The Role and Mechanism of Action of CD200:CD200R1 Interaction in Breast Cancer

by

Anna Podnos

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Institute of Medical Science University of Toronto

© Copyright by Anna Podnos 2015

The Role and Mechanism of Action of CD200:CD200R Interaction in Breast Cancer

Anna Podnos

PhD

Institute of Medical Science University of Toronto

2015 Abstract

Cancer cells can use immune inhibitory receptors to evade the host’s anti-tumour responses and establish immunosuppressive networks in the tumour microenvironment. In this thesis, we investigated the interaction between the immunosuppressive molecule CD200 with its receptor, CD200R1, in breast cancer. We found that CD200 is expressed in the local tumour microenvironment in human breast cancer patients and developed a mouse model to study the effect of CD200 on tumour growth and metastasis. Using mouse-derived EMT6 breast cancer cells and BALB/c female hosts, we explored the effects of overexpressing and silencing the expression of CD200 and CD200R1 in hosts and tumour cells.

CD200 expression by host and tumour cells enhanced tumour growth and metastasis to draining lymph nodes (DLN). Silencing CD200 expression in EMT6 tumour cells led to a reduction in primary tumour size and metastasis, as well as an increase in anti-tumour cytotoxic responses in the host. Lack of CD200R1 expression in the host resulted in a marked decrease in breast cancer development and CD200R1-/- mice were able to mount specific anti-EMT6 immune response that could be adoptively transferred to wild type naïve hosts. In addition, we extended our findings to a model in which anti-tumour immunity was explored in EMT6 tumour-bearing

ii hosts lacking CD200 expression and treated with a combination of immunotherapy with the non- conventional chemotherapeutic agent, metformin. The findings suggest that CD200 may be an important prognostic marker and a target for breast cancer treatment that could synergize with other therapies and improve outcomes in patients.

iii

Acknowledgments

First and foremost, I would like to express my gratitude to my supervisor and mentor, Reg Gorczynski, for his unconditional support, patience, generosity, and guidance. His passion for science has fueled my interest in biomedical research, and his vivacious energy and sense of humour have motivated me and made my time in his laboratory an enjoyable experience that I will never forget.

I would also like to thank my colleagues and lab mates, who have supported me through the ups and downs of research over the last five years. I would especially like to thank our lab manager and a great teacher, Ismat Khatri, who taught me everything I know in the lab and has become a close personal friend and mentor. I am grateful for the encouragement and help from lab members Olha Kos and Fang Zhu, who have been reliable and supportive throughout this journey. Also, I thank Hassan Sadozai and Ramzi Khattar for their help with planning experiments and editing my thesis.

I thank my program advisory committee members, Shannon Dunn and David Spaner, for their helpful comments and suggestions that have made this work better and for their guidance in resolving lab related problems. I also thank the Institute of Medical Science, particularly Mingyao Li, Cindy Morshead, and the administrative staff, for helping me navigate complicated situations and complete my thesis. Finally, I am thankful to Gillian Einstein, who has been my mentor for the Collaborative Graduate Program in Women’s Health and has expanded my research interests during my PhD.

I am eternally grateful to my fiancé, Tom, for believing in me, supporting me through the hard times, celebrating with me in good times, and editing countless copies of my thesis. I also want to thank my parents, who have taught me to work hard and persevere, and my little brother David, who can always put a smile on my face. Last but not least, I would like to give a shout- out to my dogs, Chuba and Xena, for their unconditional love and slobbery kisses.

iv

Contributions

Dr. Reg Gorczynski1,2 assisted with limiting dilution assays, chromium release assays, and mouse injections. Dr. Ismat Khatri2 assisted with human and mouse sCD200 ELISAs and data analysis. Dr. Kai Yu2 helped with genotyping and maintaining animal colonies. Dr. Nuray

Erin3 assisted with the mixed lymphocyte culture assays. The Ontario Tumour Bank provided the human breast cancer plasma and serum samples as well as histology sections.

1 Institute of Medical Science, University of Toronto, Toronto, ON, Canada

2 University Health Network, Toronto General Hospital, Toronto, ON, Canada

3Department of Medical Pharmacology, School of Medicine, Akdeniz University, Antalya City, Antalya, Turkey

v

Table of Contents

Acknowledgments ...... iv

Contributions ...... v

Table of Contents ...... vi

List of Tables ...... x

List of Figures ...... xi

List of Abbreviations ...... xiv

List of Appendices ...... xix

Chapter 1 Introduction and literature overview ...... 1

1 Introduction and literature overview ...... 1

1.1 Cancer and the immune system ...... 1

1.1.1 Immunity and cancer ...... 1

1.1.2 Tumour microenvironment ...... 2

1.1.3 Non-immune cells in the tumour microenvironment ...... 5

1.1.4 Innate immunity in cancer ...... 7

1.1.5 Adaptive immunity in cancer ...... 11

1.1.6 Immune inhibitory receptors in the tumour microenvironment ...... 19

1.2 Metastasis and the immune system ...... 20

1.2.1 Metastasis is a hallmark of cancer ...... 20

1.2.2 Immunity and metastasis ...... 22

1.3 Breast cancer ...... 23

1.3.1 Clinical features of breast cancer ...... 23

1.3.2 Tumour microenvironment in breast cancer ...... 28

1.3.3 Animal models of breast cancer ...... 37

vi

1.4 CD200 and CD200R ...... 42

1.4.1 CD200:CD200R overview ...... 42

1.4.2 CD200:CD200R in cancer ...... 43

1.5 EMT6 breast cancer model ...... 47

1.6 Objectives and hypotheses ...... 50

Chapter 2 Evidence for a role of tumour CD200 expression in breast cancer metastasis: decreased metastasis in CD200R1-/- mice or using EMT6siCD200 breast cancer cells ...... 52

2 Studies in WT and CD200R1-/- mice with EMT6, EMT6CD200tg, and EMT6shCD200 tumour cells ...... 52

2.1 Abstract ...... 52

2.2 Introduction ...... 53

2.3 Materials and methods ...... 55

2.3.1 Mice ...... 55

2.3.2 Monoclonal antibodies ...... 55

2.3.3 EMT6 breast tumour cells, induction of tumour growth in BALB/c mice, and limiting dilution assays ...... 55

2.3.4 Production and use of lentiviral particles encoding shRNA specific for mouse CD200 ...... 56

2.3.5 Preparation of cells and cytotoxicity, proliferation, and cytokine assays ...... 57

2.3.6 Statistics ...... 58

2.4 Results ...... 58

2.4.1 Comparison of metastasis of EMT6 or EMT6CD200tg to DLN of wild-type (WT), CD200tg or CD200R1-/- mice ...... 58

2.4.2 EMT6CD200tg tumour promotes metastasis of CD200- tumour cells to DLN ...... 61

2.4.3 Metastasis of CD200- EMT6 enhanced by EMT6CD200 depends on host CD200R1 expression ...... 62

2.4.4 Decreased primary growth and absence of metastasis to DLN following injection of EMT6siCD200 in WT BALB/c is rescued by co-injection with EMT6CD200tg ...... 64

vii

2.4.5 Improved host immunity to EMT6 tumour cells in mice following injection of both EMT6siCD200 in WT mice and control EMT6 tumour cells in CD200R1-/- mice ...... 68

2.5 Discussion ...... 72

2.6 Tables ...... 76

Chapter 3 The role of CD200 expression by tumour and host cells in regulation of immunotherapy in the mouse EMT6 breast cancer model ...... 81

3 Studies in CD200-/- and WT hosts with EMT6 and EMT6siCD200 tumour cells ...... 81

3.1 Introduction ...... 81

3.2 Materials and methods ...... 83

3.2.1 Mice ...... 83

3.2.2 Tumour cells ...... 83

3.2.3 Growth of EMT6 cells in mice ...... 84

3.2.4 Tumour resection and immunization of mice ...... 84

3.2.5 Immunostaining and flow cytometry ...... 84

3.2.6 Antibodies ...... 85

3.2.7 Proteomics assay ...... 86

3.2.8 ELISArray ...... 86

3.2.9 Cytokine ELISAs ...... 86

3.3 Results ...... 86

3.3.1 Characterization of BALB/c CD200-/- mice ...... 86

3.3.2 Analysis of cell populations harvested from tumours in WT and CD200-/- BALB/c female mice ...... 88

3.3.3 Altering CD4 and CD8 T cell signaling in WT female mice bearing EMT6 or EMT6siCD200 tumours...... 91

3.3.4 Augmentation of adaptive immunity to EMT6 in mice receiving metformin ...... 93

3.4 Discussion ...... 99

Chapter 4 Soluble CD200 in plasma and serum of human breast cancer patients ...... 108

viii

4 Soluble CD200 levels in breast cancer patients ...... 108

4.1 Abstract ...... 108

4.2 Introduction ...... 109

4.3 Materials and methods ...... 111

4.3.1 Patient Samples ...... 111

4.3.2 Antibodies ...... 112

4.3.3 ELISA of soluble CD200 in plasma ...... 112

4.3.4 Immunological staining ...... 112

4.3.5 mRNA extraction from formalin fixed paraffin embedded (FFPE) tissue sections and qRT-PCR ...... 113

4.3.6 Statistics ...... 113

4.4 Results ...... 114

4.4.1 Levels of sCD200 in plasma of 30 breast cancer patients compared with 10 healthy age-matched women ...... 114

4.4.2 Levels of sCD200 in plasma and serum samples from 100 breast cancer patients correlated with clinical characteristics of cancer ...... 115

4.4.3 Comparison of mRNA expression and cell surface staining for CD200 and CD200R in breast tumour samples and adjacent normal tissue ...... 116

4.5 Discussion ...... 117

4.6 Tables ...... 124

Chapter 5 General Discussion and Future Directions ...... 126

References ...... 134

Appendices ...... 166

Bidirectional effect of CD200 on breast cancer development and metastasis, with ultimate outcome determined by tumour aggressiveness and a cancer-induced inflammatory response ...... 166

Cure of metastatic growth of EMT6 tumour cells in mice following manipulation of CD200:CD200R signaling ...... 168

ix

List of Tables

Table 2-1: Comparative frequency of CD200+ EMT6 tumour cell clones in DLN of mice from Fig. 2-1 (pg. 76)

Table 2-2: Comparative frequency of CD200+ EMT6 tumour clones in DLN of mice from Fig. 2- 2 (pg. 77)

Table 2-3: Comparative frequency of CD200+ EMT6 tumour clones in DLN of mice from Fig. 2- 3 (pg. 78)

Table 2-4: Comparative frequency of CD200+ EMT6 tumour clones in DLN of mice from Fig. 2- 4 (pg. 79)

Table 2-5: Comparative frequency of CD200+ EMT6 tumour clones in DLN of mice from Fig. 2- 5 (pg. 80)

Table 4-1: Demographic data from OTB samples (pg. 124)

Table 4-2: Drug use in control volunteers (pg. 125)

x

List of Figures

Figure 1-1: Cells in the tumour microenvironment (pg. 4)

Figure 1-2: Anti-tumour immunity vs. tumour-induced immunosuppression (pg. 18)

Figure 1-3: The anatomy of a breast (pg. 26)

Figure 1-4: Cytokines secreted by cells in the tumour microenvironment regulate breast cancer stem cell renewal (pg. 37)

Figure 1-5: CD200 in the tumour microenvironment (pg. 46)

Figure 1-6: CD200 expression on the surface of EMT6 cell (pg. 48)

Figure 1-7: Predicted outcomes of EMT6 breast tumour growth and metastasis, according to the CD200:CD200R1 model (pg. 51)

Figure 2-1: Primary tumour growth and limiting dilution analysis of frequency of cloneable EMT6 cells in DLN of WT, CD200tg, or CD200R1-/- mice receiving either control EMT6 or EMT6CD200tg tumour cells (pg. 60)

Figure 2-2: Comparison of primary tumour mass and frequency of tumour cells cloned by limiting dilution from DLN of WT (panel b) or CD200tg BALB/c mice at 18 days post transplantation of EMT6 tumour cells (pg. 63)

Figure 2-3: Differences in primary tumour growth and DLN metastasis of mixtures of control or EMT6CD200tg tumour cells injected into WT or CD200R1-/- mice (pg. 65)

Figure 2-4: Decreased local growth and frequency of metastasis of EMT6siCD200 tumour cells in WT mice is attenuated when cells are injected instead into CD200tg recipients (pg. 67)

Figure 2-5: Attenuated local growth and metastasis of EMT6siCD200 cells to DLN in recipient WT mice is rescued by co-injection of EMT6siCD200 cells with tumour cells themselves over- expressing CD200 (EMT6CD200tg) (pg. 69)

xi

Figure 2-6: Adaptive immune response in CD200R1-/- tumour bearing hosts can be transferred to naïve WT mice (pg. 71)

Figure 2-7: Importance of CD8+ T cells in specific lysis of EMT6 tumour cells (30:1 effector:target) using DLN cells harvested at 15 days post tumour inoculation into WT or CD200R1-/- mice (pg. 73)

Figure 3-1: Characterization of CD200-/- mice in the context of EMT6 breast cancer model (pg. 87)

Figure 3-2: EMT6 and EMT6siD200 tumour microenvironment in WT and CD200-/- female mice (pg. 90)

Figure 3-3: Depleting CD8+ T cells in EMT6 and EMT6siCD200 tumour bearing WT hosts (pg. 92)

Figure 3-4: Depleting CD4+ T cells in EMT6 and EMT6siCD200 tumour bearing WT hosts (pg. 94)

Figure 3-5: Metformin augments anti-tumour immunity in CD200R1-/- EMT6 tumour bearing hosts (pg. 96)

Figure 3-6: Verteporfin negates the effect of metformin on EMT6 tumour growth and metastasis in CD200-/- mice (pg 98)

Figure 3-7: Cytokines in the EMT6 and EMT6siCD200 tumour microenvironment (pg. 100)

Figure 3-8: In vivo CD4 cell depletion in WT EMT6 tumour bearing mice (pg. 102)

Figure 3-9: Metformin does not improve anti-tumour immunity in WT EMT6 tumour bearing hosts (pg. 103)

Figure 4-1: Soluble CD200 levels in plasma of 30 breast cancer patients compared with 10 age- matched controls (pg. 115)

Figure 4-2: Soluble CD200 levels in plasma and serum of 100 breast cancer patients compared with normal plasma (pg. 117)

xii

Figure 4-3: Correlation of sCD200 levels with clinical characteristics of breast cancer in 100 patient samples (pg. 118)

Figure 4-4: Correlation of circulating sCD200 levels with ER, PR, and HER2/neu status in 100 breast cancer patients (pg. 120)

Figure 4-5: Relative CD200 expression in paired tumour and healthy FFPE tissues from breast cancer patients (pg. 122)

xiii

List of Abbreviations

ADCC Antibody-dependent cellular cytotoxicity

AKT Protein kinase B

ALDH1 Aldehyde dehydrogenase

ALL Acute lymphoid leukemia

AML Acute myeloid leukemia

B7-H1 B7-homologue 1 (PD-L1 or CD274)

BCR B cell receptor

BCSC Breast cancer stem cell

BRCA1/2 Breast cancer susceptibility gene ½

BTLA B and T lymphocyte attenuator (CD272)

CAF Cancer associated fibroblasts

CD Cluster of differentiation

C/EBP CCAAT-enhancer binding proteins

CIA Collagen-induced arthritis

CLL Chronic lymphocytic leukemia

CNS Central nervous system

CSC Cancer stem cells

CSF Colony-stimulating factor

CTL Cytotoxic T lymphocyte

CTLA-4 Cytotoxic T lymphocyte antigen 4 (CD152)

CXCL C-X-C motif chemokine ligand

CXCR C-X-C motif chemokine receptor

DC Dendritic cell

DCIS Ductal carcinoma in situ xiv

EAE Experimental autoimmune encephalomyelitis

ECM Extracellular matrix

EGF Epidermal growth factor

ELISA Enzyme-linked immunosorbent assay

EMT Epithelial to mesenchymal transition

ER Estrogen receptor

ERK Extracellular ligand-regulated kinase

FACS Fluorescence-activated cell sorting

FasL Fas ligand

FOXP3 Forkhead/winged helix box protein P3

G418 Geneticin

G-CSF Granulocyte colony stimulating factor

GAPDH Glyceraldehyde 3-phosphate dehydrogenase

GH Growth hormone

GITR Glucocorticoid-induced TNFR family related gene

GM-CSF Granulocyte macrophage colony stimulating factor

HLA Human leukocyte antigen

HSC Hematopoietic stem cell

HER2/neu Human epidermal growth factor receptor 2

HSPG Heparin sulfate proteoglycan

HVEM Herpesvirus entry mediator

IDC Invasive ductal carcinoma

IDO Indoleamine-pyrrole 2,3-dioxygenase

IFN Interferon

IGF Insulin-like growth factor

xv

IL Interleukin iNOS Inducible nitric oxide synthase

IRF Interferon regulatory factor

ITAM Immunoreceptor tyrosine-based activation motif

ITIM Immunoreceptor tyrosine-based inhibitory motif iv Intravenous

Jak Janus kinase

LCIS Lobular carcinoma in situ

LPS Lipopolysaccharide

M-CSF Macrophage colony stimulating factor

MAPK Mitogen-activated protein kinase

MCP Monocyte chemoattractant protein

MDSC Myeloid derived suppressor cells

MHC Major histocompatibility complex

MIP Macrophage inflammatory protein

MLC Mixed lymphocyte culture

MMP Matrix metalloproteinase

MMTV Mouse mammary tumour virus

MS Multiple sclerosis

MSC Mesenchymal stem cells

NFAT Nuclear factor of activated T cells

NFκB Nuclear factor κB

NK Natural killer

NKG2D Natural killer group 2 member D

NOD Non-obese diabetic

xvi

PBL Peripheral blood leukocytes pCR Pathologic complete response rate

PCR Polymerase chain reaction

PD-1 Programmed cell death 1 pDC Plasmacytoid dendritic cell

PDGF Platelet-derived growth factor

PI3K Phosphatidylinositol 3-kinase

PKC Protein kinase C

PR Progesterone receptor

RA Rheumatoid arthritis

RANTES Regulated on activation, normal T cell expressed and secreted

RNS Reactive nitrogen species

ROS Reactive oxygen species

SCID Severe combined immunodeficiency

SDF-1 Stromal cell-derived factor 1 siRNA Small interfering RNA shRNA Short hairpin RNA

SMA Smooth muscle actin

STAT Signal transducer and activator of transcription

TAA Tumour associated antigen

TAM Tumour associated macrophages

TCR T cell receptor

TGF Transforming growth factor

TIL Tumour infiltrating lymphocytes

TLR Toll-like receptor

xvii

TIMP Tissue inhibitor of metalloproteinases

TNBC Triple-negative breast cancer

TNFα Tumour necrosis factor α

VCAM Vascular cell adhesion molecule

VEGF Vascular endothelial growth factor

xviii

List of Appendices

Bidirectional effect of CD200 on breast cancer development and metastasis, with ultimate outcome determined by tumour aggressiveness and a cancer-induced inflammatory response (pg. 166)

Cure of metastatic growth of EMT6 tumour cells in mice following manipulation of CD200:CD200R signaling (pg. 168)

xix 1

Chapter 1 Introduction and literature overview

This thesis describes studies investigating the role of CD200, an immunoregulatory molecule, and its receptor, CD200R, in breast cancer. Topics in cancer immunity relevant to this thesis are discussed in section 1.1. A summary of important concepts in metastasis is presented in section 1.2. An overview of breast cancer, including clinical features, key components in the breast tumour microenvironment, and animal models of breast cancer, are considered in section

1.3. Literature on CD200, CD200R, and the role of their interaction in cancer immunology are reviewed in section 1.4. Section 1.5 provides an overview of the EMT6 mouse breast cancer model, which is used in the studies presented in the following chapters. Lastly, the objectives and hypotheses of this study are discussed in section 1.6.

1 Introduction and literature overview

1.1 Cancer and the immune system

1.1.1 Immunity and cancer

A healthy immune system is important for controlling cancer, and immune suppression associated with malignancies contributes to their progression1. Tumour infiltrating immune cells are found in most if not all neoplastic lesions, and these inflammatory cells can have both cancer-antagonistic and cancer-promoting effects2. It is thought that only a minority of all cancers are caused by germline mutations, whereas 90% are linked to somatic mutations and environmental factors3. Many environmental causes and risk factors of cancer are associated with chronic inflammation. Up to 20% of all cancers have been linked to chronic infections,

2 30% are attributed to tobacco smoking and inhaled pollutants (such as silica and asbestos), and

35% are thought to be related to dietary factors (20% of cancer burden is linked to obesity)4.

The existence of a functional relationship between inflammation and cancer was first proposed by Virchow in the 19th century, when he observed that cancer often arose at sites of chronic inflammation5, 6. The connection was further supported by findings that tumours resemble wounds that never heal7. During normal wound healing, immune inflammatory cells appear transiently and then disappear. Conversely, inflammatory cells persist at sites of chronic inflammation, and their presence has been associated in some instances with various tissue pathologies, including fibrosis, aberrant angiogenesis, and neoplasia2, 3, 8.

At least two pathways are thought to link inflammation and cancer. In the intrinsic pathway, activation of different classes of oncogenes drives the expression of inflammation- related programs, which guide the creation of an inflammatory microenvironment. This, in turn, drives oncogenesis, for example by releasing chemicals that are actively mutagenic (e.g., ROS)2.

In the extrinsic pathway, inflammatory conditions are hypothesized to promote cancer development (e.g., colitis-associated cancer of the intestine, cervical carcinoma with HPV, liver cancer with hepatitis B and C viruses)1, 9. Inflammation can contribute to many functional hallmarks of cancer through growth factors that sustain proliferative signaling, survival factors that limit cell death, and extracellular matrix-modifying enzymes that facilitate angiogenesis, invasion, and metastasis2, 3.

1.1.2 Tumour microenvironment

Solid tumours can be viewed as organ-like structures composed of various cell types, whose interaction is required to drive and promote growth and metastasis (Fig. 1-1). Neoplasia- associated angiogenesis and lymphangiogenesis help produce a chaotic vascular organization of

3 blood vessels and lymphatics, forming the tumour microenvironment5, which contains immune cells, cancer cells, and their surrounding stroma. The inflammatory component of a developing neoplasm may include a diverse leukocyte population—innate immune cells, like neutrophils,

NK cells, macrophages, eosinophils, mast cells, as well as adaptive immune cells, including B cells, T cells, and dendritic cells (DC)5. All of these cells are capable of producing an array of cytokines, chemokines, cytotoxic mediators, such as reactive oxygen species (ROS), serine and cysteine proteases, matrix metalloproteinases (MMPs) and membrane-perforating agents, as well as soluble mediators of cell killing, such as TNFα, interleukins and interferons. The expression of various immune modulators and the abundance and activation state of different cell types in the tumour microenvironment affects the balance between inflammation-promoted tumour growth and anti-tumour immunity3. On one hand, the immune system specifically detects and targets infectious agents with the adaptive immune response (B and T cells), which is supported by cells of the innate system. On the other hand, the innate system is involved in wound healing and removal of dead cells and cellular debris. Cells of the innate immune system are one of the major sources of angiogenic, epithelial, and stromal growth factors and matrix- remodeling enzymes, and these cells may be recruited and subverted to support neoplastic progression. Similarly, subclasses of B and T lymphocytes may facilitate the recruitment, activation, and persistence of wound-healing and tumour-promoting innate immune cells

(neutrophils, macrophages)2.

In addition, and perhaps just as important, there are physical barriers that limit the access of the immune cells to the tumour cells10. These include a dense extracellular matrix (ECM), high interstitial fluid pressure that is caused by a lack of lymphatic drainage, and molecular barriers that are expressed by endothelial cells, including regulator of G protein signaling 5

(RGS5) and endothelin B receptor (ETBR)11-13.

4

Figure 1-1 (Top) Cells in the tumour microenvironment. (Bottom) Cancer metastasis is a complex process that involves separation of tumour cells from the primary tumour, invasion through surrounding tissues and basement membranes, entry and survival in circulation, lymphatics, or peritoneal space, and arrest in a distant target organ. Adapted from Hanahan & Weinberg, 20112.

5 1.1.3 Non-immune cells in the tumour microenvironment

1.1.3.1 Cancer associated fibroblasts

Fibroblasts are found across the spectrum of carcinomas, and often are the preponderant cell population of the tumour stroma2. Some cancer associated fibroblasts (CAFs) are similar to the fibroblasts that create the structural foundation supporting most normal epithelial tissues.

Other CAFs are myofibroblasts, which express α-smooth muscle actin (SMA), and whose biological roles and properties differ markedly from those of tissue derived fibroblasts.

Myofibroblasts transiently increase in abundance in wounds and are also found at sites of chronic inflammation, contributing to the pathological fibrosis observed in tissues such as lung, kidney, and liver2.

Recruited myofibroblasts and reprogrammed variants of normal tissue-derived fibroblastic cells can enhance cancer cell proliferation, angiogenesis, and metastasis14, 15. CAFs secrete a variety of extracellular matrix components, which function in the formation of desmoplastic stroma that characterizes many advanced carcinomas2, 16. For example, CAFs secrete elevated levels of the cytokine SDF-1 (also known as CXCL12) that stimulates carcinoma cell proliferation in vivo, acting through the CXCR4 receptor expressed on the surface of carcinoma cells17, 18. CAFs also secrete high levels of TGF-β1, PDGF, and VEGF1 in many cancers, including head, neck, cervical, and prostate carcinomas19 (Fig. 1-2). These soluble mediators modulate an extensive array of cellular functions that can promote tumourigenesis.

6 1.1.3.2 Endothelial cells

Endothelial cells may play an important role in the tumour microenvironment through direct interaction with the tumour cells as well as through their role in the formation of tumour associated vasculature. The angiogenic switch is a process that activates quiescent endothelial cells, causing them to form new blood vessels20, 21. Tumour associated angiogenesis has been found to depend on several signaling pathways involving ligands of signal-transducing receptors expressed by endothelial cells (e.g., VEGF, angiopoietin, FGF, Notch)2, 20, 21 (Fig. 1-2). The gene expression profiles of tumour associated endothelial cells are distinct from normal endothelial cells22, 23. The initial angiogenic switch during tumour development is followed by neovascularization of variable intensity, which is controlled by both the cancer cells and the associated stromal microenvironment24, 25.

1.1.3.3 Mesenchymal cells

Mesenchymal stem and progenitor cells have been found to migrate into tumours from the marrow, where they may differentiate into various characterized stromal cell types, including osteoblasts, chondrocytes, adipocytes, fibroblasts, and myocytes26. Some of these cells may persist in an undifferentiated (or partially differentiated) state, retaining functions that their more differentiated progeny lack2. For example, MSCs present in the tumour stroma can secrete

CCL5/RANTES in response to signals released by cancer cells. CCL5 then acts reciprocally on the cancer cells to stimulate invasive behaviour27.

Pericytes are a specialized mesenchymal cell type (related to smooth muscle cells), with finger-like projections that wrap around the endothelial tubing of blood vessels. In normal tissues they provide paracrine support signals to the normally quiescent endothelium and

7 collaborate with endothelial cells to synthesize the vascular membrane that helps vessel walls to withstand the hydrostatic pressure of blood flow2, 28. Pericytes play a role in supporting the tumour endothelium28-30. For example, tumour vessels with less pericyte coverage appear more vulnerable to radiation and chemotherapy, suggesting that pericytes are critical to protect ECs and may promote therapeutic resistance31, 32. Inhibition of platelet-derived growth factor

(PDGF) receptor signaling by tumour associated pericytes results in reduced coating of tumour vessels and destabilization of vascular integrity and function29, 30, 33. This is thought to enhance epithelial-to-mesenchymal transition (EMT, see below), cancer cell intravasation into the circulatory system, and enable cancer cell dissemination2, 34. In addition, it was recently reported that in glioblastoma, glial stem cells can give rise to vascular pericytes that directly support neoangiogenesis, vessel function, and tumour growth35.

Pericyte and CAF progenitors were found to originate from the bone marrow in various mouse models of cancer26, 36, 37. Tumour-associated stromal cells in growing tumours may be derived from proliferation of pre-existing stromal cells; by differentiation in situ of local stem/progenitor cells originating in the neighboring normal tissue; or through recruitment of bone marrow-derived stem/progenitor cells2.

1.1.4 Innate immunity in cancer

According to the immune surveillance theory, innate immunity is responsible for early detection and elimination of foreign antigens and cancer cells. Innate immunity may be inefficient in patients who develop malignancy1.

8 1.1.4.1 Tumour associated macrophages

Circulating monocytes give rise to a variety of tissue-resident macrophages throughout the body, as well as to specialized cells such as dendritic cells (DCs) and osteoclasts. Monocytes originate in the bone marrow from a common myeloid progenitor (CMP) that is shared with neutrophils, and they are then released into the peripheral blood, where they circulate for several days before entering tissues and replenishing the tissue macrophage populations38, 39.

Macrophages are phagocytic cells that function in pathogen destruction, inflammation, and tissue repair. Macrophage differentiation results in mature cells polarized toward the M1 or M2 phenotype, which differ in terms of receptor expression, effector function, and cytokine and chemokine production40. For instance, arginine metabolism is characterized by high levels of inducible nitric oxide synthase (iNOS) in M1 macrophages, whereas the arginase pathway predominates in M2 cells with generation of ornithine and polyamines40. Furthermore, cells of the M1 phenotype produce cytokines like IL-12, IL-23, and tumour necrosis factor (TNF), whereas M2 macrophages typically produce IL-10, IL-1 receptor antagonist (IL-1ra) and the type II IL-1 decoy receptor (see Fig. 1-2). Chemokine receptors and ligands are differentially modulated in polarized macrophages. Differential production of chemokines that attract Th1

(e.g. CXCL9, CXCL10) and Th2 or regulatory T (Treg) cells (e.g. CCL22) integrates M1 and M2 macrophages into circuits of amplification and regulation of polarized T-cell responses40, 41.

Classically activated M1 macrophages are potent effector cells that kill microorganisms and tumour cells and produce copious amounts of proinflammatory cytokines. In contrast, M2 cells

(induced by IL-10, IL-4 or IL-13, glucocorticoid hormones, vitamin D3) are hypothesized to modulate inflammatory responses and adaptive Th1 immunity, scavenge cellular debris, and promote angiogenesis, tissue remodeling and repair40.

9 Tumour associated macrophages (TAMs) are a significant component of the inflammatory infiltrate in neoplastic tissues and are likely derived from a blood-borne macrophage precursor42. Macrophages have the potential to express pro- and anti-tumour activity, the former prevailing in established neoplasia40. Clinical studies have found that high numbers of intra-tumour macrophages correlate with high vessel density and tumour progression43. TAMs and related cell types in mouse and human tumours generally have an M2 phenotype, which is thought to be oriented towards suppressing adaptive immunity, promoting tumour growth, remodeling tissues, and promoting angiogenesis 40, 44, 45 (Fig. 1-2). TAMs are required to regulate both the angiogenic switch and tumour metastasis in human cancers and in mouse tumour models46, 47. For example, VEGF-C production by TAMs in human cervical tumours regulated peri-tumoural lymphangiogenesis and dissemination of cancer cells to lymph nodes40.

Macrophages and monocytes are recruited to tumours via chemotactic signaling.

Expression of the chemokine CCL2/MCP-1 is frequently observed in tumours (sarcomas, gliomas, melanomas, lung, breast, cervix, and ovary tumours) in keeping with its description as a tumour-derived chemotactic factor48, 49. CCL2 contributes to regulation of macrophage recruitment to tumours in mouse models of cancer and human malignancies40, 49. Other CC chemokines related to CCL2, such as CCL7 and CCL8, are also produced by tumours and can recruit monocytes43, 50. Vascular endothelial growth factor (VEGF) and macrophage-colony stimulating factor (M-CSF) also contribute to macrophage recruitment in tumours and promote their migration, proliferation, and survival43, 51-53 (Fig. 1-2).

10 1.1.4.2 Tumour associated neutrophils

Neutrophils defend the host from invading microorganisms and assist in wound healing54. Invading pathogens elicit an inflammatory response that recruits neutrophils to sites of infection, where they engulf and eliminate microorganisms using an arsenal of cytotoxic substances, proteinases, cytokines, and chemokines 54,55,56. The process of neutrophil recruitment and activation, observed in infection, is recapitulated within the tumour microenvironment. However, accumulating evidence suggests that, within the tumour microenvironment, neutrophils act to the detriment of the host.

Neutrophils are the most abundant circulating leukocyte in humans and make up a significant portion of the inflammatory cell infiltrate found in a wide variety of murine cancer models and human cancers57. Clinical studies have indicated that the presence of tumour associated neutrophils (TANs) confers a poor prognosis57. For example, in patients with renal cell carcinoma, an increased number of neutrophils correlated with increased mortality58. In addition, increased levels of neutrophils in the bronchioalveolar space of patients with bronchioalveolar carcinoma were associated with poor outcomes59. Levels of IL-8, an important chemotactic molecule for neutrophils and macrophages, have been associated with neutrophil accumulation and reduced patient survival60.

Many cell types within the tumour microenvironment are capable of secreting neutrophil chemotactic substances. However, the tumour cells themselves may mediate neutrophil recruitment to sites of tumourigenesis by secreting CXC chemokines (e.g., interleukin-8 (IL-8)), suggesting that TANs are not a means of host defense. Studies report concrete examples of tumour-mediated signals eliciting pro-tumour responses from neutrophils (e.g., in mouse breast cancer and hepatocellular carcinoma)57. Soluble factors produced by TANs include ROS, which

11 can initiate DNA damage and tumour establishment, nitric oxide, proteinases (important for

ECM remodeling), elastase, and matrix metalloproneinases57.

1.1.5 Adaptive immunity in cancer

When mice genetically engineered to be deficient for various components of the innate and adaptive immune system were assessed for the development of carcinogen-induced tumours, it was observed that tumours arose more frequently and grew more rapidly in immunodeficient mice relative to immunocompetent controls2. In particular, deficiencies in the

+ + development or function of CD8 cytotoxic T lymphocytes (CTLs), CD4 Th1 helper T cells, or

NK cells each led to demonstrable increases in tumour incidence. Moreover, mice with combined immunodeficiencies in both T cells and NK cells were even more susceptible to cancer development. These results suggested that, at least in certain experimental models, both the innate and adaptive cellular arms of the immune system were able to contribute significantly to immune surveillance and thus tumour eradication2, 61, 62. Anti-tumour immune responses are often reliant on the immunogenicity of a tumour, which varies greatly between cancer types and individuals10.

Clinical epidemiology also increasingly supports the hypothesis that an anti-tumour immune response exists in various types of human cancer63-65. For example, patients with colon and ovarian tumours that are heavily infiltrated with CTLs and NK cells often have a better prognosis than those that lack such abundant killer lymphocytes65, 66. Additionally, immunosuppressed organ transplant recipients have been observed to develop donor-derived cancers, consistent with the idea that in the ostensibly tumour-free donors, the cancer cells were held in check, in a dormant state, by a fully functional immune system2, 67 (Fig. 1-2).

12 1.1.5.1 Natural killer cells

Natural killer (NK) cells mediate innate immunity against pathogens and tumours. They were originally discovered because of their ability to kill certain tumour cells in vitro68. Studies in leukemia, lymphoma, and gastrointestinal stromal tumours support the idea that NK activation and cytotoxicity influence patient outcome69-71. Low NK cell activity has been reported in familial breast cancer patients as well as their clinically asymptomatic first degree relatives, suggesting that NK activity has an important hereditary and genetic component72.

Depletion of NK cells in vivo leads to enhanced tumour formation in mouse tumour models73.

Although several NK cell receptors have been implicated in the killing of tumours74, the activating NKG2D receptor is unique because it recognizes defined antigens (e.g. stress-induced proteins) that are frequently overexpressed on many different tumours, indicating a role for this receptor in immune surveillance against cancer.

Tumours have developed a number of strategies that might allow them to escape effective NK suveillance. Tumours can shed soluble ligands for activation receptors, secrete immunosuppressive cytokines, such as TGFβ and IL-10, which could impair NK cell or T cell effector functions, and produce the apoptotic Fas ligand, which can also lead to elimination of

NK cells. Furthermore, not all cells within a tumour mass might express NKG2D ligands, and expression of an NKG2D ligand alone on tumours might be insufficient to activate NK cells.

Finally, NK cells might not traffic to the primary tumour site, since histological examination of tumours rarely indicates the presence of NK cells68 (Fig. 1-2).

13 1.1.5.2 T cells

T cells recognize peptides that are presented by human leukocyte antigen (HLA, also known as major histocompatibility complex (MHC)) molecules10. Tumour specific antigens recognized by antibodies and T cells have been identified in cancer patients, yet their existence often does not control malignant growth45. T lymphocyte precursors capable of responding to self-antigens, which are often overexpressed in tumours, are detectable at low levels in the circulation of most individuals1 and are enriched in tumour tissues75. Despite the TIL activation phenotype, they are functionally compromised and are enriched in TAA-specific memory T cells1.

There are many cellular and molecular mechanisms that mediate tumour escape from natural immune surveillance76-78. Studies of melanoma antigen (MLANA)-specific T cells confirmed that T cells specific for tumour antigen are rendered functionally tolerant once they are present in the tumour microenvironment79, 80. Furthermore, these studies suggested that this locally induced tolerance may be reversible81. Reduced expression of the major histocompatibility complex (MHC) on the cancer cell membrane and impaired antigenic peptide expression on the tumour cell surface are examples of mechanisms that can attenuate direct recognition of tumour antigens by T cells and direct priming of an immune response by a tumour82. As a result, tumours can evade the host’s immune response by being poor stimulators of T cells and by being poor targets for tumour specific CTLs1.

Tumours also directly interfere with the host immune system through release of factors that modulate the functions of immune cells or induce of apoptosis of these cells1, 82. In addition to the wide variety of soluble immunosuppressive factors (TGFβ, IL-10, ROS, enzymes, and inhibitory ligands, such as FasL or TRAIL) that are released by tumour cells or other cells in the

14 tumour microenvironment, immune suppressor cell populations have been shown to play a key role in down-regulation of anti-tumour host immunity83, 84. Conversion of naïve T cells into adaptive Treg cells is facilitated by direct release from tumours of TGF-β, IL-10, and indoleamine 2,3-dioxygenase (IDO), an immunosuppressive enzyme that depletes tryptophan, or through a tumour derived activation of release of such molecules from myeloid-derived suppressor cells (MDSC), TAMs and/or dendritic cells (DC)85.

Tregs are a population of cells that play a critical role in the induction and maintenance of immunologic tolerance. Treg populations are heterogeneous and differ in phenotype, cytokine secretion profile, and suppressive mechanism. Subsets include Tregs that depend on cytokine

+ + + production (IL-10 or TGFβ) for their function; CD4 CD25 FOXP3 Tregs (naturally occurring or

+ antigen-induced); and various other regulatory cells, including NKT cells, CD8 Treg, γδTCR+ cells, and DN T cells.

These cells can suppress tumour-specific T cell immunity, hamper NK cell activation and cytotoxicity, and inhibit maturation of DC, thus contributing to the progression of human

86, 87 and mouse tumours . Depletion of Tregs in mice with anti-IL-2R antibody or with low dose cyclophosphamide improves T cell-mediated tumour clearance, and depletion of CD4+CD25+ T

88, 89, 90 cells has been shown to promote tumour rejection in mice . In humans, Tregs can be defined as FOXP3+, CTLA4+, GITR+ T-cell subsets, and they are enriched among TIL in human tumours and are more abundant in the peripheral circulation of patients with cancer than

91 of healthy controls . Tregs accumulate in both human and mouse tumours, as well as in secondary lymphoid organs92, and are recruited and expanded by either the proliferation of pre-

93 85, 94 existing Tregs or the conversion of CD25-negative T cells . In some tumour types, including breast cancer and hepatocellular carcinoma, increased numbers of Treg cells correlate with reduced overall survival95, 96, whereas in other types, such as colorectal cancer and some

15

97, 98 hematological malignancies, Treg cells are associated with improved survival . These conflicting results may be due to the fact that in many of these studies, the function of

FOXP3+ cells was not tested, and not all FOXP3+ T cells are functionally suppressive98. It is

+ possible that the effects of Tregs on the adaptive immunity or the differentiation of FOXP3 cells into functional Tregs differs in various cancers. Alternatively, it is possible that the appropriate T cell compartment was not studied in the context of lymphoid malignancies. The utility of

FOXP3 content and Treg function as prognostic factors in different cancers is an area of intense investigation, but much remains to be understood98. Moreover, the controversial relationship between Treg populations and patient prognosis may be further impacted by other inflammatory cell subsets in the tumour microenvironment99.

TGF-β, a cytokine produced by Tregs, has been implicated in tumour progression. The

TGF-β pathway has been linked to metastatic processes and has been shown to dramatically

15, 100, 101 impact the ability of tumour cells to spread throughout the body . Tregs may express cell surface molecules that deliver negative signals to DC, including cytotoxic T lymphocyte antigen

4 (CTLA4) and lymphocyte activation gene 3 (LAG3). These signals inhibit the maturation of dendritic cells, block their expression of MHC and co-stimulatory molecules (CD80 and

CD86)102, activate their ability to produce IDO, and indirectly suppress genes encoding IL-6 and

TNF103. In addition, the release of adenosine and the secretion of TGF-β, IL-10, and IL-35 by

102, 104 Tregs may interfere with the activation and effector functions of T cells . Finally, secretion of granzymes and perforin might have cytolytic effects on target T cells DC82 (Fig. 1-2).

1.1.5.3 B cells

B cells can mediate tumour immunity by secreting pro-tumourigenic cytokines and

105 altering Th1-to-Th2 ratios . Their importance in supporting tumour growth is evident in B cell–

16 deficient mice, which exhibit resistance to engraftment of certain syngeneic tumours106.

Antibodies to tumour associated antigens (TAA) are often detectable in patients with cancer107 and have been used as a biomarker of prognosis, as is the case with, for example, antibodies to p53 in a subgroup of patients with head and neck cancers108.

Regulatory B cells (Bregs) have been shown to play a role in tumour progression to

109 metastasis . Similar to Tregs, the regulatory function of B cells is exerted via the production of regulatory cytokines, such as IL-10 and TGF-β, and the ability to express inhibitory molecules that suppress pathogenic T cells and autoreactive B cells in a cell-to-cell contact-dependent

110 + manner . Tumour-evoked Bregs induce the TGF-β-dependent conversion of resting CD4 T

+ 111 cells to immune-suppressive FOXP3 Tregs . Tumour-promoting B cells have been shown to facilitate the conversion of M1 macrophages to a tumour-promoting M2 phenotype through IL-

10 secretion112. Additionally, B cells promote lymphangiogenesis and may actively promote metastasis, as has been shown for B cell-mediated lymphangiogenic metastasis in lymphoma and melanoma113, 114 (Fig. 1-2).

1.1.5.4 Dendritic cells

Dendritic cells (DCs) are monocytic antigen-presenting cells that are derived from the bone marrow. DCs have been detected in several tumour types, including breast, lung, prostate, kidney, and ovarian carcinomas and melanoma115-117. Several subtypes of DC have been identified to infiltrate tumours, including Langerhan-like DC (positive for the marker Langerin), plasmacytoid DC (pDC), and myeloid-derived DC43. DC can take up antigens and cross-present them to T cells, modulating their activity. However, local DC maturation is often blocked in the tumours by combinations of tumour-derived cytokines, such as M-CSF and IL-6, resulting in immunosuppresion.

17 An aberrant balance between immature and mature myeloid cells is a hallmark of cancer118. Immature myeloid cells with specific inhibitory activities, called myeloid-derived suppressor cells (MDSCs), negatively regulate the immune response by suppressing CTL and

NK cell activity119. Although initially described in cancer patients, MDSCs are also present in other inflammatory settings, including solid organ transplantation, where tolerance is thought to be dependent, in part, on MDSCs that accumulate in the allografts120, 121. In preclinical mouse models, the phenotype of MDSCs consists of co-expression of the myeloid lineage differentiation antigens Gr-1 (Ly6G) and CD11b (CR3, Mac-1)122, 123. In contrast to murine models, the phenotype of MDSCs in humans is not as well defined119. Identification of human

MDSCs has been complicated by the lack of a specific marker and by the absence of a human homologue of mouse Gr-1124. Typically, human MDSCs are positive for the markers CD33 and

CD11b but express low levels of HLA-DR125. Accumulation of MDSCs in both preclinical models and in human samples was associated with defective dendritic cell function and inhibition of antigen specific T cell responses126-128.

Several mechanisms have been described by which MDSCs suppress T cell responses.

One is the depletion of specific amino acids such as l-arginine (l-Arg), l-cysteine, or l- phenylalanine124, which can inhibit expression of the CD3ζ chain and, thereby, T cell

129 proliferation, causing arrest of T cells in the G0–G1 phase of the cell cycle . The production of

ROS, such as H2O2, is another mechanism that affects immune regulation by inhibiting T cell proliferation124. Furthermore, reactive nitrogen species (RNS) are produced when superoxide anion interacts with NO130 and nitrate aromatic amino acids (such as tyrosine residues) in the T cell receptor (TCR) and CD8, resulting in a decreased recognition of peptide–MHC (major histocompatibility complex) by the TCR131. MDSCs can also secrete immune inhibitory cytokines that inhibit the activation of CD4+132 and CD8+123, 126 T cells, attenuate the

18 cytotoxicity of NK cells133, and polarize immunity toward a tumour-promoting type 2 phenotype through downregulation of IFN-γ and upregulation of IL-10125, 134 (Fig. 1-2).

Figure 1-2 Anti-tumour immunity vs. tumour-induced immunosuppression. The balance between anti-tumour immunity and tumour-mediated immune suppression is affected by the abundance and activation state of many different cell types in the tumour microenvironment. Tumour cell–specific antigens can be recognized by cytotoxic immune cells, leading to their destruction. Fibroblasts and macrophages within the tumour microenvironment contribute to a growth-suppressive state; however, these cells may later become “educated” by the tumour to acquire pro-tumourigenic functions. TAMs support tumour growth, angiogenesis, and invasion by secreting a pro-tumourigenic proteases, cytokines and growth factors (for example, EGF, which participates in a paracrine signaling loop through tumour-secreted CSF-1). As tumours grow, immune-suppressor cells, including MDSCs, Bregs, and Tregs are mobilized into the circulation in response to activated cytokine axes that are induced by tumourigenesis. MDSCs and Treg cells infiltrate the growing tumour to disrupt immune surveillance through multiple mechanisms, including disruption of antigen presentation by DCs, inhibition of T and B cell proliferation, and modulation of NK cell cytotoxicity. CAFs, which become activated by tumour-derived factors (for example, SDF-1, TGF-β, FGF or PDGF), secrete ECM proteins and

19 basement membrane components, regulate differentiation, modulate immune responses, and contribute to deregulated homeostasis. CAFs are also a key source of VEGF, which supports angiogenesis during tumour growth. Adapted from Quail & Joyce135.

1.1.6 Immune inhibitory receptors in the tumour microenvironment

There is growing evidence for a role for immunosuppressive molecules expressed by tumour cells themselves, or by host cells under the control of these tumour cells, in activating endogenous immune inhibitory pathways leading to tumour growth and suppression of the inflammatory responses of immune cells in the tumour microenvironment136. Immune inhibitory receptors prevent overactivation and dysregulation of immune cells, which may cause extreme lymphoproliferation, chronic inflammation and autoimmunity. These pathways have recently become therapeutic targets to strengthen anti-tumour responses137, since a number of immunoregulatory molecules are overexpressed by cancer cells of different tissue origins.

The co-inhibitory molecule of the B7 family, B7-H1 (also known as PD-L1), is overexpressed in multiple myeloma, leukemia, ovarian cancer, and breast cancer138. The B7 family consists of structurally related, cell-surface protein ligands, which bind to receptors on lymphocytes that regulate immune responses. B7 ligands deliver 'costimulatory' or 'coinhibitory' signals through the CD28 family of receptors on lymphocytes139. In ovarian cancer, the inhibitory receptor for B7-H1, PD-1, was found on tumour infiltrating CD8+ T cells carrying

TCRs with specificity for tumour-associated antigens but with defective effector functions140. In breast cancer, a concurrent and abundant infiltration of T cells expressing the B7-H1 and

FOXP3 molecules in the tumour microenvironment has been observed in high-risk patients141.

20 Another co-inhibitory receptor, B and T lymphocyte attenuator (BTLA), was found to be expressed at high levels on human melanoma tumour antigen-specific effector CD8+ T cells that were susceptible to inhibition by the BTLA ligand HVEM142. BTLA+CD8+ T cells from the tumour microenvironment were shown to produce less IFNγ than their BTLA- counterparts143.

These studies support the hypothesis that immunoregulatory molecules expressed by tumour cells modulate anti-tumour immune response.

Clinical trials with blocking antibodies against inhibitory receptors, like CTLA4 (the inhibitory receptor for CD80 and CD86) and PD-1, have yielded promising results for the treatment of various cancers (e.g. melanoma, gastric and esophageal adenocarcinomas), and

CTLA4 blockade is already in use for stage IV melanoma patients144, 145. Immunoregulatory molecule blockade is a promising new approach to cancer therapy, and many other inhibitory immune interactions remain to be characterized.

This thesis focuses on characterization of another inhibitory receptor, CD200R (and its ligand CD200), in the regulation of breast cancer, in both mouse models and humans.

1.2 Metastasis and the immune system

1.2.1 Metastasis is a hallmark of cancer

Metastatic disease that is resistant to therapy is the major cause of death from cancer146.

The multistep process of invasion and metastasis has been schematized as a sequence of discrete steps, referred to as the invasion-metastasis cascade147. This succession of events begins with local invasion, then intravasation by cancer cells into nearby blood and lymphatic vessels, and transit of cancer cells through the lymphatic and hematogenous systems. This is followed by

21 extravasation of cancer cells, the formation of micrometastases, and finally the colonization of distant organs by macroscopic tumours2 (Fig. 1-1).

Epithelial-mesenchymal transition (EMT) is a process in which transformed epithelial cells lose their cell polarity and cell-cell adhesion and acquire the abilities to invade, resist apoptosis, and disseminate148-150. By co-opting a process involved in various steps of embryonic morphogenesis and wound healing, cancerous cells are hypothesized to acquire concomitantly multiple attributes that enable invasion and metastasis2. These phenotypic changes are driven by alterations in signal transduction pathways that activate EMT-associated transcription factors, such as Snail, Zeb, and Twist family members, which directly alter epithelial gene expression programs, particularly suppression of E-cadherin adhesion protein encoding gene Cdh1151, 152.

Interestingly, EMT has been reported to promote the generation of CAFs, through de- differentiation of tumour cells of epithelial origin (for example, in breast and prostate cancers) to generate a mesenchymal-like cell population that expresses CAF markers135, 153.

In addition to the alterations that occur in the primary tumour microenvironment, roles for host-derived cells and mediators at sites distal to the tumour have also been reported154. An important concept in tumour metastasis is the formation of premetastatic niches, in which malignant tumours “prepare” the environment of remote organs to receive metastatic cells by altering host homeostasis in these organs before tumour cell arrival155. MDSCs have been identified as an important component of these premetastatic niches156-158.

To metastasize, the tumour must possess certain unique characteristics, including the ability to penetrate the endothelium and acquire mobility within tissues as well as lymphatics or blood vessels159. Not surprisingly, solid tumour cells are able to adopt the phenotypic characteristics of lymphoid cells that enable them to migrate using similar mechanisms160.

22 Circulating tumour cells are particularly sensitive to lysis by NK cells or monocytes. In the presence of anti-tumour Abs, these effector cells of innate immunity are also able to mediate antibody-dependent cellular cytotoxicity (ADCC), thus efficiently eliminating tumour targets161.

A tumour cell that manages to avoid such immune intervention in the peripheral blood or lymphatic circulation and arrives at a new tissue site is dependent on the local microenvironment for growth factors and structural support by the extracellular matrix (ECM), which can be produced by immune cells responding to TAA, thus promoting metastasis formation1.

1.2.2 Immunity and metastasis

Communication between cancer cells and the stromal cells in the tumour microenvironment is involved in the acquisition of invasive growth and metastatic capabilities2,

105, 162, 163. Tumour-infiltrating immune cells secrete cytokines and chemokines that can promote

EMT, including CAFs, macrophages, MDSCs, granulocytes, and lymphocytes 109, 164-166.

Macrophages in the tumour microenvironment of pancreatic islet cancers, mammary tumours, and lung metastases were activated by IL-4 secreted by cancer cells to produce high levels of cathepsin protease, which is thought to promote invasion167. TNF-α-stimulated MSCs can secrete CCL2, which enhanced recruitment of monocytes, macrophages, and neutrophils to the tumour in a mouse lymphoma model, and this promoted malignancy through macrophage- dependent mechanisms168. Moreover, TNF-α secreted by TAMs activated NF-κB-mediated transcription of Snail1 and Zeb, leading to diminished E-cadherin expression on pancreatic carcinoma cells169. Leukocyte-secreted cytokines, such as IL-6 and IL-23, can also initiate EMT via the activation of STAT3 signaling resulting in inhibition of E-cadherin expression and loss

109, 170 of cell–cell adhesion in breast cancer cells . TAMs, MDSCs, and Tregs can also produce

23 large quantities of TGF-β and are important inducers of EMT171. Furthermore, cancer invasiveness can be regulated by interactions with immune inflammatory cells that assemble at the boundaries of the tumours, producing the extracellular matrix-degrading enzymes and other factors that enable invasive growth105, 163, 172.

1.3 Breast cancer

1.3.1 Clinical features of breast cancer

With an estimated 1.15 million new cases each year, female breast cancer is the second most common cancer in the world and the most common cancer among women, accounting for

411,093 cancer deaths per year173, 174. It is estimated that in 2014, 24,400 women will be diagnosed with breast cancer in Canada (1 in 9 women over their lifetime), which represents

26% of all new cancer cases in Canadian women175. This disease is the 2nd leading cause of death from cancer in women (following lung cancer), and it kills 1 in 29 Canadian women176.

Breast cancer can also occur in men, but it is rare – only about 1 in 150 breast cancers is diagnosed in males177. Breast cancer incidence has increased steadily in developed countries over the past few decades, but the mortality caused by breast cancer has decreased in recent years. This could be caused by earlier detection of breast cancer, improved screening techniques, surgical and radiotherapy interventions, understanding of the pathogenesis of the disease, and the use of chemotherapies in a more efficacious manner178. Despite local and systemic therapies, women with breast cancer are at risk of developing metastases (often in bones, lungs, liver, and brain) throughout their life, and approximately 30% of breast cancer

24 patients relapse179. Macroscopic metastases may erupt from dormant micrometastases decades after a primary tumour has been surgically removed or pharmacologically destroyed2, 180, 181.

Several factors have been associated with an increased risk of breast cancer, including family history, nulliparity, early menarche, advanced age, and a personal history of breast cancer182, 183. Five to 10% of all women with breast cancer may have a germline mutation of

BRCA1 and BRCA2 genes; women with BRCA1/BRCA2 mutations have a 40% to 85% lifetime risk of developing breast cancer184 and an increased risk of ovarian and other cancers185, 186. The risk of a primary breast cancer in the contralateral breast ranges from 3% to 10% at 10 years after diagnosis187, but for BRCA1/2 mutation carriers diagnosed at a young age, the risk of contralateral breast cancer can reach 50% in the following 25 years188, 189.

Breast cancer is commonly treated by various combinations of surgery, radiation therapy, chemotherapy, hormone therapy, and immunotherapy. Several clinical and pathological features can influence the combination of therapies the patient receives, such as the patient’s age and menopausal status; disease stage; tumour size; lymph node status; histologic and nuclear grade of the primary tumour; estrogen receptor (ER) and progesterone receptor (PR) status of the tumour; human epidermal growth factor type 2 receptor (HER2/neu) overexpression; and the proliferative capacity of the tumour190, 191.

About 60 to 70% of human breast cancers are ER-positive and estrogen-dependent192, and ER and PR expression are an important indicator of potential responses to hormonal therapy193. Based on the molecular classification of breast cancers, ER+ tumours fall under luminal A (~40%; ERhigh, HER2low) and luminal B subtypes (~20%; ERlow, HER2low), with luminal A subtype having a better outcome194, 195. The ER+ group is the most numerous and diverse, with several genomic tests to assist in predicting outcomes for ER+ patients receiving

25 endocrine therapy196, 197. ER is a member of the superfamily of nuclear receptors that function as ligand-inducible transcription factors198 and mediates the biological (proliferative) effects of estrogens (steroid sex hormones)199. Clinical–epidemiological studies have shown a strong correlation between the actions of ovarian steroid hormones, particularly 17β-estradiol, and carcinogenesis in the mammary gland and uterus200, 201. In addition to endogenous estrogens in the body, aromatase enzyme can aromatize androgens into estrogens. Hormone therapies include ER modulators, like tamoxifen or raloxifene, which block ERs on breast cancer cells, and aromatase inhibitors that lower the amount of estrogen in the body by stopping the conversion of other hormones into estrogens202.

Human epidermal growth factor receptor HER2 overexpression is present in approximately 20–30% of breast cancer tumours. HER2 overexpression is associated with a more aggressive disease, higher recurrence rate, and shortened survival203.The discovery of

HER2 (also called ERBB2) group204 is of interest given the effective therapeutic targeting of HER2 191. Trastuzumab is a humanized monoclonal antibody targeting the HER2 receptor, which was approved for use in 1998. The mechanisms of action of trastuzumab include extracellular mechanisms involving antibody-dependent cellular cytotoxicity (ADCC), and intracellular mechanisms involving apoptosis and cell cycle arrest. Inhibition of angiogenesis, and prevention of DNA repair following chemotherapy-induced damage may also play a role in the mechanism of action205, 206. Trastuzumab has been shown to be effective in combination with chemotherapy, for the treatment of early stage and metastatic HER2 positive breast cancer203.

The American Joint Committee on Cancer (AJCC) has designated disease staging by

TNM classification to define breast cancer207; T describes the size of the primary tumour, N describes the regional lymph node involvement, and M describes distant metastasis. The stage of breast cancer is defined by the combination of these three factors: stage 0 (Tis, N0, M0),

26 stages IA-B (T0-1, N0-1, M0), stages IIA-B (T0-3, N0-1, M0), stages IIIA-C (T0-4, N0-3, M0), stage IV (any T, any N, M1).

Figure 1-3 The anatomy of a breast. Adapted from Canadian Cancer Society.

Out of several histologic classification for breast cancer (ductal, lobular, nipple – see

Fig. 1-3 for breast anatomy), the infiltrating or invasive ductal carcinoma (IDC) is the most common type, comprising 70% to 80% of all cases208. Ductal carcinoma in situ (DCIS) (Stage 0 breast cancer) is a noninvasive condition, but it can sometimes progress to become an invasive cancer, like IDC. Surgery (breast-conserving lumpectomy or total mastectomy) is a common treatment option for patients with DCIS; there is a 25% to 50% incidence of recurrence following limited surgery for palpable tumour, with 50% of those recurrences being invasive carcinoma209, 210. Usually, surgery is accompanied by radiation therapy, although not every patient benefits from it, and researchers are looking for a way to identify these unresponsive subsets of patients to avoid unnecessary postoperative radiation therapy211-213. In addition, patients with DCIS may receive a regimen of tamoxifen214, an ER antagonist. Clinical trials

27 have shown that with tamoxifen, ipsilateral invasive breast cancer decreased from 4.2% to 2.1% at 5 years (P= .03), and the incidence of contralateral breast neoplasms (invasive and noninvasive) decreased from 0.8% per year to 0.4% per year (P = .01)215.

Lobular neoplasia, or lobular carcinoma in situ (LCIS), is often a multifocal and bilateral disease that identifies women at an increased risk for subsequent development of invasive breast cancer. Most women with LCIS have disease that can be managed without additional local therapy after biopsy. The use of tamoxifen has been shown to decrease the risk of developing subsequent breast cancers in women with LCIS216.

Stage I, II, IIIA, and operable IIIC breast cancer often requires a multimodal approach to treatment, depending on ER, PR and HER2/neu status of the primary tumour217. Treatment options include different combinations of lumpectomy/mastectomy, radiation therapy, axillary lymph node surgery, adjuvant or neoadjuvant chemotherapy, hormone therapy (tamoxifen, aromatase inhibitors), and targeted therapy like monoclonal antibodies (trastuzumab for HER2+ patients). Stage IIIB, inoperable IIIC, IV, recurrent, and metastatic breast cancer may often be responsive to different combinations of therapies, but the treatment for metastatic breast cancer is rarely curative.

Triple-negative breast cancer (TNBC) is defined by the absence of staining for ER, PR, ad HER2/neu. TNBC is unresponsive to some of the most effective therapies available for breast cancer treatment, including HER2-directed therapy, such as trastuzumab, and endocrine therapies, such as tamoxifen or the aromatase inhibitors. It is characterized by diffuse erythema and edema, no palpable mass, early age at diagnosis, poor nuclear grade, and poor survival outcome218. Combination cytotoxic chemotherapy remains the standard therapy for early-stage

TNBC219.

28 In Chapter 4 of this thesis I will describe data from our studies suggesting a novel classification of human breast cancer, which takes into account expression of the CD200 molecule. Preliminary data suggest this may highlight a previously unrecognized heterogeneity in this human cancer population, and this may open up new avenues of treatments.

1.3.2 Tumour microenvironment in breast cancer

1.3.2.1 Components of the breast tumour microenvironment

1.3.2.1.1 Cancer stem cells

Cancer stem cells (CSCs) are defined by their ability to initiate tumours in immunocompromised mice upon serial passage (a demonstration of self-renewal) as well as by their ability to differentiate into the non–self-renewing cells forming the tumour bulk220-222. Epigenetic changes that occur during carcinogenesis often affect stem cell regulatory pathways, such as Notch, Hedgehog, Wnt, PI3K, NF-κB, and Jak/STAT pathways223-227. In breast cancer, these pathways are frequently dysregulated by signals from the tumour microenvironment228. MSCs, CAFs, endothelial cells, and various immune cells interact with

CSCs via networks of growth factors and cytokines that form positive feedback loops that promote tumour cell invasion and metastasis27, 229, 230 (Fig. 1-4). Breast cancer stem cells

(BCSC) have been found to express specific cell surface markers CD44+CD25-/lo (human) and

CD24+CD29hiCD49fhi (mouse)222, 231, 232.

1.3.2.1.2 Mesenchymal stem cells

In human breast cancers, MSCs may be recruited from the bone marrow229 or from the normal breast stroma, where they integrate into the tumour-associated stroma and increase the

29 potential for metastasis27 . Mobilization of MSCs into the circulation of patients with advanced highly vascularized breast cancer is associated with chemoresistance233. Aldehyde dehydrogenase 1 (ALDH1) is one of several cancer stem cell markers found in breast cancer, and its expression also identifies MSCs that are selectively recruited to sites of growing tumour.

MSCs interact with breast cancer stem cells (BCSCs) through cytokine loops involving IL-6 and

CXCL7229, which stimulate the self-renewal of BCSCs234. Furthermore, CAFs, TAMs, and

MSCs have been shown to secrete IL-6, IL-8, and CXCL7, which in turn activate STAT3/NF-

κB signaling, leading to self-renewal of BCSCs229 (Fig. 1-4). This generates a positive feedback loop between the tumour microenvironment and tumour cells. Immunohistochemical analysis has supported the idea of the existence of such MSC:BCSC interactions in biopsies obtained from breast cancer patients229. Expression of stem cell markers such as ALDH1 in breast cancer cells was shown to be an independent predictor of poor outcome in women with breast cancer221. In addition, MSCs have the ability to differentiate into adipocytes and CAFs, which can interact with and influence tumour cells235.

1.3.2.1.3 Cancer associated fibroblasts

The gene expression profile of CAFs (activated fibroblasts within desmoplastic lesions that are associated with malignant tumours15) resembles that of wound-activated fibroblasts, and this profile is associated with poor prognosis in breast cancer236, 237. Breast CAFs confer a mesenchymal-like phenotype and can enhance the metastasis of both premalignant and malignant mammary epithelial cells, whereas normal fibroblasts promote an epithelial-like phenotype and suppress metastasis135, 238. Cytokines, such as SDF-1 produced by breast CAFs

(but not normal fibroblasts), may promote proliferation of tumour cells, which express the SDF-

30 1 receptor CXCR417. High levels of expression of SDF-1 and CXCR4 have been associated with poor survival in breast cancer patients239, 240. Additional factors produced by cells in the tumour microenvironment that regulate tumour proliferation, invasion, and metastasis include insulin- like growth factor (IGF), platelet-derived growth factor (PDGF), Wnt, Notch ligand, Hedgehog ligands, and matrix metalloproteinases (MMPs)241-246. CAFs in the breast tumour microenvironment can select for bone-specific metastatic characteristics in primary tumour cells due to selective interaction between breast cancer cells with high Src activity with CAFs that secrete CXCL12 and IGF1247.

CAFs in breast cancer can secrete exosomes, which are vesicles 30-100 nm in size that carry small signaling molecules, including mRNAs and microRNAs248. CAF-derived exosomes can promote breast cancer cell migration through WNT-PCP signaling249.

1.3.2.1.4 Endothelial cells

Endothelial cells have an important function in the tumour microenvironment by direct interaction with the tumour cells as well as by their role in blood vessel formation250, 251. Newly formed blood vessels carry oxygen and nutrients to the growing tumours, facilitating growth and metastasis. Endothelial cells also produce a variety of cytokines that influence in breast tumour progression and can directly regulate BCSCs252, 253 (Fig. 1-4). Although many pro-angiogenic factors have been identified, vascular endothelial growth factor (VEGF) is the primary mediator of angiogenesis254, and, as a result, it has been the principal target of anti-angiogenic therapies.

A humanized monoclonal antibody targeting VEGF (bevacizumab) as well as multi-kinase

VEGF inhibitors (sorafenib and sunitinib) are currently approved for clinical use in a number of tumour types, including renal and colon cancers255, 256. Bevacizumab was initially approved for use in metastatic breast cancer on the basis of reports demonstrating that it prolonged the time of

31 tumour progression when used in combination with paclitaxel257. More recently, however, it was found that the effect on tumour progression was limited, and that the addition of bevacizumab to cytotoxic chemotherapy failed to increase patient survival258.

1.3.2.1.5 Tumour associated macrophages

T cells and monocytes that differentiate into TAMs at the tumour site are the most common breast tumour infiltrating leukocytes259-263. Genes associated with leukocyte or macrophage infiltration, like CD68, have been identified as part of molecular signatures that correlate with poor prognosis in breast carcinomas264. High TAM density is a strong negative prognostic indicator in breast cancer. There is a positive relationship between high levels of

TAMs and lymph node metastases in breast carcinoma, and the density of TAMs is associated with clinical aggressiveness259, 261, 263. In addition, in human breast and oesophagus cancers,

CCL2 levels correlated with the extent of macrophage infiltration, lymph node metastasis, and clinical aggressiveness265, 266. The tumour-promoting properties of TAM may be related to their ability to express growth factors for breast cancer cells, angiogenic mediators, extracellular matrix-degrading enzymes, inflammatory cytokines, and reactive oxygen intermediates6, 48, 262,

267, 268.

In an experimental model of metastatic breast cancer, TAMs supplied epidermal growth factor (EGF) to breast cancer cells, while the cancer cells reciprocally stimulated the macrophages with CSF-1 (Fig. 1-2). The interactions of TAMs and cancer cells were hypothesized to facilitate extravasation into the circulatory system and metastatic dissemination163, 269. Blocking TAMs with clodronate liposomes or through ablation of the

CSF1 gene significantly interfered with extravasation and metastatic outgrowth of breast tumour cells in the lung270. In addition, VCAM-1–positive breast cancer cells bound to VLA-4 (also

32 called integrin α4β1)-expressing macrophages during metastasis of to the lung271. This interaction activated phosphatidylinositol 3-kinase (PI3K)-AKT signaling in breast tumour cells, protecting them from caspase-induced apoptosis, and interruption of this interaction rendered metastatic cells susceptible to apoptotic insult271. Interestingly, VCAM-1 also interacts with a different integrin partner, α4β1, in osteoclasts, which can contribute to bone metastasis272.

1.3.2.1.6 T cells

In contrast to TAMs, the correlation between the extent and type of T cell infiltration in the primary tumour and breast cancer progression is less clear263. A tumour-directed immune response involving cytolytic CD8+ T cells, Th1 cells, and NK cells appears to protect against tumour progression, while an immune response involving the activation of humoral immunity and a Th2 polarized response likely promotes tumour development178. T cell anti-tumour responses are reported to be impaired in advanced stages of breast cancer and tumour-specific T cells can undergo inhibitory regulation and become anergic in tumour-bearing hosts273-276.

Increased Treg levels have been observed in the peripheral blood, primary tumour, and draining lymph nodes (DLN) of breast cancer patients277. Studies have shown that FOXP3+ cells were associated with a more advanced disease in breast cancer, consistent with findings in many other cancers278. It was also found that intratumoural FOXP3 expression had a linear association with invasion, size, and vascularity in breast carcinoma279. Metastatic breast cancer was found to

+ high + + + be associated with an expansion of peripheral blood CD4 CD25 FoxP3 GITR CD152 Treg, whose immunosuppressive properties do not differ from those of healthy subjects178, 280.

Furthermore, FOXP3-positive Treg numbers represent an important marker for the identification of breast cancer patients at risk of late relapses95.

33 IDO, an immunosuppressive enzyme that depletes tryptophan, has been found to promote Treg differentiation and may become a suitable target to abrogate the development of T cell tolerance and to promote an effective immune response to breast cancer278. In a 4T1 mouse model of breast cancer, attenuation of the immune suppressive Treg function using a soluble homodimeric form of mouse GITRL, with adenovirus-mediated intratumoural murine GM-CSF and IL-12 gene delivery, led to elevated IFNγ production, tumour-specific cytolytic T-cell activities, tumour rejection, and long-term survival in 65% of the animals281.

1.3.2.1.7 Myeloid derived suppressor cells

Circulating MDSC levels are positively correlated with clinical breast cancer stage and metastatic tumour burden119. Immature DCs are present in more than 90% of breast cancers, whereas mature DCs are confined to peritumoural areas115. Breast cancer patients with lower numbers of circulating MDSCs had a significantly higher probability of achieving a complete response to chemotherapy282.

In mouse models of breast cancer, limiting the number/function of MDSCs may help improve the efficacy of breast cancer therapies and enhance anti-tumour immunity. For example, 4T1 cells implanted into the flanks of BALB/c mice resulted in the production of nitric oxide by MDSC and attenuation of the IFN responsiveness of T cells283. Furthermore, bone marrow derived myeloid progenitors can be recruited to the pre-metastatic lung in breast cancer284, where they can induce the metastasis of tumour cells through downregulation of

SMAD2 signaling (the canonical TGF-β pathway) and a switch to metastatic growth285. In mouse models of breast and ovarian cancers, tumours with the ability to release significant amounts of CSFs or VEGF are associated with an expansion of a population of MDSC241. This

34 may not only help the tumours suppress immune reactions, but also aid in the construction of new blood vessels for tumour growth286, 287.

1.3.2.2 Signaling in the breast tumour microenvironment

Different components of the breast tumour microenvironment described above communicate with each other by secreting cytokines and chemokines involved in inflammation, as well as by expressing immunoregulatory molecules on their surface. In most cases, inflammation associated with cancer is similar to that seen with chronic inflammation, which can be characterized by slow onset, long duration, large mononuclear cell infiltration, and tissue fibrosis and angiogenesis. Occasionally the balance seems to shift towards a more acute inflammatory process, favouring immune effector function activation45. Chronic inflammation is correlated with risk of breast cancer recurrence in women after primary therapy288 and may be mediated by cytokines including TNFα, IL-1β, IL-6, and IL-85 (Fig. 1-2, 4).

TNFα is one of the key chemical mediators implicated in inflammation-associated cancers. High doses of TNFα can cause hemorrhagic necrosis via selective destruction of tumour blood vessels and generation of specific T cell anti-tumour immunity. However, when produced in the tumour microenvironment, TNFα can act as an endogenous tumour promoter289,

290. TNFα has been linked to breast cancer development. Inhibition of TNFα and NF-κB transcription factor can protect from chemically-induced mammary gland carcinogenesis291, and in vitro activation of the TNFα/NF-κB axis can induce an invasive and malignant behaviour in breast cancer cells292. Human breast cancer cells have been shown to express TNFα261, and its chronic expression in breast tumours may support breast tumour growth293, together with other inflammatory cytokines, like IL1β, CCL2, and CCL5294.

35 The release of chemokines by breast tumour cells can mediate the migration of leukocytes, primarily monocytes, from the circulation to breast tumours. CCL2, a regulator of macrophage recruitment and differentiation, was found to be highly expressed by the breast tumour cells, and expression levels were correlated with TAM accumulation. High expression of

CCL2 was an indicator of early relapse and poor prognosis in breast cancer patients266, 295.

Inhibition of CCL2-CCR2 signaling in mouse breast cancer models prevented metastasis- associated macrophage accumulation and reduced metastasis to the lungs296. Moreover, the ability of human breast carcinoma MDA-231 cells expressing CCL2 to form lung micrometastases in mice was inhibited by neutralizing antibodies to CCL2297.

In addition to CCL2, high levels of CCL5 (RANTES) were shown to correlate with advanced breast carcinoma298, 299, and overexpression of CCL5 in MCF-7 breast tumour cells resulted in increased invasiveness300. In cell lines of breast carcinoma, the expression of CCL5 was elevated by TNFα, alone or in synergism with IFNγ265, 301.

Consistent with the data above, samples of peripheral blood cells from stage II breast cancer patients showed higher expression of the chemokine receptor CXCR4 than PBL of controls and patients in earlier stages of disease302. CXCL12 is a chemokine that binds to a

CXCR4. CXCL12 is expressed in various tumours and may be important in tumour growth and invasion303.

Blockade of IL-6:IL-6R and IL-8:CXCR1 pathways reduced the proliferation of BCSCs in mouse xenograft models of breast cancer and resulted in a reduction of tumour growth and metastasis in these models304 (Fig. 1-4). Serum levels of IL-6 and IL-8 have been associated with poor patient outcome in breast cancer234, 305. The production of these inflammatory cytokines is regulated by the NF-κB signaling pathway306. IL-6 has been shown to be a direct

36 regulator of BCSC self-renewal224 and is a key component of a positive feedback loop involving

MSCs and BCSCs229. IL-6 is also an important mediator of the expansion and recruitment of

MDSCs307, 308. IL-8 receptor CXCR1 is highly expressed on BCSCs, and IL-8 has been shown to stimulate their self-renewal304. Blocking CXCR1 in mouse xenografts significantly reduced the number of BCSCs, leading to decreased tumourigenicity and metastasis. Levels of another inflammatory cytokine, IL-1β, have been shown to be significantly higher in invasive carcinoma

(like IDC) than in DCIS or in benign lesions, implying that elevated levels of IL-1β are also directly correlated with a more advanced disease309. IL-6 and IL-1β increase the rate of accumulation and T cell suppressive activity of MDSCs310, 311. IL-1β-driven inflammation also increases MDSC suppression of innate immunity by facilitating cross-talk between MDSCs and macrophages312. Again, in a study using a 4T1 mouse model of breast cancer, the ectopic expression of IL-1β by 4T1 cells was shown to increase MDSC levels and stimulate growth and metastasis of tumours in vivo310.

GM-CSF, a cytokine that drives myeloid and specifically dendritic cell differentiation, and VEGF, which drives angiogenesis, are additional inflammation-associated molecules that induce the accumulation of MDSC84. Furthermore, GM‐CSF produced by breast cancer cells elicited the production of the IL-6–like cytokine oncostatin M by neutrophils in co-culture experiments. In turn, oncostatin M–stimulated breast cancer cells exhibited increased VEGF production and increased invasiveness in Matrigel invasion assays84.

37

Figure 1-4 Cytokines secreted by cells in the tumour microenvironment regulate breast cancer stem cell renewal. A model of cytokine networks mediating the interaction between MSCs, breast cancer cells, TAMs, CAFs, endothelial cells, and BCSC. IL-6 regulates CXCL7 production by MSCs and mediates MSC chemotaxis. The interaction between mesenchymal cells and tumour cells is regulated by a positive feedback loop that includes CXCL7 and IL-6. CAFs, TAMs, and MSCs secrete IL-6, IL-8, and CXCL7, which in turn activate Stat3/NF-κB signaling, leading to self-renewal of BCSCs, generating a positive feedback loop between the tumour microenvironment and tumour cells Adapted from Liu et. al., 2011 229 and Korkaya, 2011234.

1.3.3 Animal models of breast cancer

Mouse models for breast cancer can be characterized into four main groups: syngeneic models, genetically engineered mice, xenograft models, and chemically/virally/ionizing radiation-induced models313. Animal models have a limited role in cancer research, because the

38 biology of rodents and their tumours differs significantly from that of humans and human cancer

(e.g., size, lifespan and time of tumour progression, numbers and differentiation of targets for oncogenic transformation, transformation efficiency of cancer cells). For example, about one- half of human breast cancers are hormone-responsive at diagnosis, while the vast majority of mouse tumours are hormone-independent with much lower levels of ER and PR than human tumours314. Although the basic mutation frequency is similar in both species, sporadic cancers are quite rare in WT rodents, and cells of rodent origin are much easier to transform in vitro by oncogene transfection or chemical carcinogens315.

Chemical carcinogens, in particular polycyclic hydrocarbons (e.g. DMBA) and alkylating agents (e.g. MNU, ENU) have been widely used to study mammary tumourigenesis in mice. These models have been useful for identification of oncogenes and tumour-suppressor genes, mapping of tumour susceptibility traits, and the assessment of the carcinogenic or chemopreventative effects of various compounds316. However, environmentally-induced cancer models develop a restricted subset of tumour types and grades with incomplete penetrance and variable latency316.

Mouse strains susceptible to mammary cancer were isolated many years ago, with vertical transmission subsequently shown to be due to a mouse mammary tumour virus (Bittner milk factor)315. Mouse breast cancer cell lines that are commonly used include 4T1, 4T07, and

EMT6 cells. Every cancer cell line has unique properties. For example, 4T1 mammary carcinoma (originally derived from a spontaneous mouse mammary tumour of a BALB/C mouse) is a transplantable cell line that is highly tumourigenic and invasive and can spontaneously metastasize from the primary tumour in the mammary gland to multiple distant sites including lymph nodes, blood, liver, lung, brain, heart, and bone317. In comparison to 4T1,

EMT6 tumours are less inflammatory and aggressive. EMT6 cells secrete much lower levels of

39 IL-6 than 4T1 cells, which corresponded with lower levels of infiltrating MDSC in the primary breast tumour318 (see EMT6 Model for a full description).

The site of injection and the specific tropism of the breast cancer cell line used have largely defined primary and secondary metastatic growth. Orthotopic or ectopic implantation of cancer cells in the skin or mammary fat pad with subsequent formation of primary tumour metastases resembles the multiple stages involved in malignant breast cancer development in patients319. Tail vein injection results mainly in lung seeding, whereas portal vein injection results in colonization of the liver. Intracardiac infusion gives rise to a broader target organ spectrum, including bone. The direct introduction of cancer cells into the blood circulation should be considered an assay of organ colonization and not a true metastatic process320.

Transgenic mice that express oncogenes under the mammary-specific promoters, such as the tumour virus long terminal repeat (MMTV-LTR) or the whey acidic protein gene (Wap), were the first generation of genetically modified mice for modeling breast cancer321. Since then, hundreds of transgenic strains have been generated to test the biological relevance of several oncogenic pathways for the initiation of neoplastic transformation of mammary epithelial cells322. Cell-type specific promoters to limit gene expression to specific target tissues

(tetracycline-based systems) and promoter-specific recombinase-based (Cre-Lox) mechanisms for eliminating transgenes from specific tissues have proven to be useful methods for discovering genes and pathways that are involved in tumour initiation322.

Xenograft models are widely used in preclinical breast cancer studies. There are many human breast cancer cell lines available, most are derived from either established cancer cell lines or immortalized normal breast epithelial cells315. Among the more commonly used are the metastatic human breast adenocarcinoma cells (MCF-7, MCF10AT) and the metastatic human

40 breast ductal carcinoma cells (T-47D), which result in the formation of solid tumours after a relatively short latency post-orthotopic injection in immunocompromised recipients. MDA-MB-

231 cells (derived from the pleural effusion of a cancer patient) are a widely used model of an estrogen-independent breast cancer cell line, which is able to colonize bone, liver, lung, adrenal glands, ovary, and brain after intravenous injection into immunocompromised mice320.

Cancer cell lines, like the ones mentioned above, that have been adapted to grow in culture are likely to have different environmental requirements than primary breast tumour cells. The establishment of a primary tumour in vitro is a rare event, found in no more than 1% of primary cancers315. Cell selection in conversion to a continuous cell culture line, genetic drift, as well as viral or Mycoplasma infection are factors that can impact the validity of these xenograft models. Several other important limitations to xenograft models include the lack of immune responses in immunocompromised mice (which are key for tumour development); stromal components (like CAFs) that are not of tumour origin17; and human-mouse cell-host incompatibility320, 323.

Humanized mouse models of breast cancer have recently become a major tool for studying this disease324. Such models can be used to humanize metabolic enzymes, glycosylation enzymes, telomere structure and the immune system, which are useful in biomarker discovery as well as preclinical testing of therapeutics316. Humanizing the murine immune system has consisted of ablating the endogenous immune system followed by engraftment with human immune cells325. Advances in humanizing the mouse mammary fat pad, including reconstituting tumour-promoting immune and fibroblast compartments may improve breast xenograft models326. Further modifications that rely on transgenic expression of human immunoreceptor genes have primarily been performed in mice expressing the homologous murine receptors that could heterodimerize with the human gene products316, 327.

41 With the advent of recombinase-mediated genomic replacement, it might now be possible to knock in the human loci for the entire T-cell and B-cell receptors that span hundreds of kilobases. In an analogous fashion, specific major histocompatability complexes (MHCs) can be introduced into mice that lack endogenous MHC genes, thus making it theoretically possible to reconstitute the major components of the human immune system328.

Another mechanism of humanization is the insertion of human genes into the mouse genome. Knock-in of a human cDNA into the corresponding murine genomic locus is a frequently used strategy316, 329. A recent method called recombinase-mediated genomic replacement could be used to replace hundreds to thousands of kilobases of mouse genomic sequence with syntenic human sequence and to analyze the contribution of non-coding mutations that affect oncogene or tumour suppressor gene expression330. Furthermore, there is a fundamental difference in telomere length in mice and humans, related to the limited or absent telomerase expression in most somatic human cells331, as opposed to persistent telomerase expression in mouse tissues332. Telomeres maintain chromosomal integrity and telomere dysfunction has an important role in oncogenesis, so the replacement of murine telomere components with human is another way to better recapitulate the neoplastic process316.

The model systems employed in the thesis below have focused on mouse transplantable tumour cell lines EMT6 and 4T1, which, as will become evident, have allowed characterization of host immune/ inflammatory processes in murine breast cancers, which differ in their growths, and metastatic potentials.

42 1.4 CD200 and CD200R

1.4.1 CD200:CD200R overview

CD200 is a transmembrane glycoprotein that belongs to the type I immunoglobulin family and exists in both soluble and membrane-bound forms333. CD200 is widely expressed on leukocytes, including B cells, DCs and T cells. It is also found on endothelial cells, neurons, skin and tissue stem cells334. CD200 has a short cytoplasmic tail with no known signaling motifs, and it functions through binding to CD200R, an immune inhibitory receptor that is predominantly expressed by cells of the myeloid lineage335, resulting in an increased threshold of immune activation and attenuation of inflammatory responses336-338. In mouse, five CD200 receptors, CD200R1-R5, have been described. Of the five receptors, only two, CD200R1 and

R2, are found in human. CD200R1 is the major receptor for CD200 in both mouse and human, and CD200R1 contains the NPXY motif in its cytoplasmic tail that is responsible for delivering inhibitory signals downstream337 through Dok2 and RasGAP339. Functional properties have been attributed to the alternative receptors CD200R2-R5 in mouse. However, in human, the function of CD200R2 remains unknown340.

The immunoregulatory function of CD200:CD200R1 interaction has been demonstrated in a number of models. CD200 knock-out (CD200-/-) C57B/6 mice are phenotypically normal, but have an increased susceptibility to autoimmune diseases, such as experimental autoimmune encephalomyelitis (EAE) and collagen-induced arthritis (CIA)341. CD200-/- mice have an increased activation of CD200R1+ myeloid-derived cells, including macrophages and DCs341.

CD200R+ plasmacytoid DCs (pDCs) in mice are not tolerogenic under basal conditions, but engagement of CD200R on pDCs was induces IDO expressions, initiating the tolerogenic pathway of tryptophan catabolism with the help of type I interferons 342, 343. CD200:CD200R

43 interaction skews cytokine production by macrophages and T cells towards the TH2 arm and inhibits degranulation of mast cells in a variety of mouse models, including allergy and allotransplantation studies344.

Our lab has reported attenuation of CIA and reduced allo- and xeno- graft rejection by infusion of a soluble recombinant form of CD200 named CD200Fc (synthesized by linking the extracellular domain of CD200 to a murine IgG2aFc region)345-347. In this molecule, the Fcr and complement binding regions of the Fc region of IgG2a had been eliminated. CD200Fc suppresses T cell allostimulation and type-1 cytokine production (IL-2, IFNγ ) in vitro and in vivo348. Tolerance and long-term survival of both skin and cardiac allografts occurred in mice with systemic overexpression of a CD200 transgene (CD200tg mice)349. Allografts in CD200tg

+ 350 mice had more infiltrating FOXP3 Treg cells and non-degranulating mast cells . These results support the characterization of CD200 as a co-regulatory molecule, which controls the outcome of the TCR-antigen encounter336, 347.

Several viral homologs of CD200 have been identified, indicating that the

CD200:CD200R axis has been exploited by viruses as a mean to control host immune responses351. It was reported that CD200R1 signaling deficiency in female mice affects the clearance of viral infection. Female CD200R1-deficient mice had significantly greater immune- mediated pathology during influenza A infection than male mice352. In mice, engagement of

CD200R2 appears to skew differentiation of DCs toward a phenotype that is capable of inducing

+ + + 353 CD4 CD25 FOXP3 Treg cells .

1.4.2 CD200:CD200R in cancer

CD200 levels are elevated in patients with various solid cancers including: renal, colon, melanoma, and ovarian carcinomas as well as cancers of the hematopoietic system, including

44 myeloma, acute myeloid leukemia (AML), and chronic lymphocytic leukemia (CLL)354-358.

Elevated CD200 has a negative effect on likelihood of survival, and expression of CD200 on cancer cells has been correlated with poor chances of recovery among AML patients355. Clinical trials of an antibody to CD200 found reduced tumour size in leukemia and lymphoma patients

359 and a decrease in the number of circulating Treg . The role of CD200 in immune tolerance suggests that CD200 may be a marker for cells that are better able to evade the immune system.

Established tumours may evade the immune system by generating a tolerogenic response mediated by CD200 (Fig. 1-5).

The role of CD200 in cancer has also been supported by work in preclinical models.

CD200-/- mice were shown to be resistant to chemical-induced skin carcinogenesis360, and decreased tumour growth in these mice was accompanied by increased expression of proinflammatory cytokines by DCs in skin-draining lymph nodes, indicating that tumour growth in the presence of CD200 was likely a result of CD200-induced immunosuppression.

Administering CD200Fc to mice with EL4 thymoma tumours increased tumour growth347, 361.

Human cancer stem cells from prostate, breast, brain, and colon cancers expressed CD200 on their surface358, 362, and CD200+ breast cancer cells, but not CD200- cells, grew in SCID mice to form a tumour363.

Expression of membrane CD200 on human CLL cells caused suppression of autologous effector T cells, and CD200 blockade using a rat monoclonal anti-CD200 antibody produced a

+ + + 364, 365 reduction in the number of CD25 CD4 FOXP3 Treg in vitro . Furthermore, increased expression of CD200R, which is required for signalling following CD200 engagement, was detected on a subpopulation of CD4+ T cells in the spleen of CLL patients relative to controls.

45 Importantly, CD200 expression on tumour cells can be modulated upon immune challenge in vivo. Using a murine breast tumour cell line EMT6 with no detectable CD200 expression in vitro, we showed that CD200 expression on EMT6 cells was increased in vivo after transplantation into immunocompetent BALB/c mice366 (see EMT6 breast cancer model below).

These findings suggest that the presence of CD200 on tumour cells may identify cancer cells that are better able to evade the patient’s immune response362.

CD200:CD200R signaling may regulate anti-cancer responses by altering the proliferation and activation of regulatory cells in the tumour microenvironment. CD200 affects the cytokine production profile by immune cells344, which may influence immune cell infiltration and activation in the tumour and the DLN, enhancing metastasis via distal action. For example, overexpression of CD200 is associated with increased production of TGFβ367, an important molecule in the regulation of anti-tumour immunity368. In mice, CD200 stimulates the development of IL-10/TGFβ-producing suppressor T cells (Ts) that prevent an inflammatory immune response367, 369 and may protect breast tumour cells from the host’s anti-tumour inflammation, thus resulting in enhanced tumour growth and metastasis.

The thesis that follows demonstrates a role for CD200 expression by tumour/host on both local and metastatic growth of breast cancer cells in mice. A recent study has supported these findings in a different mouse model of squamous cell carcinoma, where CD200 expression was shown to be highly induced in cells that have metastasized to the lymph node and other solid organs. These CD200+ squamous carcinoma cells appeared to have a selective advantage to metastasize, possibly through modulating CD200R+ myeloid cells in the lymph node microenvironment370.

46

Figure 1-5 CD200 in the tumour microenvironment. (Upper) CD200 expressed by tumour or host cells interacts with CD200R either on an immune effector cell, directly suppressing anti- tumour activity, or on an APC, which delivers an indirect immunosuppressive signal. (Lower) Antibodies blocking the CD200:CD200R interaction may be a promising therapeutic approach to enhance anti-tumour immune responses in cancer. Adapted from Rygiel & Meyaard, 2012336.

47 1.4.2.1 Soluble CD200

A soluble form of CD200 (sCD200) was detected in normal human plasma, and its levels were increased in the plasma of CLL patients, where sCD200 levels were correlated with tumour burden, late stage disease, and disease aggressiveness371. High sCD200 levels enhanced engraftment of splenic human CLL cells in vivo, when NOD.SCIDIL-2Rγ-/-mice were supplemented with sCD200hi CLL plasma. Engraftment was attenuated when sCD200 was pre- absorbed from the plasma or when the mice were treated with an anti-CD200 mAb371. Recently we found that CD200Fc increased lung and liver metastasis of three human breast cancer lines tested in immunocompromised mice (MCF7; HTB19; MDA-MB-231). Furthermore, tumour metastasis was associated with changes in gene expression in tumour tissue, with significant alterations in genes encoding MMPs and transcription factors modulating inflammation372.

1.5 EMT6 breast cancer model

We have developed a mouse model to study the role of CD200 in breast cancer growth and metastasis. Murine EMT6 breast tumour cells are injected into the mammary fat pads of female BALB/c mice, where the tumours develop over 18-20 days. The EMT6 cell line was established several decades ago from a transplantable murine mammary carcinoma that arose in a BALB/cCRGL mouse after implantation of a hyperplastic mammary alveolar nodule. The resulting tumour line (named KHJJ) was propagated in BALB/cKa mice and adapted to tissue culture after the 25th animal passage, and the cell line was named EMT. EMT6 is a clonal isolate of EMT isolated in 1971 at Stanford University373.

48 EMT6 breast tumour cells grown in vitro do not express CD200 on their surface.

However, when EMT6 cells are injected into immunocompetent, wild type (WT) BALB/c mice, approximately one third of the EMT6 cells isolated from the tumour were found to express

CD200 on their surface366. The in vivo induced expression of cell-surface CD200 is lost after 7-

10 days in culture. In the presence of a compromised immune system (in NOD.SCIDIL-2Rγ-/- and

CD200tg mice), this selection for CD200+ tumour cells is not observed (Fig. 1-6)366. Thus, in female mice, we infer that a competent immune system selects for growth of breast cancer cells expressing CD200 on their surface. CD200, in turn, binds to receptor(s) on the host’s immune cells within the tumour microenvironment and inhibits generation and expression of anti-tumour immune response. This hypothesis supports the idea that cancer cells may hijack the body’s own immune system to protect themselves from being eradicated by the host’s anti-tumour immunity.

Figure 1-6 CD200 expression on the surface of EMT6 cells grown in culture, in immunocompromised NOD.SCIDIL-2Rγ-/- mice, and in immunocompetent WT BALB/c mice.

To study the effect of host- or tumour-derived CD200 on EMT6 breast cancer growth and metastasis, we have developed four animal strains and three cancer cell lines. In addition to the WT female host, our lab generated CD200R1-/- mice, CD200-/- mice, and CD200tg mice

49 (which systemically overexpress a CD200 transgene under the control of a doxycycline- inducible promoter). We also established two EMT6 cell lines in addition to the control EMT6:

EMT6CD200tg tumour cells, which express a CD200 transgene when cultured in vitro; and

EMT6siCD200 tumour cells, where the expression of CD200 was silenced with a short hairpin

RNA.

CD200 overexpression by both host and tumour cells leads to an increased rate of primary tumour growth and an increased frequency of tumour cell metastasis to DLN366. When

EMT6 tumour cells that have metastasized to tumour DLN are cloned in a limiting dilution assay, higher frequencies of CD200+ metastatic clones were observed in WT animals than in immunocompromised mice, even though the latter mice showed faster tumour growth.

Treatment with a non-cytotoxic anti-CD200 monoclonal antibody augmented EMT6 tumour resistance in control mice. Thus, in the presence of an intact immune system, CD200 expression on tumour cells produces a metastatic advantage363.

CD200+ EMT6 tumour cells can be preferentially cloned from DLN of tumour-bearing mice at a frequency 10 times higher than CD200- EMT6 cells. In addition, when DLN from

EMT6 tumour-bearing mice were stimulated with EMT6 tumour cells in vitro, cytokine production in the culture supernatants indicated that control mice have a greater immunity to

EMT6 cells than CD200tg mice. The absence of a significant cytotoxic response in CD200tg mice and the decline in cytotoxicity in control mice was associated with the presence of CD4+

+ + cells. Also, there were more Gr-1 MDSCs and FOXP3 Treg in both the DLN and TIL populations of control compared with CD200tg EMT6 tumour-bearing mice, which reflects immune resistance in control animals.

50 1.6 Objectives and hypotheses

Using the various strains of host animals, with different expression levels of CD200 and

CD200R1, and EMT6 tumour cells, over- or under-expressing CD200, a number of “proof of principle” studies can be conducted to further our understanding of the role of CD200:CD200R interaction in regulation of local and metastatic growth of breast cancer, and to explore the mechanisms behind these effects. The objective of this thesis is to:

1. investigate the role of CD200 expression by the host and tumour cells in breast

cancer growth and metastasis

2. study the role of CD200R1 signaling in immune system regulation during breast

cancer progression

3. investigate whether soluble CD200 in circulation can be a biomarker that

differentiates breast cancer patients with different prognoses

Fig. 1-7 shows hypothesized outcomes for various combinations – on one extreme side, the CD200R1-/- host bearing EMT6siCD200 tumour cells is predicted to have the slowest primary tumour growth and metastasis, while on the other side, the systemically immunocompromised

CD200tg host carrying a EMT6CD200 tumour are expected to grow the largest tumours and have the greatest extent of metastasis, following delivery of immunoregulatory CD200 signals through CD200R.

51

Figure 1-7 Predicted outcomes of EMT6 breast tumour growth and metastasis, according to the CD200:CD200R1 model

The experiments in this thesis that follows provide data to address the hypotheses depicted in this figure, and aims to characterize some of the mechanisms by which

CD200:CD200R interactions may produce the effects described.

52

Chapter 2 Evidence for a role of tumour CD200 expression in breast cancer metastasis: decreased metastasis in CD200R1-/- mice or using EMT6siCD200 breast cancer cells

A manuscript of the same title has been published in Breast Cancer Research and Treatment Journal, 2012, 136: 117-127

2 Studies in WT and CD200R1-/- mice with EMT6, EMT6CD200tg, and EMT6shCD200 tumour cells

2.1 Abstract

Previous studies reported that CD200 expression by cells of the transplantable EMT6 mouse breast cancer line was increased during growth in immunocompetent mice. Low levels of expression persisted in NOD.SCIDIL-2Rγ-/- mice or in mice with generalized over-expression of a

CD200 transgene (CD200tg mice), despite the faster tumour growth in both of the latter strains.

We also showed that CD200 expression (by the host and/or tumour cells) led to increased seeding of tumour cells to draining lymph nodes (DLN) in immunocompromised (CD200tg or

NOD.SCIDIL-2Rγ-/-) vs immunocompetent mice, using limiting dilution cloning assay of tumour cells from DLN (vs contralateral lymph nodes, CLN). Evidence for an important role for CD200 expression in this increased metastasis came from the observation that neutralization of CD200 by anti-CD200mAbs decreased tumour metastasis and increased levels of cytotoxic anti-tumour immune cells in DLN.

53 In the current studies we have extended these observations by exploring tumour growth/metastasis in CD200R1-/- mice, in which we have previously shown, in a transplant model, that expression of CD200 fails to deliver an immunosuppressive signal. In addition, we have studied local and metastatic growth in healthy control mice of EMT6 tumour cells stably transduced with shRNA able to silence CD200 expression. In both scenarios decreased metastasis was observed, with increased immunity to EMT6 detected by cytotoxicity assays.

Additionally, adoptive transfer of DLN to control mice attenuated EMT6 metastases, implying a potential therapeutic benefit from neutralizing CD200 expression in breast cancer.

2.2 Introduction

Metastatic spread of breast cancer cells is regulated by factors intrinsic to tumour cells374-

376 as well as by host associated elements111, 377, 378. Foremost amongst the factors investigated by several groups has been the expression of TGF, and indeed, polymorphisms in TGFR have been reported to play an important role in breast cancer metastasis in humans379. Expression of chemokines or chemokine receptors regulates metastatic spread in many animal models, possibly by recruiting inflammatory-type cells to the local tumour environment where they produce angiogenic factors and matrix-degrading enzymes380-383. Recruitment of cells which counter host responses normally controlling distant spread of disease, including Gr-1+CD11b+ myeloid- derived immune suppressor cells (MDSCs)276, 384 and FOXP3+ regulatory T cells (Treg)385 may facilitate metastasis, and indeed TGFis implicated in recruitment of GR-1+CD11b+ MDSCs to promote metastasis386.

Increased expression of CD200 has been suggested to contribute to cancer progression in human solid tumours356, 357 and hematological tumours354, 355, 387. Several years ago it was reported that human breast cancer stem cells expressed CD200, and that CD200+ cells, but not

54 CD200- cells, grew in SCID mice to form a tumour358. We ourselves have observed increased expression of human CD200 on several human breast cancer lines growing in NOD.SCIDIL-2R-/- mice (RMG-unpublished). These data are reminiscent of studies we reported earlier, showing that in immunocompetent mice, growth of EMT6 mouse breast cancer cells was associated with increased expression of membrane CD200 (mCD200) and increased FOXP3+ Treg and MDSCs in the draining lymph nodes (DLN)366. These effects were not observed when EMT6 tumour cells were grown in immunocompromised mice, either NOD.SCIDIL-2R-/- or CD200 transgenic mice

(CD200tg), where CD200 produced by tumour cells has negligible effect on the level of

CD200363.

Most recently we showed that increased expression of CD200 by either the host or tumour cell (using EMT6CD200tg lines), could facilitate metastatic spread to DLN of tumour cells implanted into mouse mammary fat pads. The increased metastasis was abolished by anti-CD200 antibody therapy, and was associated with an increased presence of CD200+ and CD200- tumour cells cloned from the DLN363. We showed also that production of TGF by host cells was strongly implicated in the increased metastasis. As further proof-of-principle of this suggested role for CD200-CD200R interactions in promotion of breast cancer metastasis, we herein describe three additional studies. Using EMT6 cells “silenced” for CD200 expression following transfection with a lentivirus encoding shRNA for CD200 (EMT6siCD200), we show that these cells fail to metastasize to DLN. A similar decreased rate of metastasis was seen using CD200R1-

/- mice as tumour recipients. Finally, we demonstrate that DLN from both of these latter mice showed evidence for an increased host immune response to the tumour, which could be adoptively transferred to naïve recipients.

55 2.3 Materials and methods

2.3.1 Mice

Female WT BALB/c mice were purchased from Jackson Laboratories (Bar Harbour,

Maine). Founder C57BL/6 CD200tg mice and CD200R1-/- mice, described in detail elsewhere349, 388, went through ten generations of backcrossing to BALB/c background. All mice were housed 5/cage and allowed food and water ad libitum in an accredited facility at the

University Health Network. Female mice were used at 8-12 weeks of age. All animal experimentation was performed following guidelines of animal care committee (AUP1.15).

2.3.2 Monoclonal antibodies

Antibodies were described previously366. Anti-mouse CD4 mAb (GK1.5) or anti-CD8 mAb (YTS.156.7), along with rabbit complement, were purchased from Cedarlane Labs,

Hornby, Ontario. Depletion of CD4+/CD8+ cells used mAb (1:10 dilution) for 60min at 4oC, followed by washing and treatment with complement (45min at 37oC). >95% specific depletion was observed as assessed by subsequent FACS analysis.

2.3.3 EMT6 breast tumour cells, induction of tumour growth in BALB/c mice, and limiting dilution assays

BALB/c mice received 5x105 tumour cells injected into the mammary fat pad in 100 μl

PBS. Tumours were harvested, and cells digested with mixture of collagenase, trypsin, and

DNase. Cells were centrifuged over mouse lymphopaque (Cedarlane Labs, Hornby, Ontario), and resuspended in αMEM medium with 10−6M 2-mercaptoethanol and 10% fetal bovine serum (αF10).

56 Tumour draining lymph nodes (DLN) were harvested from the same mice and single cell suspensions prepared by digestion as for the tumour mass itself. Limiting dilution analysis employed to assess the frequency of tumour cells that can be cloned in the lymph node preparations used the following dilutions for each DLN preparation in vitro: 18 wells (103 cells/well); 18 wells (104/well); 18 wells (105/well); 9 wells (106/well); 1 well (107/well). Cells were grown in microtitre plates for 22 days, fixed with glutaraldehyde, washed threefold with

PBS, and incubated with rat anti-mCD200 mAb (10A5) for 90 min at room temperature. Plates were washed exhaustively with PBS, incubated with HRP-conjugated anti-rat Ig followed by appropriate substrate, and read in an ELISA plate reader. Control wells on the plate contained eight wells/group of confluent cultures of EMT6 or eight wells of confluent EMT6CD200tg cells.

A cloned well was deemed positive (for CD200 expression) if the OD415 was >3SD greater than the mean OD for control EMT6 cells363.

2.3.4 Production and use of lentiviral particles encoding shRNA specific

for mouse CD200

shRNA plasmid DNA for CD200 was amplified from bacterial glycerol stocks

(MISSION® shRNA, Sigma Aldrich). Self-inactivating replication-incompetent viral particles were produced in packaging cells (Hek293) by co-transfection with compatible packaging and envelope plasmids (psPAX2 and pMD2.G). 48 and 72h medium containing lentiviral particles were pooled, centrifuged, and supernatants were stored at –80oC.

EMT6 cells and A20 cells (mouse lymphoma cell line - CD200+ control) were infected with the lentiviral particles and grown in the presence of puromycin for 72h. Bulk cultures were then cloned under limiting dilution to produce stably transfected EMT6siCD200 or A20siCD200 cells.

The cloned cells were analyzed by real time PCR for knockdown of CD200 (>90% loss of

57 CD200 mRNA expression in A20shCD200). Confirmation of continued expression of CD200shRNA in the EMT6shCD200 cells (EMT6 cells themselves only express CD200 following growth in immunocompetent mice366) was obtained using RT-PCR analysis with primer pairs designed to generate an amplicon from the viral insert, which included the shRNA. No such amplicon was seen using cDNA from control EMT6 cells.

2.3.5 Preparation of cells and cytotoxicity, proliferation, and cytokine

assays

Single cell suspensions were prepared aseptically from DLN of mice and cells were resuspended in αF10. In stimulation assays used to assess induction of cytotoxic T cells and/or cytokine production, 5×106 responder DLN from control or tumour-bearing mice were stimulated with 2 × 104 irradiated (2500 Rads) cultured EMT6 tumour stimulator cells in duplicate in αF10. Supernatants were pooled at 40 hrs from replicate wells and assayed in triplicate in ELISA assay, with capture and biotinylated detection mAbs as reported348. Varying volumes of supernatant were bound in triplicate at 4°C to plates pre-coated with 100 ng/ml mAb, washed thrice, and biotinylated detection antibody was added. After washing, plates were incubated with streptavidin–horseradish peroxidase (Cedarlane Labs), developed with appropriate substrate and OD405 determined using an ELISA plate reader. Recombinant cytokines for standardization were obtained from Pharmingen (USA), with assay sensitivity in the range ~40 pg/ml. When cytotoxicity was assayed, cultures were harvested at 6 days, and cells were titrated at various effector:target ratios for killing (18 hrs at 37°C) of 51Cr-labeled

EMT6 targets.

In cultures CD4+ T cells isolated from tumour-bearing mice were used as a regulatory pool source. CD4+ cells were purified from spleen cell preparations by negative selection using

58 a CD4+ T Cell Isolation Kit according to the manufacturer’s instructions (StemCell

Technologies, Vancouver, Canada). 5×106 responder lymph node cells were used from mice challenged 14 days earlier with 10 × 106 irradiated cultured EMT6 tumour cells subcutaneously, and 1 × 106 regulatory DLN were added as described. Mixed cultures received stimulation with

2 × 104 irradiated EMT6 tumour stimulator cells in duplicate as described above366.

2.3.6 Statistics

The frequency of cloneable tumour cells was determined by a least squares linear regression plotting the frequency of wells without tumour cell growth versus the log10 of the number of cells plated, and the number of cells containing one cloneable unit was estimated at

36.7% negative wells. This value and its standard error were determined using the method of

Snedecor and Cochrane363. Within-experiment comparison between groups used ANOVA, with subsequent paired Student’s t-tests as indicated.

2.4 Results

2.4.1 Comparison of metastasis of EMT6 or EMT6CD200tg to DLN of wild-

type (WT), CD200tg or CD200R1-/- mice

In our first study 6 mice/group of WT BALB/c, CD200tg, or CD200R1-/- females received 5x105 tumour cells subcutaneously in the mammary fat pad, with all mice maintained on Dox throughout the experiment in order to activate the transgene. The challenge tumour cells were themselves harvested at day 14 post injection from WT BALB/c mice previously inoculated with normal EMT6 or EMT6CD200tg (see Materials and Methods). As shown in Fig. 2-

1a, both control and EMT6CD200tg tumour cells grew to a larger mass in WT and CD200tg recipients than in CD200R1-/- recipients. All primary tumour recipients were sacrificed at day

59 18, axillary DLN pooled from individual mice, and single cell suspensions were prepared for each animal. Each suspension was plated at varying dilutions (from 103/well to 107/well), and tumour clones were assessed in all wells at day 22. Fig. 2-1b/c shows pooled data (% non- responding wells) from all 6 mice/group, for both tumour cell lines in each of the three recipient strains. Cells in all clones were stained (~100% positive) with an antibody against the intracellular tumour-amplified kinase BTAK (data not shown).

As reported previously363, few cloneable tumour cells were detected in DLN from WT mice challenged with the parent EMT6 tumour (Fig. 2-1a) - frequency ~1 in 8x105 cells.

Injection of those same tumour cells into CD200tg mice produced ~10-fold increase in the frequency of tumour clones (now ~1 in 7x104 cells) from DLN (Fig. 2-1b). Importantly, injection into CD200R1-/- mice led to even fewer detectable metastatic clones than were seen in

WT mice (1 in 7x106). This difference was maintained when EMT6CD200tg cells were injected into CD200R1-/- animals (Fig. 2-1c), although, as described in our earlier report, a much higher frequency of metastasis was observed even in WT mice using this CD200-overexpressing tumour line. These data confirm our earlier observation that overexpression of CD200 by both host and tumour cells can contribute to increased metastasis of EMT6 to the DLNs. Moreover, they establish a key role for host CD200R1 expression in the signaling pathway which contributes to this effect, a phenomenon previously reported in the context of an immunosuppressive circuit responsible for attenuation of transplant rejection388.

After measuring the frequency of cloneable tumour cells in DLN of all mice, positive tumour wells were fixed with glutaraldehyde and expression of CD200 estimated by ELISA.

- Clones were scored positive if the OD415 was >3SD above the mean for control (CD200 ) cultured EMT6 cells. Table 2-1 indicates that EMT6 cells cloned from DLN of WT mice were

60

Figure 2-1 Primary tumour growth and limiting dilution analysis of frequency of cloneable EMT6 cells in DLN of WT, CD200tg, or CD200R1-/- mice receiving either control EMT6 or EMT6CD200tg tumour cells. 5x105 tumour cells were injected subcutaneously into the mammary

61 fat pads of a total of 6 mice per group. All mice were maintained on Dox throughout to ensure induction of CD200 transgene expression in CD200tg animals. Primary tumour mass was weighed and axillary DLN were pooled from individual mice at d18 of tumour growth, single cell suspensions made, and cells cloned at different dilutions to determine the frequency of outgrowth of tumour cells (100% stained, after permeabilization, with BTAK antibody). Data show pooled results from all 6 mice of each group. In panel a, ** indicates significant increase, and * significant decrease (p<0.05) relative to WT control

. predominantly mCD200+ (~75%), while cells recovered from CD200tg mice were predominantly mCD200- (~75%). All tumour cells cloned from DLN of mice receiving

EMT6CD200tg, including CD200R1-/- mice, were, as predicted, mCD200+ (see bottom two rows of Table 2-1).

Applying the frequency of cloneable tumour cells (Fig. 2-1) to Table 2-1, we calculated the number of CD200+ and CD200- clones in 107 lymph node cells. In WT BALB/c mice receiving EMT6 tumour, most clones were CD200+, whereas in CD200tg BALB/c, there were five times more CD200+ clones and 24 times more CD200- clones. This effect of host CD200 expression to promote outgrowth of CD200- tumour cells was seen also in a previous study364.

2.4.2 EMT6CD200tg tumour promotes metastasis of CD200- tumour cells

to DLN

The data above confirm an important role for CD200 engagement of host CD200R1 both in regulation of primary tumour growth and in promotion of metastasis of EMT6 to DLN, as well as that host CD200 expression can promote metastasis even of CD200- tumour cells. In order to assess whether CD200+ tumour cells could also enhance metastasis of CD200- EMT6 to

62 DLN, we compared local tumour growth and the frequency (and CD200 expression) of clonal outgrowths of tumour cells obtained from DLN of WT or CD200tg mice receiving EMT6 or

EMT6CD200tg alone, or mixtures of these two tumour cells (3:1 ratio of control EMT6 vs

EMT6CD200tg). Data obtained pooled from 2 independent studies, each using a total of 6 mice/group, are shown in Fig. 2-2 and Table 2-2.

Data in Fig. 2-2a confirm the increased primary growth of EMT6CD200tg tumour cells in both WT and CD200tg recipients, with mixtures of the two tumour cells growing to a similar mass. Fig. 2-2b shows that the frequency of tumour clones obtained from DLN of WT mice receiving the mixture of cells was increased by more than 10 times compared with mice receiving only control EMT6 cells, to a tumour frequency akin to that seen using EMT6CD200tg alone, or that when control EMT6 cells were implanted in CD200tg recipients (Fig. 2-1c). This effect was less apparent in CD200tg mice because of the increased frequency of clonal tumour growth (Fig. 2-1c/2c). However, comparison of CD200 expression by the tumour cells cloned from DLN (Table 2-2) emphasizes that increased clonal outgrowth of CD200- tumour cells was seen both using CD200tg recipients of control EMT6 cells and using WT mice receiving mixtures of both EMT6 and EMT6CD200tg cells. These data confirm that expression of CD200 by either the recipient or by the injected tumour cells can promote outgrowth of CD200- tumour cells.

2.4.3 Metastasis of CD200- EMT6 enhanced by EMT6CD200 depends on

host CD200R1 expression

To investigate whether the effect of tumour CD200 expression on promoting metastasis of CD200- tumour cells was attributable to engagement of recipient CD200R1 on host cells, we repeated the study shown in Fig. 2-2, using CD200R1-/- mice as recipients. These data are

63

Figure 2-2 Comparison of primary tumour mass (panel a) and frequency of tumour cells cloned by limiting dilution from DLN of WT (panel b) or CD200tg (panel c) BALB/c mice at 18 days

64 post transplantation of 5x105 tumour cells. 6 recipients were used per group. The tumour inocula used for each recipient group were either control EMT6, EMT6S, or a mixture of these two cells (with 75% control cells in the mixture). In panel a, * indicates significant difference from control EMT6 tumour cells (p<0.05).

shown in Fig. 2-3 and Table 2-3. Note again (Fig. 2-3a) the attenuated growth of both EMT6 and EMT6CD200tg cells (and mixtures of these cells) in CD200R1-/- recipients (see also Fig. 2-1a).

Once again, when growth and metastasis was explored in WT BALB/c, expression of

CD200 by EMT6 cells themselves increased the frequency of detectable CD200- tumours derived from recipient DLN using mixtures of EMT6CD200tg and control EMT6 (see also Fig. 2-

2). In contrast, no such effect was observed in CD200R1-/- mice, with few tumour clones derived from the DLN, regardless of the tumour inoculum used.

2.4.4 Decreased primary growth and absence of metastasis to DLN

following injection of EMT6siCD200 in WT BALB/c is rescued by co-

injection with EMT6CD200tg

As further confirmation of an important role for tumour expression of CD200 in promotion of local and metastatic outgrowth of CD200- tumour cells in DLN, we studied an

EMT6 cell line carrying a silencer for CD200 gene expression (EMT6siRNA) following transfection with a lentiviral vector expressing CD200 shRNA (see Materials and Methods).

Since cultured EMT6 cells fail to express CD200 at high levels366, we used a control cell line, the murine A20 lymphoma, to monitor effective functional silencing by the lentiviral vector used. In control studies >90% suppression of constitutive CD200 mRNA expression was seen

65

Figure 2-3 Differences in primary tumour growth (panel a) and DLN metastasis of mixtures of control or EMT6CD200tg tumour cells injected into WT or CD200R1-/- mice. The study described is similar to that shown in Fig. 2-2, except that tumour mixtures were in this case injected into

66 WT or CD200R1-/- BALB/c recipients. Data show the measured frequency of tumour cells cloned by limiting dilution from DLN of mice at 18 days post transplantation of 5x105 tumour cells (see Fig. 2-1 and text). In panel a, * indicates significant decrease and ** significant increase (p<0.05) compared with control EMT6 tumour cells in WT recipients.

following transfection and puromycin selection of these cells, with identical conditions used to derive EMT6siCD200 cells. In preliminary studies we compared growth of these latter cells with control EMT6 cells in WT or CD200tg mice. Data shown in Fig. 2-4 and Table 2-4 show that unlike the control EMT6 cells, these cells grow locally and metastasize to a lesser degree in WT mice than control EMT6, with the few DLN clones detected all CD200- (unlike the ~75% positive clones seen using control cells). However, a more similar local growth and frequency of metastasis to control EMT6 cells was seen when EMT6siCD200 were injected into CD200tg mice, suggesting that failure of local growth/metastasis in WT mice did not reflect an unexpected intrinsic biological defect in these silenced cells.

As an alternative approach to “rescue” DLN metastatic growth following injection of

EMT6siCD200 cells into the mammary fat pad, we used a co-injection stategy with EMT6CD200tg, as described in Fig. 2-2. These data, derived from pooled studies using 8 mice/group are shown in Fig. 2-5 and Table 2-5. It is clear again that bystander CD200 expression, in this case from co-injected CD200+ tumour cells, fostered both local growth and metastatic spread of CD200- cells (from the EMT6siCD200 injected) in DLN, although the frequency of metastasis from this mixture was significantly less that seen using admixture of EMT6CD200tg and control EMT6 cells. This may be explained by the fact that about 1/3 EMT6 cells express CD200 and with gene silencing, there was less CD200 compared to the other mixture.

67

Figure 2-4 Decreased local growth (panel a) and frequency of metastasis of EMT6siCD200 tumour cells in WT mice (panel b) is attenuated when cells are injected instead into CD200tg recipients (panels a/c). Frequencies were established by limiting dilution cloning using DLN harvested at 18d from mice receiving the primary tumour inocula shown. In panel a, * indicates

68 significant decrease and ** significant increase (p<0.05) compared with control EMT6 tumour cells in WT recipients.

2.4.5 Improved host immunity to EMT6 tumour cells in mice following

injection of both EMT6siCD200 in WT mice and control EMT6 tumour

cells in CD200R1-/- mice

In our earlier studies we reported evidence for the development of T-cell-mediated immunity to EMT6 tumour cells in BALB/c mice receiving EMT6 tumour, though at later stages of tumour growth (~d14-18 post injection of tumour cells) this immunity was suppressed in the host, likely by a combination of both myeloid-suppressor cells and FOXP3+CD4+ T cells389. Given that two independent approaches, namely growth of tumour in CD200R1-/- mice and transfer of EMT6siCD200 cells into WT mice, had clearly established evidence that attenuation of CD200 expression by tumour, or of CD200-mediated signaling in the host, reduced both local tumour growth and DLN metastasis, we sought evidence in both situations for an active immune process which was restraining tumour growth in such mice. For this study groups of 4 WT or CD200R1-/- mice each received either control or EMT6siCD200 cells as before.

All mice were sacrificed at day 15 post tumour inoculation, and DLN cells were pooled from individual animals. Tumour size in these mice is shown in Fig. 2-6a. DLN cells were incubated with 51Cr-labeled EMT6 tumour target cells at 30:1 effector:target ratio for 12hr at 37oC, and specific lysis was enumerated by counting in a TopCounter. Data in Fig. 2-6b, are pooled from two such studies, and show clearly that following injection of EMT6siCD200 into WT mice, or even control cells into CD200R1-/- mice, there was evidence for increased killing of control

69

Figure 2-5 Attenuated local growth (panel a) and metastasis of EMT6siCD200 cells to DLN in recipient WT mice (panel b) is rescued by co-injection of EMT6siCD200 cells with tumour cells

70 themselves over-expressing CD200 (EMT6CD200tg)-panels a/c). In panel a, * indicates significant decrease and ** significant increase (p<0.05) compared with control EMT6 tumour cells in WT recipients.

EMT6 cells using cells harvested from DLN of recipient animals. In additional studies (sup. Fig.

2-1) killing was attenuated (>80% loss of activity) following depletion of CD8+ cells in all groups.

To determine if cells in DLN from mice in Fig. 2-6 would show an ability to attenuate tumour growth in vivo, in a separate series of studies we explore the effect of adoptive transfer of 107 DLN cells harvested at d15 from these mice to WT BALB/c injected with control EMT6 cells in the mammary fat pad. Control groups of mice received no DLN cells or cells from non- tumour bearing WT animals. Primary tumour mass was measured in all recipients at d18, and

DLN were used to explore evidence for tumour metastasis (frequency of tumour clones developing after 22d in vitro). Data showing the mean tumour size and the number of tumour cells cloned/106 DLN cells are shown in Fig. 2-6c (mean +SD from a total of 6 mice/group).

It is clear from these data that, in accord with in vitro results shown in panel a, DLN cells taken from CD200R1-/- tumour bearers, or even WT mice receiving EMT6siCD200 cells, contained a population of cells able to suppress both primary EMT6 tumour growth at the implantation site (mammary fat pad) and secondary metastases in WT recipients of those control tumour cells.

71

Figure 2-6 Adaptive immune response in CD200R1-/- tumour bearing hosts can be transferred to naïve WT mice a) Tumour size (day 15) in WT or CD200R1-/- mice receiving 5x105 control EMT6 or EMT6siCD200 tumour cells. Data show mean (±SD) of 8 mice/group. b) Specific lysis of EMT6 tumour cells in vitro (30:1 effector:target) using DLN cells harvested from mice in panel a. c) Tumour volume (day 18 post tumour injection) or DLN clones/106 harvested DLN, in WT mice sacrificed at 18d post control EMT6 tumour inoculation. Mice received in addition either no DLN cells, cells from WT (tumour-free mice) or mice from the groups shown in panel a. Data represent mean (±SD) of 6 mice/group.

72 2.5 Discussion

Multiple mechanisms are thought to be implicated in regulating cancer growth, progression, invasion, and metastasis, including inflammatory cytokines, chemokines, and chemokine receptors ,as well as populations of regulatory T cells (Tregs) and other myeloid suppressors cells, all of which may act primarily by altering host resistance mechanisms119, 276,

381, 382, 390-392. The cytokine TGFis thought to play a central role in altering tumour growth

386, 393- directly, and indirectly, again through its effects on altered immunity and Treg induction

395. We and others have presented evidence suggesting an important role for CD200:CD200R interactions in regulating inflammation and immunity in a variety of model systems, including autoimmune activation, viral infection, allergy, and transplant rejection. In an earlier report we also observed that EMT6 breast cancer cells growing in immunocompetent mice were apparently selected for over-expression of CD200, a molecule not normally expressed by the same cells in vitro, or by EMT6 tumours growing in immunocompromised animals389. We speculated that such expression was a key factor in enabling growth in immunocompetent mice, and showed that such growth was diminished by a non-depleting (Fab) anti-CD200 mAb.

Additional studies have characterized an important role for CD200 expression in aiding metastasis of EMT6 cells to DLN following implantation into the mammary fat pad of recipient mice. Few such tumour clones are derived from DLN in WT mice, but these clones show

73

Figure 2-7 Importance of CD8+ T cells in specific lysis of EMT6 tumour cells (30:1 effector:target) using DLN cells harvested at 15 days post tumour inoculation into WT or CD200R1-/- mice. Tumour cells inoculated were either control EMT6 or EMT6siCD200 cells. DLN cells were treated with anti-CD4 or anti-CD8 mAbs before use in 51Cr-release assay. Untreated cells were treated with rabbit complement only. Data represent mean (±SD) of 4 mice/group. * p<0.05 compared with corresponding untreated or anti-CD4-treated groups.

overexpression of CD200 (>70%). In contrast, a higher frequency of clones was observed in

DLN following tumour growth in CD200tg mice, but these tumour clones were predominantly

CD200-363. EMT6 cells themselves transfected to overexpress a CD200 transgene (EMT6CD200tg) also showed increased metastasis to DLN in normal recipient mice.

The studies described above have extended these observations in several important ways.

We showed that simple admixture of CD200+ tumour cells (EMT6CD200tg) with control (CD200-) cultured EMT6 cells increased the ability of these latter cells to grow locally and metastasize in

WT mice, with these metastatic cells remaining predominantly CD200- (Fig. 2-2 and Table 2-2).

The evidence that this reflects an effect on a host immune response, which normally acts to

74 control growth and tumour metastasis, comes from similar studies in mice lacking the primary inhibitory receptor for CD200, CD200R1. In such animals attenuated growth and metastasis to

DLN was seen following inoculation of either control or EMT6CD200tg tumour cells (Fig. 2-1) and admixture of the two cells also failed to reverse these effects (Fig. 2-3). Thus, failure to attenuate host immunity by CD200:CD200R interactions concomitantly failed to augment local growth or DLN metastasis, regardless of any additional CD200 expression by tumour cells. As it is the clonogenic subpopulation of a tumour that is responsible for its maintenance, and as the frequency of these putative tumour stem cells is low, staining for CD200 and CD200R1 along with BTAK would not provide pathological confirmation of our findings. However, if a marker for tumour stem cells were eventually developed, it could become possible to provide an alternative quantitation of tumour stem cell frequency. Given the small number of clonogenic cells in the DLN, for in situ quantitation, the entire node (from independent groups of animals) would need to be serially sectioned, and stained using adjacent sections from as a control.

Alternatively, it could become more feasible to stain tumour stem cells in the DLN suspension and perform a FACS analysis. Until new technology becomes available, we believe the limiting dilution analysis approach described above provides a useful methodology to investigate such populations.

As further evidence for the importance of both host and tumour CD200 expression on growth and DLN metastasis in this breast cancer model, we studied EMT6 cells stably transfected with a lentiviral vector carrying a silencer short hairpin RNA (shRNA) specific for

CD200. Using such cells as tumour inoculum we again saw attenuation of local growth and metastasis in WT mice, although more than 10-fold higher frequencies of DLN metastases were observed in CD200tg mice (Fig. 2-4). In addition, rescue of CD200- tumour metastatic cells from

75 these CD200-silenced tumour cells occurred following co-injection with EMT6 overexpressing the CD200 transgene, EMT6CD200tg (Fig. 2-5, Table 2-5).

Finally, we have asked whether the manipulations described, namely the use of CD200- silenced tumour cells and growth in CD200R1-/- mice, which had such a profound impact on decreasing DLN metastasis of EMT6, was reflective of improved host immunity in such tumour-inoculated mice. In one such study, shown in Fig. 2-6a/b, we used DLN cells as a source of cytotoxic cells for radiolabeled EMT6 targets, and showed that both CD200R1-/- recipients of control EMT6 and WT recipients of EMT6siCD200 silenced cells developed augmented cytotoxicity to EMT6 cells, with evidence for attenuation of growth of primary tumour (Fig. 2-

6a). More interesting perhaps are the data of Fig. 2-6c, which show that adoptive transfer of these DLN cells to secondary WT mice receiving control EMT6 tumour cells markedly reduces both primary tumour size (volume) and DLN metastasis in these secondary hosts. The cell(s) responsible for function in this latter assay have yet to be characterized. Nevertheless, our observations suggest that therapy aimed at interfering with CD200:CD200R1 interactions in breast tumour bearing hosts, regardless of the site (tumour/host) of CD200 expression, may not only decrease local and distant (metastatic) growth of tumour, but may also, perhaps through a linked mechanism, lead to improved host anti-tumour immunity.

Acknowledgements: Supported by a grant from the Canadian Cancer Society (CCS) to RMG &

DAC.

76

2.6 Tables

Footnotes:

a. Tumour cells were cloned at limiting dilution as described in Fig. 2-1 from DLN of mice

injected with EMT6 tumour cells into the mammary fat pad and sacrificed at d18. Cells

were fixed at d22 of culture, and mCD200 expression assayed using rat anti-CD200 (see

Materials and Methods). For rows 1 versus 2/3, and rows 4 versus 5/6 clone frequency

differs by P < 0.01.

+ b. CD200 clones were those with OD415 >3SD above the mean for control EMT6 carried

- in culture (CD200 ); mean OD415 for control wells was 0.12+0.01. Total number of

clones scored in ELISA shown in parentheses.

+ c. Arithmetic mean (+SD) OD415 for mCD200 clones

d. Number of cloneable cells per 107 DLN cells.

77

Footnotes:

a-d as for Table 2-1. Groups of control or CD200tg mice received control EMT6,

EMT6CD200tg or a mixture of these two tumour cells (3:1 control EMT6 cells). Clone

frequencies were not significantly different in row 2 vs 3, but both were significantly higher

at P < 0.01 from row 1. There was no significant difference in overall frequency detected

(ANOVA) between rows 4-6, but there were differences in CD200+/CD200- EMT6 clones

detected in different groups.

78

Footnotes:

a-d as for Tables 1/2. Groups of WT or CD200R1-/- mice received control EMT6,

EMT6CD200tg or a mixture of these two tumour cells (3:1). Clone frequencies were

significantly higher in rows 1, 2, and 5, with few tumour clones derived from any of the

CD200R1-/- mice. Note also the difference in CD200+/CD200- EMT6 clones detected in

the different groups of rows 1, 2, and 5 (see also Table 2-2).

79

Footnotes:

a-d as for Tables 1-3. Groups of WT or CD200tg mice received control EMT6, or

EMT6siCD200. Clone frequencies were significantly higher in rows 1 vs 2, with

attenuation of this difference in comparison between rows 3 and 4. No CD200+EMT6

clones were detected in mice receiving EMT6siCD200, confirming the efficacy of the

lentiviral transfection system.

nd = no detectable signal above background (from CD200- EMT6 carried in vitro)

80

Footnotes:

a-d as for Tables 1-4. Mice received control EMT6, EMT6CD200tg or EMT6siCD200, and in

some cases a mixture of tumour cells, in which one-quarter of the partner cell mixture

was EMT6CD200tg. Clone frequencies were significantly higher in row 2 vs 1, with again

few tumour clones derived from mice receiving EMT6siCD200 cells (row 3). Failure of

metastasis from EMT6shCD200 was rescued by co-injection with EMT6CD200tg (rows 4 and

5). Note again the difference in CD200+/CD200- EMT6 clones detected in the different

groups.

nd = no detectable signal above background (from CD200- EMT6 carried in vitro)

81

Chapter 3 The role of CD200 expression by tumour and host cells in regulation of immunotherapy in the mouse EMT6 breast cancer model

Manuscript will be submitted to Nature Publishing Group Breast Cancer Journal

3 Studies in CD200-/- and WT hosts with EMT6 and EMT6siCD200 tumour cells

3.1 Introduction

Mouse models of breast cancer have provided insights into the mechanisms of immune responses to tumour cells, and some of these findings are being translated into effective cancer immunotherapies for humans. EMT6 is a transplantable breast cancer cell line derived from a

BALB/c female mouse. It is considered to be a less aggressive type of breast cancer compared with other cell lines, like 4T1, which may be a closer model of rare human inflammatory breast cancer. In the previous chapter, we reported that cell-surface CD200 expression by mouse

EMT6 breast tumour cells increased primary tumour growth and metastasis to the draining lymph nodes (DLN) in both WT and CD200tg BALB/c female recipients. Lack of CD200R1 expression in the CD200R1-/- host negated this effect. Furthermore, silencing CD200 expression in EMT6siCD200 tumour cells reduced their ability to grow and metastasize in WT animals396.

These data were consistent with the hypothesis that CD200 expression, through engagement of

CD200R1, leads to attenuation of a protective anti-tumour response and was important for controlling metastasis. In this chapter, the mechanism of CD200:CD200R1 interaction in breast cancer was investigated further.

82 CD200 can be expressed by both host cells and breast tumour cells in vivo. In this study, we characterized tumour infiltrating and DLN cells in mice either lacking CD200 expression

(CD200-/- BALB/c mice), or receiving tumour cells lacking CD200 expression (EMT6siCD200 cells). To determine the role of T cells in the anti-EMT6 tumour immune response, we explored the TILs and DLN cells in WT mice challenged with EMT6 and EMT6siCD200 tumours and treated with CD8 and CD4 depleting antibodies.

We extended these findings to a model in which anti-EMT6 tumour immunity was explored in CD200-/- mice with immunotherapy in combination with chemotherapy. CD200-/- mice served as a perfect host for testing the synergistic effect of this combination, because immunotherapy alone, which included altered CD200 signaling, irradiated EMT6 cells as a source of antigen, and CpG as adjuvant, was able to decrease EMT6 metastasis, but was not sufficient for generating a long-lasting anti-tumour immune response397. We chose metformin for chemotherapy, because recent studies in mouse models have demonstrated that metformin may target cancer-initiating cells, affect metabolism, and alter immune responses in the host.

For example, metformin inhibited the growth of a subpopulation of breast cancer initiating cells in culture and reduced their ability to form tumours in mice398. Metformin was reported to inhibit angiogenesis and metastatic growth of breast cancer by targeting both the tumour cells and the white adipose tissue endothelial progenitor cells in the tumour microenvironment399.

When combined with trastuzumab, metformin reduced the cancer-initiating cell population in

HER2/neu-amplified breast cancer cells400. Metformin also reduced the growth of a variety of tumour xenografts in mice, including those established from breast and prostate cancer cells401,

402, and suppressed the development of breast, colon, and other tumours in transgenic mice403,

404.

83 3.2 Materials and methods

3.2.1 Mice

Female WT BALB/c stock mice were purchased from Jackson Laboratories (Bar

Harbour, Maine). CD200-/- (BL/6 background) mice were derived commercially by Caliper Life

Science, with deletion of exons 2–4 encoding murine CD200. CD200-/- BALB/c mice were derived from founder stock (on BL/6 background) and backcrossed through ten generations with

BALB/c mice obtained from Jax Labs before intercrossing for use in subsequent studies

(genotyping performed by Kai Yu). All mice were housed 5/cage and allowed food and water ad libitum in an accredited facility at the University Health Network (UHN). Female mice were used at 8-12 weeks of age. All animal experimentation was performed following guidelines of an accredited animal care committee (protocol No. AUP1.15/6).

3.2.2 Tumour cells

The BALB/c-derived EMT6 breast tumour cell line, used to model hormonal, radiation, and chemotherapeutic effects on primary and metastatic breast cancer growth405, was obtained from ATCC and passaged in vitro in α-minimal essential medium (α-MEM) with 10% fetal calf serum (αF10), supplemented with penicillin (10 units/ml) and streptomycin (10µg/ml).

siCD200 396 EMT6 cells were derived as described previously and passaged in vitro in αF10, supplemented with penicillin (10 units/ml) and streptomycin (10µg/ml), as well as puromycin

(1μg/ml) for selection and maintenance of silenced cells. Cells were incubated in a humidified

o CO2 incubator at 37 C.

84 3.2.3 Growth of EMT6 cells in mice

5 × 105 EMT6 tumour cells in 150 μl of PBS were injected into the mammary fat pad of recipient mice. Seventeen to 22 days post tumour cell injection, mice were sacrificed via cervical dislocation, tumours were excised, and draining lymph nodes (DLN) and contralateral lymph nodes (CLN) were harvested. Single-cell suspensions from the lymph nodes were prepared by passing the tissues through 70 µm filters and resuspending the cells in PBS.

Tumour cells were digested with a mixture of collagenase and trypsin for 30 min at 37oC in a bottle with a magnetic stirrer. Tumours cells were then centrifuged over mouse lymphopaque

(Cedarlane Labs, Hornby, Ontario, Canada) and resuspended in PBS without calcium and magnesium.

3.2.4 Tumour resection and immunization of mice

EMT6 and EMT6siCD200 tumours were surgically resected from WT BALB/c and

CD200-/- mice 12-15 days post tumour cell inoculation397. As discussed elsewhere397, mice receiving immunotherapy were treated by intraperitoneal immunization with 2x106 EMT6 tumour cells (irradiated with 2500 rad) mixed with 100 µg CpG ODN in 100 µl PBS and emulsified with an equal volume of Incomplete Freund’s adjuvant, 1-2 days after surgery. In some studies, metformin, an agent reported to be able to attenuate breast cancer cell growth, was also administered IP daily (60 mg/kg IP injections daily) for an additional 6 weeks. Verteporfin

(autophagy inhibitor) was used in some studies (4.5 mg/kg IP injections 3 times/week), and mice were kept in the dark 24 hrs post injections.

3.2.5 Immunostaining and flow cytometry

Single-cell suspensions from DLN, CLN, and tumours were washed twice with 1 mL

FACS buffer (PBS, 5% FBS, 5mM EDTA) and incubated with cell-surface antibodies (or

85 isotype controls) purchased from Biolegend (see below) at concentrations recommended by the supplier for 20 min at 4oC in the dark. Cells were washed again and fixed with 1% paraformaldehyde (PFA) for 30 min. Samples were assessed by a LSR flow cytometer and the data were analyzed using FlowJo software. Analysis gates were set with the aid of “fluorescence minus one” isotype controls. Forward scatter pulse height and side scatter analyses were performed to exclude cell clusters.

3.2.6 Antibodies

Anti-CD45, anti-CD3, anti-CD4, anti-CD8, anti-CD11b, anti-GR-1, anti-F4/80, and anti-

CD200 antibodies conjugated to PE, FITC, APC, Pe-Cy7, or APC-Cy7 fluorochromes were purchased from Biolegend and used at concentrations determined recommended by the supplier for immunostaining and flow cytometric analysis of DLN, CLN, and tumour cells.

Anti-CD4 (L3/T4, clone YTS) mAb ascites purchased from Cedarlane was used for depleting CD4+ T cells in tumour-bearing mice at concentrations determined to be effective.

Mice were injected intravenously (IV) at 5 day intervals starting at day 6 post tumour inoculation.

Anti-CD8a monoclonal antibody (clone 53-6.7) purchased from Biolegend was used for depleting CD8+ cells in tumour-bearing mice. Mice were injected IV at 4 day intervals starting at day 6 post tumour inoculation.

Intracellular staining for IFN-γ and FOXP3 was performed using the BD Biosciences

Cytofix/Cytoperm Kit (Cat # 555028) according to the manufacturer’s instructions after incubating DLN cells with BD GolgiPlug for 8 hrs (Cat#555029).

86 3.2.7 Proteomics assay

EMT6 and EMT6siCD200 tumour lysates were assayed for the expression of 40 proteins associated with inflammation using Proteome Profiler Mouse Cytokine Array Panel A purchased from R&D Systems (catalog #ARY006), following the manufacturer’s instructions.

3.2.8 ELISArray

Supernatants of EMT6 and EMT6siCD200 tumours collected after 24 hrs in αF10 culture were assayed for the expression of 12 mouse inflammatory chemokines and cytokines using

Multi-Analyte ELISArray Kit purchased from Qiagen.

3.2.9 Cytokine ELISAs

Supernatant samples harvested from tumours cultured in αF10 medium for 24 hrs were assayed for GM-CSF and IL-1β concentration using ELISA kits purchased from Biolegend. A standard curve was obtained in each assay to quantify cytokines present in the supernatant.

3.3 Results

3.3.1 Characterization of BALB/c CD200-/- mice

The gross cellular phenotype of male and female CD200-/- BALB/c mice ~12 weeks old was assessed by flow cytometry of immune cell populations and ELISA of sCD200 in peripheral blood. In comparison to WT mice, with ~5-8 ng/ml of circulating sCD200 in their peripheral blood, CD200-/- mice had undetectable levels of sCD200 (Fig. 3-1a). Flow cytometric analyses of spleen, thymus, as well as axillary and mesenteric lymph nodes confirmed the lack of cell-surface CD200 expression in CD200-/- mouse organs (Fig. 3-1b). As reported elsewhere

-/- 341 -/- for CD200 BL/6 mice , the frequency of immune cells in CD200 BALB/c mice , including

87

Figure 3-1 Characterization of CD200-/- mice in the context of EMT6 breast cancer model. a) Levels of sCD200 in the peripheral blood of WT mice and CD200-/- mice were assessed by ELISA. Data show mean (+SD) of 3 mice/group. b) Flow cytometric analysis of cell-surface

88 expression of CD200 by cells in spleen, thymus, and lymph nodes in WT (n=3) and CD200-/- (n=6) mice. Shaded curves show tissues staining from CD200-/- mice. Representative plots are shown. c) Tumour size (day 22) in WT (n=3) and CD200-/- (n=4) female mice receiving 5x105 control EMT6 or EMT6siCD200 tumour cells. Data show mean (+SD). *p<0.05 (2-tailed Student’s t test) d) Comparison of EMT6 and EMT6siCD200 tumour growth in WT (top panel) and CD200-/- (bottom panel) female mice. Each point represents mean of 3 mice/group.

B cells, CD4+ and CD8+ T cells, NK cells, macrophages, and myeloid cells, did not differ significantly from WT BALB/c mice. No abnormalities were detected in reproductive cycles and the health of litters from CD200-/- female BALB/c mice.

The rate of growth of EMT6 breast tumour cells was similar in WT and CD200-/- mice, and EMT6siCD200 tumours were consistently smaller in both WT and CD200-/- hosts at the time of sacrifice (Fig. 3-1c,d). On average, control EMT6 tumours growing in immunocompetent WT female BALB/c mice reach upwards of 0.7 g by day 20 post tumour cell injection, and this remained approximately the same in CD200-/- mice. EMT6siCD200 tumours at that time-point were only 0.4 g in both WT and CD200-/- mice. There was no obvious difference in the ability of

EMT6 and EMT6siCD200 tumours to metastasize to lung and liver in WT mice, which may reflects a greater importance for host rather than tumour CD200 expression in this phenomenon.

3.3.2 Analysis of cell populations harvested from tumours in WT and CD200-/- BALB/c female mice

Tumour infiltrating lymphocytes (TIL) and immune cells in the DLN and contralateral lymph nodes (CLN) were investigated by cell-surface staining and flow cytometry. There were fewer cells infiltrating EMT6siCD200 tumours but a higher percentage of CD45+ immune cells than in control EMT6 tumours (Fig. 3-2a). These CD45+ cells included T cells (CD3+) and myeloid cells, including macrophages (GR-1+F4/80+) and myeloid-derived suppressor cells

89

Figure 3-2 EMT6 and EMT6siD200 tumour microenvironment in WT and CD200-/- female mice. a) EMT6 and EMT6siCD200 tumours were harvested from WT mice at day 20 post tumour cell

90 injection. Tumours were digested, stained for CD45 cell-surface molecule, and analyzed by flow cytometry. Data show mean (±SD) of 4 mice/group. *p<0.0005 b) EMT6 and EMT6siCD200 tumours grown in WT and CD200-/- mice were digested and stained for CD45, GR-1, and CD11b cell-surface markers. Panels show representative dot plots from 4 mice/group after gating on CD45+ cells. NS c) EMT6 and EMT6siCD200 tumours grown in WT and CD200-/- mice were digested and stained for CD45, CD3, and CD8 cell-surface markers d) EMT6 and EMT6siCD200 tumours grown and DLN from WT mice were digested and stained for CD45, CD3, and CD8 cell-surface markers. Panels show representative dot plots from 4 mice/group after gating on CD45+ cells. e) Relative levels of G-CSF, GM-CSF, IL-1β, MCP-5, MIP-1β, and RANTES in EMT6 and EMT6siCD200 tumours grown in WT mice were assessed by a proteomics array. Data show mean (±SD) optical density (OD) of 3 samples/tumour. f) Levels of GM-CSF in 24hr supernatant from EMT6 and EMT6siCD200 tumours cultured in αF10 were assessed by ELISA. Data show mean (±SD) of 4 mice/group. *p<0.05 g) Levels of IL-1β in 24hr supernatant from EMT6 and EMT6siCD200 tumours cultured in αF10 were assessed by ELISA. Data show mean (±SD) of 4 mice/group. NS

(MDSC) (GR-1+CD11b+). The TIL in EMT6siCD200 tumours were generally more enriched for

CD8+ T lymphocytes with less MDSC than control EMT6 TIL, which corresponded with the lower numbers of CD8+ T cells found in the DLN of mice bearing EMT6siCD200 tumours (Fig. 3-

+ + 2b-d). We detected more CD4 CD25 Treg cells in the DLN of WT mice bearing EMT6 tumours compared to EMT6siCD200 tumours (Fig. 3-7a). These results are consistent with the hypothesis that adaptive immunity is enhanced in hosts carrying tumours that have reduced expression of the immunoregulatory CD200 molecule.

The relative levels of 40 different cytokines, chemokines, and acute phase proteins present in the microenvironment EMT6 and EMT6siCD200 tumours growing in WT female mice were determined using a commercial proteomics array with tumour lysates. There was a decrease in the concentration of several soluble mediators in EMT6siCD200 tumour lysates

91 compared with EMT6 lysates, including G-CSF, GM-CSF, IL-1β, MCP-5, MIP-1β, and

RANTES (Fig. 3-2d and Fig.3-7b). Given the correlation with the increased immune cell infiltration and decreased tumour size in EMT6siCD200 tumours, these data imply a potential role for these cytokines in promoting EMT6 tumour growth in female WT BALB/c mice.

Supernatant samples from cultures of EMT6 and EMT6siCD200 tumours kept in αF10 medium for 24 hrs were assayed for cytokines and chemokines detected by a commercial mouse inflammatory protein ELISA. Consistent with the results from the proteomics array, GM-CSF and IL-1β levels were decreased in EMT6siCD200 tumours compared with control EMT6 tumours

(Fig. 3-7 c-h). Using quantitative ELISAs of mouse IL-1β and GM-CSF, EMT6 tumours were found to express more IL-1β than EMT6siCD200 tumours from WT mice (93 ± 66 pg/ml vs. 58 ±

10 pg/ml; Fig. 3-2e), and more GM-CSF than EMT6siCD200 tumours cultured for 24 hrs (3.74 ±

1.01 ng/ml vs. 1.7 ± 0.9 ng/ml; Fig. 3-2f), though there was significant variability in the data collected.

3.3.3 Altering CD4 and CD8 T cell signaling in WT female mice bearing EMT6 or EMT6siCD200 tumours

Reduced expression of CD200 by EMT6 tumour cells resulted in a smaller tumour size

(Fig. 3-1c), which may reflect an enhanced adaptive anti-tumour immune response to

EMT6siCD200 cells compared with control EMT6 cells in WT BALB/c mice. To investigate the role of CD8+ and CD4+ T cells in mediating anti-tumour responses against EMT6 cells, we treated EMT6 and EMT6siCD200 tumour bearing WT BALB/c female mice with either a CD8 or

CD4 depleting antibody every 4/5 days starting at day 6 post tumour cell injection until day 20.

DLNs of tumour bearing mice treated with anti-CD8 had, as expected, few CD8+ T cells

(>90% depletion; Fig. 3-3a), and the weight of EMT6siCD200 tumours growing in WT mice for 20 days increased in size following CD8+ cells depletion from, ~0.4 g to ~0.75 g (Fig. 3-3b).

92

Figure 3-3 Depleting CD8+ T cells in EMT6 and EMT6siCD200 tumour bearing WT hosts. a) Single cell suspensions of DLN from EMT6 and EMT6siCD200 tumour bearing WT mice treated with a CD8-depleting antibody or PBS were stained for CD45, CD3, and CD8 cell surface markers. Data show mean (±SD) of 3 mice/group. b) Tumour size (day 19) in WT mice receiving 5x105 control EMT6 or EMT6siCD200 tumour cells and treated with CD8-depleting antibody or PBS. Data show mean (±SD) of 3 mice/group. *p<0.05 c) EMT6 and EMT6siCD200 tumours were harvested from WT mice treated with anti-CD8 or PBS at day 19 post tumour cell injection. Tumours were digested, stained for CD45 cell-surface expression, and analyzed by flow cytometry. Data show mean (±SD) of 3 mice/group. NS

Although the EMT6siCD200 tumours were smaller, they contained 2-3 times more CD45+ infiltrating cells than EMT6 tumours (Fig. 3-2a, 3-3c). When tumour bearing WT mice were treated with anti-CD8, the level of CD45+ TIL dropped to baseline, as seen in control EMT6

93 tumours (Fig, 3-3c), consistent with the notion that CD8+ T cells play a critical role in anti- tumour immunity in the EMT6 model.

When CD4+ T cells in EMT6 and EMT6siCD200 tumour bearing WT hosts were depleted, a significant reduction in EMT6 and EMT6siCD200 tumour growth was observed (Fig. 3-4b).

Depletion of CD4+ T cells (Fig. 3-4a) corresponded with an increase in the number of CD45+ cells infiltrating the tumour, although the total number of cells in CD4-depleted hosts was significantly smaller (Fig. 3-4c). Treatment with anti-CD4 also increased the number of CD8+ T cells in the DLN of both EMT6 and EMT6siCD200 tumour bearing mice (Fig. 3-4d).

CD45+CD3+CD8+ cells in the DLN of EMT6siCD200 tumour bearing hosts showed increased

IFNγ expression in the absence of ex vivo stimulation (Fig. 3-4e) in comparison with control

EMT6 tumour bearing hosts. Treatment of EMT6 tumour bearing mice with CD4-depleting antibody increased the number of IFNγ+ CD8+ T cells in the DLN. These results support the hypothesis that CD4+ T cells play a tumour-promoting and immunoregulatory role in the EMT6 tumour model.

3.3.4 Augmentation of adaptive immunity to EMT6 in mice receiving metformin

In previous studies we reported that using conventional chemotherapy with a combination of paclitaxel and anti-VEGF antibody, we were able to cure mouse EMT6 primary tumours406. However, this treatment did not lead to development of any resistance to re- challenge with the same tumour, leaving treated mice at risk of metastasis406. In contrast to WT hosts, metastatic growth of EMT6 cells could be cured in CD200R1-/- mice after surgical resection of the primary tumour followed by immunization with irradiated EMT6 cells and CpG

94

Figure 3-4 Depleting CD4+ T cells in EMT6 and EMT6siCD200 tumour bearing WT hosts. a) Single cell suspensions of DLN from EMT6 and EMT6siCD200 WT tumour bearing mice treated

95 with a CD4-depleting ascites or PBS were stained for CD45, CD3, and CD4 cell surface expression. Data show mean (±SD) of 5 mice/group. b) Tumour size (day 19) in WT mice receiving 5x105 control EMT6 or EMT6siCD200 tumour cells and treated with CD4-depleting ascites or PBS. Data show mean (±SD) of 8 mice/group from 2 pooled experiments. *p<0.005 c) EMT6 and EMT6siCD200 tumours were harvested from WT mice treated with anti-CD4 or PBS at day 19 post tumour cell injection. Tumours were digested, stained for CD45 cell-surface expression, and analyzed by flow cytometry. Data show mean (±SD) of 3 mice/group. *p<0.005 d) Single cell suspensions of DLN from EMT6 and EMT6siCD200 WT tumour bearing mice treated with a CD4-depleting ascites or PBS were stained for CD45, CD3, and CD8 cell surface expression. Data show mean (±SD) of 5 mice/group. *p<0.005 e) DLN cells from EMT6 and EMT6siCD200 WT tumour bearing mice treated with control IgG or anti-CD4 were incubated with GolgiPlug for 8 hours and stained with CD45, CD3, and CD8 cell-surface antibodies as well as IFNγ intracellular antibody. Data show mean (±SD) of 3 mice/group. *p<0.05

as adjuvant397, and these survived metastasis-free for over 300 days. In CD200-/- mice, although the outcome could be improved, lung and liver metastases could be detected after ~150 days

(Fig. 3-5a,b).

Importantly, metastasis-free survival in mice with altered CD200:CD200R1 axis was correlated with increased anti-tumour CTL activity in the DLN of tumour bearing mice in vitro

(Fig. 3-5c) and protection from subsequent re-challenge with EMT6 cells in vivo397. We suggested that this was consistent with an immunosuppressive role for CD200 expressed on tumour cells in such mice. In order to explore whether it might be possible to develop a model in which both immunotherapy and chemotherapy were used to enhance tumour resistance, we investigated the effect of treating CD200-/- mice with immunotherapy, as before, in addition to daily administration of metformin after immunization. Metformin is a non-conventional

96

Figure 3-5 Metformin augments anti-tumour immunity in CD200-/- EMT6 tumour bearing hosts. a,b) Published in Gorczynski et al, Breast Cancer Research and Treatment, 2013. a) Lung

97 and liver metastases in control, CD200-/-, and CD200R1-/- BALB/c mice receiving 5x105 tumour cells subcutaneously into the mammary fat pads, following surgical resection 15 days later, with subsequent immunization with EMT6 with CpG as adjuvant . b) Control* shows frequency of tumour cells in DLN at surgery (d14). * indicates significant difference with WT controls. The remaining histograms show frequency of tumour cells in DLN at times shown (post surgery) for mice with surgery alone, or surgery + immunotherapy (irradiated EMT6+CpG). Experiment terminated at d90 (no surviving CD200-/-). *indicates significant difference with mice receiving surgery only. c) Specific lysis of 2x104 EMT6 (18 hrs) by EMT6-restimulated DLN cells of control, CD200-/-, or CD200R1-/- mice, compared with control BALBc. All mice (except WT-no EMT6 group) received EMT6, surgical resection, (d14), and immunization with EMT6 in CFA (d16). Mice were sacrificed 14 d after surgery. Data shown are arithmetic mean ±SD for each group. The number of mice in each group is indicated in paretheses. d) Decreased tumour metastasis to lung/liver in immunized CD200-/- mice after metformin treatment (x21d), 4 mice/group. e) Single cell suspensions of DLN from CD200-/- EMT6 tumour bearing mice treated with irradiated EMT6 cells after primary tumour resection with and without daily administration of metformin were stained for CD45, CD3, and CD8 cell surface expression. Data show mean (±SD) of 2-4 mice/group. *p<0.005). f) Increased CTL activity in splenocytes of CD200-/- mice after metformin treatment (x21d), 4 mice/group.

anticancer agent, and, when administered together with standard chemotherapy, it has been reported to improve the outcome in a number of human cancers407.

Primary tumours in female CD200-/- BALB/c mice were resected 15 days after EMT6 tumour cell injection. Two days later, mice were immunized with irradiated EMT6 cells and treated with i.p. metformin injections daily. The combination of metformin with immunotherapy and lack of CD200 expression in the host resulted in increased survival and prevention of lung/liver metastasis compared with either treatment alone (Fig. 3-5d). This was correlated with an increase in the number of CD8+ T cells in the DLN of CD200-/- mice treated with surgery,

98

Figure 3-6 Verteporfin negates the effect of metformin on EMT6 tumour growth and metastasis in CD200-/- mice. a) Single cell suspensions of DLN from CD200-/- EMT6 tumour bearing mice treated with irradiated EMT6 cells after primary tumour resection with and without daily

99 administration of metformin and verteporfin (an autophagy inhibitor) were stained for CD45, CD3, and CD8 cell surface expression. b) CTL activity against EMT6 tumour cells in splenocytes of CD200-/- mice (50:1 effector:target) in mice treated with metformin is decreased by treatment with verteporfin. c) Increased tumour metastasis to lungs in immunized CD200-/- mice treated with metformin as well as with verteporfin. Data show mean (±SD) of 2-4 mice/group. *p<0.05; **p<0.005; ***p<0.0005.

immunotherapy, and metformin (Fig. 3-5e). Moreover, CD8+ T cells from metformin-treated mice showed significantly enhanced tumour cell cytotoxicity (Fig. 3-5f). Importantly, the combination of treatments was effective in increasing host anti-tumour immunity only in

CD200-/- mice and not in WT mice, which succumbed to EMT6 breast cancer metastasis regardless of treatment (Fig 3-9). We suggest this reflects the continued importance of an increased immune response in CD200-/- mice as well as the supportive effect of metformin in protection. Treatment of mice with an autophagy blocking agent, verteporfin, negated the effect of metformin on EMT6 metastasis to lungs in CD200-/- mice and decreased the CTL activity to baseline (Fig. 3-6). These data suggests that metformin’s mechanism of action in this model involves autophagy by host or tumour cells, which may synergize with immunotherapy to produce an effective anti-tumour immune response.

3.4 Discussion

CD200:CD200R1 interaction has been shown to suppress immune responses in autoimmune disorders, infectious diseases, transplantation, and cancer. CD200-/- mice were first generated on a C57BL/6 (B6) background in 2000341. These mice had an increased susceptibility to autoimmune diseases, like EAE and CIA, consistent with the hypothesis of a general

100

Figure 3-7 Cytokines in the EMT6 and EMT6siCD200 tumour microenvironment a) DLN cells from EMT6 and EMT6siCD200 WT tumour bearing mice were stained for CD45, CD3, CD4, and

101 FOXP3 cell surface expression. Data show mean (±SD) of 5 mice/group. b) Relative levels of 40 inflammatory cytokines in EMT6 and EMT6siCD200 tumours grown in WT mice were assessed by a proteomics array. Data show mean (±SD) optical density (OD) of 3 samples/tumour. c-h) Relative levels of IL-1β, GM-CSF, IL-1α, IL-10, IL-6, and G-CSF in 24hr supernatant from EMT6 and EMT6siCD200 tumours from WT mice cultured in αF10 for 24 hrs were assessed by ELISarray. Data show mean (±SD) of 4 mice/group. *p<0.005.

immunoregulatory role for CD200. CD200-/- B6 mice were also found to be more susceptible to viral infections, such as influenza, where a dose causing non-fatal disease in WT mice now resulted in death408. In the studies described herein, we used newly generated CD200-/- BALB/c mice and WT BALB/c mice as EMT6 and EMT6siCD200 tumour bearing hosts to investigate how the lack of CD200 expression by host and tumour cells affected EMT6 breast cancer progression.

EMT6 tumour growth in CD200-/- hosts did not differ significantly from WT mice, indicating that the lack of CD200 expression in the host is not detrimental to tumour progression, as long as CD200 can be expressed by the tumour cells (Fig. 3-1c,d). Lack of

CD200 expression by EMT6 tumour cells resulted in reduced cancer progression in both WT and CD200-/- mice. This suggests that CD200 expression by tumour cells is more important than host CD200 expression for control of local tumour growth in this breast cancer model, as long as CD200R1 is expressed in the host.

Even though EMT6siCD200 tumours were smaller than control EMT6 tumours (Fig. 3-1c), they had a greater percentage of CD45+ infiltrating immune cells (Fig. 3-2a), suggesting an enhanced immune response in hosts bearing tumours with reduced CD200 expression. CD45+

TILs were enriched in myeloid cells, and there was an increase in CD8+ cells infiltrating the

102

Figure 3-8 In vivo CD4 cell depletion in WT EMT6 tumour bearing mice. a) DLN cells from EMT6 and EMT6siCD200 WT tumour bearing mice treated with a CD4-depleting ascites or PBS were stained for CD45, CD3, CD8, and CD200 cell surface expression. Data show mean (±SD) of 5 mice/group. *p<0.0001. b) DLN cells from EMT6 and EMT6siCD200 WT tumour bearing mice treated with a CD4-depleting ascites or PBS were stained for CD45, GR-1, and CD11b cell surface expression. Data show mean (±SD) of 5 mice/group.

EMT6siCD200 tumours. EMT6siCD200 tumour bearing mice had lower numbers of CD8+ T cells in the DLN, implying an increased trafficking to the periphery and tumours (Fig. 3-2b). In addition, the EMT6 tumour microenvironment contained more inflammatory cytokines, including IL-1β and GM-CSF, than EMT6siCD200 tumours (Fig. 3-2d), which may suggest a role for these cytokines in promoting tumour growth in the EMT6 breast cancer model.

103

Figure 3-9 Metformin does not improve anti-tumour immunity in WT EMT6 tumour bearing hosts. a) Lungs and livers from WT EMT6 tumour bearing mice treated with immunotherapy and/or metformin after surgery were fixed in Bouin's fixative and macroscopic metastases were counted. Data show mean (±SD) of 3-5 mice/group. b) Single cell suspensions of DLN from WT EMT6 tumour bearing mice treated with irradiated EMT6 cells after primary tumour resection with and without daily administration of metformin were stained for CD45, CD3, and CD8 cell surface expression. Data show mean (±SD) of 3-6 mice/group.

104 GM-CSF is a soluble mediator secreted by a variety of cells, including macrophages, granulocytes, T cells, endothelial cells, and fibroblasts, with pleiotropic effects. There is conflicting evidence for tumour promoting and protective roles for this cytokine in breast cancer progression under different conditions409. GM-CSF has been shown to be capable of stimulating an immune response, improving the antigen presenting capacity of DCs, as well as of suppressing an immune response by favouring the development of immature DCs that recruit

+ + + 410-412 CD4 CD25 FOXP3 Tregs . A recent study in breast cancer patients reported an association of elevated GM-CSF levels and EMT and poor prognosis413. In co-culture systems and humanized mice, mesenchymal-like breast cancer cells activate macrophages to a TAM-like phenotype by GM-CSF, and, in turn, CCL18 derived from TAMs can induce cancer cell EMT in a positive feedback loop413. These findings support the hypothesis that GM-CSF has a tumour- promoting role in the EMT6 mouse breast cancer model, although the mechanism of action remains to be investigated.

IL-1β is an inflammation-associated cytokine secreted mainly by macrophages localized in the breast tumour microenvironment414, and its overexpression is thought to be tumour- promoting in many cancer models415. In murine models of breast cancer it has been shown that

IL-1β is involved in various stages of tumour development, invasiveness, and metastasis416, 417.

In addition, IL-1β was previously shown to negatively regulate anti-cancer immune responses through induction of MDSC expansion418. The function IL-1β in the EMT6 breast cancer model is unknown.

To investigate the role of T cells in the adaptive anti-tumour immune response, we studied the progression of EMT6 and EMT6siCD200 tumours in WT mice treated with CD4 and

CD8 depleting antibodies 6 days after tumour cell injection, before adaptive immunity was fully developed. In the previous chapter we reported increased killing of EMT6 tumour cells by

105 lymphocytes from the DLN of EMT6siCD200 tumour bearing mice as well as CD200R1-/- mice.

Consistent with the previous reports, we found that CD8+ T cells are critical in the immune response against tumours that lacked CD200 expression, since depletion of the CD8+ cells caused an increase in tumour size and a decrease in immune cell infiltration (Fig. 3b).

In contrast, depletion of CD4+ T cells resulted in a significant reduction in both EMT6 and EMT6siCD200 tumour size accompanied by an increase in CD45+ TILs (Fig. 3-4b,c), suggesting that the depleted cells worked in an immunosuppressive capacity. We also found a trend to increased numbers of IFN-γ expressing CD8+ T cells in the DLN of EMT6 tumour- bearing mice treated with CD4-depleting antibody with no ex vivo stimulation of the cells. This supports the hypothesis that CD4+ T cells play a regulatory role in EMT6 breast cancer development, which results in a reduced cytotoxic CD8+ T cell mediated anti-tumour responses.

As mentioned above, we previously showed that female CD200R1-/- BALB/c mice could be cured from EMT6 breast cancer by immunization with irradiated EMT6 cells and CpG after surgical resection of the primary tumour406. WT mice treated the same way succumb to EMT6 lung and liver metastases within 50 days. Although CD200-/- mice were able to survive longer than WT hosts, they did not develop full anti-tumour immunity and died within 150 days (Fig.

3-5a-c). Conventional chemotherapies, like paclitaxel and anti-VEGF, can cure EMT6 tumours in WT mice, but these treatments do not produce an effective anti-cancer immune response in the hosts406.

Metformin is a drug commonly used by diabetic patients, but it also has a history of use in oncology419. Numerous observational studies reported decreased cancer incidence and cancer- related mortality in diabetics receiving standard doses of metformin (1500 to 2250 mg/day in adults)420-422. Studies examining all forms of cancer have reported reduced cancer risk in

106 diabetics on metformin (vs. no metformin treatment421, 423) and lower cancer-related mortality in patients receiving metformin compared to those receiving other standard diabetic therapies407,

424. An epidemiological study of 2,529 women with breast cancer reported higher pathologic complete response rates (pCRs; considered a surrogate for overall survival in this setting) to neoadjuvant systemic therapy in diabetic patients receiving metformin compared to diabetic patients not receiving metformin and non-diabetic patients not receiving metformin425. Note that despite reports that metformin can increase pCR, it has not been reported to improve the estimated 3-year relapse-free survival rate.

Our data show an additive protective effect on long term anti-tumour immunity by combining an immunotherapeutic approach with daily metformin administration in WT and

CD200-/- EMT6 tumour bearing mice (Fig. 3-5 and Fig. 3-9). While WT mice still developed lung and liver metastases, CD200-/- mice had cancer-free livers and lungs as well as a significant increase in the number and cytotoxicity of CD8+ T cell in the lymph nodes. This suggests that metformin can help increase anti-tumour responses against EMT6 cells in hosts lacking CD200 expression. One possible mechanism of CD8+ T cell immune response augmentation is through induction of autophagy mediated by metformin. Multiple studies report that metformin activates

AMPK, inhibits mTOR, and promotes autophagy. A recent paper suggests enhanced autophagy is essential for CD8+ T cell memory cell survival426, which could explain the significant increase in the number of CD8+ cells in the DLN of CD200-/- mice treated with metformin after tumour resection and immunization. This notion is supported by our data from an experiment using verteporfin, a drug that inhibits early stages of autophagy without light activation427 (Fig. 3-6).

The mechanism of action of metformin in this CD200-/- model is currently under investigation.

In summary, our data support the hypothesis that CD200 blockade may be an effective therapeutic target in breast cancer. Currently, the majority of immunotherapies in clinical use

107 are adjuvant treatments to chemotherapy or radiation therapy, but there are over 1000 ongoing clinical trials of cancer immunotherapies428. There is a strong rationale to continue developing combinatorial strategies targeting mechanisms of immunosuppression that synergize with standard treatments as well as unconventional drugs, like metformin to improve the outcome for cancer patients.

108

Chapter 4

Soluble CD200 in plasma and serum of human

breast cancer patients

4 Soluble CD200 levels in breast cancer patients

4.1 Abstract

The composition of the tumour microenvironment is important for regulation of metastasis in breast cancer, and targeting immune inhibitory receptors has recently become a focus for activating anti-tumour immune responses in the host. In the EMT6 mouse breast cancer model, expression of the immunoregulatory molecule CD200 on the surface of a subpopulation of cancer cells results in a suppression of anti-tumour immunity and increased metastasis to local draining lymph nodes (DLN). CD200 can also be released by host or tumour cells into the circulation in a soluble form.

In this study, we investigated the levels of sCD200 in human breast cancer patient plasma using an ELISA. We correlated our findings with clinical characteristics of breast cancer and with cell surface CD200 expression by the cells in the breast tumour, as detected by immunostaining. Our data support the hypothesis that CD200 may be important in the progression of breast cancer.

109 4.2 Introduction

Breast cancer is the most common malignancy among Canadian women, accounting for

28% of all new cancer cases, and affecting 1 in 9 women over their lifetime. Thirty percent of women with breast cancer develop metastases despite local and systemic therapies179. Metastasis of breast cancer remains a largely incurable disease and is the major cause of mortality among breast cancer patients429.

Metastatic spread has traditionally been considered a clonal event late in malignant progression, occurring after a single cell within the original tumour acquires genetic mutations that allow for the development of cells with a more aggressive phenotype430. Recent studies, however, show that dissemination of primary cancer cells to distant sites, primarily bone, brain, lungs, and liver, often occurs early during breast cancer development431. Epidemiological analysis of more than 12,000 breast cancer patients indicated that metastasis might be initiated

5–7 years before diagnosis of the primary tumour432. Disseminated tumour cells could be detected in the circulation and bone marrow of patients with the early tumour type DCIS years before metastatic manifestation431, 433.

Many factors regulate metastatic spread of breast cancer cells, both those that are intrinsic to tumour cells and host-associated elements in the tumour microenvironment. The immune system plays an important role in regulating the tumour microenvironment, which contains innate and adaptive immune cells, cancer cells, and their surrounding stroma. The expression of various immune modulators and the abundance and activation state of different cell types in the tumour microenvironment affect the balance between inflammation-promoted tumour growth and anti-tumour immunity3.

110 Regulation of host’s anti-tumour immune responses can occur via immune inhibitory receptors, which deliver signals that override co-stimulation of effector cells138. CD200 is an immunoregulatory molecule that has been found to be overexpressed in various human cancers354-356. Our laboratory described a soluble variant of CD200 (sCD200) in the serum of healthy individuals, which was likely present in the serum following ectodomain shedding371, and which was shown to bind and induce phosphorylation of CD200R1434. We reported that

CLL patients have significantly more sCD200 in their circulation than healthy volunteers, and that sCD200 was critical for engraftment of CLL cells in immunocompromised mice371.

Murine EMT6 breast carcinoma model showed that CD200+ tumour cells exhibit a preferential growth advantage in mice with competent immune systems363, 366, 396. Importantly,

CD200- clonogenic tumour cells grew better in wild type (WT), immunocompetent mice when

CD200+ tumour cells were present363. The CD200+ subset of clonogenic tumour cells was less likely to be rejected. Thus, we hypothesized that CD200+ tumour cells may facilitate the growth of CD200- clonogenic cells. Silencing the expression of CD200 in EMT6 breast cancer cells decreased their rate of tumour growth and metastasis and increased the hosts’ anti-tumour immune responses396. This suggests that CD200:CD200R1 interaction has an important function in immune responses to breast cancer cells in the EMT6 model.

A preliminary assessment of the levels of soluble CD200 and expression of CD200 by host or tumour cells in 30 breast cancer patients suggested that there may be an elevation of sCD200 levels in patients compared with healthy age-matched controls. We now report on a larger, more definitive study using 100 breast cancer patients’ plasma/serum samples obtained from the Ontario Tumour Bank. Circulating levels of sCD200 in plasma/serum samples were tested using an ELISA, and sections of the primary tumour and DLN (if tumour cells were present) were stained for cell surface CD200 expression. The results were correlated with

111 various clinical markers, including node involvement, tumour grade and size, as well as positivity for estrogen receptor (ER), progesterone receptor (PR), and HER2/neu markers.

4.3 Materials and methods

4.3.1 Patient Samples

A small pilot study was performed with 30 breast cancer patients’ sera/plasmas under the approval by the McMaster Faculty of Health Sciences/Hamilton Health Sciences Research

Ethics Board. Signed informed consent was obtained for use of left-over plasma/serum for each patient, and for immunostaining studies of sections of the patients’ breast cancer tissues. Thirty women with documented breast cancer were recruited one month after surgery at the Juravinski

Cancer Centre. Ten healthy volunteer women ranging from 35 to 70 years of age were recruited as age-matched controls. In some instances, paired serum was also obtained.

Encouraged by the apparent evidence that elevated levels of sCD200 may be present in breast cancer patients compared with controls, we extended our observations in a more definitive study using 100 breast cancer patients’ sera/plasmas obtained from the Ontario

Tumour Bank under the approval by the University Health Network (REB: 13-6897-CE).

Normal human plasma was purchased from Bioreclamation (Catalog #HMPLNAHP). Twenty nine healthy volunteers ranging from 39 to 75 years of age were recruited as age-matched controls.

As an internal control, serum from one CLL patient, known to have high levels of sCD200, was used in all ELISA assays at a set dilution of 1:4.

112 4.3.2 Antibodies

The rat monoclonal anti-hCD200 antibody 1B9 was raised against the CD200 fusion protein (CD200Fc), as described previously 435. The polyclonal rabbit anti-hCD200 serum

(Rb846) was generated following immunization of rabbits with CD200Fc and subsequent boosting with the lysate from HEK293 cells stably transfected with human CD200364. Both antibodies were exhaustively absorbed to remove any anti-Fc activity as assessed by ELISA.

4.3.3 ELISA of soluble CD200 in plasma

ELISA was performed as described previously333. Briefly, ELISA plates were coated with the monoclonal anti-CD200 Ab 1B9 at 1.25 ng/l in a Tris HCL coating buffer (pH 8).

Plasma samples at a 1:2 and 1:4 dilutions in blocking buffer were incubated O/N at 4oC with gentle rocking. A standard curve was prepared by serial dilutions of CD200Fc, beginning from

8000 pg/ml. The rabbit polyclonal anti-CD200 Ab Rb846 was used as detection at a 1:2000 dilution, for 45 min at 37oC. Anti-rabbit IgG-horseradish peroxidase (IgG-HRP) was used at

1:15000 dilution (30 min at 37oC), and TMB substrate was used for developing. The reaction was stopped with 2M HCl, and the spectrophotometric absorbance was measured at 456 nm. To account for inter-assay variability, serum from a chronic lymphocytic leukemia (CLL) patient with high concentration of soluble CD200 was used in each assay for normalization of data. The result of simultaneous assay of serum and plasma gave the same result, and as plasma was easier to obtain from material left over after routine blood tests in breast cancer patients, we used plasma for this study.

4.3.4 Immunological staining

Formalin-fixed and paraffin-embedded (FFPE) tissue slides (10µm) from paired breast cancer and normal tissues were purchased from the Ontario Tumour Bank. Staining with the rabbit anti-CD200 (Rb846) Ab was performed at Mount Sinai Hospital. The immunostaining

113 technique was validated using a human lymphoma/leukemia sample known to be strongly positive in flow cytometry for CD200364.

4.3.5 mRNA extraction from formalin fixed paraffin embedded (FFPE) tissue sections and qRT-PCR

mRNA was extracted from FFPE tissue sections of primary breast cancer as well as from slides coated with CLL cells (as positive control) using the RecoverAll Total Nucleic Acid

Isolation kit for FFPE (from Life Technologies Cat. # AM1975). Quality of the mRNA was assessed by 260:280 and 260:230 ratios (above 1.5). RNA was converted to cDNA, and qPCR

(Roche Machine) was done with up to 10 ng of cDNA template using the following primers with Sybr Green:

HPRT (housekeeping) sense 5’-CAAGCTTGCTGGTGAAAAGGA-3’

antisense 5’-TGAAGTATTCATTATAGTCAAGGGCATA-3’

CD200 sense 5’-GGAAGCCCTCATTGTGACAT-3’

antisense 5’-TATAGGCAGGCTGGATCACC-3’

CD200R1: sense 5’-CCATTTGACTGGCAACAAGA-3’

antisense 5’-GCAGCCATTGACTTTCAACA-3’

4.3.6 Statistics

Duplicate patient values were used to calculate the mean. The significance of correlations was determined using the unpaired two-tailed t-test, one-way ANOVA, or Fisher’s

Exact test where appropriate. All tests were done with Prism 5.0d software (Graph Pad

Software, Inc, San Diego, CA, USA).

114 4.4 Results

4.4.1 Levels of sCD200 in plasma of 30 breast cancer patients

compared with 10 healthy age-matched women

Plasma from 10 healthy female volunteers ages 35 to 70 was used to establish a baseline level of sCD200 in healthy women. As shown in Fig. 4-1a, the level of sCD200 in healthy subjects was determined to be 516 ± 236 pg/ml, while plasma of breast cancer patients (n=30) had elevated sCD200 levels (1066 pg/ml ± 365 pg/ml; p<0.0001). However, in a subsequent study using the same sCD200 ELISA and commercially available normal pooled plasma from healthy subjects, the concentration of sCD200 was found to be 1333.75 ± 284.5 pg/ml.

In this pilot study with 30 patients, no significant correlation was found between levels of sCD200 and breast cancer stage, tumour size, node positivity, and ER, PR, or HER2/neu positivity (Fig. 4-1b). Additional studies (unpublished) showed no detectable CD200R1 mRNA could be extracted from slides of tumour samples, despite the presence of CD200 mRNA (using

HPRT as housekeeping control). There was, in addition, data supporting the notion that positive tumour CD200 staining correlated with elevated levels of sCD200 in patient plasma (p=0.0079).

These combined observations of sCD200 plasma levels and CD200 tumour expression provided the impetus for us to extend these studies to a larger sample of breast cancer patients from the

Ontario Tumour Bank.

115

Figure 4-1 Soluble CD200 levels in plasma of 30 breast cancer patients compared with 10 age- matched controls. a) Ten healthy women (average level of sCD200 516 ± 236 pg/ml) had significantly lower levels of sCD200 than 30 breast cancer patients (average level of sCD200 1066 ± 365 pg/ml) (p<0.0001) when tested at 1:4 dilution. b) Breast tumour size, and ER, PR, and HER2/neu positivity were correlated with the levels of sCD200 in plasma of patients (NS)

4.4.2 Levels of sCD200 in plasma and serum samples from 100 breast

cancer patients correlated with clinical characteristics of cancer

Although an apparent difference in sCD200 expression levels was seen between patient and control subjects, there was a significant caveat to the observations in the pilot study.

Importantly, only 10 control subjects were available for testing, and sCD200 levels in a commercial plasma sample pooled across a large number of subjects did not differ significantly from the patient samples. Accordingly, we expanded the study to include 90 plasma and 10 serum samples from patients with invasive ductal breast carcinoma from the Ontario Tumour

116 Bank, using 29 new plasma samples from age-matched healthy control volunteers. In this repeat analysis, the level of sCD200 in 100 breast cancer patients was 1168 ± 484 pg/ml, similar to that recorded in the pilot study with 30 patients (Figs. 4-1a/4-2a). However, the level of sCD200 detected in the plasma of 29 control subjects was now 1475 ± 708 pg/ml, a level not significantly different from the breast cancer patients’ plasma (Fig. 4-2a,b).

There was a large variability in the levels of sCD200 within our sample of 100 breast cancer patients, with concentrations ranging from 210 pg/ml to 2812 pg/ml (Fig. 4-2a). To investigate whether sCD200 levels correlate with any clinical characteristics of cancer, we segregated the patients by age, tumour size, cancer grade, lymph node involvement, as well as

ER, PR, and HER2/neu status. There was a general increase in sCD200 concentration in plasma/serum of younger women with breast cancer (p=0.0332, Fig. 4-3a). No significant difference in the levels of sCD200 was found between T1, T2, and T3 tumours (Fig. 4-3b).

Correlation of sCD200 levels with tumour grade, node involvement, and metastasis did not show any significant differences (Fig. 4-3c,d,e). Segregation of patients by positivity for ER,

PR, and HER2/neu expression also did not indicate any significant differences in levels of circulating sCD200 (Fig. 4-4).

4.4.3 Comparison of mRNA expression and cell surface staining for CD200 and CD200R in breast tumour samples and adjacent normal tissue

The presence of CD200 mRNA in patient tumour FFPE tissue sections was confirmed by qPCR, using HPRT as a housekeeping control (Fig. 4-5a). There was no detectable

CD200R1 mRNA in the tumour sections. There was no significant association between levels of sCD200 the circulation and relative CD200 mRNA expression in the breast tumour (Fig. 4-5b).

117

Figure 4-2 Soluble CD200 levels in plasma/serum of 100 breast cancer patients compared with normal plasma. a) Twenty nine plasma samples from healthy age-matched controls had 1475 ± 708 pg/ml of sCD200. Ten serum samples and 90 plasma samples from breast cancer patients tested at 1:2 dilution had 1168 ± 484 pg/ml sCD200. b) sCD200 levels in normal plasma from the initial study (n=10, 516 ± 236 pg/ml) were lower compared with normal plasma from the larger study (n=29, 1475 ± 708 pg/ml).

4.5 Discussion

In this study, we explored whether the level of soluble CD200 in the plasma of breast cancer patients was increased compared to healthy controls, and whether these levels were correlated with the tumour characteristics and disease stage. Our initial hypothesis was that an elevated level of sCD200 in plasma of cancer patients may define a subset of patients with a more aggressive disease course.

118

Figure 4-3 Correlation of sCD200 levels with clinical characteristics of breast cancer in 100 patient samples. a) Levels of sCD200 in patients’ plasma/serum segregated by women’s ages showed that younger breast cancer patients tend to have higher levels of sCD200 (p<0.05 by one-way ANOVA with Bonferroni’s post-hoc test). b) Patients with T1, T2, and T3 breast tumours had similar levels of sCD200 (T1, 1161 ± 429 pg/ml; T2, 1141 ± 526 pg/ml; T3, 1200 ± 249 pg/ml). c) Patients with Grade I, II, and III breast tumours had similar levels of sCD200 (Grade I, 1050 ± 361 pg/ml; Grade II, 1136 ± 583 pg/ml; Grade III, 1185 ± 444 pg/ml). d)

119 Patients with N0, N1, N2, and N3 breast tumours had similar levels of sCD200 (N0, 1102 ± 546 pg/ml; N1, 1245 ± 450 pg/ml; N2, 1169 ± 433 pg/ml; N3, 1260 ± 254 pg/ml). e) Patients with M0, M1, and MP breast tumours had slightly increasing levels of sCD200 (M0, 1095 ± 493 pg/ml; M1, 1201 ± 261 pg/ml; MP, 1311 ± 260 pg/ml).

In humans, CD200 is overexpressed in different cancers, including renal, colon, and ovarian carcinomas, melanoma, multiple myeloma, acute myeloid leukemia (AML), and chronic lymphocytic leukemia (CLL)354-356, 371. The expression of CD200 on malignant cells is correlated with poor prognosis in AML355, and a clinical trial of anti-CD200 monoclonal antibody treatment in leukemia and lymphoma patients decreased the size of CD200+ tumours

359 and reduced Treg levels . Human cancer stem cells from prostate, breast, brain, and colon cancers express CD200 on their surface358.

In the murine EMT6 breast cancer model, expression of CD200 by tumour cells promotes the development of tumour deposits in the draining lymph nodes (DLN). CD200tg mice, which systemically overexpress CD200, show increased metastasis of tumour cells to

DLN, and CD200- cells had a metastatic advantage when CD200+ cells were present366.

However, it is still not known whether the systemic overexpression of CD200 in the host favours the metastasis of clonogenic tumour cells to DLN.

CD200Fc, a soluble chimeric protein in which the extracellular domains of CD200 were linked with a murine IgG2aFc region, increased lung and liver metastasis of three human breast cancer lines tested in immunocompromised mice (MCF7; HTB19; MDA-MB-231).

Furthermore, tumour metastasis was associated with changes in gene expression in tumour tissue, with significant alterations in genes encoding matrix metalloproteinases (MMPs) and transcription factors modulating inflammation372. Therefore, we hypothesized that CD200+

120

Figure 4-4 Correlation of circulating sCD200 levels with ER, PR, and HER2/neu status in 100 breast cancer patients. a) Patients with ER+ breast tumours have a slightly higher level of sCD200 than patients with ER- tumours (ER+, 1196 ± 496 pg/ml; ER-, 1019 ± 481 pg/ml). b) Patients with PR+ and PR- breast tumours have similar levels of sCD200 (PR+, 1188 ± 522 pg/ml; PR-, 1136 ± 414 pg/ml). c) Patients with HER2/neu+ and HER2/neu- breast tumours have similar levels of sCD200 (HER2/neu+, 1134 ± 458 pg/ml; HER-, 1131 ± 476 pg/ml).

breast cancer cells in women might also release sCD200 to attenuate the anti-tumour immune response.

121 While our initial pilot study suggested that our hypothesis might have some merit, a more extensive analysis using 100 patient samples and 29 control subjects showed no difference in sCD200 levels, even in subsets of patients with aggressive disease and metastasis. However, analysis of CD200 and CD200R expression in the tumour microenvironment by cancer cells or tumour infiltrating cells may indicate that local CD200 expression and not circulating sCD200 levels in plasma is a better indicator of disease (Fig. 4-5).

There are several recent studies reporting on the expression of sCD200 in the peripheral circulation of patients with various diseases. A recent paper by Gumuslu et al supports our findings by also reporting no differences in sCD200 levels from patients with invasive ductal carcinoma (pre- and post- treatment) compared with healthy volunteers436. Our lab showed that sCD200 may be involved in bone remodeling under bed rest conditions437. In SLE patients, the number of CD200+ cells and the level of sCD200 were higher than in healthy controls, but serum CD200 level did not correlate with the clinical disease activity markers438. The level of sCD200 was elevated in serum of patients with type 2 diabetic foot439 and bullous pemphigoid440 compared with healthy controls. A small study in patients with non-small cell lung cancer reported that the overall survival correlated with sCD200 levels441. In patients with brain tumours, glioblastoma patients’ sCD200 level was significantly elevated and correlated with MDSC expansion442. A recent abstract reported that sCD200 levels in peritoneal fluids from metastatic endometrial and ovarian carcinomas and in serum from metastatic breast cancer were not significantly different443. These inconsistent data indicate that, other than in blood- related cancers (like CLL and AML), the association of sCD200 with disease severity is unclear, and that in solid tumours local expression of cell-surface CD200 may be a better prognostic marker in some cases.

122

Figure 4-5 Relative local CD200 expression in breast tumours. a) Real time PCR performed on tumour FFPE tissues from breast cancer patients. Sections from patients with low sCD200 levels (<700 pg/ml, n=4) and high sCD200 levels (>1700 pg/ml, n=4). CLL cells serve as a positive control for CD200 expression. The relative gene expression was normalized against HPRT housekeeping gene. Data represented as geometric mean. b) Levels of sCD200 plotted against CD200 mRNA expression in tumours of breast cancer patients from panel a). Slope does not significantly deviate from zero (r2 = 0.1149, p=0.4). c) Relative CD200 mRNA expression in paired tumour and adjacent normal tissue from three representative patients

123 It is important to note that other factors besides tumour growth and affect sCD200 levels.

Inflammation, infection, and other abnormalities in immune homeostasis, including drug ingestion, can alter CD200 expression. We have no detailed independent information on the medical status of either the cancer patients or the control subjects in this regard (Table 4-1).

124

4.6 Tables

125

126

Chapter 5

General Discussion and Future Directions

An immunosuppressive role for CD200:CD200R interaction has been demonstrated in a variety of mouse models, including allograft transplantation, autoimmunity, infection, and cancer. The data in this thesis provide further evidence for an immunoregulatory function of

CD200, a widely expressed cell-surface and soluble immunoregulatory molecule, and answer several key questions about the mechanism of action of CD200:CD200R interaction in breast cancer.

Firstly, we showed that removing CD200 from the surface of EMT6 tumour cells leads to a reduction of tumour growth and metastasis in BALB/c female mice. Secondly, we found that removing CD200R1 from host cells inhibits the CD200-induced immunosuppression and allows for a more robust anti-EMT6 tumour immune response that can be adoptively transferred to naïve tumour hosts. Thirdly, we determined that depletion of CD4+ T cells resulted in increased protection from EMT6 tumour growth, suggesting that they may play an immunoregulatory role in this breast cancer model. Finally, we showed that mice lacking

CD200 expression can be cured from EMT6 breast cancer metastasis after treatment with immunotherapy and metformin.

Patients with solid tumours, such as breast cancer, are not usually systemically immunocompromised (unless they are getting chemotherapy and/or radiation treatment) and can mount immunological responses against infectious antigens and vaccinations444. Inside the tumour microenvironment, however, antigen processing and presentation by APCs as well as anti-tumour functions of effector cells are often suppressed by signaling cascades between

127 tumour cells and the surrounding stroma (Fig. 1-2). Although elevated levels of circulating sCD200 are associated with immunosuppression and poor prognosis in patients with hematopoietic cancers like AML and CLL, we did not find the levels of sCD200 in the circulation of breast cancer patients to be significantly different from healthy age-matched volunteers, or to be associated with clinical characteristics of breast cancer (Fig. 4-2a, 4-3, 4-4).

However, CD200 was expressed in the local breast tumour microenvironment (Fig. 4-5) at higher levels than in normal breast tissue, which suggested that it may play a role in mediating suppression of local anti-tumour immune responses. To investigate this phenomenon, we developed several mouse models of breast cancer with altered CD200 or CD200R1 expression by BALB/c female hosts and by the mouse-derived EMT6 breast tumour cells, which normally upregulate CD200 cell-surface expression in immunocompetent mice (Fig. 1-6).

It was previously reported that overexpression of CD200 transgene in EMT6 cells

(EMT6CD200tg) increases tumour growth and metastasis to DLN in WT BALB/c mice. In addition, overexpression of CD200 in the host (CD200tg mice) led to increased growth of EMT6 tumours regardless of CD200 expression by tumour cells363. These results supported our hypothesis that CD200 expressed by the host or tumour cells may have immunosuppressive functions that help control tumour progression and metastasis.

The lack of CD200R1 expression in the host (CD200R1-/- mice) resulted in a significant decrease in local EMT6 tumour growth and metastasis to the DLN, even when all of the EMT6 cells in the tumour mixture were overexpressing CD200 (Fig. 2-1), which indicated a critical role for CD200R1 in CD200-mediated immunosuppression in this breast cancer model. Even when only 25% of the tumour cell mixture was expressing CD200 in vivo, we observed a significant increase in primary tumour growth and metastasis to the DLN in WT BALB/c mice,

128 suggesting that a subset of CD200+ cells in the tumour is capable of improving the outgrowth of the whole tumour cell population (Fig. 2-3, 2-5).

The lack of CD200 expression by EMT6 tumour cells (EMT6siCD200) led to a growth disadvantage in both WT, CD200tg, and CD200-/- mice (Fig. 2-3, Fig. 3-1), but these cells still metastasized to the DLN. The lack of CD200 expression in the host (CD200-/- mice) did not result in a significant alteration of EMT6 and EMT6siCD200 tumour growth compared with WT mice (Fig. 3-1). This suggests that, in this model, CD200 expression by EMT6 tumour cells has a greater influence on tumour growth than CD200 expression by host cells. In the future, it would be informative to study the growth of EMT6 tumours expressing various levels of CD200 in CD200R1-/- hosts overexpressing CD200 (CD200R1-/-CD200tg mice) to gain further insight into the mechanism of action of CD200 in EMT6 breast cancer.

EMT6siCD200 tumours had a greater amount of infiltrating lymphocytes, which correlated with a decrease in the number of CD8+ T cells in the DLN of EMT6siCD200 tumour bearing mice, implying increased trafficking of T cells to tumours not expressing CD200 (Fig. 3-2). CD200 has been found to alter cytokine production towards the TH2 response by lymphocytes and macrophages in a variety of mouse models344, which may influence immune cell recruitment to the tumour site and enhance metastasis to the DLN. In addition, CD200R mRNA is expressed by cells of the high endothelial venules (HEVs), which may directly affect CD200+ cell migration in and out of the breast tumour microenvironment, and this phenomenon needs to be investigated further. Control EMT6 tumours had a higher concentration of several inflammation associated cytokines and chemokines, including GM-CSF and IL-1β, but the role of these soluble mediators in EMT6 breast cancer progression has not yet been identified.

129 Depletion of CD4+ T cells in WT BALB/c mice resulted in a significant reduction in

EMT6 and EMT6siCD200 tumour growth and an increased immune cell infiltration (Fig. 3-4), and

EMT6 tumours had a greater amount of infiltrating CD4+CD25+ T cells in WT hosts (Fig. 3-7a).

These data support the hypothesis that CD4+ T cells have a regulatory role in this model, but these cells have yet to be characterized for FOXP3 expression and immunosuppressive function.

The protective role of CD4+ T cells in cancer has been controversial. CD4+ T cells are known to facilitate the optimal expansion, trafficking, effector, and memory functions of CD8+ T cells.

This can happen through direct cell interaction (through CD40:CD154) as well as by secreting soluble factors and conditioning DC to increase their ability to stimulate CD8+ T cells445, 446. In addition to helping CD8+ T cells, CD4+ T cells are also essential for B cell antibody class switching and activating phagocytic activity of macrophages to induce and maintain destructive immune responses to self-antigens, including tumour antigens. On the other hand, the intensity and duration of T cell responses is negatively modulated by certain regulatory subsets of CD4+

T cells, such as Treg and TH17 cells, and these cells have been shown to compromise anti-tumour responses. Regulatory T cells (CD4+CD25+FOXP3+) can prevent the activation and expansion

+ + of both CD4 (TH1) and CD8 T cells through secretion of suppressive cytokines, such as IL-10 and TGF-β, or by restraining the antigen-presenting capacity of DCs in the tumour

+ microenvironment. In breast cancer, FOXP3 Tregs in the peripheral circulation are associated with advanced disease278, 280, and local expression of FOXP3 inside the breast tumours has been linked to invasion, size, and vascularity279. The data from human patients support the findings from our EMT6 mouse model that CD4+ have a tumour protective role in breast cancer. The immunoregulatory mechanism of action of CD4+ T cells in this breast cancer model remains to be clarified.

130 The adaptive anti-tumour response to EMT6 and EMT6siCD200 tumours was found to be dependent on CD8+ T cells, since depletion of these cells resulted in an increase of tumour size and decreased immune cell infiltration (Fig. 3-3). DLN cells from EMT6siCD200 tumour bearing

WT mice had a greater tumour specific lysis ability than DLN cells from EMT6 tumour bearing mice, indicating increased cytotoxic effector function of cells from the former (Fig. 2-6).

Importantly, DLN cells from CD200R1-/- EMT6 and EMT6siCD200 tumour bearing hosts had significantly greater cytotoxic activity against labeled EMT6 cells in vitro, and adoptive transfer of these DLN cells to secondary WT mice receiving control EMT6 tumour cells markedly reduced both primary tumour size and DLN metastasis in these secondary hosts (Fig. 2-6). The cells responsible for transferring the immunity from CD200R1-/- tumour bearing hosts to naïve recipients have not been characterized.

CD200R1-/- EMT6 tumour bearing hosts were successfully cured (no sign of metastasis or disease past 300 days) after surgical resection of the primary tumour and immunotherapy consisting of irradiated EMT6 cells with CpG as adjuvant397. CD200-/- tumour bearing hosts receiving the same treatment survived longer than WT mice but still succumbed to metastasis and did not survive past 150 days. Therefore, CD200-/- mice became an ideal host for studying the synergistic effect of immunotherapy with chemotherapy for curing breast cancer.

Conventional chemotherapy, such as paclitaxel and anti-VEGF, was effective in curing WT mice bearing EMT6 tumours but did not produce an anti-tumour immune response, and mice succumbed to breast cancer after re-challenge with EMT6 cells406.

Metformin is a non-conventional chemotherapeutic agent that, in addition to effects on host metabolism, may affect host immune cells. When CD200-/- EMT6 tumour bearing mice were treated with metformin after surgical tumour resection and immunotherapy, there was an an additive protective effect on long term anti-tumour immunity (Fig. 3-5). CD200-/- mice

131 treated with this combinatorial approach were metastasis-free and had increased numbers of

CD8+ T cells in the DLN that exhibited enhanced cytotoxic activity against labeled EMT6 tumour cells. These data support the notion that disrupting mechanisms of immunosuppression can synergize with standard treatments as well as with unconventional drugs, like metformin, to improve outcomes in breast cancer. It is unknown whether a similar reduction in EMT6 tumour growth and metastasis in WT mice may be achieved with immunotherapy, metformin, and anti-

CD200 mAb (to mimic the environment in the CD200-/- host) after primary tumour resection.

Also, there may be an association between metformin use by breast cancer patients and CD200 expression levels in circulation or in the local tumour microenvironment.

The mechanism of action of metformin in this model is still unclear. Our preliminary data show that verteporfin, an early stage autophagy inhibitor, negates the effect of metformin on EMT6 growth and metastasis. This is in line with the idea that metformin-induced autophagy may have an effect on CD8+ memory T cell activation426 in CD200-/- EMT6 tumour-bearing mice. Autophagy is upregulated in response to stresses like starvation, hypoxia, and accumulation of damaged organelles or misfolded proteins. The catabolic activity of autophagy, which involves the engulfment and delivery of cytosolic contents to lysosomes for degradation, is essential for cellular homeostasis and has been suggested to be inversely correlated with cell growth and proliferation426, 447. Autophagy has been linked to various pathologies including cancer and to the cellular response to anti-cancer therapies. Several preclinical studies have demonstrated that genetic or pharmacological inhibition of autophagy can enhance drug- and radiation-induced cytotoxicity in cell culture and in vivo in poorly vascularized and hypoxic tumours, including pancreatic, breast, and colon cancers427, 448, 449. In contrast to this paradigm, it has been reported (mostly from in vitro studies) that autophagy is upregulated in proliferating T cells, in the presence of positive mTOR signaling following TCR stimulation445. Little is known

132 about the in vivo autophagy activity in antigen-specific T cells during the course of the differentiation of effector and memory T cells in cancer, but a recent paper reported that autophagy was required for the formation of memory CD8+ T cells during an acute viral infection426. It is possible that in our EMT6 breast cancer model, metformin-induced increase in autophagy also enhances the activation of memory T cells and allows for the formation of a more robust anti-tumour immune response.

The goal of cancer immunotherapy is to induce active immune responses against tumour antigens leading to the elimination of disseminated tumours and induction of immunologic memory to prevent tumour recurrence450. Most tumours have evolved mechanisms to evade immune recognition, including downregulation of tumour antigen expression, impaired antigen presentation, inhibition of T-cell signaling, and secretion of immunosuppressive cytokines that skew the immune response toward tolerance rather than activation137, 451, 452. Multiple strategies to overcome this tumour-associated immune tolerance and promote optimal B and T cell activation have been explored.

Breast cancer cells continuously undergo immunosurveillance, as evidenced by altered expression of MHC genes in cancer cells and by the association of MHC-I expression with breast cancer452, 453. Lymphocyte infiltration into breast tumours is correlated with improved overall survival454, and peripheral blood of breast cancer patients shows evidence of cellular and humoral immunity to tumour associated antigens found in human breast cancer, including

MUC-1 and HER2/neu455, 456. Cancer vaccines with HER2/neu peptides and other antigenic moieties have yielded limited success, and this is thought to be linked with the induction of Tregs after vaccination457.

133 A newer approach to cancer immunotherapy targets T cell inhibitory pathways, such as

CTLA4 and PD-1458. Blocking immune inhibitory interactions can skew the immune responses towards activation instead of tolerance and interfere with the evasion of tumour cells from immune effector cells. Based on the data in this thesis, blocking the CD200:CD200R1 pathway may be an important component of achieving an anti-tumour immune response in breast cancer patients, combined with additional antigenic stimulation and adjuvants (like CpG or BCG).

Compounding the complexity of understanding the role of immunotherapy in breast cancer treatment is the potential effect of concomitant chemotherapy on the immune system of the tumour host406. Our studies using metformin as a non-conventional chemotherapeutic in combination with cancer immunotherapy, which included irradiated EMT6 cells as a source of antigen, CpG as adjuvant, and altered CD200 signaling (in CD200-/- mice), resulted in a synergistic protective effect on long term anti-tumour immunity. Therefore, there is a strong rationale to continue the bench-to-bedside development of combinatorial treatment strategies that use chemotherapeutic drugs to target the tumour cells directly as well as cancer immunotherapies, which can lift the tumour-induced immune suppression and improve outcome for breast cancer patients.

134

References

1. Whiteside TL. Immune suppression in cancer: effects on immune cells, mechanisms and future therapeutic intervention. Semin Cancer Biol 2006;16:3-15.

2. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. Volume 144. United States: 2011 Elsevier Inc, 2011:646-74.

3. Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and cancer. Cell 2010;140:883-99.

4. Aggarwal BB, Vijayalekshmi RV, Sung B. Targeting inflammatory pathways for prevention and therapy of cancer: short-term friend, long-term foe. Clin Cancer Res 2009;15:425-30.

5. Coussens LM, Werb Z. Inflammation and cancer. Nature 2002;420:860-7.

6. Balkwill F, Mantovani A. Inflammation and cancer: back to Virchow? Lancet 2001;357:539-45.

7. Dvorak HF. Tumors: wounds that do not heal. Similarities between tumor stroma generation and wound healing. N Engl J Med 1986;315:1650-9.

8. Karin M, Lawrence T, Nizet V. Innate immunity gone awry: linking microbial infections to chronic inflammation and cancer. Cell 2006;124:823-35.

9. Sica A. Role of tumour-associated macrophages in cancer-related inflammation. Exp Oncol 2010;32:153-8.

10. Blankenstein T, Coulie PG, Gilboa E, et al. The determinants of tumour immunogenicity. Nat Rev Cancer 2012;12:307-13.

11. Tredan O, Galmarini CM, Patel K, et al. Drug resistance and the solid tumor microenvironment. J Natl Cancer Inst 2007;99:1441-54.

12. Hamzah J, Jugold M, Kiessling F, et al. Vascular normalization in Rgs5-deficient tumours promotes immune destruction. Nature 2008;453:410-4.

13. Buckanovich RJ, Facciabene A, Kim S, et al. Endothelin B receptor mediates the endothelial barrier to T cell homing to tumors and disables immune therapy. Nat Med 2008;14:28-36.

14. Bhowmick NA, Neilson EG, Moses HL. Stromal fibroblasts in cancer initiation and progression. Nature 2004;432:332-7.

15. Kalluri R, Zeisberg M. Fibroblasts in cancer. Nat Rev Cancer 2006;6:392-401.

135 16. Shimoda M, Mellody KT, Orimo A. Carcinoma-associated fibroblasts are a rate-limiting determinant for tumour progression. Semin Cell Dev Biol 2010;21:19-25.

17. Orimo A, Gupta PB, Sgroi DC, et al. Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF- 1/CXCL12 secretion. Cell 2005;121:335-48.

18. Ao M, Franco OE, Park D, et al. Cross-talk between paracrine-acting cytokine and chemokine pathways promotes malignancy in benign human prostatic epithelium. Cancer Res 2007;67:4244-53.

19. !!! INVALID CITATION !!!

20. Ahmed Z, Bicknell R. Angiogenic signalling pathways. Methods Mol Biol 2009;467:3- 24.

21. Carmeliet P, Jain RK. Angiogenesis in cancer and other diseases. Nature 2000;407:249- 57.

22. Nagy JA, Chang SH, Shih SC, et al. Heterogeneity of the tumor vasculature. Semin Thromb Hemost 2010;36:321-31.

23. Ruoslahti E. Specialization of tumour vasculature. Nat Rev Cancer 2002;2:83-90.

24. Baeriswyl V, Christofori G. The angiogenic switch in carcinogenesis. Seminars in Cancer Biology 2009;19:329-337.

25. Bergers G, Benjamin LE. Tumorigenesis and the angiogenic switch. Nat Rev Cancer 2003;3:401-410.

26. Bergfeld S, DeClerck Y. Bone marrow-derived mesenchymal stem cells and the tumor microenvironment. Cancer and Metastasis Reviews 2010;29:249-261.

27. Karnoub AE, Dash AB, Vo AP, et al. Mesenchymal stem cells within tumour stroma promote breast cancer metastasis. Nature 2007;449:557-63.

28. Bergers G, Song S. The role of pericytes in blood-vessel formation and maintenance. Neuro Oncol 2005;7:452-64.

29. Gaengel K, Genove G, Armulik A, et al. Endothelial-mural cell signaling in vascular development and angiogenesis. Arterioscler Thromb Vasc Biol 2009;29:630-8.

30. Pietras K, Ostman A. Hallmarks of cancer: interactions with the tumor stroma. Exp Cell Res 2010;316:1324-31.

31. Bergers G, Song S, Meyer-Morse N, et al. Benefits of targeting both pericytes and endothelial cells in the tumor vasculature with kinase inhibitors. The Journal of Clinical Investigation 2003;111:1287-1295.

136 32. Franco M, Roswall P, Cortez E, et al. Pericytes promote endothelial cell survival through induction of autocrine VEGF-A signaling and Bcl-w expression. Blood 2011;118:2906- 2917.

33. Raza A, Franklin MJ, Dudek AZ. Pericytes and vessel maturation during tumor angiogenesis and metastasis. Am J Hematol 2010;85:593-8.

34. Gerhardt H, Semb H. Pericytes: gatekeepers in tumour cell metastasis? J Mol Med (Berl) 2008;86:135-44.

35. Cheng L, Huang Z, Zhou W, et al. Glioblastoma stem cells generate vascular pericytes to support vessel function and tumor growth. Cell 2013;153:139-52.

36. Fang S, Salven P. Stem cells in tumor angiogenesis. Journal of Molecular and Cellular Cardiology 2011;50:290-295.

37. Lamagna C, Bergers G. The bone marrow constitutes a reservoir of pericyte progenitors. Journal of Leukocyte Biology 2006;80:677-681.

38. Gordon S, Taylor PR. Monocyte and macrophage heterogeneity. Nat Rev Immunol 2005;5:953-64.

39. Volkman A, Gowans JL. THE ORIGIN OF MACROPHAGES FROM BONE MARROW IN THE RAT. Br J Exp Pathol 1965;46:62-70.

40. Mantovani A, Sozzani S, Locati M, et al. Macrophage polarization: tumor-associated macrophages as a paradigm for polarized M2 mononuclear phagocytes. Trends Immunol 2002;23:549-55.

41. Mills CD, Kincaid K, Alt JM, et al. M-1/M-2 Macrophages and the Th1/Th2 Paradigm. The Journal of Immunology 2000;164:6166-6173.

42. Franklin RA, Liao W, Sarkar A, et al. The cellular and molecular origin of tumor- associated macrophages. Science 2014;344:921-5.

43. Sica A, Schioppa T, Mantovani A, et al. Tumour-associated macrophages are a distinct M2 polarised population promoting tumour progression: potential targets of anti-cancer therapy. Eur J Cancer 2006;42:717-27.

44. Mantovani A, Allavena P, Sica A, et al. Cancer-related inflammation. Nature 2008;454:436-44.

45. Mantovani A, Romero P, Palucka AK, et al. Tumour immunity: effector response to tumour and role of the microenvironment. Lancet 2008;371:771-83.

46. Lin EY, Pollard JW. Tumor-associated macrophages press the angiogenic switch in breast cancer. Cancer Res 2007;67:5064-6.

47. Lewis CE, Hughes R. Inflammation and breast cancer. Microenvironmental factors regulating macrophage function in breast tumours: hypoxia and angiopoietin-2. Breast Cancer Res 2007;9:209.

137 48. Mantovani A, Bottazzi B, Colotta F, et al. The origin and function of tumor-associated macrophages. Immunol Today 1992;13:265-70.

49. Nesbit M, Schaider H, Miller TH, et al. Low-level monocyte chemoattractant protein-1 stimulation of monocytes leads to tumor formation in nontumorigenic melanoma cells. J Immunol 2001;166:6483-90.

50. Van Damme J, Proost P, Lenaerts JP, et al. Structural and functional identification of two human, tumor-derived monocyte chemotactic proteins (MCP-2 and MCP-3) belonging to the chemokine family. J Exp Med 1992;176:59-65.

51. Barleon B, Sozzani S, Zhou D, et al. Migration of human monocytes in response to vascular endothelial growth factor (VEGF) is mediated via the VEGF receptor flt-1. Blood 1996;87:3336-43.

52. Duyndam MC, Hilhorst MC, Schluper HM, et al. Vascular endothelial growth factor-165 overexpression stimulates angiogenesis and induces cyst formation and macrophage infiltration in human ovarian cancer xenografts. Am J Pathol 2002;160:537-48.

53. Lin EY, Nguyen AV, Russell RG, et al. Colony-stimulating factor 1 promotes progression of mammary tumors to malignancy. J Exp Med 2001;193:727-40.

54. Nathan C. Neutrophils and immunity: challenges and opportunities. Nat Rev Immunol 2006;6:173-182.

55. Smith JA. Neutrophils, host defense, and inflammation: a double-edged sword. Journal of Leukocyte Biology 1994;56:672-686.

56. Pham CTN. Neutrophil serine proteases: specific regulators of inflammation. Nat Rev Immunol 2006;6:541-550.

57. Gregory AD, Houghton AM. Tumor-associated neutrophils: new targets for cancer therapy. Cancer Res 2011;71:2411-6.

58. Jensen HK, Donskov F, Marcussen N, et al. Presence of Intratumoral Neutrophils Is an Independent Prognostic Factor in Localized Renal Cell Carcinoma. Journal of Clinical Oncology 2009;27:4709-4717.

59. Wislez M, Rabbe N, Marchal J, et al. Hepatocyte Growth Factor Production by Neutrophils Infiltrating Bronchioloalveolar Subtype Pulmonary Adenocarcinoma: Role in Tumor Progression and Death. Cancer Research 2003;63:1405-1412.

60. Bellocq A, Antoine M, Flahault A, et al. Neutrophil alveolitis in bronchioloalveolar carcinoma: induction by tumor-derived interleukin-8 and relation to clinical outcome. Am J Pathol 1998;152:83-92.

61. Teng MWL, Swann JB, Koebel CM, et al. Immune-mediated dormancy: an equilibrium with cancer. Journal of Leukocyte Biology 2008;84:988-993.

138 62. Kim R, Emi M, Tanabe K. Cancer immunoediting from immune surveillance to immune escape. Immunology 2007;121:1-14.

63. Bindea G, Mlecnik B, Fridman W-H, et al. Natural immunity to cancer in humans. Current Opinion in Immunology 2010;22:215-222.

64. Ferrone C, Dranoff G. Dual Roles for Immunity in Gastrointestinal Cancers. Journal of Clinical Oncology 2010;28:4045-4051.

65. Nelson BH. The impact of T-cell immunity on ovarian cancer outcomes. Immunological Reviews 2008;222:101-116.

66. Pages F, Galon J, Dieu-Nosjean MC, et al. Immune infiltration in human tumors: a prognostic factor that should not be ignored. Oncogene 2009;29:1093-1102.

67. Strauss DC, Thomas JM. Transmission of donor melanoma by organ transplantation. The Lancet Oncology 2010;11:790-796.

68. Cerwenka A, Lanier LL. Natural killer cells, viruses and cancer. Nat Rev Immunol 2001;1:41-9.

69. Ruggeri L, Capanni M, Urbani E, et al. Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants. Science 2002;295:2097-100.

70. O'Hanlon LH. Natural born killers: NK cells drafted into the cancer fight. J Natl Cancer Inst 2004;96:651-3.

71. Borg C, Terme M, Taieb J, et al. Novel mode of action of c-kit tyrosine kinase inhibitors leading to NK cell-dependent antitumor effects. J Clin Invest 2004;114:379-88.

72. Strayer D, Carter W, Brodsky I. Familial occurrence of breast cancer is associated with reduced natural killer cytotoxicity. Breast Cancer Research and Treatment 1986;7:187- 192.

73. Smyth MJ, Godfrey DI, Trapani JA. A fresh look at tumor immunosurveillance and immunotherapy. Nat Immunol 2001;2:293-9.

74. Moretta A, Bottino C, Vitale M, et al. ACTIVATING RECEPTORS AND CORECEPTORS INVOLVED IN HUMAN NATURAL KILLER CELL-MEDIATED CYTOLYSIS. Annual Review of Immunology 2001;19:197-223.

75. Albers AE, Ferris RL, Kim GG, et al. Immune responses to p53 in patients with cancer: enrichment in tetramer+ p53 peptide-specific T cells and regulatory T cells at tumor sites. Cancer Immunol Immunother 2005;54:1072-81.

76. Dunn GP, Bruce AT, Ikeda H, et al. Cancer immunoediting: from immunosurveillance to tumor escape. Nat Immunol 2002;3:991-8.

77. Pardoll D. Does the immune system see tumors as foreign or self? Annu Rev Immunol 2003;21:807-39.

139 78. .

79. Mortarini R, Piris A, Maurichi A, et al. Lack of terminally differentiated tumor-specific CD8+ T cells at tumor site in spite of antitumor immunity to self-antigens in human metastatic melanoma. Cancer Res 2003;63:2535-45.

80. Zippelius A, Batard P, Rubio-Godoy V, et al. Effector function of human tumor-specific CD8 T cells in melanoma lesions: a state of local functional tolerance. Cancer Res 2004;64:2865-73.

81. Broderick L, Brooks SP, Takita H, et al. IL-12 reverses anergy to T cell receptor triggering in human lung tumor-associated memory T cells. Clin Immunol 2006;118:159-69.

82. Cavallo F, De Giovanni C, Nanni P, et al. 2011: the immune hallmarks of cancer. Cancer Immunol Immunother 2011;60:319-26.

83. Shevach EM. Fatal attraction: tumors beckon regulatory T cells. Nat Med. Volume 10. United States, 2004:900-1.

84. Gabrilovich D. Mechanisms and functional significance of tumour-induced dendritic-cell defects. Nat Rev Immunol 2004;4:941-52.

85. Colombo MP, Piconese S. Regulatory T-cell inhibition versus depletion: the right choice in cancer immunotherapy. Nat Rev Cancer 2007;7:880-887.

86. Beyer M, Schultze JL. Regulatory T cells in cancer. Blood 2006;108:804-11.

87. Smyth MJ, Teng MW, Swann J, et al. CD4+CD25+ T regulatory cells suppress NK cell- mediated immunotherapy of cancer. J Immunol 2006;176:1582-7.

88. Gallimore A, Sakaguchi S. Regulation of tumour immunity by CD25+ T cells. Immunology 2002;107:5-9.

89. Onizuka S, Tawara I, Shimizu J, et al. Tumor Rejection by in Vivo Administration of Anti-CD25 (Interleukin-2 Receptor α) Monoclonal Antibody. Cancer Research 1999;59:3128-3133.

90. Turk MJ, Guevara-Patino JA, Rizzuto GA, et al. Concomitant tumor immunity to a poorly immunogenic melanoma is prevented by regulatory T cells. J Exp Med 2004;200:771-82.

91. Schaefer C, Kim GG, Albers A, et al. Characteristics of CD4+CD25+ regulatory T cells in the peripheral circulation of patients with head and neck cancer. Br J Cancer 2005;92:913-20.

92. Curiel TJ, Coukos G, Zou L, et al. Specific recruitment of regulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survival. Nat Med 2004;10:942-949.

140 93. Ghiringhelli F, Puig PE, Roux S, et al. Tumor cells convert immature myeloid dendritic cells into TGF-β–secreting cells inducing CD4+CD25+ regulatory T cell proliferation. The Journal of Experimental Medicine 2005;202:919-929.

94. Zou W. Regulatory T cells, tumour immunity and immunotherapy. Nat Rev Immunol 2006;6:295-307.

95. Bates GJ, Fox SB, Han C, et al. Quantification of regulatory T cells enables the identification of high-risk breast cancer patients and those at risk of late relapse. J Clin Oncol 2006;24:5373-80.

96. Fu J, Xu D, Liu Z, et al. Increased regulatory T cells correlate with CD8 T-cell impairment and poor survival in hepatocellular carcinoma patients. Gastroenterology 2007;132:2328-39.

97. Frey DM, Droeser RA, Viehl CT, et al. High frequency of tumor-infiltrating FOXP3(+) regulatory T cells predicts improved survival in mismatch repair-proficient colorectal cancer patients. Int J Cancer 2010;126:2635-43.

98. Curiel TJ. Tregs and rethinking cancer immunotherapy. J Clin Invest 2007;117:1167-74.

99. Mahmoud SM, Paish EC, Powe DG, et al. An evaluation of the clinical significance of FOXP3+ infiltrating cells in human breast cancer. Breast Cancer Res Treat 2011;127:99- 108.

100. Akhurst RJ, Derynck R. TGF-beta signaling in cancer--a double-edged sword. Trends Cell Biol 2001;11:S44-51.

101. Pollard JW. Tumour-educated macrophages promote tumour progression and metastasis. Nat Rev Cancer 2004;4:71-8.

102. Rochman Y, Spolski R, Leonard WJ. New insights into the regulation of T cells by [gamma]c family cytokines. Nat Rev Immunol 2009;9:480-490.

103. Wing K, Sakaguchi S. Regulatory T cells exert checks and balances on self tolerance and autoimmunity. Nat Immunol 2010;11:7-13.

104. Ernst PB, Garrison JC, Thompson LF. Much Ado about Adenosine: Adenosine Synthesis and Function in Regulatory T Cell Biology. The Journal of Immunology 2010;185:1993-1998.

105. Joyce JA, Pollard JW. Microenvironmental regulation of metastasis. Nat Rev Cancer 2009;9:239-52.

106. de Visser KE, Korets LV, Coussens LM. De novo carcinogenesis promoted by chronic inflammation is B lymphocyte dependent. Cancer Cell 2005;7:411-23.

107. Sahin U, Tureci O, Pfreundschuh M. Serological identification of human tumor antigens. Curr Opin Immunol 1997;9:709-16.

141 108. Hoffmann TK, Donnenberg AD, Finkelstein SD, et al. Frequencies of tetramer+ T cells specific for the wild-type sequence p53(264-272) peptide in the circulation of patients with head and neck cancer. Cancer Res 2002;62:3521-9.

109. Smith HA, Kang Y. The metastasis-promoting roles of tumor-associated immune cells. J Mol Med (Berl) 2013;91:411-29.

110. Lundy SK. Killer B lymphocytes: the evidence and the potential. Inflamm Res 2009;58:345-57.

111. Olkhanud PB, Damdinsuren B, Bodogai M, et al. Tumor-evoked regulatory B cells promote breast cancer metastasis by converting resting CD4(+) T cells to T-regulatory cells. Cancer Res 2011;71:3505-15.

112. Sica A, Porta C, Riboldi E, et al. Convergent pathways of macrophage polarization: The role of B cells. Eur J Immunol 2010;40:2131-3.

113. Harrell MI, Iritani BM, Ruddell A. Tumor-induced sentinel lymph node lymphangiogenesis and increased lymph flow precede melanoma metastasis. Am J Pathol 2007;170:774-86.

114. Ruddell A, Harrell MI, Furuya M, et al. B lymphocytes promote lymphogenous metastasis of lymphoma and melanoma. Neoplasia 2011;13:748-57.

115. Bell D, Chomarat P, Broyles D, et al. In breast carcinoma tissue, immature dendritic cells reside within the tumor, whereas mature dendritic cells are located in peritumoral areas. J Exp Med 1999;190:1417-26.

116. Vermi W, Bonecchi R, Facchetti F, et al. Recruitment of immature plasmacytoid dendritic cells (plasmacytoid monocytes) and myeloid dendritic cells in primary cutaneous melanomas. J Pathol 2003;200:255-68.

117. Zou W, Machelon V, Coulomb-L'Hermin A, et al. Stromal-derived factor-1 in human tumors recruits and alters the function of plasmacytoid precursor dendritic cells. Nat Med 2001;7:1339-46.

118. Kusmartsev S, Gabrilovich DI. Immature myeloid cells and cancer-associated immune suppression. Cancer Immunol Immunother 2002;51:293-8.

119. Diaz-Montero CM, Salem ML, Nishimura MI, et al. Increased circulating myeloid- derived suppressor cells correlate with clinical cancer stage, metastatic tumor burden, and doxorubicin-cyclophosphamide chemotherapy. Cancer Immunol Immunother 2009;58:49-59.

120. Dugast AS, Haudebourg T, Coulon F, et al. Myeloid-derived suppressor cells accumulate in kidney allograft tolerance and specifically suppress effector T cell expansion. J Immunol 2008;180:7898-906.

121. Ochando JC, Chen SH. Myeloid-derived suppressor cells in transplantation and cancer. Immunol Res 2012;54:275-85.

142 122. Bronte V, Wang M, Overwijk WW, et al. Apoptotic death of CD8+ T lymphocytes after immunization: induction of a suppressive population of Mac-1+/Gr-1+ cells. J Immunol 1998;161:5313-20.

123. Kusmartsev SA, Li Y, Chen SH. Gr-1+ myeloid cells derived from tumor-bearing mice inhibit primary T cell activation induced through CD3/CD28 costimulation. J Immunol 2000;165:779-85.

124. Solito S, Marigo I, Pinton L, et al. Myeloid-derived suppressor cell heterogeneity in human cancers. Ann N Y Acad Sci 2014;1319:47-65.

125. Markowitz J, Wesolowski R, Papenfuss T, et al. Myeloid-derived suppressor cells in breast cancer. Breast Cancer Res Treat 2013;140:13-21.

126. Almand B, Clark JI, Nikitina E, et al. Increased production of immature myeloid cells in cancer patients: a mechanism of immunosuppression in cancer. J Immunol 2001;166:678-89.

127. Almand B, Resser JR, Lindman B, et al. Clinical significance of defective dendritic cell differentiation in cancer. Clin Cancer Res 2000;6:1755-66.

128. Bronte V, Apolloni E, Cabrelle A, et al. Identification of a CD11b(+)/Gr-1(+)/CD31(+) myeloid progenitor capable of activating or suppressing CD8(+) T cells. Blood 2000;96:3838-46.

129. Rodríguez PC, Ochoa AC. Arginine regulation by myeloid derived suppressor cells and tolerance in cancer: mechanisms and therapeutic perspectives. Immunological Reviews 2008;222:180-191.

130. Beckman JS, Koppenol WH. Nitric oxide, superoxide, and peroxynitrite: the good, the bad, and ugly. Am J Physiol 1996;271:C1424-37.

131. Nagaraj S, Gupta K, Pisarev V, et al. Altered recognition of antigen is a mechanism of CD8+ T cell tolerance in cancer. Nat Med 2007;13:828-835.

132. Sinha P, Clements VK, Ostrand-Rosenberg S. Reduction of Myeloid-Derived Suppressor Cells and Induction of M1 Macrophages Facilitate the Rejection of Established Metastatic Disease. The Journal of Immunology 2005;174:636-645.

133. Suzuki E, Kapoor V, Jassar AS, et al. Gemcitabine Selectively Eliminates Splenic Gr- 1+/CD11b+ Myeloid Suppressor Cells in Tumor-Bearing Animals and Enhances Antitumor Immune Activity. Clinical Cancer Research 2005;11:6713-6721.

134. Sinha P, Clements VK, Bunt SK, et al. Cross-Talk between Myeloid-Derived Suppressor Cells and Macrophages Subverts Tumor Immunity toward a Type 2 Response. The Journal of Immunology 2007;179:977-983.

135. Quail DF, Joyce JA. Microenvironmental regulation of tumor progression and metastasis. Nat Med 2013;19:1423-37.

143 136. Pardoll DM. Immunology beats cancer: a blueprint for successful translation. Nat Immunol 2012;13:1129-32.

137. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer 2012;12:252-64.

138. Zou W, Chen L. Inhibitory B7-family molecules in the tumour microenvironment. Nat Rev Immunol 2008;8:467-77.

139. Collins M, Ling V, Carreno BM. The B7 family of immune-regulatory ligands. Genome Biol 2005;6:223.

140. Matsuzaki J, Gnjatic S, Mhawech-Fauceglia P, et al. Tumor-infiltrating NY-ESO-1- specific CD8+ T cells are negatively regulated by LAG-3 and PD-1 in human ovarian cancer. Proc Natl Acad Sci U S A 2010;107:7875-80.

141. Ghebeh H, Barhoush E, Tulbah A, et al. FOXP3+ Tregs and B7-H1+/PD-1+ T lymphocytes co-infiltrate the tumor tissues of high-risk breast cancer patients: Implication for immunotherapy. BMC Cancer 2008;8:57.

142. Derre L, Rivals JP, Jandus C, et al. BTLA mediates inhibition of human tumor-specific CD8+ T cells that can be partially reversed by vaccination. J Clin Invest 2010;120:157- 67.

143. Fourcade J, Sun Z, Pagliano O, et al. CD8(+) T cells specific for tumor antigens can be rendered dysfunctional by the tumor microenvironment through upregulation of the inhibitory receptors BTLA and PD-1. Cancer Res 2012;72:887-96.

144. Ribas A. Clinical development of the anti-CTLA-4 antibody tremelimumab. Semin Oncol 2010;37:450-4.

145. Brahmer JR, Drake CG, Wollner I, et al. Phase I study of single-agent anti-programmed death-1 (MDX-1106) in refractory. J Clin Oncol 2010;28:3167-75.

146. Weigelt B, Peterse JL, van 't Veer LJ. Breast cancer metastasis: markers and models. Nat Rev Cancer 2005;5:591-602.

147. Talmadge JE, Fidler IJ. AACR centennial series: the biology of cancer metastasis: historical perspective. Cancer Res 2010;70:5649-69.

148. Klymkowsky MW, Savagner P. Epithelial-mesenchymal transition: a cancer researcher's conceptual friend and foe. Am J Pathol 2009;174:1588-93.

149. Polyak K, Weinberg RA. Transitions between epithelial and mesenchymal states: acquisition of malignant and stem cell traits. Nat Rev Cancer 2009;9:265-73.

150. Yilmaz M, Christofori G. EMT, the cytoskeleton, and cancer cell invasion. Cancer Metastasis Rev 2009;28:15-33.

151. Thiery JP, Acloque H, Huang RY, et al. Epithelial-mesenchymal transitions in development and disease. Cell 2009;139:871-90.

144 152. Kang Y, Massague J. Epithelial-mesenchymal transitions: twist in development and metastasis. Cell 2004;118:277-9.

153. Petersen OW, Nielsen HL, Gudjonsson T, et al. Epithelial to mesenchymal transition in human breast cancer can provide a nonmalignant stroma. Am J Pathol 2003;162:391- 402.

154. Kaplan RN, Rafii S, Lyden D. Preparing the “Soil”: The Premetastatic Niche. Cancer Research 2006;66:11089-11093.

155. Vadrevu SK, Chintala NK, Sharma SK, et al. Complement c5a receptor facilitates cancer metastasis by altering T-cell responses in the metastatic niche. Cancer Res 2014;74:3454-65.

156. Gao D, Joshi N, Choi H, et al. Myeloid Progenitor Cells in the Premetastatic Lung Promote Metastases by Inducing Mesenchymal to Epithelial Transition. Cancer Research 2012;72:1384-1394.

157. Kaplan RN, Riba RD, Zacharoulis S, et al. VEGFR1-positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche. Nature 2005;438:820-827.

158. Sceneay J, Chow MT, Chen A, et al. Primary Tumor Hypoxia Recruits CD11b+/Ly6Cmed/Ly6G+ Immune Suppressor Cells and Compromises NK Cell Cytotoxicity in the Premetastatic Niche. Cancer Research 2012;72:3906-3911.

159. Bogenrieder T, Herlyn M. Axis of evil: molecular mechanisms of cancer metastasis. Oncogene 2003;22:6524-36.

160. Nathanson SD. Insights into the mechanisms of lymph node metastasis. Cancer 2003;98:413-23.

161. Mellstedt H. Monoclonal antibodies in human cancer. Drugs Today (Barc) 2003;39 Suppl C:1-16.

162. Egeblad M, Nakasone ES, Werb Z. Tumors as organs: complex tissues that interface with the entire organism. Dev Cell 2010;18:884-901.

163. Qian BZ, Pollard JW. Macrophage diversity enhances tumor progression and metastasis. Cell 2010;141:39-51.

164. Kalluri R. EMT: when epithelial cells decide to become mesenchymal-like cells. J Clin Invest 2009;119:1417-9.

165. Chaffer CL, Weinberg RA. A perspective on cancer cell metastasis. Science 2011;331:1559-64.

166. Singh A, Settleman J. EMT, cancer stem cells and drug resistance: an emerging axis of evil in the war on cancer. Oncogene 2010;29:4741-51.

145 167. Gocheva V, Wang HW, Gadea BB, et al. IL-4 induces cathepsin protease activity in tumor-associated macrophages to promote cancer growth and invasion. Genes Dev 2010;24:241-55.

168. Ren G, Zhao X, Wang Y, et al. CCR2-dependent recruitment of macrophages by tumor- educated mesenchymal stromal cells promotes tumor development and is mimicked by TNFalpha. Cell Stem Cell 2012;11:812-24.

169. Maier HJ, Schmidt-Strassburger U, Huber MA, et al. NF-kappaB promotes epithelial- mesenchymal transition, migration and invasion of pancreatic carcinoma cells. Cancer Lett 2010;295:214-28.

170. Sullivan NJ, Sasser AK, Axel AE, et al. Interleukin-6 induces an epithelial-mesenchymal transition phenotype in human breast cancer cells. Oncogene 2009;28:2940-7.

171. Fuxe J, Karlsson MC. TGF-beta-induced epithelial-mesenchymal transition: a link between cancer and inflammation. Semin Cancer Biol 2012;22:455-61.

172. Kessenbrock K, Plaks V, Werb Z. Matrix metalloproteinases: regulators of the tumor microenvironment. Cell 2010;141:52-67.

173. Wang Y, Zhou BP. Epithelial-mesenchymal Transition---A Hallmark of Breast Cancer Metastasis. Cancer Hallm 2013;1:38-49.

174. Kamangar F, Dores GM, Anderson WF. Patterns of cancer incidence, mortality, and prevalence across five continents: defining priorities to reduce cancer disparities in different geographic regions of the world. J Clin Oncol 2006;24:2137-50.

175. Walters S, Maringe C, Butler J, et al. Breast cancer survival and stage at diagnosis in Australia, Canada, Denmark. Br J Cancer 2013;108:1195-208.

176. Greif J, Pezzi C, Klimberg V, et al. Gender Differences in Breast Cancer: Analysis of 13,000 Male Breast Cancers from the National Cancer Data Base. Abstract presented at the American Society of Breast Surgeons 13th Annual Meeting. Phoenix, AZ.: Canadian Cancer Statistics 2012, Toronto, ON: Canadian Cancer Society, 2012*Leading Causes of Death in Canada, 2008, Ottawa, ON: Statistics Canada, released 2011., 2012.

177. Muir D, Kanthan R, Kanthan SC. Male versus female breast cancers. A population-based comparative immunohistochemical analysis. Arch Pathol Lab Med 2003;127:36-41.

178. Watanabe MA, Oda JM, Amarante MK, et al. Regulatory T cells and breast cancer: implications for immunopathogenesis. Cancer Metastasis Rev 2010;29:569-79.

179. Jemal A, Siegel R, Xu J, et al. Cancer statistics, 2010. CA Cancer J Clin 2010;60:277- 300.

180. Barkan D, Green JE, Chambers AF. Extracellular matrix: a gatekeeper in the transition from dormancy to metastatic growth. Eur J Cancer 2010;46:1181-8.

146 181. Townson JL, Chambers AF. Dormancy of solitary metastatic cells. Cell Cycle 2006;5:1744-50.

182. Claus EB, Risch N, Thompson WD. Autosomal dominant inheritance of early-onset breast cancer. Implications for risk prediction. Cancer 1994;73:643-51.

183. Gail MH, Brinton LA, Byar DP, et al. Projecting individualized probabilities of developing breast cancer for white females who are being examined annually. J Natl Cancer Inst 1989;81:1879-86.

184. Offit K, Gilewski T, McGuire P, et al. Germline BRCA1 185delAG mutations in Jewish women with breast cancer. Lancet 1996;347:1643-5.

185. Miki Y, Swensen J, Shattuck-Eidens D, et al. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science 1994;266:66-71.

186. Cancer risks in BRCA2 mutation carriers. J Natl Cancer Inst 1999;91:1310-6.

187. Gustafsson A, Tartter PI, Brower ST, et al. Prognosis of patients with bilateral carcinoma of the breast. J Am Coll Surg 1994;178:111-6.

188. Graeser MK, Engel C, Rhiem K, et al. Contralateral breast cancer risk in BRCA1 and BRCA2 mutation carriers. J Clin Oncol 2009;27:5887-92.

189. Garber JE, Golshan M. Contralateral breast cancer in BRCA1/BRCA2 mutation carriers: the story of the other side. J Clin Oncol. Volume 27. United States, 2009:5862-4.

190. Simpson JF, Gray R, Dressler LG, et al. Prognostic value of histologic grade and proliferative activity in axillary node-positive breast cancer: results from the Eastern Cooperative Oncology Group Companion Study, EST 4189. J Clin Oncol 2000;18:2059- 69.

191. Comprehensive molecular portraits of human breast tumours. Nature 2012;490:61-70.

192. Masood S. Estrogen and progesterone receptors in cytology: a comprehensive review. Diagn Cytopathol 1992;8:475-91.

193. Trichopoulos D, MacMahon B, Cole P. Menopause and breast cancer risk. J Natl Cancer Inst 1972;48:605-13.

194. Perou CM, Sorlie T, Eisen MB, et al. Molecular portraits of human breast tumours. Nature 2000;406:747-52.

195. Malhotra GK, Zhao X, Band H, et al. Histological, molecular and functional subtypes of breast cancers. Cancer Biol Ther 2010;10:955-60.

196. Paik S, Shak S, Tang G, et al. A multigene assay to predict recurrence of tamoxifen- treated, node-negative breast cancer. N Engl J Med 2004;351:2817-26.

197. van 't Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002;415:530-6.

147 198. Tsai MJ, O'Malley BW. Molecular mechanisms of action of steroid/thyroid receptor superfamily members. Annu Rev Biochem 1994;63:451-86.

199. Mohibi S, Mirza S, Band H, et al. Mouse models of estrogen receptor-positive breast cancer. J Carcinog 2011;10:35.

200. Pike MC, Spicer DV, Dahmoush L, et al. Estrogens, progestogens, normal breast cell proliferation, and breast cancer risk. Epidemiol Rev 1993;15:17-35.

201. Boyd J. Estrogen as a carcinogen: the genetics and molecular biology of human endometrial carcinoma. Prog Clin Biol Res 1996;394:151-73.

202. Advani P, Moreno-Aspitia A. Current strategies for the prevention of breast cancer. Breast Cancer (Dove Med Press) 2014;6:59-71.

203. Mitri Z, Constantine T, O'Regan R. The HER2 Receptor in Breast Cancer: Pathophysiology, Clinical Use, and New Advances in Therapy. Chemother Res Pract 2012;2012:743193.

204. Slamon DJ, Clark GM, Wong SG, et al. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 1987;235:177-82.

205. Nuti M, Bellati F, Visconti V, et al. Immune effects of trastuzumab. J Cancer 2011;2:317-23.

206. Spector NL, Blackwell KL. Understanding the Mechanisms Behind Trastuzumab Therapy for Human Epidermal Growth Factor Receptor 2–Positive Breast Cancer. Journal of Clinical Oncology 2009;27:5838-5847.

207. Singletary SE, Allred C, Ashley P, et al. Revision of the American Joint Committee on Cancer staging system for breast cancer. J Clin Oncol 2002;20:3628-36.

208. Yeatman TJ, Cantor AB, Smith TJ, et al. Tumor biology of infiltrating lobular carcinoma. Implications for management. Ann Surg 1995;222:549-59; discussion 559- 61.

209. Fonseca R, Hartmann LC, Petersen IA, et al. Ductal carcinoma in situ of the breast. Ann Intern Med 1997;127:1013-22.

210. Lagios MD, Westdahl PR, Margolin FR, et al. Duct carcinoma in situ. Relationship of extent of noninvasive disease to the frequency of occult invasion, multicentricity, lymph node metastases, and short-term treatment failures. Cancer 1982;50:1309-14.

211. Page DL, Lagios MD. Pathologic analysis of the National Surgical Adjuvant Breast Project (NSABP) B-17 Trial. Unanswered questions remaining unanswered considering current concepts of ductal carcinoma in situ. Cancer 1995;75:1219-22; discussion 1223- 7.

212. Holland R, Peterse JL, Millis RR, et al. Ductal carcinoma in situ: a proposal for a new classification. Semin Diagn Pathol 1994;11:167-80.

148 213. Silverstein MJ, Lagios MD, Craig PH, et al. A prognostic index for ductal carcinoma in situ of the breast. Cancer 1996;77:2267-74.

214. Fisher B, Dignam J, Wolmark N, et al. Tamoxifen in treatment of intraductal breast cancer: National Surgical Adjuvant Breast and Bowel Project B-24 randomised controlled trial. Lancet 1999;353:1993-2000.

215. Houghton J, George WD, Cuzick J, et al. Radiotherapy and tamoxifen in women with completely excised ductal carcinoma in situ of the breast in the UK, Australia, and New Zealand: randomised controlled trial. Lancet 2003;362:95-102.

216. Fisher B, Costantino JP, Wickerham DL, et al. Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 1998;90:1371-88.

217. Fisher B, Fisher ER, Redmond C, et al. Tumor nuclear grade, estrogen receptor, and progesterone receptor: their value alone or in combination as indicators of outcome following adjuvant therapy for breast cancer. Breast Cancer Res Treat 1986;7:147-60.

218. Maughan KL, Lutterbie MA, Ham PS. Treatment of breast cancer. Am Fam Physician 2010;81:1339-46.

219. Mehta RS. Dose-dense and/or metronomic schedules of specific chemotherapies consolidate the chemosensitivity of triple-negative breast cancer: a step toward reversing triple-negative paradox. J Clin Oncol. Volume 26. United States, 2008:3286-8; author reply 3288.

220. Korkaya H, Paulson A, Iovino F, et al. HER2 regulates the mammary stem/progenitor cell population driving tumorigenesis and invasion. Oncogene 2008;27:6120-30.

221. Ginestier C, Hur MH, Charafe-Jauffret E, et al. ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell 2007;1:555-67.

222. Al-Hajj M, Wicha MS, Benito-Hernandez A, et al. Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A 2003;100:3983-8.

223. Korkaya H, Paulson A, Charafe-Jauffret E, et al. Regulation of mammary stem/progenitor cells by PTEN/Akt/beta-catenin signaling. PLoS Biol 2009;7:e1000121.

224. Iliopoulos D, Hirsch HA, Struhl K. An epigenetic switch involving NF-kappaB, Lin28, Let-7 MicroRNA, and IL6 links inflammation to cell transformation. Cell 2009;139:693- 706.

225. Liu S, Dontu G, Mantle ID, et al. Hedgehog signaling and Bmi-1 regulate self-renewal of normal and malignant human mammary stem cells. Cancer Res 2006;66:6063-71.

226. Dontu G, Jackson KW, McNicholas E, et al. Role of Notch signaling in cell-fate determination of human mammary stem/progenitor cells. Breast Cancer Res 2004;6:R605-15.

149 227. Iliopoulos D, Jaeger SA, Hirsch HA, et al. STAT3 activation of miR-21 and miR-181b-1 via PTEN and CYLD are part of the epigenetic switch linking inflammation to cancer. Mol Cell 2010;39:493-506.

228. Ma XJ, Dahiya S, Richardson E, et al. Gene expression profiling of the tumor microenvironment during breast cancer progression. Breast Cancer Res 2009;11:R7.

229. Liu S, Ginestier C, Ou SJ, et al. Breast cancer stem cells are regulated by mesenchymal stem cells through cytokine networks. Cancer Res 2011;71:614-24.

230. Sansone P, Storci G, Tavolari S, et al. IL-6 triggers malignant features in mammospheres from human ductal breast carcinoma and normal mammary gland. J Clin Invest 2007;117:3988-4002.

231. Stingl J, Eirew P, Ricketson I, et al. Purification and unique properties of mammary epithelial stem cells. Nature 2006;439:993-7.

232. Boyle ST, Kochetkova M. Breast Cancer Stem Cells and the Immune System: Promotion, Evasion and Therapy. J Mammary Gland Biol Neoplasia 2014.

233. Roodhart JM, Daenen LG, Stigter EC, et al. Mesenchymal stem cells induce resistance to chemotherapy through the release of platinum-induced fatty acids. Cancer Cell 2011;20:370-83.

234. Korkaya H, Liu S, Wicha MS. Breast cancer stem cells, cytokine networks, and the tumor microenvironment. J Clin Invest 2011;121:3804-9.

235. Pittenger MF, Mackay AM, Beck SC, et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284:143-7.

236. Farmer P, Bonnefoi H, Anderle P, et al. A stroma-related gene signature predicts resistance to neoadjuvant chemotherapy in breast cancer. Nat Med 2009;15:68-74.

237. Finak G, Bertos N, Pepin F, et al. Stromal gene expression predicts clinical outcome in breast cancer. Nat Med 2008;14:518-27.

238. Dumont N, Liu B, Defilippis RA, et al. Breast fibroblasts modulate early dissemination, tumorigenesis, and metastasis through alteration of extracellular matrix characteristics. Neoplasia 2013;15:249-62.

239. Burger JA, Kipps TJ. CXCR4: a key receptor in the crosstalk between tumor cells and their microenvironment. Blood 2006;107:1761-7.

240. Chu QD, Panu L, Holm NT, et al. High chemokine receptor CXCR4 level in triple negative breast cancer specimens predicts poor clinical outcome. J Surg Res 2010;159:689-95.

241. Dufraine J, Funahashi Y, Kitajewski J. Notch signaling regulates tumor angiogenesis by diverse mechanisms. Oncogene 2008;27:5132-7.

150 242. Singer CF, Kronsteiner N, Marton E, et al. MMP-2 and MMP-9 expression in breast cancer-derived human fibroblasts is differentially regulated by stromal-epithelial interactions. Breast Cancer Res Treat 2002;72:69-77.

243. Pietras K, Rubin K, Sjoblom T, et al. Inhibition of PDGF receptor signaling in tumor stroma enhances antitumor effect of chemotherapy. Cancer Res 2002;62:5476-84.

244. LeBedis C, Chen K, Fallavollita L, et al. Peripheral lymph node stromal cells can promote growth and tumorigenicity of breast carcinoma cells through the release of IGF- I and EGF. Int J Cancer 2002;100:2-8.

245. Hiscox S, Parr C, Nakamura T, et al. Inhibition of HGF/SF-induced breast cancer cell motility and invasion by the HGF/SF variant, NK4. Breast Cancer Res Treat 2000;59:245-54.

246. Dale TC, Weber-Hall SJ, Smith K, et al. Compartment switching of WNT-2 expression in human breast tumors. Cancer Res 1996;56:4320-3.

247. Zhang XH, Jin X, Malladi S, et al. Selection of bone metastasis seeds by mesenchymal signals in the primary tumor stroma. Cell 2013;154:1060-73.

248. Valadi H, Ekstrom K, Bossios A, et al. Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells. Nat Cell Biol 2007;9:654-9.

249. Luga V, Zhang L, Viloria-Petit Alicia M, et al. Exosomes Mediate Stromal Mobilization of Autocrine Wnt-PCP Signaling in Breast Cancer Cell Migration. Cell 2012;151:1542- 1556.

250. Duda DG, Cohen KS, Kozin SV, et al. Evidence for incorporation of bone marrow- derived endothelial cells into perfused blood vessels in tumors. Blood 2006;107:2774-6.

251. Lyden D, Hattori K, Dias S, et al. Impaired recruitment of bone-marrow-derived endothelial and hematopoietic precursor cells blocks tumor angiogenesis and growth. Nat Med 2001;7:1194-201.

252. Bhati R, Patterson C, Livasy CA, et al. Molecular characterization of human breast tumor vascular cells. Am J Pathol 2008;172:1381-90.

253. Keith B, Simon MC. Hypoxia-inducible factors, stem cells, and cancer. Cell 2007;129:465-72.

254. Holash J, Wiegand SJ, Yancopoulos GD. New model of tumor angiogenesis: dynamic balance between vessel regression and growth mediated by angiopoietins and VEGF. Oncogene 1999;18:5356-62.

255. Okines A, Cunningham D. Current perspective: bevacizumab in colorectal cancer--a time for reappraisal? Eur J Cancer 2009;45:2452-61.

151 256. Yang JC, Haworth L, Sherry RM, et al. A randomized trial of bevacizumab, an anti- vascular endothelial growth factor antibody, for metastatic renal cancer. N Engl J Med 2003;349:427-34.

257. Miller K, Wang M, Gralow J, et al. Paclitaxel plus bevacizumab versus paclitaxel alone for metastatic breast cancer. N Engl J Med 2007;357:2666-76.

258. Petrelli F, Barni S. Bevacizumab in advanced breast cancer: an opportunity as second- line therapy? Med Oncol 2012;29:1-4.

259. Stewart TH, Heppner GH. Immunological enhancement of breast cancer. Parasitology 1997;115 Suppl:S141-53.

260. Hadden JW. The immunology and immunotherapy of breast cancer: an update. Int J Immunopharmacol 1999;21:79-101.

261. Leek RD, Landers R, Fox SB, et al. Association of tumour necrosis factor alpha and its receptors with thymidine phosphorylase expression in invasive breast carcinoma. Br J Cancer 1998;77:2246-51.

262. Crowther M, Brown NJ, Bishop ET, et al. Microenvironmental influence on macrophage regulation of angiogenesis in wounds and malignant tumors. J Leukoc Biol 2001;70:478- 90.

263. Ben-Baruch A. Host microenvironment in breast cancer development: inflammatory cells, cytokines and chemokines in breast cancer progression: reciprocal tumor- microenvironment interactions. Breast Cancer Res 2003;5:31-6.

264. Dave SS, Wright G, Tan B, et al. Prediction of survival in follicular lymphoma based on molecular features of tumor-infiltrating immune cells. N Engl J Med 2004;351:2159-69.

265. Azenshtein E, Luboshits G, Shina S, et al. The CC chemokine RANTES in breast carcinoma progression: regulation of expression and potential mechanisms of promalignant activity. Cancer Res 2002;62:1093-102.

266. Saji H, Koike M, Yamori T, et al. Significant correlation of monocyte chemoattractant protein-1 expression with neovascularization and progression of breast carcinoma. Cancer 2001;92:1085-91.

267. van Netten JP, Ashmed BJ, Cavers D, et al. 'Macrophages' and their putative significance in human breast cancer. Br J Cancer 1992;66:220-1.

268. Leek RD, Lewis CE, Whitehouse R, et al. Association of macrophage infiltration with angiogenesis and prognosis in invasive breast carcinoma. Cancer Res 1996;56:4625-9.

269. Wyckoff JB, Wang Y, Lin EY, et al. Direct visualization of macrophage-assisted tumor cell intravasation in mammary tumors. Cancer Res 2007;67:2649-56.

270. Qian B, Deng Y, Im JH, et al. A distinct macrophage population mediates metastatic breast cancer cell extravasation, establishment and growth. PLoS One 2009;4:e6562.

152 271. Chen Q, Zhang XH, Massague J. Macrophage binding to receptor VCAM-1 transmits survival signals in breast cancer cells that invade the lungs. Cancer Cell 2011;20:538-49.

272. Lu X, Mu E, Wei Y, et al. VCAM-1 promotes osteolytic expansion of indolent bone micrometastasis of breast cancer by engaging alpha4beta1-positive osteoclast progenitors. Cancer Cell 2011;20:701-14.

273. Staveley-O'Carroll K, Sotomayor E, Montgomery J, et al. Induction of antigen-specific T cell anergy: An early event in the course of tumor progression. Proc Natl Acad Sci U S A 1998;95:1178-83.

274. Wolf AM, Wolf D, Steurer M, et al. Increase of regulatory T cells in the peripheral blood of cancer patients. Clin Cancer Res 2003;9:606-12.

275. Cuenca A, Cheng F, Wang H, et al. Extra-lymphatic solid tumor growth is not immunologically ignored and results in early induction of antigen-specific T-cell anergy: dominant role of cross-tolerance to tumor antigens. Cancer Res 2003;63:9007-15.

276. Huang B, Pan PY, Li Q, et al. Gr-1+CD115+ immature myeloid suppressor cells mediate the development of. Cancer Res 2006;66:1123-31.

277. Liyanage UK, Moore TT, Joo HG, et al. Prevalence of regulatory T cells is increased in peripheral blood and tumor microenvironment of patients with pancreas or breast adenocarcinoma. J Immunol 2002;169:2756-61.

278. Mansfield AS, Heikkila PS, Vaara AT, et al. Simultaneous Foxp3 and IDO expression is associated with sentinel lymph node metastases in breast cancer. BMC Cancer 2009;9:231.

279. Gupta S, Joshi K, Wig JD, et al. Intratumoral FOXP3 expression in infiltrating breast carcinoma: Its association with clinicopathologic parameters and angiogenesis. Acta Oncol 2007;46:792-7.

280. Audia S, Nicolas A, Cathelin D, et al. Increase of CD4+ CD25+ regulatory T cells in the peripheral blood of patients with metastatic carcinoma: a Phase I clinical trial using cyclophosphamide and immunotherapy to eliminate CD4+ CD25+ T lymphocytes. Clin Exp Immunol 2007;150:523-30.

281. .

282. Montero AJ, Diaz-Montero CM, Deutsch YE, et al. Phase 2 study of neoadjuvant treatment with NOV-002 in combination with doxorubicin and cyclophosphamide followed by docetaxel in patients with HER-2 negative clinical stage II-IIIc breast cancer. Breast Cancer Res Treat 2012;132:215-23.

283. Mundy-Bosse BL, Lesinski GB, Jaime-Ramirez AC, et al. Myeloid-derived suppressor cell inhibition of the IFN response in tumor-bearing mice. Cancer Res 2011;71:5101-10.

284. Psaila B, Lyden D. The metastatic niche: adapting the foreign soil. Nat Rev Cancer 2009;9:285-93.

153 285. Gao D, Joshi N, Choi H, et al. Myeloid progenitor cells in the premetastatic lung promote metastases by inducing mesenchymal to epithelial transition. Cancer Res 2012;72:1384-94.

286. Melani CCCFGCMP. Myeloid cell expansion elicited by the progression of spontaneous mammary carcinomas in c-erbB-2 transgenic BALB/c mice suppresses immune reactivity. Blood 2003;102:2138-2145.

287. Curiel TJ, Cheng P, Mottram P, et al. Dendritic Cell Subsets Differentially Regulate Angiogenesis in Human Ovarian Cancer. Cancer Research 2004;64:5535-5538.

288. Pierce BL, Ballard-Barbash R, Bernstein L, et al. Elevated biomarkers of inflammation are associated with reduced survival among breast cancer patients. J Clin Oncol 2009;27:3437-44.

289. Balkwill F. TNF-alpha in promotion and progression of cancer. Cancer Metastasis Rev 2006;25:409-16.

290. Balkwill F, Joffroy C. TNF: a tumor-suppressing factor or a tumor-promoting factor? Future Oncol 2010;6:1833-6.

291. Connelly L, Barham W, Onishko HM, et al. Inhibition of NF-kappa B activity in mammary epithelium increases tumor latency and decreases tumor burden. Oncogene 2011;30:1402-12.

292. Balkwill F. Tumour necrosis factor and cancer. Nat Rev Cancer 2009;9:361-71.

293. Kamel M, Shouman S, El-Merzebany M, et al. Effect of tumour necrosis factor-alpha on estrogen metabolic pathways in breast cancer cells. J Cancer 2012;3:310-21.

294. Soria G, Ofri-Shahak M, Haas I, et al. Inflammatory mediators in breast cancer: coordinated expression of TNFalpha & IL-1beta with CCL2 & CCL5 and effects on epithelial-to-mesenchymal transition. BMC Cancer 2011;11:130.

295. Ueno T, Toi M, Saji H, et al. Significance of macrophage chemoattractant protein-1 in macrophage recruitment, angiogenesis, and survival in human breast cancer. Clin Cancer Res 2000;6:3282-9.

296. Qian BZ, Li J, Zhang H, et al. CCL2 recruits inflammatory monocytes to facilitate breast-tumour metastasis. Nature 2011;475:222-5.

297. Salcedo R, Ponce ML, Young HA, et al. Human endothelial cells express CCR2 and respond to MCP-1: direct role of MCP-1 in angiogenesis and tumor progression. Blood 2000;96:34-40.

298. Luboshits G, Shina S, Kaplan O, et al. Elevated expression of the CC chemokine regulated on activation, normal T cell expressed and secreted (RANTES) in advanced breast carcinoma. Cancer Res 1999;59:4681-7.

154 299. Niwa Y, Akamatsu H, Niwa H, et al. Correlation of tissue and plasma RANTES levels with disease course in patients with breast or cervical cancer. Clin Cancer Res 2001;7:285-9.

300. Mira E, Lacalle RA, Gonzalez MA, et al. A role for chemokine receptor transactivation in growth factor signaling. EMBO Rep 2001;2:151-6.

301. Ali S, Kaur J, Patel KD. Intercellular cell adhesion molecule-1, vascular cell adhesion molecule-1, and regulated on activation normal T cell expressed and secreted are expressed by human breast carcinoma cells and support eosinophil adhesion and activation. Am J Pathol 2000;157:313-21.

302. do Val Carneiro JL, Nixdorf SL, Mantovani MS, et al. Plasma malondialdehyde levels and CXCR4 expression in peripheral blood cells of breast cancer patients. J Cancer Res Clin Oncol 2009;135:997-1004.

303. Muller A, Homey B, Soto H, et al. Involvement of chemokine receptors in breast cancer metastasis. Nature 2001;410:50-6.

304. Ginestier C, Liu S, Diebel ME, et al. CXCR1 blockade selectively targets human breast cancer stem cells in vitro and in xenografts. J Clin Invest 2010;120:485-97.

305. Yao C, Lin Y, Chua MS, et al. Interleukin-8 modulates growth and invasiveness of estrogen receptor-negative breast cancer cells. Int J Cancer 2007;121:1949-57.

306. Hoffmann A, Natoli G, Ghosh G. Transcriptional regulation via the NF-kappaB signaling module. Oncogene 2006;25:6706-16.

307. Gabrilovich DI, Ostrand-Rosenberg S, Bronte V. Coordinated regulation of myeloid cells by tumours. Nat Rev Immunol 2012;12:253-68.

308. Gabrilovich DI, Nagaraj S. Myeloid-derived suppressor cells as regulators of the immune system. Nat Rev Immunol 2009;9:162-74.

309. Jin L, Yuan RQ, Fuchs A, et al. Expression of interleukin-1beta in human breast carcinoma. Cancer 1997;80:421-34.

310. Bunt SK, Sinha P, Clements VK, et al. Inflammation induces myeloid-derived suppressor cells that facilitate tumor progression. J Immunol 2006;176:284-90.

311. Song X, Krelin Y, Dvorkin T, et al. CD11b+/Gr-1+ immature myeloid cells mediate suppression of T cells in mice bearing tumors of IL-1beta-secreting cells. J Immunol 2005;175:8200-8.

312. Bunt SK, Clements VK, Hanson EM, et al. Inflammation enhances myeloid-derived suppressor cell cross-talk by signaling through Toll-like receptor 4. J Leukoc Biol 2009;85:996-1004.

313. Wagner KU. Models of breast cancer: quo vadis, animal modeling? Breast Cancer Res. Volume 6. England, 2004:31-8.

155 314. Nandi S, Guzman RC, Yang J. Hormones and mammary carcinogenesis in mice, rats, and humans: a unifying hypothesis. Proc Natl Acad Sci U S A 1995;92:3650-7.

315. Kim JB, O'Hare MJ, Stein R. Models of breast cancer: is merging human and animal models the future? Breast Cancer Res. Volume 6. England, 2004:22-30.

316. Frese KK, Tuveson DA. Maximizing mouse cancer models. Nat Rev Cancer 2007;7:654-658.

317. Pulaski BA, Ostrand-Rosenberg S. Mouse 4T1 breast tumor model. Curr Protoc Immunol 2001;Chapter 20:Unit 20.2.

318. Oh K, Lee OY, Shon SY, et al. A mutual activation loop between breast cancer cells and myeloid-derived suppressor cells facilitates spontaneous metastasis through IL-6 trans- signaling in a murine model. Breast Cancer Res 2013;15:R79.

319. Khanna C, Hunter K. Modeling metastasis in vivo. Carcinogenesis 2005;26:513-23.

320. Fantozzi A, Christofori G. Mouse models of breast cancer metastasis. Breast Cancer Res 2006;8:212.

321. Stewart TA, Pattengale PK, Leder P. Spontaneous mammary adenocarcinomas in transgenic mice that carry and express MTV/myc fusion genes. Cell 1984;38:627-37.

322. Matulka LA, Wagner K-U. Models of breast cancer. Drug Discovery Today: Disease Models 2005;2:1-6.

323. Kuperwasser C, Dessain S, Bierbaum BE, et al. A mouse model of human breast cancer metastasis to human bone. Cancer Res 2005;65:6130-8.

324. Vargo-Gogola T, Rosen JM. Modelling breast cancer: one size does not fit all. Nat Rev Cancer 2007;7:659-72.

325. Shultz LD, Ishikawa F, Greiner DL. Humanized mice in translational biomedical research. Nat Rev Immunol 2007;7:118-130.

326. Proia DA, Kuperwasser C. Reconstruction of human mammary tissues in a mouse model. Nat Protoc 2006;1:206-14.

327. Chen Z, De Kauwe AL, Keech C, et al. Humanized transgenic mice expressing HLA DR4-DQ3 haplotype: reconstitution of phenotype and HLA-restricted T-cell responses. Tissue Antigens 2006;68:210-219.

328. Madsen L, Labrecque N, Engberg J, et al. Mice lacking all conventional MHC class II genes. Proceedings of the National Academy of Sciences 1999;96:10338-10343.

329. Liu K, Hipkens S, Yang T, et al. Recombinase-mediated cassette exchange to rapidly and efficiently generate mice with human cardiac sodium channels. Genesis 2006;44:556-64.

156 330. Wallace HA, Marques-Kranc F, Richardson M, et al. Manipulating the mouse genome to engineer precise functional syntenic replacements with human sequence. Cell 2007;128:197-209.

331. Harley CB, Futcher AB, Greider CW. Telomeres shorten during ageing of human fibroblasts. Nature 1990;345:458-60.

332. Prowse KR, Greider CW. Developmental and tissue-specific regulation of mouse telomerase and telomere length. Proc Natl Acad Sci U S A 1995;92:4818-22.

333. .

334. Barclay AN, Wright GJ, Brooke G, et al. CD200 and membrane protein interactions in the control of myeloid cells. Trends Immunol. Volume 23. England, 2002:285-90.

335. Minas K, Liversidge J. Is the CD200/CD200 receptor interaction more than just a myeloid cell inhibitory signal? Crit Rev Immunol. Volume 26. United States, 2006:213- 30.

336. Rygiel TP, Meyaard L. CD200R signaling in tumor tolerance and inflammation: A tricky balance. Curr Opin Immunol 2012;24:233-8.

337. Wright GJ, Cherwinski H, Foster-Cuevas M, et al. Characterization of the CD200 receptor family in mice and humans and their. J Immunol 2003;171:3034-46.

338. Rijkers ES, de Ruiter T, Baridi A, et al. The inhibitory CD200R is differentially expressed on human and mouse T and B. Mol Immunol 2008;45:1126-35.

339. Mihrshahi R, Barclay AN, Brown MH. Essential roles for Dok2 and RasGAP in CD200 receptor-mediated regulation of human myeloid cells. J Immunol. Volume 183. United States, 2009:4879-86.

340. Gorczynski R, Chen Z, Kai Y, et al. CD200 is a ligand for all members of the CD200R family of immunoregulatory molecules. J Immunol 2004;172:7744-9.

341. Hoek RM, Ruuls SR, Murphy CA, et al. Down-regulation of the macrophage lineage through interaction with OX2 (CD200). Science 2000;290:1768-71.

342. Fallarino F, Asselin-Paturel C, Vacca C, et al. Murine plasmacytoid dendritic cells initiate the immunosuppressive pathway of tryptophan catabolism in response to CD200 receptor engagement. J Immunol. Volume 173. United States: 2004 The American Association of Immunologists, Inc., 2004:3748-54.

343. Fallarino F, Orabona C, Vacca C, et al. Ligand and cytokine dependence of the immunosuppressive pathway of tryptophan catabolism in plasmacytoid dendritic cells. Int Immunol. Volume 17. England, 2005:1429-38.

344. Gorczynski, M. R. CD200:CD200R-Mediated Regulation of Immunity. Volume Vol 2012. ISRN Immunology, 2012:18.

157 345. Gorczynski RM, Hu J, Chen Z, et al. A CD200FC immunoadhesin prolongs rat islet xenograft survival in mice. Transplantation 2002;73:1948-53.

346. Gorczynski RM, Chen Z, Yu K, et al. CD200 immunoadhesin suppresses collagen- induced arthritis in mice. Clin Immunol 2001;101:328-34.

347. Gorczynski RM, Chen Z, Hu J, et al. Evidence of a role for CD200 in regulation of immune rejection of leukaemic tumour cells in C57BL/6 mice. Clin Exp Immunol. Volume 126. England, 2001:220-9.

348. Gorczynski RM, Cattral MS, Chen Z, et al. An immunoadhesin incorporating the molecule OX-2 is a potent immunosuppressant that prolongs allo- and xenograft survival. J Immunol 1999;163:1654-60.

349. Gorczynski RM, Chen Z, He W, et al. Expression of a CD200 transgene is necessary for induction but not maintenance of tolerance to cardiac and skin allografts. J Immunol 2009;183:1560-8.

350. Gorczynski RM, Chen Z, Khatri I, et al. Graft-infiltrating cells expressing a CD200 transgene prolong allogeneic skin graft survival in association with local increases in Foxp3(+)Treg and mast cells. Transpl Immunol 2011;25:187-93.

351. Foster-Cuevas M, Wright GJ, Puklavec MJ, et al. Human herpesvirus 8 K14 protein mimics CD200 in down-regulating macrophage activation through CD200 receptor. J Virol 2004;78:7667-76.

352. Karnam G, Rygiel TP, Raaben M, et al. CD200 receptor controls sex-specific TLR7 responses to viral infection. PLoS Pathog 2012;8:e1002710.

353. Gorczynski R, Khatri I, Lee L, et al. An interaction between CD200 and monoclonal antibody agonists to CD200R2 in development of dendritic cells that preferentially induce populations of CD4+CD25+ T regulatory cells. J Immunol 2008;180:5946-55.

354. Moreaux J, Hose D, Reme T, et al. CD200 is a new prognostic factor in multiple myeloma. Blood 2006;108:4194-7.

355. Tonks A, Hills R, White P, et al. CD200 as a prognostic factor in acute myeloid leukaemia. Leukemia. Volume 21. England, 2007:566-8.

356. Siva A, Xin H, Qin F, et al. Immune modulation by melanoma and ovarian tumor cells through expression of the. Cancer Immunol Immunother 2008;57:987-96.

357. Petermann KB, Rozenberg GI, Zedek D, et al. CD200 is induced by ERK and is a potential therapeutic target in melanoma. J Clin Invest 2007;117:3922-9.

358. Kawasaki BT, Mistree T, Hurt EM, et al. Co-expression of the toleragenic glycoprotein, CD200, with markers for cancer. Biochem Biophys Res Commun 2007;364:778-82.

158 359. Kretz-Rommel A, Bowdish KS. Rationale for anti-CD200 immunotherapy in B-CLL and other hematologic malignancies: new concepts in blocking immune suppression. Expert Opin Biol Ther 2008;8:5-15.

360. Rygiel TP, Karnam G, Goverse G, et al. CD200-CD200R signaling suppresses anti- tumor responses independently of CD200 expression on the tumor. Oncogene. Volume 31. England, 2012:2979-88.

361. Gorczynski RM. Evidence for an immunoregulatory role of OX2 with its counter ligand (OX2L) in the regulation of transplant rejection, fetal loss, autoimmunity and tumor growth. Arch Immunol Ther Exp (Warsz) 2001;49:303-9.

362. Kawasaki BT, Farrar WL. Cancer stem cells, CD200 and immunoevasion. Trends Immunol 2008;29:464-8.

363. Gorczynski RM, Clark DA, Erin N, et al. Role of CD200 expression in regulation of metastasis of EMT6 tumor cells in mice. Breast Cancer Res Treat 2011;130:49-60.

364. Wong KK, Khatri I, Shaha S, et al. The role of CD200 in immunity to B cell lymphoma. J Leukoc Biol 2010;88:361-72.

365. Pallasch CP, Ulbrich S, Brinker R, et al. Disruption of T cell suppression in chronic lymphocytic leukemia by CD200 blockade. Leuk Res 2009;33:460-4.

366. Gorczynski RM, Chen Z, Diao J, et al. Breast cancer cell CD200 expression regulates immune response to EMT6 tumor cells. Breast Cancer Res Treat 2010;123:405-15.

367. Gorczynski RM, Chen Z, Kai Y, et al. Induction of tolerance-inducing antigen- presenting cells in bone marrow cultures. Transplantation 2004;77:1138-44.

368. Yang J, Weinberg RA. Epithelial-mesenchymal transition: at the crossroads of development and tumor metastasis. Dev Cell. Volume 14. United States, 2008:818-29.

369. Yu K, Chen Z, Khatri I, et al. CCR4 dependent migration of Foxp3+ Treg cells to skin grafts and draining lymph. Immunol Lett 2011;141:116-22.

370. Stumpfova M, Ratner D, Desciak EB, et al. The immunosuppressive surface ligand CD200 augments the metastatic capacity of squamous cell carcinoma. Cancer Res 2010;70:2962-72.

371. Wong KK, Brenneman F, Chesney A, et al. Soluble CD200 is critical to engraft chronic lymphocytic leukemia cells in immunocompromised mice. Cancer Res 2012;72:4931-43.

372. Gorczynski R, Chen Z, Khatri I, et al. Increased lung/liver metastasis of human breast cancer cells in NOD. SCIDIL2rγ-/-mice after treatment with soluble CD200Fc (TUM7P. 922). The Journal of Immunology 2014;192:203-4.

373. Rockwell SC, Kallman RF, Fajardo LF. Characteristics of a serially transplanted mouse mammary tumor and its tissue-culture-adapted derivative. J Natl Cancer Inst 1972;49:735-49.

159 374. Pandit TS, Kennette W, Mackenzie L, et al. Lymphatic metastasis of breast cancer cells is associated with differential gene expression profiles that predict cancer stem cell-like properties and the ability to survive, establish and grow in a foreign environment. Int J Oncol 2009;35:297-308.

375. Pfeffer U, Romeo F, Noonan DM, et al. Prediction of breast cancer metastasis by genomic profiling: where do we stand? Clin Exp Metastasis 2009;26:547-58.

376. Chen X, Xu Z, Wang Y. Recent advances in breast cancer metastasis suppressor 1. Int J Biol Markers 2011;26:1-8.

377. Pollard JW. Macrophages define the invasive microenvironment in breast cancer. J Leukoc Biol 2008;84:623-30.

378. Lu X, Kang Y. Chemokine (C-C motif) ligand 2 engages CCR2+ stromal cells of monocytic origin to promote breast cancer metastasis to lung and bone. J Biol Chem 2009;284:29087-96.

379. Chen T, Jackson CR, Link A, et al. Int7G24A variant of transforming growth factor-beta receptor type I is associated with invasive breast cancer. Clin Cancer Res 2006;12:392-7.

380. Liang Z, Yoon Y, Votaw J, et al. Silencing of CXCR4 blocks breast cancer metastasis. Cancer Res 2005;65:967-71.

381. Takahashi M, Miyazaki H, Furihata M, et al. Chemokine CCL2/MCP-1 negatively regulates metastasis in a highly bone marrow-metastatic mouse breast cancer model. Clin Exp Metastasis 2009;26:817-28.

382. Ma X, Norsworthy K, Kundu N, et al. CXCR3 expression is associated with poor survival in breast cancer and promotes metastasis in a murine model. Mol Cancer Ther 2009;8:490-8.

383. Chen J, Yao Y, Gong C, et al. CCL18 from tumor-associated macrophages promotes breast cancer metastasis via PITPNM3. Cancer Cell 2011;19:541-55.

384. Yang L, DeBusk LM, Fukuda K, et al. Expansion of myeloid immune suppressor Gr+CD11b+ cells in tumor-bearing host directly promotes tumor angiogenesis. Cancer Cell 2004;6:409-21.

385. Qin FX. Dynamic behavior and function of Foxp3+ regulatory T cells in tumor bearing host. Cell Mol Immunol 2009;6:3-13.

386. Yang L, Huang J, Ren X, et al. Abrogation of TGF beta signaling in mammary carcinomas recruits Gr-1+CD11b+ myeloid cells that promote metastasis. Cancer Cell. Volume 13. United States, 2008:23-35.

387. McWhirter JR, Kretz-Rommel A, Saven A, et al. Antibodies selected from combinatorial libraries block a tumor antigen that plays a key role in immunomodulation. Proc Natl Acad Sci U S A. Volume 103. United States, 2006:1041-6.

160 388. Boudakov I, Liu J, Fan N, et al. Mice lacking CD200R1 show absence of suppression of lipopolysaccharide-induced tumor necrosis factor-alpha and mixed leukocyte culture responses by CD200. Transplantation 2007;84:251-7.

389. Skalova K, Mollova K, Michalek J. Human myeloid dendritic cells for cancer therapy: Does maturation matter? Vaccine 2010.

390. Walser TC, Ma X, Kundu N, et al. Immune-mediated modulation of breast cancer growth and metastasis by the chemokine Mig (CXCL9) in a murine model. J Immunother 2007;30:490-8.

391. Baumgartner J, Wilson C, Palmer B, et al. Melanoma induces immunosuppression by up-regulating FOXP3(+) regulatory T cells. J Surg Res 2007;141:72-7.

392. Hilchey SP, De A, Rimsza LM, et al. Follicular lymphoma intratumoral CD4+CD25+GITR+ regulatory T cells potently suppress CD3/CD28-costimulated autologous and allogeneic CD8+CD25- and CD4+CD25- T cells. J Immunol 2007;178:4051-61.

393. Bierie B, Moses HL. Transforming growth factor beta (TGF-beta) and inflammation in cancer. Cytokine Growth Factor Rev. Volume 21. England, 2010:49-59.

394. Hori S. Developmental plasticity of Foxp3+ regulatory T cells. Curr Opin Immunol 2010;22:575-82.

395. deLeeuw RJ, Kost SE, Kakal JA, et al. The prognostic value of FoxP3+ tumor- infiltrating lymphocytes in cancer: a critical review of the literature. Clin Cancer Res 2012;18:3022-9.

396. Podnos A, Clark DA, Erin N, et al. Further evidence for a role of tumor CD200 expression in breast cancer metastasis: decreased metastasis in CD200R1KO mice or using CD200-silenced EMT6. Breast Cancer Res Treat 2012;136:117-27.

397. Gorczynski RM, Chen Z, Khatri I, et al. Cure of metastatic growth of EMT6 tumor cells in mice following manipulation of CD200:CD200R signaling. Breast Cancer Res Treat 2013;142:271-82.

398. Hirsch HA, Iliopoulos D, Tsichlis PN, et al. Metformin selectively targets cancer stem cells, and acts together with chemotherapy to block tumor growth and prolong remission. Cancer Res 2009;69:7507-11.

399. The biguanides metformin and phenformin inhibit angiogenesis, local and metastatic growth of breast cancer by targeting both neoplastic and microenvironment cells. Int J Cancer 2015;136:E534-44.

400. Vazquez-Martin A, Oliveras-Ferraros C, Del Barco S, et al. The anti-diabetic drug metformin suppresses self-renewal and proliferation of trastuzumab-resistant tumor- initiating breast cancer stem cells. Breast Cancer Res Treat 2011;126:355-64.

161 401. Ben Sahra I, Laurent K, Loubat A, et al. The antidiabetic drug metformin exerts an antitumoral effect in vitro and in vivo through a decrease of cyclin D1 level. Oncogene 2008;27:3576-86.

402. Liu B, Fan Z, Edgerton SM, et al. Metformin induces unique biological and molecular responses in triple negative breast cancer cells. Cell Cycle 2009;8:2031-40.

403. Anisimov VN, Berstein LM, Egormin PA, et al. Effect of metformin on life span and on the development of spontaneous mammary tumors in HER-2/neu transgenic mice. Exp Gerontol 2005;40:685-93.

404. Dowling RJ, Goodwin PJ, Stambolic V. Understanding the benefit of metformin use in cancer treatment. BMC Med 2011;9:33.

405. Wu L, Tannock IF. Repopulation in murine breast tumors during and after sequential treatments with cyclophosphamide and 5-fluorouracil. Cancer Res 2003;63:2134-8.

406. Gorczynski RM, Chen Z, Erin N, et al. Comparison of Immunity in Mice Cured of Primary/Metastatic Growth of EMT6 or 4THM Breast Cancer by Chemotherapy or Immunotherapy. PLoS One 2014;9:e113597.

407. Kasznicki J, Sliwinska A, Drzewoski J. Metformin in cancer prevention and therapy. Ann Transl Med 2014;2:57.

408. Snelgrove RJ, Goulding J, Didierlaurent AM, et al. A critical function for CD200 in lung immune homeostasis and the severity of influenza infection. Nat Immunol 2008;9:1074- 83.

409. Gutschalk CM, Yanamandra AK, Linde N, et al. GM-CSF enhances tumor invasion by elevated MMP-2, -9, and -26 expression. Cancer Med 2013;2:117-29.

410. Parmiani G, Castelli C, Pilla L, et al. Opposite immune functions of GM-CSF administered as vaccine adjuvant in cancer patients. Ann Oncol 2007;18:226-32.

411. Sheng JR, Muthusamy T, Prabhakar BS, et al. GM-CSF-induced regulatory T cells selectively inhibit anti-acetylcholine receptor-specific immune responses in experimental myasthenia gravis. J Neuroimmunol 2011;240-241:65-73.

412. O'Keeffe M, Hochrein H, Vremec D, et al. Effects of administration of progenipoietin 1, Flt-3 ligand, granulocyte colony-stimulating factor, and pegylated granulocyte- macrophage colony-stimulating factor on dendritic cell subsets in mice. Blood 2002;99:2122-30.

413. Su S, Liu Q, Chen J, et al. A positive feedback loop between mesenchymal-like cancer cells and macrophages is essential to breast cancer metastasis. Cancer Cell 2014;25:605- 20.

414. Perez-Yepez EA, Ayala-Sumuano JT, Lezama R, et al. A novel beta-catenin signaling pathway activated by IL-1beta leads to the onset of epithelial-mesenchymal transition in breast cancer cells. Cancer Lett 2014;354:164-71.

162 415. Krelin Y, Voronov E, Dotan S, et al. Interleukin-1beta-driven inflammation promotes the development and invasiveness of chemical carcinogen-induced tumors. Cancer Res 2007;67:1062-71.

416. Apte RN, Dotan S, Elkabets M, et al. The involvement of IL-1 in tumorigenesis, tumor invasiveness, metastasis and tumor-host interactions. Cancer Metastasis Rev 2006;25:387-408.

417. Voronov E, Dayan M, Zinger H, et al. IL-1 beta-deficient mice are resistant to induction of experimental SLE. Eur Cytokine Netw 2006;17:109-16.

418. Bunt SK, Yang L, Sinha P, et al. Reduced inflammation in the tumor microenvironment delays the accumulation of myeloid-derived suppressor cells and limits tumor progression. Cancer Res 2007;67:10019-26.

419. Quinn BJ, Kitagawa H, Memmott RM, et al. Repositioning metformin for cancer prevention and treatment. Trends Endocrinol Metab 2013;24:469-80.

420. Evans JM, Donnelly LA, Emslie-Smith AM, et al. Metformin and reduced risk of cancer in diabetic patients. Bmj 2005;330:1304-5.

421. Libby G, Donnelly LA, Donnan PT, et al. New users of metformin are at low risk of incident cancer: a cohort study among people with type 2 diabetes. Diabetes Care 2009;32:1620-5.

422. Landman GW, Kleefstra N, van Hateren KJ, et al. Metformin associated with lower cancer mortality in type 2 diabetes: ZODIAC-16. Diabetes Care 2010;33:322-6.

423. Monami M, Lamanna C, Balzi D, et al. Sulphonylureas and cancer: a case-control study. Acta Diabetol 2009;46:279-84.

424. Bowker SL, Majumdar SR, Veugelers P, et al. Increased cancer-related mortality for patients with type 2 diabetes who use sulfonylureas or insulin. Diabetes Care 2006;29:254-8.

425. Jiralerspong S, Palla SL, Giordano SH, et al. Metformin and pathologic complete responses to neoadjuvant chemotherapy in diabetic patients with breast cancer. J Clin Oncol 2009;27:3297-302.

426. Xu X, Araki K, Li S, et al. Autophagy is essential for effector CD8(+) T cell survival and memory formation. Nat Immunol 2014;15:1152-61.

427. Donohue E, Tovey A, Vogl AW, et al. Inhibition of autophagosome formation by the benzoporphyrin derivative verteporfin. J Biol Chem 2011;286:7290-300.

428. U.S. National Institutes of Health. ClinicalTrials.gov. Volume 2015.

429. El Saghir NS, Tfayli A, Hatoum HA, et al. Treatment of metastatic breast cancer: state- of-the-art, subtypes and perspectives. Crit Rev Oncol Hematol 2011;80:433-49.

430. Nowell PC. The clonal evolution of tumor cell populations. Science 1976;194:23-8.

163 431. Husemann Y, Geigl JB, Schubert F, et al. Systemic spread is an early step in breast cancer. Cancer Cell 2008;13:58-68.

432. Engel J, Eckel R, Kerr J, et al. The process of metastasisation for breast cancer. European Journal of Cancer 2003;39:1794-1806.

433. Klein CA. The systemic progression of human cancer: a focus on the individual disseminated cancer cell--the unit of selection. Adv Cancer Res 2003;89:35-67.

434. Wong K. CD200: A Novel Therapeutic Target for Chronic Lymphocytic Leukemia. 2013.

435. Chen DX, Gorczynski RM. Discrete monoclonal antibodies define functionally important epitopes in the. Transplantation 2005;79:282-8.

436. Celik B, Yalcin AD, Gizem EG, et al. Proteomics pattern of peritoneal sApo-2L but not CD200 (OX-2) as a possible screening biomarker for stage 4 metastatic ovarian, endometrial, ad breast carcinoma. Journal of Balkan Union of Oncology: Antalya Training Hospital, 2015.

437. Kos O, Hughson RL, Hart DA, et al. Elevated serum soluble CD200 and CD200R as surrogate markers of bone loss under bed rest conditions. Bone 2014;60:33-40.

438. Li Y, Zhao LD, Tong LS, et al. Aberrant CD200/CD200R1 expression and function in systemic lupus erythematosus contributes to abnormal T-cell responsiveness and dendritic cell activity. Arthritis Res Ther 2012;14:R123.

439. Arik HO, Yalcin AD, Celik B, et al. Evaluation of soluble CD200 levels in type 2 diabetic foot and nephropathic patients: association with disease activity. Med Sci Monit 2014;20:1078-81.

440. Akman-Karakas A, Yalcin AD, Koc S, et al. Serum soluble CD200 level was higher in patients with bullous pemphigoid during the active phase of the disease than for healthy individuals. Clin Lab 2014;60:1237-40.

441. Bisgnin A, Yalcin AD, Gumuslu S, et al. Correlation of serum proteomics patterns of sCD200 (OX-2), sApo-2L (sTRAIL), vitamin-D and homocysteine to quantitative FDG- PET/CT findings in newly diagnosed non-small cell lung cancer. J buon 2013;18:1099- 100.

442. Moertel CL, Xia J, LaRue R, et al. CD200 in CNS tumor-induced immunosuppression: the role for CD200 pathway blockade in targeted immunotherapy. J Immunother Cancer 2014;2:46.

443. Celik B, Yalcin AD, Genc GE, et al. Possible usage of soluble CD200 and soluble tumor necrosis factor-related apoptosis inducing ligand as a screening biomarker of metastatic ovarian, endometrial and breast carcinoma. Conference, EAACI, Italy: Antalya Training and Research Hospital, 2015.

164 444. Pedrazzoli P, Baldanti F, Donatelli I, et al. Vaccination for seasonal influenza in patients with cancer: recommendations of the Italian Society of Medical Oncology (AIOM). Ann Oncol 2014;25:1243-7.

445. Walsh CM, Edinger AL. The complex interplay between autophagy, apoptosis, and necrotic signals promotes T-cell homeostasis. Immunol Rev 2010;236:95-109.

446. Bourgeois C, Rocha B, Tanchot C. A role for CD40 expression on CD8+ T cells in the generation of CD8+ T cell memory. Science 2002;297:2060-3.

447. Wang RC, Levine B. Autophagy in cellular growth control. FEBS Lett 2010;584:1417- 26.

448. Qadir MA, Kwok B, Dragowska WH, et al. Macroautophagy inhibition sensitizes tamoxifen-resistant breast cancer cells and enhances mitochondrial depolarization. Breast Cancer Res Treat 2008;112:389-403.

449. Sasaki K, Tsuno NH, Sunami E, et al. Resistance of colon cancer to 5-fluorouracil may be overcome by combination with chloroquine, an in vivo study. Anticancer Drugs 2012;23:675-82.

450. Hege KM, Jooss K, Pardoll D. GM-CSF gene-modifed cancer cell immunotherapies: of mice and men. Int Rev Immunol 2006;25:321-52.

451. Heath WR, Kurts C, Miller JF, et al. Cross-tolerance: a pathway for inducing tolerance to peripheral tissue antigens. J Exp Med 1998;187:1549-53.

452. Marincola FM, Jaffee EM, Hicklin DJ, et al. Escape of human solid tumors from T-cell recognition: molecular mechanisms and functional significance. Adv Immunol 2000;74:181-273.

453. Chaudhuri S, Cariappa A, Tang M, et al. Genetic susceptibility to breast cancer: HLA DQB*03032 and HLA DRB1*11 may represent protective alleles. Proc Natl Acad Sci U S A 2000;97:11451-4.

454. Menard S, Tomasic G, Casalini P, et al. Lymphoid infiltration as a prognostic variable for early-onset breast carcinomas. Clin Cancer Res 1997;3:817-9.

455. Disis ML, Calenoff E, McLaughlin G, et al. Existent T-cell and antibody immunity to HER-2/neu protein in patients with breast cancer. Cancer Res 1994;54:16-20.

456. Jerome KR, Domenech N, Finn OJ. Tumor-specific cytotoxic T cell clones from patients with breast and pancreatic adenocarcinoma recognize EBV-immortalized B cells transfected with polymorphic epithelial mucin complementary DNA. J Immunol 1993;151:1654-62.

457. Zhou G, Drake CG, Levitsky HI. Amplification of tumor-specific regulatory T cells following therapeutic cancer vaccines. Blood 2006;107:628-36.

165 458. Duraiswamy J, Kaluza KM, Freeman GJ, et al. Dual blockade of PD-1 and CTLA-4 combined with tumor vaccine effectively restores T-cell rejection function in tumors. Cancer Res 2013;73:3591-603.

166

Appendices Bidirectional effect of CD200 on breast cancer development and metastasis, with ultimate outcome determined by tumour aggressiveness and a cancer- induced inflammatory response

Erin N, Podnos A, Tanriover G, Duymuş O, Cote E, Khatri I, Gorczynski RM

Abstract

CD200 acts through its receptor (CD200R) to inhibit excessive inflammation. The role of CD200:CD200R1 interaction in tumour immunity is poorly understood. In this study, we examined the role of CD200-CD200R1 interaction in the progression and metastasis of highly aggressive 4THM murine breast carcinoma using CD200 transgenic (CD200tg) and CD200R1 knock-out (CD200R1-/-) BALB/c mice. 4THMcells induce extensive visceral metastasis and neutrophil infiltration in affected tissues. CD200 overexpression in the host was associated with decreased primary tumour growth and metastasis, while lack of CD200R1 expression by host cells was associated with enhanced visceral metastasis. Absence of CD200R1 expression led to decreased tumour infiltrating cytotoxic T cells and increased the release of inflammatory cytokines, such as TNF-α and IL-6. In contrast, CD200 overexpression led to increased tumour- induced IFN-γ and IL-10 response and decreased TNF-α and IL-6 release. Neutrophil infiltration of tissues was markedly decreased in CD200tg animals and increased in CD200R1-/- mice. These findings are contrary to what has been reported in the EMT6 mouse breast cancer model. Other distinguishing features of tumour elicited by EMT6 and 4THM cell injections were also examined. Visceral tissues from mice bearing EMT6 tumours showed a lack of neutrophil infiltration and decreased IL-6 release in CD200R1-/- mice. EMT6 and 4THM cells also differed in vimentin expression, in vitro migration rate, IL-6 and TNFα secretion, which

167 was markedly lower in EMT6 tumours. These results support the hypothesis that CD200 expression can alter immune responses, and can inhibit metastatic growth of tumour cells that induce systemic and local inflammatory response. Increasing CD200 activity/signaling might be an important therapeutic strategy for treatment of aggressive breast carcinomas.

168

Cure of metastatic growth of EMT6 tumour cells in mice following manipulation of CD200:CD200R signaling

Gorczynski RM, Chen Z, Khatri I, Podnos A, Yu K

Abstract

In previous studies, we observed that regulation of expression of CD200, both on cells of a transplantable breast cancer, EMT6, and of the host, as well as of the receptor, CD200R in host mice, regulated local tumour growth and metastasis in immunocompetent animals. This in turn led to an improved ability to document immunity to EMT6 in CD200R1-/- mice. In the current study, we have explored the ability to cure BALB/c CD200-/- or CD200R1-/- mice of tumours ≤1 cm3 in size by surgical resection of localized tumour, followed by immunization with irradiated EMT6 cells along with CpG as adjuvant. While control animals treated in this fashion developed significant pulmonary and liver metastases within 30 days of surgery, significant protection was seen in both CD200-/- or CD200R1-/- mice, with no macroscopic lung/liver metastases observed in CD200R1-/- mice on sacrifice at day 300. Following surgical resection and immunization, draining lymph nodes from control mice contained tumour cells cloned at limiting dilution in vitro even before pulmonary and hepatic metastasis was seen. In contrast, within the limits of detection of the assay used (sensitivity ~1 in 107 cells), no tumour cells were detected at limiting dilution in similarly treated CD200R1-/- mice, and significant reductions were seen in CD200-/- mice. Infusion of anti-CD4, but less so anti-CD8, mAb into surgically treated and immunized CD200R1-/- mice attenuated protection from both macroscopic

(liver/lung) and microscopic (assayed by limiting dilution of DLN) metastasis. Adoptive transfer of lymphocytes from treated CD200R1-/- mice to surgically treated control mice also attenuated metastatic growth of tumour, which was abolished by pretreatment of transferred

169 cells with anti-CD4 mAb. Our data suggest that CD200:CD200R attenuates a potentially tumour-protective CD4 host response to breast cancer.