcomes have not firmly been established in all model cases, there is increasing evidence that TLR signaling in macrophages has a profound impact on tumour progression.

TLR Signaling Causes Tumour Regression In the late nineteenth century, William Coley read and observed that postoperative infections after resection of tumours often lead to a better clinical outcome. To determine whether this was due to live microorganisms or their products, Coley generated a combination of killed Gram-positive and Gram-negative bacteria, which included Streptococcus pyogenes and Serratia marcescens, respectively (Hennessy et al. 2010). Intratumoural injection of these “toxins” was observed to have antitumour responses (Hallam et al. 2009). In the mid-twentieth century, Shear and Turner determined that LPS was the active antitumour component in Coley’s toxins (RakoffNahoum and Medzhitov 2009). Therefore, TLR signaling likely mediated these original observations of antitumour response to TLR ligand injections. Proof for the antitumour efficacy of TLR ligands comes in the form of two agents approved by the US Food and Drug Administration for the treatment of cancers. First, the Bacillus Calmette-Guerin (BCG) vaccine, a live attenuated strain of Mycobacterium bovis, is approved for the treatment of primary and relapsed superficial transitional cell bladder carcinoma. The use of this agent is mucosal via intravesicular administration into the bladder. The effect of the BCG vaccine may be through

128

O.R. Colegio and R. Medzhitov

TLR2 and TLR4 signaling as abrogation of signaling through these TLRs either by knockout or by blocking antibodies inhibited the maturation of and proinflammatory cytokine production by myeloid DCs in model systems (Uehori et al. 2003). Second, imiquimod is an imidazoquinoline analog of ssRNA that signals through TLR7. Imiquimod is approved by the US Food and Drug Administration for the treatment of superficial basal cell carcinomas and actinic keratoses (intraepidermal neoplasm of dysplastic keratinocytes). Imiquimod induces maturation and proinflammatory cytokine production in DCs and leads to a Th1 antitumour lymphocyte response (Clark et al. 2008; Schon and Schon 2008). TLR7 is also expressed on macrophages and keratinocytes. The critical responsive cell(s) by which this TLR7 ligand acts has yet to be characterized. TLR signaling from endogenous ligands has recently been determined to have a profound impact on the clinical response to chemotherapy and radiation therapy (Apetoh et al. 2007). Treatment of solid tumours often includes chemotherapy and/ or radiation therapy. The efficacy of these therapies has been attributed to their direct elimination of tumour cells. However, an adaptive immune response stemming from these destructive therapies was determined to affect clinical outcome. In mice and humans, dying tumour cells treated with chemotherapy or radiotherapy release the nuclear protein HMGB1, which activates the TLR4–MyD88 pathway in DCs leading to tumour antigen-specific T-cell immunity. The activation of TLR4 by HMGB1 is essential for efficient cross-presentation of tumour antigens and effective antitumour response. Further, a cohort of breast cancer patients carrying a sequence polymorphism of TLR4, which is hypo-responsive to LPS, was found to relapse more quickly after chemotherapy or radiation therapy than the wild-type allele (Apetoh et al. 2007). Various TLR ligands are in clinical trials for the treatment of cancer, infections, and autoimmune diseases. Currently, CpG-based oligonucleotides that target TLR9 are in clinical trials for the treatment of non-small cell lung cancer (Krieg 2008). Other TLR9 ligands are in clinical trials for the treatment of non-Hodgkin’s lymphoma as well as prostate and colorectal cancer (Hennessy et al. 2010). Specific to the effect of TLR signaling in TAMs, injection of M13 bacteriophage into an ectopic tumour model of melanoma resulted in a MyD88-dependent shift from the alternatively activated M2 phenotype to the classically activated M1 phenotype. This shift to the M1 phenotype correlated with an increase production of proinflammatory cytokines and an influx of neutrophils into the tumour (Eriksson et al. 2009).

TLR Signaling Promotes Tumourigenesis In contrast to the relatively protective effects resulting from TLR signaling described above, numerous studies have revealed that TLR signaling can conversely lead to tumour progression and metastasis. In a murine model of breast cancer metastasis, 4T1 mammary adenocarcinoma cells injected intravenously resulted in increased angiogenesis, vascular permeability, and tumour cell invasion in the group receiving

9

TLR Signaling and Tumour-Associated Macrophages

129

intraperitoneal LPS compared with the control groups (Harmey et al. 2002). Analogously, in a murine model of colon carcinoma metastasis, CT26 colon adenocarcinoma cells injected intravenously developed an increased number of tumour nodules and lung weights in the group receiving intraperitoneal LPS than the control mice. In this case, TLR-signaling was dependent on a supportive cell type such as a TAM. This is likely because NF-kB signaling within the tumour cells which was found critical to the anti-apoptotic effects was observed in the wild-type but not the TLR4 mutant mice (Luo et al. 2004). Whereas the initial studies on the tumour-promoting effects of TLR signaling used isolated TLR ligands, more recent studies have studied the effect of bacterial infection on tumourigenesis. The clearest example of how bacterial infection can promote tumourigenesis is the association of chronic infection with Helicobacter pylori and the development of gastric cancer. To study the role of infection on tumour progression, Listeria moncytogenes, a Gram-positive facultative intracellular bacterium, was injected into ectopically implanted H22 hepatocellular carcinoma cells. The Listeria survived in larger but not smaller tumours. Further, Listeria increased growth of these tumours, which could be abrogated with silencing of TLR2 but not TLR4 expression (Huang et al. 2007). MyD88 signaling is crucial for tumour progression in both spontaneous Apcmin/+ model and the azoxymethane chemical carcinogenesis model of intestinal tumourigenesis. In the first model, Apcmin/+ mice lack one copy of the adenomatous polyposis coli (APC) tumour suppressor gene. Therefore, upon loss of the working copy of the gene, development of a focus of atypical epithelia begins. Apcmin/+ MyD88−/− mice had fewer and smaller polyps than Apcmin/+ mice; however, the number of microadenomas was not different between the Apcmin/+ MyD88−/− mice and the Apcmin/+ mice. Similarly, in azoxymethane-induced carcinogenesis, the incidence of tumour formation was decreased in the MyD88−/− mice. Therefore, MyD88 signaling is essential for tumour progression but not initiation (Rakoff-Nahoum and Medzhitov 2007). TAMs are associated with a worse clinical outcome in a majority of tumour types. Yet, few tumour-derived TLR ligands that activate TAMs to promote tumour progression and metastasis have been identified and characterized. One clear example was the product of a biochemical screen of macrophage-activating factors secreted by tumours. The extracellular matrix protein versican produced by Lewis lung carcinoma cells induced the secretion of IL-6 and TNFD by macrophages through TLR2/TLR6 signaling. Further, direct binding of TLR2 to versican was demonstrated. Abrogation of signaling through TLR2/TLR6 and MyD88 resulted in increased survival and fewer metastases in a murine model of metastatic lung cancer (Kim et al. 2009).

Conclusions Tumour progression has long been associated with inflammation. In particular, increased macrophage density in solid tumours as well as lymphomas has been associated with a poorer prognosis. Inherent to cancer is the loss of normal tissue

130

O.R. Colegio and R. Medzhitov

architecture, which may provide signals or perpetual tissue stress and damage. Cancers have been compared with wounds that do not heal (Dvorak 1986) and they both have significant overlap in their gene expression signatures (Chang et al. 2004). Among the signals for tissue damage are endogenous TLR ligands. As macrophages have the physiologic role of restoring tissue integrity and homeostasis after tissue damage, cancers exploit them as a rich source of growth factors and angiogenic factors (Pollard 2009). Future therapeutic opportunities to modulate the progression of cancer will come from understanding the signals produced by tumours, and the pathways they educate macrophages into trophic TAMs. In addition, early studies suggest that TLR signaling may enable the production of an acquired immune response and, therefore, a more durable remission after treatment. Dissecting and modulating both the trophic and acquired immune responses from TLR signaling will be challenging but will have great therapeutic promise.

References Alexopoulou L, Holt AC, Medzhitov R, Flavell RA (2001) Recognition of double-stranded RNA and activation of NF-kappaB by Toll-like receptor 3. Nature 413:732–738 Apetoh L, Ghiringhelli F, Tesniere A, Obeid M, Ortiz C, Criollo A, Mignot G, Maiuri MC, Ullrich E, Saulnier P et al (2007) Toll-like receptor 4-dependent contribution of the immune system to anticancer chemotherapy and radiotherapy. Nat Med 13:1050–1059 Balkwill F, Mantovani A (2001) Inflammation and cancer: back to Virchow? Lancet 357:539–545 Bernasconi NL, Traggiai E, Lanzavecchia A (2002) Maintenance of serological memory by polyclonal activation of human memory B cells. Science 298:2199–2202 Biragyn A, Ruffini PA, Leifer CA, Klyushnenkova E, Shakhov A, Chertov O, Shirakawa AK, Farber JM, Segal DM, Oppenheim JJ et al (2002) Toll-like receptor 4-dependent activation of dendritic cells by beta-defensin 2. Science 298:1025–1029 Blander JM, Medzhitov R (2004) Regulation of phagosome maturation by signals from toll-like receptors. Science 304:1014–1018 Blander JM, Medzhitov R (2006a) On regulation of phagosome maturation and antigen presentation. Nat Immunol 7:1029–1035 Blander JM, Medzhitov R (2006b) Toll-dependent selection of microbial antigens for presentation by dendritic cells. Nature 440:808–812 Carty M, Goodbody R, Schroder M, Stack J, Moynagh PN, Bowie AG (2006) The human adaptor SARM negatively regulates adaptor protein TRIF-dependent Toll-like receptor signaling. Nat Immunol 7:1074–1081 Chang HY, Sneddon JB, Alizadeh AA, Sood R, West RB, Montgomery K, Chi JT, van de Rijn M, Botstein D, Brown PO (2004) Gene expression signature of fibroblast serum response predicts human cancer progression: similarities between tumors and wounds. PLoS Biol 2:E7 Clark RA, Huang SJ, Murphy GF, Mollet IG, Hijnen D, Muthukuru M, Schanbacher CF, Edwards V, Miller DM, Kim JE et al (2008) Human squamous cell carcinomas evade the immune response by down-regulation of vascular E-selectin and recruitment of regulatory T cells. J Exp Med 205:2221–2234 Coban C, Ishii KJ, Kawai T, Hemmi H, Sato S, Uematsu S, Yamamoto M, Takeuchi O, Itagaki S, Kumar N et al (2005) Toll-like receptor 9 mediates innate immune activation by the malaria pigment hemozoin. J Exp Med 201:19–25 Condeelis J, Pollard JW (2006) Macrophages: obligate partners for tumor cell migration, invasion, and metastasis. Cell 124:263–266

9

TLR Signaling and Tumour-Associated Macrophages

131

Dvorak HF (1986) Tumors: wounds that do not heal. Similarities between tumor stroma generation and wound healing. N Engl J Med 315:1650–1659 Ellerman JE, Brown CK, de Vera M, Zeh HJ, Billiar T, Rubartelli A, Lotze MT (2007) Masquerader: high mobility group box-1 and cancer. Clin Cancer Res 13:2836–2848 El-Omar EM, Ng MT, Hold GL (2008) Polymorphisms in Toll-like receptor genes and risk of cancer. Oncogene 27:244–252 Eriksson F, Tsagozis P, Lundberg K, Parsa R, Mangsbo SM, Persson MA, Harris RA, Pisa P (2009) Tumor-specific bacteriophages induce tumor destruction through activation of tumor-associated macrophages. J Immunol 182:3105–3111 Ermolaeva MA, Michallet MC, Papadopoulou N, Utermohlen O, Kranidioti K, Kollias G, Tschopp J, Pasparakis M (2008) Function of TRADD in tumor necrosis factor receptor 1 signaling and in TRIF-dependent inflammatory responses. Nat Immunol 9:1037–1046 Ganguly D, Chamilos G, Lande R, Gregorio J, Meller S, Facchinetti V, Homey B, Barrat FJ, Zal T, Gilliet M (2009) Self-RNA-antimicrobial peptide complexes activate human dendritic cells through TLR7 and TLR8. J Exp Med 206:1983–1994 Gerondakis S, Grumont RJ, Banerjee A (2007) Regulating B-cell activation and survival in response to TLR signals. Immunol Cell Biol 85:471–475 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 219:143–152 Harmey JH, Bucana CD, Lu W, Byrne AM, McDonnell S, Lynch C, Bouchier-Hayes D, Dong Z (2002) Lipopolysaccharide-induced metastatic growth is associated with increased angiogenesis, vascular permeability and tumor cell invasion. Int J Cancer 101:415–422 Hennessy EJ, Parker AE, O’Neill LA (2010) Targeting Toll-like receptors: emerging therapeutics? Nat Rev Drug Discov 9:293–307 Huang B, Zhao J, Shen S, Li H, He KL, Shen GX, Mayer L, Unkeless J, Li D, Yuan Y et al (2007) Listeria monocytogenes promotes tumor growth via tumor cell toll-like receptor 2 signaling. Cancer Res 67:4346–4352 Imai Y, Kuba K, Neely GG, Yaghubian-Malhami R, Perkmann T, van Loo G, Ermolaeva M, Veldhuizen R, Leung YH, Wang H et al (2008) Identification of oxidative stress and Toll-like receptor 4 signaling as a key pathway of acute lung injury. Cell 133:235–249 Jiang D, Liang J, Fan J, Yu S, Chen S, Luo Y, Prestwich GD, Mascarenhas MM, Garg HG, Quinn DA et al (2005) Regulation of lung injury and repair by Toll-like receptors and hyaluronan. Nat Med 11:1173–1179 Jiang D, Liang J, Noble PW (2007) Hyaluronan in tissue injury and repair. Annu Rev Cell Dev Biol 23:435–461 Karin M, Lawrence T, Nizet V (2006) Innate immunity gone awry: linking microbial infections to chronic inflammation and cancer. Cell 124:823–835 Kawai T, Akira S (2010) The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors. Nat Immunol 11:373–384 Kim S, Takahashi H, Lin WW, Descargues P, Grivennikov S, Kim Y, Luo JL, Karin M (2009) Carcinoma-produced factors activate myeloid cells through TLR2 to stimulate metastasis. Nature 457:102–106 Kluwe J, Mencin A, Schwabe RF (2009) Toll-like receptors, wound healing, and carcinogenesis. J Mol Med 87:125–138 Krieg AM (2008) Toll-like receptor 9 (TLR9) agonists in the treatment of cancer. Oncogene 27: 161–167 Lande R, Gregorio J, Facchinetti V, Chatterjee B, Wang YH, Homey B, Cao W, Su B, Nestle FO, Zal T et al (2007) Plasmacytoid dendritic cells sense self-DNA coupled with antimicrobial peptide. Nature 449:564–569 Lin EY, Nguyen AV, Russell RG, Pollard JW (2001) Colony-stimulating factor 1 promotes progression of mammary tumors to malignancy. J Exp Med 193:727–740 Liu YY, Sun LC, Wei JJ, Li D, Yuan Y, Yan B, Liang ZH, Zhu HF, Xu Y, Li B et al (2010) Tumor cell-released TLR4 ligands stimulate Gr-1+ CD11b+ F4/80+ cells to induce apoptosis of activated T cells. J Immunol 185:2773–2782

132

O.R. Colegio and R. Medzhitov

Luo JL, Maeda S, Hsu LC, Yagita H, Karin M (2004) Inhibition of NF-kappaB in cancer cells converts inflammation-induced tumor growth mediated by TNFalpha to TRAIL-mediated tumor regression. Cancer Cell 6:297–305 Macedo L, Pinhal-Enfield G, Alshits V, Elson G, Cronstein BN, Leibovich SJ (2007) Wound healing is impaired in MyD88-deficient mice: a role for MyD88 in the regulation of wound healing by adenosine A2A receptors. Am J Pathol 171:1774–1788 Mancuso G, Gambuzza M, Midiri A, Biondo C, Papasergi S, Akira S, Teti G, Beninati C (2009) Bacterial recognition by TLR7 in the lysosomes of conventional dendritic cells. Nat Immunol 10:587–594 Medzhitov R, Preston-Hurlburt P, Janeway CA Jr (1997) A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature 388:394–397 O’Neill LA, Bowie AG (2007) The family of five: TIR-domain-containing adaptors in Toll-like receptor signalling. Nat Rev Immunol 7:353–364 Oyama J, Blais C Jr, Liu X, Pu M, Kobzik L, Kelly RA, Bourcier T (2004) Reduced myocardial ischemia-reperfusion injury in toll-like receptor 4-deficient mice. Circulation 109:784–789 Palm NW, Medzhitov R (2009) Pattern recognition receptors and control of adaptive immunity. Immunol Rev 227:221–233 Parroche P, Lauw FN, Goutagny N, Latz E, Monks BG, Visintin A, Halmen KA, Lamphier M, Olivier M, Bartholomeu DC et al (2007) Malaria hemozoin is immunologically inert but radically enhances innate responses by presenting malaria DNA to Toll-like receptor 9. Proc Natl Acad Sci USA 104:1919–1924 Pobezinskaya YL, Kim YS, Choksi S, Morgan MJ, Li T, Liu C, Liu Z (2008) The function of TRADD in signaling through tumor necrosis factor receptor 1 and TRIF-dependent Toll-like receptors. Nat Immunol 9:1047–1054 Pollard JW (2004) Tumour-educated macrophages promote tumour progression and metastasis. Nat Rev Cancer 4:71–78 Pollard JW (2009) Trophic macrophages in development and disease. Nat Rev Immunol 9: 259–270 Rakoff-Nahoum S, Medzhitov R (2007) Regulation of spontaneous intestinal tumorigenesis through the adaptor protein MyD88. Science 317:124–127 Rakoff-Nahoum S, Medzhitov R (2009) Toll-like receptors and cancer. Nat Rev Cancer 9:57–63 Rakoff-Nahoum S, Paglino J, Eslami-Varzaneh F, Edberg S, Medzhitov R (2004) Recognition of commensal microflora by toll-like receptors is required for intestinal homeostasis. Cell 118: 229–241 Scaffidi P, Misteli T, Bianchi ME (2002) Release of chromatin protein HMGB1 by necrotic cells triggers inflammation. Nature 418:191–195 Schaefer L, Babelova A, Kiss E, Hausser HJ, Baliova M, Krzyzankova M, Marsche G, Young MF, Mihalik D, Gotte M et al (2005) The matrix component biglycan is proinflammatory and signals through Toll-like receptors 4 and 2 in macrophages. J Clin Invest 115:2223–2233 Schon MP, Schon M (2008) TLR7 and TLR8 as targets in cancer therapy. Oncogene 27:190–199 Seki E, Tsutsui H, Iimuro Y, Naka T, Son G, Akira S, Kishimoto T, Nakanishi K, Fujimoto J (2005) Contribution of Toll-like receptor/myeloid differentiation factor 88 signaling to murine liver regeneration. Hepatology 41:443–450 Sims GP, Rowe DC, Rietdijk ST, Herbst R, Coyle AJ (2010) HMGB1 and RAGE in inflammation and cancer. Annu Rev Immunol 28:367–388 Takeuchi O, Akira S (2010) Pattern recognition receptors and inflammation. Cell 140:805–820 Tang SC, Arumugam TV, Xu X, Cheng A, Mughal MR, Jo DG, Lathia JD, Siler DA, Chigurupati S, Ouyang X et al (2007) Pivotal role for neuronal Toll-like receptors in ischemic brain injury and functional deficits. Proc Natl Acad Sci USA 104:13798–13803 Tian J, Avalos AM, Mao SY, Chen B, Senthil K, Wu H, Parroche P, Drabic S, Golenbock D, Sirois C et al (2007) Toll-like receptor 9-dependent activation by DNA-containing immune complexes is mediated by HMGB1 and RAGE. Nat Immunol 8:487–496 Tsan MF, Gao B (2009) Heat shock proteins and immune system. J Leukoc Biol 85:905–910

9

TLR Signaling and Tumour-Associated Macrophages

133

Tsung A, Hoffman RA, Izuishi K, Critchlow ND, Nakao A, Chan MH, Lotze MT, Geller DA, Billiar TR (2005a) Hepatic ischemia/reperfusion injury involves functional TLR4 signaling in nonparenchymal cells. J Immunol 175:7661–7668 Tsung A, Sahai R, Tanaka H, Nakao A, Fink MP, Lotze MT, Yang H, Li J, Tracey KJ, Geller DA et al (2005b) The nuclear factor HMGB1 mediates hepatic injury after murine liver ischemiareperfusion. J Exp Med 201:1135–1143 Uehori J, Matsumoto M, Tsuji S, Akazawa T, Takeuchi O, Akira S, Kawata T, Azuma I, Toyoshima K, Seya T (2003) Simultaneous blocking of human Toll-like receptors 2 and 4 suppresses myeloid dendritic cell activation induced by Mycobacterium bovis bacillus Calmette-Guerin peptidoglycan. Infect Immun 71:4238–4249 Uematsu S, Fujimoto K, Jang MH, Yang BG, Jung YJ, Nishiyama M, Sato S, Tsujimura T, Yamamoto M, Yokota Y et al (2008) Regulation of humoral and cellular gut immunity by lamina propria dendritic cells expressing Toll-like receptor 5. Nat Immunol 9:769–776 Wang H, Bloom O, Zhang M, Vishnubhakat JM, Ombrellino M, Che J, Frazier A, Yang H, Ivanova S, Borovikova L et al (1999) HMG-1 as a late mediator of endotoxin lethality in mice. Science 285:248–251 West MA, Wallin RP, Matthews SP, Svensson HG, Zaru R, Ljunggren HG, Prescott AR, Watts C (2004) Enhanced dendritic cell antigen capture via toll-like receptor-induced actin remodeling. Science 305:1153–1157 Wiktor-Jedrzejczak W, Bartocci A, Ferrante AW Jr, Ahmed-Ansari A, Sell KW, Pollard JW, Stanley ER (1990) Total absence of colony-stimulating factor 1 in the macrophage-deficient osteopetrotic (op/op) mouse. Proc Natl Acad Sci USA 87:4828–4832 Willingham SB, Allen IC, Bergstralh DT, Brickey WJ, Huang MT, Taxman DJ, Duncan JA, Ting JP (2009) NLRP3 (NALP3, Cryopyrin) facilitates in vivo caspase-1 activation, necrosis, and HMGB1 release via inflammasome-dependent and -independent pathways. J Immunol 183:2008–2015 Wu H, Chen G, Wyburn KR, Yin J, Bertolino P, Eris JM, Alexander SI, Sharland AF, Chadban SJ (2007) TLR4 activation mediates kidney ischemia/reperfusion injury. J Clin Invest 117: 2847–2859 Xia ZP, Sun L, Chen X, Pineda G, Jiang X, Adhikari A, Zeng W, Chen ZJ (2009) Direct activation of protein kinases by unanchored polyubiquitin chains. Nature 461:114–119 Yanai H, Ban T, Wang Z, Choi MK, Kawamura T, Negishi H, Nakasato M, Lu Y, Hangai S, Koshiba R et al (2009) HMGB proteins function as universal sentinels for nucleic-acidmediated innate immune responses. Nature 462:99–103 Zhang D, Zhang G, Hayden MS, Greenblatt MB, Bussey C, Flavell RA, Ghosh S (2004) A Toll-like receptor that prevents infection by uropathogenic bacteria. Science 303:1522–1526 Zhang B, Ramesh G, Uematsu S, Akira S, Reeves WB (2008) TLR4 signaling mediates inflammation and tissue injury in nephrotoxicity. J Am Soc Nephrol 19:923–932

Chapter 10

SHIP and Tumour-Associated Macrophages Victor W. Ho, Melisa J. Hamilton, Etsushi Kuroda, Jens Ruschmann, Frann Antignano, Vivian Lam, and Gerald Krystal

Abbreviations ALL BMmIs BMMCs CML COX-2 DCs DLBCL Doks ES F.t. GvHD HA HSCs iNOS IP4 INPP4B ITAMs ITIMs LPS M-CSF mIs MDSCs

acute lymphocytic leukemia bone marrow-derived macrophages bone marrow mast cells chronic myelogenous leukemia cyclooxygenase-2 dendritic cells diffuse large B-cell lymphoma downstream of tyrosine kinases embryonic stem Francisella tularensis graft-versus-host disease hyaluronic acid hematopoietic stem cells inducible nitric oxide synthase inositol-1,3,4,5-tetrakisphosphate inositol polyphosphate 4-phosphatase type II immunoreceptor tyrosine-based activation motifs immunoreceptor tyrosine-based inhibitory motifs lipopolysaccharide macrophage colony-stimulating factor macrophages myeloid-derived suppressor cells

67(Os-*(AMILTONs%+URODAs*2USCHMANNs&!NTIGNANOs6,AMs'+RYSTAL*) The Terry Fox Laboratory, B.C. Cancer Agency, Vancouver, BC V5Z 1L3, Canada e-mail: [email protected] T. Lawrence and T. Hagemann (eds.), Tumour-Associated Macrophages, DOI 10.1007/978-1-4614-0662-4_10, © Springer Science+Business Media, LLC 2012

135

136

M)2S .+ NO PGE2 PGN PH PIAS1 0)+ PIP3 0+" PTB PTEN 2/3 SH2 SHIP sSHIP 4,2S 542

V.W. Ho et al.

MICRO2.!S NATURALKILLER nitric oxide prostaglandin E2 peptidoglycan pleckstrin homology protein inhibitor of activated STAT1 PHOSPHATIDYLINOSITOL  KINASE phosphatidylinositol 3,4,5-trisphosphate PROTEINKINASE" phosphotyrosine binding phosphatase and tensin homologue deleted on chromosome ten REACTIVEOXYGENSPECIES Src homology 2 Src homology 2-containing inositol 5c-phosphatase stem cell SHIP 4OLL LIKERECEPTORS UNTRANSLATEDREGION

Introduction The biological properties of the SH2-containing inositol 5c-phosphatase, SHIP (also known as SHIP1 or SHIP1D) are inherently linked with the phosphatidylinositol 0)  KINASE0)+ PATHWAY4HISPATHWAYPLAYSACENTRALROLEINREGULATINGMANY biological processes, including proliferation, survival, differentiation, activation, and chemotaxis through the generation of the critical second messenger PI-3,4,5-P3 (PIP3 +RYSTAL 2000). This membrane-associated phospholipid, which is present at very low levels in resting cells, is transiently synthesized from PI-4,5-P2 by 0)+INRESPONSETOADIVERSEARRAYOFEXTRACELLULARSTIMULIANDATTRACTSPLECKSTRIN homology (PH)-containing proteins such as the survival/proliferation enhancing SERINETHREONINEKINASE !KT ALSOKNOWNASPROTEINKINASE"0+" TOMEDIATEITS effects. Consistent with its role in promoting proliferation and survival, the activaTIONOFTHE0)+PATHWAYHASBEENSHOWNTOBEACRITICALEVENTINTUMOURDEVELOPMENT WITHMANYONCOGENESEG %'&2 (ER.EU AND+ 2AS STIMULATINGTUMOUR GROWTH BY ENHANCING THE 0)+ PATHWAY &RY 2001). The fact that many cancers possess activating mutations or gene amplifications of the catalytic p110D subunit OF0)+#AMPBELLETAL2004) or Akt (Hennessy et al. 2005), while others have MUTATIONS IN THE P SUBUNIT OF 0)+ THAT ACTIVATE THE P D, E, or G subunits *AISWALETAL2009; Martin-Berenjeno and Vanhaesebroeck 2009), highlights the IMPORTANCEOFTIGHTLYREGULATINGTHE0)+PATHWAY4OENSURETHATTHEACTIVATIONOF this pathway is appropriately suppressed/terminated in normal cells, the ubiquitously expressed 54-kDa tumour suppressor PTEN (i.e., phosphatase and tensin homologue deleted on chromosome ten) hydrolyzes PIP3 back to PI-4,5-P2 while the

10

SHIP and TAMs

137

145-kDa hematopoietic-restricted SHIP and the more widely expressed 155-kDa SHIP2 hydrolyze it to PI-3,4-P2 (reviewed by Sly et al. 2003). The fact that PI-3,4-P2 can also act as a second messenger (Gewinner et al. 2009; Ma et al. 2008; Scheid et al. 2002; Zhang et al. 2009b) complicates the notion of SHIP and SHIP2 as simPLETERMINATORSOFTHE0)+PATHWAYANDTHISWILLBEDISCUSSEDINMOREDETAILLATER As far as the relative contribution of these phosphatases is concerned, there is some evidence that PTEN functions primarily to maintain PIP3 levels below a critical threshold level in unstimulated/resting cells (reviewed by Antignano et al. 2009), while SHIP and SHIP2 are thought to be inactive until cells are stimulated, at which time they are recruited to the plasma membrane, and likely other membranes, via TYROSINE PHOSPHORYLATED PROTEINS TO ANTAGONIZE STIMULUS INDUCED 0)+ SIGNALING (Downes et al. 2007). Consistent with this, PTEN is transiently inactivated by many EXTRACELLULARSTIMULI VIATHEIRINDUCTIONOFREACTIVEOXYGENSPECIES2/3 THATCAUSE disulfide bridge formation in the active site of PTEN (Salmeen and Barford 2005) and/or by their triggering the serine/threonine phosphorylation of residues within the C-terminus of PTEN (Leslie et al. 2008; Lu et al. 2003). 4HEIMPORTANTROLETHAT04%.AND3()0PLAYINKEEPINGTHE0)+PATHWAYIN check is underscored by the fact that almost 50% of human cancers contain biallelic inactivating mutations of PTEN (Cantley and Neel 1999) and that SHIP has been shown to be mutated or markedly reduced in many leukemias and lymphomas (Costinean et al. 2009; Lo et al. 2009; O’connell et al. 2009; Pedersen et al. 2009; Yamanaka et al. 2009). In contrast, a tumour suppressor role for SHIP2 has not as yet been demonstrated and preliminary studies suggest that it may in fact promote tumour progression (Prasad et al. 2008a, b) but further studies are needed to corroborate these findings. There are several excellent reviews that have been published recently on PTEN, SHIP, and/or SHIP2 (Astle et al. 2007; Frimberger et al. 2001; Gloire et al. 2007; Gratacap et al. 2008; Harris et al. 2008+ASHIWADAETAL2007; Leung et al. 2009; Ooms et al. 20092OHRSCHNEIDERETAL 2000; Sasaoka et al. 2006). In this review, we concentrate on how the levels of SHIP protein are regulated and on SHIP’s biological properties, with special emphasis on SHIP’s role in repressing the skewing of macrophages (mIs) to an alternatively activated (M2) phenotype.

SHIP’s Structure and Binding Partners The gene for SHIP, INPP5D (EC 3.1.3.56), which is located on chromosome 2q36/37 in humans and contains 29 exons and 1C5 in mice and contains 27 exons, encodes a 145-kDa protein that possesses, as shown in Fig. 10.1, an amino-terminal Src homology 2 (SH2) domain that binds, such as SHIP2’s SH2 domain, to the sequence pY(Y/S)(L/Y/M)(L/M/I/V), but with faster association and dissociation rates (Sweeney et al. 2005; Wavreille et al. 2007; Zhang et al. 2009c). SHIP has been shown to bind, via its SH2 domain, to the tyrosine-phosphorylated forms of Shc, SHP-2, downstream of tyrosine kinases (Doks), Gabs, CD150, PECAM, Cas,

138

V.W. Ho et al.

Shc, SHP-2, ITIMs, ITAMs, Doks, Gabs, CD150, PECAM, Cas, c-Cbl

Grb2, Src, Abl, PLCYI, Tec, PIAS1

PI-3,4-P2 Shc, Dok1, 2 p85, SHIP2

SHIP1A

NPNY Y

SH2

C2

SHIP1B

SH2

C2

SHIP1G

SH2

C2

NPLY Y Pro-rich

1188

110 135 125 NPNYNPLY Y Y Pro-rich 1130

NPNY Y

sSHIP

C2

959

NPNY NPLY Y Y Pro-rich

928

Fig. 10.1 The structure and binding partners of the different forms of SHIP. The NPNY and NPLY motifs (Y917 and Y1020 in murine SHIP and Y915 and Y1022 in human SHIP, respectively) in their phosphorylated state can bind proteins with PTB domains. There is also evidence for phosphorylation on Y867 and Y944 of mouse SHIP, at least in IgE+Ag-treated BMMCs (Cao et al. 2007). A C2 domain reported to bind PI-3,4-P2 (Ong et al. 2007) is indicated. PEST sequences are shown as black arrows under each protein. Putative protein cleavage sites are shown as white arrows under, and proline-rich motifs as open triangles above, full-length SHIP. The total number of amino acids is shown to the right of each protein

c-Cbl (Yogo et al. 2006), certain immunoreceptor tyrosine-based inhibitory motifs (ITIMs) and certain immunoreceptor tyrosine-based activation motifs (ITAMs) (reviewed by Sly et al. 2003). SHIP also contains a centrally located phosphoinositol phosphatase domain of 400–500 amino acids that selectively hydrolyzes the 5c-phosphate from PIP3, PI-4,5-P2 (Pesesse et al. 2006), and inositol-1,3,4,5tetrakisphosphate (IP4 *UST# TERMINALTOTHISPHOSPHATASEDOMAINISA#DOMAIN that Alice Mui’s lab has recently identified and shown to be an allosteric activating site when bound by SHIP’s enzymatic product, PI-3,4-P2 (Ong et al. 2007). Within the proline-rich C terminus are two NPXY motifs that, when phosphorylated, bind proteins with a phosphotyrosine binding (PTB) (e.g., Shc, Dok 1, and Dok 2) or an SH2 (e.g., p85D and SHIP2) domain. As well, the proline-rich C terminus binds, via its 4 PxxP motifs, a subset of SH3-containing proteins including Grb2, Src, Lyn,

10

SHIP and TAMs

139

Hck, Abl, PLCJ AND 0)!3 REVIEWED BY 2OHRSCHNEIDER ET AL 2000; Sly et al. 2003; Song et al. 2005; Wisniewski et al. 1999). In addition to full-length SHIP, there appears to be a low abundance 160-kDa form in some cells (Geier et al. 1997; Ono et al. 1996) which recent work suggests MAYBEPOLY UBIQUITINATED3()02USCHMANNETAL2010). As well, there are two alternate splice forms: a 135-kDa SHIPE and a 110-kDa SHIPG (Fig. 10.1). SHIPE is the result of an in-frame deletion of 183 nucleotides immediately after the first NPXY motif that yields a protein missing 61 amino acids and changes the amino acids downstream of the first NPXY from NPXYIGM to NPXYIAN so that, following TYROSINEPHOSPHORYLATION ITNOLONGERBINDSTHEPSUBUNITOF0)+'UPTAETAL 1999 ,UCAS AND 2OHRSCHNEIDER 1999). SHIPG is the product of an out-of-frame splice that yields a protein with a unique C-terminal 41 amino acids and missing the second NPXY motif and all four C-terminal proline-rich motifs. Very little is known about this protein other than it is poorly tyrosine phosphorylated and not associated with Shc after M-CSF stimulation of FD-Fms cells (Dybedal et al. 2001). As well, 135, 125, and 110 kDa C-terminally truncated forms of SHIP have been identified in NP40- or TX100-solubilized cell lysates, and there has been some controversy about whether they occur in intact cells (Damen et al. 1998; Horn et al. 2001). However, recent studies suggest that they are, in fact, present in intact cells and may BEINTERMEDIATESFORMEDDURINGPROTEASOMALDEGRADATION2USCHMANNETAL2010). In addition to full-length SHIP, its two splice forms and proteasomal degradation products, a 104-kDa stem cell SHIP (sSHIP) has been identified (Fig. 10.1). This sSHIP, originally cloned from a human placental cDNA library and called SIP-110 +AVANAUGH ET AL 1996), is the only form of SHIP expressed in embryonic stem (ES) cells (Tu et al. 2001) and is coexpressed, albeit at low levels (Dr. Connie Eaves, personal communication), with full-length SHIP in hematopoietic stem cells (HSCs) (Tu et al. 2001). In both cases, it disappears with differentiation and is replaced, as (3#SDIFFERENTIATE WITHFULL LENGTH3()0+ALESNIKOFFETAL2003 S3()0M2.! is transcribed from a promoter within the intron between exons 5 and 6 of the SHIP gene and is likely restricted to a subset of stem/progenitor cells during mouse embryo DEVELOPMENT2OHRSCHNEIDERETAL2005). Since the sSHIP protein is truncated at its N terminus, it lacks an SH2 domain and does not become tyrosine phosphorylated or associated with Shc following stimulation (Tu et al. 2001). It does, however, bind constitutively to Grb2 and so may be recruited via Grb2’s SH2 domain to the plasma membrane to regulate PIP3 levels in stem cells (Tu et al. 2001).

The Phenotype of SHIP Knockout Mice A number of ubiquitous (Collazo et al. 2009; Helgason et al. 1998) as well as B cell-, T cell-, and mI-specific SHIP knockout mice (Leung et al. 2009; Liu et al. 1998a; Tarasenko et al. 2007) have been generated during the past 12 years. Although germline, ubiquitous SHIP−/− C57BL/6 mice are viable, they have a shortened

140

V.W. Ho et al.

lifespan, likely due, at least in part, to chronic allergic inflammation of their lungs (Helgason et al. 1998; Oh et al. 2007 2ELATEDTOTHIS (ADDONETALRECENTLYSHOWED that these mice have elevated numbers of hyper-responsive mast cells and that these cells likely play a major role in the elevated serum levels of IL-6, TNFD, and IL-5 as well as the lung inflammation that characterize these mice (Haddon et al. 2009). These SHIP−/− mice also suffer from severe osteoporosis, due to an increased number of hyper-resorptive osteoclasts (Takeshita et al. 2002 AND ABNORMAL .+ CELLS characterized by an abnormally high expression of inhibitory receptors. This latter property likely plays a role in the reduced acute graft-versus-host disease (GvHD) and absence of allograft rejection of SHIP−/− mice (Wang et al. 2002). However, these SHIP−/− mice also have elevated numbers of Gr-1+CD11b+ myeloid-derived suppressor cells (MDSCs) and Tregs (Collazo et al. 2009+ASHIWADAETAL2006), both of which very likely contribute to the suppression of GvHD and allograft acceptance (Ghansah et al. 2004). Importantly, the myeloid progenitors from ubiquitous SHIP−/− mice are far more responsive to suboptimal levels of cytokines, growth factors, and chemokines than their SHIP+/+ counterparts (Helgason et al. 1998+IMETAL1999) and this likely explains why SHIP−/− mice overproduce granulocytes, mast cells (Haddon et al. 2009), and mIs, and force erythropoiesis out of the bone marrow and into the spleen and elsewhere (Helgason et al. 1998). As a result, these mice develop splenomegaly, due to both extramedullary erythropoiesis and myelopoiesis (Helgason et al. 1998). In addition, while circulating erythrocytes and platelets are only slightly reduced in number, despite an elevated number of primitive erythroid and megakaryocytic progenitors (Mason et al. 2002; Perez et al. 2008), B cells are significantly reduced as the SHIP−/− mice age because the elevated numbers of mIs produce high levels of IL-6 in response to increasing IgG (Maeda et al. 2010), and this inhibits B cell and enhances myeloid cell development (Nakamura et al. 2004). As far as T cells are concerned, ubiquitous SHIP−/− mice have slightly reduced peripheral T cell numbers but markedly elevated numbers of CD4+CD25+ Tregs, as mentioned above, and CD62Llow effector T cells (Collazo et al. 2009+ASHIWADAETAL2006). However, studies with T cell-specific and mI/ granulocyte-specific SHIP−/− mice suggest that this is not due to the absence of SHIP within T cells (Tarasenko et al. 2007) but to the lack of SHIP in mIs and granulocytes and the resulting inflammatory environment this generates (Leung et al. 2009; Tarasenko et al. 2007). This contrasts somewhat with more recent studies suggesting that SHIP within CD4+ T helper cells is essential for normal Th17 development and the absence of SHIP skews these Th cells towards Tregs (Locke et al. 2009). Since many of the characteristics of ubiquitous SHIP−/− mice appear to worsen with age, it is possible that they are the result of compensatory mechanisms to offset the damage induced by the absence of SHIP. For example, the increase in MDSCs (Ghansah et al. 2004) and Tregs (Tarasenko et al. 2007) may occur to counter the inflammatory environment created by the overproduction and hyperactivity of neutrophils and monocyte/mIS2ELATEDTOTHIS ANDDESCRIBEDINMOREDETAILLATER one of the most interesting phenotypic properties of SHIP−/− mice is that their mIs tend to be skewed away from a killer M1 phenotype and towards a healer M2 pheNOTYPEASTHEMICEREACHADULTHOOD2AUHETAL2005). Since Savage et al. recently

10

SHIP and TAMs

141

showed that human anti-inflammatory mIs induce Tregs (Savage et al. 2008), this M2 skewing in SHIP−/− mice might represent an early compensatory event to prevent SHIP−/− mice from going into pro-inflammatory cytokine-induced shock.

The Mechanism of Action of SHIP SHIP, present in unstimulated cells primarily within the cytoplasm and also readily detectable in cytoskeletal and membrane fractions, is constitutively associated with 'RB AND #). +OWANETZ ET AL 2004; Busche et al. 2011), at least in some hematopoietic cells. In response to extracellular stimulation by cytokines, growth factors, antibodies, chemokines, and integrin ligands (Maxwell et al. 2004; Sly et al. 2003), as well as hypertonic and oxidative stressors, SHIP becomes both tyrosine phosphorylated, typically by members of the Src family (Baran et al. 2003), and bound to Shc or SHP-2. As well, there is often a slight but detectable increase in membrane association (Cox et al. 2001; Phee et al. 2000, 2001) and both the SH2 and C-terminal domains of SHIP are required for this translocation and for SHIP’s biological effects (Aman et al. 2000; Damen et al. 2001; Liu et al. 1997). Since SHIP’s 5c-phosphatase activity does not change significantly following extracellular stimulation (Damen et al. 1996) or subsequent tyrosine phosphorylation (Phee et al. 2000), it is thought that SHIP mediates its effects primarily by translocating to sites of PIP3, and perhaps IP4, synthesis. The mechanism of translocation to tyrosine phosphorylated cell surface receptors appears to vary, sometimes directly involving SHIP’s own SH2 domain or indirectly, via Shc’s PTB or SH2 domain (Bone and Welham 2000; Galandrini et al. 2001, 2002; Marchetto et al. 1999; 2AMSHAWETAL2007; Tridandapani et al. 1999, 2002; Velazquez et al. 2000). In contrast to tyrosine phosphorylation of SHIP, Zhang et al. recently reported that 0+! MEDIATED SERINETHREONINE PHOSPHORYLATION OF 3()0 INCREASES ITS CATALYTIC activity two- to threefold (Zhang et al. 2009a). Also, as mentioned earlier, recent evidence suggests that SHIP can be recruited to the plasma membrane via binding of its C2 domain to its product, PI-3,4-P2, and that this induces an allosteric activation of SHIP (Ong et al. 2007). SHIP has also been shown to translocate to the actin cytoskeleton, perhaps via actin-binding filamins (Cox et al. 2001; Dyson et al. 2001) and this might inhibit, for example, IgE-induced signaling in mast cells (Lesourne et al. 2005). Tying these last two phenomena together, it has been shown that PI-3,4-P2 slowly accumulates to very high levels in platelets following thrombin stimulation, and there is a striking correlation between the time course of this PI-3,4-P2 accumulation and the tyrosine phosphorylation and relocation of SHIP to the cytoskeleton (Gratacap et al. 2008). As mentioned earlier, there is also evidence that PI-3,4-P2 acts as a second MESSENGERINMANYCELLSBYATTRACTING0( CONTAININGPROTEINS SUCHAS0$+ !KT Phox47, Bam32, TAPP1 and -2, and lamellipodin (Allam and Marshall 2005+RAHN et al. 2004+RAUSEETAL2004; Marshall et al. 2002; Scheid et al. 2002; Zhang et al. 2009b). In fact, there is growing evidence that both PIP3 and PI-3,4-P2 are required

142

V.W. Ho et al.

for full phosphorylation/activation of Akt (Ma et al. 2008; Scheid et al. 2002). The fact that Steel Factor stimulation of SHIP−/− BMMCs, which results in an almost 100-fold increase in PIP3 levels but only half the increase in PI-3,4-P2 levels seen with WT cells, gives a dramatic increase in Akt phosphorylation (Scheid et al. 2002), suggests that half the usual level of PI-3,4-P2 is still sufficient to contribute to Akt activation. Further support for a signaling role for PI-3,4-P2 comes from recent work in Lewis Cantley’s lab showing that inositol polyphosphate 4-phosphatase type II (INPP4B), which preferentially hydrolyzes PI-3,4-P2 to PI-3P, reduces Akt activation and that knockdown of this enzyme increases and prolongs Akt activation (Gewinner et al. 2009). These results suggest that a coordinated generation of PIP3 and PI-3,4-P2 is important for the appropriate triggering of many intracellular events and thus one might expect that SHIP−/− cells, with elevated PIP3 but reduced PI-3,4-P2 (Huber et al. 1998, 1999), would be qualitatively different from PTEN−/− cells (with elevated PIP3 and PI-3,4-P2). Although most studies to date have concentrated on SHIP’s ability to hydrolyse PIP3, SHIP can hydrolyze IP4 in vitro (Damen et al. 1996) and may do so in intact cells as well (Valderrama-Carvajal et al. 2002). This could affect the levels of IP6, WHICHPLAYANESSENTIALROLEINTRANSPORTINGM2.!OFTHENUCLEUSFORTRANSLATION (Feng et al. 2001) and IP7, which compete with PIP3 for PH domain-containing proteins (Luo et al. 2003a). IP7 and IP8 have also been shown to phosphorylate multiple proteins via direct transfer of their E phosphate (Saiardi et al. 2004) and to regulate vesicle trafficking, apoptosis, DNA repair, telomere maintenance, and the stress response (Bennett et al. 2006). SHIP has also been shown to mediate some of its effects by acting as an adaptor (Forestell et al. 1995+ONCZETAL2001; Song et al. 2005). For example, there is evidence that during FcJ2))" MEDIATEDINHIBITIONOF" CELLACTIVATION 3()0REDUCES 2AS ACTIVITY BY RECRUITING PDok IE $OK TO 2AS'!0 &ORESTELL ET AL 1995). !SWELL 3()0HASBEENSHOWNTOASSOCIATEWITH$OKIN"CELLSINRESPONSETO"#2 engagement and this Dok3/SHIP complex appears to suppress B-cell activation via THEINHIBITIONOFTHE*NKPATHWAY2OBSONETAL2004). There is also evidence that 3()0 AND 3()0 DIRECTLY REDUCE 2AS ACTIVATION IN SOME CELLS BY COMPETING WITH Grb2 for Shc (Coggeshall 1998; Ishihara et al. 1999). As well, SHIP, via its proline-rich MOTIFS APPEARS TO BIND THE 3( DOMAIN OF 4EC +INASE AND INHIBIT THE MEMBRANE localization of Tec (Tomlinson et al. 2004). In addition, PIAS1 (protein inhibitor of activated STAT1) has been shown to bind the C terminus of SHIP in unstimulated cells and be released upon M-CSF stimulation. Although the biological significance of this interaction is not yet clear, released PIAS1 may then interact with STAT1 to modify - #3& STIMULATED34!4 INDUCEDTRANSCRIPTION2OHRSCHNEIDERETAL2000).

The Regulation of SHIP’s Protein Levels As mentioned earlier, sSHIP is the only form expressed during the very early stages OFMURINEDEVELOPMENT WITHFULL LENGTH3()0BEINGlRSTDETECTABLE BY24 0#2 IN day 7.5 embryos, coincident with the onset of hematopoiesis. SHIP’s protein expres-

10

SHIP and TAMs

143

sion pattern, both in the embryo and in the adult mouse, is restricted to hematopoietic cells, spermatids (Liu et al. 1998b), and endothelial cells (Tanigaki et al. 2009; Zippo et al. 2004). Within the hematopoietic system, protein levels of full-length SHIP and its two splice forms vary considerably (Geier et al. 1997; Liu et al. 1998b; Lucas AND2OHRSCHNEIDER19992OHRSCHNEIDERETAL2000; Wolf et al. 2000). For example, SHIP is lost when erythroid cells become Ter119++ALESNIKOFFETAL2003) and this may be required for terminal erythroid differentiation (Siegel et al. 1999). In contrast, SHIP levels increase substantially in both CD4+ and CD8+ T cells after positive selection (Liu et al. 1998b), show a bimodal expression pattern during B-cell development and are dramatically increased when resting B cells are activated (Brauweiler et al. 2001). SHIP is also present in mature neutrophils, basophils, monocyte/mIs, denDRITICCELLS$#S MASTCELLS PLATELETS ANDNATURALKILLER.+ CELLS#OXETAL2001; Geier et al. 1997; Giuriato et al. 1997; Maresco et al. 1999; Trotta et al. 2005). In terms of how SHIP levels are regulated, it has been shown that two potent inhibitors of hematopoietic cell proliferation and survival, transforming growth factor-E (TGFE) and activin, elicit their effects, in part, by dramatically upregulating 3()0M2.!ANDPROTEINLEVELS6ALDERRAMA #ARVAJALETAL2002 2ELATEDTOTHIS lipopolysaccharide (LPS) has also been found to increase SHIP, but not SHIP2 or PTEN, in bone marrow-derived macrophages (BMmIs) and mast cells (BMMCs), via the induction of autocrine-acting TGFE (Sly et al. 2004). Consistent with this, Cuschieri et al. reported that a 24-h pre-treatment with insulin reduces LPS-induced TNFD production from a human promonocytic cell line via the induction of autocrineacting activin A, which then increases SHIP levels (Cuschieri et al. 2007). Also, as far as SHIP levels are concerned, it is well established that there is an INVERSE RELATIONSHIP BETWEEN THE EXPRESSION OF "#2 !", AND 3()0 *IANG ETAL 2003; Martino et al. 2001), suggesting that reduced SHIP activity might be a prerequisite for the proliferative advantage of some chronic myelogenous leukaemic (CML) clones. Interestingly, a similar inverse relationship between a constitutively ACTIVEONCOGENICC KITRECEPTORIE +)4+%) and SHIP (but not SHIP2 or PTEN) has been reported (Vanderwinden et al. 2006) and the inhibition of c-kit’s intrinsic tyrosine kinase activity with Gleevec reversibly raises SHIP levels (Vanderwinden et al. 2006 2ELATEDTOTHIS WORKINOURLAB3LYETAL2004) and others (Cuschieri et al. 2007) suggests that SHIP expression levels likely play an important role in determining its activity and we have been exploring the pathway(s) by which it is DEGRADED3PECIlCALLY SINCE3ATTLERETALSHOWEDTHAT"#2 !",EXPRESSIONLEADS to a reduction in SHIP levels (Sattler et al. 1997), we have been examining the MECHANISMS INVOLVEDANDHAVEFOUNDTHAT"#2 !", DIRECTLYORVIAA3RCFAMILY member, tyrosine phosphorylates SHIP, which is constitutively associated with c-Cbl and Cbl-b, and this phosphorylation leads to SHIP’s poly-ubiquitination AND SUBSEQUENT PROTEOSOMAL DEGRADATION 2USCHMANN ET AL 2010). We have also recently found that IL-4, which skews mIs to an M2 phenotype, triggers the tyrosine PHOSPHORYLATIONANDPROTEASOMALDEGRADATIONOFBOTH3()0AND3()02USCHMANN et al. 2010). It will be interesting to determine whether phosphorylation of the two NPXY motifs is what triggers the poly-ubiquitination or whether distinct tyrosine phosphorylations, which have been reported to occur (Cao et al. 2007), are involved.

144

V.W. Ho et al.

! RECENT EXCITING DISCOVERY IS THAT 3()0 LEVELS CAN BE REGULATED BY MICRO2.!S MI2S #OSTINEANETAL2009; Cremer et al. 2009; O’connell et al. 2009; Pedersen et al. 2009; Yamanaka et al. 2009 -I2SAREAPPROXIMATELYBASE PAIREDUNTRANSLATED2.! strands that interfere with both the transcription and translation of specific genes by binding the 3c UNTRANSLATEDREGION542 OFM2.!S/NEOFTHESEMI2S MI2  which is derived from the BIC gene, has emerged as an important regulator of immune cell development and function as well as oncogenesis and appears to carry out this function, at least in part, by binding and repressing the translation of SHIP M2.!/CONNELLETAL2009; Pedersen et al. 2009 )NTERESTINGLY MI2 TRANSgenic mice develop acute lymphoblastic leukemia (ALL)/high-grade lymphoma (Costinean et al. 2009 ANDTHELEVELSOFMI2 HAVEBEENFOUNDTOBEELEVATEDIN HUMAN.+ CELLLYMPHOMALEUKEMIA9AMANAKAETAL2009) and diffuse large B cell lymphoma (DLBCL) (Pedersen et al. 2009), lending further support to SHIP being a tumour suppressor (Costinean et al. 2009). In DLBCL, autocrine-acting TNFDAPPEARSTOBERESPONSIBLEFORTHEINCREASEDMI2 ANDSUBSEQUENTREDUCED SHIP that leads to the increased proliferation of these abnormal B cells. Interestingly, +OHLHAASETALHAVESHOWNTHATMI2  WHICHISHIGHLYEXPRESSEDIN4REGS ISA DIRECT TARGET OF &OXP AND THAT MI2 −/− mice have reduced numbers of Tregs +OHLHAASETAL2009). This is consistent with studies showing that SHIP represses Treg development (Locke et al. 2009). As an indicator of the complexity involved, ,03 HAS BEEN SHOWN TO INDUCE MI2  EXPRESSION WITHIN MIS 2UGGIERO ET AL 2009) while, at the same time, the activation of Akt by LPS, reduces it (Androulidaki et al. 2009). Lastly, Susheela Tridandapani’s lab has found that a highly virulent strain of Francisella tularensis (F.t.), type A SCHU S4, which grows within phagosomes in monocyte/mIS INDUCESLESSMI2 PRODUCTIONTHANAFARLESSVIRULENT strain, F.t. novicida. This virulent strain, therefore, does not downregulate SHIP as much, resulting in less PIP3-dependent phagosomal–lysosomal fusion and thus less bacterial killing. Interestingly, this may account for the greater virulence of type A F.t. (Cremer et al. 2009).

The Biological Roles of SHIP SHIP knockout studies have revealed that SHIP generally acts as a negative regulator OFTHE0)+PATHWAYANDSO ASONEMIGHTEXPECT ITREDUCESTHESURVIVAL PROLIFERAtion, and end cell activation of many hematopoietic cell types (Antignano et al. 2009). Consistent with this, missense mutations of SHIP have been reported in the blast cells of Chinese patients with AML and acute lymphoblastic leukemia (ALL) (Luo et al. 2003b, 2004), suggesting that SHIP may normally function as a tumour SUPPRESSORINHEMATOPOIETICPROGENITORS!DDITIONALLY ASMENTIONEDEARLIER "#2 ABL reduces SHIP levels and this might be a prerequisite for the proliferative advantage of some CML clones. In terms of human allergies, MacDonald and Vonakis have shown that basophils from a subset of highly allergic people have significantly lower SHIP levels than basophils from normal volunteers (MacDonald

10

SHIP and TAMs

145

and Vonakis 2001). As well, basophils from a subgroup of patients with chronic idiopathic urticaria that do not degranulate upon IgE stimulation have elevated levels of SHIP and SHIP2 (MacDonald and Vonakis 2001). In addition, MacGlashan recently showed that IgE-induced degranulation correlates to some degree with SHIP expression levels in the basophils of 36 different subjects (MacGlashan 2007). These studies are all consistent with SHIP being a negative regulator of IgE-induced mast cell/basophil degranulation. However, there are a growing number of reports showing that SHIP can act under certain circumstances as a positive regulator. For example, several early studies demonstrated a positive role for SHIP in proliferation and survival including studies FROM0AUL2OTHMANSLAB SHOWINGTHATTHEEXPRESSIONOFACATALYTICALLYACTIVE3()0 IN $ MYELOID CELLS EXPRESSING )23  ENHANCED ),  INDUCED PROLIFERATION (Giallourakis et al. 2000 ANDFROM*EFFREY2AVETCHSLAB SHOWINGTHAT3()0ATTENUated FcJ2)) INDUCED" CELLAPOPTOSIS0EARSEETAL1999 -ORERECENTLY IN*URKAT cells, which are immortalized T cells that do not express SHIP, reintroduction of 3()0WASSHOWNTOBOTHPROMOTETHE)++COMPLEX MEDIATEDSERINEPHOSPHORYLATION of INBD and subsequent activation of NFNB and to make the cells more resistant to H2O2-induced apoptosis (Gloire et al. 2006  2ELATED TO THIS +AMEN ET AL HAVE shown that SHIP positively regulates the oxidative burst and phagosome maturation in mIs, perhaps by increasing PI-3,4-P2LEVELS+AMENETAL2008). As well, Charlier et al. elegantly showed that SHIP, in a phosphatase-independent manner, reduces Fas-induced T-cell death by promoting CD95 glycosylation in the endoplasmic reticulum. Specifically, the glycosylated CD95 subsequently fails to oligomerize upon FasL stimulation, protecting the T cells from FasL-induced cell death (Zhang et al. 2009b). As well, Nishio et al. have shown that SHIP is essential for the generation of a leading edge in neutrophil chemotaxis by hydrolyzing PIP3 everywhere else around the cell membrane (Nishio et al. 2007). Moreover, SHIP has been shown to play a positive role in platelet–platelet interactions and thrombus growth in suspension (Gratacap et al. 2008). In addition, recent studies in our lab have shown that SHIP positively regulates DC maturation (Antignano et al. 2010) and bacterial clearance (Bishop et al. 2008) in mice. We have also shown that SHIP, likely as an ADAPTOR PLAYSAPOSITIVEROLEIN4,2LIGAND INDUCEDINmAMMATORYCYTOKINEPRODUCtion from connective tissue mast cells.

The Role of SHIP in M1 Versus M2 Macrophage Skewing Simplistically speaking, mIs can be subdivided into “killer,” classically activated, M1 mIs and “healer,” alternatively activated, M2 mIs (Mills 2001). M1 mIs rapidly upregulate inducible nitric oxide synthase (iNOS, also called NOS2) in response TO DANGER SIGNALS 2AUH ET AL 2005; Sinha et al. 2005), to convert the substrate l-arginine into nitric oxide (NO) to kill bacteria, viruses, and tumour cells (Gordon 2003; Mantovani et al. 2002). M2 mIs, on the other hand possess constitutively high levels of arginase 1, which sequesters l-arginine away from iNOS to generate

146

V.W. Ho et al.

ornithine, and likely plays an important role in phagocytosing cellular debris, stimulating host cell proliferation (via ornithine-derived polyamines) and collagen synthesis (via ornithine-derived proline) after an infectious agent has been destroyed. Given that successful tumours manipulate infiltrating mIs to acquire an M2-like phenotype, which supports tumour growth (Lewis and Pollard 2006), there is considerable interest in trying to understand what regulates M1 versus M2 skewing. Our work suggests that elevated PIP3 levels drive mI progenitors towards M2 difFERENTIATIONANDTHAT3()0BLOCKSTHISSKEWING2AUHETAL2005). Consequently, all of the mIs in SHIP−/− mice possesses an M2 phenotype. Moreover, the fact that mIs in SHIP−/− mice only develop an M2 phenotype after 5 weeks of age suggests that this skewing is not due solely to a cell autonomous deletion of SHIP in mIS2AUH et al. 2005). Subsequent studies into SHIP−/− mI M2 skewing have revealed that skewing is not cell-intrinsic, since SHIP−/− BMmIs do not become M2 when cultured under standard in vitro conditions (i.e., with macrophage colony-stimulating factor, M-CSF, also called CSF-1). Indeed, there are two required components for M2 skewing SHIP−/− mIs during differentiation: a lin+ cellular component in BM and a soluble component(s) contained in mouse plasma. We have recently identified the lin+ cellular component as the CD49b-expressing basophil and basophil progenitor +URODAETAL2009), both of which, when activated by stimuli, such as IL-3 and )G%!G SECRETEHIGHLEVELSOF), +URODAETAL2009), which is the canonical TH2/M2 cytokine which skews both T cells and mIs to their respective type-2 fates (reviewed by Sokol and Medzhitov 2010; Varin and Gordon 2009). Intriguingly, SHIP−/− basophils are much more sensitive to such stimuli, secreting substantially more IL-4 than their similarly stimulated SHIP+/+COUNTERPARTS+URODAETAL2009). &URTHERMORE THE0)+INHIBITORS ,9ANDWORTMANNIN POTENTLYINHIBIT),  SECRETIONFROMBASOPHILS+URODAETAL2009 SUGGESTINGTHAT0)+ISAPOSITIVE regulator of basophil IL-4 production and that SHIP is a negative regulator via its ability to hydrolyse PIP3ANDDAMPEN0)+PATHWAYACTIVITY4HESElNDINGSAGREE with the in vivo phenotypes of the SHIP+/+ and SHIP−/− mice: mIs from SHIP+/+ mice, with normal basophils, are M1, while the mIs from SHIP−/− mice, with hyperactive basophils, are M2. Thus, the increased IL-4 production by SHIP−/− basophils is likely an important part of SHIP−/− M2 skewing. Although necessary for skewing SHIP−/− mIs to M2, basophils alone are not sufficient for M2 skewing, since SHIP−/− mIs derived in M-CSF in the presence OFBASOPHILSARE-+URODAETAL20092AUHETAL2005). However, the in vivo M2 phenotype can be recapitulated by supplementing mI cultures with mouse PLASMA2AUHETAL2005), which contains factors present in the in vivo milieu, including basophil survival/activating factors (e.g., IL-3). Taken together, these data suggest that the addition of plasma to the standard culture conditions that are used to derive mIs from BM (6–10 days with M-CSF) more closely mimics in vivo differentiation conditions in the SHIP−/− mouse and that basophil stimulatory factors may comprise the soluble components required for M2 skewing in vivo (Fig. 10.2).

10

SHIP and TAMs

147

Fig. 10.2 Model of M2 skewing of SHIP−/− macrophages. We propose that basophils, upon stimulation by factors such as IL-3, GM-CSF, HA, PGN, and mouse plasma, skew mIs and T cells to become M2 and TH RESPECTIVELY IN AN ),  DEPENDENT MANNER -OREOVER 0)+ IS A POSITIVE regulator of this skewing, and SHIP, through its dampening of this pathway, is a negative regulator of basophil-mediated M2 and TH2 skewing

It is surprising, however, that IL-3 and GM-CSF, both of which elicit substantial IL-4 secretion from SHIP−/− BASOPHILS +URODA ET AL 2009), are not necessary in mouse plasma-induced M2 skewing, since neutralizing antibodies to these cytokines do not reduce M2 skewing. In addition, IL-3, GM-CSF, and IL-4 are not detectable in appreciable quantities in mouse plasma by cytokine ELISA. Furthermore, TGFE, which is found abundantly in plasma, seems to be necessary but not sufficient for M2 SKEWING2AUHETAL2005). These data suggest that, in vivo, the soluble component(s) of SHIP−/− mIs skewing is a combination of TGFE and one or more additional factors. )NTERESTINGLY WEHAVEEVIDENCETOSUGGESTTHAT4OLL LIKERECEPTOR 4,2 MAY play a role in SHIP−/−-SKEWING!NTIBODYNEUTRALIZATIONOF4,2PARTIALLYBLOCKS mouse plasma-induced M2 skewing in vitro, and hyaluronic acid (HA) and peptidoGLYCAN0'. BOTHOFWHICHCANSIGNALTHROUGH4,2 EACHSKEW3()0−/− mIs to M2. In addition, cyclooxygenase-2 (COX-2), a critical enzyme in the prostaglandin E2 (PGE2 SYNTHESISPATHWAYTHATHASBEENIMPLICATEDIN4,2 INDUCEDSIGNALING

148

V.W. Ho et al.

(Villamon et al. 2005), may play a role in skewing. Inhibition of COX-2, with the selective inhibitor SC58125, dose-dependently reduces the level of plasma M2 skewing in SHIP−/− mIs, and the addition of PGE2 is sufficient to skew SHIP−/− mI phenoTYPETO-4HUS THE4,2 0'%2 axis may be central to SHIP−/− M2 skewing. Not only do SHIP−/− basophils play a role in M2 skewing, but they are also implicated in T-cell TH2 skewing. Activated SHIP−/− basophils, through their aberrantly high level of IL-4 secretion, skew cocultured naïve T cells to a TH2 phenotype, while activated SHIP+/+BASOPHILSDONOT+URODAETAL2011). Interestingly, ex vivo SHIP−/− splenic CD4+ T cells are TH2 skewed; they produce more IL-4 in response to polyclonal activation, and a greater proportion express the TH2 marker T1/ST2, when compared with their SHIP+/+COUNTERPARTS+URODAETALSUBMITTED #URRENT evidence indicates that SHIP is an important regulator of basophil IL-4 production, which, in turn, controls both M2 and TH2 skewing (Fig. 10.2). Acknowledgments 7ETHANK#HRISTINE+ELLYFORTYPINGTHEMANUSCRIPT4HISWORKWASSUPPORTED by the NCI-C, with funds from the Terry Fox Foundation and core support from the BC Cancer Foundation and BC Cancer Agency.

References !LLAM! -ARSHALL!* )MMUNOL,ETTn !MAN-* 7ALK3& -ARCH-%ETAL -OL#ELL"IOLn Androulidaki A, Iliopoulos D, Arranz A et al (2009) Immunity 31:220–231 !NTIGNANO& 2USCHMANN* (AMILTON-ETAL 4HE3RCHOMOLOGYCONTAININGINOSITOLc PHOSPHATASES)N"RADSHAN2! $ENNIS%!EDS (ANDBOOKOFCELLSIGNALLING %LSEVIER)NC San Diego, CA !NTIGNANO& )BARAKI- +IM#ETAL *)MMUNOLn !STLE-6 (ORAN+! /OMS,-ETAL "IOCHEM3OC3YMPn "ARAN#0 4RIDANDAPANI3 (ELGASON#$ETAL *"IOL#HEMn Bennett M, Onnebo SM, Azevedo C et al (2006) Cell Mol Life Sci 63:552–564 "ISHOP*, 3LY,- +RYSTAL'ETAL )NFECT)MMUNn "ONE( 7ELHAM-* #ELL3IGNALn "RAUWEILER! 4AMIR) -ARSCHNER3ETAL *)MMUNOLn "USCHE4 (ORRAS. ,ENFERT%ETAL -OLCELLPROTEOMICS*ULY;%PUB!HEADOF0RINT= #AMPBELL)' 2USSELL3% #HOONG$9ETAL #ANCER2ESn Cantley LC, Neel BG (1999) Proc Natl Acad Sci USA 96:4240–4245 #AO, 9U+ "ANH#ETAL *)MMUNOLn #OGGESHALL+- #URR/PIN)MMUNOLn #OLLAZO-- 7OOD$ 0ARAISO+(ETAL "LOODn #OSTINEAN3 3ANDHU3+ 0EDERSEN)-ETAL "LOODn #OX$ $ALE"- +ASHIWADA-ETAL *%XP-EDn #REMER4* 2AVNEBERG$( #LAY#$ETAL 0,O3/NEE #USCHIERI* "ULGER% 'RINSELL2ETAL 3HOCKn $AMEN*% ,IU, 2OSTEN0ETAL 0ROC.ATL!CAD3CI53!n $AMEN*% ,IU, 7ARE-$ETAL "LOODn $AMEN*% 7ARE-$ +ALESNIKOFF*ETAL "LOODn $OWNES#0 ,ESLIE.2 "ATTY)(ETAL "IOCHEM3OC4RANSn

10

SHIP and TAMs

149

Dybedal I, Bryder D, Fossum A et al (2001) Blood 98:1782–1791 $YSON*- /-ALLEY#* "ECANOVIC*ETAL *#ELL"IOLn &ENG9 7ENTE32 -AJERUS07 0ROC.ATL!CAD3CI53!n &ORESTELL30 "OHNLEIN% 2IGG2* 'ENE4HERn &RIMBERGER!% -C!ULIFFE#) 7ERME+!ETAL "R*(AEMATOLn &RY-* "REAST#ANCER2ESn 'ALANDRINI2 4ASSI) -ORRONE3ETAL %UR*)MMUNOLn 'ALANDRINI2 4ASSI) -ATTIA'ETAL "LOODn 'EIER3* !LGATE0! #ARLBERG+ETAL "LOODn 'EWINNER# 7ANG:# 2ICHARDSON!ETAL #ANCER#ELLn 'HANSAH4 0ARAISO+( (IGHlLL3ETAL *)MMUNOLn 'IALLOURAKIS# +ASHIWADA- 0AN09ETAL *"IOL#HEMn 'IURIATO3 0AYRASTRE" $RAYER!,ETAL *"IOL#HEMn 'LOIRE' #HARLIER% 2AHMOUNI3ETAL /NCOGENEn 'LOIRE' %RNEUX# 0IETTE* "IOCHEM3OC4RANSn 'ORDON3 .AT2EV)MMUNOLn 'RATACAP-0 3EVERIN3 #HICANNE'ETAL !DV%NZYME2EGULn 'UPTA. 3CHARENBERG!- &RUMAN$!ETAL *"IOL#HEMn (ADDON$* !NTIGNANO& (UGHES-2ETAL *)MMUNOLn (ARRIS3* 0ARRY26 7ESTWICK*ETAL *"IOL#HEMn (ELGASON#$ $AMEN*% 2OSTEN0ETAL 'ENES$EVn (ENNESSY"4 3MITH$, 2AM04ETAL .AT2EV$RUG$ISCOVn (ORN3 -EYER* (EUKESHOVEN*ETAL ,EUKEMIAn (UBER- (ELGASON#$ 3CHEID-0ETAL %-"/*n (UBER- (ELGASON#$ $AMEN*%ETAL 4HEROLEOF3()0IN&CH2) INDUCEDSIGNALLING In: Daeron M, Vivier E (eds) Current topics in microbiology and immunology, vol 244. Springer, Berlin )SHIHARA( 3ASAOKA4 (ORI(ETAL "IOCHEM"IOPHYS2ES#OMMUNn *AISWAL"3 *ANAKIRAMAN6 +LJAVIN.-ETAL #ANCER#ELLn *IANG8 3TUIBLE- #HALANDON9ETAL "LOODn +ALESNIKOFF* 3LY,- (UGHES-2ETAL 4HEROLEOF3()0INCYTOKINE INDUCEDSIGNALING )N!MARA3' "AMBERG% "LAUSTEIN-0ETALEDS 2EVIEWSOFPHYSIOLOGY BIOCHEMISTRYAND pharmacology. Springer, Heidelberg +AMEN,! ,EVINSOHN* #ADWALLADER!ETAL *)MMUNOLn +ASHIWADA- #ATTORETTI' -C+EAG,ETAL *)MMUNOLn +ASHIWADA- ,U0 2OTHMAN0" )MMUNOL2ESn +AVANAUGH7- 0OT$! #HIN3-ETAL #URR"IOLn +IM#( (ANGOC' #OOPER3ETAL *#LIN)NVESTn +OHLHAAS3 'ARDEN/! 3CUDAMORE#ETAL *)MMUNOLn +ONCZ' 4OTH'+ "OKONYI'ETAL %UR*"IOCHEMn +OWANETZ+ (USNJAK+ (OLLER$ETAL -OL"IOL#ELLn +RAHN!+ -A+ (OU3ETAL *)MMUNOLn +RAUSE- ,ESLIE*$ 3TEWART-ETAL $EV#ELLn +RYSTAL' 3EMIN)MMUNOLn +URODA% (O6 2USCHMANN*ETAL *)MMUNOLn +URODA% !NTIGNANO& (O67ETAL *)MMUNOLn ,ESLIE.2 "ATTY)( -ACCARIO(ETAL /NCOGENEn ,ESOURNE2 &RIDMAN7( $AERON- *)MMUNOLn ,EUNG7( 4ARASENKO4 "OLLAND3 )MMUNOL2ESn ,EWIS#% 0OLLARD*7 #ANCER2ESn ,IU, $AMEN*% 7ARE-$ETAL *"IOL#HEMn ,IU1 /LIVEIRA $OS 3ANTOS!* -ARIATHASAN3ETALA *%XP-EDn ,IU1 3HALABY& *ONES*ETALB "LOODn ,O4# "ARNHILL,- +IM9ETAL ,EUK2ESn

150

V.W. Ho et al.

,OCKE.2 0ATTERSON3* (AMILTON-*ETAL *)MMUNOLn ,U9 9U1 ,IU*(ETAL *"IOL#HEMn ,UCAS$- 2OHRSCHNEIDER,2 "LOODn ,UO(2 (UANG9% #HEN*#ETALA #ELLn ,UO*- 9OSHIDA( +OMURA3ETALB ,EUKEMIAn ,UO*- ,IU:, (AO(,ETAL :HONGHUA8UE9E8UE:A:HIn -A+ #HEUNG3- -ARSHALL!*ETAL #ELL3IGNALn MacDonald SM, Vonakis BM (2001) Mol Immunol 38:1323–1327 -AC'LASHAN$7*R *!LLERGY#LIN)MMUNOLn -AEDA+ -EHTA( $REVETS$!ETAL "LOODn Mantovani A, Sozzani S, Locati M et al (2002) Trends Immunol 23:549–555 Marchetto S, Fournier E, Beslu N et al (1999) Leukemia 13:1374–1382 -ARESCO$, /SBORNE*- #OONEY$ETAL *)MMUNOLn -ARSHALL!* +RAHN!+ -A+ETAL -OL#ELL"IOLn Martin-Berenjeno I, Vanhaesebroeck B (2009) Cancer Cell 16:449–450 -ARTINO2 #ABALLERO-$ #ANALS#ETAL "R*(AEMATOLn -ASON*- (ALUPA! (YAM$ETAL "LOODA -AXWELL-* 9UAN9 !NDERSON+%ETAL *"IOL#HEMn -ILLS#$ #RIT2EV)MMUNOLn .AKAMURA+ +OURO4 +INCADE07ETAL *%XP-EDn .ISHIO- 7ATANABE+ 3ASAKI*ETAL .AT#ELL"IOLn /CONNELL2- #HAUDHURI!! 2AO$3ETAL 0ROC.ATL!CAD3CI53!n /H39 :HENG4 "AILEY-,ETAL *!LLERGY#LIN)MMUNOLn /NG#* -ING ,UM! .ODWELL-ETAL "LOODn Ono M, Bolland S, Tempst P et al (1996) Nature 383:263–266 /OMS,- (ORAN+! 2AHMAN0ETAL "IOCHEM*n 0EARSE2. +AWABE4 "OLLAND3ETAL )MMUNITYn 0EDERSEN)- /TERO$ +AO%ETAL %-"/-OL-EDn Perez LE, Desponts C, Parquet N et al (2008) PLoS One 3:e3565 0ESESSE8 "ACKERS+ -OREAU#ETAL !DV%NZYME2EGULn 0HEE( *ACOB! #OGGESHALL+- *"IOL#HEMn 0HEE( 2ODGERS7 #OGGESHALL+- -OL#ELL"IOLn 0RASAD.+ 4ANDON- "ADVE3ETALA #ARCINOGENESISn 0RASAD.+ 4ANDON- (ANDA!ETALB 4UMOUR"IOLn 2AMSHAW(3 'UTHRIDGE-! 3TOMSKI&#ETAL "LOODn 2AUH-* (O6 0EREIRA#ETAL )MMUNITYn 2OBSON*$ $AVIDSON$ 6EILLETTE! -OL#ELL"IOLn 2OHRSCHNEIDER,2 &ULLER*& 7OLF)ETAL 'ENES$EVn 2OHRSCHNEIDER,2 #USTODIO*- !NDERSON4!ETAL $EV"IOLn 2UGGIERO4 4RABUCCHI- $E3ANTA&ETAL &!3%"*n 2USCHMANN* (O6 !NTIGNANO&ETAL %XP(EMATOLn 3AIARDI! "HANDARI2 2ESNICK!#ETAL 3CIENCEn 3ALMEEN! "ARFORD$ !NTIOXID2EDOX3IGNALn Sasaoka T, Wada T, Tsuneki H (2006) Pharmacol Ther 112:799–809 3ATTLER- 3ALGIA2 3HRIKHANDE'ETAL /NCOGENEn 3AVAGE.$ DE"OER4 7ALBURG+6ETAL *)MMUNOLn 3CHEID-0 (UBER- $AMEN*%ETAL *"IOL#HEMn 3IEGEL* ,I9 7HYTE0 /NCOGENEn 3INHA0 #LEMENTS6+ -ILLER3ETAL #ANCER)MMUNOL)MMUNOTHERn 3LY,- 2AUH-* +ALESNIKOFF*ETAL %XP(EMATOLn 3LY,- 2AUH-* +ALESNIKOFF*ETAL )MMUNITYn 3OKOL#, -EDZHITOV2 #URR/PIN)MMUNOLn 3ONG- +IM-* (A3ETAL %XP-OL-EDn 3WEENEY-# 7AVREILLE!3 0ARK*ETAL "IOCHEMISTRYn

10

SHIP and TAMs

4AKESHITA3 .AMBA. :HAO**ETAL .AT-EDn 4ANIGAKI+ -INEO# 9UHANNA)3ETAL #IRC2ESn 4ARASENKO4 +OLE(+ #HI!7ETAL 0ROC.ATL!CAD3CI53!n 4OMLINSON-' (EATH6, 4URCK#7ETAL *"IOL#HEMn 4RIDANDAPANI3 0RADHAN- ,A$INE*2ETAL *)MMUNOLn 4RIDANDAPANI3 7ANG9 -ARSH#"ETAL *)MMUNOLn 4ROTTA2 0ARIHAR2 9U*ETAL "LOODn 4U: .INOS*- -A:ETAL "LOODn Valderrama-Carvajal H, Cocolakis E, Lacerte A et al (2002) Nat Cell Biol 4:963–969 6ANDERWINDEN*- 7ANG$ 0ATERNOTTE.ETAL #ELL3IGNALn Varin A, Gordon S (2009) Immunobiology 214:630–641 Velazquez L, Gish GD, van der Geer P et al (2000) Blood 96:132–138 6ILLAMON% 2OIE0 'IL-,ETAL 2ES-ICROBIOLn 7ANG*7 (OWSON*- 'HANSAH4ETAL 3CIENCEn Wavreille AS, Garaud M, Zhang Y et al (2007) Methods 42:207–219 Wisniewski D, Strife A, Swendeman S et al (1999) Blood 93:2707–2720 Wolf I, Lucas DM, Algate PA et al (2000) Genomics 69:104–112 Yamanaka Y, Tagawa H, Takahashi N et al (2009) Blood 114:3265–3275 9OGO+ -IZUTAMARI- -ISHIMA+ETAL %NDOCRINOLOGYn :HANG* 7ALK3& 2AVICHANDRAN+3ETALA *"IOL#HEMn :HANG44 ,I( #HEUNG3-ETALB )MMUNOL2EVn :HANG9 7AVREILLE!3 +UNYS!2ETALC "IOCHEMISTRYn :IPPO! $E2OBERTIS! "ARDELLI-ETAL "LOODn

151

Chapter 11

NF-KappaB-Mediated Regulation of Tumour-Associated Macrophages: Mechanisms and Significance Antonio Sica and Alberto Mantovani

Introduction Recent efforts have shed new light on molecular and cellular pathways linking inflammation and cancer (Balkwill and Mantovani 2001; Mantovani et al. 2008a; Coussens and Werb 2002; Mantovani 2009). Two pathways link inflammation and cancer. In the intrinsic pathway, activation of different classes of oncogenes drives the expression of inflammation-related programs which guide the construction of an inflammatory microenvironment. In the extrinsic pathway, inflammatory conditions promote cancer development (e.g., colitis-associated cancer of the intestine). Key orchestrators at the intersection of the intrinsic and extrinsic pathways include transcription factors (e.g., NFNB, Stat3, and HIF) (Rius et al. 2008; Karin 2006; Yu et al. 2007), cytokines (e.g., TNF), and chemokines (Fig. 11.1). Thus, inflammation is a key component of the tumour microenvironment and a target for pharmacologic intervention. Tumour-associated macrophages (TAM) are a major component of leukocytic infiltrate of tumours and have served as a paradigm for cancer-related inflammation (Mantovani et al. 2008a; Pollard 2009; De Palma et al. 2007). Macrophages are a double-edged sword, with the potential to express pro- and anti-tumour activity [the macrophage balance, (Mantovani et al. 1992, 2002)], the former prevailing in established neoplasia. Macrophages and some of their products (IL-1, TNF, and IL-6) have long been known to increase metastasis (Mantovani et al. 2008a) and recent work has provided new evidence (Hagemann et al. 2005, 2006; Kim et al. 2009; A. Sica (*) Istituto Clinico Humanitas IRCCS, via Manzoni 56, 20089 Rozzano, Italy DiSCAFF, University of Piemonte Orientale A. Avogadro, 28100 Novara, Italy e-mail: [email protected] A. Mantovani Istituto Clinico Humanitas IRCCS, via Manzoni 56, 20089 Rozzano, Italy Department of Translational Medicine, University of Milan, Milan, Italy T. Lawrence and T. Hagemann (eds.), Tumour-Associated Macrophages, DOI 10.1007/978-1-4614-0662-4_11, © Springer Science+Business Media, LLC 2012

153

154

A. Sica and A. Mantovani Intrinsic pathways

Extrinsic pathways Chronic inflammation

Inflammatorycells Inflammatorycells Mc

Mø Eo

Mø/MDSC

PMN

Carcinogenesis

Oncogenetic events ras / raf Nuclear Tyrosine WnT/Bcatenin; Myc Kinases Oncosuppressor (RET/PTC; EGF-R) (TGFB; VHL/HIF; RelB PTEN;Acatenin)

Infections Autoimmune (Virus, HBV, Autoinflammatory HCV,HPV; (IBD) Bacteria, H. pylori, Irritants (COPD) Protozoa, schistosoma) Protozoa S. Mansony)

Oxidative stress, DNA damage NF-kB, STAT3, HIFs (inflammation, cell proliferation, antiapoptosis, invasion)

4

Fig. 11.1 Inflammation and cancer connection. Irrespective of the trigger for the development both intrinsic (driven by genetic alteration) and extrinsic (driven by inflammatory cells and mediators) pathways result in inflammation and neoplasia. Both neoplastic cells and leukocytes, mainly belonging to the myelomonocytes lineage, contribute to the “smoldering” inflammation associated with tumour initiation and progression. The transcription factors NF-NB, HIF-1D, and STAT-3 are key modulators of the inflammatory response that promotes cancer development through different mechanisms including the induction of genomic instability, alteration in epigenetic events and subsequent inappropriate gene expression, enhanced proliferation and resistance to apoptosis of initiated cells, induction of tumour angiogenesis and tissue remodeling with consequent promotion of tumour cells invasion, and metastasis (Mø macrophages, Mc mast cells, MDSC myeloid-derived suppressor cells, Eo eosinophil, PMN polymorphonuclear cells, T T lymphocytes)

Kulbe et al. 2007; Wyckoff et al. 2007). In particular, cells of hematopoietic origin and specifically myelomonocytic cells have been shown to home and condition the premetastatic niche, to favor secondary localization of cancer (Kaplan et al. 2005; Erler et al. 2009; Wels et al. 2008). Here we will focus on selected molecular pathways underlying TAM recruitment and polarization, emphasizing the dual potential of cancer-related inflammation (Mantovani et al. 2008b) and the diversity of pathways and functions in tumour development.

Influence of Extracellular Signals on TAM Function and Polarization Plasticity is a hallmark of mononuclear phagocytes. In response to diverse signals, macrophages undergo polarized activation. Classically activated (M1) macrophages, following exposure to interferon J (IFNJ), have tumouricidal activity and elicit tissue destructive reactions. In response to IL-4 or IL-13, macrophages undergo alternative (M2) activation (Mantovani et al. 2002; Martinez et al. 2009). In general, M2 cells obtained in response to IL-4 are oriented to tissue repair and remodeling, immunoregulation, and tumour promotion (Mantovani et al. 2002). Interestingly,

11

NF-KappaB-Mediated Regulation of Tumour-Associated Macrophages…

155

recent evidence indicates that polymorphonuclear leukocytes can also polarize in a type I or type II direction in cancer (Fridlender et al. 2009). In most, but not all (Torroella-Kouri et al. 2009) tumours investigated, TAM have an M2-like phenotype. Tumour cell products, including extracellular matrix components, IL-10, CSF1, activate macrophages and set them in an M2, cancer promoting mode (Hagemann et al. 2006; Kim et al. 2009; Erler et al. 2009; Kuang et al. 2007). Chemokines (CCL17 and CCL22) that interact with CCR4 have been shown to promote M2 polarization (Wallace et al. 2009). CCL2 has been shown to play a role not only in attracting tumour-promoting macrophages in carcinoma of the prostate but also in promoting their survival and M2 polarization (Roca et al. 2009). Anti-CCL2 antibodies are undergoing evaluation in humans. Tumour products can instruct dendritic cells to prime IL-13 producing T cells which promote breast cancer (Aspord et al. 2007). New evidence indicates that adaptive immune responses can shape the infiltration and function of TAM. In a primary mammary carcinoma model, CD4+ T cells promote metastasis by causing M2 activation via IL-4 (DeNardo et al. 2009). Interestingly, by pursuing previous studies on multi-stage epithelial carcinogenesis in the skin driven by transgenic expression of HPV16, the same group identified a different pathway of orchestration of inflammation-mediated tumour promotion by adaptive immune responses. Here, T cell-dependent humoral antibody production against extracellular matrix components orchestrates by remote control cancer-related inflammation (de Visser et al. 2005). Autoantibodies act via FcJR and regulate recruitment and M2-like polarization of myelomonocytic cells which promote tumour progression (Andreu et al. 2010).

Molecular Determinants of TAM Functions: Role of NF-kappaB As mentioned above, TAM generally have phenotype and functions similar to alternative or M2 macrophages (Mantovani et al. 2008a; Biswas et al. 2006). For example, TAM express low levels of the major histocompatibility complex class II and reduced antimicrobial and tumouricidal activity, while increasing the production of mediators that promote angiogenesis, such as vascular endothelial growth factor and cyclooxygenase-2 (COX-2)-derived prostaglandin E2, as well as the antiinflammatory cytokine IL-10 (Mantovani et al. 2008a). Another hallmark feature of alternative activation expressed by TAM is the low expression of IL-12 and upregulated levels of M2-specific genes, such as arginase-1 (Arg-1), macrophage galactosetype C-type lectin-2 (Mgl2), Fizz1, and Ym1 (Biswas et al. 2006; Ghassabeh et al. 2006). Recent addition to the molecular repertoire of TAM includes Sema4D and Gas6, which are, respectively, involved in promoting tumour angiogenesis (Sierra et al. 2008) and cancer cell proliferation (Loges et al. 2010). Analysis of the molecular basis of the TAM phenotype has identified the transcriptional factors NF-NB and HIF-1 as master regulators of their transcriptional programs and indicates these factors as central regulators of tumour progression and metastasis (Sica and Bronte 2007).

156

A. Sica and A. Mantovani

NF-NB induces several cellular alterations associated with tumourigenesis and more aggressive phenotypes, including: self-sufficiency in growth signals, insensitivity to growth inhibition, resistance to apoptotic signals, immortalization, angiogenesis, tissue invasion, and metastasis (Karin 2006). Constitutive NF-NB activation often observed in cancer cells may be either promoted by genetic alterations or by microenvironmental signals, including cytokines, hypoxia, and reactive oxygen intermediates (ROI) (Karin 2006). In particular, proinflammatory cytokines (e.g., IL-1 and TNF), expressed by infiltrating leukocytes, can activate NF-NB in cancer cells and contribute to their proliferation and survival (Karin 2006). In addition to target cancer cell functions, NF-NB has been described as a key regulator of inflammation and resolution, whose activation is subject to multiple levels of regulation including inhibitory (Bonizzi and Karin 2004). New recent evidence indicate that key components of the NF-NB system, including signaling transduction molecules and NF-NB subunits play a central role in the modulation of macrophage functions, as well as TAM activities (Hagemann et al. 2008; Porta et al. 2009). Members of the NF-NB/Rel family regulate many genes involved in immunity and inflammation (Bonizzi and Karin 2004). Two major signaling pathways control the activation of the NF-NB (Bonizzi and Karin 2004). The classical pathway is stimulated by proinflammatory cytokines, such as tumour necrosis factor-D (TNF-D) and interleukin-1 (IL-1), as well as by the recognition of pathogen-associated molecular patterns (PAMPs), and is mostly involved in innate immunity (Bonizzi and Karin 2004). In addition, an alternative pathway of NF-NB activation, mainly involved in adaptive immunity, is activated by certain members of the TNF cytokine family, but not by TNF-D itself (Bonizzi and Karin 2004). The NF-NB family consists of five members: NF-NB1 (p105/p50), NF-NB2 (p100/p52), RelA (p65), RelB, and c-Rel (Bonizzi and Karin 2004; Verma 2004), which may form different homo- and heterodimers associated with differential regulation of target genes (Verma 2004). For example, the p50 and p52 homodimers act as repressors, as these proteins lack a transcription activation domain, present in RelA, RelB, v-Rel, and c-Rel (Bonizzi and Karin 2004). Accumulation of p50 homodimers has been observed in endotoxin tolerant macrophages (Ziegler-Heitbrock et al. 1997) as well as in TAM (Biswas et al. 2006; Saccani et al. 2006), suggesting an important role in the control and extinction of the inflammatory response (Porta et al. 2009; Ziegler-Heitbrock et al. 1997). Other negative regulators of NF-NB activation have been identified in disease and include the LPS-inducible splice variant of myeloid differentiation 88 (MyD88), termed MyD88s, the single immunoglobulin IL-1 receptor-related molecule (SIGIRR)/ TIR8, ST2, IRAK-M, SOCS1 (Liew et al. 2005; O’Neill and Bowie 2007), and the Src homology 2-containing inositol-5c-phosphatase, SHIP (Kuroda et al. 2009). Exposure to microbial components such as bacterial lipopolysaccharide (LPS) has long been known to induce tolerance to the same agonist in terms of in vivo toxicity and macrophage production of inflammatory mediators, such as TNF (Ziegler-Heitbrock et al. 1997). This phenomenon has been sometimes referred to as immunoparalysis (Cavaillon et al. 2005), which might play a role in sepsis mortality, and it was recently pointed out that in septic patients a reprogramming of the macrophage, rather than an “immunoparalysis,” might occur (Cavaillon et al. 2005; Cavaillon and Adib-Conquy 2006).

11

NF-KappaB-Mediated Regulation of Tumour-Associated Macrophages…

157

Carcinogen induced Genetic alterations (eg. P53 mutation)

Switch in Macrophage phenotypes

PP .& K"

CD4 NKT CD4

TNF IL-23 RNI IL-1 IL-12high ROI IL-6 IL-10low

M1

CD8

NF-KB

1) Initiation/promotion CD4

2) Elimination CD8

CD8

NK

NK

M2

NK CD8

CD4

PP .& K"

Polyamine IL-12low MCSF IL-10high MMPs TGFb VEGF Tumour growth/survival Metastasis Angiogenesis Immunosuppression

4) Escape Normal tissue

3) Equilibrium initiated tumour cells

Lysed tumour cell

tumouricidal function

Macrophage

blood vessels

tumour cell variants

carcinogenic signal

Fig. 11.2 It is proposed that a gradual switching of TAM polarization, from M1 to M2, is paralleled by the gradual inhibition of NF-NB, during different stages of tumour progression (Dunn et al. 2002). Early in carcinogenesis T cell driven M1-activated macrophages may contribute to elimination (Smyth et al. 2006). Ensuing regulatory mechanisms of results in M2-polarized TAM which orchestrate smoldering, non-resolving tumour promoting inflammation

To the extent they have been investigated, TAM display a defective NF-NB activation in response to different pro-inflammatory signals (Allavena et al. 2008), suggesting their tolerant phenotype. Inhibition of NF-NB activation in TAM correlates with impaired expression of NF-NB-dependent inflammatory functions [e.g., expression of cytotoxic mediators, such as NO and cytokines (TNFD, IL-1, and IL-12) (Mantovani et al. 2008a; Sica et al. 2008)]. Hence, there is an apparent contrast with inflammatory functions expressed by TAM during early steps of carcinogenesis. Possible explanations for this discrepancy are based on the functional feature of macrophages, which are considered highly plastic cells able to finely modulate their programs in response to different microenvironmental conditions (Sica et al. 2008). It has been speculated that dynamic changes of the tumour microenvironment may occur during the transition from early neoplastic events toward advanced tumour stages (Sica and Bronte 2007). These events would drive an M1 towards M2 switch of TAM functions (Fig. 11.2) (Sica and Bronte 2007) and are likely connected to the profound changes occurring in the tumour microphysiol-

158

A. Sica and A. Mantovani

ogy (e.g., hypoxia, glucose levels, pH) (Vaupel 2008). Thus, while full activation of NF-NB in leukocytes would favor M1 inflammation and tumourigenesis, tumour growth and progression may drive the inhibition of NF-NB in infiltrating leukocytes, as reported in both myeloid and lymphoid cells associated with advanced tumour stages (Sica and Bronte 2007). Interestingly, TAM from p50 NF-NB−/− tumour-bearing mice express cytokines characteristic of M1 macrophages (e.g., IL-12high/IL-10low) and their splenocytes produce increased levels of Th1 cytokines (e.g., IFN-J), which are associated with a delay in tumour growth (Saccani et al. 2006). By searching for the microenvironmental signals promoting accumulation of the p50 homodimer in macrophages, we demonstrated that IL-10, PGE2, and TGF-E, which are expressed by TAMs and promote M2-type polarized inflammation (Mantovani et al. 2004), promote p50 NF-NB homodimer activity (Saccani et al. 2006). A detailed analysis of the role of p50 NF-NB homodimer in macrophage functions revealed that its nuclear accumulation, both in TAM and LPS-tolerant macrophages, not only mediates a status of unresponsiveness (tolerance) towards pro-inflammatory signals, but actually plays as key regulator of M2-driven inflammatory reactions, acting through inhibition of NF-NB-driven M1-polarizing IFNE production and STAT1 phosphorylation (Porta et al. 2009) (Fig. 11.3). A recent study by Hagemann et al. demonstrated the requirement for IKKE to maintain the IL-10high/IL-12low phenotype of TAM in a mouse model of ovarian cancer (Hagemann et al. 2008). These authors showed that targeted deletion or inhibition of IKKE in TAM increased their tumouricidal activity through elevated NOS2 expression and IL-12-dependent recruitment and activation of NK cells. Because increased expression of IL-12, NOS2, and enhanced tumouricidal activity are associated with M1 characteristics, these data imply that inhibition of the IKKE/ NF-NB pathway promotes an M1-like phenotype in TAM. Additional efforts are required to clarify the role of NF-NB in TAM polarization, but both studies indicate that the inhibition of either p50 NF-NB or IKKE result in the restoration of STAT1 activity, which appears a convergent and necessary function to promote M1 macrophage polarization (Hagemann et al. 2008; Porta et al. 2009). A dynamic view of this scenario would also suggest that an early IKKE-dependent NF-NB activation may trigger cancer-related inflammation and the subsequent activation of the p50dependent regulatory pathway may tune and promote M2-associated smoldering inflammation. A microenvironmental condition that appears to impact on NF-NB signaling in TAM is hypoxia (low oxygen tension). The presence of many areas of hypoxia is a hallmark feature of most forms of solid tumour (Vaupel 2008) and TAM have been shown to accumulate in these areas where hypoxia promotes their pro-tumour phenotype (Murdoch et al. 2008). Hypoxia-inducible factor 1 (HIF-1) has been shown to control the cellular response to hypoxia. Hypoxia stabilizes HIF-1D, preventing post-translational hydroxylation and subsequent degradation via the proteasome. More recently, short-term exposure of murine bone marrow-derived macrophages to hypoxia has been shown to upregulate NF-NB activity, which in turn upregulates

11

NF-KappaB-Mediated Regulation of Tumour-Associated Macrophages…

159

LBP TLR4 LPS MD-2 CD14

TIRAP

MyD88

TRIF

TRAF6 IKK IFN-B IRF-3

P-STAT1

LP

S-t

d an ce M ran TA tole S-

LP

ole

ran

ce

PP .& K"

IFN-BR

P.& K" P.& K" P.& K"

Fig. 11.3 Different mechanisms of p50 NF-NB-mediated M2 polarization in TAM versus LPS-tolerant macrophages. In LPS tolerance, p50 NF-NB acts by suppressing NF-NB driven, M1 polarizing, IFND/E production, thus preventing STAT1 phosphorylation. In contrast, TAM display only impaired NF-NB activation, but functional STAT1 activity

HIF-1D levels (Rius et al. 2008). This study used macrophages from IKKE−/− mice to show that NF-NB is a critical transcriptional activator of HIF-1D and that basal NF-NB activity is required for HIF-1D protein accumulation under hypoxia. Overall, this evidence indicate a considerable plasticity in NF-NB in TAM and suggest that the modulation of its activity in these cells maintains their immunosuppressive, tumour-promoting phenotype. Further studies addressing the relative contribution of individual NF-NB members (p65, c-Rel, p50, and BCL3) and their combinatorial transcriptional partners, such as STATs and IRF3 (Kawai and Akira 2007), will likely contribute to fully clarify its role in cancer-related inflammation.

160

A. Sica and A. Mantovani

Tuning the Balance Certain forms of inflammation are protective in a preventive or therapeutic setting (Mantovani et al. 2008a; Kryczek et al. 2009; Nickoloff et al. 2005). An immune response has long been known to contribute to the outcome of chemotherapy. It has now been shown that dying tumour cells can be cross presented by dendritic cells and trigger a protective immune response via a TLR4–MyD88 pathway (Apetoh et al. 2007). Inflammasome activation and IL-1E production underlie ultimate the activation of protective immunity (Ghiringhelli et al. 2009). Association of TLR4 and PrX7 polymorphisms with clinical outcome are consistent with the relevance of these pathways. Thus, it appears that a main therapeutic challenge will be to promote inflammatory programs with anticancer properties. Within this perspective, direct activation of selected forms of innate immunity cells (macrophage) polarization may represent a new alternative or complementary antitumour strategy (Duluc et al. 2009). As an example, the M1-polarizing cytokine IFNJ has been shown to reeducate TAM (Duluc et al. 2009) and there is proof of principle evidence for antitumour activity of this molecule in minimal residual disease in humans (Colombo et al. 1992; Windbichler et al. 2000; Berek et al. 2003). Further, given the protumour function of TAM in many cancer, strategies have been directed at blocking recruitment (see above) targeting chemokines or CSF-1 or inhibiting TAM directly (Ritchie and Smyth 2009; Song et al. 2009; Luo et al. 2006). Noteworthy, CSF-1 is also known to promote strong M2 polarization of macrophages (Pollard 2009). As more evidence emerges for the signaling pathways involved in the “switch” of macrophage polarization states (i.e., M1–M2) in the early stages of tumour progression (Sica and Bronte 2007), it may be possible to develop new therapies aimed at preventing this and/or reorientating M2 TAM in favor of a more antitumoural phenotype. In this regard, several lines of evidence support the idea that pharmacological skewing of TAM polarization, from M2 to a full M1 phenotype, may sustain an antitumour activity. Indeed, combination of CpG plus an anti-IL-10 receptor antibody switched infiltrating macrophages from M2 to M1 and triggered innate response debulking large tumours within 16 h (Guiducci et al. 2005; Colombo and Mantovani 2005). Interestingly, both CpG and anti-IL-10 treatments lead to the restoration of NF-NB activity (Kawai and Akira 2007; Sica et al. 2000). Moreover, TAMs lacking STAT6−/−, the major mediator of IL-4 and IL-13 biological functions, display an M1 phenotype, with low level of ARG and high level of NO (OstrandRosenberg et al. 2000). As a result, these mice rejected spontaneous mammary carcinoma by a process requiring adaptive immunity to cancer (Ostrand-Rosenberg et al. 2000). In analogy, the inhibition of STAT3 activity, which is required for biological functions and gene transcription of the M2 cytokine IL-10, resulted in both restored expression of pro-inflammatory mediators, including IL-12 and TNF-D, by infiltrating leukocytes and tumour inhibition (Kortylewski et al. 2005).

11

NF-KappaB-Mediated Regulation of Tumour-Associated Macrophages…

161

All these findings concur to suggest that M2-inflammation may fuel cancer progression and propose a dynamic model of gradual NF-NB inhibition associated with the orientation of M2-polarized functions in TAM. It is likely that whereas the full activation of NF-NB in inflammatory leukocytes resident in preneoplastic sites may exacerbate local M1-inflammation (TNF-Dhigh, IL-1high, IL-12high, IL-10low, and TGFElow) and favor tumourigenesis (Pikarsky et al. 2004; Greten et al. 2004; Moore et al. 1999), tumour growth may result in progressive inhibition of NF-NB in infiltrating leukocytes, as observed in both myeloid (Biswas et al. 2006; Saccani et al. 2006) and lymphoid (Ghosh et al. 1994) cells from tumour bearers and in the progressive development of M2-inflammation (TNF-Dlow, IL-1low, IL-12low, IL-10high, and TGFEhigh). If so, therapeutic efficacy of anti-NF-NB strategies against cancers may be subject to both tumour stage and polarization status of infiltrating leukocytes. In this regard, along with proinflammatory cytokines (e.g., TNF-D and IL-1), endogenous ligands of TLRs present into the tumour microenvironment are fibrinogen, HSP, and HMGB proteins (Chen et al. 2007).

Concluding Remarks Recent results have highlighted a striking similarity in terms of transcriptional profile, functional properties, and underlying transcription factors between TAM and “tolerant” macrophage (Porta et al. 2009). Tolerant macrophages belong to the wide spectrum and universe of M2-like polarized phagocytes. The finding that PMN have unsuspected plasticity and are subject to a P1–P2 balance in the tumour microenvironment (Fridlender et al. 2009) calls for a reappraisal of their role and significance in cancer-related inflammation. Inflammatory reactions, and macrophages in particular, can exert a dual influence on tumour growth and progression (Mantovani et al. 1992, 2008b). In this balance, progression through the three “E” (elimination, equilibrium, and escape) is likely mirrored by shift in this ying yang balance (Smyth et al. 2006). A functional reprogramming of TAM functions appears to be linked to the dynamic changes of the tumour microenvironment during the transition from early neoplastic events to advanced tumour stages, which would result in progressive modulation of their NF-NB activity and progressive conversion of the TAMs from an M1 to M2 macrophage phenotype (Fig. 11.2). Certain chemotherapeutic agents or inflammatory conditions can likewise activate a protective macrophage response. Diversity is a hallmark of cancer-related inflammation. Recent results have highlighted how adaptive immunity can orchestrate cancer-promoting inflammation and TAM through distinct molecular pathways (DeNardo et al. 2009; Andreu et al. 2010). Thus, the development of effective strategies to tip the balance by blocking cancer-promoting inflammation and/or activating protective innate immunity may require definition of actors at play at different anatomical sites and in different tumours.

162

A. Sica and A. Mantovani

References Allavena P, Garlanda C, Borrello MG, Sica A, Mantovani A (2008) Pathways connecting inflammation and cancer. Curr Opin Genet Dev 18:3–10 Andreu P, Johansson M, Affara NI, Pucci F, Tan T, Junankar S, Korets L, Lam J, Tawfik D, DeNardo DG et al (2010) FcRgamma activation regulates inflammation-associated squamous carcinogenesis. Cancer Cell 17(2):121–134 Apetoh L, Ghiringhelli F, Tesniere A, Obeid M, Ortiz C, Criollo A, Mignot G, Maiuri MC, Ullrich E, Saulnier P et al (2007) Toll-like receptor 4-dependent contribution of the immune system to anticancer chemotherapy and radiotherapy. Nat Med 13:1050–1059 Aspord C, Pedroza-Gonzalez A, Gallegos M, Tindle S, Burton EC, Su D, Marches F, Banchereau J, Palucka AK (2007) Breast cancer instructs dendritic cells to prime interleukin 13-secreting CD4+ T cells that facilitate tumor development. J Exp Med 204:1037–1047 Balkwill F, Mantovani A (2001) Inflammation and cancer: back to Virchow? Lancet 357:539–545 Berek JS, Schultes BC, Nicodemus CF (2003) Biologic and immunologic therapies for ovarian cancer. J Clin Oncol 21:168s–174s Biswas SK, Gangi L, Paul S, Schioppa T, Saccani A, Sironi M, Bottazzi B, Doni A, Bronte V, Pasqualini F et al (2006) A distinct and unique transcriptional programme expressed by tumorassociated macrophages: defective NF-kB and enhanced IRF-3/STAT1 activation. Blood 107:2112–2122 Bonizzi G, Karin M (2004) The two NF-kappaB activation pathways and their role in innate and adaptive immunity. Trends Immunol 25:280–288 Cavaillon JM, Adib-Conquy M (2006) Bench-to-bedside review: endotoxin tolerance as a model of leukocyte reprogramming in sepsis. Crit Care 10:233 Cavaillon JM, Adrie C, Fitting C, Adib-Conquy M (2005) Reprogramming of circulatory cells in sepsis and SIRS. J Endotoxin Res 11:311–320 Chen K, Huang J, Gong W, Iribarren P, Dunlop NM, Wang JM (2007) Toll-like receptors in inflammation, infection and cancer. Int Immunopharmacol 7:1271–1285 Colombo MP, Mantovani A (2005) Targeting myelomonocytic cells to revert inflammation-dependent cancer promotion. Cancer Res 65:9113–9116 Colombo N, Peccatori F, Paganin C, Bini S, Brandely M, Mangioni C, Mantovani A, Allavena P (1992) Anti-tumor and immunomodulatory activity of intraperitoneal IFN-gamma in ovarian carcinoma patients with minimal residual tumor after chemotherapy. Int J Cancer 51:42–46 Coussens LM, Werb Z (2002) Inflammation and cancer. Nature 420:860–867 De Palma M, Murdoch C, Venneri MA, Naldini L, Lewis CE (2007) Tie2-expressing monocytes: regulation of tumor angiogenesis and therapeutic implications. Trends Immunol 28:519–524 de Visser KE, Korets LV, Coussens LM (2005) De novo carcinogenesis promoted by chronic inflammation is B lymphocyte dependent. Cancer Cell 7:411–423 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 16:91–102 Duluc D, Corvaisier M, Blanchard S, Catala L, Descamps P, Gamelin E, Ponsoda S, Delneste Y, Hebbar M, Jeannin P (2009) Interferon-gamma reverses the immunosuppressive properties and prevents the generation of human tumor-associated macrophages. Int J Cancer 125(2): 367–373 Dunn GP, Bruce AT, Ikeda H, Old LJ, Schreiber RD (2002) Cancer immunoediting: from immunosurveillance to tumor escape. Nat Immunol 3:991–998 Erler JT, Bennewith KL, Cox TR, Lang G, Bird D, Koong A, Le QT, Giaccia AJ (2009) Hypoxiainduced lysyl oxidase is a critical mediator of bone marrow cell recruitment to form the premetastatic niche. Cancer Cell 15:35–44 Fridlender ZG, Sun J, Kim S, Kapoor V, Cheng G, Ling L, Worthen GS, Alberlda SM (2009) Polarization of tumor-associated neutrophil (TAN) phenotype by TGF-beta: “N1” versus “N2” TAN – a new paradigm? Cancer Cell 16:183–194

11

NF-KappaB-Mediated Regulation of Tumour-Associated Macrophages…

163

Ghassabeh GH, De Baetselier P, Brys L, Noel W, Van Ginderachter JA, Meerschaut S, Beschin A, Brombacher F, Raes G (2006) Identification of a common gene signature for type II cytokineassociated myeloid cells elicited in vivo in different pathologic conditions. Blood 108:575–583 Ghiringhelli F, Apetoh L, Tesniere A, Aymeric L, Ma Y, Ortiz C, Vermaelen K, Panaretakis T, Mignot G, Ullrich E et al (2009) Activation of the NLRP3 inflammasome in dendritic cells induces IL-1beta-dependent adaptive immunity against tumors. Nat Med 15:1170–1178 Ghosh P, Sica A, Young HA, Ye J, Franco JL, Wiltrout RH, Longo DL, Rice NR, Komschlies KL (1994) Alterations in NF kappa B/Rel family proteins in splenic T-cells from tumor-bearing mice and reversal following therapy. Cancer Res 54:2969–2972 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 118:285–296 Guiducci C, Vicari AP, Sangaletti S, Trinchieri G, Colombo MP (2005) Redirecting in vivo elicited tumor infiltrating macrophages and dendritic cells towards tumor rejection. Cancer Res 65: 3437–3446 Hagemann T, Wilson J, Kulbe H, Li NF, Leinster DA, Charles K, Klemm F, Pukrop T, Binder C, Balkwill FR (2005) Macrophages induce invasiveness of epithelial cancer cells via NF-kappa B and JNK. J Immunol 175:1197–1205 Hagemann T, Wilson J, Burke F, Kulbe H, Li NF, Pluddemann A, Charles K, Gordon S, Balkwill FR (2006) Ovarian cancer cells polarize macrophages toward a tumor-associated phenotype. J Immunol 176:5023–5032 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-kappaB. J Exp Med 205:1261–1268 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 Karin M (2006) Nuclear factor-kappaB in cancer development and progression. Nature 441: 431–436 Kawai T, Akira S (2007) Signaling to NF-kappaB by Toll-like receptors. Trends Mol Med 13: 460–469 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 via TLR2 to stimulate metastasis. Nature 457:102–106 Kortylewski M, Kujawski M, Wang T, Wei S, Zhang S, Pilon-Thomas S, Niu G, Kay H, Mule J, Kerr WG et al (2005) Inhibiting Stat3 signaling in the hematopoietic system elicits multicomponent antitumor immunity. Nat Med 11:1314–1321 Kryczek I, Banerjee M, Cheng P, Vatan L, Szeliga W, Wei S, Huang E, Finlayson E, Simeone D, Welling TH et al (2009) Phenotype, distribution, generation, and functional and clinical relevance of Th17 cells in the human tumor environments. Blood 114:1141–1149 Kuang DM, Wu Y, Chen N, Cheng J, Zhuang SM, Zheng L (2007) Tumor-derived hyaluronan induces formation of immunosuppressive macrophages through transient early activation of monocytes. Blood 110:587–595 Kulbe H, Thompson R, Wilson JL, Robinson S, Hagemann T, Fatah R, Gould D, Ayhan A, Balkwill F (2007) The inflammatory cytokine tumor necrosis factor-alpha generates an autocrine tumorpromoting network in epithelial ovarian cancer cells. Cancer Res 67:585–592 Kuroda E, Ho V, Ruschmann J, Antignano F, Hamilton M, Rauh MJ, Antov A, Flavell RA, Sly LM, Krystal G (2009) SHIP represses the generation of IL-3-induced M2 macrophages by inhibiting IL-4 production from basophils. J Immunol 183:3652–3660 Liew FY, Xu D, Brint EK, O’Neill LA (2005) Negative regulation of toll-like receptor-mediated immune responses. Nat Rev Immunol 5:446–458 Loges S, Schmidt T, Tjwa M, Van Geyte K, Lievens D, Lutgens E, Vanhoute D, Borgel D, Plaisance S, Hoylaerts M et al (2010) Malignant cells fuel tumor growth by educating infiltrating leukocytes to produce the mitogen Gas6. Blood 115(11):2264–2273

164

A. Sica and A. Mantovani

Luo Y, Zhou H, Krueger J, Kaplan C, Lee SH, Dolman C, Markowitz D, Wu W, Liu C, Reisfeld RA et al (2006) Targeting tumor-associated macrophages as a novel strategy against breast cancer. J Clin Invest 116:2132–2141 Mantovani A (2009) Cancer: inflaming metastasis. Nature 457:36–37 Mantovani A, Bottazzi B, Colotta F, Sozzani S, Ruco L (1992) The origin and function of tumorassociated macrophages. Immunol Today 13:265–270 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 Mantovani A, Sica A, Sozzani S, Allavena P, Vecchi A, Locati M (2004) The chemokine system in diverse forms of macrophage activation and polarization. Trends Immunol 25:677–686 Mantovani A, Allavena P, Sica A, Balkwill F (2008a) Cancer-related inflammation. Nature 454: 436–444 Mantovani A, Romero P, Paluka A, Marincola F (2008b) Tumor immunity: effector response to tumor and the influence of the microenvironment. Lancet 371:771–783 Martinez FO, Helming L, Gordon S (2009) Alternative activation of macrophages: an immunologic functional perspective. Annu Rev Immunol 27:451–483 Moore RJ, Owens DM, Stamp G, Arnott C, Burke F, East N, Holdsworth H, Turner L, Rollins B, Pasparakis M et al (1999) Mice deficient in tumor necrosis factor-alpha are resistant to skin carcinogenesis. Nat Med 5:828–831 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 Nickoloff BJ, Ben-Neriah Y, Pikarsky E (2005) Inflammation and cancer: is the link as simple as we think? J Invest Dermatol 124:x–xiv O’Neill LA, Bowie AG (2007) The family of five: TIR-domain-containing adaptors in Toll-like receptor signalling. Nat Rev Immunol 7:353–364 Ostrand-Rosenberg S, Grusby MJ, Clements VK (2000) Cutting edge: STAT6-deficient mice have enhanced tumor immunity to primary and metastatic mammary carcinoma. J Immunol 165: 6015–6019 Pikarsky E, Porat RM, Stein I, Abramovitch R, Amit S, Kasem S, Gutkovich-Pyest E, Urieli-Shoval S, Galun E, Ben-Neriah Y (2004) NF-kappaB functions as a tumour promoter in inflammation-associated cancer. Nature 431:461–466 Pollard JW (2009) Trophic macrophages in development and disease. Nat Rev Immunol 9:259–270 Porta C, Rimoldi M, Raes G, Brys L, Ghezzi P, Di Liberto D, Dieli F, Ghisletti S, Natoli G, De Baetselier P et al (2009) Tolerance and M2 (alternative) macrophage polarization are related processes orchestrated by p50 nuclear factor kappaB. Proc Natl Acad Sci USA 106: 14978–14983 Ritchie DS, Smyth MJ (2009) A new therapeutic target for leukemia comes to the surface. Cell 138:226–228 Rius J, Guma M, Schachtrup C, Akassoglou K, Zinkernagel AS, Nizet V, Johnson RS, Haddad GG, Karin M (2008) NF-kappaB links innate immunity to the hypoxic response through transcriptional regulation of HIF-1alpha. Nature 453:807–811 Roca H, Varsos ZS, Sud S, Craig MJ, Ying C, Pienta KJ (2009) CCL2 and IL-6 promote survival of human CD11b+ − peripheral blood mononuclear cells and induce M2-type macrophage polarization. J Biol Chem 284:34342–34354 Saccani A, Schioppa T, Porta C, Biswas SK, Nebuloni M, Vago L, Bottazzi B, Colombo MP, Mantovani A, Sica A (2006) p50 nuclear factor-kappaB overexpression in tumor-associated macrophages inhibits M1 inflammatory responses and antitumor resistance. Cancer Res 66:11432–11440 Sica A, Bronte V (2007) Altered macrophage differentiation and immune dysfunction in tumor development. J Clin Invest 117:1155–1166 Sica A, Saccani A, Bottazzi B, Polentarutti N, Vecchi A, van Damme J, Mantovani A (2000) Autocrine production of IL-10 mediates defective IL-12 production and NF-kappa B activation in tumor-associated macrophages. J Immunol 164:762–767

11

NF-KappaB-Mediated Regulation of Tumour-Associated Macrophages…

165

Sica A, Larghi P, Mancino A, Rubino L, Porta C, Totaro MG, Rimoldi M, Biswas SK, Allavena P, Mantovani A (2008) Macrophage polarization in tumour progression. Semin Cancer Biol 18:349–355 Sierra JR, Corso S, Caione L, Cepero V, Conrotto P, Cignetti A, Piacibello W, Kumanogoh A, Kikutani H, Comoglio PM et al (2008) Tumor angiogenesis and progression are enhanced by Sema4D produced by tumor-associated macrophages. J Exp Med 205:1673–1685 Smyth MJ, Dunn GP, Schreiber RD (2006) Cancer immunosurveillance and immunoediting: the roles of immunity in suppressing tumor development and shaping tumor immunogenicity. Adv Immunol 90:1–50 Song L, Asgharzadeh S, Salo J, Engell K, Wu HW, Sposto R, Ara T, Silverman AM, DeClerck YA, Seeger RC et al (2009) Valpha24-invariant NKT cells mediate antitumor activity via killing of tumor-associated macrophages. J Clin Invest 119:1524–1536 Torroella-Kouri M, Silvera R, Rodriguez D, Caso R, Shatry A, Opiela S, Ilkovitch D, Schwendener RA, Iragavarapu-Charyulu V, Cardentey Y et al (2009) Identification of a subpopulation of macrophages in mammary tumor-bearing mice that are neither M1 nor M2 and are less differentiated. Cancer Res 69:4800–4809 Vaupel P (2008) Hypoxia and aggressive tumor phenotype: implications for therapy and prognosis. Oncologist 13(Suppl 3):21–26 Verma IM (2004) Nuclear factor (NF)-kappaB proteins: therapeutic targets. Ann Rheum Dis 63(Suppl 2):ii57–ii61 Wallace KL, Marshall MA, Ramos SI, Lannigan JA, Field JJ, Strieter RM, Linden J (2009) NKT cells mediate pulmonary inflammation and dysfunction in murine sickle cell disease through production of IFN-gamma and CXCR3 chemokines. Blood 114:667–676 Wels J, Kaplan RN, Rafii S, Lyden D (2008) Migratory neighbors and distant invaders: tumorassociated niche cells. Genes Dev 22:559–574 Windbichler GH, Hausmaninger H, Stummvoll W, Graf AH, Kainz C, Lahodny J, Denison U, Muller-Holzner E, Marth C (2000) Interferon-gamma in the first-line therapy of ovarian cancer: a randomized phase III trial. Br J Cancer 82:1138–1144 Wyckoff JB, Wang Y, Lin EY, Li JF, Goswami S, Stanley ER, Segall JE, Pollard JW, Condeelis J (2007) Direct visualization of macrophage-assisted tumor cell intravasation in mammary tumors. Cancer Res 67:2649–2656 Yu H, Kortylewski M, Pardoll D (2007) Crosstalk between cancer and immune cells: role of STAT3 in the tumour microenvironment. Nat Rev Immunol 7:41–51 Ziegler-Heitbrock HW, Petersmann I, Frankenberger M (1997) p50 (NF-kappa B1) is upregulated in LPS tolerant P388D1 murine macrophages. Immunobiology 198:73–80

Chapter 12

Role of Hypoxia-Inducible Transcription Factors in TAM Function Nadine Rohwer and Thorsten Cramer

Tumour progression is characterized by massive cellular proliferation associated with alterations of the tumour microenvironment. Hence, the tumour microenvironment is considered to be of great importance for tumourigenesis and, as a consequence, might influence the response to antitumour therapy. The microenvironmental alterations comprise hypoxia, acidosis, nutrient starvation, as well as increased interstitial fluid pressure (Denko 2008; Milosevic et  al. 2004; Pouyssegur et  al. 2006; Vaupel et al. 1989) and are largely the result of a defective and/or inadequate tumour vasculature which develops during rapid tumour growth (Bertout et al. 2008; Brown and Giaccia 1998; Vaupel et al. 1989; Vaupel 2004). Hypoxia is present in virtually every solid tumour and probably represents the most persistent of the microenvironmental hallmarks that sway tumour progression. Adaptation of tumour cells to hypoxia is a decisive driving force in the clonal selection that, ultimately, results in a more invasive and aggressive tumour phenotype (Vaupel 2004). In addition, alterations in gene expression with subsequent changes in the proteome as well as genomic instability are involved in hypoxia-induced tumour progression (Vaupel et al. 2004). In this context, hypoxia causes a number of crucial effects on various cellular and physiologic functions, including angiogenesis, cell proliferation, immunosurveillance, metabolism, DNA replication, and protein turnover (Bertout et  al. 2008; Harris 2002; Hockel and Vaupel 2001). Moreover, tumour hypoxia is associated with resistance to radio- and chemotherapy and, ultimately, poor patient prognosis in the majority of human cancers (Bertout et al. 2008; Gray et al. 1953; Teicher et al. 1981, 1990; Teicher 1994; Vaupel et al. 2004). Tumour cells respond to these microenvironmental conditions and co-opt adaptive mechanisms to adjust their oxygen consumption to the limited supply (Denko et  al. 2003; Hockel and Vaupel 2001). One of the most important aspects of how tumour cells respond to this complex N. Rohwer • T. Cramer (*) Medizinische Klinik mit Schwerpunkt Hepatologie und Gastroenterologie and Molekulares Krebsforschungszentrum, Charité – Universitätsmedizin Berlin, Campus Virchow-Klinikum, Augustenburger Platz 1, 13353 Berlin, Germany e-mail: [email protected] T. Lawrence and T. Hagemann (eds.), Tumour-Associated Macrophages, DOI 10.1007/978-1-4614-0662-4_12, © Springer Science+Business Media, LLC 2012

167

168

N. Rohwer and T. Cramer

microenvironment is the activation of the hypoxia-inducible factor (HIF) isoforms (Iyer et al. 1998). HIFs, in turn, initiate several adaptive survival processes such as metabolic changes, angiogenesis, erythropoiesis, cell survival, resistance to apoptosis, and pH homeostasis, thereby promoting malignant progression.

HIF-a as Master Regulators of Adaptation to Hypoxia The transcriptional activator HIF has emerged as the pivotal regulator of oxygen homeostasis. HIF-mediated pathways influence development, physiology, and numerous diseases by modulating crucial cellular and physiologic aspects as mentioned above (Semenza 2007). HIF is a heterodimeric transcription factor consisting of two basic helix-loop-helix (bHLH) proteins of the PAS family (PER, ARNT, and SIM family): the constitutively expressed b-subunit [also known as ARNT (aryl hydrocarbon receptor nuclear translocator)] and a O2-regulated a-subunit (Wang and Semenza 1995; Wang et al. 1995). In mammals, three genes have been shown to encode HIF-a isoforms by distinct genetic loci. Since HIF-1a and HIF-2a share high-sequence homology and a similar domain architecture, proteolytic regulation appears concordant in many aspects. However, HIF-2a exhibits a more restricted tissue expression than HIF-1a, which is widely expressed in tissues (Wiesener et al. 2003). In addition, HIF-1a and HIF-2a activate overlapping but distinct sets of target genes, even in a single cell type (Hu et al. 2007; Lau et al. 2007). Targeted inactivation of HIF-1a and HIF-2a leads to significantly different phenotypes in mouse models (Mole and Ratcliffe 2008). The precise function of HIF-3a is not yet fully understood. However, alternative splicing of HIF-3a generates an inhibitory PAS domain protein (IPAS) that inhibits HIF-1 by forming transcriptionally inactive heterodimers with HIF-1a (Makino et  al. 2001). Under normoxic conditions, HIF-a subunits have a very short half-life due to continuous degradation by the ubiquitin-proteasome system (Jewell et al. 2001). In the presence of oxygen, prolyl hydroxylases (PHDs) catalyze the hydroxylation of two specific prolyl residues within the oxygen-dependent degradation (ODD) domain of the a-subunits (Ivan et  al. 2001; Jaakkola et  al. 2001). Once hydroxylated, the von Hippel–Lindau tumour suppressor protein (pVHL) interacts with HIF-a and recruits an E3 ubiquitin ligase complex that targets HIF-a for ubiquitination and subsequent proteasomal degradation (Maxwell et al. 1999; Ohh et al. 2000). Under hypoxic conditions, PHD activity decreases and prolyl hydroxylation of HIF-a is inhibited. In turn, HIF-a rapidly accumulates, translocates to the nucleus, and heterodimerizes with the b-subunit. The heterodimeric transactivating complex HIF then binds to the core pentanucleotide sequence (RCGTG) in the hypoxia response elements (HRE) of target genes and recruits coactivators to induce target gene expression. The inter­ action of HIF-a with transcriptional coactivators such as p300 and CBP is also inhibited by hydroxylation, in this case that of a C-terminal asparagine residue by an enzyme termed factor-inhibiting HIF (FIH-1) (Lando et al. 2002a, b; Mahon et al. 2001). In addition to hypoxic induction, HIF-a activity can be induced by various

12  Role of Hypoxia-Inducible Transcription Factors in TAM Function

169

growth factors, cytokines, activated oncogenes, or loss-of-function mutated tumour suppressor genes (Frede et al. 2007). HIF isoforms play crucial and distinct roles in tumour progression. Immuno­ histochemical analyses have demonstrated that HIF-1a and/or HIF-2a are overexpressed in the majority of human cancers and their metastases (Semenza 2010; Talks et al. 2000; Zhong et al. 1999). Increased levels of HIF-1a and HIF-2a in cancer and stromal cells can be due to intratumoural hypoxia or other HIF-a inducers such as reactive oxygen and nitrogen species (Brune and Zhou 2007; Dewhirst et al. 2008; Vaupel et al. 2004). Furthermore, HIF-a protein levels can also be induced as a result of genetic and epigenetic alterations such as activated oncogenes or loss-of-function mutated tumour suppressor genes (Semenza 2010). In general, tumoural HIF-1a and HIF-2a overexpression has been associated with a more aggressive tumour phenotype and poor patient survival reflecting the tumour-promoting impact of HIFs (Bertout et al. 2008; Semenza 2010). In line with the clinical data, a vast number of experimental studies have shown that the inhibition of HIF-1a and HIF-2a has proven antitumoural efficacy in various murine tumour models (Semenza 2010). Recently, a role for HIF-1a in promoting pulmonary and osteolytic bone metastasis using a transgenic model of breast cancer was also described (Hiraga et al. 2007; Liao et al. 2007). According to the aforementioned differing patterns of tissue expression and gene activation, it appears that HIF-1a and HIF-2a can also have distinct effects on tumourigenesis. For example, HIF-1a overexpression was associated with decreased patient mortality for head and neck cancer and non-small cell lung cancer, whereas HIF-2a overexpression correlated with increased patient mortality for either cancer type (Beasley et al. 2002; Giatromanolaki et al. 2001; Koukourakis et al. 2002; Volm and Koomagi 2000). Similarly, HIF-2a expression increases, while HIF-1a protein levels gradually decreases as renal cell carcinomas develop in VHL disease patients (Mandriota et al. 2002). Concordant with this, HIF-2a overexpression enhanced the growth of renal cell carcinoma xenografts, whereas overexpression of HIF-1a resulted in inhibited xenograft growth (Raval et  al. 2005). In contrast, inhibiting HIF-1a decreased colorectal carcinoma xenograft growth, whereas HIF-2a inhibition resulted in increased tumour growth (Imamura et al. 2009). Interestingly, the loss of HIF-2a expression was significantly associated with advanced tumour stage in human colon cancer (Imamura et al. 2009). The overall conclusion of these observations is that HIF-1a and HIF-2a generally promote cancer progression, but the two a-subunits can also have opposing effects on tumour growth depending on the tumour entity. These data also demonstrate that the contribution of the a-subunits to tumourigenesis is tissue- and context-specific. Based on the crucial role for a wide range of common human cancers, targeting HIF-a is commonly considered as an innovative and effective approach for cancer therapy (Semenza 2009). Just like all systemically administered pharmacological substances, inhibitors of HIF-a will affect not only neoplastic cells but also any cellular component of the tumour microenvironment. The function of macrophages, for example, is markedly determined by the HIF-a isoforms (Dehne and Brune 2009). Myeloid cell-specific deletion of HIF-1a results in robust inhibition of macrophage (and granulocyte) function, leading to reduced tissue inflammation and bacterial killing (Peyssonnaux et al. 2005; Cramer et al. 2003). In addition, the transcriptional response that mediates cellular adaptation to hypoxia is strongly

170

N. Rohwer and T. Cramer

depending on HIF-a. For the majority of genes, functional inactivation of either HIF-a isoform blocks hypoxia-induced expression in macrophages to a great extent (Fang et al. 2009). Other, as yet unidentified factors are necessary for hypoxia-induced gene expression in macrophages as abrogation of HIF-a does not result in complete loss of the hypoxic response. Interestingly, for a small subset of genes, e.g., the adenosine A2a receptor (ADORA2A) and the intercellular adhesion molecule 1 (ICAM1), HIF-2 represents the definite regulator of enhanced expression in response to hypoxia (Fang et al. 2009). The distinct importance of hypoxia and inflammation for tumour progression together with the above outlined role of HIF-a for accurate macrophage function in these contexts raises the question as to the significance of HIF-a for TAM biology, discussed in the next section.

Evidence for HIF-a Expression in Tumour-Associated Macrophages HIF-2a is expressed on protein and mRNA level in CD68-positive cells in the tumour stroma as evidenced by the analysis of a wide array of tumour samples including breast, colon, ovarian, prostate, and renal carcinomas (Talks et al. 2000). Interestingly, the expression of HIF-2a in TAM is, in most cases, more frequent than in neoplastic cells, arguing for a pivotal importance of this HIF-a isoform for TAM biology. Furthermore, HIF-2a in breast cancer-infiltrating TAM is indicative of high-grade tumours, increased angiogenesis, and, ultimately, poor patient prognosis (Leek et al. 2002). Of note, in some tumour types, for example, head and neck cancer, no correlation between the number of HIF-2a-positive TAM and patient outcome, can be noted (Beasley et al. 2002). This latter result further stresses the above outlined importance of tissue- and microenvironment-specific factors for the regulation of HIF-a biology. Similar to HIF-2a, HIF-1a is found in TAM of various malignant tumours, for example, breast, ovarian, and prostate cancer (Burke et  al. 2002). HIF-1a protein expression in TAM is highly variable within the same tumour: The number of  HIF-1a-positive TAM range from 5 to 50% (Burke et  al. 2002). Both HIF-a isoforms are found predominantly in TAM around necrotic tumour regions, pointing towards hypoxia in the tumour microenvironment as the pivotal factor controlling HIF-a stability in TAM (Burke et al. 2002; Talks et al. 2000).

Functional Importance of HIF-a for TAM Macrophages display a high degree of plasticity depending on function, mode of activation, and tissue distribution (Sica et  al. 2008). Assorting macrophages in a classically activated, pro-inflammatory polarization (termed M1) and an alternatively activated, anti-inflammatory one (M2) represents a simplified but nonetheless useful approach to understand macrophage plasticity (for details see Chap. 1). Experimental results from independent research groups have established the alternatively activated M2 phenotype as the archetypical form of TAM (Mantovani et  al. 2002).

12  Role of Hypoxia-Inducible Transcription Factors in TAM Function LPS, IFNg SOCS1, −3 notch

unpolarized macrophage

171 IL-4, IL-13

IL-4Rα STAT6

HIF-1a

M1 macrophage

iNOS, IL-1, IL-12

HIF-2a

M2 macrophage

arginase, IL-10, CD206

Fig. 12.1  HIF-a determines macrophage polarization. Archetypical Th1 cytokines result in HIF-1a stabilization via the activation of SOCS1 and -3 and possibly notch, ultimately skewing macrophages towards M1. Conversely, central Th2 mediators, e.g., IL-4 and IL-13, result in M2 polarization via IL-4Ra- and potentially STAT6-mediated HIF-2a activation

As outlined above, TAMs are predominantly found in hypoxic areas of tumours. These are characterized by hostile microenvironmental conditions due – in addition to the lack of oxygen – to high lactate concentration, nutrient deprivation, and low pH (Vaupel 2008). Interestingly, not only hypoxia but also any one of these micro­ environmental conditions is able to induce a biologically significant stabilization of HIF-a in vitro (Ruscher et al. 1998; Lu et al. 2002). It was, therefore, both reasonable and important to investigate the functional importance of HIF-a ­isoforms for TAM biology in more detail (Fig. 12.1).

HIF-a and TAM Polarization Differential expression of HIF-1a versus HIF-2a by Th1 versus Th2 cytokines constitutes a pivotal requirement for polarization of macrophages into M1 versus M2 (Takeda et al. 2010). Treatment of macrophages with LPS or interferon-g results in HIF-1a stabilization and subsequent M1 polarization, while HIF-1a-deficient macrophages display significantly reduced ability for differentiation into M1 upon treatment with LPS or interferon-g (Werno et al. 2010; Takeda et al. 2010). On the other hand, archetypical Th2 cytokines, such as interleukin-4, elicit an M2 response via activation of HIF-2a (Takeda et  al. 2010). Th1 cytokines further aggravate the M1-skewing by the inhibition of HIF-2a, while Th2 cytokines fail to show a concomitant reduction of HIF-1a gene expression. These results are especially intriguing because HIF-1a and HIF-2a share pronounced homologies with respect to sequence, activation, and degradation as well as DNA binding and, hence, the

172

N. Rohwer and T. Cramer

differential roles of the HIF-a isoforms have traditionally been difficult to determine. Today, however, it is clear that they have a unique tissue distribution and play pivotal but distinct parts in various pathophysiological conditions, e.g., malignant progression and inflammation (Gordan and Simon 2007). In contrast to HIF-1a, which is ubiquitously expressed in organisms, HIF-2a mRNA is expressed in a tissue-specific pattern predominated by endothelial cells, kidney and pulmonary epithelial cells, and bone marrow macrophages (Qing and Simon 2009). A number of HIF-2a-specific target genes have been identified, for example, erythropoietin, Oct4, and ADFP (adipose differentiation-related protein), suggesting a special importance of HIF-2a for such pivotal processes as erythropoiesis, stem cell function/ differentiation and lipid homeostasis (Rankin et al. 2007, 2009; Covello et al. 2006). While the molecular mechanisms that underlie this specificity remained elusive for quite some time, the analysis of macrophage polarization via HIF-a induction provided some reasonable angles. Both IL-4 and IL-13 share a common receptor (IL-4Ra) that signals through the JAK/STAT6 pathway and the murine HIF-2a promoter sequence has a putative STAT6 binding site (−448 nucleotides upstream of its transcription start site) (Yin et  al. 1994; Malabarba et  al. 1996; Takeda et  al. 2010). However, thus far IL-4/-13-induced activation of HIF-2a via STAT6 remains a reasonable hypothesis as experimental evidence for functional STAT6 binding to  the HIF-2a promoter is still lacking. Another important player in determining macrophage polarization is the SOCS (suppressor of cytokine signaling) pathway: Expression of the isoforms SOCS1 and SOCS3 in macrophages is induced by M1 order Th1 cytokines, while IL-4 results exclusively in SOCS1 activation (Liu et al. 2008). SOCS3 is pivotal for M1 polarization, its functional inactivation in murine macrophages induces typical M2 features: Enhanced expression of macrophage mannose receptor (CD206) and arginase, restoration of IL-4 responsiveness, decreased synthesis of the proinflammatory mediators NO and IL-6, and reduced expression of the co-stimulatory factor CD86. Whether HIF-a is able to distinctively modify the SOCS pathway has not been elucidated thus far, however, indirect evidence for a role of HIF-a in this setting was obtained by analyzing the Notch pathway (Wang et al. 2010). HIF-1a is a critical component of Notch activation as the Notch intracellular domain interacts with HIF-1a, thereby recruiting HIF-1a to Notch-responsive promoters upon Notch activation (Gustafsson et al. 2005). Furthermore, the negative HIF-1a-regulator FIH-1 (factor-inhibiting HIF-1, see above) also modifies Notch activity (Zheng et al. 2008). M2-skewed TAMs have reduced levels of Notch and forced activation of Notch in macrophages is sufficient to establish M1 polarization (Wang et al. 2010). The inhibition of M2 polarization by Notch is partly mediated by SOCS3, further strengthening the importance of SOCS3 for macrophage polarization.

The Importance of HIF-a for Pivotal TAM Properties As comprehensively outlined in other chapters in this textbook, TAMs accelerate tumour progression by multiple means, for example, via induction of angiogenesis and inflammation as well as inhibition of tumour-directed cellular immunity.

12  Role of Hypoxia-Inducible Transcription Factors in TAM Function angiogenesis

migration

173

NF-kB activity

HIF-α

macrophage polarization

activation by dying tumour cells

interplay with cellular immunity

Fig. 12.2  HIF-a and TAM biology. Schematic overview of TAM properties that are critically influenced by HIF-a

The specific role of HIF-a isoforms for particular pro-tumourigenic properties will be described in the following sections (Fig. 12.2).

HIF-a and Macrophage Migration The majority of solid tumours show a dense infiltration with macrophages, clearly outnumbering the tissue of origin. Occasionally, macrophages can make up the majority of a tumours cellular mass (Pollard 2004). Hence, the ability for directed migration into tumours constitutes a basic and indispensable feature to instigate the TAM phenotype. Both HIF-a isoforms have been attributed to the migratory capacity of macrophages: HIF-1a regulates macrophage migration (and invasion of extracellular matrix) in different murine models of myeloid cell-mediated inflammation (Cramer et al. 2003). HIF-2a is essential for the movement of macrophages into neoplastic tissues as analyzed in murine liver and colon tumour models (Imtiyaz et al. 2010). Interestingly, the molecular mechanisms differ quite substantially as HIF-1a exerts its role predominantly via the control of glycolytic ATP synthesis, while HIF-2a directly controls the expression of membrane-bound receptors pivotal for migration (M-CSFR and CXCR4) without effecting cellular ATP levels (Imtiyaz et al. 2010; Cramer et al. 2003). Taken together, even though the functional importance of HIF-1a for migration and tissue invasion of TAMs still needs to be addressed by future work, it is suffice to say that HIF-a constitutes a central regulator of macrophage motility.

174

N. Rohwer and T. Cramer

The Role of HIF-a for TAM-Induced Tumoural Angiogenesis Macrophages are essential cellular components of the angiogenic switch in tumours, mainly due to their ability to enhance the local abundance of pro-angiogenic factors. This is either achieved by direct secretion of molecules such as vascular endothelial growth factor A (VEGF-A) and basic fibroblast growth factor (bFGF) or by means of matrix metalloproteinase (MMP) 9-mediated liberation of pro-angiogenic factors from the extracellular matrix (for details, refer to Chap. 2). Transcriptional activity of the majority of pro-angiogenic factors is tightly controlled by HIF-a in cells of mesenchymal as well as of epithelial origin (Rey and Semenza 2010). In macrophages, both HIF-1a and -2a are able to induce the expression of endothelial mitogens. However, exemplified by analyzing VEGF-A, HIF-1a is of markedly greater importance for hypoxia-induced upregulation of pro-angiogenic factor expression in vitro than HIF-2a (Imtiyaz et al. 2010; Cramer et al. 2003). The importance of HIF-1a for macrophage-stimulated blood vessel growth is further strengthened by experiments using three dimensional spheroids of human breast tumour cells infiltrated with macrophages: The supernatant of spheroids infiltrated with HIF-1a-deficient macrophages fails to induce differentiation of embryonic stem cells to endothelial precursor cells (Werno et al. 2010). In summary, HIF-a holds the potential to significantly influence TAM-induced tumoural angiogenesis. It should be noted, however, that the majority of the hitherto published experiments were performed with macrophages (as opposed to TAMs) in rather artificial experimental conditions. While the scientific significance of these results is undisputed, the validity for the in vivo situation clearly needs clarification by future experiments.

HIF-a in TAM Is Strongly Activated by Dying Tumour Cells Besides their crucial role in host defense via phagocytosis of invading microorganisms, macrophages are of pivotal importance for tissue integrity by removing damaged and decomposed cells (Henson and Hume 2006). In addition, macrophages (and other phagocytes) are able to engulf and lyse dysplastic as well as fully transformed cells, thereby inhibiting tumour formation at early stages. While this phagocyte property clearly does not represent a sufficient barrier against tumour formation, macrophages are believed to confer substantial growth advantages to neoplastic cells at later stages of tumour development (for details refer to Chaps. 3 and 4). The emergence of dying cells is a typical observation during tumour formation and mainly explained by the hostile microenvironmental conditions that accompany the rapid and chaotic growth of tumours: Hypoxia, low nutrient availability, low pH, and high interstitial pressure. When exposed to these conditions, tumour cells respond either by terminal growth arrest (senescence), autophagy or demise (mainly via apoptosis or necrosis, occasionally by mitotic catastrophe). Importantly, both apoptotic and necrotic tumour cell death constitutes a strong attractant for macrophage migration

12  Role of Hypoxia-Inducible Transcription Factors in TAM Function

175

into the  tumour microenvironment (Henson and Hume 2006). Besides engulfing dead or dying tumour cells, these macrophages secrete a plethora of tumourpromoting factors (e.g., VEGF-A and MMPs), ultimately converting the perdition of a few neoplastic cells into a substantial growth advantage for the entire tumour. HIF-1a is partially involved in this process as the supernatant from apoptotic tumour cells strongly induces HIF-1a activity in macrophages followed by proangiogenic ­signaling (Herr et al. 2009). While this impact of tumour cell apoptosis is certainly an intriguing observation, its significance for tumour progression awaits confirmation in vivo.

Interdependency of HIF-a and NF-kB: Relevance for Macrophage Polarization The family of NF-kB transcription factors is a pivotal regulator of inflammation, immunity, and cancer (Karin 2006). The notion that these conditions are also markedly influenced by HIF-a, together with the observation that NF-kB under certain circumstances is inducible by hypoxia soon raised the intriguing question, if these two factors interact in a biologically relevant manner. It soon became clear that NF-kB regulates HIF-1a activation in response to various stimuli, for example, proinflammatory cytokines [e.g., tumour necrosis factor alpha (TNF-a) and interleukin 1 beta (IL-1b)] and growth factors [e.g., hepatocyte growth factor (HGF) and the microtubule-disrupting agent colchicine] (Tacchini et al. 2004; Jung et al. 2003a, b, c). The effect of NF-kB on HIF-1a is in large part mediated by an NF-kB consensus site in the HIF-1a promoter at −197 to −188 bp (Bonello et al. 2007). The NF-kB subunits p65 and p50 bind to the HIF-1a promoter and control HIF-1a gene expression under conditions of oxidative stress and short-term hypoxia, while NF-kB seems to be dispensable for HIF-1a activation under prolonged periods of hypoxia (Belaiba et al. 2007; Bonello et al. 2007; Fang et al. 2009). Taken together, NF-kB is a crucial transcriptional activator of HIF-1a and basal NF-kB activity is indispensable for HIF-1a protein stabilization under hypoxia. Of note, studies with murine macrophages under both short- and long-term exposure to hypoxia failed to show a significant role of NF-kB isoforms for the regulation of HIF-2a activity under these conditions (Fang et al. 2009). Vice versa, regulation of NF-kB activity by HIFs was first noted in murine neutrophil granulocytes. Like other cells of the immune system, granulocytes need to perform their specialized functions under conditions of hostile microenvironmental conditions, e.g., hypoxia. To achieve this, immune cells have developed an arsenal of distinct molecular and cell biological responses for proper function under hypoxia. This is illustrated by the ability of granulocytes and macrophages to defer the induction of apoptosis that normally occurs in non-transformed cells under hypoxic conditions. Functional inactivation of HIF-1a results in the loss of apoptosis-resistance of neutrophil granulocytes mediated partly by a failure to activate NF-kB and its well-described anti-apoptotic properties (Walmsley et  al. 2005). Control of NF-kB by HIF-1a is not restricted to immune cells as the induction

176

N. Rohwer and T. Cramer

of NF-kB upon treatment of gastric cancer cells with chemotherapeutic agents also depends on HIF-1a (Rohwer et  al. 2010). Taken together, the crosstalk between HIF-1a and NF-kB is of pivotal importance for various aspects of biology and pathobiology. The fact that NF-kB, similar to HIF-a, plays a central role in macrophage polarization and emergence of the TAM phenotype (for details see Chaps. 10 and 11) warrants the perception of inflammation-associated cancer as the principal model for HIF-a/NF-kB interaction. To date, however, the majority of studies on the interaction of HIF-a and NF-kB was performed in primary murine macrophages or immortalized monocytic cell lines. Undoubtedly, in (or at least ex) vivo experiments with appropriate tumour models are required to definitely clarify relevance and druggability of the HIF-a/NF-kB crosstalk for TAM function.

Myeloid HIF-1a Suppresses Tumour-Associated Adaptive Immunity The adaptive immune system is able to modify tumour progression by multiple means, a phenomenon called “cancer immunosurveillance” by Brunet and Thomas, later refined to “cancer immunoediting” by Schreiber and colleagues (Dunn et  al. 2006). Among the cellular effectors of the adaptive immune system, T lymphocytes received special attention by oncologists due to their ability to kill neoplastic cells upon recognition of specific “tumour antigens” on the cell surface. The relevance of this phenomenon for the progression of human tumours long remained obscure as cancer is a frequent disease and the majority of tumours do not show signs of immunological control. Interestingly, while lymphocytes isolated from the spleen or peripheral blood of cancer patients frequently display robust antitumour properties, lymphocytes isolated directly from tumour tissues fail to do so. It has long been hypothesized that the specific composition of the tumour microenvironment is able to suppress antitumour adaptive immunity. In fact, hypoxia represents a pivotal factor in this setting as a lack of oxygen in the tumour microenvironment results in marked inhibition of immune surveillance (Corzo et al. 2010; Doedens et al. 2010). HIF-1a is the principle molecular mediator here as its functional inactivation in TAM or myeloid-derived suppressor cells (MDSC) abolishes the immune-suppressive function of hypoxia in murine tumour models (Corzo et al. 2010; Doedens et al. 2010).

Future Directions The majority of our current knowledge on the role of HIF-a for TAM function was obtained by analyzing in  vitro polarized macrophages. For a comprehensive and detailed understanding, experimental settings that more closely mirror the in vivo situation are needed. This is especially true for HIF-a, as these transcription factors

12  Role of Hypoxia-Inducible Transcription Factors in TAM Function

177

are markedly influenced by microenvironmental conditions that cannot be sufficiently imitated in  vitro. The availability of genetically engineered mice harboring ­macrophage-specific inactivation of HIF-1a and HIF-2a, respectively, will result in further disclosure of the precise role of HIF-a for TAM (and macrophage) function (Imtiyaz et al. 2010; Cramer et al. 2003). As a final caveat it should be noted that, due to relevant interspecies divergence between the human and murine mononuclear phagocyte system, a direct transfer from mouse model results to the human context cannot be achieved (Martinez et al. 2006). This is exemplified by the missing induction of the HIF-2a target and murine M2 marker arginase-1 in human M2 macrophages (Raes et al. 2005; Scotton et al. 2005). Hence, findings obtained with murine methodology call for meticulous confirmation in the human system before any potential therapeutic implications can be deducted.

References Beasley NJ, Leek R, Alam M et al (2002) Hypoxia-inducible factors HIF-1alpha and HIF-2alpha in head and neck cancer: relationship to tumor biology and treatment outcome in surgically resected patients. Cancer Res 62:2493–2497 Belaiba RS, Bonello S, Zahringer C et al (2007) Hypoxia up-regulates hypoxia-inducible factor1alpha transcription by involving phosphatidylinositol 3-kinase and nuclear factor kappaB in pulmonary artery smooth muscle cells. Mol Biol Cell 18:4691–4697 Bertout JA, Patel SA, Simon MC (2008) The impact of O2 availability on human cancer. Nat Rev Cancer 8:967–975 Bonello S, Zahringer C, Belaiba RS et al (2007) Reactive oxygen species activate the HIF-1alpha promoter via a functional NFkappaB site. Arterioscler Thromb Vasc Biol 27:755–761 Brown JM, Giaccia AJ (1998) The unique physiology of solid tumors: opportunities (and problems) for cancer therapy. Cancer Res 58:1408–1416 Brune B, Zhou J (2007) Nitric oxide and superoxide: interference with hypoxic signaling. Cardiovasc Res 75:275–282 Burke B, Tang N, Corke KP et al (2002) Expression of HIF-1alpha by human macrophages: implications for the use of macrophages in hypoxia-regulated cancer gene therapy. J Pathol 196: 204–212 Corzo CA, Condamine T, Lu L et al (2010) HIF-1alpha regulates function and differentiation of myeloid-derived suppressor cells in the tumor microenvironment. J Exp Med 207:2439–2453 Covello KL, Kehler J, Yu H et al (2006) HIF-2alpha regulates Oct-4: effects of hypoxia on stem cell function, embryonic development, and tumor growth. Genes Dev 20:557–570 Cramer T, Yamanishi Y, Clausen BE et al (2003) HIF-1alpha is essential for myeloid cell-mediated inflammation. Cell 112:645–657 Dehne N, Brune B (2009) HIF-1 in the inflammatory microenvironment. Exp Cell Res 315: 1791–1797 Denko NC (2008) Hypoxia, HIF1 and glucose metabolism in the solid tumour. Nat Rev Cancer 8:705–713 Denko NC, Fontana LA, Hudson KM et al (2003) Investigating hypoxic tumor physiology through gene expression patterns. Oncogene 22:5907–5914 Dewhirst MW, Cao Y, Moeller B (2008) Cycling hypoxia and free radicals regulate angiogenesis and radiotherapy response. Nat Rev Cancer 8:425–437 Doedens AL, Stockmann C, Rubinstein MP et al (2010) Macrophage expression of hypoxia-inducible factor-1 alpha suppresses T-cell function and promotes tumor progression. Cancer Res 70: 7465–7475

178

N. Rohwer and T. Cramer

Dunn GP, Koebel CM, Schreiber RD (2006) Interferons, immunity and cancer immunoediting. Nat Rev Immunol 6:836–848 Fang HY, Hughes R, Murdoch C et al (2009) Hypoxia-inducible factors 1 and 2 are important transcriptional effectors in primary macrophages experiencing hypoxia. Blood 114:844–859 Frede S, Berchner-Pfannschmidt U, Fandrey J (2007) Regulation of hypoxia-inducible factors during inflammation. Methods Enzymol 435:405–419 Giatromanolaki A, Koukourakis MI, Sivridis E et al (2001) Relation of hypoxia inducible factor 1 alpha and 2 alpha in operable non-small cell lung cancer to angiogenic/molecular profile of tumours and survival. Br J Cancer 85:881–890 Gordan JD, Simon MC (2007) Hypoxia-inducible factors: central regulators of the tumor phenotype. Curr Opin Genet Dev 17:71–77 Gray LH, Conger AD, Ebert M et al (1953) The concentration of oxygen dissolved in tissues at the time of irradiation as a factor in radiotherapy. Br J Radiol 26:638–648 Gustafsson MV, Zheng X, Pereira T et al (2005) Hypoxia requires notch signaling to maintain the undifferentiated cell state. Dev Cell 9:617–628 Harris AL (2002) Hypoxia – a key regulatory factor in tumour growth. Nat Rev Cancer 2:38–47 Henson PM, Hume DA (2006) Apoptotic cell removal in development and tissue homeostasis. Trends Immunol 27:244–250 Herr B, Zhou J, Werno C et al (2009) The supernatant of apoptotic cells causes transcriptional activation of hypoxia-inducible factor-1alpha in macrophages via sphingosine-1-phosphate and transforming growth factor-beta. Blood 114:2140–2148 Hiraga T, Kizaka-Kondoh S, Hirota K et al (2007) Hypoxia and hypoxia-inducible factor-1 expression enhance osteolytic bone metastases of breast cancer. Cancer Res 67:4157–4163 Hockel M, Vaupel P (2001) Tumor hypoxia: definitions and current clinical, biologic, and molecular aspects. J Natl Cancer Inst 93:266–276 Hu CJ, Sataur A, Wang L et al (2007) The N-terminal transactivation domain confers target gene specificity of hypoxia-inducible factors HIF-1alpha and HIF-2alpha. Mol Biol Cell 18: 4528–4542 Imamura T, Kikuchi H, Herraiz MT et al (2009) HIF-1alpha and HIF-2alpha have divergent roles in colon cancer. Int J Cancer 124:763–771 Imtiyaz HZ, Williams EP, Hickey MM et  al (2010) Hypoxia-inducible factor 2alpha regulates macrophage function in mouse models of acute and tumor inflammation. J Clin Invest 120(8): 2699–2714 Ivan M, Kondo K, Yang H et al (2001) HIFalpha targeted for VHL-mediated destruction by proline hydroxylation: implications for O2 sensing. Science 292:464–468 Iyer NV, Kotch LE, Agani F et al (1998) Cellular and developmental control of O2 homeostasis by hypoxia-inducible factor 1 alpha. Genes Dev 12:149–162 Jaakkola P, Mole DR, Tian YM et  al (2001) Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science 292:468–472 Jewell UR, Kvietikova I, Scheid A et al (2001) Induction of HIF-1alpha in response to hypoxia is instantaneous. FASEB J 15:1312–1314 Jung Y, Isaacs JS, Lee S et al (2003a) Hypoxia-inducible factor induction by tumour necrosis factor in normoxic cells requires receptor-interacting protein-dependent nuclear factor kappa B activation. Biochem J 370:1011–1017 Jung YJ, Isaacs JS, Lee S et  al (2003b) IL-1beta-mediated up-regulation of HIF-1alpha via an  NFkappaB/COX-2 pathway identifies HIF-1 as a critical link between inflammation and oncogenesis. FASEB J 17:2115–2117 Jung YJ, Isaacs JS, Lee S et al (2003c) Microtubule disruption utilizes an NFkappa B-dependent pathway to stabilize HIF-1alpha protein. J Biol Chem 278:7445–7452 Karin M (2006) Nuclear factor-kappaB in cancer development and progression. Nature 441: 431–436 Koukourakis MI, Giatromanolaki A, Sivridis E et al (2002) Hypoxia-inducible factor (HIF1A and HIF2A), angiogenesis, and chemoradiotherapy outcome of squamous cell head-and-neck cancer. Int J Radiat Oncol Biol Phys 53:1192–1202

12  Role of Hypoxia-Inducible Transcription Factors in TAM Function

179

Lando D, Peet DJ, Gorman JJ et  al (2002a) FIH-1 is an asparaginyl hydroxylase enzyme that regulates the transcriptional activity of hypoxia-inducible factor. Genes Dev 16:1466–1471 Lando D, Peet DJ, Whelan DA et al (2002b) Asparagine hydroxylation of the HIF transactivation domain a hypoxic switch. Science 295:858–861 Lau KW, Tian YM, Raval RR et al (2007) Target gene selectivity of hypoxia-inducible factor-alpha in renal cancer cells is conveyed by post-DNA-binding mechanisms. Br J Cancer 96: 1284–1292 Leek RD, Talks KL, Pezzella F et al (2002) Relation of hypoxia-inducible factor-2 alpha (HIF-2 alpha) expression in tumor-infiltrative macrophages to tumor angiogenesis and the oxidative thymidine phosphorylase pathway in Human breast cancer. Cancer Res 62:1326–1329 Liao D, Corle C, Seagroves TN et al (2007) Hypoxia-inducible factor-1 alpha is a key regulator of metastasis in a transgenic model of cancer initiation and progression. Cancer Res 67:563–572 Liu Y, Stewart KN, Bishop E et al (2008) Unique expression of suppressor of cytokine signaling 3 is essential for classical macrophage activation in rodents in  vitro and in  vivo. J Immunol 180:6270–6278 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 Mahon PC, Hirota K, Semenza GL (2001) FIH-1: a novel protein that interacts with HIF-1alpha and VHL to mediate repression of HIF-1 transcriptional activity. Genes Dev 15:2675–2686 Makino Y, Cao R, Svensson K et al (2001) Inhibitory PAS domain protein is a negative regulator of hypoxia-inducible gene expression. Nature 414:550–554 Malabarba MG, Rui H, Deutsch HH et al (1996) Interleukin-13 is a potent activator of JAK3 and STAT6 in cells expressing interleukin-2 receptor-gamma and interleukin-4 receptor-alpha. Biochem J 319(Pt 3):865–872 Mandriota SJ, Turner KJ, Davies DR et al (2002) HIF activation identifies early lesions in VHL kidneys: evidence for site-specific tumor suppressor function in the nephron. Cancer Cell 1:459–468 Mantovani A, Sozzani S, Locati M et  al (2002) Macrophage polarization: tumor-associated macrophages as a paradigm for polarized M2 mononuclear phagocytes. Trends Immunol 23:549–555 Martinez FO, Gordon S, Locati M et al (2006) Transcriptional profiling of the human monocyteto-macrophage differentiation and polarization: new molecules and patterns of gene expression. J Immunol 177:7303–7311 Maxwell PH, Wiesener MS, Chang GW et al (1999) The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 399:271–275 Milosevic M, Fyles A, Hedley D et al (2004) The human tumor microenvironment: invasive (needle) measurement of oxygen and interstitial fluid pressure. Semin Radiat Oncol 14:249–258 Mole DR, Ratcliffe PJ (2008) Cellular oxygen sensing in health and disease. Pediatr Nephrol 23:681–694 Ohh M, Park CW, Ivan M et al (2000) Ubiquitination of hypoxia-inducible factor requires direct binding to the beta-domain of the von Hippel-Lindau protein. Nat Cell Biol 2:423–427 Peyssonnaux C, Datta V, Cramer T et al (2005) HIF-1alpha expression regulates the bactericidal capacity of phagocytes. J Clin Invest 115:1806–1815 Pollard JW (2004) Tumour-educated macrophages promote tumour progression and metastasis. Nat Rev Cancer 4:71–78 Pouyssegur J, Dayan F, Mazure NM (2006) Hypoxia signalling in cancer and approaches to enforce tumour regression. Nature 441:437–443 Qing G, Simon MC (2009) Hypoxia inducible factor-2alpha: a critical mediator of aggressive tumor phenotypes. Curr Opin Genet Dev 19:60–66 Raes G, Van den BR, De BP et al (2005) Arginase-1 and Ym1 are markers for murine, but not human, alternatively activated myeloid cells. J Immunol 174:6561–6562 Rankin EB, Biju MP, Liu Q et  al (2007) Hypoxia-inducible factor-2 (HIF-2) regulates hepatic erythropoietin in vivo. J Clin Invest 117:1068–1077

180

N. Rohwer and T. Cramer

Rankin EB, Rha J, Selak MA et  al (2009) Hypoxia-inducible factor 2 regulates hepatic lipid metabolism. Mol Cell Biol 29:4527–4538 Raval RR, Lau KW, Tran MG et al (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 Rey S, Semenza GL (2010) Hypoxia-inducible factor-1-dependent mechanisms of vascularization and vascular remodelling. Cardiovasc Res 86:236–242 Rohwer N, Dame C, Haugstetter A et  al (2010) Hypoxia-inducible factor 1alpha determines gastric cancer chemosensitivity via modulation of p53 and NF-kappaB. PLoS One 5:e12038 Ruscher K, Isaev N, Trendelenburg G et al (1998) Induction of hypoxia inducible factor 1 by oxygen glucose deprivation is attenuated by hypoxic preconditioning in rat cultured neurons. Neurosci Lett 254:117–120 Scotton CJ, Martinez FO, Smelt MJ et al (2005) Transcriptional profiling reveals complex regulation of the monocyte IL-1 beta system by IL-13. J Immunol 174:834–845 Semenza GL (2007) Hypoxia-inducible factor 1 (HIF-1) pathway. Sci STKE 2007:cm8 Semenza GL (2009) HIF-1 inhibitors for cancer therapy: from gene expression to drug discovery. Curr Pharm Des 15:3839–3843 Semenza GL (2010) Defining the role of hypoxia-inducible factor 1 in cancer biology and therapeutics. Oncogene 29:625–634 Sica A, Larghi P, Mancino A et al (2008) Macrophage polarization in tumour progression. Semin Cancer Biol 18:349–355 Tacchini L, De PC, Matteucci E et al (2004) Hepatocyte growth factor-activated NF-kappaB regulates HIF-1 activity and ODC expression, implicated in survival, differently in different carcinoma cell lines. Carcinogenesis 25:2089–2100 Takeda N, O’Dea EL, Doedens A et  al (2010) Differential activation and antagonistic function of  HIF-{alpha} isoforms in macrophages are essential for NO homeostasis. Genes Dev 24: 491–501 Talks KL, Turley H, Gatter KC et 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 157:411–421 Teicher BA (1994) Hypoxia and drug resistance. Cancer Metastasis Rev 13:139–168 Teicher BA, Lazo JS, Sartorelli AC (1981) Classification of antineoplastic agents by their selective toxicities toward oxygenated and hypoxic tumor cells. Cancer Res 41:73–81 Teicher BA, Holden SA, al-Achi A et al (1990) Classification of antineoplastic treatments by their differential toxicity toward putative oxygenated and hypoxic tumor subpopulations in vivo in the FSaIIC murine fibrosarcoma. Cancer Res 50:3339–3344 Vaupel P (2004) The role of hypoxia-induced factors in tumor progression. Oncologist 9(Suppl 5): 10–17 Vaupel P (2008) Hypoxia and aggressive tumor phenotype: implications for therapy and prognosis. Oncologist 13(Suppl 3):21–26 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, Mayer A, Hockel M (2004) Tumor hypoxia and malignant progression. Methods Enzymol 381:335–354 Volm M, Koomagi R (2000) Hypoxia-inducible factor (HIF-1) and its relationship to apoptosis and proliferation in lung cancer. Anticancer Res 20:1527–1533 Walmsley SR, Print C, Farahi N et al (2005) Hypoxia-induced neutrophil survival is mediated by HIF-1alpha-dependent NF-kappaB activity. J Exp Med 201:105–115 Wang GL, 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-helix-PAS heterodimer regulated by cellular O2 tension. Proc Natl Acad Sci USA 92:5510–5514

12  Role of Hypoxia-Inducible Transcription Factors in TAM Function

181

Wang YC, He F, Feng F et al (2010) Notch signaling determines the M1 versus M2 polarization of macrophages in antitumor immune responses. Cancer Res 70:4840–4849 Werno C, Menrad H, Weigert A et al (2010) Knockout of Hif-1alpha in tumor-associated macrophages enhances M2 polarization and attenuates their pro-angiogenic responses. Carcinogenesis 31(10):1863–1872 Wiesener MS, Jurgensen JS, Rosenberger C et al (2003) Widespread hypoxia-inducible expression of HIF-2alpha in distinct cell populations of different organs. FASEB J 17:271–273 Yin T, Tsang ML, Yang YC (1994) JAK1 kinase forms complexes with interleukin-4 receptor and 4PS/insulin receptor substrate-1-like protein and is activated by interleukin-4 and interleukin-9 in T lymphocytes. J Biol Chem 269:26614–26617 Zheng X, Linke S, Dias JM et al (2008) Interaction with factor inhibiting HIF-1 defines an additional mode of cross-coupling between the Notch and hypoxia signaling pathways. Proc Natl Acad Sci USA 105:3368–3373 Zhong H, De Marzo AM, Laughner E et al (1999) Overexpression of hypoxia-inducible factor 1 alpha in common human cancers and their metastases. Cancer Res 59:5830–5835

Index

A Adaptive immunity HIF–1D, 176 response, TLR signaling, 123 Angiogenesis. See Tumour angiogenesis Anti-angiogenic therapy, 108, 112 Antigen-presenting cells (APCs) cysteine, from T cells, 99–100 MDSC, 100–101 Arginase, 78, 82–83, 85 Arginase 1 (ARG1) and nitric oxide synthase 2 (NOS2), 56, 57 Arginine and cysteine deprivation, in tumoral immune escape, 98–99 L-Arginine metabolism, 77 cancer cells and tumour microenvironment arginase, 82–83 inducible nitric oxide synthase (iNOS), 83–84 M1 and M2, 81 cationic amino acid transporter (CAT), 80 enzymes regulation, 78 LPS activated macrophages, 80, 81 pharmacological modulation, malignancy, 88 iNOS/arginase, 87 liposome-encapsulated clodronate (CLIP), 86, 87 targeting TAMs, 86–87 synthesis, 78, 79 Th1, Th2-type cytokines, 79, 80 tumoural metabolism argininosuccinate synthetase (ASS1), 84–85 iNOS/arginase, 85 Argininosuccinate synthetase (ASS1), 77, 84–85

B Bacillus Calmette-Guerin (BCG) vaccine, 128 Basophils, SHIP, 144–145 Biglycan, 125 Bisphosphonate, 63, 64

C CAT. See Cationic amino acid transporter (CAT) Cathelicidin, 124 Cathepsins, 40 Cationic amino acid transporter (CAT), 80 CCAAT-enhancer-binding protein-beta (C/ EBPE), 54 CD97 molecule, 10–11 Chemokines CXCR4/CXCL12, 37, 38 Tumour-promoting signaling cascades, 38 Clinical target. See Therapeutic strategies, TAM Colony-stimulating factor–1 (CSF–1), 32, 33, 35–36 CX3CL1, 65–66 Cyclooxygenase–2 (COX–2), 147 Cysteine APC, T-cell activation, 99–100 deprivation MDSC, 100–101 in tumoral immune escape, 98–99

D E-Defensin–2, 124 Dendritic cells (DCs) cytokine, 53 hematopoietic progenitors cells (HPCs), 50

T. Lawrence and T. Hagemann (eds.), Tumour-Associated Macrophages, DOI 10.1007/978-1-4614-0662-4, © Springer Science+Business Media, LLC 2012

183

184 Dendritic cells (DCs) (cont.) iDCs maturation, 50, 51 Janus-activated kinase 2 (JAK2), 54 signal transducer and activator of transcription 3 (STAT3), 54 Difluoromethylornithine (DFMO), 87

E ECM. See Extracellular matrix (ECM) EGF. See Epidermal growth factor (EGF) EGF-TM7. See Epidermal growth factor (EGF)-seven-span transmembrane (7TM) receptors (EGF-TM7) EMR2 molecule, 11–13 Endogenous ligands, of TLR extracellular matrix, 125 induction of, 124–125 intracellular-derived, 123–124 Epidermal growth factor (EGF), 35–36 Epidermal growth factor (EGF)-seven-span transmembrane (7TM) receptors (EGF-TM7) CD97, 10–11 characteristics, 7–8 EMR2, 11–13 F4/80, 8–9 Epithelial-mesenchymal transition (EMT), 32 Extracellular matrix (ECM), 19, 20, 32 ligands, of TLR, 125 proteases role and, 40–41

F F4/80 molecule, 8–9 Fractalkine. See CX3CL1 Functional plasticity, 18

G Granulocyte–monocyte colony stimulating factor (GM-CSF), 53

H Hematopoietic progenitors cells (HPCs), 50, 53 High-mobility group box 1 (HMGB1), 123–124 Hyaluronan, 125 Hypoxia, 21, 167 NF-NB signaling, 158–159 signalling, 22–23 VEGF, 107–108

Index Hypoxia-inducible factor-D (HIF-D) adaptation, regulators of, 168–170 adaptive immunity, 176 dying tumour cells, 174–175 expression, evidence for, 170 functional importance, 170–171 genetically engineered mice, 177 HIF–1D and HIF–2D, 168, 169 macrophage migration, 173 microenvironmental alterations, 167, 168 and NF-NB, interdependency, 175–176 Notch pathway, 172 pivotal TAM properties, importance of, 172–173 suppressor of cytokine signaling (SOCS) pathway, 172 TAM polarization, 171–172 tumoral angiogenesis, 174

I IKKE, 158 Imiquimod, 128 Indoleamine 2, 3-dioxygenase amino acid metabolism APC cysteine, from T cells, 99–100 MDSC, 100–101 cysteine/cystine, 99 immune escape arginine and cysteine deprivation, 98–99 in cancer, 94–95 genetically synonymous pathways, 96–98 and immune tolerance, 93–94 immunoediting and oncogenesis, integrated model of, 92–93 and inflammatory, in tumour microenvironment, 91–92 inflammatory skin carcinogenesis, 95–96 Inducible nitric oxide synthase (iNOS), 83–85 Inflammation and cancer, 153, 154 Intracellular-derived endogenous ligands, TLR, 123–124

J Janus-activated kinase 2 (JAK2), 54

L Leukocyte trafficking, CD11b and CD18, 65 Lipopolysaccharide (LPS)-tolerant macrophages, p50NF-NB

Index homodimer-mediated M2 polarization, 158, 159 Liposomal clodronate, 64

M Macrophage ablation, 63–64 phenotype, in tumours CD97, 7, 10–11 cellular ligand, 8–10 EGF-TM7 receptors and, 6–8 EMR2, 11–13 F4/80, 7–9 heterogeneity in, 4–5 membrane markers for, 5–6 metastasis, 3 micro-environment, 3, 5, 13 monocytes and, 3–4 Macrophage-induced invasion experimental evidence for resident macrophages role, 33–34 tumour cell–macrophage interactions, 32–33 hypothetical model of, 42 signaling chemokine connection, 37–38 EGF/CSF–1 loop, 35–36 tumour necrosis factor D (TNFD), 36–37 Wnt signaling, 38–39 Malignant invasion cell motility, 31, 32 epithelial-mesenchymal transition (EMT), 32 extracellular matrix (ECM), 40–41 hypothetical model, 42 infiltrating leukocytes, 31 macrophage-induced invasion, experimental evidence for resident macrophages role, 33–34 tumour cell-macrophage interactions, 32–33 migration, 32, 37, 41 proteases role, 40–41 signaling chemokine connection, 37–38 EGF/CSF–1 loop, 35–36 tumour necrosis factor D (TNFD), 36–37 Wnt signaling, 38–39 Matrix metalloproteinases (MMPs), 19, 40 MDSCs. See Myeloid-derived suppressor cells (MDSCs) Medical Research Council Myeloma IX Study, 64 Membrane markers, macrophage, 5–6

185 MicroRNAs (miRs), 143–144 Monocytes and macrophage, 3–4 recruitment inhibition CX3CL1, 65–66 leukocyte trafficking, CD11b and CD18, 65 Myeloid cells dendritic cells (DCs) cytokine, 53 failure, 51 hematopoietic progenitors cells (HPCs), 50 immature DCs (iDCs), 50, 51 Janus-activated kinase 2 (JAK2), 54 pluripotent cell types, 51 signal transducer and activator of transcription 3 (STAT3), 54 immune response mechanisms, 50 MDSCs, 51, 52 ARG1 and NOS2 in, 56 CCAAT-enhancer-binding protein-beta (C/EBPE), 54 GM-CSF, 53 tumour-associated antigens, identification of, 49 tumour-derived soluble factors (TDSFs), 52, 54 tumour escape, 50 tumour permissive microenvironment adaptive immunity regulation, 55 ARG1/NOS2, 56, 57 balance needle model, 57, 58 and inflammation, 57 lymphoid organs, 55, 56 Treg expansion, 58 VEGF, 52, 53, 110, 111 Myeloid-derived suppressor cells (MDSCs), 100–101 ARG1 and NOS2, 56, 57 CCAAT-enhancer-binding protein-beta (C/ EBPE), 54 GM-CSF, 53 Myeloid differentiation factor 88 (MyD88) signaling, 121–122

N Nanoparticles, 67, 68 NF-KappaB (NF-NB), 23 extracellular signals, influence of, 154–155 HIF-D, 175–176 inflammation and cancer, 153, 154 intrinsic and extrinsic pathways, 153

186 NF-KappaB (NF-NB) (cont.) molecular determinants, of TAM cellular alterations, 156 defective activation in, 157 functions, 155 homo-and heterodimers, 156 hypoxia, 158–159 IKKE, 158 vs. LPS-tolerant macrophages, p50 homodimer-mediated M2 polarization, 158, 159 M2 to M1 phenotype, 157, 160, 161 preventive/therapeutic setting, 160 tumour growth and progression, 161 Nitric oxide synthase (NOS), 78 Nitric oxide synthase 2 (NOS2), 56, 57 Notch pathway, 172

P Pathogen-associated molecular patterns (PAMPs), 120 Phosphatase and tensin homologue deleted on chromosome ten (PTEN), 137 Phosphatidylinositol (PI)–3-kinase (PI3K) pathway, 136–137 Plasmacytoid dendritic cells (pDCs), 95 Prokineticin–2, 110 Proteases role and ECM, 40–41

R Radiotherapy, 67 Rituximab, 67 Rovelizumab, 65

S SH2-containing inositol 5’-phosphatase (SHIP) biological roles of basophils, 144–145 killer and healer, 145–146 M1 vs. M2 macrophage skewing, 145–148 as positive regulator, 145 T-cell, 148 TLR2 and COX–2, 147 cancer, 136, 137 knockout mice, phenotype of, 139–140 mechanism of action, 141–142 phosphatidylinositol (PI)–3-kinase (PI3K) pathway, 136–137, 141–142 PI–3,4-P2, 137

Index protein levels, regulation of BCR-ABL, 143 miR–155, 144 protein expression pattern, 142–143 PTEN, 137 structure and binding partners forms of, 137, 138 SHIPE and SHIPG, 139 stem cell SHIP (sSHIP), 139 Signal transducer and activator of transcription 3 (STAT3), 54 Sildenafil, 87 Skin carcinogenesis, indoleamine 2, 3-dioxygenase, 95–96 Suppressor of cytokine signaling (SOCS) pathway, 172

T Therapeutic strategies, TAM adoptive transfer, 68–69 macrophage ablation, 63–64 monocyte recruitment inhibition CX3CL1, 65–66 leukocyte trafficking, CD11b and CD18, 65 re-education nanoparticles, 67, 68 phenotypic plasticity, 66 rituximab/radiotherapy, 67 TIR domain-containing adaptor inducing interferon beta (TRIF) signaling, 122 Tissue repair and TLR signaling ischemia–reperfusion, pathology, 126–127 models, 126 regeneration, 127 Toll-like receptors (TLR) signaling adaptive immune response, 123 in cancer and TAM tumourigenesis, 129 tumour regression, 127–128 description of, 120–121 endogenous ligands of extracellular matrix, 125 induction of, 124–125 intracellular-derived, 123–124 in host defense, 122–123 PAMPs, 120 pathways MyD88, 121–122 TRIF, 122 and tissue repair ischemia–reperfusion, pathology, 126–127

Index models, 126 regeneration, 127 TLR2, 147 Transcription regulation, 22–23 Tumour cell–macrophage interactions colony-stimulating factor–1 (CSF–1), 32, 33 MMTV-PyMT mouse, 32, 33 tumour cell motility, 33 tumour progression steps, 33 Tumour-derived soluble factors (TDSFs), 52, 54 Tumour-draining lymph nodes (TDLNs), 94, 95 Tumourigenesis, TLR, 129 Tumour-induced immunosuppression. See Myeloid cells Tumour microenvironment L-arginine metabolism arginase, 82–83 inducible nitric oxide synthase (iNOS), 83–84 M1 and M2 TAMs, 81 HIF-D, 167, 168 inflammatory and immune escape mechanisms in, 91–92 Tumour necrosis factor D (TNFD), 36–37 Tumour regression, TLR, 127–128 Tumour angiogenesis, 105, 106, 108 HIF-D, 174 regulation functional plasticity, 18 hypoxic signalling, 22–23 M2 macrophages, 18 murine tumour models, pro-angiogenic effect of, 17 pro-angiogenic functions, 19–20

187 TIE2-expressing monocyte/ macrophages, 23–24 transcription regulation, 22–23 tumour-derived signals, 21–22 tumour vasculature formation, 9, 20 Tumour-derived signals, 21–22 Tumour vasculature formation, 19, 20 Tyrosine kinase with Ig and EGF homology domain–2 (TIE2) receptor, 23–24

V Vascular endothelial growth factor (VEGF), 52 anti-angiogenic therapy, 108, 112 anti-VEGF treatment, 109 Bv8, 110 gene array analysis, 109 hypoxia, 107–108 M1 (classically activated)/M2 (alternatively activated) phenotypes, 106 monocytes, 107 myeloid cells, 110, 111 tumour angiogenesis, 105, 106, 108 VEGFR tyrosine kinases, 106 Vascular endothelial growth factor A (VEGF-A), 19, 20 Versican, 125

W Wnt signaling, 38–39

Z Zoledronic acid, 64