Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Manipulating the tumor microenvironment to slow cancer growth Ho, Victor Wing Heng 2012

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2012_fall_ho_victor.pdf [ 4.23MB ]
Metadata
JSON: 24-1.0072765.json
JSON-LD: 24-1.0072765-ld.json
RDF/XML (Pretty): 24-1.0072765-rdf.xml
RDF/JSON: 24-1.0072765-rdf.json
Turtle: 24-1.0072765-turtle.txt
N-Triples: 24-1.0072765-rdf-ntriples.txt
Original Record: 24-1.0072765-source.json
Full Text
24-1.0072765-fulltext.txt
Citation
24-1.0072765.ris

Full Text

Manipulating the Tumor Microenvironment to Slow Cancer Growth  by  Victor Wing Heng Ho  B.Sc. (Honours), The University of British Columbia, 2005  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  The Faculty of Graduate Studies (Experimental Medicine)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April 2012 © Victor Wing Heng Ho, 2012  Abstract The tumor microenvironment encompasses all of the factors and accessory cells which interact with a tumor and, more often than not, are co-opted by tumorsecreted factors to help the cancer grow. In this thesis, we examined elements of the tumor microenvironment, specifically the cancer-promoting M2 macrophages (Ms)and tumor glucose supply and metabolism. In the vast majority of advanced cancer patients and tumor-bearing animals, their tumors contain Ms that are profoundly skewed to a cancer-promoting M2 phenotype, which often correlates with a poor prognosis. As well, these same tumor tissues are more dependent on high glucose levels for energy and survival than most normal tissues. Although M phenotype and tumor cell glucose metabolism are quite disparate fields of study, we employed similar strategies to uncover what regulates them in order to explore mechanisms to reduce tumor growth. In our M2 M studies, we assessed the phenotypic plasticity of mature IL-4induced M2 Ms and found their phenotype (i.e., cell-surface markers, cytokine secretion, and T cell stimulatory properties) could be fully reversed with IL-4 withdrawal and subsequent IFN- priming, demonstrating that M2 Ms, indeed, can be reprogrammed even if they are phenotypically polarized. As well, we uncovered a circuit of M2-M generation which may be relevant in vivo in the M2-skewed SHIP-/mouse model. This circuit involves the sensitization of Ms and M progenitors, via IgG and TGF--containing mouse plasma, to low levels of constitutive IL-4 uniquely secreted by SHIP-/- basophils. In a similar vein, and based on the Warburg effect, which describes the propensity of cancer cells to consume glucose via glycolysis rather than oxidative phosphorylation (OXPHOS), we designed low carbohydrate (CHO) diets to limit the glucose supply to tumors. We found that mice fed our low CHO diets had lowered blood glucose (BG), insulin (Ins), and lactate, and this correlated with a slower growth rate of implanted tumors, and a lower cancer incidence in a spontaneous mouse mammary carcinoma model. Furthermore, our diets worked additively with known anti-cancer agents (i.e.,Temsirolimus, Celebrex) to slow tumor growth.  ii  Preface I was responsible for all of the experimental design, data collection, data analysis, thesis preparation and editing, under the supervision of Dr. Gerald Krystal at the Terry Fox Laboratory in the British Columbia Cancer Research Centre, except for the parts stated below. Certain parts of Chapter 4 are based on the work that has been published in The Journal of Immunology (Kuroda et al., 2009; Kuroda et al. 2011). In the 2011 publication, Dr. Etsushi Kuroda was the lead author, under the supervision of Dr. Gerald Krystal, and did the majority of the in vitro work and manuscript writing, with Dr. Frann Antignano (Biomedical Research Centre) and Dr. Michael Hughes (Biomedical Research Centre) being primarily responsible for the in vivo mouse model, which was provided by Dr. McNagny (Biomedical Research Centre). Dr. Toshiaki Kawakami provided reagents (La Jolla Institute for Allergy & Immunology). I was  responsible  for the  DX5+  and  DX5-  BM  fractionation  studies  and  characterization of basophilia in the SHIP-/- mouse. All the other authors, including the ones already acknowledged above had intellectual input in the study and manuscript writing. In the 2009 publication (Kuroda et al., 2009), Dr. Etsushi Kuroda, under the supervision of Dr. Gerald Krystal, was responsible for the majority of the experimental work, analysis and manuscript writing. I was responsible for the macrophage progenitor IL-4 sensitivity studies, as well as some experimental repeats. Drs. Antov and Flavell provided reagents. All other authors, including myself, contributed intellectually to data analyses and discussion. Also in Chapter 4, the basophil IgG stimulation studies (unpublished) were carried out with Dr. Jens Ruschmann, who provided both intellectual and technical guidance. A version of Chapter 6 is published in Cancer Research (Ho et al., 2011), and has been cited twice by the faculty of 1000: Ho VW, Leung K, Hsu A, Luk B, Lai J, Shen SY, Minchinton AI, Waterhouse D, Bally MB, Lin W, Nelson BH, Sly LM, Krystal G. (2011). A low carbohydrate, high protein diet slows tumor growth and prevents cancer initiation. Cancer Res. 2011 Jul 1;71(13):4484-93. iii  Under the supervision of Dr. Gerald Krystal, I was the lead author in this study and was responsible for the technical work, data gathering, analysis, and figure and manuscript preparation and editing. The design of the diets was a joint effort between Dr. Gerald Krystal and myself. Kelvin Leung, Anderson Hsu, Beryl Luk, June Lai, and Sung Yuan Shen provided technical assistance, under my supervision. Dr. Minchinton provided some tumor cell lines, some tumor injection equipment, and expertise with in vivo tumor studies. Drs. Waterhouse and Bally cosupervised the human xenograph studies, which were performed by members of their labs, and Dr. Nelson (Deeley Research Centre) supervised the spontaneous cancer model, and animal monitoring, necropsy, and plasma extraction for those studies were performed by Wendy Lin. All authors contributed intellectually to discussion. All of this work was co-supervised by Dr. Gerald Krystal and myself. Vivian Lam (Terry Fox Laboratory) provided technical assistance in biochemical protocols (i.e., mouse plasma immunoglobulin depletion, mouse plasma ELISAs) and tumor cell-line maintenance. Kelvin Leung, Anderson Hsu, Beryl Luk, June Lai, and Sung Yuan Shen provided general technical assistance, under my supervision. Mice were housed in the Animal Resource Centre (ARC) with the assistance of ARC staff or at the Deeley Research Centre. Mouse studies were performed under UBC Animal Care protocols A07‐0221, A10-0039 (Dr. Gerald Krystal).  Publications Arising from Work in this Thesis: (Chapter 6) Ho VW, Leung K, Hsu A, Luk B, Lai J, Shen SY, Minchinton AI, Waterhouse D, Bally MB, Lin W, Nelson BH, Sly LM, Krystal G. (2011) A low carbohydrate, high protein diet slows tumor growth and prevents cancer initiation. Cancer Research. 2011 Jul 1;71(13):4484-93.  iv  Publications on which this Thesis is Partially Based: (Chapter 4) Kuroda E, Ho V, Ruschmann J, Antignano F, Hamilton M, Rauh MJ, Antov A, Flavell RA, Sly LM, Krystal G. (2009). SHIP represses the generation of IL-3-induced M2 macrophages by inhibiting IL-4 production from basophils. Journal of Immunology. 2009 Sep 15;183(6):3652-60.  (Chapter 4) Kuroda E, Antignano F, Ho VW, Hughes MR, Ruschmann J, Lam V, Kawakami T, Kerr WG, McNagny KM, Sly LM, Krystal G. (2011). SHIP represses Th2 skewing by inhibiting IL-4 production from basophils. Journal of Immunology. 2011 Jan 1;186(1):323-32.  v  Table of Contents Abstract ............................................................................................................ ii Preface ............................................................................................................. iii Table of Contents ........................................................................................... vi List of Tables.................................................................................................. xii List of Figures ............................................................................................... xiii List of Symbols and Abbreviations ............................................................. xvi Acknowledgements ...................................................................................... xxi Dedication .................................................................................................... xxii 1.  Introduction ............................................................................................... 1 1.1  Overview of the Thesis ............................................................................. 1  1.2  Hallmarks of Cancer ................................................................................. 1  1.3  The Tumor Microenvironment ................................................................... 2  1.4  The History of Cancer and Immunotherapy............................................... 3  1.5  Cancer Immunotherapy and the Immune Microenvironment ..................... 4  1.5.1 Monocyte and Macrophage Phenotypes .................................................... 6 1.5.1.1 Macrophage Polarization ................................................................... 9 1.5.1.2 Macrophage Polarization and Cancer ...............................................10 1.5.1.3 SHIP and M2 Macrophages ..............................................................14 1.6  Energy Metabolism and Glycolysis ..........................................................14  1.6.1 The Warburg Effect .................................................................................. 17 1.6.2 Aerobic Glycolysis and Cell Proliferation .................................................. 18 1.6.3 Lactate Metabolism and Cancer ............................................................... 19 1.7  Glucose and Dietary Carbohydrate Metabolism .......................................21  1.7.1 Blood Glucose, Insulin and Cancer .......................................................... 22 1.8  2.  Manipulating the Tumor Microenvironment ..............................................23  Materials and Methods ........................................................................... 25 2.1  Mice .........................................................................................................25  2.2  Cell Cultures ............................................................................................25  vi  2.2.1 Macrophage Cultures ............................................................................... 25 2.2.2 Bone Marrow Environment Assay ............................................................ 26 2.2.3 Macrophage Harvests .............................................................................. 27 2.2.4 Cell Lysis .................................................................................................. 27 2.2.5 BM-Derived Basophils .............................................................................. 28 2.2.6 BMB and BMM Co-cultures ................................................................... 29 2.2.7 Peritoneal Ms......................................................................................... 29 2.3  Reagents, Inhibitors, Cytokines ...............................................................30  2.4  Quantitative Assays .................................................................................30  2.4.1 Arginase Assay ........................................................................................ 30 2.4.2 Bradford Assay......................................................................................... 31 2.4.3 BCA Protein Assay ................................................................................... 31 2.4.4 M-Produced Cytokines .......................................................................... 32 2.4.5 Measurement of Blood Glucose, Insulin, and Lactate ............................... 32 2.4.6 T Cell Proliferation Assay ......................................................................... 33 2.5  SDS-PAGE and Western Blotting ............................................................33  2.5.1 Polyacrylamide Gels................................................................................. 33 2.5.2 Electrophoresis and Immunoblotting ........................................................ 34 2.6  Flow Cytometry ........................................................................................35  2.7  Antibodies ................................................................................................35  2.7.1 Western Blotting ....................................................................................... 35 2.7.2 Flow Cytometry and Cell Enrichment........................................................ 36 2.7.3 Functional Assays .................................................................................... 36  3.  2.8  IgE and IgG Depletion..............................................................................36  2.9  Depletion of Basophils in vivo and IL-4 Neutralization ..............................37  2.10  In vivo Tumor Cell Studies .......................................................................37  2.11  Mouse Diets .............................................................................................38  2.12  Statistical Analyses ..................................................................................38  Plasticity of Macrophage Polarization ................................................... 39 3.1  Introduction ..............................................................................................39  3.1.1 Surface Markers of M2 Macrophages ....................................................... 39 3.1.2 Arginine Metabolism ................................................................................. 39 3.1.3 Macrophage Immune Regulation.............................................................. 40  vii  3.1.4 Macrophage Phenotype Plasticity ............................................................ 41 3.2  Results ....................................................................................................42  3.2.1 IL-4 Induces an M2, Arg1+ M Phenotype ............................................... 42 3.2.2 M2 Activation Leads to a Wound-Healing, Arg1+ M Phenotype .............. 44 3.2.3 M1 and M2 Ms Display Unique Cell-Surface Antigens ........................... 45 3.2.4 IL-4 Withdrawal is Sufficient to Reverse M2 Arginine Metabolic and Cell-Surface Antigen Expression Phenotypes .......................................... 47 3.2.5 M2-primed BMMs can be M1 programmed with IFN-γ ........................... 49 3.2.6 M2 Priming Shifts M Cytokine Production Profile to a Proinflammatory M1-like Phenotype. ............................................................. 53 3.2.7 M2-Primed BMMs Induce Less Antigen-Specific T Cell Proliferation than M1-Primed BMMs.......................................................................... 57 3.2.8 M2 M Phenotype of SHIP-/- BMMs ...................................................... 59 3.3  Discussion ...............................................................................................67  3.3.1 Surface Marker Phenotype ....................................................................... 67 3.3.2 Arginine Metabolism and M2 Programming .............................................. 69 3.3.3 Immune Regulatory M Phenotype.......................................................... 72 3.3.4 M2 Phenotype of SHIP-/- Ms .................................................................. 75 3.3.5 M M2 Phenotype Reversibility ............................................................... 75  4.  Identifying the Factors that Promote the SHIP-/- M2 Macrophage  Phenotype ...................................................................................................... 77 4.1  Introduction ..............................................................................................77  4.1.1 In vivo Derived Arg1+ M2 Ms ................................................................. 77 4.1.2 M2 Ms in SHIP-/- Mice ............................................................................ 78 4.1.3 Elucidation of the Factors Responsible for the M2 Skewing of SHIP-/Ms ......................................................................................................... 79 4.2  Results ....................................................................................................79  4.2.1 IL-4 is Essential for Plasma-Induced M2-Skewing of SHIP-/- BMMs ....... 79 4.2.2 Stimulated Basophils Secrete IL-4 and are Required for M2-Skewing in vitro ...................................................................................................... 81 4.2.3 MP Does Not Stimulate IL-4 Secretion from Intact Adh- BM Cells ............ 85 4.2.4 MP Does Not Increase IL-4 Production More than M-CSF from DX5+ SHIP-/- BM Cells ....................................................................................... 86 viii  4.2.5 MP Drives SHIP-/- M2 Skewing Via a Distinct Mechanism from IL-3 ......... 87 4.2.6 IgG, but not IgE, is Necessary for MP-Induced M2-Skewing of SHIP-/BM ........................................................................................................... 89 4.2.7 MP and IgG Sensitize BM Progenitors and BMMs to IL-4 ...................... 93 4.2.8 M2-skewing is induced by DX5+ BM-produced IL-4 and MP-induced Sensitization of BMM Progenitors ......................................................... 95 4.2.9 Targeting the Basophil-IL-4 Axis Does Not Reduce SHIP-/- M2Skewing in vivo ........................................................................................ 96 4.3  Discussion ...............................................................................................99  4.3.1 IL-4-Producing DX5+ Cells are Necessary and Sufficient for M2Skewing ................................................................................................... 99 4.3.2 MP-Induced M2-skewing of SHIP-/- BM is Mediated not by an Increase in IL-4 Levels, but an Increase in IL-4 Sensitivity ..................... 101 4.3.3 IgG in MP Promotes M2 Skewing by Increasing the Sensitivity of BM progenitors to IL-4.................................................................................. 102 4.3.4 Differences Between SHIP+/+ and SHIP-/- BM Contributing to Their Difference in M2-Skewing Behavior in vitro ............................................ 103 4.3.5 Reversing SHIP-/- M2-Skewing in vivo .................................................... 106  5.  The Role of COX-2 and TGF- Pathways in M2-Skewing of  Macrophages ................................................................................................ 107 5.1  Introduction ............................................................................................107  5.1.1 M2-Skewing in TAMs ............................................................................. 107 5.1.2 Factors Influencing M2-Skewing............................................................. 107 5.2  Results ..................................................................................................109  5.2.1 TGF- is Not Sufficient to Recapitulate Plasma-Induced SHIP-/- M2 MSkewing .......................................................................................... 109 5.2.2 Prostaglandin E2 (PGE2) Induces the SHIP-/- M2 Phenotype .................. 110 5.2.3 COX-2 Activity is Necessary for Plasma-Induced M2-Skewing of SHIP-/- BMMs ...................................................................................... 114 5.2.4 PGE2 and TGF- Sensitize BMM Progenitors to IL-4 ........................... 117 5.2.5 COX-2 Inhibition Ameliorates in vivo M2-skewing in the SHIP-/Mouse .................................................................................................... 122 5.3  Discussion .............................................................................................123  ix  5.3.1 PGE2 and COX-2 Synergize with TGF- to Skew SHIP-/- BM to M2 Ms ....................................................................................................... 123 5.3.2 The Roles of PGE2 and TGF-in in vivo M2-Skewing ............................ 125 5.3.3 COX-2, TGF-, and MP-Mediated IL-4 Sensitization .............................. 126  6.  A Low Carbohydrate, High Protein Diet Slows Tumor Growth and  Prevents Cancer Initiation ........................................................................... 129 6.1  Introduction ............................................................................................129  6.1.1 Glucose .................................................................................................. 129 6.1.2 Diet and Carbohydrates.......................................................................... 130 6.1.3 Dietary Intervention ................................................................................ 130 6.2  Results ..................................................................................................132  6.2.1 Tumors Grow Slower in Mice on an 8% CHO, 69% Protein, 23% Fat (8% CHO) Diet, but the Mice Lose Weight ............................................. 132 6.2.2 A 15% High Amylose CHO, 58% Protein, 26% Fat (15% CHO) Diet Reduces Fasting and Constitutive BG, and Tumor Growth .................... 135 6.2.3 A 10% CHO Diet Slows Tumor Growth More than a 15% CHO Diet without Significant Weight Loss ............................................................. 138 6.2.4 Low CHO Diets Cause a Drop in Plasma Insulin and Lactate ................. 140 6.2.5 Low CHO Diets Act Additively with Known Cancer Therapeutic Agents to Reduce Tumor Growth ........................................................... 143 6.2.6 The 15% CHO Diet Reduces the Incidence of Tumors in a Spontaneous Mouse Model of Breast Cancer ........................................ 145 6.2.7 Our Low CHO Diets Do Not Increase Ketosis......................................... 147 6.3  Discussion .............................................................................................147  6.3.1 Diet-Induced Biochemical Changes Associated with Reduced Cancer Growth ................................................................................................... 148 6.3.2 Low CHO Diets as Complementary Treatments ..................................... 149 6.3.3 Blood Glucose Regulation in Humans .................................................... 149 6.3.4 Dietary Macronutrient Proportions .......................................................... 151  7.  Summary and Conclusions .................................................................. 152 7.1  Targeted Therapy and the Tumor Microenvironment .............................152  7.2  Reversing Pro-Tumor M2 Ms ..............................................................154  7.3  Manipulating M2 M Generation in vivo ................................................155 x  7.4  Manipulating the Metabolic Environment................................................159  7.5  Limitations of This Thesis ......................................................................160  7.5.1 In Translation ......................................................................................... 162 7.6  Future Directions ...................................................................................163  References ................................................................................................... 166  xi  List of Tables Table 6.1 Macronutrient breakdown of diets used. ........................................ 133  xii  List of Figures Figure 1.1 Model of the reversibility of M2 functional phenotypes. ......................... 13 Figure 1.2 Intracellular fate of glucose. ................................................................... 16 Figure 3.1 IL-4 induces M2 M skewing of mature BMMs. .................................. 43 Figure 3.2 M1 and M2 programmed BMMs display different L-arginine metabolic strategies........................................................................................ 45 Figure 3.3 IFN--induced M1 priming increases CD204 and IL-4 ± IL-10 M2 priming increases CD11c on BMMs............................................................. 46 Figure 3.4 The IL-4-induced phenotype can be reversed by withdrawing IL-4. ....... 48 Figure 3.5 M2-primed BMMs can be reprogrammed with IFN-. .......................... 50 Figure 3.6 M2-primed BMMs can be reprogrammed............................................ 52 Figure 3.7 M2-activated BMMs produce more IL-10 and less IL-12 or TNF-. .... 53 Figure 3.8 M2-primed BMMs produce less IL-10 and TNF- but more IL-12 in response to subsequent LPS stimulation than control BMMs. ..................... 54 Figure 3.9 M2-activated BMMs can be reverse activated and are primed for M1-activation. ................................................................................................. 56 Figure 3.10 M1-primed BMMs are better inducers of Ag-specific T cell proliferation than M2-primed BMMs. ............................................................ 58 Figure 3.11 SHIP-/- M2 BMM phenotype is reversible. ......................................... 60 Figure 3.12 The SHIP-/- M2 BMM surface marker phenotype is reversible. ......... 62 Figure 3.13 LPS-stimulated M2-primed SHIP-/- BMMs produce more IL-12 and IL-10 compared to unprimed BMMs............................................................. 63 Figure 3.14 Functional phenotype of M2-primed SHIP-/- BMMs is reversible and similar to that of SHIP+/+ BMMs............................................................. 65 Figure 3.15 SHIP-/- BMMs induce less Arg1 in response to IL-4. ......................... 66 Figure 3.16 Model of the reversibility of the M2 cell surface and L-arginine metabolic phenotypes..................................................................................... 71 Figure 3.17 Model of the reversibility of M2 functional phenotypes. ....................... 74  xiii  Figure 4.1 MP-induced skewing of SHIP-/- BM during differentiation to M2 BMMs is dependent on IL-4. ........................................................................ 81 Figure 4.2 DX5+ BM cells are required for MP- or IL-3-mediated M2-skewing of SHIP-/- BM. ..................................................................................................... 82 Figure 4.3 Although MP stimulates IL-4 production from SHIP-/-, but not SHIP+/+, BMBs, it M2-skews Ms in mature BMMs co-cultured with either SHIP+/+ or SHIP-/- BMBs. ............................................................................................. 84 Figure 4.4 MP induces Arg1 in differentiating SHIP-/- BMMs without stimulating a significant increase in IL-4. .......................................................................... 85 Figure 4.5 MP does not stimulate more IL-4 production from SHIP-/- DX5+ cells then M-CSF alone. ......................................................................................... 87 Figure 4.6 IL-3 is not responsible for the M2-skewing activity of MP. ..................... 88 Figure 4.7 IgE is not responsible for MP-induced M2-skewing of SHIP-/- BM. ........ 90 Figure 4.8 IgG in MP induces M2-skewing of SHIP-/- BM........................................ 92 Figure 4.9 MP increases the responsiveness of DX5- BM cells and mature BMMs to IL-4. .............................................................................................. 94 Figure 4.10 IgG sensitizes BMMs to IL-4 during and after maturation. ................ 95 Figure 4.11 SHIP-/- DX5+ cells enable SHIP+/+ or SHIP-/- DX5- BM to be skewed to an M2 phenotype during BMM differentiation. ......................................... 96 Figure 4.12 Targeting the basophil-IL-4 axis in vivo does not reduce M2 skewing. ......................................................................................................... 98 Figure 4.13 The SHIP-/- peritoneal Arg1+ M phenotype is not reversible. ............. 99 Figure 4.14 Model of M2-skewing in maturing Ms in vivo and in vitro: Roles of SHIP and basophil/basophil progenitors on M progenitors. ....................... 105 Figure 5.1 TGF- fails to induce a similar level of arginase activity as plasma. .... 110 Figure 5.2 PGE2 and TGF- induce the M2 M phenotype in SHIP-/- BMMs during maturation.......................................................................................... 112 Figure 5.3 PGE2 and TGF- act synergistically to induce arginase activity in maturing SHIP-/- BMMs. ............................................................................. 113  xiv  Figure 5.4 COX-2 activity and TGF- signaling are both crucial to Arg1 induction by plasma in maturing SHIP-/- BMMs. ........................................................ 115 Figure 5.5 Both Celebrex and SC-58125 repress MP-induced M2 skewing. ........ 116 Figure 5.6 PGE2 + TGF--induced M2-skewing of SHIP-/- BMMs during differentiation is dependent on IL-4. ............................................................. 117 Figure 5.7 The combination of PGE2 and TGF- does not increase IL-4 production from BMBs or DX5+ BM. ............................................................. 118 Figure 5.8 PGE2 and TGF- sensitize BMMs to IL-4, and COX-2 and TGF- signaling are necessary for MP-sensitized BMM response to IL-4 ............ 121 Figure 5.9 COX-2 inhibition reduces the ex vivo and in vivo Arg1+ phenotype of SHIP-/- PMs. ............................................................................................... 123 Figure 5.10 Model of M2-skewing in mature Ms and M progenitors: the roles of MP, COX-2, and TGF- ............................................................................ 127 Figure 6.1 Tumors grow slower in mice on an 8% CHO diet than those on a Western diet, but the mice weigh less. ......................................................... 134 Figure 6.2 Tumors grow slower in mice on a 15% CHO diet than a Western diet, and the mice weigh the same. ...................................................................... 137 Figure 6.3 A 10% CHO diet is more effective than the 15% CHO diet at slowing tumor growth with only a slight effect on mouse weight. ............................... 139 Figure 6.4 Low CHO diets reduce plasma insulin and lactate levels..................... 142 Figure 6.5 Low CHO diets act additively with current treatments for cancer. ........ 144 Figure 6.6 The 15% CHO diet reduces the incidence of tumors in a spontaneous mouse model of breast cancer. ............................................... 146 Figure 6.7 Plasma -hydroxybutyrate levels in low CHO-fed mice are the same as those in 5058-fed mice. ........................................................................... 147 Figure 7.1 Model of M2-skewing in maturing Ms in vivo and in vitro. ................. 158 Figure 7.2 Unified Model of M2-skewing in vivo. ................................................... 161  xv  List of Symbols and Abbreviations 15DPGJ2 = 15-Deoxy-Δ12,14-Prostaglandin J2 2-FDG = 18Fluorodeoxyglucose 3 [H]Thymidine = Tritiated-Thymidine Ab = Antibody Adh- = Adherence-Depleted Ag = Antigen AGE = Advanced Glycation End Products Akt = Protein Kinase-B APC = Antigen Presenting Cell APhC = Allophycocyanin Arg1 = Arginase 1 ATIR = Anti-Tumor Immune Response ATP = Adenosine Triphosphate BAD = Bcl2 Agonist of Cell Death BCM = Basophil Complete Medium BG = Blood Glucose BLP = Bacterial Lipoprotein BM = Bone Marrow BMM = Bone Marrow-Derived Macrophage BMB = Bone Marrow-Derived Basophil BMENV = Bone marrow Environment Assay BMM = Bone Marrow-Derived Macrophage Medium BSA = Bovine Serum Albumin BSM = Basophil Starve Medium CAT = Cationic Amino Acid Transporter CD = Cluster of Differentiation CD1a = Cluster of Differentiation-1a CD3 = Cluster of Differentiation-3 CD4 = Cluster of Differentiation-4 CD8 = Cluster of Differentiation-8 CD11b = Cluster of Differentiation-11b CD14 = Cluster of Differentiation-14 CD16 = Cluster of Differentiation-16 CD49b = Cluster of Differentiation-49b CD68 = Cluster of Differentiation-68 (LAMP) CD204 = Cluster of Differentiation-204 (SR-A) CD206 = Cluster of Differentiation-206 (MR) CD163 = Cluster of Differentiation-163 CDB = Cell Dissociation Buffer c-fms = MCSFR CCR2 = Chemokine (C-C Motif) Receptor-2 CX3CR1 = Chemokine (C-X3-C Motif) Receptor-1 CHO = Carbohydrate CMP = Common Myeloid Progenitor xvi  CO2 = Carbon Dioxide COX = Cyclooxygenase COX-1 = Cyclooxygenase-1 COX-2 = Cyclooxygenase-2 cpm = Counts Per Minute CR = Caloric Restriction CSF-1 = Colony Stimulating Factor-1 (M-CSF) CSF-2 = Colony Stimulating Factor-2 (GM-CSF) CTL = Cytotoxic T Lymphocyte CTLA-4 = Cytotoxic T-Lymphocyte-Associated Protein-4 DC = Dendritic Cell ddH2O = Deionized Distilled Water DMSO = Dimethyl Sulfoxide DX5+ = DX5-Enriched DX5- = DX5-Depleted ECM = Extracellular Matrix EGF = Epidermal Growth Factor ELISA = Enzyme-Linked Immunosorbent Assay EMR-1 = EGF-like Module Containing Mucin-Like Hormone Receptor-1 ERK = Extracellular Signal-Regulated Kinase ETC = Electron Transport Chain EtOH = Absolute Ethanol FCS = Fetal Calf Serum FG = Fermentative Glycolysis FITC = Fluorescein Isothiocyanate GAPDH = Glyceraldehyde 3-Phosphate Dehydrogenase GI = Glycemic Index GLUT = Glucose Transporter GM-CSF = Granulocyte Macrophage Colony Stimulating Factor (CSF-2) Grb2 = Growth Factor Receptor-Bound Protein-2 GSH = Glutathione HBSS = Hank's Balanced Salt Solution HF = Hank's Balanced Salt Solution + 2% FCS HFN = HF + 0.02% NaN3 HIF1- = Hypoxia-Induced Factor-1- IFN- = Interferon- Ig = Immunoglobulin IGF-1 = Insulin-Like Growth Factor-1 IL = Interleukin IMDM = Iscove‘s Modified Dulbecco‘s Media iNOS = Inducible Nitric Oxide Synthase (NOS-2) Ins = Insulin IP = Intraperitoneal IR = Insulin Receptor IRS-1 = Insulin-Regulated Substrate-1 IV = Intravenous  xvii  LAMP = Lysosome Associated Membrane Glycoprotein (CD68) LAP = Latency Associated Protein LDH = Lactate Dehydrogenase LDH-A = Type-A LDH Subunits LDH-B = Type-B LDH Subunits LDL = Low-Density Lipoprotein Lin- = Lineage Depleted LPS = Lipopolysaccharide MAPK = Mitogen-Activated Protein Kinase M-CSF = Macrophage Colony Stimulating Factor (CSF-1) M-CSFR = M-CSF Receptor MC = Mast Cell MCP-1 = Monocyte Chemotactic Protein-1 MCT = Monocarboxylate Transporter MCT-1 = Monocarboxylate Transporter-1 MCT-3 = Monocarboxylate Transporter-3 MCT-4 = Monocarboxylate Transporter-4 MDP = M/DC Progenitor MHC-I = Major Histocompatibility Complex-1 MHC-II = Major Histocompatibility Complex-2 M = Macrophage Mo = Monocyte MMP = Matrix Metalloproteinase MMTV = Mouse Mammary Tumor Virus MP = Mouse Plasma MR = Mannose Receptor (CD206) MSP = Macrophage Stimulating Protein MTG = Monothioglycerol mTOR = Mammalian Target of Rapamycin MyD88 = Myeloid Differentiation Primary Response Gene (88) NF-B = Nuclear Factor Kappa-b NAD+ = Nicotinamide Adenine Dinucleotide NADH = Reduced Nicotinamide Adenine Dinucleotide NADPH = Nicotinamide Adenine Dinucleotide Phosphate NCKD = No Carbohydrate Ketogenic Diet NK = Natural Killer NO = Nitric Oxide NP40 =Nonyl Phenoxylpolyethoxylethanol NSAID = Non-Steroidal Anti-Inflammatory Drug NSCLC = Non-Small Cell Lung Cancer O2 = Oxygen OVA = Ovalbumin OXPHOS = Oxidative Phosphorylation P/S = Penicillin + Streptomycin PAMP = Pathogen-Associated Molecular Pattern PBS = Phosphate Buffered Saline xviii  PDC = Pyruvate Dehydrogenase Complex PDK1 = 3-Phosphoinositide-Dependent Protein Kinase-1 PE = R-Phycoerythrin PET = Positron Emission Tomography PG = Prostaglandin PGE2 = Prostaglandin E2 PI = Propidium Iodide PI3K = Phosphatidylinositol 3-Kinase PI-4,5-P2 = Phosphatidylinositol 4,5-Bisphosphate PI-3,4-P2 = Phosphatidylinositol 3,4-Bisphosphate PIP3 = Phosphatidylinositol 3,4,5-Trisphosphate PM = Peritoneal M Medium PM = Peritoneal Macrophage PolyI:C = Polyinosinic-polycytidylic Acid PPAR- = Peroxisome Proliferator-Activated Receptor- P/S = Penicillin and Streptomycin P/T = PGE2 + TGF- PSA = Prostate-Specific Antigen PSB =Phosphate Solubilzation Buffer pSTAT-1 = Phosphorylated Signal Transducer and Activator of Transcription-1 pSTAT-6 = Phosphorylated Signal Transducer and Activator of Transcription-6 PVDF = Polyvinylidene Fluoride RAGE = Receptors for Advanced Glycation End Products RBC = Red Blood Cell RCC = Renal Cell Carcinoma rh = Recombinant Human rm = Recombinant Mouse rhTGF- = Recombinant Human TGF- rmIFN- = Recombinant Mouse IFN- rmIL-3 = Recombinant Mouse IL-3 rmIL-4 = Recombinant Mouse IL-4 rmIL-10 = Recombinant Mouse IL-10 RNS = Reactive Nitrogen Species ROS = Reactive Oxygen Species RPMI = Roswell Park Memorial Institute Medium 1640 SC = Subcutaneous SCCVII = Squamous Cell Carcinoma VII SDS = Sodium Dodecyl Sulfate SHC = Src Homology 2 Domain-Containing-Transforming Protein C1 SHIP = SH2-Containing Inositol 5'-Phosphatase SR-A = Scavenger Receptor Type A (CD204) STAT = Signal Transducer and Activator of Transcription STAT-1 = Signal Transducer and Activator of Transcription-1 STAT-3 = Signal Transducer and Activator of Transcription-3 STAT-6 = Signal Transducer and Activator of Transcription-6 T:T = Tris-Buffered 0.1% Triton X-100 Buffer xix  TA = Tumor Antigen TBS = Tris-Buffered Saline TCA = Tri-Carboxylic Acid TEMED = Tetramethylethylenediamine TH-1 = T Helper-1 TH-2 = T Helper-2 TGF- = Transforming Growth Factor- TKR = Tyrosine Kinase Receptor TLR = Toll-Like Receptor TNF- = Tumor Necrosis Factor- Treg = T Regulatory Cell TRIF = TIR-Domain-Containing Adapter-Inducing Interferon-β VEGF = Vascular Endothelial Growth Factor V/V = Volume Per Volume W/V = Weight Per Volume W/W = Weight Per Weight  xx  Acknowledgements I would like to sincerely thank Dr. Gerald Krystal for giving me a once in a lifetime opportunity to work in a world-class research laboratory, especially since I am not and never was a model student. I would like to thank him for his patience, mentorship, guidance, and for demonstrating to me the important nuances in verbal and written communication, which I had previously overlooked. Most of all, I would like to thank Gerry for showing me that success and good, ethical science are not mutually exclusive concepts, and that nice guys do not always finish last. Thanks also to my supervisory committee: Drs. Neil Reiner, Andrew Minchinton, and Kim Chi. Thank you for your guidance and patience, and for taking the time out of your busy schedules to supervise me. As well, I would like to thank my colleagues, past and present, in no particular order: Vivian Lam, Drs. Melisa Hamilton-Valenski, Jens Ruschmann, Frann Antignano, Etsushi Kuroda, and Michael Rauh. Thanks for the numerous discussions about science, food, life, and other things. Thank you for showing me new perspectives in science and life, and for keeping me sane through the fair weather and turbulence on this long but enjoyable voyage. Thanks also to my present and former students Kelvin Leung, Anderson Hsu, Beryl Luk, June Lai, Kevin Shen, Natalie Firmino, and Brian Hsu: your patience, dedication and understanding were invaluable. I would also like to acknowledge my parents for their limitless love and support, and for leaving their own up and coming careers and bright futures so that I have mine. Thanks also to the rest of my family for their constant support, encouragement, and affirmation. Lastly, I would like to thank my wife, Eveline. Thank you for your perpetual love and support. This hasn‘t been easy, but we‘re finally nearing the end. I started on this journey alone, but I‘m immensely grateful that you are crossing the finish line with me now. This would not mean nearly as much, were it not for you.  xxi  Dedication I dedicate this to my wife, family, friends, and to anyone who believes that our world can be explained through observation and reason.  xxii  1. Introduction 1.1  Overview of the Thesis The overarching goal of the work presented in this thesis was to gain insights  into how to manipulate the local tumor microenvironment to slow tumor growth. Because the tumor microenvironment encompasses a vast number of interacting cells and factors, we took two approaches to do this, specifically aiming at cellular and non-cellular components. Firstly, since the vast majority of successful tumors co-opt the immune system to help tumors grow and metastasize, we investigated the phenotypic plasticity of these pro-tumor immune cells to see if these phenotypes could be reversed in vitro and in vivo. Secondly, based on the heightened requirement of most tumor cells for glucose, we tried to reduce the levels of blood glucose (BG), via diet changes, to restrict tumor growth. Since a comprehensive background of these two, somewhat disparate, fields would be beyond a reasonable length for the Introduction of a PhD thesis, we have opted to concentrate only on those areas that directly relate to the studies we have undertaken.  1.2  Hallmarks of Cancer In their seminal review, Hanahan and Weinberg (2000) proposed six hallmarks  of cancer that normal cells must acquire in order to become highly malignant cancer cells. These six requirements are the ability of the cells to 'evade apoptosis', 'sustain angiogenesis', become 'insensitive to anti-growth signals', invade and metastasize', have 'limitless replicative potential', and be 'self-sufficient with regard to growth signals'. Concurrent with the emergence of this paradigm, basic research into cancer therapeutics have focused on reversing these hallmark capabilities, upon which tumor initiation and progression are thought to rely (Hanahan and Weinberg, 2000). In a more recent review, Hanahan and Weinberg (2011) have updated their six classic hallmarks of cancer in accordance with key findings of the last decade, and have added 2 emerging hallmarks. Specifically, the tumor microenvironment, which includes the milieu of cells, extracellular matrix (ECM), and soluble factors surrounding the tumor, is now being recognized as a major player in determining the onset and outcome of cancer. In fact, two elements involving the tumor 1  microenvironment, namely 'tumor-promoting inflammation' and 'deregulating cellular energetics', have been dubbed an 'enabling characteristic' and an 'emerging hallmark', respectively (Hanahan and Weinberg, 2011). This thesis focuses on these two emerging themes, two of the newest hallmarks of cancer. Specifically, the experiments herein were aimed at investigating cancer-promoting immune celltypes, the key factors for their generation, and the potential of reversing such phenotypes, as well as the role of glucose metabolism in cancer incidence and progression, and whether or not it could be manipulated to slow and prevent the growth and incidence of cancer.  1.3  The Tumor Microenvironment In a crude conceptual bifurcation, tumors are composed of a tumor mass,  which is made of up cancer cells, and a stroma made up of the array of different cells and factors surrounding the mass. The tumor microenvironment encompasses all of the non-tumor-cell parts of a primary tumor site, including cellular components such as blood vessels (endothelial cells and pericytes), fibroblasts, and resident or recruited immune cells, including antigen (Ag)-presenting macrophages (Msand dendritic cells (DCs), natural killer (NK) cells, granulocytes (e.g., neutrophils, basophils, and eosinophils), and T and B lymphocytes, as well as molecular components, such as the ECM, hormones and growth factors, and a milieu of nutrients and biomolecules (glucose, amino acids, lactate, etc.) (Mbeunkui and Johann, 2009; Mathupala et al., 2010). Although the tumor microenvironment can induce spontaneous tumor rejection, more often than not they support and facilitate the tumor's acquisition of various procancerous cellular adaptations (i.e., the hallmarks of cancer), as well as prime secondary sites for metastasis (Josson et al., 2010; Wang et al., 2011). Because of the roles played by each component of the tumor microenvironment in influencing tumor progression, viable therapeutics against cancer-promoting features of the microenvironment are currently of great interest to the scientific and pharmaceutical communities. In particular, with regard to the cellular components of the tumor microenvironment, tumor-associated immune cells and the anti-tumor immune  2  response (ATIR) are topics of special interest. As well, as far as the non-cellular components of the tumor microenvironment are concerned, glucose metabolism via aberrant (aerobic) glycolysis, is a topic of great interest.  1.4  The History of Cancer and Immunotherapy The involvement of the immune system in cancer has long been observed. In  fact, the correlation of acute inflammation, in the form of fever, and spontaneous tumor regression was documented as early as 1742 and is still being observed today (Thomas and Badini, 2011; Hoption Cann et al., 2002). More concrete developments linking the immune system and cancer came in 1909 when Paul Ehrlich proposed that the immune system actively prevents the formation and progression of tumors in a healthy host (Ichim, 2005). This hypothesis implied that tumor development represented a breakdown in the immune system's ability to destroy incipient malignancies and led to research aimed at exploring the roles in cancer played by many components of the immune system, both adaptive and innate, in the destruction of tumor cells. A number of significant milestones in immunotherapy research were achieved as a result, including the discovery that carcinogen-induced cancer cells are often highly immunogenic (Prehn and Prehn, 2008; Prehn and Main, 1957) and that T cells and the adaptive immune response are critical for tumor rejection (Vessiere et al., 1982). Specifically, it was demonstrated that an interferon- (IFN-)-mediated T helper-1 (TH-1) response, via cluster of differentitaion (CD)8+ cytotoxic T lymphocytes (CTLs), as opposed to an IL-4-mediated T helper-2 (TH-2) antibody (Ab) response, is important in the ATIR (Vessiere et al., 1982). Since the identification of the roles of CTLs and the adaptive immune system in the ATIR, it was recognized that innate immune cells play a prominent part as well. IFN--activated Ms, for example, are important anti-tumor effectors (Buhtoiarov et al., 2007). Also of note, NK cells, like CTLs, were shown to mediate antitumor immunity, via perforin-dependent cell lysis, and be important players in the ATIR (Sanchez et al., 2011; Ruggeri et al., 2002; van den Broek et al., 1995). Instead of recognizing specific tumor antigens (TAs), however, NK cells detect and kill cancer cells expressing low levels of type-1 major histocompatibility complex (MHC-I) 3  molecules, which escape the MHC-dependent Ag-specific CTL response (Sanchez et al., 2011). Although these were not the only significant advances, they highlighted the critical role of the immune system in preventing tumor formation and helped to establish this fundamental principle upon which much of modern immunotherapy is based: a CTL-mediated tumor Ag-specific TH-1 response is the type of ATIR that leads to tumor rejection. Furthermore, it opened an entirely new avenue of cancer therapy, which aims to mobilize the immune system against tumors with vaccines, not unlike those that are commonly used effectively against viral diseases.  1.5  Cancer Immunotherapy and the Immune Microenvironment Early cancer vaccines attempting to immunize hosts against TAs, though based  on sound immunological principles, have been disappointing in vivo, and do not seem to have the curative effect of most viral vaccines (Schreiber et al., 2010; Yanelli and Wroblewski, 2004; Vasievich and Huang, 2011). However, a recent study has demonstrated that immunizing prostate cancer patients with a vaccinia virus expressing prostate-specific antigen (PSA) increased overall survival (Kantoff et al., 2010b). What is confounding is that sometimes, despite evidence indicating a robust TH-1 response, vaccines somehow fail to induce tumor regression (Triozzi et al., 2010; Yanelli and Wroblewski, 2004; Vasievich and Huang, 2011). To further add to the confusion, and quite counter-intuitively, the presence of immune cells, such as T cells and Ms, which is commonplace in the vast majority of tumors, correlated with a poor prognosis in most cases (Ruffell et al., 2011; Makitie et al., 2001; Steidl et al., 2010; Sugihara et al., 2009). To help understand this phenomenon, it is important to discuss the cancer immunosurveillance model of tumor initiation and progression. This model suggests that tumors arise during the 'initiation' phase from one or more normal cells transformed via carcinogen exposure, age-related mutations, and/or inherited cancer-promoting genetic variation (Dunn et al., 2002; Vesely et al., 2011). These cells, which may or may not express specific TAs, are subject to a spontaneous ATIR (Vesely et al., 2011; Czeh et al., 2010; Kandalaft et al., 2011). Importantly, like all immune responses, the early stages entail the  4  infiltration of innate leukocytes, most prominently NK cells, DCs, and Ms, after which come the effectors of the adaptive immune system, TA-specific T cells and tumor-specific Abs (Faget et al., 2011; Kobukai et al., 2011; Nagashio et al., 2008). An effective response comes about when Ag-presenting cells (APCs), such as DCs and Ms, detect and present TAs to T cells within draining lymph nodes to induce Ag-dependent T cell clonal expansion, and differentiation into CD4+ TH-1 cells and CD8+ CTLs, which in turn, find and eliminate all TA-expressing tumor cells, leaving tumor cells expressing low levels of MHC-I for NK cells to eradicate. The tumor goes through the 'elimination' phase of the immunosurveillance model if the ATIR eradicates all of the tumor cells (Dunn et al., 2002; Vesely et al., 2011). In many cases, however, ATIR fails to completely eliminate the cancer and exerts a selective pressure on tumor cells to evade the immune system by lowering their TA-expression and releasing NK cell inhibitory ligands to suppress the NK response (Vesely et al., 2011; Ashiru et al., 2010; Dunn et al., 2002). Concurrent with this, tumor cells actively derail other elements of the ATIR by producing, or by eliciting the production from stromal cells, of cytokines, such as interleukin-(IL)-10, transforming growth factor-(TGF)-, macrophage colony stimulating factor (M-CSF or CSF-1), granulocyte macrophage colony stimulating factor (GM-CSF or CSF-2), and vascular endothelial growth factor (VEGF), to skew infiltrating leukocytes towards an anergic or tumor-promoting phenotype (Whiteside, 2010). For example, infiltrating lymphocytes exposed to IL-10 or TGF- become T regulatory cells (Tregs), while infiltrating myeloid cells become anergic DCs via VEGF, M-CSF, or GM-CSF stimulation, myeloid-derived suppressor cells (MDSCs) via PGE2 or GMCSF stimulation, or anti-inflammatory Ms via IL-10 stimulation (Ochoa et al., 2007b; Whiteside, 2010; Fichtner-Feigl et al., 2008; Gordon and Martinez, 2010; Almand et al., 2000). Briefly, all of these cell phenotypes dampen down a robust ATIR through direct or indirect mechanisms that prevent the induction of an effective, CTL-mediated TH-1 ATIR (Ochoa et al., 2007b; Whiteside, 2010; Fichtner-Feigl et al., 2008; Gordon and Martinez, 2010; Almand et al., 2000). For example, immature or anergic DCs as well as M2 Ms are poor Ag presenters (i.e., fail to induce a productive Ag-specific T cell 5  proliferation response), and do not, therefore, initiate an effective T H-1 response (Gordon and Martinez, 2010; Almand et al., 2000). Similarly, the many subsets of Tregs and MDSCs suppress T cell proliferation through a variety of contactdependent or bystander mechanisms (Whiteside, 2010; Ochoa et al., 2007b). This tug-of-war between the host ATIR and tumor cell adaptation is known as the 'cancer immunoediting' phase of immunosurveillance, during which there is significant immune infiltration of the tumor (Vesely et al., 2011; Dunn et al., 2002). The tumor goes through the 'immune escape' phase if it succeeds in suppressing the ATIR to allow it to grow in an uninhibited fashion. This local immunosuppressive tumor microenvironment, as well as a systemic immune suppression that is often found in advanced cancer patients may explain why cancer vaccines have not been more effective (De Boniface et al., 2011). However, based on the advances in our understanding of the innate immune activation of T H cells, recent cancer immunotherapies designed to augment host immunity against tumor-mediated immune suppression, such as Sipuleucel-T (i.e., autologous APC activation therapy) and ipilimumab (i.e., an Ab against the negative co-receptor, cytotoxic T-lymphocyteassociated protein-4 [CTLA-4] on T cells), have demonstrated clinical effectiveness, which underscores the importance of the ATIR as a therapeutic target (Kantoff et al., 2010a; Margolin et al., 2012). Because of the sequential nature of the immune response, it seems, therefore, that the fate of a developing tumor depends largely on the nature of the infiltrating leukocytes, such as Ms, during the early stages of tumor development. Specifically, whether they adopt an activated immunogenic phenotype or succumb to the influence of the tumor to adopt a non-immunogenic phenotype, determines if they stimulate a robust ATIR (elimination) or help to create an immunosuppressive tumor microenvironment (escape). Since Ms are both initiators and effectors of the ATIR, we focused on their function and programming.  1.5.1  Monocyte and Macrophage Phenotypes  Ms are myeloid mononuclear white blood cells first identified for their large size and phagocytic properties (Takahashi, 2001). They are sentinels in the innate 6  immune system, located throughout different tissues, including the lungs, gut, and spleen, and perform a number of immune and non-immune-related roles (Stefater et al., 2011; Takahashi, 2001). Although Ms are a heterogeneous and phenotypically plastic cell-type, they all originate in the bone marrow (BM) from a common myeloid progenitor (CMP) (Geissmann et al., 2010). This CMP becomes a M/DC progenitor (MDP), and eventually a monocyte (Mo), at which point it leaves the BM and enters the bloodstream where it is carried to destination tissues (Geissmann et al., 2010; Pixley and Stanley, 2004). While the CMP and MDP are not very well defined, it is known that there are two main Mo subsets in the blood: CD16+ and CD16- in humans, and Ly-6Clo and Ly-6CHigh in mice, respectively (Ingersoll et al., 2010; Geissmann et al., 2003). The Ly-6Clo subset expresses chemokine (C-X3-C motif) receptor-1 (CX3CR1) but not chemokine (C-C motif) receptor-2 (CCR2), while the Ly-6Chigh subset expresses CCR2 but not CX3CR1 (Geissmann et al., 2003). Although the issue is still controversial, it has been suggested that the Ly-6Chigh subset may represent an earlier stage in M maturation, since they give rise to Ly-6Clo Mos in vivo (Yona and Yung, 2010). The last step in becoming a M occurs when the Mo extravasates out of the bloodstream and into tissues to become a fully matured M (Takahashi, 2001). Interestingly, it has recently been shown that the Ly-6Clo subset patrols the endothelium and is the first responder to extravasate into tissues following a proximal wound or injury (Auffray et al., 2007). Furthermore, while these Mos primarily differentiate into Ms, Ly-6Chigh Mos, which infiltrate after, tend to differentiate into inflammatory DCs (Auffray et al., 2007). This process is regulated by many growth factors, but most prominently, by MCSF, which stimulates progenitor cells expressing the M-CSF receptor (M-CSFR) to survive and differentiate into mature Ms (Pixley and Stanley, 2004). While the MCSFR is a good cell-surface marker for the M lineage, most mature Ms also express  the  adhesion  molecule  CD11b,  lysosome-associated  membrane  glycoprotein (LAMP or CD68), the hemoglobin receptor CD163, scavenger receptorA (SR-A or CD204), mannose receptor (MR or CD206), and the Toll-like receptor (TLR)-4 co-receptor molecule CD14 (Gordon, 1999; Mantovani et al., 2004). In 7  murine Ms, epidermal growth factor (EGF)-like module containing mucin-like hormone receptor (EMR)-1, which is recognized by the F4/80 monoclonal Ab, is a marker for mature Ms (Lin et al., 2010). Ms can be divided into two subsets: resident or tissue Ms and inflammatory Ms. Resident Ms, including alveolar, splenic, and peritoneal Ms (PMs), reside in tissues for an extended period of time and are responsible for a whole host of vital homeostatic processes including: red blood cell (RBC), iron, and hemoglobin turnover; apoptotic-cell clearance; lipid metabolism; and wound healing (Stefater et al., 2011). It has been demonstrated that resident Ms can arise from Ly-6Clo Mos in the circulation, and that this pattern of migration and differentiation is not significantly disturbed during an infection (Geissmann et al., 2003). Inflammatory Ms, on the other hand, seem to differentiate from the CCR2-expressing Ly-6Chigh Mo, only arising during inflammation (Geissmann et al., 2003). Immunologically, resident Ms are the first defenders of the innate immune system against pathogens through direct phagocytosis and/or bactericidal reactive nitrogen species (RNS) release (Stefater et al., 2011). Thereafter, inflammatory Ms, arising from patrolling Mos extravasating into the inflamed tissue, perpetuate the inflammation until the pathogen is eliminated (Stefater et al., 2011; Auffray et al., 2007). As part of the innate immune system, M responses were once thought to be non-specific, contrasting with the Ag-specific responses of the adaptive immune system. With the discovery of TLR expression on Ms, however, it became apparent that even the responses of the innate immune system were specific, at least to some degree, and complex. TLRs are encoded by genes homologous to the Drosophila Toll genes, which in mammals play a vital role in the immune response (Takeda and Akira, 2005). Essentially, Ms express a repertoire of TLRs, each recognizing specific motifs, known as pathogen-associated molecular patterns (PAMPs), on viruses, yeast, or bacteria thus allowing Ms to 'identify' the pathogen and respond accordingly (Kawai and Akira, 2011). For example, PAMPs such as lipopolysaccharide (LPS) on the intracellular bacteria, Salmonella, is recognized by TLR-4. This TLR, in turn, signals via the adapter proteins myeloid differentiation  8  primary response gene (88) (MyD88) and TIR-domain-containing adapter-inducing interferon-β (TRIF), to drive the production of inflammatory cytokines, such as IL-12, to skew the adaptive immune response towards a T H-1, cell-mediated response, which is poised to clear intracellular bacteria and viruses (Kawai and Akira, 2011). Thus, how a M responds during the initial stages of an immune response has a profound effect on the subsequent acquired immune response. Furthermore, aside from the immune receptor repertoire on a M, its activation or polarization state is also a vital determinant of how it directs subsequent immune responses (Gordon and Martinez, 2010; Mosser and Edwards, 2008).  1.5.1.1  Macrophage Polarization  While it has long been known that Ms can be 'classically' (M1) activated when exposed to TLR ligands and the TH-1 cytokine IFN-, it is only recently that Ms have been shown to possess an 'alternative' (M2) activation program in response to the TH-2 cytokines IL-4 or IL-13 (Stein et al., 1992; Mills, 2001). Far from being a homogeneous subset, it is now known that there is a spectrum of alternative, M2, activation, which directs non- or anti-inflammatory M programs (Mantovani et al., 2004; Mosser and Edwards, 2008). Specifically, IL-4-stimulated Ms are also known as M2a or wound healing Ms, immune-complex- plus TLR ligand- co-stimulated Ms are known as M2b, and Ms stimulated by deactivating factors (e.g., IL-10, TGF-) are known as M2c, or regulatory Ms (Mantovani et al., 2004; Mosser and Edwards, 2008) (Fig 1.1 A). When M1 activated, Ms become more cytotoxic by increasing the production of phagolysosomes, nitric oxide (NO), and pro-inflammatory cytokines (IL-12, tumor necrosis factor (TNF)-, etc.), while producing very little anti-inflammatory cytokines (IL-10, etc.) (Mosser and Edwards, 2008). M2 activated Ms, on the other hand, produce relatively low levels of NO and pro-inflammatory cytokines but higher levels of anti-inflammatory cytokines (Mantovani et al., 2004; Mosser and Edwards, 2008). Uniquely, murine M2a or wound healing Ms constitutively express the enzyme arginase-1 (Arg1), which competes with the NO-producing enzyme, inducible nitric  9  oxide synthase (iNOS or NOS-2), for L-arginine to limit NO production in Ms through the exhaustion of the finite L-arginine supply (Gordon and Martinez, 2010; Mills, 2001; Mosser and Edwards, 2008) (Fig 1.1 A). Because of their characteristic production of cytotoxic NO and pro-inflammatory cytokines, M1 Ms are thus ―killer‖ Ms, which are effective at killing infected cells, pathogens, and tumor cells directly, or by promoting a cell-mediated TH-1 response (Keller et al., 1990; Mills, 2001). M2 Ms are ―healer‖ Ms: their anti-inflammatory cytokine repertoire dampens the TH-1 response, and constitutive Arg1 expression reduces the available L-arginine for cytotoxic NO (Mills, 2001). Furthermore, Arg1 promotes wound healing by metabolizing L-arginine into ornithine, and ultimately proline and polyamines, which enhance collagen synthesis (fibrosis) and cell division, respectively (Minois et al., 2011; Mills, 2001). Interestingly, recent data have shown that while M1 Ms are recruited to sites of intracellular bacterial or viral infections, M2 Ms are not recruited but are, instead, a result of local resident M proliferation in response to IL-4 secreted by eosinophils or basophils recruited in response to the infection (Jenkins et al., 2011).  1.5.1.2  Macrophage Polarization and Cancer  The literature is replete with evidence of M association with different tumors in both mice and humans (Gordon and Martinez, 2010; Ruffell et al., 2011; Makitie et al., 2001; Steidl et al., 2010; Sugihara et al., 2009). In fact, it has been shown that tumor cells produce chemotatic factors, such as monocyte chemotatic protein-1 (MCP-1), to recruit Mos to become tumor-associated Ms (TAMs) (Sica et al., 2008a; Bottazzi et al., 1983). With the emergence of the M2 M activation paradigm, it has become clear that the phenotype of TAMs needs to be established to understand the role of TAMs in tumors. Because of their contrasting biochemical, killer and healer, properties, M1 and M2 Ms have opposing effects on the progression of cancer. For example, it is well established that Ms mediate tumor cell-killing via NO and TNF- (Keller et al., 1990; Buhtoiarov et al. 2007), both of which are abundantly produced by M1 Ms.  10  This makes them potent anti-tumor effectors. M2 Ms, on the other hand, are poor producers of NO and TNF-, and enhance tumor cell growth by expressing Arg1, as observed in murine Ms, which further suppresses NO production through substrate competition and, like in wound healing, promotes cell division via downstream polyamine production (Bernacki et al., 1995; Minois et al., 2011). In addition to direct biochemical repercussions, M1 and M2 Ms also have contrasting influences on the ATIR. M1 M-produced IL-12 not only skews towards an anti-tumor TH-1, cell-mediated adaptive response, it also stimulates the production of IFN-from NK cells, which increases T cell proliferation for a robust T cell response (Trinchieri and Sher, 2007). M2 Ms, on the other hand, produce IL10 to skew to a humoral TH-2 response, which is less effective at killing tumor cells (Ding et al., 1993; Fiorentino et al., 1991). As an autocrine-acting cytokine, IL-10 reduces MAg presentation, as well as NO and pro-inflammatory cytokine production through signal transducer and activation of transcription-(STAT)-3mediated inhibition of nuclear factor kappa-B (NF-B) activation, to limit the ability of Ms to initiate a CTL-mediated or M-mediated ATIR (Beissert et al., 1995). Furthermore, since L-arginine is required for optimal T cell proliferation, Arg1 expression will stunt the TA-specific CTL response via L-arginine depletion in the tumor microenvironment (Bansal and Ochoa 2003; Rauh et al., 2005; Sinha et al., 2005a; Popovic et al., 2007). TAMs, therefore, may be critical, if they are M1, or detrimental, if they are M2, to mounting an ATIR. In accordance with predictions, it was recently confirmed that TAMs typically correlate with poor prognosis or faster tumor growth, in both mice and humans, and are indeed M2 or M2-like Ms (Rauh et al., 2005; Tjiu et al., 2009; Kurahara et al., 2011). In experimental mouse models with impaired IL-4 signaling (STAT-6-/-), it has also been shown that there is more M1 M activation and an absence of M2 Ms and Arg1, and this promotes the rejection of syngeneic 4T1 mammary carcinomas as well as limits metastasis (Sinha et al., 2005b). Furthermore, it was also recently demonstrated that the recruitment of M2 Ms and the VEGF that these Ms secrete  11  mediate of both cancer growth and metastasis in mice (Qian et al., 2011; Lin et al., 2007). From this evidence implicating the M2 M in tumor growth, ameliorating or reversing the M2 phenotype may prove to be a viable immunotherapeutic option in targeting tumor-promoting cellular elements of the tumor microenvironment. Although we discussed M2 Ms as a single population for the sake of convenience, M2 Ms are a diverse population of cells classified into different subtypes, as mentioned above. Although taken together, this heterogeneous population promotes the effects on cancer cell survival/proliferation and immune system suppression as already discussed, it is now known that different subsets of M2 Ms carry out distinct aspects of tumor progression and immunosuppression (Mosser and Edwards, 2008; Martinez et al., 2008). A summary of the effects of each M2 M subset on cancer cells and the immune response is presented in Figure 1.1 B.  12  Figure 1.1 Model of the reversibility of M2 functional phenotypes. A) Consensus model of M activation, M1 and various M2 subsets, by different stimuli. B) Summary of the effects of various M subsets on Ms, T cells, and cancer cells. Solid lines represent a change in activation state and dotted lines represent the positive (arrows) or negative (blunted arrows) effects exerted by one cell type on another. This figure is a summary of the ideas and concepts presented in Martinez et al. (2008), and Mosser and Edwards (2008).  13  1.5.1.3  SHIP and M2 Macrophages  A cell-signaling pathway critical to M2-skewing is the phosphatidylinositol 3kinase (PI3K) pathway, which regulates cell growth and differentiation (Franke et al., 1997). When stimulated by growth factors, cell-surface growth factor receptors recruit PI3K to the plasma membrane where it converts phosphatidylinositol 4,5bisphosphate (PI-4,5-P2) into phosphatidylinositol 3,4,5-trisphosphate (PIP3), which in turn attracts and activates downstream kinases, such as protein kinase-B (Akt) (Franke et al., 1997). Under normal circumstances, the hemopoietic-specific SH2containing inositol 5'-phosphatase (SHIP) negatively regulates PI3K activation by hydrolyzing PIP3 to phosphatidylinositol 3,4-bisphosphate (PI-3,4-P2) to stop downstream kinase activation (Damen et al., 1998). We found in earlier studies that in the SHIP knockout (SHIP-/-) mouse, peritoneal and alveolar Ms exhibited a profound M2-skewing that is partially dependent on TGF- (Rauh et al., 2005). In keeping with the literature, ectopic tumors implanted into these M2-skewed SHIP-/- mice grew faster than ones implanted into their more M1-skewed SHIP+/+ littermates (Rauh et al., 2005). Furthermore, the increased tumor growth rate correlated with the infiltration of M2like TAMs into the tumors in SHIP-/- mice (Rauh et al., 2005).  1.6  Energy Metabolism and Glycolysis Like tumor-infiltrating immune cells, the surrounding milieu of the tumor is  crucial to its growth because it contains the substances (i.e., nutrients, biomolecules) that cancer cells must use to sustain rapid proliferation and a positive energy balance. Energy production and management are important to any cell, normal or malignant alike, and is especially critical in the case of the latter, since rapid division and biosynthesis puts a significant demand on cellular energy supplies (Fox et al., 2005; Buttgereit and Brand, 1995). All cellular processes (i.e., biosynthesis, maintenance, or cell-signaling) draw from a common supply of energy, the intracellular adenosine triphosphate (ATP) stores. The ATP supply, in turn, can be replenished through the catabolism of three principle energy producing substrates: glucose, amino acids, and fatty acids. These 14  substrates are broken down into carbon skeletons, which are oxidized to generate ATP (Nelson and Cox, 2000). Glucose is the primary substrate for energy production and, upon its import into cells via glucose transporters (GLUTs), is phosphorylated into glucose-6-phosphate to retain glucose inside the cell and to begin glycolysis (also called the EmbdenMeyerhof-Parnas pathway), which, through a series of enzymatic steps, catabolizes one glucose molecule into 2 pyruvate molecules for a net gain of 2 ATPs and 2 reduced nicotinamide adenine dinucleotide (NADH) molecules (Nelson and Cox, 2000). At this point, the pyruvate dehydrogenase complex (PDC) can combine each pyruvate with coenzyme A (CoA) to form a thioester acetyl-CoA, which enters the tricarboxylic acid (TCA) cycle (also called the citric acid or Krebs cycle) in the mitochondria to generate substrates, such as NADH and succinate; these serve as electron donors for the oxidative phosphorylation (OXPHOS) pathway, which uses the electron transport chain (ETC), and oxygen (O2) as a terminal electron acceptor to create a proton (H+) gradient for ATP production from ATP synthases, which are powered by the electrochemical energy of the H+ gradient (Holness and Sugden, 2003; Dudkina et al., 2010). An alternate fate for pyruvate is fermentation, where it is metabolized into lactate by lactate dehydrogenase (LDH), in an O2-independent process (Diaz-Ruiz et al., 2009). Glycolysis and fermentation will be herein referred to as fermentative glycolysis (FG). Lactate is an end product of FG and, unlike acetyl-CoA, it is not used for further ATP production (Diaz-Ruiz et al., 2009). The intracellular fate of glucose is summarized in Figure 1.2.  15  Figure 1.2 Intracellular fate of glucose. Thin solid arrows represent the movement of molecules from one compartment to another. Dotted black arrows represent glycolysis, the shared initial steps of OXPHOS and FG. Solid black arrows represent the OXPHOS pathway, and the solid gray arrow represents fermentation. FG is glycolysis and fermentation. Curved gray arrows represent the consumption and production of molecules with each step.  16  Amino acids and fatty acids can also be directly catabolized for energy but are not processed through FG, nor do they have FG equivalents. In the case of amino acids, their catabolism converges at -ketoglutarate, acetyl-CoA, and, through nonATP producing transamination reactions, pyruvate. Similarly, fatty acids are primarily catabolized for energy through -oxidation to form acetyl-CoA (Nelson and Cox, 2000). All of these energy producing substrates produced by non-glucose energy production molecules are metabolized for ATP exclusively by OXPHOS. In other words, a distinguishing feature unique to glucose among energy production molecules is that it can be a source of energy under anaerobic conditions, since FG generates ATP in an O2-independent manner. From an ATP standpoint, FG is inefficient, yielding a net gain of only 2 ATPs per glucose molecule, and leaving much potential energy in lactate, the waste product of FG. The TCA cycle and OXPHOS, on the other hand, are highly efficient, yielding 30-34 ATPs per glucose molecule, and the final products, oxaloacetate and carbon dioxide (CO2), are recombined with acetyl-CoA to re-enter the TCA cycle, and efficiently removed by RBCs, respectively (Nelson and Cox, 2000). As such, normal cells under normoxic conditions typically use OXPHOS, and use FG only when O2 becomes unavailable (Kroemer and Pouyssegur, 2008). 1.6.1  The Warburg Effect  The vast majority of cancer cells, unlike most normal cells, have a profound dependence on glucose for optimal growth and survival (Warburg et al., 1927; Vander Heidan et al., 2009). In fact, Otto Warburg showed more than 80 years ago that cancer cells generate the majority of their ATP through glucose catabolism via FG, as indicated by the accumulation of lactate in the tumor and tumor-efferent circulation (Warburg et al., 1927). Most interestingly, tumor cells do so even under normoxic conditions, a phenomenon that has become known as aerobic glycolysis (Warburg et al., 1927). While Warburg hypothesized that the 'Warburg effect', which describes the propensity of cancer cells to take up glucose and generate ATP through aerobic glycolysis, was due to irreversible mutations in mitochondrial enzymes necessary for OXPHOS, recent studies have shown that mitochondrial 17  defects are rare and the Warburg effect is, more often than not, reversible (Warburg, 1956; Fantin et al., 2006; Bragoszewski et al., 2008). Nevertheless, the ‗Warburg effect‘ describes the very real propensity of tumor cells for aerobic glycolysis and enhanced glucose uptake, and this is why an increased uptake of the glucose analog,  18  fluorodeoxyglucose (2-FDG), detected by positron emission tomography  (PET) is being used today to locate and follow the progression of most human tumors (Gambhir, 2002). As well, because of this propensity, aerobic glycolysis and the Warburg effect have long been considered potential therapeutic targets.  1.6.2  Aerobic Glycolysis and Cell Proliferation  Given the stark contrast in ATP-generating efficiency between FG and OXPHOS, the preferential use of aerobic glycolysis for rapid cell proliferation seems to be unlikely, given the harsh conditions in which cancers evolve would likely put substantial selective pressure on cancer cells to be efficient. While once thought to be a unique feature of cancer, transient aerobic glycolysis has since been observed in highly proliferative non-cancerous tissues, such as normal embryonic cells and clonally expanding T cells (Sitkovsky and Lukashev, 2005). Therefore, to explain aerobic glycolysis, the current consensus is that, despite the low ATP yield, FG gives an advantage to proliferating cells (Vander Heidan et al., 2009). A distinguishing feature of FG is that it has offshoots, which bypass the energy producing steps of this pathway to divert carbon towards the synthesis of important macromolecules. Specifically, the pentose phosphate shunt short circuits canonical glycolysis by using glucose-6-phosphate to generate both ribose and nicotinamide adenine dinucleotide phosphate (NADPH), the former, a component of nucleic acids, and the latter, a co-factor for fatty acid synthesis and an intermediate for generating glutathione, which protects cells from reactive oxygen species (ROS)-induced oxidative damage to ensure DNA replication fidelity and prevent DNA-damagedinduced apoptosis or senescence (Fico et al., 2004; Kroemer and Pouyssegur, 2008; Vander Heidan et al., 2009); pyruvate and acetyl-CoA can also be diverted away from energy production, to be converted to alanine for protein synthesis and citrate for fatty acid synthesis, respectively (Kroemer and Pouyssegur, 2008; Vander  18  Heidan et al., 2009). The prevailing hypothesis as to why rapidly proliferating cells use FG over OXPHOS, therefore, is that it provides them with macromolecules necessary for rapid cell division (Vander Heidan, et al., 2009). Another unique advantage of FG is that because glucose can diffuse farther than O2, it allows cells to survive and proliferate beyond the diffusion limit of O2. Since neoplasias often grow beyond this limit, FG but not OXPHOS allows them to generate both the building blocks and energy necessary for rapid proliferation in a hypoxic environment, provided that there is sufficient supplies of BG (Gatenby and Gillies, 2004). In fact, hypoxia-induced factor-1- (HIF-1) is a critical factor that is upregulated in hypoxia and responsible for the skewing to FG and for angiogenesis in tumors (Kim et al., 2006). Even after the development of tumor vasculature, the disorganized and chaotic nature of tumor blood vessels creates patchy and transient hypoxia, which potentially requires FG for the tumor to maintain its ATP levels (Minchinton and Tannock, 2006).  1.6.3  Lactate Metabolism and Cancer  A metabolic consequence of FG is the production of lactate during fermentation. In resting muscle and most other tissues, lactate production is low because pyruvate is typically consumed via OXPHOS, a higher-yield but slower energy-producing pathway (Berg, 2002). In rapidly contracting white skeletal muscle or highly proliferative tissues, FG takes over as the primary mode of ATP production to meet the demand for a rapid source of energy (Bauer et al., 2004; Berg, 2002). The exclusive use of FG as a means of ATP production creates a deficit in nicotinamide adenine dinucleotide (NAD+), a critical oxidizer during the energygenerating phase of FG, and the conversion of pyruvate to lactate becomes a necessary step to regenerate this critical co-factor (Feron, 2009). However, because lactate is acidic, specialized monocarboxylate transporters (MCTs) for lactate efflux are required to maintain intracellular pH (Ganapathy et al., 2009) (Fig 1.2). Because tumors rely heavily upon FG, they produce significant amounts of lactate, which is then exported, typically via increased levels of MCT-4, to the extracellular milieu (Pinheiro et al., 2008; Halestrap and Meredith, 2004). This  19  creates an acidic extracellular local environment, which is exacerbated by the failure of disorganized tumor vasculature to efficiently remove extracellular lactate (Minchinton and Tannock, 2006). This can reduce the pH of the tumor microenvironment from 7.4 to as low as 6.0 (Warburg et al., 1927; Dang et al., 2011; Minchinton and Tannock, 2006; Gillies et al., 2002). While the buildup of a waste product may seem like a hindrance, it is becoming clear that lactate buildup and lactate-mediated acidification of the extracellular milieu may be advantageous to the tumor. In fact, most successful tumors exhibit adaptations which increase the production of lactate (Feron, 2009). Specifically, both murine and human tumor cells have been shown to increase their expression of lactate exporters (i.e., MCT-1, MCT-3, MCT-4), and the enzyme LDH-5, comprised of 4 type-A LDH subunits (LDH-A), which favor the conversion of pyruvate to lactate (Zhuang et al., 2010; Serganova et al., 2011; Pinheiro et al., 2008). Moreover, both the oncoprotein c-Myc, and the cancer promoting transcription HIF1-, which skews metabolism towards FG and away from OXPHOS, induce the expression of this particular isoform of LDH and not other isoforms containing type-B LDH subunits (LDH-B), which favor the conversion of lactate to pyruvate (Gatensby and Gilles, 2004; Shim et al., 1997). Recent studies have shown that lactate benefits tumors by inhibiting ATIR and promoting metastasis (Feron, 2009). To inhibit the adaptive arm of ATIR, high lactate concentrations in the tumor microenvironment create a steep gradient for lactate efflux from normal cells, which renders tumor-infiltrating CTLs unable to export lactate, thus impairing their tumoricidal functions (Singer et al., 2011; Fischer et al., 2007). In terms of the innate immune response, lactate inhibits IL-12 production from DCs and down-regulates presentation molecules, such as CD1a, thus interfering with Ag presentation (Gottfried et al., 2006). In more developed tumors, the lactate-induced pH change in the tumor microenvironment has been shown to enhance the activity of certain extracellular proteases, such as matrix metalloproteinases (MMPs), to promote ECM breakdown and thus facilitate metastasis (Yin et al., 2009; Martinez-Zaguilan et al., 1996; Turner, 1979).  20  1.7  Glucose and Dietary Carbohydrate Metabolism Like normal cells, cancer cells rely on circulating BG as a primary glucose  source (Warburg et al., 1927). Glucose can enter the bloodstream through one, or a combination of three principal routes: from glycogen breakdown within the liver, gluconeogenesis from lactate and/or certain amino acids (e.g. alanine), and dietary absorption. While the first two routes allow an organism to maintain a constant BG supply between meals, dietary absorption is the only route that brings new glucose into the system. Therefore, dietary intake of glucose and complex carbohydrates (CHOs), the latter being broken down to glucose, is critical, because it determines the amount of glucose flowing into the system. When CHOs are ingested, they are broken down mechanically, by mastication, and enzymatically, by salivary amylases, in the oral cavity. The majority of absorption occurs in the small intestine, through the gut epithelium (enterocytes) and then into the bloodstream, after pancreatic amylases and intestinal epithelium-bound disaccharases (e.g. maltase, sucrase, lactase) have broken complex CHOs (i.e., starch, oligosaccharides) into six-carbon sugar monomers (e.g., glucose, galactose, fructose). After a meal rich in CHOs, BG increases rapidly, and this post-prandial BG spike stimulates the release of the hormone insulin (Ins) to activate glucose uptake (Henquin et al., 2009). Ins is a peptide hormone produced in pancreatic -cells, and triggered to be released from storage granules when BG is abundant (Henquin et al., 2009). Ins then circulates in the bloodstream and binds primarily to the ubiquitously expressed insulin receptor (IR) to cause the autophosphorylation of this tyrosine kinase receptor (TKR), ultimately triggering the phosphorylation of insulin regulated substrate-1 (IRS-1) (Godsland, 2010). IRS-1, in turn, initiates downstream responses by recruiting and activating PI3K and extracellular signal-regulated kinase (ERK), via Ras in the case of the latter (Siddle, 2011). IRS-1 activated PI3K is a critical step in the Ins signaling pathway. It is responsible for the increase in a cell's glucose influx capacity by stimulating the transport of the glucose importer GLUT-4 from storage vesicles to the plasma membrane as well as the de novo synthesis of GLUT-1 (Taha et al., 1995; Hatakeyama and Kanzaki, 2011). Imported glucose can  21  be consumed for energy (i.e. muscle contraction) or to generate biosynthetic molecules for proliferation, or it can be polymerized and stored as glycogen in the liver or converted to fat for storage in adipocytes. PI3K-activated Akt phosphorylates and inactivates glycogen synthase kinase-3 to increase glycogen synthesis in an Ins-stimulated cell (Taha and Klip, 1999).  1.7.1  Blood Glucose, Insulin and Cancer  Although glucose absorption and Ins regulation of BG are fundamental mechanisms of glucose homeostasis, high levels of both BG and Ins have been correlated with cancer (Pollak, 2008). In fact, in a recent meta-analysis of six individual studies, which examined leukemias and solid tumors (i.e., cancers of the colon, lung, prostate, ovary, and breast), the authors found a statistically significant trend positively correlating BG to both cancer incidence and fatality (Stocks et al., 2009). Ins and IR expression have also been shown to correlate to poor prognosis in non-small cell lung cancer (NSCLC), and the use of therapeutic Ins or Ins analogues appears to increase cancer risk (Yang et al., 2004; Hemkens et al., 2009; Kim et al., 2009). Interestingly, type-2 diabetes mellitus, which is characterized by Ins resistance and abnormally high fasting BG and Ins, is also associated with higher cancer risk (Braun et al., 2011; Yuhara et al., 2011). Molecular studies have shown that the downstream effects of high IR levels seem to be key to the tumor-promoting effects of BG and Ins, and may explain their correlation with disease (Pollak, 2008, Osborne et al., 1976). Specifically, Ins acts as a mitogen and has been shown in vitro to increase the proliferation of tumor cell lines (Pollak, 2008, Osborne et al., 1976). The proliferative effect of the IR is mediated through PI3K and mitogen-activated protein kinase (MAPK) cascades: Akt, which is downstream of PI3K, phosphorylates mammalian target of rapamycin (mTOR) and Bcl2 agonist of cell death (BAD), to facilitate cell proliferation and cell survival, respectively (Siddle, 2011); likewise, the activation of the ERK cascade increases cell proliferation by promoting G1/S transitions in the cell cycle (Rivard et al., 1999; Aliaga et al., 1999). Furthermore, insulin-like growth factor 1 (IGF-1), which is secreted from the liver when Ins levels are high, also signals through IRS-1,  22  either via IR or the related IGF1 receptor, and has been implicated in prostate cancer progression in experimental models (Pollak, 2008; Saikali et al., 2008; Venkateswaran et al., 2007). High BG, by itself, has also been implicated in various cancers since it leads to higher levels of glycation (Rojas et al., 2010). This non-enzymatic covalent attachment of sugars to proteins, peptides and amino acids leads to advanced glycation end products (AGE) on the surface, for example, of the endothelial lining of blood vessels. These changed cell-surface proteins are recognized by immune cells via receptors for advanced glycation end products (RAGE) and trigger inflammation, which can increase atherosclerosis and lead to chronic inflammation that predisposes to many cancers (Rojas et al., 2010).  1.8  Manipulating the Tumor Microenvironment The tumor microenvironment is a complex multi-factorial system comprised of  cellular and non-cellular components that can have a dramatic effect on cancer progression. Cancer cells themselves seem to be selected for adaptations, such as cytokine secretion (i.e., M-CSF, IL-10) and aerobic glycolysis, which skew elements of the microenvironment to subvert ATIR and augment their growth. In terms of cellular components of the tumor microenvironment, the M2 Ms, through their anti-inflammatory cytokine repertoire and wound-healing biochemical properties have been shown to both benefit tumor growth as well as inhibit host ATIR. However, the tumoricidal potential of M1 Ms demonstrates that they can play an opposite role. Given also the pervasiveness of M infiltration in solid tumors, it stands to reason that the reversal of the M2 phenotype or the induction of the M1 phenotype may be viable therapeutic goals. Glucose metabolism, via FG, not only provides tumor cells with cellular building blocks (i.e., ribose) and energy, it produces lactate to inhibit the ATIR. Furthermore, the very presence of high BG, which leads to the secretion of Ins, may be yet another benefit to tumor growth, since Ins enhances cell proliferation and survival. However, the continued use of FG requires large supplies of BG, because of the inherent inefficiencies of ATP production via FG and the rapid consumption of 23  glucose for biosynthesis. Reducing the BG supply in the tumor microenvironment, therefore, may limit the tumor's ability to exploit FG and Ins signaling to augment its own growth. In this thesis, we assessed the feasibility of altering or reversing tumor adaptations  that  manipulate  cellular  and  non-cellular  components  of  the  microenvironment to augment tumor growth. Specifically, we have investigated M plasticity to see if polarized M2 Ms can be skewed to a more immune-stimulatory M1 phenotype, and the pathways and cell-types involved in M2 M skewing in a genetically predisposed M2 M-skewed mouse model. Furthermore, of the noncellular components of the tumor microenvironment, we investigated the feasibility of reducing BG, Ins, and other tumor-promoting elements of glucose metabolism through the limitation of dietary CHO intake in mice.  24  2. Materials and Methods 2.1  Mice SHIP+/+ and SHIP-/- F2 mice on a mixed C57Bl/6 x 129Sv background were  derived as described (Helgason et al., 1998), and similar to RAG2M mice, were bred and housed in specific pathogen-free facilities at the British Columbia Cancer Research Centre (Vancouver, British Columbia). For the diet studies, the C3H/HeN mice were obtained from Simonsen Laboratories (Gilroy, CA, USA) or bred in-house. The B6.Cg-KitW-sh/HNihrJaeBsmJ (Wsh/Wsh) mice were purchased from Jackson Laboratory (Bar Harbor, ME, USA), and maintained at The Biomedical Research Centre at the University of British Columbia (Vancouver, BC, Canada). Wsh/Wsh mice were crossed with B6-congenic inducible SHIP-/- mice (i.e., SHIPFL/FL.Mx1Cre mice [Wang et al., 2002]) to obtain the mast cell (MC)-free, inducible SHIP-/- mice (Wsh.SHIPFL/FL.MxCre1+), which were used in our experiments. For our spontaneous cancer model studies, female NOP mice, which express the Trp53 minigene and a HER2/Neu-Ovalbumin fusion protein under the control of the mouse mammary tumor virus (MMTV) promoter (Wall et al., 2007) were bred and housed in the Deeley Research Centre (Victoria, BC, Canada). OT-II mice, which contain only ovalbumin-specific CD4+ T cells (Barnden et al., 1998), were initially bought from Jackson Laboratory and bred in-house.  2.2  Cell Cultures Unless otherwise stated, all cell cultures were incubated at 37°C in a 20% O2  and 5% CO2 environment using a Forma Scientific (Markham, ON, Canada) CO2 water jacketed, HEPA filtered incubator.  2.2.1  Macrophage Cultures  Using a 10 mL syringe and 26 gauge needle, the femurs and tibias from one mouse were flushed with 30 mL of peritoneal M medium (PM), which is Iscove‘s Modified Dulbecco‘s medium (IMDM) (StemCell Technologies; Vancouver, BC, Canada) + 10% fetal calf serum (FCS) (Hyclone; South Logan, UT, USA) + 25  0.00125% monothioglycerol (MTG) (Sigma-Aldrich) + 100 U/mL penicillin + streptomycin (P/S) (StemCell Technologies), through a cell strainer (BD Falcon) with 70 m or 100 m pores to make a single cell BM suspension into a 50 mL conical tube (BD Falcon; Mississauga, ON, Canada). The cell strainers were washed with an additional 5 mL of PM, twice, making the total final volume 40 mL. To adherencedeplete the mature Ms, the BM suspension was cultured in a 75 cm2 tissue culture flask (Nunc; Rochester, NY, USA) for 3 - 18 h. In 175 cm2 flasks (Nunc), nonadherent BM progenitors were cultured in bone marrow-derived M Medium (BMM), which is PM + 5 or 10 ng/mL recombinant mouse (rm) M-CSF (StemCell Technologies), at 5 x 105 – 1 x 106 cells/mL on day 0. After three days (day 3), the medium was replaced with fresh BMM, and the old medium was centrifuged and the non-adherent cells added back into their respective cultures. On day 7, the medium from each flask was aspirated and replaced with fresh BMM. Ms were typically harvested and used on day 10, but were usable until day 16. For all mature bone marrow-derived Ms (BMMs) stimulation experiments, PM was used. For BMM generation from BM, BMM was used.  2.2.2  Bone Marrow Environment Assay  The bone marrow environment assay (BMENV) was used to test the various candidate factors and cytokines which skew differentiating BMMs to M2. As described in Rauh et al. (2005), we harvested BM as indicated above and cultured the cells in BMM, after adherence depletion in PM, at a concentration of 0.5 x 106 cells per mL for SHIP-/- BM and 0.5 or 1.0 x 106 cells/mL for SHIP+/+ BM. SHIP+/+ BM was sometimes seeded at a higher cell concentration to offset the increased growth rate of SHIP-/- BM cells. Similar yields of mature BMMs could thus be obtained from SHIP+/+ and SHIP-/- BM cultures. Cells were harvested 6 days after seeding. M2-skewing factors were either added immediately after BM seeding or as late as 24 h after seeding. Transwell plates (0.4 m membrane pore size) used were purchased from Corning (Lowell, MA, USA).  26  2.2.3  Macrophage Harvests  Ms from bulk culture flasks or in wells after stimulation were harvested using cell dissociation buffer (CDB) (Life Technologies Gibco; Burlington, ON, Canada), which is a non-enzymatic buffer used to extract adherent Ms from tissue culture flasks or plates. Before CDB application, cell-culture medium was aspirated and adherent cells were washed with PM to remove non-adherent cells. After wash medium was aspirated, CDB was added at 5 – 10 mL for tissue culture flasks or 100 L – 1 mL for well-cultured Ms, depending on the surface area. After 3 – 5 min of incubation at 23C, flasks were agitated by light tapping against an open palm, and wells were agitated through repeated aspiration and expulsion of cell suspensions into the well. The CDB cell suspension was then transferred using an electronic pipette (BD Falcon) to polystyrene conical tubes (BD Falcon). The tissue culture flask was then washed with PM and the wash combined with the CDB cell suspension, which was centrifuged at 1200 rpm in an X-12R Allegra centrifuge (Beckman Coulter; Mississauga, ON, Canada) for 6 – 10 min to pellet the cells, which were then resuspended, after supernatant aspiration, with the desired medium at the desired volume for counting, plating, or lysis.  2.2.4  Cell Lysis  Cells were harvested as described above and the cell pellets suspended in 500 – 1000 L 4C phosphate buffered saline (PBS) in 1.5 mL polystyrene microcentrifuge tubes on ice. For lysis using T:T, which is a 1:1 mixture of 0.1% Triton X-100 (Sigma-Aldrich) and 25 mM Tris (Fisher) in de-ionized distilled water (ddH2O) or 0.5% NP40 (nonyl phenoxylpolyethoxylethanol) (Calbiochem; San Diego, CA, USA), cells were centrifuged at 2000 – 3500 rpm in a Heraeus Biofuge pico microcentrifuge (Buckinghamshire, England) for 7 – 10 min to pellet the cells. Supernatants were aspirated and pellets resuspended in the desired volume of T:T or NP40. For lysis with 1x sodium dodecyl sulfate (SDS) buffer, a 4x stock of the SDS buffer, which is 4.6 mL 20% SDS (BioRad; Hercules, CA, USA) + 4 mL glycerol (Sigma-Aldrich) + 2 mL -mercaptoethanol (Sigma-Aldrich), was first diluted to 1x in 27  phosphate solublization buffer (PSB), which is 50 mM HEPES (Sigma-Aldrich) + 100 mM NaF (Sigma-Aldrich) + 10 mM Na4P2O7 (Sigma-Aldrich) + 2 mM Na3VO4 (Sigma-Aldrich) + 2 mM EDTA (USB; Cleveland, OH, USA) + 2 mM NaMoO4 (Sigma-Aldrich), and kept at 23C. Then, 1/5 of the PBS cell suspensions were transferred to a fresh set of microcentrifuge tubes, and both sets of tubes were centrifuged at 2000 – 3500 rpm in a microcentrifuge (Heraeus) for 7 – 10 min at 4°C. The set of tubes containing cell-pellets of 1/5 of the original suspension were suspended in T:T or NP40 for protein quantification using the Bradford and BCA assays, respectively. The other set of tubes containing cell-pellets of 4/5 of the original cell suspension were suspended in 1x SDS buffer and heated at 100°C for 4 min. Cell pellets from the two sets of tubes were suspended in proportional volumes, such that the protein concentration from the protein quantification assays were reflective of the protein concentration in the tubes suspended in SDS buffer. The SDS buffer lysates were passed through a 26 gauge needle to shear the DNA to reduce viscosity. In addition to the above procedure, cells in tissue culture wells for BMENV assays were harvested by washing with PBS after culture medium aspiration, and T:T was added directly into the wells at desired volumes. After incubation at 23C for 10 min and gentle scraping of the well surface to thoroughly remove cellular material, these T:T lysates were transferred into microcentrifuge tubes for - 20C storage.  2.2.5  BM-Derived Basophils  To derive bone marrow-derived basophils (BMBs) BM from each mouse was flushed in a similar manner as used for M bulk cultures, using HF, which is Hank's buffered salt solution (HBSS) (StemCell Technologies) + 2% FCS in place of PM. The BM suspension was then centrifuged at 1200 rpm in a X-12R Allegra centrifuge (Beckman Coulter) for 7 - 10 min and the cells resuspended in 1 mL HF after the supernatant was aspirated. RBCs were lysed by adding 4 mL NH4Cl (StemCell Technologies) to the BM suspension and incubating on ice for 8 min. Twenty-five mL of HF was added into the BM suspension after this incubation. Cells were 28  centrifuged at 300-500 x g and washed with basophil starve medium (BSM) (i.e., IMDM + 20% FCS + 100 U/mL P/S). Washed BM was cultured in 50 mL basophil complete medium (BCM), which is BSM + 10 ng/mL rmIL-3 (StemCell Technologies), in a 175 cm2 tissue culture flask at a concentration of 1x105 – 5x105 cells/mL on day 0. Non-adherent cells were harvested and cultured in a new 75 cm 2 flask with fresh BCM on day 7. Mature basophils were enriched using EasySep (StemCell Technologies), which is a magnetic bead bound Ab-linked cell-enrichment procedure, to positively select for DX5+ cells before use on day 9-12. Basophil function was assessed at a cell concentration of 2 x 10 4 cells/100 L in PM, unless otherwise stated.  2.2.6  BMB and BMM Co-cultures  Unless otherwise stated, 2 x 104 DX5+ BMBs derived and enriched as described above were cultured with 2.5 x 105 BMMs in 500 L PM in 24-well tissue culture plates (BD Falcon). Where applicable, rmIL-4 was added (0.5, 1.0, 5.0, or 10.0 ng/mL) for 72 h. After stimulation, wells were washed thoroughly to ensure the removal of non-adherent BMBs. BMMs were lifted off their wells with CDB and subjected to SDS lysis and/or protein quantification with the Bradford or BCA assay.  2.2.7  Peritoneal Ms  PMs were obtained via peritoneal flushes and adherence selection. Briefly, 5 mL PM was injected into and aspirated out of the peritoneal cavity of a mouse using a 5 mL or 10 mL disposable syringe and 22 gauge needle. This procedure was repeated 2 additional times. Ms were counted with a hemocytometer, discriminating them from other cells by size and morphology. To adherence select for PMs, cells were plated at 0.5 x 106 cells/mL or 1.0 x 106 cells/mL for 3 h – 16 h in PM, after which the wells were washed twice with PM to get rid of the nonadherent cells, leaving adherent Ms.  29  2.3  Reagents, Inhibitors, Cytokines The cytokines used to stimulate Ms, including recombinant mouse rmIL-4,  rmIL-10, recombinant human (rh) TGF-1 (rhTGF-1), rmIFN-γ and rmIL-3 were purchased from StemCell Technologies. Mouse plasma (MP) was obtained primarily from SHIP+/+ or SHIP+/- mice via cardiac puncture and centrifugation of the blood at 3500 rpm for 7 min in Biofuge pico microcentrifuge (Heraeus). Collected MP was stored at -20C. Commercially available MP (Innovative Research; Novi, MI, USA) was also used. Human sera and plasma were obtained from Sigma-Aldrich. All blood products were filter sterilized with 0.22 m filters (Millipore; Billerica, MA, USA) before use, with the exception of sheep RBCs (Lampire Biological Products; Pipersville, PA, USA). Sera and plasma from mice and humans were all used as a blood-factor-induced M2 control in BMENV assays because there was no qualitative difference between their M2-skewing behavior, as previously reported (Rauh et al., 2005). Prostaglandin (PG) E2 (PGE2) was purchased from Sigma-Aldrich. With regard to COX-2 inhibitors, Celebrex was purchased from LC Laboratories (Woburn, MA, USA) or from Pfizer (Kirkland, Quebec, Canada), and SC-58125 was purchased from Cayman Chemical. The TGF- signaling inhibitor SB-505124 was from Sigma-Aldrich.  2.4  Quantitative Assays All quantitative assays described below were colorimetric, and the absorbances  of the reactions were measured using an ELX808 Ultra Microplate Reader (BioTek, Winooski, VT, USA) at the indicated wavelengths.  2.4.1  Arginase Assay  Cultured cells were washed with PBS to deplete the protein present in the culture media before lysing with T:T. Using 100 L of cell lysate, or a dilution of the lysate up to 100 L with T:T, arginase activity was determined by a colorimetric enzyme assay outlined in Morrison and Correll (2002). Briefly, each sample was heat-activated at 55C for 10 min with 10 L of 10 mM MnCl2. Then, 100 L 0.5 M Larginine (pH 9.8) (Sigma-Aldrich) was added to each sample and incubated at 37C 30  for 1 h. After this arginine hydrolysis step, the reaction was stopped using 800 L of a 1:3:7 H2SO4:H3PO4:H2O (Fisher). The production of urea was determined by adding 40 L 9% (w/v in absolute ethanol [EtOH]) -isonitrosopropiophenone (Sigma-Aldrich) and heating at 100C for 30 min. Urea concentration was determined by reading the absorbance at 550 nm and comparing it to a urea standard curve. Arginase activity was expressed as g urea/g protein/h. Protein concentration was determined using the Bradford Assay. For arginase activity quantification in live cells, culture medium was aspirated and wells were washed with HBSS. Equal volumes of HBSS containing 0.5 M Larginine was added to the wells for 24 - 48 h. Urea in these samples was quantified as stated above.  2.4.2  Bradford Assay  For the Bradford assay (BioRad), cells were lysed in T:T and diluted in PBS to give a lysate to PBS ratio of 1:3. A protein standard curve was generated using bovine serum albumin (BSA) in T:T, serially diluted by half in PBS from 200 g/mL to 3.125 g/mL for a total of 7 concentrations. PBS alone served as a 0 g/mL control. All standards and samples were assayed in a 96 well non-tissue culture plate (BD Falcon) at a final volume of 20 L, and 200 L of assay reagent, diluted and filtered according to BioRad‘s instructions in ddH2O, was then added for 5 min at 23C. Absorbance was measured at 595 nm.  2.4.3  BCA Protein Assay  For BCA assays (Thermo Scientific Pierce; Rockford, IL, USA), cells were lysed in 0.5% NP40 lysis buffer. Cell lysates were assayed in a 96 well polystyrene, nontissue culture plate (BD Falcon). Samples were serially diluted by half in PBS, up to 3 times, in a final volume of 10 L and all the dilutions were assayed. Assay reagent was prepared as per the kit instructions, using 190 L of reagent for each assay well. The assay reactions were incubated either at 37°C for 60 min, or at 55°C for 40 min, for lysates of many or few cells, respectively. Protein standards, using BSA in  31  PBS, were serially diluted by half in PBS starting from 2000 g/mL to 31.25 g/mL for a total of 7 concentrations with the addition of a PBS only control to give a 0 g/mL point. Absorbance was measured at 595 nm.  2.4.4  M-Produced Cytokines  Levels of cytokines produced in culture were measured using cell-culture supernatants harvested from cell suspensions after centrifugation, or directly from M cultures, where cells were adherent. Collected supernatants were stored for no more than 2 months at -20°C before analysis. Cytokines were quantified using enzyme-linked immunosorbent assays (ELISAs), as per the manufacturers' instructions. Mouse-specific IL-4, IL-12p70, IL-10, and TNF- ELISAs were purchased from BD Biosciences (Mississauga, ON, Canada). Absorbance was measured at 450 nm. Similarly, supernatant NO was assayed using the Griess Assay at 23C (Kleinbongard et al., 2002). In brief, 50 L 1% (w/v) sulfanilamide (Sigma-Aldrich) + 2.5% H3PO4 (85%) (Fisher) in ddH2O was added to 50 L culture supernatants in a 96 well plate polystyrene plate. Immediately after, 50 L of 0.1% (w/v) naphthylethylenediamine diHCl + H3PO4 (85%) 2.5% in ddH2O was added. Reactions were incubated for 5 min at 23C in the dark. The concentration of NO2, the stable reaction product of secreted NO, was proportional to the absorbance at 550 nm.  2.4.5  Measurement of Blood Glucose, Insulin, and Lactate  BG was measured via tail vein by using LifeScan test strips (LifeScan; Burnaby, BC, Canada) and OneTouch Ultraglucose meter (LifeScan; Burnaby, BC, Canada). Insulin and lactate levels were determined by ELISA (Mercodia; Winston Salem, NC, USA) and lactate assay kits (BioVision; Mountain View, CA, USA), respectively, using plasma from CO2-euthanized mice.  32  2.4.6  T Cell Proliferation Assay  Ag-specific T cell proliferation assays were performed to assess the Ag presentation capacity of various M subtypes. Ms were plated in a 96-well plate at indicated cell numbers in PM and activated with 50 - 100 ng/mL LPS + 0.5 - 1.0 g/mL OVA peptide (amino acid 323-339) (GenScript; Piscataway, NJ, USA). Splenic T cells were extracted from the spleens of C57/B6 OT-II transgenic mice (bred in-house) using an EasySep kit (StemCell Technologies), which enriches for CD4+ T cells via negative selection. Enriched splenic CD4+ T cells were then added to the Mcontaining wells. Cells were co-cultured for 54 h, after which 1 Ci (2 Ci/mM; PerkinElmer) of Tritiated-thymidine ([3H]thymidine) was added at to each well. Cells were co-cultured for another 18 h before harvesting with a 96-well harvester (Molecular Devices; Sunnyvale, CA, USA) onto a filtermat (PerkinElmer). The T cell incorporation of [3H]thymidine, which is directly proportional to T cell proliferation, was analyzed using a betaplate liquid scintillation counter (Wallac, Waltham, MA, USA), which records radioactive events as counts per minute (cpm).  2.5  SDS-PAGE and Western Blotting  2.5.1  Polyacrylamide Gels  Resolution gel buffer concentrate (4x), which is 1.5 M Tris + 0.4% SDS + 12 mL 12 N HCl in a final volume of 500 mL in ddH2O, was diluted in ddH2O and 40% acrylamide solution (BioRad) to a 1x concentration, such that it contained the desired percentage of acrylamide, typically 10% or 12.5%. To speed up polymerization, 10% APS (BioRad) in ddH2O and Tetramethylethylenediamine (TEMED) (BioRad) were added to the resolution gel mixture at prescribed concentrations (Sambrook and Russell, 2001). For a large gel, 40 mL of the resolution gel solution was poured into a gasket-sealed gel-setting apparatus between glass plates (BioRad) separated by spacers for a 1.5 mm thick gel. For a small gel, 7.5 mL of the same solution was poured into a smaller gel-setting apparatus with two glass plates (BioRad), one of which contained built-in spacers for 33  a 1.5 mm thick gel, set upon a gasket-stopped stand (BioRad). Isopropanol was carefully layered on top of the gel solution after pouring to ensure the formation of crisp top edge. The gel was allowed to set for at least 30 min, after which the isopropanol was poured off, and the gel was washed thoroughly in the apparatus with ddH2O. For the stacking gel, stacking gel buffer concentrate (4x) (i.e.,0.5 M Tris + 0.4% SDS + 7.5 mL 12 N HCl) was diluted in ddH2O and 40% acrylamide solution (BioRad) to a 1x concentration containing 5% acrylamide. To speed up polymerization, 10% APS (BioRad) and TEMED (BioRad) were added to the stacking gel mixture at prescribed concentrations (Sambrook and Russell, 2001). For large and small gels, 15 mL and 2.5 mL of this solution was poured on top of the set resolution gel, respectively. Plastic combs (1.5 mm thick) (Biorad) were placed on top of the solution before it hardened to make sample wells. After the stacking gel was allowed to polymerize for at least 30 min, or up to 16 h at 4°C, the comb was taken off, and the wells were washed thoroughly with ddH2O. 2.5.2  Electrophoresis and Immunoblotting  Small gels were subjected to electrophoresis at 100 volts for 1.5 - 2.0 h, and large gels for 16 - 18 h. Proteins were transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore) for 70 - 90 min (small gels) or 180 min (large gels). Membranes were incubated in PSB + 5% BSA (Fisher Scientific; Ottawa, ON, Canada) + 0.05% NaN3 (Sigma-Aldrich) for 1 h at 23C or 16 - 18 h at 4°C, then in primary Ab for 16 - 18 h at 4°C. After this incubation, the blots were washed 5x, 5 min per wash, with Tris-buffered saline (TBS), which is 0.05% Tween-20 (SigmaAldrich) + 160 g NaCl + 4 g KCl + 60 g Tris + 37.5 mL HCL in 1 L of ddH 2O, and incubated at 23C for 45 min in horseradish peroxidase-conjugated secondary Ab solution. Blots were washed in a similar fashion as above with TBS before visualization. Western Lightning Plus enhanced chemiluminesence substrate (PerkinElmer; Waltham, MA, USA) and Kodak X-Omat film (PerkinElmer) were used to visualize proteins.  34  2.6  Flow Cytometry Ms harvested with CDB, as indicated above, were washed with HF + 0.02%  NaN3 (HFN) and resuspended at a concentration of 1 x10 5 - 1 x 106 cells/100 L HFN in a round bottom 5 mL polystyrene tube (BD Falcon). The Fc R-specific 2.4g2 (BD Biosciences) monoclonal Ab was added for 15 min at 4°C to each sample to block IgG receptors on cells. Fluorochrome-conjugated Abs were then added at their recommended concentrations and incubated at 4°C for 15 min. Cells were then washed at least twice to remove the unbound Ab, before being resuspended in HFN at 300 L. Flow cytometry was carried out on these samples using FACSCalibur machines (BD Bioscience). Propidium iodide (PI) (BioRad) was incorporated before the final wash at 1 g/mL in HFN. Cells were gated for PI-exclusion to exclude dead cells.  2.7  Antibodies 2.7.1  Western Blotting  The following primary mouse-specific mouse monoclonal Abs were used: Arg1 (BD Biosciences), -glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (Fitzgerald; Concord, MA, USA ) as a loading control, -phosphorylated signal transducer and activator of transcription-1 (pSTAT-1) (Y701) (Zymed Laboratories; San Fransisco, CA, USA), and -phosphorylated signal transducer and activator of transcription-6 (pSTAT-6) (Y641) (Abcam; Cambridge, MA, USA). The following primary mouse-specific rabbit polyclonal Abs were used: -iNOS (Santa Cruz; Santa Cruz, CA, USA), -growth factor receptor-bound protein-2 (Grb2) (Santa Cruz) as a loading control, -M-CSFR (c-fms) (Santa Cruz), -SHIP (Santa Cruz), -src homology 2 domain-containing-transforming protein C1 (SHC) (Transduction Laboratories; Mississauga, ON, Canada) as a loading control, -cyclooxygenase-2 (COX-2) (Cayman Chemical; Ann Arbor, MI, USA), and -Ym1 (StemCell Technologies).  35  2.7.2  Flow Cytometry and Cell Enrichment  For flow cytometric analyses, the recommended Ab/cell ratio was used, as outlined in the specific data sheets. The following fluorochrome-conjugated Abs were used: Fluorescein isothiocynate (FITC)-conjugated -CD204 (Serotec; Raleigh, NC, USA), R-Phycoerythrin (PE)-conjugated F4/80 (Caltag Labs; Burlingame, CA, USA), Allophycocyanin (APhC)-conjugated -CD206 (Serotec), PE-conjugated CD11c (StemCell Technologies), and APhC-conjugated DX5 (BioLegend; San Diego, CA, USA). Mouse-specific -CD16/CD32, the 2.4g2 clone (BD Biosciences), was used to prevent binding of Abs via their Fc portions to IgG receptors on Ms, as mentioned earlier.  2.7.3  Functional Assays  The following Abs were used with Ms during functional or M2-skewing assays, in vitro and in vivo: neutralizing -IL-4 (R&D Systems; Burlington, ON, Canada), -FcRI (MAR-1) Ab (eBioscience; San Diego, CA, USA) neutralizing -IL3 (R&D Systems), reagent grade mouse IgG (Sigma-Aldrich), 2.4g2 (BD Biosciences), -Rat immunoglobulin (Ig)-G (Jackson ImmunoResearch; West Grove, PA, USA) and -sheep RBC Ab (Sigma-Aldrich), -CD3 (eBioscience). For immune complexes, sheep RBCs were incubated at 1 x 108 cells/mL in 1:500 -sheep RBC Ab (v/v) for 40 min in HBSS at 23C complexes were washed twice with HBSS before addition into cell cultures.  2.8  IgE and IgG Depletion For the depletion of IgE, frozen MP collected from cardiac punctures was  warmed to 23C and incubated with -mouse IgE (Biolegend) coupled to Protein G Agarose beads (0.5 mg to 100 L beads) for 1 h at 23C on a nutator. The suspension was then centrifuged at 13,200 rpm for 1 min in a Biofuge pico microcentrifuge (Heraeus). Supernatants were transferred to a fresh tube and frozen until use.  36  Similarly, for IgG depletion, thawed MP was incubated with Protein G Agarose beads (Pierce) which were pre-washed with PBS. At a ratio of 1:1, by volume, the MP was incubated on a nutator for 1 h at 23C and then the mixture was centrifuged at 13,200 rpm for 1 min as above, and the supernatant transferred to a fresh tube and frozen until use. To quantify Ig‘s in MP we used ELISAs for IgG (eBioscience) and IgE (BioLegend).  2.9  Depletion of Basophils in vivo and IL-4 Neutralization Before  we  depleted  basophils,  we  induced  SHIP  deletion  in  the  Wsh.SHIPFL/FL.Mx1Cre+ with 200 g/mouse intraperitoneal (IP) injections of polyinosinic-polycytidylic acid (poly I:C), which we purchased from Sigma-Aldrich. We injected a total of 4 doses, once every 48 h. Wsh.SHIP FL/FL.Mx1Cre- were similarly injected as a control. Next, as per Charles et al. (2009), basophils were depleted in Wsh.SHIPFL/FL.Mx1Cre+ and Wsh.SHIPFL/FL.Mx1Cre- mice with 10 IP injections of either MAR-1 (10 g/mouse) or an isotype control Ab (10 g/mouse), one injection every 3 days. After this regimen and a 3 day rest, these mice were injected with 50 g/mL of either MAR-1 or control Ab intravenously (IV), for 3 consecutive days. Mice were euthanized for analysis 3 days after the final IV injection. For IL-4 neutralization studies, -IL-4 neutralizing Ab or isotype control Ab (R&D systems), were injected at 500 g/mouse every 3 days for 2 weeks. Mice were euthanized 3 days after the last injection for analysis.  2.10 In vivo Tumor Cell Studies Five- to 8-week-old C3H/HeN and Rag2M mice were housed 2 - 4 mice/cage in high-top Allentown cages on static racks and bedding was changed twice/week. Unless otherwise stated, 2 × 105 murine squamous cell carcinoma VII (SCCVII) cells (from James W. Evans, Threshold Pharmaceuticals, South San Francisco, CA, USA) were cultured in vitro in Roswell Park Memorial Institute medium 1640 (RPMI) (StemCell Technologies) + 10% FCS + 50 U/mL penicillin and 50 mg/mL streptomycin, were injected subcutaneously (SC) into the backs of shaved C3H/HeN 37  mice. Similarly, 8 × 106 human colorectal carcinoma (HCT-116) cells (ATCC; Manassas, VA, USA) were injected into Rag2M mice. Tumors were measured 2 to 3 times per week by using manual calipers, and their volumes were determined by the formula—(Length × Width × Height) × π/6—except for the Rag2M study, where the following formula was used—(Length × Width × Width)/2. Female NOP mice were put on Western (5058) or 15% CHO diets at 8 weeks of age and monitored for tumor development. They were euthanized when tumors were palpable (with subsequent confirmation by necropsy) or when age-associated idiopathic dermatitis developed.  2.11 Mouse Diets All diets were from TestDiet. Unless otherwise stated, diets were switched 7 days before tumor implantation. Celebrex (Pfizer) was formulated into the diets by TestDiet. CCI-779 (LC Labs) was diluted from a 100% ethanol stock into the vehicle (5% Tween-80 + 5% PEG400 in PBS) used for IP injections into mice at 1.5 mg/kg on days 4 and 7 after tumor implantation.  2.12 Statistical Analyses GraphPad Prism (GraphPad Software, Inc.; San Diego, CA, USA) was used for statistical analyses. Briefly, tumor sizes, insulin and BG were tested for statistical significant differences by using a 1-tailed t test, and regressions were tested by using the Spearman rank correlation and F tests. A log-rank (Mantel–Cox) test was used to determine the significance of the difference between the survival curves in the spontaneous tumor study. For cytokine ELISA results, and arginase activity comparisons, 2 tailed t tests were used. Numbers were considered statistically significant if p-values were 0.05 or less. Densitometry was measured using ImageQuant 5.2 (GE Healthcare; Piscataway, NJ, USA). Although it is not ideal, we presented the error in our data as SEM rather than SD because of convention. Since we base statistical significance on calculated p-values, which are independent of SEM, and we make no conclusions based on the SEM, we feel justified in doing so.  38  3. Plasticity of Macrophage Polarization 3.1  Introduction  3.1.1  Surface Markers of M2 Macrophages  The renaissance of myeloid immunology, initiated in part on by the elucidation of TLR-biology, was further buoyed by the discovery of M2 M activation. This served to underscore the newfound specificity of the innate immune system. To identify and discriminate between different M activation states, extensive studies using flow cytometry have been undertaken to gain insight into cell-surface Ag expression of M2 Ms and the functions they perform. Two cell-surface receptors of particular interest are SR-A (CD204) and MR (CD206), both of which are expressed by Ms, and increased in M2 Ms (Mantovani et al., 2002; Sica et al., 2008b; Stein et al., 1992). SR-A is a unique receptor in that it recognizes many ligands and, as a result, has many biological roles (Gough and Gordon, 2000). Of note, SR-A is necessary for the recognition and uptake of lipoproteins, such as low density lipoprotein (LDL) and its modified counterparts, and most importantly, it mediates the phagocytosis of apoptotic cells and cellular debris, which is a canonical function of M2 (i.e., M2a or wound healing) Ms (Wolfs et al., 2011; Platt et al., 1996). MR, on the other hand, recognizes bacteria expressing mannose residues on its cell walls and has a role in Ag delivery from the cell surface to MHC-containing vesicles in APCs (Apostolopoulos et al., 2000). Although traditionally associated with M2 activation, MR has been implicated in the promotion of a TH-1 response in certain contexts (Zenaro et al., 2009). 3.1.2  Arginine Metabolism  The M1-M2 M activation paradigm was originally defined by cytokine production and L-arginine metabolic strategy (Mills, 2001; Stein et al., 1992). Over the last decade, much has developed in M research, and it comes as no surprise that the M1-M2 model is now considered to be an oversimplification of a spectrum of M activation states (Mosser et al., 2008; Mantovani et al., 2004). More extensive  39  definitions of Ms, particularly M2 activation phenotypes (i.e., M2a, which are also known as wound healing, and M2b and M2c, which are also known as regulatory Ms) have since evolved, highlighting the degree of heterogeneity that Ms possess. In these newer paradigms, L-arginine metabolism still remains a defining feature of murine Ms (i.e., M2a and wound healing Ms). Interestingly, M2a Ms, which arise in response to IL-4 or IL-13, show a more pronounced Arg1 phenotype when induced with IL-10, which alone induces M2c or regulatory Ms, than with IL-4 alone, suggesting that different M2 stimuli may cooperate to enhance certain aspects of the M2 phenotype (Stout and Suttles, 2004). L-arginine metabolism is a good differentiator of M phenotype in mice because of its profound impact on the immune response. L-arginine, while not an essential amino acid, can become limiting under certain circumstances, most notably in wound healing, where tissues express high levels of L-arginine catabolic enzymes, such as Arg1 (Mills, 2001). Because of enzyme kinetics, Arg1 is a particularly good competitor for free L-arginine, when compared to other L-arginine catabolic enzymes, such as iNOS. Thus, in a tumor or during inflammation, Arg1-expressing M2 Ms outcompete iNOS-expressing M1 Ms for L-arginine and thus limit tumoricidal NO production (Chang et al., 1998). Furthermore, L-arginine is essential for a robust adaptive response not only because it is necessary for T cell proliferation, but the limiting L-arginine also causes the production of peroxynitrite, via iNOS-produced superoxide, and the subsequent apoptosis-inducing nitration of essential T cell signaling proteins (Brito et al., 1999). Since iNOS-expressing Ms and Arg1-expressing Ms are both found in early stage tumors, it is conceivable that their coexistence may be a source of immunosuppression in a tumor scenario (Massi et al., 2007).  3.1.3  Macrophage Immune Regulation  The M cytokine production phenotype is also of particular importance because of its downstream effects on adaptive immunity. The cytokines IL-12 and IL-10, which are associated with M1 and M2 Ms, respectively, have potent but  40  antagonizing effects on direct M cytotoxic function and T cell differentiation (Mills, 2001). M-produced IL-12 elicits IFN- secretion from NK and CD4+ TH-1 cells (Xu et al., 2010). IFN-, together with TLR agonists, potentiate full M1 activation to maximize a M's cytotoxic abilities, which include NO production, pathogen phagocytosis and lysosomal degradation; these are cornerstones of the Mmediated immune response. Moreover, IL-12 also seems to have a direct cytotoxic effect on IL-12-receptor-expressing tumor cells (Gorelik et al., 2004). In terms of T cell regulation, IL-12, via IFN-, skews T cells to TH-1 by inducing T-bet transcription in naive T cells (Xu et al., 2010). IFN- is also responsible for inducing IL-2, a T cell mitogenic cytokine required for TH-1 and CTL proliferation during clonal expansion (Xu et al., 2010). Furthermore, IFN- suppresses TH-2 development by inducing apoptosis in TH-2 cells (Novelli et al., 1997). In this manner, IL-12 is a master controller of effective TH-1 adaptive and M cytotoxic responses. IL-10 production has traditionally been associated with M2 Ms and TH-2 immunity (Mills, 2001). It is produced by TH-2 cells as well as M2 Ms and acts in a paracrine and autocrine manner to inhibit the immune actions of Ms (Conti et al., 2003). Specifically, IL-10 treatment has been shown to not only blunt the NF-B response in M1 Ms, which results in a reduction in M cytotoxic responses (i.e., NO production, lysosomal degradation), it also represses T H-1 skewing by down regulating IL-12 transcription (Conti et al., 2003). Furthermore, IL-10 is an important cytokine in the generation of Tregs, which inhibit the proliferation of T H-1 cells and CTLs. Given the disparate effects of IL-12 and IL-10, it is not surprising that M1 Ms are better than M2 Ms at initiating a robust Ag-specific T cell proliferation response (Edwards et al., 2006).  3.1.4  Macrophage Phenotype Plasticity  The potential for a vast array of phenotypic activation states has been demonstrated for Ms (Martinez et al., 2008). Whether or not these activation states represent terminal cell differentiation steps, however, has not been established. 41  While cytokine production profiles have been shown to be plastic after polarization, to some degree, the plasticity of other functional aspects critically relevant to the definition of a M, such as L-arginine metabolism and Ag presentation, have not been thoroughly investigated (Stout and Suttles, 2004). PI3K and SHIP play a role in this plasticity, as can be seen in the SHIP -/mouse, wherein in vivo-derived Ms (i.e., alveolar, peritoneal) are profoundly M2skewed (Rauh et al., 2005). While it was shown that SHIP-/- M are not innately M2skewed, it is not known if they are more sensitive to M2-skewing stimuli or are somehow more intrinsically prone to M2 programming. In this chapter, we have used mouse BMMs as a model, focusing on M2a or wound healing Arg1+ Ms, which we have called ‗M2‘ Ms for the sake of simplicity. To investigate if M2-polarized Ms are a permanent, terminally differentiated celltype, we exposed mature BMMs to an M2-skewing cytokine environment, and assessed if M2 characteristics could be reversed with cytokine withdrawal or subsequent M1 activation. In the latter parts of this chapter, we assessed the plasticity of SHIP-/- BMMs in a similar manner.  3.2  Results  3.2.1  IL-4 Induces an M2, Arg1+ M Phenotype  Arg1 is expressed by in vivo derived M2 TAMs, which are associated with the progression  of  many  murine  cancers,  and  may  be  a  key  factor  in  immunosuppression in the tumor microenvironment (Martinez et al., 2008). To ascertain the potential mode of M2-skewing in such an in vivo scenario, we wanted to assess, in vitro, other modes of M2 M induction to see if they also induce Arg1. To this end, we stimulated mature BMMs with various M2-skewing factors such as IL-4 (i.e. M2a activation), immune complexes with sheep RBCs or FcR ligation (i.e. M2b activation), IL-10 (i.e. M2c activation), and TGF-(i.e. M2c activation). We found that of all of the factors tested, only IL-4 induced expression of Arg1 (and Ym1, another M2 marker associated with IL-4 stimulation). On the other  42  hand, TGF-, sheep RBC immune complexes, and 2.4g2-induced FcR ligation, but not IL-4 nor IL-10, induced COX-2 expression (Fig 3.1 A). Since only IL-4 induced the Arg1+ phenotype that is typical of many TAMs, we focused on this particular mode of M2 skewing, and found that the Arg1+ phenotype was fully induced after only two days of IL-4 treatment (Fig 3.1 B). Agreeing with increased Arg1 expression, we also found that M2 (i.e. M2a) Ms had a reduced capacity for LPSinduced NO production after M2 activation (Fig 3.1 C).  Figure 3.1 IL-4 induces M2 M skewing of mature BMMs. A) Western blots of BMMs cultured for 72 h ± IL-4 (10 ng/mL), FcR ligation (1 g/mL 2.4g2 + 1 g/mL Rat IgG) (Imm), Ab-coated sheep RBCs (20:1 RBCs to Ms) (SR), IL-10 (20 ng/mL), TGF- (10 ng/mL). B) Western blots from mature BMMs stimulated with IL-4 (10 ng/mL) for 2, 4, and 6 days. C) Time course of LPS (100 ng/mL)-stimulated NO production, of BMMs cultured for 7 days ± IL-4 (10 ng/mL). Results are representative of 3 independent experiments.  43  3.2.2  M2 Activation Leads to a Wound-Healing, Arg1+ M Phenotype  To establish whether or not the M2 phenotype was reversible, we wanted to first establish phenotypic baselines for M2 activation, starting with the L-arginine metabolic phenotype. In this thesis, M2 activation is defined as the co-stimulation of IL-4 with a TLR agonist (i.e., LPS), and M1 activation as the co-stimulation of IFN- with LPS. The stimulation of cells with IFN- or IL-4 without co-stimulation with LPS is referred to as 'priming'. In keeping with the literature, we found that IFN-stimulated the phosphorylation of STAT-1 but did not upregulate the M2 markers Arg1 or Ym1, while IL-4 stimulated STAT-6 phosphorylation and substantial Arg1 and Ym1 expression (Fig 3.2 A). We also found, in keeping with recent work from our laboratory showing  that IL-4 induces the proteasomal degradation of SHIP  protein, that IL-4 reduced SHIP levels (Ruschmann et al., 2010) (Fig 3.2 A). Furthermore, M1-activated BMMs showed significant production of NO, while their M2-activated counterparts produced very little NO (Fig 3.2 B).  44  Figure 3.2 M1 and M2 programmed BMMs display different L-arginine metabolic strategies. A) Western blots of mature BMMs cultured for 72 h with IL-4 (10 ng/mL) or IFN- (100 ng/mL) for SHIP, pSTAT-6 (Y641), pSTAT-1 (Y701), Ym1, and Arg1. B) NO production was assessed in 24 h culture supernatants from BMMs stimulated with LPS (100 ng/mL), M1-activated with LPS (100 ng/mL) + IFN- (100 ng/mL) (M1), or M2-activated with LPS (100 ng/mL) + IL-4 (10 ng/mL) (M2). Results are pooled data from 3 independent experiments. * = p ≤ 0.05 compared to control and M2-activated BMMs 3.2.3  M1 and M2 Ms Display Unique Cell-Surface Antigens  To assess the surface marker phenotype of M1 and M2 Ms, mature BMMs were M1 or M2 primed for 3 days in vitro. An interesting observation we made when we M2-primed our BMMs was that they became less adherent than control cells. Concerned that they may not be fully mature BMMs, we used flow cytometry to analyze CD11c and F4/80 expression and found that while M2-primed BMMs did increase the expression of CD11c (Fig 3.3 A), typically a DC marker, they were fully mature F4/80-expressing BMMs (Fig 3.3 B). We also analyzed the primed BMMs by flow cytometry with Abs against SR-A and MR, which are cell-surface receptors traditionally associated with M2 Ms (Mantovani et al., 2004). While MR (CD206) expression was unchanged in either 45  M1- or M2 -primed Ms (data not shown), we were surprised to find that SR-A (CD204) expression was markedly increased in M1-primed BMMs (Fig 3.3 C).  Figure 3.3 IFN--induced M1 priming increases CD204 and IL-4 ± IL-10 M2 priming increases CD11c on BMMs. A) CD11c expression in unstimulated (Control), M1-primed (100 ng/mL IFN- for 72 h), and M2-primed (10 ng/mL IL-4 for 72 h) BMMs. B) and C) F4/80 and CD204, expression, respectively, in unstimulated (Control), M1- primed (100 ng/mL IFN- for 72 h), and M2-primed (10 ng/mL IL-4 + 10 ng/mL IL-10 for 72 h) BMMs. CD204 and F4/80 results of M2 BMMs with IL-4 alone were the same as shown with IL-4 + IL-10 (data not shown). All cells were controlled with autofluorescence of unstained cells, which was at 101. Results shown are representative of 3 independent experiments.  46  3.2.4  IL-4 Withdrawal is Sufficient to Reverse M2 Arginine Metabolic  and Cell-Surface Antigen Expression Phenotypes Having established that priming with M2-skewing cytokines dramatically alters the Arg1+ and cell-surface Ag expression, we were interested in testing the plasticity of this phenotype. To do so, we skewed BMMs to an M2 phenotype and withdrew the skewing cytokine by washing the cells and leaving them in culture for an additional period of time. We found that after IL-4 withdrawal, Ym1 expression returned to levels similar to those seen in control cells, and Arg1 protein was also dramatically reduced (Fig 3.4 A). In keeping with this, LPS-stimulated NO production was also restored after IL-4 withdrawal (Fig 3.4 B). The expression of CD11c in M2primed BMMs was similarly reduced with IL-4 withdrawal (Fig 3.4 C).  47  Figure 3.4 The IL-4-induced phenotype can be reversed by withdrawing IL-4. A) Western blots of mature BMMs cultured for 12 days ± IL-4 (10 ng/mL) for 7 days, then washed and replated with IL-4 (IL-4) or without IL-4 (Wash) for another 5 days. B) Corresponding NO production of cells in panel A) after 24 h stimulation with LPS (100 ng/mL). C) CD11c expression of mature BMMs cultured for 7 days, ± 10 ng/mL IL-4 (IL-4 Day 7 or control), or IL-4 for 3 days, and then washed and cultured for 4 days without IL-4 (washed), or cells stimulated with IL-4 for the last 3 days (IL-4 Day 3). All cells were controlled with autofluorescence of unstained cells, which was at 101. Results are representative of 2 independent experiments.  48  3.2.5  M2-primed BMMs can be M1 programmed with IFN-γ  Having shown that the cell-surface marker phenotype of M2 Ms can be dampened down via IL-4 withdrawal, we were interested in whether or not M2primed Ms could be reverse-primed with IFN- to an M1-primed cell-surface marker phenotype. To do this, we washed M2-primed BMMs out of IL-4 and primed them with IFN-. Because we doubled the culture time by first priming, then reverse priming, we used flow cytometry to check the M status of BMMs after and found that M1 and M2-primed BMMs retained their CD204 and F4/80 status during this extended culture period (Fig 3.5 A). Interestingly, when we reverse primed M2-primed BMMs, we found that, like unstimulated BMMs, they retained F4/80 expression and were capable of upregulating CD204 in response to IFN- (Fig 3.5 B).  49  Figure 3.5 M2-primed BMMs can be reprogrammed with IFN-. A) Expression of CD204 (left) and F4/80 (right) on M2-primed (10 ng/mL IL-4 + IL10), on M1-primed (100 ng/mL IFN-), or on unstimulated BMMs for 6 days in culture. B) Expression of CD204 (left) and F4/80 (right) on BMMs cultured for 6 days (Control) with 10 ng/mL IL-4 + 10 ng/mL IL-10 (M2), or under these same conditions for 3 days and then switching to 100 ng/mL IFN- (M2M1) for 3 days. All cells were controlled with autofluorescence of unstained cells, which was at 101. Results are representative of 3 independent experiments.  50  In terms of the L-arginine metabolic phenotype, reverse-primed BMMs, like IL-4-withdrawn BMMs, exhibited low Arg1 and Ym1 protein levels, similar to those seen in control or M1-primed BMMs (Fig 3.6 A). Of note, while M2-primed BMMs displayed little NO production upon LPS stimulation, reverse-primed BMMs, surprisingly, showed a substantial increase in LPS-induced NO compared to control or M1-primed BMMs (Fig 3.6 B). LPS-stimulated, reverse-primed BMMs also displayed low levels of Arg1 expression and a significant increase in iNOS expression, agreeing with our NO production results (Fig 3.6 C). Furthermore, in keeping with the low Arg1 expression, reverse-primed BMMs produced less urea, which is a measure of Arg1 activity (Morrison and Correll, 2002), than M2-primed BMMs (Fig 3.6 D).  51  Figure 3.6 M2-primed BMMs can be reprogrammed. A) Western blots of cell lysates from BMMs cultured for 8 days ± 100 ng/mL IFN- (M1-primed) or 10 ng/mL IL-4 + 10 ng/mL IL-10 (M2-primed), or M2-primed for 4 days then washed and M1-primed for the remaining 4 days (M2M1). B) 24 h LPSinduced NO secretion, expressed as a percent of control BMM production (mean ± SEM). 24 h LPS-induced NO secretion of BMMs cultured and primed in a similar manner as in A). Black arrow indicates the absence of NO production in M2-primed BMMs. Results shown in panel (B) are pooled from 2 independent experiments C) Representative Western blots of cell lysates of BMMs primed in a similar manner as those in B) 48 h after stimulation with 100 ng/mL LPS. All lanes come from the same gel and the same time exposed film. Dotted lines indicate where irrelevant lanes have been cropped out. D) urea production from 24 h to 48 h cultures after LPS (100 ng/mL) stimulation as a measure of Arg1 activity in live BMMs from (C). Black arrow indicates the absence of detectable urea production. Results are representative of 2 independent experiments * = p ≤ 0.05 comparing control with M2-primed BMMs  52  3.2.6  M2 Priming Shifts M Cytokine Production Profile to a Pro-  inflammatory M1-like Phenotype. Next, we assessed cytokine production from M1 and M2-activated BMMs. In keeping with the literature, we found that M2-activated BMMs produced more IL-10 than M1-activated BMMs, but surprisingly, less than their control LPS-stimulated counterparts (Fig 3.7 A). For the pro-inflammatory cytokines IL-12 and TNF-, M2activated BMMs produced the least, while M1-activated BMMs produced the most (Fig 3.7 B and C).  Figure 3.7 M2-activated BMMs produce more IL-10 and less IL-12 or TNF-. Levels of A) IL-10, B) IL-12p70, and C) TNF-in 24 h stimulated BMM cultures. Control cells were stimulated with 100 ng/mL LPS alone; M1-activated cells were costimulated with 100 ng/mL IFN-; and M2-activated cells were co-stimulated with 10 ng/mL IL-4. Results are representative of 2 independent experiments. Data shown are means ± SEM of assay triplicates. * = p ≤ 0.05 compared to any other experimental condition  53  Interestingly, when M2-primed BMMs were challenged with LPS, we found that, unlike M2-activated BMMs, LPS-stimulated M2-primed BMMs behaved more like M1-activated BMMs and produced less IL-10 and more IL-12 than unprimed BMMs (Fig 3.8 A and B). In contrast to this pro-inflammatory trend, however, M2-primed BMMs produced less TNF- upon LPS-stimulation than control or M1-activated BMMs (Fig 3.8 C).  Figure 3.8 M2-primed BMMs produce less IL-10 and TNF- but more IL-12 in response to subsequent LPS stimulation than control BMMs. Levels of A) IL-10, B) IL-12p70, and C) TNF- were assessed in culture supernatants from 24h LPS (100 ng/mL)-stimulated BMMs pre-cultured ± 10 ng/mL IL-4 for 3 days. BMMs cultured without IL-4 were M1-activated with 100 ng/mL LPS + 100 ng/mL IFN- for comparison. The black arrow indicates the absence of any detectable IL-10 in the culture medium. Results shown are means ± SEM of assay triplicates pooled from 2 independent studies, and representative of 4 independent experiments. * = p ≤ 0.05 compared to the control. 54  We next assessed the capacity of M2-primed cells to respond to M1 activation and found that 3-day M2-primed BMMs, similar to M1-activated unprimed BMMs, produced less IL-10 than LPS-stimulated unprimed BMMs after M1-activation (Fig 3.9 A left panel). We also found that extended priming for 3 additional days still produced levels of IL-10 similar to M1-activated unprimed BMMs (Fig 3.9 A right panel). In terms of pro-inflammatory cytokines, both 3-day and 6-days M2-primed BMMs produced significantly more IL-12 than either LPS-stimulated or M1activated unprimed BMMs (Fig 3.9 B). While TNF- production was also increased in M1-activated M2-primed cells at 3 days of priming, this increase was absent with 6 days of priming (Fig 3.9 C).  55  Figure 3.9 M2-activated BMMs can be reverse activated and are primed for M1-activation. A) Levels of IL-10 in culture supernatants of M1-activated (100 ng/mL LPS + 100 ng/mL IFN-) BMMs cultured ±10 ng/mL IL-4 for 3 days (left panels), or 6 days (right panels). Unconditioned BMMs were stimulated with LPS alone as control. B) IL-12 and C) TNF-Levels of culture supernatants from BMMs cultured in the same manner described in (A). Data shown are means ± SEM of assay triplicates (IL-12) and duplicates (IL-10, TNF-). Results are representative of 2 independent experiments. The black arrow indicates the absence of detectable IL-10. ** = p ≤ 0.01 compared to the control. 56  3.2.7  M2-Primed  BMMs  Induce  Less  Antigen-Specific  T  Cell  Proliferation than M1-Primed BMMs Another critical aspect of M function is the ability to present Ag and initiate a T cell response. We tested the ability of M1- and M2-primed BMMs to induced Agspecific T cell proliferation by using freshly isolated splenic CD4 + T cells, from OT-II mice (Barnden et al., 1998), that possess TCRs that specifically recognize ovalbumin (OVA) peptide (amino acid 323-339). In an LPS and OVA peptide stimulated co-culture system with BMMs and CD4+ T cells, using [3H]thymidine incorporation as an indicator of proliferation, we found that M2-primed BMMs induced T cell proliferation to a lesser degree than M1-primed BMMs, but to a greater degree than unprimed BMMs (Fig 3.10 A). To look at the reversibility of this phenotype, we primed BMMs for an extended period in culture and reverse-primed the M2-primed BMMs with IFN- for the second half of the extended culture period to see if the M2-primed BMM-T cell stimulation phenotype could be reversed. We found that even with extended priming, M1- and M2-primed BMMs retained their respective T cell stimulation phenotype as shown previously (Fig 3.10 B). As well, similar to other aspects of M function, a subsequent exposure to IFN- was sufficient for the induction of an M1-primed T cell stimulation phenotype in M2-primed BMMs, albeit to a lesser degree than in unprimed BMMs (Fig 3.10 B).  57  Figure 3.10 M1-primed BMMs are better inducers of Ag-specific T cell proliferation than M2-primed BMMs. 3 T cell proliferation, indicated as counts per minute (cpm), of [ H]thymidine+ incorporated ovalbumin (OVA)-specific CD4 T cells isolated from the spleens of OTII mice stimulated with LPS (50 ng/mL) and OVA peptide (0.5 g/mL) for 72 h, and 3 with [ H]thymidine (1 Ci/well) for the last 18 h. A) T cells were cultured alone (T cell) or with control, 3 day M1-primed (10 ng/mL IFN-), or 3 day M2-primed (10 ng/mL IL-4) BMMs at a T cell:BMMratio of 40:1. B) T cells were cultured alone (T cell) or with control, 14 day M1-primed (10 ng/mL IFN-) (M1),14 day M2-primed (10 ng/mL IL-4) (M2), or 7 days M2-primed then 7 days M1-primed BMMs (M2M1) at a T cell:BMMratio of 20:1. Data shown are means ± SEM of experimental triplicates. Results are representative of 2 independent experiments. # = p ≤ 0.055 compared to M2-primed cells. * = p ≤ 0.05 compared to control BMMs. ** = p ≤ 0.01 compared to all other conditions.  58  3.2.8  M2 M Phenotype of SHIP-/- BMMs  Since in vivo derived SHIP-/- Ms are heavily M2-skewed, we wanted to know if the absence of SHIP affected the plasticity of BMMs. To do this, we first tested M2a, 2b, and 2c stimuli and found that, similar to what we obtained in SHIP+/+ Ms (Fig 3.1 A), only IL-4 stimulation led to an Arg1+ M2 phenotype (Fig 3.11 A). Also, SHIP-/- BMMs that were M1-primed (IFN-) expressed pSTAT-1, while those that were M2-primed (IL-4) expressed pSTAT-6, Ym1, and Arg1 (weakly) (Fig 3.11 B). As seen by the marked reduction in protein expression, Ym1 was readily reversed within 72 h of IL-4 withdrawal and Arg1 was also reversible, albeit to a lesser degree than Ym1, in both SHIP+/+ and SHIP-/- mature BMMs (Fig 3.11 C). Unlike SHIP+/+ BMMs, LPS-stimulated, IL-4-withdrawn, SHIP-/- M2-primed BMMs did not produce the same levels of NO as LPS- stimulated unprimed SHIP-/- BMMs or even M2-primed SHIP-/- BMMs (Fig 3.11 D); however, IL-4-withdrawn SHIP-/- M2primed BMMs produced more NO between 24 and 48 h after LPS stimulation than M2-primed SHIP-/- BMMs (Fig 3.11 D).  59  Figure 3.11 SHIP-/- M2 BMM phenotype is reversible. A) Western blots of BMMs cultured for 72 h ± IL-4 (10 ng/mL), FcR ligation (1 g/mL 2.4g2 + 1 g/mL -Rat IgG) (Imm), Ab-coated sheep RBCs (20:1 RBCs to Ms) (SR), IL-10 (20 ng/mL), or TGF- (10 ng/mL). B) Western blots of mature -/SHIP BMMs cultured for 72 h with IL-4 (10 ng/mL) or IFN- (100 ng/mL) and probed for pSTAT-1 (Y701), pSTAT-6 (Y641), Ym1, and Arg1. C) Western blots of +/+ -/SHIP and SHIP mature BMMs pre-treated with 10 ng/mL IL-4 for 72 h, then washed and re-cultured with (IL-4) or without (C) IL-4 for another 72 h. All lanes come from the same gel and same time exposed film. Dotted lines indicate where irrelevant lanes have been cropped out. D) 24 h NO production induced by 100 +/+ -/ng/mL LPS + 100 ng/mL IFN- from mature SHIP (left panel) and SHIP (middle panel) BMMs pre-treated with IL-4 (10 ng/mL) for 72 h, then washed and recultured with (IL-4) or without IL-4 (Wash) for another 120 h. The NO production of each genotype between 24 h and 48 h is shown in the right panel. Results are representative of 2 independent experiments.  60  Looking at surface markers, we found that SHIP-/- BMMs were similar to their SHIP+/+ counterparts in that CD204 and CD11c appeared to be reliable markers of M1 and M2 priming, respectively (Fig 3.12 A). Also, the increase in CD11c in M2primed SHIP-/- BMMs could be reduced to control levels with cytokine withdrawal, much like that observed with M2-primed SHIP+/+ BMMs (Fig 3.12 B). Furthermore, like the surface marker phenotype of SHIP+/+ BMMs, M2-primed SHIP-/- BMMs, when reverse primed with IFN- increased CD204 expression (Fig 3.12 C).  61  Figure 3.12 The SHIP-/- M2 BMM surface marker phenotype is reversible. -/A) CD11c expression in SHIP unstimulated (Control), M1-primed (100 ng/mL IFN- for 72 h), and M2-primed (10 ng/mL IL-4 for 72 h) BMMs (left panel). CD204 expression in Naïve (Control), M1- primed (100 ng/mL IFN- for 72 h), and M2primed (10 ng/mL IL-4 + 10 ng/mL IL-10 for 72 h ) BMMs (right panel). CD204 expression of M2 BMMs activated with IL-4 ± IL-10 were the same (data not -/shown). B) CD11c expression of SHIP BMMs cultured for 7 days, with or without 10 ng/mL IL-4 (IL-4 or control), or cultured in IL-4 or 3 days, then washed and cultured for 4 days without (washed). Green peak indicates expression of CD11c of -/IL-4-stimulated cells at 3 days. C) CD204 expression of SHIP BMMs cultured for 6 days with 10 ng/mL IL-4 + 10 ng/mL IL-10 (M2-primed), or M2-primed for 3 days then M1 conditioned (100 ng/mL IFN-) for the remaining 3 days. All cells were controlled with autofluorescence of unstained cells, which was at 101. Results shown are representative 3 independent experiments.  62  We wanted to determine if M2-priming had the same effect on cytokine secretion from SHIP-/- BMMs as with SHIP+/+ BMMs. We found that while M2primed SHIP-/- BMMs, like SHIP+/+ cells, produced more IL-12 in response to LPS compared to unprimed BMMs, they also produced higher levels of IL-10, in contrast to SHIP+/+ BMMs (Fig 3.13 A and B).  Figure 3.13 LPS-stimulated M2-primed SHIP-/- BMMs produce more IL-12 and IL-10 compared to unprimed BMMs. A) Levels of IL-12p70 and B) IL-10 in supernatants from 24 h LPS (100 ng/mL) +/+ -/stimulated SHIP and SHIP control or M2-primed (10 ng/mL IL-4 for 3 days) BMMs. Results shown are means ± SEM of experimental triplicates pooled from 2 independent experiments. * ** *** @  = p ≤ 0.05 compared to control cells of the same genotype. = p ≤ 0.01 compared to control cells of the same genotype. = p ≤ 0.001 compared to control cells of the same genotype. = p ≤ 0.06 compared to control cells of the same genotype. To see if this cytokine secretion profile of SHIP-/- BMMs was reversible, we  compared the cytokine production profiles of control LPS-stimulated, M1-activated, M2-activated, and M2-primed then M1-activated BMMs. Since the differences between the secretion of IL-10 and IL-12 in the differently stimulated SHIP+/+ BMMs were more pronounced than that of TNF-, we focused on IL-12 and IL-10 secretion. We found that the IL-10 production pattern of LPS-stimulated, M1-activated, and M2activated SHIP-/- BMMs was similar to that of SHIP+/+ BMMs, and when M263  primed SHIP-/- BMMs were M1 activated, they secreted low levels of IL-10 similar to those of M1-activated BMMs (Fig 3.14 A). For IL-12, M2-activated SHIP-/BMMs produced significantly less than LPS-stimulated unprimed or M1-activated BMMs, and like in SHIP+/+ BMMs, M2-priming seemed to sensitize SHIP-/BMMs to M1-activation-induced IL-12 production (Fig 3.14 B). Notable differences between the SHIP+/+ and SHIP-/- cells in this context was that while SHIP+/+ M1activated, M2-primed BMMs produced significantly less IL-10 than M2-activated BMMs, and SHIP+/+ M1-activation significantly increased IL-12 production compared to control LPS-stimulated cells, such differences were not observed in similarly treated SHIP-/- BMMs (Fig 3.14 A and B). These cytokine production differences between similarly treated SHIP+/+ and SHIP-/- were consistently observed, although not always as dramatic as those shown in this figure. In terms of T cell activation, M1-primed SHIP-/- BMMs induced more T cell proliferation than M2-primed SHIP-/- BMMs, or unprimed BMMs (Fig 3.14 C). In contrast to the SHIP+/+ BMMs, M2-priming did not increase the ability of BMMs to induce T cell proliferation compared to control (Fig 3.14 C). Reverse priming of M2primed SHIP-/- BMMs showed that the M1-primed T cell stimulation phenotype could be induced despite prior M2-priming (Fig 3.14 C).  64  Figure 3.14 Functional phenotype of M2-primed SHIP-/- BMMs is reversible and similar to that of SHIP+/+ BMMs. +/+ A) Levels of IL-10 and B) IL-12p70 in 24h-post stimulation supernatants of SHIP -/and SHIP BMMs, which were LPS (100 ng/mL) stimulated (C), M1-activated (LPS + 100 ng/mL IFN-M2-activated (LPS + 10 ng/mL IL-4), or M2-primed for 3 days then M1-activated. Results were normalized to total protein and shown in panels (A) and (B) as means ± SEM of assay triplicates pooled from 2 independent experiments. C) T cell proliferation, indicated as counts per minute (cpm), of 3 + [ H]thymidine-incorporated ovalbumin (OVA)-specific CD4 T cells isolated from the spleens of OT-II mice stimulated with LPS (50 ng/mL) ± OVA peptide (0.5 g/mL) for 3 72 h, and with [ H]thymidine (1 Ci/well) for the last 18 h. The different BMMs were conditioned as indicated in Fig. 3.10 at a T cell:BMMratio of 40:1. Data are means ± SEM of experimental triplicates, and representative of 2 independent experiments. * = p ≤ 0.05 compared to any other condition. ** = p ≤ 0.01 compared to any other condition. ***= p ≤ 0.001 compared to any other condition. # = p ≤ 0.05 in compared to M1 of the same genotype. ¥ = p ≤ 0.05 compared to C and M2 of the same genotype. @ = p ≤ 0.075 compared to C and M2 of the same genotype. † = p ≤ 0.05 compared to similarly conditioned cells of the other genotype.  65  Since we only observed a weak Arg1 upregulation in IL-4-stimulated SHIP-/BMMs (Fig 3.11 A), we wanted to compare the sensitivity of SHIP+/+ and SHIP-/BMMs to M2-skewing. Interestingly, while IL-4 induced Ym1 to similar levels, Arg1 upregulation was lower in SHIP-/- BMMs in response to the same dose of IL-4, and also, SHIP-/- BMMs stimulated with M2b- or M2c-skewing stimuli (i.e. 2.4g2induced FcR ligation, sheep RBC immune complexes, and TGF-) appeared to have less COX-2 protein than similarly stimulated SHIP+/+ BMMs (Fig 3.15).  Figure 3.15 SHIP-/- BMMs induce less Arg1 in response to IL-4. +/+ Uncropped version of Fig 3.1 A and Fig 3.11 A (same gel) showing SHIP and -/SHIP BMMs stimulated with M2-skewing factors for 72 h. Results are representative of 5 independent experiments.  66  3.3  Discussion Since M phenotypes are often hallmarks of various diseases, especially  cancer, we were interested in the permanency of M M2 activation. In this chapter, we explored the plasticity of M phenotypes, and asked if the M2 M phenotype, which often characterizes TAMs and predicts a poor prognosis, could be manipulated. Specifically, we assessed in vitro the M2 M phenotype, focusing on aspects relevant to tumor immunology, by examining the stability of surface marker expression, L-arginine metabolic strategies, and T cell stimulatory properties. Furthermore, we investigated the M2 M phenotype of SHIP-/- BMMs to see if the reason they are M2-skewed in vivo may be due to a cell-intrinsic propensity towards M2 programming. While we used M-CSF-containing BMM for BMM generation, all experiments were carried out using M-CSF-free PM. Repeat experiments using BMM yielded similar results.  3.3.1  Surface Marker Phenotype  We were surprised at first to find that SR-A was induced by IFN- priming, because it is traditionally associated with the cellular debris clearance function of M2 Ms (Mantovani et al., 2004). However, Since SR-A, despite its name, is an antimicrobial  pattern  recognition  receptor  as  well  as  a  scavenger  receptor  (Mukhopadhyay et al., 2004), it is not surprising that M1-priming induces its expression. Furthermore, it has also been shown in Ms from Balb/C mice that IFN induces expression of SR-A (Mukhopadhyay et al., 2004), suggesting that perhaps SR-A is a marker of both M1- and M2-activated Ms. CD11c is a myeloid integrin that mediates cell migration, adhesion and phagocytosis (Sadhu et al., 2007). While it is thought to be expressed primarily on DCs, certain tissue-specific Ms, such as alveolar and adipose tissue Ms, also express CD11c (Moon et al., 2007; Fischer-Posovszky et al., 2011; Lloyd et al., 2008). Interestingly, both adipose tissue and alveolar Ms exhibit other M2 markers, and in the lungs in particular, infiltrating eosinophils are known to produce IL-4, especially during inflammation (Moon et al., 2007; Fischer-Posovszky et al., 2011).  67  Given this evidence of M2-like Ms expressing CD11c, our results showing that we find priming with IL-4 leads to CD11c upregulation fits with the literature. Indeed, our results support the idea that CD11c may be a context-dependent marker of M2 M activation. When investigating the surface marker phenotype of IL-4-stimulated M2 BMMs, we did not detect a significant increase in MR with IL-4-primed BMMs (data not shown). While this appears to contradict established dogma, a closer look at previous reports reveals that this may be a phenotypic variation between the M model we used versus those used in studies to determine the regulation of MR by IL-4. Two key papers identifying the increase of MR mRNA and protein expression in IL-4-stimulated Ms used thioglycolate or BioGel-elicited PMs as a model, while we used BMMs. While they are both legitimate M models, it has been shown that their functions differ from resident peritoneal Ms; thioglycolate, for example, seems to select for a more Mo-like M by ablating resident peritoneal Ms, and encouraging Mo infiltration and differentiation (Zhang et al., 2008; Ghosn et al., 2010). It is therefore, conceivable that MR regulation might be different between these Ms and BMMs. In the context of phenotypic reversibility, it seems that the IL-4-induced upregulation of CD11c returns to control levels with cytokine withdrawal, and M2primed cells can be reverse-primed with IFN- to express SR-A at the same level as unprimed BMMs (Fig 3.4, Fig 3.5). This suggests that while IL-4 induces a strong and distinct surface marker phenotype, it is highly context-dependent and seems to require the continued presence of IL-4. In fact, we have preliminary evidence that even in vitro GM-CSF + IL-4-derived myeloid DCs lose CD11c expression when IL-4 is withdrawn from culture (Ho and Krystal, unpublished data). It is interesting to note that IL-4-induced CD11c expression in both SHIP+/+ and SHIP-/- BMMs was higher after IL-4-stimulation for 3 days than at 7 days (Fig 3.4, Fig 3.12). Although we did not examine IL-4 consumption in culture, the reduction of CD11c expression with time may be due to decreasing levels of IL-4, as it is being consumed by BMMs in vitro.  68  3.3.2  Arginine Metabolism and M2 Programming  Amongst the M2 M subtypes proposed by Mantovani and Mosser, only IL-4induced Ms were previously reported to express Arg1 and Ym1, and not surprisingly, this is what we found as well (Mosser and Edwards, 2008; Martinez et al., 2008). We also found, in accordance with previous reports (Mills, 2001; Stein et al., 1992), that NO production from M2-activated Ms was significantly lower than that from M1-activated Ms. Although not a major focus of this thesis, we also found that immune complex-activated BMMs have a COX-2+ phenotype, which others have also shown (Zhang et al., 2009). Consistent with previous studies from our laboratory, we found that Arg1+ M2 Ms have impaired NO production between 2448 h after LPS stimulation, which is the time during which Arg1-mediated L-arginine depletion begins to limit NO production (Rauh et al., 2005). The IL-4-induced Arg1+ phenotype was readily reversed in BMMs, since levels of Arg1 and Ym1 were markedly reduced following either IL-4 withdrawal or reversed priming with IFN-. Because the Arg1+ phenotype was fully induced within 2 days at the dose of IL-4 used and that its reversibility was apparent in BMMs stimulated for either 3 or 6 days in IL-4 suggests that continuous IL-4 stimulation may be required to maintain Arg1 and Ym1 expression; this observed inclination of BMMs to return to their unstimulated state also suggests that Arg1 + M2-activation may be a transient state induced by T H-2 inflammation and not a permanent differentiation step. Related to this, low NO production in IL-4-induced M2 Ms was also reversible. This was not unexpected given that IL-4 dramatically increased Arg1 expression, and Arg1-mediated L-arginine depletion is the primary mechanism by which Ms inhibit NO production (Bogden et al., 1994; Rutschman et al., 2001). Thus, since IL-4 withdrawal or reverse-priming both reduced Arg1 protein, we expected LPS-induced iNOS expression and NO production to be restored. What was intriguing, however, was our finding that LPS-induced NO production was increased over control BMMs in reverse-primed cells, where pre-treatment with IL-4 and its subsequent withdrawal seemed to potentiate NO production. A  69  possible explanation for this is that while IL-4 is a potent inducer of NO-limiting Arg1, it also upregulates cationic amino acid transporters (CATs), through which Larginine, the common substrate for both iNOS and Arg1, enters the cell (Niese et al., 2010). Thus, IL-4 treatment, by increasing a M's capacity to import L-arginine, could increase the substrate supply for iNOS after LPS stimulation. Although the continued presence of increased CAT transporters is required to confirm this hypothesis, the fact that reverse-primed BMMs produced more urea in culture than M1-primed or control BMMs, even though there was no detectable increase in Arg1 protein, supports the notion that IL-4 primed cells have a greater L-arginine flux. Another possible explanation for IL-4 pretreatment sensitizing the reverseprimed BMMs for NO production is that because it reduces SHIP protein levels, since LPS-induced NO production has been shown to be PI3K-dependent and SHIP would restrict this (Sakai et al., 2006; Rauh et al., 2005). A model summarizing our reversibility study results in terms of cell surface and L-arginine metabolism parameters of M2 BMMs is shown in Fig 3.16.  70  Figure 3.16 Model of the reversibility of the M2 cell surface and L-arginine metabolic phenotypes. A) Cell surface marker expression of M1 and M2 Ms and the reversibility of these phenotypes. B) Arg1, Ym1, and NO production status of differently activated Ms and the reversibility of those phenotypes. Solid lines represent previously known relationships, while dotted lines represent ways the M phenotype can be altered based on the work presented herein.  71  3.3.3  Immune Regulatory M Phenotype  Having demonstrated that the IL-4-induced inhibition of BMM NO secretion could be ameliorated by IL-4 withdrawal or IFN- addition, we were interested in whether this held true for other aspects of the M2 phenotype. Our results suggest that IL-12, which is a critical TH-1 cytokine produced by M and DCs and implicated in tumor regression, is produced at higher levels from M1-activated than from M2activated BMMs, in keeping with the literature (Martinez et al., 2008; Kilinc et al., 2008). Surprisingly, however, M2-priming potentiated BMM IL-12 secretion upon subsequent  M1-activation.  Although  this  may  seem  counter-intuitive,  this  phenomenon has been shown previously to occur with both Ms and DCs, and the authors of these studies have attributed the increased IL-12 production to an IL-4mediated silencing of the IL-10 promoter (Stout and Suttles, 2005; Yao et al., 2005; Varin et al., 2010). This is consistent with our results, since we found that M2-primed BMMs produced less IL-10. Although we did not directly demonstrate that the reduced IL-10 production we observed was the causal factor in the increase in IL-12 production in M2-primed BMMs, our results are consistent with studies showing that IL-10-induced STAT-3 activation inhibits the transcription of the p35 subunit of IL-12 in Ms (Rahim et al., 2005; Kortylewski et al., 2009). Interestingly, the IL-12/IL-10 profile of M1-activated BMMs and M2-primed, and then LPS or LPS + IFN--stimulated BMMs, were very similar. As a result, there was no reversibility to assess. Our results do suggest, however, that IL-4 pretreatment may enhance M1-type responses upon subsequent stimulation, since LPS-stimulation of control BMMs gave an IL-12Mid IL-10Hi cytokine profile, while similarly stimulated M2-primed BMMs had an M1-activation-like, IL-12Hi IL-10Lo cytokine profile and M1-activated M2-primed BMMs made more IL-12 than BMMs that were just M1-activated. These findings are summarized in a model (Fig 3.17 A). While cytokine secretion represents an important component of M immune regulation, an equally important component is Ag-presentation, since it is the bridge that connects innate and adaptive immunity. In the work of Edwards et al. (2006), 72  the authors demonstrated that IL-4-stimulated M2 Ms are poor Ag presenters compared to M1-activated Ms. Herein, we have shown that this is indeed the case, and adding to this finding, we discovered that M2-primed BMMs can be reverseprimed with IFN- to increase their capacity to induce Ag-specific T cell proliferation. Summarized in Fig 3.17 B, our data also suggest that although M2-primed BMMs are poor T cell stimulators compared to their M1-primed counterparts, they induced more T cell proliferation than unprimed BMMs. Since it is known that both IL-4 and IFN- increase M expression of accessory molecules critical to Ag presentation, such as major histocompatibility complex-II (MHC-II) (Martinez et al., 2009; Paludan, 1998), it is not surprising that both M1 and M2-primed Ms have a greater T cell stimulatory capacity than unstimulated Ms, especially since naïve Ms,  via  NO-mediated  inhibition  of  T  cell  proliferation,  possess  an  immunosuppressive and not an immunostimulatory phenotype (Hamilton et al., 2010).  73  Figure 3.17 Model of the reversibility of M2 functional phenotypes. A) Summary of the cytokine secretion profile (i.e., the IL-12 and IL-10 balance) of differently stimulated Ms. B) Summary of the T cell stimulatory phenotype of M1 and M2 Ms in terms of how strongly they increase Ag-specific T cell proliferation. Solid lines represent previously known relationships, while dotted lines represent ways M phenotype can be altered based on the work presented herein. * denotes the hyper M1 phenotype displayed by M2-primed then M1-activated Ms.  74  3.3.4  M2 Phenotype of SHIP-/- Ms  Because of the profound M2-skewing of in vivo derived Ms in the SHIP-/mouse (Rauh et al., 2005), we wanted to examine the phenotype of SHIP-/- M2 Ms to see if it was somehow fundamentally different from that of SHIP+/+ M2 Ms. While we did find that M2-priming increased LPS-induced IL-10 secretion in SHIP-/- but not SHIP+/+ BMMs, in general, our data suggest that the M2-assocated phenotypes examined were present and as reversible in SHIP-/- as in SHIP+/+ BMMs. This implies that the M2 phenotype of in vivo derived SHIP-/- Ms is likely not due to an innate difference in plasticity or affinity towards an M2 phenotype. In fact, since we consistently observed that IL-4-stimulated SHIP-/- BMMs induced less Arg1 than similarly stimulated SHIP+/+ BMMs, this implies that not only are SHIP-/- BMMs not predisposed to the Arg1+ M2 phenotype, they may actually be more resistant to it than SHIP+/+ BMMs.  3.3.5  M M2 Phenotype Reversibility  In this chapter, we examined the stability of the IL-4-induced M2 M phenotype using BMMs as a model to determine, as a proof-of-principle, if M phenotype manipulation is a possible avenue of therapy. Our work extends the findings of Stout et al. (2005) who showed that M cytokine secretion profiles are very much dictated by their immediate cytokine environment by demonstrating that cell-surface receptors, L-arginine metabolism, and T cell stimulatory properties are all reversible. We also determined that of all of the modes of M2 programming tested, only IL-4 induced the Arg1+ phenotype, which has been associated with cancer promotion (Sharda et al., 2011; Rauh et al., 2005). While we've demonstrated here that the in vitro IL-4-induced M2 M phenotype is reversible, it would be interesting to know if this is the case in vivo. From previous work, we have found that the SHIP-/- Arg1+ M2 phenotype is dependent on PI3K, IL4, and partially dependent on MP containing TGF- (Kuroda et al., 2009; Rauh et al., 2005); however, we have also found no in vivo evidence of increased circulating IL-4 or IL-13 in SHIP-/- mice, despite their Arg1+ M2-skewed Ms. From this and the work  75  presented in this chapter, the M2 M phenotype observed in the SHIP-/- may not be the result of IL-4 stimulation alone but a combination of stimuli (i.e., IL-4, TGF-), or a unique stimulus. Perhaps in the case of the Arg1 phenotype, a combinatorial stimulus is resistant to reversal, while the M2 phenotype of Ms stimulated with IL-4 alone is reversible. Indeed, it has been recently shown that in vivo derived cancerpromoting Arg1+ M2 Ms may be regulated by factors other than IL-4 (Sharda et al., 2011). Therefore, the study of in vivo derived M2 Ms using an M2-skewed mouse model may provide greater insight into how M2 Ms arise in vivo.  76  4. Identifying the Factors that Promote the SHIP-/- M2 Macrophage Phenotype 4.1  Introduction In the previous chapter, we investigated the plasticity of the Arg1 + M2  Mphenotype to assess its reversibility as a proof-of-principle that the M phenotype can be manipulated for therapy in diseases, such as cancer. Although we demonstrated the phenotypic reversibility of Arg1 + M2 Ms, which current models of murine M activation define as those induced by IL-4 or IL-13 (Mosser and Edwards, 2008; Martinez et al., 2008), we wanted to identify the factors that drive this Arg1+ M2 Mphenotype in vivo. Since we found that addition of mouse plasma (MP) to standard in vitro cultures mimicked the phenotypes of SHIP+/+ and SHIP-/Ms obtained in vivo, we hypothesized that the factors present in MP that promote M2 skewing in vitro might also be the ones responsible for skewing in vivo and thus set out to identify these factors.  4.1.1  In vivo Derived Arg1+ M2 Ms  The terms M1 and M2 Ms were originally coined to highlight the parallels between M1 - M2 activation and TH-1 - TH-2 immunity, and to indicate that T H-1associated IFN- and TH-2-associated IL-4 drive M1 and M2 Ms, respectively (Mills, 2001). Thus, it was not a surprise to find that two common situations in which Arg1+ M2 Ms arise in vivo are during TH-2-driven inflammation, characterized by high IL-4 levels in the peripheral blood during, for example, a parasitic infection (Satoh et al., 2010), and in wound healing. The major effector cells that produce IL-4 to drive M2 and TH-2 immunity during parasitic infections have been shown to be CD4+ TH-2 cells, eosinophils, and basophils (Voehringer et al., 2004). Basophils, especially, seem to be responsible for the initial production of IL-4 to drive M2 and TH-2 immunity (Min et al., 2004). In wound-healing, on the other hand, Arg1+ M2 Ms, which replace NO-producing M1 Ms as a wound heals, are thought to be driven by HIF-1 (Albina and Reichner, 2003).  77  4.1.2  M2 Ms in SHIP-/- Mice  In 2005 (Rauh et al.), we found that Ms in SHIP-/- mice were profoundly M2skewed and that this correlated with an increased rate of tumor growth. Interestingly, we also found that SHIP-/- BM was not innately predisposed to M2 programming, since in vitro culturing with M-CSF alone (i.e., standard in vitro culture conditions) did not generate M2 Ms from SHIP-/- BM (Rauh et al., 2005). However, we could mimic the in vivo differentiation pattern of Ms in SHIP+/+ (i.e., M1) and SHIP-/- (i.e., M2) by adding either MP, mouse serum, human plasma or human serum to standard in vitro cultures (Rauh et al., 2005). In this system we found that the generation of M2 Ms from SHIP-/- BM was partially dependent on TGF- and PI3K activity (Rauh et al., 2005). Also of interest, neither mouse nor human plasma or sera was capable of skewing mature Ms to an M2 phenotype (Rauh et al., 2005). More recent studies from our laboratory have implicated basophils and basophil progenitors as critical players in M2 and TH-2 programming in SHIP-/- mice (Kuroda et al., 2009; Kuroda et al., 2011). Specifically, we found that the addition of IL-3 to lineage depleted (lin-) BM resulted in M2 skewing, with resulting SHIP-/- Ms being more M2 skewed than their SHIP+/+ counterparts (Kuroda et al., 2009; Kuroda et al., 2011). Importantly, we found that IL-3 induced this M2 skewing by stimulating basophils and basophil progenitors, which express the leukocyte integrin, CD49b (detected by the DX5 monoclonal Ab), to survive, proliferate and secrete IL-4 (Kuroda et al., 2009). Based on these results we hypothesized that IL-3-induced more M2-skewing of SHIP-/- than SHIP+/+ BMMs because the IL-3-stimulated IL-4 production from basophils/basophil progenitors was PI3K-dependent and so, because SHIP-/- basophils lack the negative PI3K regulator SHIP, they would produce more IL-4 in response to IL-3 (Kuroda et al., 2011). Related to this, SHIP-/basophils and basophil progenitors were also found to be hypersensitive to IgE stimulation compared to their SHIP+/+ counterparts (Kuroda et al., 2011). Specifically, IgE stimulated abnormally high IL-4 production from SHIP-/-, but not SHIP+/+, basophils, and may be the major reason for the observed TH-2 skewing in the SHIP-/mouse (Kuroda et al., 2011).  78  4.1.3  Elucidation of the Factors Responsible for the M2 Skewing of  SHIP-/- Ms In this chapter, because SHIP-/- M2 TAMs correlated with faster tumor growth (Rauh et al., 2005), we wanted to determine both the cellular and non-cellular elements that skewed maturing SHIP-/- Ms to an M2 phenotype so that we could reverse this skewing in vivo. Because of the wealth of literature supporting the link between Arg1 and the M2 phenotype in our model system, we focused on the induction of Arg1, using it as a surrogate for M2 M programming. To address the differences between in vivo and the standard M-CSF-driven in vitro BMM culture differentiation environments, we modified our in vitro cultures. As before (Rauh et al., 2005), to account for the fact that M progenitors spend a portion of their lifespan in circulation, we supplemented with plasma or serum (human or mouse) to mimic in vivo differentiation: the plasma/serum supplemented derivation of SHIP-/- BMMs will from hereon be known as BM environment (BMENV) cultures/assays. Also for the majority of our experiments, we used adherence-depleted (adh-) BM, instead of lin- BM to set up our cultures in order to preserve some of the native cellular interactions we would expect in vivo. We felt it necessary, however, to use non-adherent progenitor cells rather than whole BM in order to eliminate adherent mature or semi-mature Ms, which would confound our results.  4.2  Results  4.2.1  IL-4 is Essential for Plasma-Induced M2-Skewing of SHIP-/-  BMMs As mentioned previously, we showed in earlier studies (Rauh. et al., 2005) that MP and human plasma/serum skewed SHIP-/-, but not SHIP+/+ BM to M2 BMMs during in vitro differentiation. Interestingly, we also found that this wasn‘t the case with mature BMMs, since the addition of MP did not skew either SHIP +/+ or SHIP-/BMMs to an M2 phenotype. IL-4, on the other hand, skewed these mature SHIP +/+  79  or SHIP-/- BMMs to an M2 phenotype and did so to a similar degree (Rauh et al., 2005). These findings with mature BMMs suggested that MP and human serum did not contain sufficient levels of IL-4 to induce an M2 program, a result we've confirmed by ELISA (data not shown). Our discovery that basophils generated IL-4 when IL-3 was added to our SHIP-/- BMM cultures (Kuroda et al., 2011) raised the possibility that basophil production of IL-4 may also play a role in MP-induced SHIP-/M M2-skewing during differentiation. Furthermore, since we found in the previous chapter that, of all the factors tested, only IL-4 led to the generation of Arg1+ M2 Ms, we were interested in testing the IL-4-dependence of SHIP-/- M2 BMMs generated in our BMENV assay. Using a neutralizing -IL-4 Ab, we found that, to an ever greater degree than observed for IL-3-driven SHIP-/- BMMs (Kuroda et al., 2009), MP-driven M2-skewing of SHIP-/- BMMs in the BMENV assay was dependent on IL-4, as can be seen by the marked reduction in arginase activity (Fig 4.1 A) and Arg1 expression in MP-induced SHIP-/- BMMs derived in the presence of this neutralizing -IL-4 Ab (Fig 4.1 B). Because the differences between the IL-3stimulated cultures with control Ab and neutralizing -IL-4 Ab wasn't immediately obvious, we compared their densitometry ratios (Arg1/SHC) and confirmed that, indeed, the neutralizing -IL-4 Ab reduced Arg1 expression in IL-3 stimulated BMM cultures (Fig 4.1 B).  80  Figure 4.1 MP-induced skewing of SHIP-/- BM during differentiation to M2 BMMs is dependent on IL-4. -/A) Arginase activity of SHIP BMMs derived ± IL-3 (10 ng/mL) or MP (8% final volume) ± a neutralizing Ab to IL-4 (2.5 g/mL). All data are means ± SEM of experimental triplicates, representative of 3 independent experiments. B) Western blots of the same cells. All lanes come from the same gel and the same time exposed film. Dotted lines indicate where irrelevant lanes have been cropped out. Numbers indicate densitometry ratios (Arg1/SHC) for those lanes. Results are representative of 3 independent experiments. *** = p ≤ 0.001 compared to MP-derived cells without neutralizing Ab. 4.2.2  Stimulated Basophils Secrete IL-4 and are Required for M2-  Skewing in vitro Since IL-4 was necessary for the skewing of SHIP-/- BM to M2 BMMs in our BMENV assay, and we had shown previously that DX5+ cells (i.e., basophils and basophil progenitors) were the sole producers of IL-4 in IL-3-cultured BM cells (Kuroda et al., 2011), we next wanted to test whether or not DX5 + cells were sufficient and necessary for the MP-induced M2-skewing of BMMs in our BMENV assay. To do this we fractionated adh- BM into DX5-enriched (DX5+) (i.e., basophil containing) and DX5-depleted (DX5-) BM using EasySep, and cultured DX5- BM with or without the DX5+ (basophil-containing) fraction in the presence or absence of IL-3 or MP. We found that neither MP nor IL-3 was capable of M2-skewing SHIP+/+ or SHIP-/- DX5- BM, consistent with the need for basophils for M2 skewing. However,  81  after re-introducing the DX5+ fraction, IL-3 stimulation was sufficient for the generation of M2 Ms from adh- BM of either genotype, and, importantly, MP stimulation was sufficient for M2 M generation in SHIP-/- but not SHIP+/+ adh- BM (Fig 4.2).  Figure 4.2 DX5+ BM cells are required for MP- or IL-3-mediated M2-skewing of SHIP-/- BM. + Western blots of BMMs derived from DX5 adh BM ± DX5 cells, ± MP (8%) or IL-3 + (10 ng/mL). The ratio of DX5 to DX5 cells was 10:1 in this experiment. Equal amounts of protein were loaded into each lane (assessed with a Bradford assay), except for the first two lanes, which had very little protein. This experiment is representative of 2 independent experiments. Parts of this figure have been published, and are reproduced here with permission from The Journal of Immunology and The American Association of Immunologists (Kuroda et al., 2011). Since only SHIP-/- and not SHIP+/+ Ms were M2 skewed when derived in vivo, or with MP in our in vivo mimicking BMENV assay, we also wanted to see if SHIP -/DX5+ cells were more potent, by virtue of their increased sensitivity, to stimulusinduced IL-4-production (Kuroda et al., 2011), than SHIP+/+ DX5+ cells at activating the M2 phenotype. Furthermore, we wanted to see if there was a difference between SHIP+/+ and SHIP-/- DX5+ BM-derived basophils (BMBs) in their response to MP, and if such a difference may account for MP being an effective M2-skewing agent only in SHIP-/- BM. To test this, we used IL-3-cultured BM cells, as before (Kuroda et al., 2009; Kuroda et al., 2011), and selected for DX5+ BMBs using EasySep. We found that SHIP-/- BMBs, indeed, produced more IL-4 upon stimulation with IL-3 compared to their SHIP+/+ counterparts, and that while IL-3 elicited increased IL-4 production from BMBs of both genotypes, MP elicited an increase in IL-4 production only from SHIP-/- BMBs (Fig 4.3 A). We then co-cultured in vitro-derived SHIP+/+ and SHIP-/-  82  BMBs with mature M-CSF-derived SHIP+/+ BMMs and found that with either of these BMBs present, IL-3 M2-skewed co-cultured SHIP+/+ BMMs (Fig 4.3 B, top panel), but this did not occur in the absence of BMBs (Fig 4.3 B, lower panel). This was consistent with IL-3 eliciting a sufficient production of IL-4 from both SHIP+/+ and SHIP-/- BMBs for M2 skewing of SHIP+/+ mature BMMs. Interestingly, however, even though MP did not increase IL-4 production from SHIP+/+ BMBs beyond that of unstimulated BMBs (Fig 4.3 A), MP-stimulated BMBs of both genotypes skewed cocultured mature WT BMMs to a similar degree (Fig 4.3 C). Nevertheless, we established that DX5+ cells, when IL-3 or MP stimulated, skewed co-cultured mature SHIP+/+ BMMs to an Arg1+ M2 phenotype.  83  Figure 4.3 Although MP stimulates IL-4 production from SHIP-/-, but not SHIP+/+, BMBs, it M2-skews Ms in mature BMMs co-cultured with either SHIP+/+ or SHIP-/- BMBs. 4 A) IL-4 production by mature BMBs (2 x 10 cells/100 L) at 24 h after stimulation with either IL-3 (10 ng/mL), or MP (10%). The data, pooled from three independent experiments, are represented as means ± SEM of experimental triplicates. B) (Top +/+ 5 Panel) Western blot of a co-culture of mature SHIP BMMs (2.5 x 10 cells) and +/+ -/4 SHIP or SHIP BMBs (2 x 10 cells) for 72 h ± IL-3 (10 ng/mL). IL-4 (10 ng/mL) was used as a positive control. Cultures were thoroughly washed to remove nonadherent BMBs before BMM lysates were prepared. (Bottom Panel) Western blots of mature BMMs, not co-cultured with BMBs, and stimulated with IL-3 (10 ng/mL) +/+ 5 or IL-4 (10 ng/mL) for 72 h. C) Western blot of mature SHIP BMMs (2.5 x 10 ) +/+ -/4 co-cultured with SHIP or SHIP BMBs (2 x 10 ) for 72 h ± MP (5%). Cultures were thoroughly washed to remove BMBs, which are non-adherent, before BMM lysates were prepared. All lanes represented in each panel come from the same gel and time exposed film. Dotted lines indicate where irrelevant lanes have been cropped out. Results are representative of 2 independent experiments. * = p ≤ 0.05 compared to control. † = p > 0.05 compared to control.  84  4.2.3  MP Does Not Stimulate IL-4 Secretion from Intact Adh- BM Cells  Although our results to this point suggested that MP triggers substantially more IL-4 production from SHIP-/- BMBs than from their SHIP+/+ counterparts, these experiments were carried out with arbitrary ratios of BMBs:BMMs. Also, we were concerned that the increased IL-4 production from SHIP-/- BMBs had no practical impact on the M2-skewing of co-cultured BMMs. Thus, we wanted to use intact adh- BM to test whether or not MP stimulation leads to IL-4 production and whether this accounts for the M2-skewing of SHIP-/- BMMs in our BMENV assay. To investigate this, we cultured maturing SHIP+/+ and SHIP-/- adh- BM cells with M-CSF alone, or in combination with IL-3 or MP and measured IL-4 levels in the conditioned medium. To our surprise, only IL-3 induced a substantial increase in IL-4 in the culture supernatants of maturing BMMs, even though both IL-3 and MP were sufficient to generate Arg1+ M2 Ms from SHIP-/-BM (Fig 4.4).  Figure 4.4 MP induces Arg1 in differentiating SHIP-/- BMMs without stimulating a significant increase in IL-4. (Top panel) IL-4 levels (means ± SEM of experimental duplicates) in culture supernatants after 48 h from M-CSF-cultured adh BM ± MP (10%) or IL-3 (10 ng/mL), and (bottom panel) Western blots of corresponding mature BMMs harvested at the end of the BMENV assay. All lanes represented in each panel come from the same gel and time exposed film. These results are representative of 3 independent experiments. ** = p ≤ 0.01 compared to control. † = p > 0.05 compared to control. 85  4.2.4  MP Does Not Increase IL-4 Production More than M-CSF from  DX5+ SHIP-/- BM Cells Although we did not see an increase in IL-4 secreted by MP-stimulated SHIP-/adh- BM cultures compared to those stimulated with M-CSF alone, the low concentration of IL-4-producing DX5+ cells in BM (Kuroda et al., 2009) may have obscured a slight increase in IL-4 production sufficient for M2-skewing. To test this, we separated adh- BM into DX5+ and DX5- fractions as before, and cultured the DX5+ cells at a higher cell concentration than that found in intact BM to better examine their response to stimuli. We measured IL-4 levels in cell culture supernatants of these DX5+ and DX5- BM cells stimulated with M-CSF alone, or together with IL-3 or MP, as we would in our BMENV assay, 24 h after the addition of the stimulus. We found that M-CSF-stimulated adh- SHIP+/+ and SHIP-/- DX5+ cells secreted more IL-4 upon IL-3 stimulation than when stimulated with M-CSF alone, and in SHIP+/+ cells, MP increased IL-4 production modestly over M-CSF alone (Fig 4.5). SHIP-/- cells, on the other hand, secreted an equally high level of IL-4 with MCSF alone as with M-CSF + MP, even though, as seen in Figure 4.4, only MP treated cultures caused M2-skewing of maturing Ms (Fig 4.5).  86  Figure 4.5 MP does not stimulate more IL-4 production from SHIP-/- DX5+ cells then M-CSF alone. + IL-4 production (mean ± SEM of experimental triplicates) from DX5 and DX5 adh BM cells at 24 h after stimulation with IL-3 (10 ng/mL) or MP (5%). These results are representative of 2 independent experiments. * = p ≤ 0.05 compared to M-CSF control. +/+ ¥ = p ≤ 0.05 compared to the SHIP cells under the same condition. † = p > 0.05 compared to M-CSF control. 4.2.5  MP Drives SHIP-/- M2 Skewing Via a Distinct Mechanism from IL-3  Since IL-3 was a potent stimulator of IL-4 secretion from in vitro-derived BMBs and DX5+ BM, and especially from SHIP-/- BM, we wanted to test if MP-induced M2skewing was due to IL-3 in MP. This, however, seemed unlikely given the different abilities of IL-3 and MP to elicit IL-4 production from adh- and DX5+ BM, shown in Figures 4.4 and 4.5, and the fact that IL-3 in MP was not present at detectable levels using a standard ELISA (data not shown). Nevertheless, the possibility remained that IL-3 may be present at very low but sufficient levels for inducing low levels of IL4 that M2-skew SHIP-/- adh- BM into BMMs. To test this, we added neutralizing -IL-3 Abs to our MP-induced SHIP-/- BMM cultures and found that although it reduced IL-3-induced IL-4 secretion from BMBs (Fig 4.6 A), -IL-3 did not reduce MP-induced M2-skewing of SHIP-/- BMMs (Fig 4.6 B). We obtained similar results using a different IL-3 neutralizing Ab (data not 87  shown). To further compare IL-3 and MP in our BMENV assay, we compared their abilities to M2-skew maturing SHIP-/- BMMs from adh- and lin- BM. The addition of IL-3, in keeping with previous results (Kuroda et al., 2009), was sufficient for the generation of Arg1+ M2 Ms from SHIP-/- adh- and lin- BM, while MP was only capable of M2 skewing SHIP-/- adh- BM (Fig 4.6 C). This suggested that, unlike IL-3, MP only acted on more mature M progenitors and/or BM basophils.  Figure 4.6 IL-3 is not responsible for the M2-skewing activity of MP. 4 A) IL-4 levels (mean ± SEM of assay duplicates) from BMBs (4 x 10 cells /200 L) cultured for 24 h in the presence of IL-3 (10 ng/mL) ± neutralizing Ab to IL-3 (2.5 -/g/mL). B) Arginase activity (mean ± SEM of experimental triplicates) of SHIP BMMs derived in M-CSF ± MP (8%) ± neutralizing Ab to IL-3 (2.5 g/mL) or IL-4 -/(2.5 g/mL). C) Western blots of SHIP BMMs derived in M-CSF from either lineage-depleted (lin ) or adherence-depleted (adh ) BM ± IL-3 (10 ng/mL) or MP (5%). All lanes represented in the figure come from the same gel and exposed film. Dotted lines indicate where irrelevant lanes have been cropped out. These results are representative of 3 independent experiments. * = p ≤ 0.05 compared to its respective control. ** = p ≤ 0.01 compared to its respective control. *** = p ≤ 0.001 compared to its respective control.  88  4.2.6  IgG, but not IgE, is Necessary for MP-Induced M2-Skewing of  SHIP-/- BM Apart from IL-3, we previously reported that IgE also stimulates IL-4 production from basophils, especially in SHIP-/- BMBs and in SHIP-/- DX5+ BM, and that it is an important mediator of SHIP-/- TH-2 skewing in vitro and in vivo (Kuroda et al., 2011). We, therefore, examined the role of IgE in MP-induced M2-skewing in vitro. To do so, we first added IgE to our SHIP-/- BMM derivation cultures and found that it dose-dependently increased the arginase activity of the derived BMMs (Fig 4.7 A). Unexpectedly, however, we found that MP depleted of IgE induced higher arginase activity than SHIP-/- BMMs derived with intact MP (Fig 4.7 B & C). IgE depletion of MP using a different method (i.e., biotinylated IgE-specific Abs and streptavidin beads) showed similar results (data not shown).  89  Figure 4.7 IgE is not responsible for MP-induced M2-skewing of SHIP-/- BM. -/A) Arginase activity (mean ± SEM of experimental triplicates) of SHIP BMMs derived in vitro with M-CSF ± IgE at 2 doses. B) The levels of IgE of MP (50% diluted in culture medium and filter sterilized) ± IgE-depletion determined by ELISA (results from pool MP-2 are shown) and C) the arginase activity of SHIP-/- BMMs derived with the addition of 2 pools of MP (5%). MP-1 and MP-2 represent two distinct MP pools, which were independently depleted of IgE using anti-mouse IgE Abs on Protein G-coupled agarose beads. These results are representative of 3 independent experiments. * = p ≤ 0.05 compared to any other condition. ** = p ≤ 0.01 compared to any other condition.  90  We next tested IgG in a similar manner and found that IgG, unlike IgE (Kuroda et al., 2011), did not stimulate IL-4 production from BMBs (Fig 4.8 A). Despite this, however, when IgG was added during the maturing of SHIP -/- BMMs, it dosedependently induced arginase activity in the derived mature BMMs (Fig 4.8 B). Interestingly, unlike IgE, the depletion of IgG from MP markedly reduced the ability of MP to stimulate arginase activity in SHIP-/- BMMs during differentiation, and adding IgG back to IgG-depleted MP increased its potency at inducing arginase activity (Fig 4.8 C & D). The same MP pool from Fig 4.8 D, analyzed for IgG levels by ELISA, demonstrated that the depletion was effective, and that MP contained levels of IgG in the g range (Fig 4.8 E), justifying our use of IgG in our assays at this concentration.  91  Figure 4.8 IgG in MP induces M2-skewing of SHIP-/- BM. 5 A) IL-4 production from SHIP+/+ and SHIP-/- BMBs (1 x 10 cells/mL) after 24 h ± IgG -/(1 g/mL) or IL-3 (10 ng/mL). B) Arginase activity of SHIP BMMs (mean ± SEM experimental triplicates) derived with the addition of 1 g/mL (IgG), 10 g/mL IgG Hi +/+ (IgG ). C) Arginase activity (mean ± SEM experimental triplicates) of SHIP and -/SHIP BMMs cultured with the addition of MP (5%) or 5% IgG-depleted MP (IgG -/MP). D) From the same experiment as (B), arginase activity of SHIP BMMs (mean ± SEM experimental triplicates) derived with the addition 5% MP, 5% IgGdepleted MP (IgG MP), or 5% IgG MP + 1 g/mL IgG (+IgG). E) IgG levels determined by ELISA (mean ± SEM of assay duplicates) in MP (50% diluted in culture medium and filter sterilized) ± IgG depletion. These experiments are representative of 3 independent experiments. † # * **  = p > 0.10 compared to control. = p ≤ 0.10 compared to M-CSF control. = p ≤ 0.05 compared to M-CSF control. = p ≤ 0.01 compared to IgG-depleted MP condition.  92  4.2.7  MP and IgG Sensitize BM Progenitors and BMMs to IL-4  Thus far, we have demonstrated that in SHIP -/- BM, DX5+ cells are the sole producers of the IL-4 necessary in IL-3 or MP-induced M2-skewing. However, MPinduced skewing, which we have shown occurs via an IL-3-independent mechanism, appears capable of skewing maturing SHIP-/- BM cells into M2 Ms without increasing the levels of IL-4 in the supernatant beyond those triggered by M-CSF alone (which does not M2 skew). We thus hypothesized that if IL-4 was necessary and sufficient, and MP addition led to the generation of M2 Ms without increasing IL-4 levels over that generated by M-CSF alone, then perhaps MP was enhancing the sensitivity of SHIP-/- M progenitors to IL-4. Furthermore, since MP only M2 skewed SHIP-/- BM, perhaps this increased sensitivity only occurred with SHIP-/- BM cells. To test these possibilities, we first compared the sensitivity to IL-4 of SHIP+/+ and SHIP-/- BM progenitors and found that, similar to what was found with mature BMMs in the previous chapter, SHIP-/- lin- progenitors were not more sensitive to IL-4 (Fig 4.9 A). Moreover, when DX5- BM cells were cultured in IL-4, the addition of MP increased Arg1 protein expression in BMMs derived from both SHIP+/+ and SHIP-/- BM to a similar degree (Fig 4.9 B). As well, we found that mature SHIP-/BMMs were not more sensitive to M2 skewing with IL-4 + MP than SHIP+/+ mature BMMs (Fig 4.9 C). In fact, our data suggest the opposite to be true (Fig 4.9 C).  93  Figure 4.9 MP increases the responsiveness of DX5- BM cells and mature BMMs to IL-4. A) Western blots of BMMs derived from lin BM progenitors with increasing doses of IL-4. B) Western blots of BMMs derived from DX BM in M-CSF + IL-4 (5 ng/mL) ± MP (5%). C) Western blots of mature BMMs stimulated for 72 h with increasing doses of IL-4 and MP. All lanes contained equal amounts of protein (assessed by BCA assay). These results are representative of 2 independent experiments. (A) is a figure reprinted with permission from The Journal of Immunology and the American Association of Immunologists (Kuroda et al., 2009). Since we had shown that IgG was sufficient to M2 skew SHIP -/- BM and was necessary in MP-induced M2-skewing of SHIP-/- BM, we then tested the ability of IgG to sensitize cells to IL-4. When we added IgG to mature BMMs, we found that both SHIP+/+ and SHIP-/- BMMs were more sensitive to IL-4 stimulation, as evidenced by an increase in Arg1 protein expression compared to mature BMMs stimulated with the same dose of IL-4 without IgG (Fig 4.10 A). To test if IgG in MP was the sensitizing factor during BMM maturation, we derived BMMs with IL-4 ± MP or IgG-depleted MP from DX5- BM cells (to not confound our results with the possibility of IL-4 produced by DX5+ cells). We found that while MP sensitized DX5BM to IL-4 during their maturation into BMMs in both SHIP+/+ and SHIP-/- cells, IgG-  94  depleted MP sensitized cells to IL-4 to a lesser degree (Fig 4.10 B). Importantly, as shown in the samples treated with 0.5 ng/ml IL-4, SHIP-/- BMMs were more M2skewed in the presence of MP but not IgG-depleted MP (Fig 4.10 B).  Figure 4.10 IgG sensitizes BMMs to IL-4 during and after maturation. +/+ -/A) Western blots of mature SHIP and SHIP BMMs stimulated for 72 h with IgG +/+ -/(0.1 g/mL) ± IL-4. B) Western blots of SHIP (left panel) and SHIP (right panel) BMMs derived from DX5 BM with IL-4 alone (C), or with IL-4 + 5% MP or 5% IgGdepleted MP(MP-G). Equal amounts of protein were loaded in each lane. These results are representative of 2 independent experiments. 4.2.8  M2-skewing is induced by DX5+ BM-produced IL-4 and MP-  induced Sensitization of BMM Progenitors Our experiments above suggested that the ability of MP to skew SHIP -/- BM to M2 BMMs during maturation was due to its ability to sensitize BMMprogenitors to IL-4. However, our results also showed that SHIP+/+ progenitors have a similar capacity to be sensitized to IL-4 by MP, yet SHIP+/+ BMMs derived in the presence of MP do not become Arg1+ M2 Ms. Since SHIP-/- DX5+ BM cells constitutively produce IL-4, which isn't the case with SHIP+/+ DX5+ BM cells (Fig 4.5), we 95  hypothesized that during BMM derivation, SHIP-/- DX5+ BM cells produce the requisite IL-4 to enable MP-sensitized BM cells to become Arg1+ M2 Ms. To test this, we derived BMMs from fractionated BM, mixing SHIP+/+ or SHIP-/DX5+ cells with SHIP+/+ or SHIP-/- DX5- cells. To ensure that no SHIP-/- cells contaminated the SHIP+/+ BM, we used a transwell system to prevent contact, separating the DX5+ cells above the DX5- cells with a cell-impermeable membrane. We found that SHIP-/- DX5+ cells induced M2 skewing of both SHIP+/+ and SHIP-/DX5- BM in the presence of MP, while SHIP+/+ DX5+ cells did not skew either genotype of DX5- BM, even in the presence of MP (Fig 4.11). Also of note, SHIP-/DX5+ BM was capable of stimulating some M2 skewing of SHIP -/- DX5- cells in the absence of MP, but this was greatly enhanced with MP.  Figure 4.11 SHIP-/- DX5+ cells enable SHIP+/+ or SHIP-/- DX5- BM to be skewed to an M2 phenotype during BMM differentiation. + Western blots of BMMs derived from DX5 BM in a transwell system with DX5 + cells co-cultured in a contact-independent manner. N indicates the absence of DX5 cells. Equal amounts of protein were loaded onto each lane and SHC was used as a loading control. 4.2.9  Targeting the Basophil-IL-4 Axis Does Not Reduce SHIP-/- M2-  Skewing in vivo Since we had established the importance of basophils and IL-4 to M2-skewing, we wanted to test if this could be manipulated in vivo to reduce M2-skewing in SHIP/-  mice. We first tried neutralizing Abs to IL-4, which we injected IP into SHIP-/- mice.  After a 2 week regimen, we harvested PMs and measured their arginase activity to determine the degree of M2-skewing. Despite the effectiveness of IL-4 inhibition in vitro, we found that our neutralizing IL-4 Ab treatment was insufficient to reduce M2-  96  skewing to any significant degree compared to isotype control Ab-injected mice (Fig 4.12 A). As an alternate strategy, we decided to deplete basophils, since they appeared to be the major producers of IL-4. Related to this, Rivera's group had just found that by targeting the IgE receptor, FcRI (present on mouse mast cells and basophils), with the IgE receptor Ab, MAR-1, they could deplete basophils in vivo, and they showed that this reduced a Toxoplasma gondii-induced TH-2 response (Charles et al., 2009). Employing a similar strategy, we injected our SHIP-/- mice intravenously (IV) with MAR-1 but found it caused our SHIP-/- mice to go into anaphylactic shock, as indicated by a marked drop in body temperature (data not shown). We postulated that because SHIP-/- mast cells (MCs) were hypersensitive to various stimuli (Gimborn et al., 2005; Huber et al., 1998), this Ab might be causing unwanted MC activation, resulting in anaphylactic shock. To avoid this, we used MC-deficient Wsh/Wsh mice with a Cre-lox inducible deletion of SHIP under the control of a broad viral response element, the Mx1 promoter. We injected poly I:C into Cre-expressing mice homozygous for a SHIP1 gene flanked by lox sequences. This gave us a MCfree SHIP-/- mouse, which we could treat with MAR-1 Ab injections. Similar to the results of Charles et al. (2009), we found that basophil depletion reduced the T H-2 skewing of splenic T cells, as evidenced by reduced IL-4 production from -CD3 Abactivated spleen cells ex vivo from the basophil-depleted SHIP-/- mice (Fig 4.12 B). Intriguingly, even though TH-2 skewing was reduced, the M2-skewing of the PMs was not (Fig 4.12 C).  97  Figure 4.12 Targeting the basophil-IL-4 axis in vivo does not reduce M2 skewing. +/+ -/A) Arginase activity of PMs from SHIP and SHIP mice injected twice/week with a neutralizing -IL-4 Ab (500 g/mL) or an isotype control Ab at the same dose for 2 weeks. B) IL-4 secretion from splenocytes, stimulated for 18 h with -CD3 Ab, from +/+ -/mast cell deficient (Wsh/Wsh) SHIP and SHIP mice ± basophil depletion with FcR Ab (MAR-1) injection regimens. C) Western blots of PMs harvested from the -/-  SHIP mice in panel (B). Panel (B) was reproduced with permission from The Journal of Immunology and the American Association of Immunologists (Kuroda et al., 2011). † = p > 0.10 compared to the isotype control. * = p ≤ 0.05 between compared groups.  98  Since we had shown in the previous chapter that the IL-4-induced Arg1+ M2 phenotype was reversible with cytokine withdrawal, we were curious to see if the SHIP-/- PM Arg1+ phenotype could be similarly reversed. Interestingly, we found that while the IL-4-induced Arg1+ M2 M phenotype of SHIP-/- BMMs was reduced with cytokine withdrawal (Fig 4.13 A), the Arg1+ M2 phenotype of SHIP-/- PMs, could not be reversed, even in the presence of an neutralizing -IL-4 Ab (Fig 4.13 B).  Figure 4.13 The SHIP-/- peritoneal Arg1+ M phenotype is not reversible. +/+ -/A) Western blots of SHIP and SHIP BMMs cultured for 12 days ± IL-4 (10 ng/mL), or with IL-4 for 6 days, washed and cultured without IL-4 for the remaining 6 -/days. B) Western blots of SHIP PMs cultured ex vivo for 18 h and lysed (D0), or cultured for 6 days in 2.5 g/mL non-specific Ab (C), IL-4-neutralizing Ab (4), or culture medium alone. Equal protein was loaded onto each lane. These results are representative of 3 independent experiments. 4.3  Discussion In this chapter, we sought to identify the factors that influence in vivo M  skewing, in particular those that skew Ms to an M2 phenotype in the SHIP-/mouse. Another specific aim was to elucidate the differences between SHIP +/+ and SHIP-/- cells which contribute to M2-skewing in SHIP-/- but not SHIP+/+ BMMs. And lastly, we wanted to exploit whatever we found to see if we could reverse the in vivo M2 skewing of the SHIP-/- mouse.  4.3.1  IL-4-Producing DX5+ Cells are Necessary and Sufficient for M2-  Skewing In this chapter we have attempted to identify the cell-types in SHIP-/- BM necessary for M2-skewing by MP, and whether or not they mediate skewing through  99  IL-4 production. The IL-4-dependence of MP-induced M2-skewing of SHIP-/- BMMs during differentiation suggested that perhaps, similar to IL-3-induced M2-skewing (Kuroda et al., 2009), MP also requires the secretion of IL-4 by BM cells to mediate their M2-skewing effects. In our BMB experiments, we demonstrated that activated, IL-4 producing BMBs are sufficient to generate Arg1+ M2 Ms from mature BMMs in a co-culture system. Similarly, DX5+ BM cells, which our data indicated were the sole producers of IL-4 in BM, were absolutely critical to the M2-skewing of not only IL-3- and MPstimulated SHIP-/- BM but with IL-3-stimulated SHIP+/+ BM as well. These experiments highlighted the importance of DX5+ cells in M2 programming. While we demonstrated previously (Kuroda et al., 2011) that only basophils (FcRI+ c-kit- cells) produce IL-4 upon stimulation from BMB cultures, BM is a mixture of cells, including immature NK cells, NKT cells, and memory T cells, all of which can express CD49b (the Ag detected by the DX5 Ab) (Gregoire et al., 2007; Han et al., 2010). Of note, BM-associated T cells and NKT cells are avid producers of IL-4, even in the absence of stimuli (Kronenberg and Gapin, 2002; Monteiro et al., 2005) and our experiments do not discriminate between these IL-4-producing subsets. To determine which of these cell-types are most crucial to the M2-skewing of BM, other selection methods will have to be employed to complement DX5 selection, such as selecting for the NK markers, NK1.1 or NKG2D, to determine the role of NKT cells. On a related note, IL-3- and MP-induced M2-skewing of SHIP-/- BM differ in various ways, but especially in regard to the M2-skewing of lin- and adh- BM. In our previous work, we identified basophil progenitors as the responders to IL-3 in linBM, which mediate M2-skewing (Kuroda et al., 2009). Since IL-3 not only enhances the survival of basophils but also acts as a stimulator and maturation factor, it is possible that basophil progenitors in lin- BM are being stimulated to survive, mature, and secrete IL-4 in response to IL-3. In the case of MP, however, there is no IL-3 to stimulate the maturation of basophils, so perhaps the constitutive level of IL-4 production in the basophil progenitors contained in lin - BM is insufficient for MPinduced M2-skewing, while the adh- BM pool, which contains mature basophils, contains more cells constitutively producing IL-4 to sufficiently M2-skew Ms. 100  4.3.2  MP-Induced M2-skewing of SHIP-/- BM is Mediated not by an  Increase in IL-4 Levels, but an Increase in IL-4 Sensitivity In this chapter, we have shown that an increase in IL-4 is not always necessary for M2-skewing, as demonstrated by secreted IL-4 levels of MP-induced cultures. In fact, this unique feature of MP-driven M2 M generation from SHIP-/- BM distinguishes them from IL-3-induced M2 M generation, which triggers a marked increase in secreted IL-4. Our data further suggest that instead of increasing IL-4 production, MP increases the sensitivity of progenitor cells or mature BMMs to IL4, allowing these cells to become M2 Ms with a dose of IL-4 that would normally not be sufficient. IL-13 is a TH-2 cytokine similar to IL-4, which can induce Arg1 and an M2 phenotype in Ms even if IL-4 is absent (Doyle et al., 1994). While we did not specifically test whether it plays a role in MP-induced M2-skewing, it likely does not since an IL-4 neutralizing Ab completely abrogated M2-skewing by MP. Another explanation for the absence of an IL-4 increase in culture supernatants of MPstimulated SHIP-/- BMM cultures is that the rate of IL-4 consumption may be concurrently increased with production, resulting in no net increase of IL-4 accumulation compared to control (M-CSF only) cultures. Our experiments did not account for this possibility, but whether or not MP induces an increase in IL-4 consumption by BM cells can be determined by culturing adh- SHIP-/- BM with an IL4R antagonist, preventing its uptake, or by measuring IL-4 levels in cultures of DX5adh- SHIP-/- BM given a fixed amount of IL-4 with or without MP. Nevertheless, given both SHIP-/- DX5+ BM cells alone and adh- SHIP-/- BM cultures did not show increased IL-4 production after MP stimulation strongly suggests that MP does not increase IL-4 production beyond that of M-CSF treated cells. Nonetheless, as shown in Fig 4.6, SHIP-/- DX5+ cells spontaneously secrete far higher IL-4 levels than their SHIP+/+ counterparts.  101  4.3.3  IgG in MP Promotes M2 Skewing by Increasing the Sensitivity of  BM progenitors to IL-4 Having discovered that MP was inducing M2-skewing in SHIP-/- adh- BMMs during differentiation by sensitizing these cells to IL-4, we carried out studies to identify the factors in MP responsible for this and found BMMs primed with IgG exhibited a more robust response to IL-4 than those not treated with IgG, and IgGdepleted MP displayed a marked drop in its ability to sensitize DX5-, non-IL-4producing, BM to IL-4. We used protein G, which is a bacterial IgG binding protein, to deplete IgG (Akerstrom and Bjorck, 1986). Because protein G is selective for binding IgG, showing only very weak or no affinity for other major Ig species in MP, such as IgA or IgM (Sant'Anna et al., 1985; Pilcher et al., 1991), we did not determine if IgA or IgM levels were affected with protein G treatment. Even so, our IgG add-back experiments showing that IgG-depleted MP regains its ability to skew SHIP-/- adh- BM to M2 BMMs after IgG re-incorporation is strong evidence that IgG is the important factor in MP induced M2-skewing. Because of its ability to trigger IL-4 production (Kuroda et al., 2011), IgE is a potentially important Ig species in plasma. Our IgE-depletion experiments, however, indicate that IgE is not a necessary component of MP-induced SHIP-/- M2-skewing and its depletion may, in fact, be increasing the potency of MP to M2-skew SHIP-/BM in vitro. This seems contradictory at first glance, since we also show that IgE alone can skew SHIP-/- BMMprogenitors to an M2 phenotype in a dose-dependent manner. However, the concentrations of IgE we tested were approximately 1000 fold higher than the levels known to be present in MP under homeostatic conditions (Kuroda et al., 2011; Lehrer, 1976). Although lower concentrations of IgE have not been tested, it is likely that concentrations of IgE present in MP would be too low to M2-skew SHIP-/- Mprogenitors. Our finding that IgE and IgG depletions gave opposing results, despite both factors being sufficient for M2-skewing, is interesting and may be related to the affinity of mouse IgE for the inhibitory receptor, FcRIIB. IgG signals primarily through the activating receptor, FcRI, since this receptor has a 100 fold higher affinity for IgG compared to other FcR's (Ravetch and Bolland, 2001). While IgE 102  signals primarily through FcRI, which is found on basophils and MCs, but not Ms or their progenitors, it has also been shown to bind to the inhibitory receptor, Fc RIIB (Takizawa et al., 1992; Nimmerjahn and Ravetch, 2007; Hirano et al., 2007). Thus, the removal of IgE and IgG from MP may be the equivalent of removing inhibitory and activating signals, respectively. Though we have not done the experiments to show this, the possibility that IgE in MP is inhibiting IgG-mediated sensitization is consistent with our data.  4.3.4  Differences Between SHIP+/+ and SHIP-/- BM Contributing to Their  Difference in M2-Skewing Behavior in vitro One of our specific aims was to determine why only SHIP-/- BM was M2 skewed with MP and we were surprised to find that MP did not induce a significant increase in IL-4 in vitro and that both SHIP+/+ and SHIP-/- DX5- BM and BMMs could be sensitized by MP for a more intense IL-4 response. In these respects, SHIP+/+ and SHIP-/- mature and maturing BMMs behaved almost identically. The only major difference we detected was that SHIP-/- DX5+ BM produced IL-4 constitutively, while their SHIP+/+ counterparts did not. In an earlier publication (Rauh et al., 2005), we hypothesized that there was an inherent difference between differentiating SHIP-/- BMM progenitors and mature Ms in their response to MP. While we started the work in this chapter with a similar hypothesis, our finding that MP skews via sensitization of cells to IL-4, which is something that occurs in both SHIP-/- DX5- M progenitors and mature BMMs, demonstrates that there is no difference between the innate ability of differentiating BMM progenitors and mature BMMs to respond to MP. In fact, the whole of our data suggests that the only reason why SHIP-/- BM, but not mature BMMs, respond to MP is because the latter population of cells lack the DX5 + IL-4 producers to supply the requisite IL-4 to skew MP-sensitized BMMs. Furthermore, our MP-stimulated SHIP+/+ DX5- + SHIP-/- DX5+ BM transwell studies demonstrated that SHIP-/- DX5+ cells, by virtue of their constitutive IL-4 production, supply the requisite IL-4, which in the presence of MP, for the M2-skewing of maturing BMMs, SHIP+/+ and SHIP-/-  103  alike. This suggests that the reason that SHIP+/+ BM does not skew to M2 in response to MP in vitro is because its DX5+ fraction, which lacks constitutive IL-4 production, does not produce sufficient IL-4 to skew SHIP+/+ M progenitors even with MP-induced IL-4 sensitization. This supports the hypothesis that the SHIP-/- is M2-skewed in vivo because its DX5+ cells constitutively produce IL-4 sufficient for M2-skewing MP-sensitized M progenitors in vivo, but at levels too low to detect by ELISA in MP. A model summarizing our findings is presented in Figure 4.14.  104  Figure 4.14 Model of M2-skewing in maturing Ms in vivo and in vitro: Roles of SHIP and basophil/basophil progenitors on M progenitors. +/+ -/Proposed model of M2-skewing, in vivo and in vitro, of SHIP and SHIP maturing Ms, summarizing the results from chapter 4. Solid lines represent progenitors differentiating into Ms after the influence exerted by factors added or secreted by cells represented by dotted lines. Black lines represent interactions discovered through BMENV assays and proposed as a model of the in vivo interactions. Gray lines represent stimuli supplemented in vitro, which M2 skew cells. 105  4.3.5  Reversing SHIP-/- M2-Skewing in vivo  Given the importance of basophils and IL-4 in TH-2 skewing, as indicated both by our earlier work (Kuroda et al., 2011) and that of Rivera‘s group (Charles et al., 2009), it was disappointing that basophil-depletion with MAR-1 did not reduce the M2 phenotype of SHIP-/- in vivo-derived PMs. While the ineffectiveness of basophil-depletion may indicate that basophils are simply not an important part of SHIP-/- M2-skewing in vivo, since, as stated above, other DX5+ IL-4-producing cell types may be present in vivo to produce the requisite IL-4, our data support an alternative explanation. Specifically, since PMs are long lived (49 - 100 days) (Takahashi, 2000; Melnicoff et al., 1988), and our depletion procedure spanned only ~30 days, it is conceivable that although the supply of IL-4 was reduced, the PMs, which were already M2-skewed, linger in the peritoneum. Another finding in this chapter was that the Arg1+ phenotype of SHIP-/- in vivo derived PMs cannot be readily reversed with cytokine withdrawal, even in the presence of a neutralizing -IL-4 Ab. This suggests that long-lived Ms already skewed to M2 before the treatment regimen would retain their Arg1+ phenotype. Therefore, even if our treatment prevented the generation of new M2 Ms, it is possible that we were seeing the lingering SHIP-/- PMs, which are still Arg1+, in our studies.  106  5. The Role of COX-2 and TGF- Pathways in M2-Skewing of Macrophages 5.1  Introduction In the previous chapter, we discovered, using our BMENV assay, that IL-4 and  DX5+ BM cells, which are most likely basophils or basophil progenitors, play a role in MP-induced M2-skewing of SHIP-/- BM. Our in vivo interventions, however, were not effective. We were, therefore, still looking for factors important to in vivo SHIP-/- M2skewing which we could manipulate as a therapy, for example, to reduce the M2skewing of TAMs.  5.1.1  M2-Skewing in TAMs  A situation in which Arg1+ M2 Ms have been observed in vivo is in mouse TAMs (Sharda et al., 2011; Biswas et al., 2006). Intriguingly, although Arg1+ M2 TAMs are commonplace in tumors, studies evaluating serum cytokine levels in tumor-bearing mice have revealed that while cytokines such as IL-6, MCP-1, IL-10 and TGF- are often elevated (and many are tumor-derived), a similar increase in IL4 or IL-13 has not typically been detected (Tanaka et al., 1993; Cosper and Leinwand, 2011). In fact, while an increase in Arg1+ myeloid cells has been observed in both tumor-bearing mice and in cancer patients (Rodriguez et al., 2005; Sharda et al., 2011; Ochoa et al., 2007a), the primary driver of the M2 M phenotype in model cancer systems may not be IL-4 but other M-stimulating inflammatory mediators, such as PGE2 and macrophage stimulating protein (MSP) (Morrison and Correll, 2002; Rodriguez et al., 2005; Sharda et al., 2011). Furthermore, HIF-1-mediated Arg1 induction has also been observed in TAMs infiltrating into hypoxic areas of tumors (Sica et al., 2008a).  5.1.2  Factors Influencing M2-Skewing  PGE2 is a lipid metabolite of arachidonic acid under the control of two major rate-limiting enzymes: cyclooxygenase (COX) and PG synthase (Williams and Shacter, 1997). There are two major COX enzymes in mice and humans, a  107  constitutive COX-1 and an inducible COX-2 form, the latter of which is upregulated by various immune stimuli (Simmons et al., 2004). PGE2 modulates the immune response in many cell-types, including Ms, where it is known to induce a regulatory phenotype, partially by enhancing the secretion of IL-10 and inhibiting the secretion of IL-12 (Edwards and Emens, 2010). Attesting to its many functions, PGE2 has also been shown to positively regulate the production of NO, which is traditionally associated with an inflammatory M phenotype (Sakata et al., 2011). Moreover, because of the vital role that PGE2 plays in pain and fever, COX enzymes, especially COX-2, have been successfully targeted pharmacologically by a class of COX-inhibitory drugs known as non-steroidal anti-inflammatory drugs (NSAIDs) (Simmons et al., 2004). MSP, which binds the RON receptor, is another immunemodulating factor involved in the resolution of inflammation, with a role in limiting IL12 and NO production by increasing IL-10 secretion in Ms (Kretschmann et al., 2010). It is produced in the liver and secreted into circulation, where it remains inactive until it is cleaved by the enzyme matriptase, which is present on the extracellular surface of Ms and other cells (Kretschmann et al., 2010). TGF- is another interesting factor that is linked to both mouse and human cancers (Bierie and Moses, 2010; Alshaker and Matalka, 2011). Though it is one of many anti-inflammatory factors, TGF- is of particular interest to us because it is also an important factor in SHIP-/- M2-skewing by MP (Rauh et al., 2005). Furthermore, TGF- acts on cancer cells and immune cells alike, and to further complicate things, it can have opposing effects on the same cell type depending on the context (Bierie and Moses, 2010; Alshaker and Matalka, 2011). In terms of Ms, TGF- has generally been demonstrated to be a negative regulator of M inflammatory functions (Ashcroft, 1999). In this chapter, we set out to test if these factors, which have been shown to play a role during in vivo M2-skewing in the literature, are important in SHIP-/- M2-skewing, and more importantly, if they can be exploited to reverse SHIP-/- M2-skewing in vivo.  108  5.2  Results  5.2.1  TGF- is Not Sufficient to Recapitulate Plasma-Induced SHIP-/- M2  MSkewing Since in the previous chapter we showed that the in vivo manipulation of the IL4 axis may not be efficacious, we felt that in order to reverse in vivo SHIP-/- M2skewing, it might be important to identify other key causal elements that contribute to this skewing. We thus set out to do this using our BMENV assay and employing, once again, Arg1 protein and/or activity upregulation as our M2 biomarker. To follow up on our previous results demonstrating that TGF- was necessary for plasmainduced M2-skewing of maturing SHIP-/- BMMs (Rauh et al., 2005), we compared the relative potencies of human plasma and TGF- in the M2-skewing of SHIP-/BMMs during maturation in M-CSF to establish whether or not TGF- plays a major role in M2 skewing. We found that while plasma dose-dependently increased Arg1 activity in M-CSF cultured, maturing SHIP-/- BMMs, the effect of TGF- plateaus at the higher doses tested and was much less potent than plasma (Fig 5.1). Thus while the TGF- pathway appeared to be critical for M2 skewing of maturing SHIP-/- BMMs, it was not the only factor involved.  109  Figure 5.1 TGF- fails to induce a similar level of arginase activity as plasma. -/Arginase activity of SHIP BMMs derived in the presence of M-CSF + varying doses of TGF- or human plasma. All data points are the mean ± SEM of experimental triplicates. Results are representative of 3 independent experiments. * = p ≤ 0.05 compared to all other points on the same dosage curve. ** = p ≤ 0.01 compared to all other points on the same dosage curve. † = p > 0.05 comparing the 10 ng/mL condition to the 1 ng/mL condition on the TGF- dosage curve. 5.2.2  Prostaglandin E2 (PGE2) Induces the SHIP-/- M2 Phenotype  Since our TGF- results suggested there were other factors in plasma that were contributing to the M2-skewing of SHIP-/- BMMs, we scoured the literature for plasma factors shown to upregulate M Arg1. Specifically, we were looking for factors that might synergize with TGF- to induce Arg1 in SHIP-/- BM progenitors in our BMENV assay. As mentioned earlier, Arg1+ Ms and immune cells were found to infiltrate tumors in a variety of mouse tumor models (Rodriguez et al., 2005; Sharda et al., 2011). We focused on cancer in particular because our results to date (Kuroda et al., 2009; Rauh et al., 2005) suggested the in vivo SHIP-/- M2-skewing was a result of elevated PI3K activity in immune cells, which is a commonality shared between the 110  SHIP-/- model and most tumor environments: the culprit being the absence of the critical negative regulator of PI3K, SHIP, in the former, and TAM or tumor cellderived chemokines (e.g., CCL2), which recruit immune cells and stimulate PI3K, in the latter (Negus et al., 1995; Karnoub and Weinberg, 2006). The two factors that we tested initially were MSP and PGE 2, since they had recently been implicated in Arg1 upregulation in cancer (Rodriguez et al., 2005; Sharda et al., 2011). However, while MSP induced Arg1, we excluded it as an important factor in plasma, since its effect was very mild, and in contrast to plasma/serum, it induced Arg1 in SHIP+/+ and SHIP-/- cells to the same degree (data not shown). PGE2, on the other hand, induced Arg1 specifically in maturing SHIP-/BMMs, an effect which was increased with the addition of TGF- (Fig 5.2 A). We also compared the effect of PGE2 ± TGF- on mature BMMs, and similar to our results with MP (Rauh et al., 2005), none of PGE2, TGF-, or PGE2 + TGF- (P/T) induced arginase activity (Fig 5.2 B) or Arg1 protein expression (Fig 5.2 C) in these mature cells.  111  Figure 5.2 PGE2 and TGF- induce the M2 M phenotype in SHIP-/- BMMs during maturation. A) Arginase activity of BMMs derived with M-CSF ± PGE2 (2 g/mL), ± TGF- (100 pg/mL) or both, using human serum (4%) as a positive control (all data points are the mean ± SEM of duplicates), and B) fold control arginase activity of mature BMMs stimulated with IL-4 (10 ng/mL) or PGE2 ± TGF- (100 pg/mL) for 72 hours. C) Western blots of mature BMMs stimulated with PGE2 (1 g/mL) or IL-4 (10 ng/mL) for 72 h. SHC was used to confirm equal loading. All lanes shown are from the same gel, and the same exposure time for each blot. Dotted lines indicate where irrelevant lanes have been cropped out. Results are representative of 3 independent experiments. * = p ≤ 0.05 compared to the M-CSF control. ** = p ≤ 0.01 compared to the M-CSF control. *** = p ≤ 0.001 compared to the M-CSF control. We next tested the ability of P/T to M2-skew SHIP-/- BMMs during differentiation using varying doses of PGE2 ± a fixed dose of TGF- and vice versa, and found that while PGE2 alone showed a plateau similar to that of TGF- in its ability to induce Arg1, its effect was synergistic with a very low dose of TGF- (100 pg/mL) (Fig 5.3 A and B). This synergy was further demonstrated when both PGE2 and TGF- were tested at different doses (Fig 5.3 C).  112  Figure 5.3 PGE2 and TGF- act synergistically to induce arginase activity in maturing SHIP-/- BMMs. -/Arginase activity of SHIP BMMs derived with A) increasing doses of PGE2 ± 100 pg/mL TGF-, B) increasing doses of TGF- ± 10 ng/mL PGE2, or C) increasing doses of TGF- and PGE2. All data points are the mean ± SEM of experimental triplicates. Results are representative of 2 independent experiments. * ** *** ¥ †  = p ≤ 0.05 compared to all other points on the same dosage curve. other points on the same dosage curve. = p ≤ 0.01 compared to all other points on the same dosage curve. = p ≤ 0.001 compared to all. = p ≤ 0.05 compared to the same dose on the other dosage curves. = p > 0.05 comparing the 10 ng/mL condition to the 1 ng/mL condition on the TGF- and TGF- + PGE2 dosage curve.  113  5.2.3  COX-2 Activity is Necessary for Plasma-Induced M2-Skewing of  SHIP-/- BMMs Having demonstrated that PGE2 could induce the Arg1+ SHIP-/- M phenotype in our BMENV assay, we then asked if it played a role in plasma/serum-induced SHIP-/- M2 programming. Since the in vivo half-life of PGE2 is extremely short, approximately 30s (Bygdeman, 2003; Watzer et al., 2009), and thus would be difficult to measure in MP, we asked what would happen if we prevented its synthesis using a COX-2 inhibitor. Specifically, we tested the COX-2 inhibitor SC58125 in our BMENV assay of MP-induced SHIP-/- BMMs and found that it reduced MP-induced Arg1 activity in SHIP-/- BMMs, and that this inhibitory effect was additive with that of the TGF- signaling inhibitor, SB-505124, demonstrating that COX-2 activity and TGF- signaling were both necessary for MP-induced arginase upregulation during the development of SHIP-/- BMMs (Fig 5.4).  114  Figure 5.4 COX-2 activity and TGF- signaling are both crucial to Arg1 induction by plasma in maturing SHIP-/- BMMs. -/Arginase activity of SHIP BMMs derived with M-CSF alone ± a COX-2 inhibitor (SC-58125 at 20 M in EtOH), a TGF- inhibitor (SB-505124 at 1M in dimethyl sulfoxide [DMSO]), or both (same concentrations), or with M-CSF + MP (4%) ± these same inhibitors and inhibitor combinations. All data points are the mean ± SEM; data are pooled from 2 independent triplicate experiments. * = p ≤ 0.05 compared to vehicle + MP ¥ = p ≤ 0.05 compared to SC-58125 + MP # = p ≤ 0.10 compared to vehicle + MP  115  We found that human plasma-induced arginase activity was similarly inhibited with SC-58125 (data not shown), but decided to focus on MP, since our major interest was the in vivo factors responsible for SHIP-/- M2-skewing. To try and rule out off-target effects of this COX-2 inhibitor, we tested another COX-2 inhibitor, Celebrex, in our MP-incorporated BMENV assay of SHIP-/- BMMs and found that it also reduced Arg1 protein expression in MP-derived SHIP-/- BMMs (Fig 5.5).  Figure 5.5 Both Celebrex and SC-58125 repress MP-induced M2 skewing. -/Western blot of SHIP BMMs derived ± MP (5% final concentration) ± Celebrex (Cel) or SC-58125 (SC). Loading was normalized to total protein using the BCA assay. All lanes were from the same gel and exposed film. Dotted lines indicate where irrelevant lanes have been cropped out. Results are representative of 2 independent experiments.  116  5.2.4  PGE2 and TGF- Sensitize BMM Progenitors to IL-4  Having demonstrated an alternate pathway to SHIP -/- M2-skewing from the one revealed in Chapter 4, we wanted to determine if P/T-mediated skewing and IL-4induced skewing were connected. To do this, we tested the role of IL-4 in P/Tmediated M2-skewing using a neutralizing -IL-4 Ab and found that like MP-induced M2-skewing, P/T-mediated M2-skewing of SHIP-/- BMMs during maturation was also dependent on IL-4 (Fig 5.6).  Figure 5.6 PGE2 + TGF--induced M2-skewing of SHIP-/- BMMs during differentiation is dependent on IL-4. -/Arginase activity of SHIP BMMs derived ± PGE2 (1 g/mL) + TGF- (100 pg/mL) (P/T) ± a neutralizing Ab to IL-4 (2.5 g/mL). All data are means ± SEM of experimental triplicates, representative of 2 independent experiments. ** = p ≤ 0.01 compared to MP-derived cells without neutralizing Ab. Since we had shown in the previous chapter that DX5 + BM cells, which we hypothesize are basophil or basophil progenitors, are solely responsible for IL-4 production, we examined the influence of P/T on IL-4 secretion, since it was possible that their ability to skew SHIP-/- BM might be because they promoted IL-4 secretion from DX5+ BM. We found that unlike IL-3, P/T did not significantly increase IL-4 production from SHIP-/- BMBs or DX5+ BM cells (Fig 5.7 A). As well, P/T did not induce a significant IL-4 spike from SHIP-/- adh- BM in our BMENV assay (Fig 5.7 B Top Panel), but it induced an increase in Arg1 protein expression (Fig 5.7 B Bottom Panel) and arginase activity in the derived BMMs (Fig 5.7 C).  117  Figure 5.7 The combination of PGE2 and TGF- does not increase IL-4 production from BMBs or DX5+ BM. -/4 A) (left panel) IL-4 levels produced by SHIP BMBs (2 x 10 cells/100 L) ± IL-3 (10 ng /mL) or PGE2 (9 g/mL) + TGF- (100 pg/mL) (P/T) at 24 h. (right panel) IL-4 +  -  5  levels produced by DX5 and DX5 BM cells (1 x 10 cells/200 L) stimulated with IL3 (2.5 ng/mL) or PGE2 (1.5 g/mL) + TGF- (100 pg/mL) (P/T) at 24 h. B) (Top -/-  panel) IL-4 levels in culture supernatants 24 h after in vitro SHIP BMM culturing ± IL-3 (10 ng/mL) or PGE2 (1 g/mL) + TGF- (500 pg/mL). (Lower panel) Corresponding Western blot of BMMs derived in these cultures. All lanes are from the same gel and same time exposed film. Equal loading was confirmed with SHC. Dotted lines indicate where irrelevant lanes have been cropped out. C) arginase activity of corresponding BMMs in (B). Results are representative of 2 independent experiments. † = p > 0.10 compared to M-CSF controls. 118  Because P/T, which mimicked the behaviour of MP, skewed SHIP -/- BM to an Arg1+ phenotype without increasing IL-4 production, we wanted to see if P/T increased the sensitivity of BMM to IL-4. Since we showed that MP sensitized both SHIP+/+ and SHIP-/- BMMs to IL-4, we tested both genotypes and found that, indeed, when P/T was added to mature BMMs with IL-4 or to mature BMMs cocultured with syngeneic BMBs, the resultant Arg1 protein expression was significantly increased in BMMs of both genotypes (Fig 5.8 A). Since it had been reported that TGF- can upregulate the protein expression of COX-2 in human cancer cell lines (Sheng et al., 2000; Rodriguez-Barbero et al., 2006), we asked if could do the same in mature BMMs. Interestingly, we found that it did and this TGF--induced increase was far greater in SHIP+/+ cells (Fig 5.8 B). To determine if the COX-2 and TGF- pathways were somehow involved in MPmediated sensitization, we used their respective inhibitors SC-58125 and SB505124, and found that the IL-4-induced upregulation of Arg1 protein in MPsensitized BMMs of both genotypes was reduced with SB-505124 but not with SC58125 (Fig 5.8 C). However, it appeared to act synergistically, at least in SHIP-/BMMs, to reduce Arg1 levels, (Fig 5.8 C). Furthermore, since MP-containing TGF increased COX-2 protein levels, especially in SHIP+/+ mature BMMs, we asked if MP also upregulated COX-2 and if this was dependent on TGF-. As shown in Fig 5.8 C, we indeed found that MP increased COX-2 expression in IL-4 stimulated Ms and that this increase was dependent on TGF- signalling (i.e., it was reduced with SB-505124). Since we knew COX-2 inhibition inhibited MP-induced M2-skewing of SHIP-/BMMs during maturation, which is IL-4 and MP-sensitization-dependent, but COX2 inhibition alone did not seem to inhibit the MP-induced sensitization of BMMs to IL-4, we next wanted to ascertain if the IL-4 induction of Arg1 without MP was dependent on COX-2. Since Arg1 induction in response to IL-4 was not different between SHIP+/+ and SHIP-/- BMMs, we used only SHIP+/+ BMMs for these experiments. Using two different COX-2 inhibitors, SC-58125 and Celebrex, we found both inhibitors reduced IL-4 induced Arg1 protein upregulation, and as shown 119  with Celebrex, the IL-4-induced upregulation of BMM arginase activity was decreased as well (Fig 5.8 D).  120  Figure 5.8 PGE2 and TGF- sensitize BMMs to IL-4, and COX-2 and TGF- signaling are necessary for MP-sensitized BMM response to IL-4 +/+ -/5 A) Western blot of mature SHIP and SHIP BMMs (2.5 x 10 cells) cultured with +/+ 4 -/IL-4 (+4) (10 ng/mL) or co-cultured with SHIP (2 x 10 cells) and SHIP BMBs (+B) 4 (2 x 10 cells), respectively, for 72 h ± PGE2 (9 g/mL) + TGF- (100 pg/mL) (P/T). B) Western blot of mature BMMs ± TGF- (10 ng/mL) for 72 h. C) Western blot of +/+ -/mature SHIP and SHIP BMMs cultured for 72 h with IL-4 (500 pg/mL) ± MP (5%), the COX-2 inhibitor SC-58125 (SC) (20 M), TGF- signaling inhibitor SB505124 (SB) (1 M), or both (B). All lanes contained equal amounts of protein as +/+ assessed by a BCA assay. D) Western blots of mature SHIP BMMs ± IL-4 (10 ng/mL) in the presence of (left panel) SC-58125 (20 M) for 24 h or (middle panel) Celebrex for 72 h, and (right panel) arginase activity (mean ± SEM of assay duplicates) of cells similarly stimulated for 96 h. All lanes are from the same gel and same time exposed film. Equal loading was confirmed with SHC. Dotted lines indicate where irrelevant lanes have been cropped out. Results are representative of 2 independent experiments. # = p ≤ 0.10 compared to IL-4 control. 121  5.2.5  COX-2 Inhibition Ameliorates in vivo M2-skewing in the SHIP-/-  Mouse Since we showed above that reducing PGE2 and TGF--signaling inhibited IL-4 or MP-sensitized M2-skewing, we asked if this could be exploited as potential in vivo interventions. To test this, we first looked at the effectiveness of COX-2 and TGF-signaling inhibitors at modifying SHIP-/- PMs ex vivo. Given that the M2-phenotype of these PMs could not be reduced with cytokine withdrawal or IL-4 neutralization, we were not confident this would work. Interestingly, however, we found that the ex vivo expression of Arg1 protein in SHIP-/- PMs was reduced with the addition of either of the two COX-2 inhibitors tested, while surprisingly, TGF- signaling inhibition without concurrent COX-2 inhibition did not (Fig 5.9 A). This is consistent with previous studies showing that COX-2 inhibition downregulates Arg1 in Arg1+ cells (Rodriguez et al., 2005). Since COX-2 inhibition was capable of reversing the SHIP-/- PM Arg1+ phenotype ex vivo, we incorporated the COX-2 inhibitor, Celebrex, at 0.1% (w/w) into our standard mouse chow and fed it to our SHIP -/- mice for 2 weeks. We found that SHIP-/- mice treated with this Celebrex-incorporated chow showed reduced M2-skewing, as demonstrated by the lower Arg1 protein expression in their PMs compared to those isolated from SHIP-/- mice fed normal chow (Fig 5.9 B).  122  Figure 5.9 COX-2 inhibition reduces the ex vivo and in vivo Arg1+ phenotype of SHIP-/- PMs. -/A) Western blot of SHIP PMs incubated in 10 ng/mL M-CSF for six days, ± the COX-2 inhibitors SC-58125 (SC) and Celebrex (Cel), or the TGF- signaling inhibitor SB-505124 (SB), or both SC + SB. The D0 lane represents ex vivo PMs that were not cultured in vitro. All lanes contained equal amounts of total protein and Grb2 was used as a loading control. All lanes are from the same gel and same time exposed film. The dotted line indicates where irrelevant lanes have been cropped out. B) -/Western blot of PMs from SHIP mice fed standard chow or standard chow supplemented with 0.1% (w/w) Celebrex for 14 days. Each lane represents an -/individual SHIP mouse. Results are representative of trends from 2 independent experiments. 5.3  Discussion  5.3.1  PGE2 and COX-2 Synergize with TGF- to Skew SHIP-/- BM to M2  Ms Since our results from the previous chapter revealed that an alternate method might be needed to treat SHIP-/- M2-skewing in vivo, we decided to follow up on earlier studies in our laboratory showing a role for TGF- in MP-induced M2 skewing (Rauh et al., 2005). Specifically, we were looking for other pathways or factors that  123  skewed SHIP-/- BMMs during development to an M2 phenotype, focusing particularly on ones with potential for in vivo intervention. PGE2 and COX-2 emerged in the literature as regulators of Arg1 in cancerinduced immunosuppressive M2 Ms and MDSCs (Eruslanov et al., 2010; Rodriguez et al., 2005), which prompted us to test whether or not PGE2 was a critical factor in MP for SHIP-/- M2-skewing. Our finding that PGE2 was sufficient and synergistic with TGF- in skewing SHIP-/-, but not SHIP+/+, BM to M2 Ms was evidence suggesting that it may be an important factor in MP-induced skewing. However, while the short in vivo half-life of PGE2 (Bygdeman, 2003; Watzer et al., 2009) made it difficult to measure PGE2 in MP, we found that PGE2 production was likely critical for MP-induced skewing, since the activity of COX-2, which is an enzyme critical for PGE2 synthesis, was necessary for MP-induced skewing of SHIP/-  BM into M2 Ms. It is important to note that COX-2 is an enzyme upstream of the PGE2 synthase  enzymes directly responsible for PGE2 synthesis. Therefore, although our experiments show that COX-2 activity is necessary, we cannot conclude from our results that PGE2 itself is the molecule responsible, since COX-2 produces PGG2 and PGH2, which are precursors of thromboxanes and other PGs (Simmons et al., 2004). Another caution in interpreting our results is the known off-target effects of Celebrex and SC-58125. In particular, Celebrex is known to not only inhibit COX-2, but also the activity 3-phosphoinositide-dependent protein kinase-1 (PDK1), a requisite activator of Akt (Kulp et al., 2004). Because Akt is a downstream effector of the PI3K pathway, which is upregulated in SHIP-/- cells and implicated in M2-skewing (Rauh et al., 2005), it was a concern that the effect of Celebrex may be mediated by its off-target inhibition of the PI3K pathway. However, since the concentration required for PDK1 inhibition is about 30-48 M (Kulp et al., 2004) and we show an effect at a much lower concentration, it is unlikely that Celebrex is mediating its effects through PDK1 inhibition in our experiments. As well, we obtained similar results with a second COX-2 inhibitor, SC-58125. Although this inhibitor does not seem to have an effect on the PI3K pathway, at least at the concentrations we used, it may induce oxidative stress, via glutathione (GSH) depletion, in a non-COX-2 124  dependent manner (Ryan et al., 2008). However, although this has been shown in B cells it may not be the case in Ms and preliminary results in our laboratory suggest that GSH depletion is likely not the cause of the SC-58125-dependent repression of the M2-phenotype in MP-induced SHIP-/- M2 BMMs (data not shown). Moreover, the fact that both SC-58125 and Celebrex are effective suggests that our observed repression of the M2-phenotype is attributed to COX-2 inhibition and not off-target effects.  5.3.2  The Roles of PGE2 and TGF-in in vivo M2-Skewing  Another finding from our work was that PGE 2 and TGF- behaved much like MP in skewing SHIP-/- but not SHIP+/+ BMMs during differentiation. Indeed, we also found that COX-2 activity and TGF- signaling were both necessary for optimal MPinduced SHIP-/- M2-skewing. Given that P/T-induced M2-skewing of SHIP-/- BMMs during differentiation is also dependent on IL-4, it stands to reason that DX5+ BMderived IL-4 is also necessary. An intriguing question, therefore, is the mechanism through which IL-4 and the COX-2 pathway are connected. While we have no data directly supporting this, the literature suggests that peroxisome proliferator-activated receptor- (PPAR-) may be a viable avenue of investigation. PPAR-is a nuclear receptor that binds many ligands associated with low-density lipoprotein (LDL) and its modified counterparts, including some oxidized fatty acids, and 15-deoxy-Δ12,14-prostaglandin J2 (15DPGJ2), a breakdown product of PGD2 and PGJ2 (Schupp and Lazar, 2010). Under normal conditions, PPAR- is highly expressed in adipose tissue, where M2 Ms play a role in lipid homeostasis in lean animals, and the switch from M2 to M1 may be a marker for diet-induced obesity (Lumeng et al., 2007; Schupp and Lazar, 2010). Importantly, PPAR- seems to be necessary for the induction of Arg1 by IL-4 and oxidized-LDL in BMMs (Odegaard et al., 2007; Gallardo-Soler et al., 2008). Since PG breakdown products (e.g. 15DPGJ2) may be vital to the upregulation of Arg1 by IL-4, via PPAR- (Odegaard et al., 2007; Gallardo-Soler et al., 2008), COX-2 activity may, therefore, be part of the IL-4 signaling cascade via its production of PPAR- ligands.  125  With regard to TGF-, what is most intriguing is our finding that TGF- and MP both induce COX-2 protein expression in Ms. Furthermore, the induction of COX-2 by MP appears to be dependent on TGF- signaling. This result is consistent with earlier studies with human cells and cancer cell lines showing that TGF- induces COX-2 expression (Sheng et al., 2000; Rodriguez-Barbero et al., 2006). Related to this, it is interesting that our results suggest that TGF- induces COX-2 to a lesser degree in SHIP-/- BMMs. Since COX-2 seems to be necessary for IL-4-induced Arg1 expression, this perhaps explains why, in Chapters 3 and 4, SHIP+/+ BMMs appeared to be more sensitive to IL-4 than SHIP-/- BMMs.  5.3.3  COX-2, TGF-, and MP-Mediated IL-4 Sensitization  Because we were focusing on identifying a pathway which might be treatable in vivo, we did not delve too deeply into the mechanisms through which P/T might be acting to induce skewing in vitro or in vivo. Our data do indicate, however, that COX2 activity may be required for optimal IL-4 induction of Arg1 protein and arginase activity, and TGF- signaling may be necessary for MP-mediated IL-4 sensitivity. Furthermore, similar to previous reports (Rodriguez et al., 2005), we demonstrate that Arg1 expression in both SHIP-/- ex vivo and in vivo PMs requires COX-2 activity to maintain, since the use of COX-2 inhibitors reduced Arg1 expression. Although more experiments will have to be carried out to confirm this, it is tempting to speculate that MP-induced IL-4 sensitivity and M2-skewing are due to TGF- and its upregulation of COX-2, which is required for optimal IL-4 induction. A model summarizing this is presented in Fig 5.10 A & B.  126  Figure 5.10 Model of M2-skewing in mature Ms and M progenitors: the roles of MP, COX-2, and TGF- A) Proposed model of M2-skewing of Ms, in vivo and in vitro, summarizing our results, showing the possible interaction of TGF- in MP. B) Model of possible interactions of TGF- contained in MP, and how it upregulates COX-2 to sensitize Ms to respond to IL-4.  127  We have shown in this chapter that COX-2 and TGF- are 2 additional elements that are involved in the in vitro and in vivo M2-skewing of SHIP-/- Ms. Since we showed in Chapter 4 that IgG in MP was a critical factor, our results from both chapters beg the question of how IgG, COX-2, and TGF- are related. While the possible link between COX-2 and TGF- was discussed above, it seems that circulating IgG and TGF- often form complexes, which are thought to bind to Fc Rs (Stach and Rowley, 1993; Rowley and Stach, 1998). Importantly, while this circulating TGF- is predominantly associated with latency associated protein (LAP), which renders TGF-biologically inactive, Mo/Ms seem to take up IgG-TGF- complexes via FcRs and release active TGF- (Stach and Rowley, 1993; Rowley and Stach, 1998; Ashcroft, 1999). In the context of our work, IgG may be the carrier for TGF-, which when activated induces COX-2 expression to enhance IL-4induced Arg1. However, more work will have to be carried out to test this hypothesis.  128  6. A Low Carbohydrate, High Protein Diet Slows Tumor Growth and Prevents Cancer Initiation 6.1  Introduction In the previous chapters, we concentrated on the normal cells within the tumor  microenvironment, focusing specifically on Ms polarized to the pro-cancer Arg1+ M2 phenotype. In this final data chapter, we examine the cancer cells themselves within the tumor and focused on their CHO metabolism. Similar to the strategy we employed in the previous chapters, we asked if the pro-tumor aspects of CHO metabolism exhibited by most cancer cells could be manipulated or reversed.  6.1.1  Glucose  A cancer cell's survival, like that of a parasite, is inexorably linked to its host's nutrition supply. In fact, the ability of a tumor to absorb nutrients from its environment has been shown to be superior to that of normal tissues (Warburg et al., 1927; Vander Heidan et al., 2009) and since both cancer cells and normal cells are supplied through the circulation, nutrient supply to normal tissues becomes limiting as tumors progress until it eventually results in cancer-induced cachexia (Dhanapal et al., 2011). As stated in the Introduction to this thesis, tumors typically take up BG more rapidly than normal cells and tend to use FG more than OXPHOS to generate ATP, since it allows for higher levels of cellular building blocks and immunosuppressive, extracellular lactate to augment its growth and metastasis, respectively (Warburg, 1956; Fantin et al., 2006; Bragoszewski et al., 2008; Feron, 2009). Most normal cells, which are not rapidly dividing, use FG sparingly, and instead fulfill their ATP needs via OXPHOS (Kroemer and Pouyssegur, 2008). Since OXPHOS can generate ATP through fatty acid or amino acid catabolism we reasoned that limiting BG might be a way to specifically limit the growth and survival of cancer cells.  129  6.1.2  Diet and Carbohydrates  Glucose is the basic building block of starch, the predominant food source in most Western human diets (Cordain et al., 2005). The onset of agriculture allowed for a dramatic increase in the human population (Pijl, 2011). However, this steady food supply from farming changed our pre-agrarian diet, which consisted primarily of animal-derived foods (~50%) and seasonal CHOs into a CHO-based (~50%) diet that is prevalent in the world today (Cordain et al., 2005; Pijl, 2011). In terms of biochemistry, this increased consumption of CHOs leads to an increase in BG and this, in turn, triggers the release of Ins (Henquin et al., 2009). This post-prandial Ins and BG spike is intensified with the ease with which an ingested CHO foodstuff is digested (Behall and Hallfrisch, 2002). With modern refining techniques, whole grains and sugar-containing plants are now milled and stripped into more readily absorbable forms, which result in even higher postprandial BG and insulin spikes (Cordain et al., 2005; Brand-Miller and Buyken, 2011). A regular Western diet is thus associated with BG and Ins extremes with every meal. Interestingly, high BG and Ins are associated with obesity and type-2 diabetes, both of which are in turn associated with an increased cancer risk (Braun et al., 2011; Yuhara et al., 2011). This suggests that a CHO-based Western diet may, in fact, be conducive to tumor growth. Since the use of low glycemic index (GI) diets to control post-prandial BG and Ins have proven effective against obesity (Gogenbakan et al., 2011; Parillo et al., 2011; Gannon and Nuttall, 2004), we wanted to change the dietary composition of the major nutrients (i.e., fat, CHO, and protein) to lower BG and Ins to see if this could reduce the growth and incidence of cancer.  6.1.3  Dietary Intervention  Food consumption is the route through which nutrients are assimilated into an animal. Lipids, CHO, and protein are digested and absorbed as their monomers (i.e., fatty acids, sugars, and amino acids). Cellular building blocks and ATP are, in turn, generated from these monomers (Nelson and Cox, 2000). Biochemically speaking, a  130  typical Western diet is composed of ~50% CHO, ~35% fat and ~15% protein (expressed as % calories) (Cordain et al., 2005). The most direct way of limiting BG and Ins is to limit the amount of CHO in the diet. Since the maintenance of caloric consumption is especially important in cancer, any dietary alteration should be isocaloric, which means that for every calorie of CHO that is taken away, a calorie of either fat or protein needs to replace it. In the literature, other groups have used no CHO ketogenic diets (NCKDs), which are typically high in fat. This lowers BG and initiates ketogenesis, which supplements BG with ketone bodies in the circulation to serve as energy carrier molecules (Nebeling et al., 1995; Gannon and Nuttall, 2004). Recent evidence, however, implicates high fat as a possible cancer risk (Khalid et al., 2010; VanSaun et al., 2009) and so we did not want to replace CHO with high fat. Aside from reducing total CHO, the sharp increase in BG and Ins can also be mitigated by altering the type of CHO in the diet (Behall and Hallfrisch, 2002). Typically, plant stores of CHO come in the form of starch, which is composed primarily of highly branched amylopectin and, to a lesser degree, less-branched amylose (Wheeler and Pi-Sunyer, 2008). While amylopectin is rapidly catabolized into glucose and absorbed, the less branched amylose starch, which packs tightly to restrict enzyme access, is more resistant to digestion (Behall and Hallfrisch, 2002; Wheeler and Pi-Sunyer, 2008). As a result, consumption of amylopectin starch results in rapid and pronounced post-prandial spikes in BG and Ins, while amylose consumption yields slower and more muted BG and Ins spikes (Behall and Hallfrisch, 2002). In the studies described in this chapter, we compared the effects of low CHO, high protein diets to an isocaloric Western-like diet, on BG, Ins, and lactate levels in mice. We also assessed the impact of these diets on the growth of ectopically implanted tumors as well as on the incidence of a spontaneous breast cancer model.  131  6.2  Results  6.2.1  Tumors Grow Slower in Mice on an 8% CHO, 69% Protein, 23% Fat  (8% CHO) Diet, but the Mice Lose Weight As it is well established that most human and murine tumors take up more glucose than normal tissues (Gambhir, 2002), we asked if we could decrease BG levels sufficiently, by decreasing dietary CHO, to significantly reduce tumor growth rates. We considered this possible because NCKDs have recently been shown to reduce tumor growth rates in mice and rats (Freedland et al., 2008). However, as it would be extremely difficult for humans to maintain such a NCKD, we asked if a more moderate, CHO-reduced diet could decrease BG levels and reduce tumor growth rates. To test this, we first designed a mouse diet containing 8% CHO (% of total calories consumed), because this level is used in the Atkins diet (Anderson and Moore, 2004). However, we kept fat levels in the range of a Western diet (23%) rather than the 50% used in the Atkins diet because of the tumor-promoting effects of high fat (Khalid et al., 2010; VanSaun et al., 2009), and raised the level of protein instead (Table 6.1). Comparing the effect of this diet, given ad libitum, with an isocaloric Western diet (TestDiet 5058; Table 6.1) on BG levels in non-tumorbearing Rag2M mice revealed that BG, indeed, dropped significantly after 4 to 7 days on the 8% CHO diet to a new, stable plateau (Fig 6.1 A), in keeping with previous reports showing that BG drops within 7 days on a ketogenic diet (Nebeling et al., 1995). Interestingly, this drop was more pronounced in male mice, consistent with the reported BG buffering effects of estrogen (Matsuda and Mori, 1996). On the basis of these results, we carried out the majority of our studies with male mice. In our first tumor studies, we acclimatized 5- to 6-week-old male C3H/HeN mice to either 5058 or the 8% CHO diet for 1 week, then injected, SC, SCCVII cells and monitored tumor growth. As shown in Figure 6.1 B, tumors in the mice on the 8% CHO diet grew significantly slower, with the mean tumor size of the 8% CHO group (130.9 ± 21.76 mm3) being less than half that of the 5058 group (364.3 ± 85.01 mm 3) at 16 days after tumor implantation. Also, BG levels in the 8% group were  132  significantly lower (Fig 6.1 C). Similar results were obtained in Rag2M mice injected with human colorectal cancer cells (HCT-116 cells; data not shown). Although mice on 8% CHO diet had slower growing tumors, they lost weight, weighing, on average, 20% less than mice on 5058 diet (Fig 6.1 D). This was consistent with the mice eating less than the 5058 group (data not shown), likely because the 8% CHO pellets were significantly harder to chew. This confounded our results because caloric restriction (CR), which is known to cause cells to switch, via AMPK activation, to OXPHOS to generate more ATP for survival (Guarente, 2008), has been shown to slow tumor growth (Hursting et al., 2010). Thus, we could not rule out the possibility that the slower tumor growth rates were due to the effects of CR rather than to reduced dietary CHO.  Table 6.1 Macronutrient breakdown of diets used. Reprinted with the permission of Cancer Research and the American Association for Cancer Research (Ho et al., 2011). † Starch content is 70% amylose starch.  133  Figure 6.1 Tumors grow slower in mice on an 8% CHO diet than those on a Western diet, but the mice weigh less. A) BG time course of male and female Rag2M mice after switching to an 8% CHO diet. B), growth of SCCVII tumors in C3H/HeN mice on the 8% CHO versus 5058 diets (n = 8 for both groups). C) BG and D) body weights of these mice on the 8% CHO diet 6 days after diet switch. Reprinted with the permission of Cancer Research and the American Association for Cancer Research (Ho et al., 2011). * = p ≤ 0.05 comparing the 8% CHO group to its respective 5058 group # = p ≤ 0.10 comparing the 8% CHO and 5058 groups. † = p > 0.10 comparing the 8% CHO group to its respective 5058 group.  134  6.2.2  A 15% High Amylose CHO, 58% Protein, 26% Fat (15% CHO) Diet  Reduces Fasting and Constitutive BG, and Tumor Growth To prevent CR, we formulated a new diet consisting of 15.6% CHO, 58.2% protein, and 26.2% fat. Instead of sucrose, which was in our 8% CHO diet, this diet contained cornstarch with 70% amylose because it allowed for a pellet consistency similar to 5058, and because amylose is digested more slowly than sucrose or amylopectin (in 5058), which results in less pronounced postprandial BG and Ins spikes (Behall and Hallfrisch, 2002). We found that mice ate this chow at the same rate as 5058, and that, after a short fasting period, it did not increase BG to the same extent as 5058, two hours after feeding (Fig 6.2 A). Moreover, mice on this 15% CHO diet had lower constitutive BG levels than mice on the 5058 diet (Fig 6.2 B). We then compared SCCVII tumor growth in C3H/HeN mice on the 15% CHO versus 5058 diets and found that tumors grew significantly slower in the 15% CHO group, with an average volume of 321.0 ± 79.79 mm3 versus 542.9 ± 78.80 mm3 in the 5058 group, 16 days after implantation (Fig 6.2 C). Significantly, there was no difference in caloric intake (data not shown), average body weight (Fig 6.2 D, left), or rate of weight gain (Fig 6.2 D, right) between these diet groups. We also compared the effect of this 15% CHO diet with 5058 on the growth of HCT-116 tumors in Rag2M mice and found that 15% CHO mice had significantly smaller tumors, with a mean size of 255.6 ± 10.50 mm 3 versus 401.7 ± 35.21 mm3 in the 5058 group, 21 days after tumor implantation (Fig 6.2 E). Once again, there was no difference in the average body weight or rate of weight gain between the 2 groups (Fig 6.2 F).  135  136  Figure 6.2 Tumors grow slower in mice on a 15% CHO diet than a Western diet, and the mice weigh the same. A) post-prandial BG in C3H/HeN mice fed 5058 or 15% CHO diets after a 6-hour fast (n = 5 for 5058, n = 4 for 15% CHO). B) Constitutive BG of mice at 3 different times of day. C) SCCVII tumor growth in mice fed the 15% CHO versus 5058 diets (n = 5 for both groups); D) their body weights on final measurement (left) and weight change (right). E) HCT-116 tumor growth in Rag2M mice on the 15% CHO versus 5058 diet (n = 8 for both groups). F) Weights of mice on final measurement (left) and weight change (right) of these Rag2M mice. Results are given as mean ± SEM. Reprinted with the permission of Cancer Research and the American Association for Cancer Research (Ho et al., 2011). * = p ≤ 0.05 comparing the low CHO group to its respective 5058 group. # = p ≤ 0.10 comparing the low CHO and 5058 groups. † = p > 0.10 comparing the low CHO and 5058 groups.  137  6.2.3  A 10% CHO Diet Slows Tumor Growth More than a 15% CHO Diet  without Significant Weight Loss To see if we could further reduce tumor growth rates by decreasing dietary CHO levels even more, we tested another isocaloric diet containing 10% high amylose CHO, 64% protein, and 26% fat (Table 1). Comparing tumor growth in male C3H/HeN mice, we found that tumors in mice fed this 10% CHO diet were significantly smaller (572.3 ± 215.7 mm3) than those on 5058 (1153.0 ± 108.0 mm 3), 22 days postimplantation (Fig 6.3 A). This difference was more pronounced than with the 15% CHO diet and on par with the 8% CHO diet. As expected, the BG of mice on the 10% CHO was lower than those on 5058 (Fig 6.3 B). Even though the mice on the 10% CHO diet gained weight throughout the study and ate the same amount of food (data not shown), their average body weight at the end was slightly (∼7%) lower (Fig. 6.3 C), raising the concern that the smaller tumors in the 10% CHO group might be because of a smaller body size. To investigate this, we carried out a meta-analysis of pooled data from 3 independent experiments and found no significant positive correlation between body weight and tumor size for either the 5058 (Fig 6.3 D, left) or 10% CHO groups (Fig 6.3 D, right). This indicated that, within the range of mouse weights tested, smaller body sizes were not related to smaller tumors. Nonetheless, we cannot say with absolute certainty that the slightly lower weights of the 10% CHO-fed mice had no impact on tumor size.  138  Figure 6.3 A 10% CHO diet is more effective than the 15% CHO diet at slowing tumor growth with only a slight effect on mouse weight. 5 A) Tumor growth in male C3H/HeN mice receiving 1 × 10 SCCVII cells on the 10% CHO versus 5058 diet (n = 5 for both groups). B) BG measurements of 10% CHO and 5058 groups at sacrifice. C) Changes in body weight. D) Linear regression of tumor size versus body weight for the 5058 (left) and the 10% CHO groups (right). Except for the linear regression, all results are given as mean ± SEM. Reprinted with the permission of Cancer Research and the American Association for Cancer Research (Ho et al., 2011). * = p ≤ 0.05 comparing the low CHO and 5058 groups # = p ≤ 0.10 comparing the low CHO and 5058 groups. † = p > 0.10 for both an F test for a >0 slope and a positive correlation in a Spearman rank test.  139  6.2.4  Low CHO Diets Cause a Drop in Plasma Insulin and Lactate  To gain some insight into how the low CHO diets were reducing tumor growth rates, we measured plasma Ins levels and found that all the low CHO diets reduced plasma Ins, with the 8% and 10% CHO having a more marked effect than the 15% CHO diet (Fig 6.4 A). As high BG triggers Ins release from pancreatic β-cells, and the released Ins then enhances cellular uptake of BG via Ins receptor-mediated upregulation and activation of glucose transporters (Shaw, 2006), these Ins results suggest that low CHO diets can reduce Ins-mediated glucose uptake into tumor cells. Consistent with this and our hypothesis that glucose supply is related to tumor growth, we found a positive correlation between plasma Ins levels and tumor size (Fig 6.4 B). We also compared plasma lactate levels in 5058 versus 10% CHO mice and found the 5058-fed mice had significantly higher lactate levels (0.713 ± 0.03 mmol/L versus 0.572 ± 0.03 mmol/L; Fig 6.4 C), consistent with reduced glycolysis in the low CHO-fed mice. Once again, we found a positive correlation between plasma lactate levels and tumor size (Fig 6.4 D).  140  141  Figure 6.4 Low CHO diets reduce plasma insulin and lactate levels. A) Plasma insulin levels of SCCVII tumor-bearing male C3H/HeN mice on the low CHO (8%, 15%, 10%) versus their respective 5058 experimental controls. Results of the 10% CHO diet are from data pooled from 3 experiments. B) Linear regression of final plasma insulin levels versus final tumor volumes on all mice; data pooled from 3 experiments by using the 10% CHO and 5058 diets. C) Lactate levels in the plasma of these mice on the 10% CHO and 5058 diets. D) Linear regression of final plasma lactate levels versus final tumor volumes on all mice; data pooled from 3 experiments by using the 10% CHO and 5058 diets. Except for the linear regression, all results are given as mean ± SEM. Reprinted with the permission of Cancer Research and the American Association for Cancer Research (Ho et al., 2011). * = p ≤ 0.05 comparing the low CHO and 5058 groups. ‡ = p < 0.05 for both an F test for a >0 slope and a positive correlation in a Spearman rank test. # = p ≤ 0.10 for an F test for a >0 slope and p < 0.05 for a positive correlation in a Spearman rank test.  142  6.2.5  Low CHO Diets Act Additively with Known Cancer Therapeutic  Agents to Reduce Tumor Growth Having shown that the 10% and 15% CHO diets slowed tumor growth without significant weight loss, we asked if they might be additive with known cancer therapeutic agents. To test this, we first compared the growth of SCCVII (Fig 6.5 A, left) and Lewis lung carcinoma (data not shown) tumors in mice on the 10% CHO or 5058 diets ± the mTOR inhibitor, CCI-779, and found that, in both, combining the 10% CHO diet with CCI-779 resulted in an additive effect, with a negligible effect on mouse weights (Fig 6.5 A, right). Most exciting, however, were the results obtained with the 15% CHO diet containing the COX-2 inhibitor, Celebrex. Not only was tumor growth significantly reduced with the 15% CHO diet containing 1 g/kg Celebrex, but the overall slope of the tumor growth was lower (Fig 6.5 B, left). Once again, there were negligible effects on mouse weights (Fig 6.5 B, right), and although the Celebrex-treated mice weighed slightly less than the mice not treated with Celebrex, they did not fall outside the range tested in the meta-analysis, suggesting that the effect of Celebrex was not related to lower body weights.  143  Figure 6.5 Low CHO diets act additively with current treatments for cancer. A) growth of SCCVII tumors in male C3H/HeN mice on the 10% CHO (n = 3) or 5058 diets (n = 6) ± CCI-779 (n = 3 for both CCI groups; left) and body weight versus days after tumor injection (right). B) Growth of SCCVII tumors in male C3H/HeN mice on the 15% CHO (n = 5) or 5058 (n = 5) diet ± 0.1% w/w Celebrex (Cel; 5058 + Cel n = 10; 15% CHO + Cel n = 6; left) and body weight versus days after tumor injection (right). All results are given as mean ± SEM. Reprinted with the permission of Cancer Research and the American Association for Cancer Research (Ho et al., 2011). * = p ≤ 0.05 comparing the 5058 to any other diet group. ‡ = p ≤ 0.05 comparing drug treated (5058 or low CHO) with their respective untreated control group (same diet). # = p ≤ 0.10 comparing the untreated low CHO and 5058 groups. a = p < 0.10 for a t test comparing drug treated groups (5058 versus low CHO).  144  6.2.6  The 15% CHO Diet Reduces the Incidence of Tumors in a  Spontaneous Mouse Model of Breast Cancer We then asked if our low CHO diets could reduce cancer incidence in a spontaneous cancer model by using female NOP mice, which express a dominantnegative allele of p53 and the HER2/Neu oncogene under the control of the MMTV promoter, thus mimicking human breast cancers (Wall et al., 2007). These mice have a 70% to 80% chance of developing mammary tumors over their lifetime (Wall et al., 2007). Mice were switched onto the 15% CHO or 5058 diets when they reached adulthood (8 weeks), and, 9 weeks later, we found that BGs were significantly lower in the 15% CHO group (Fig 6.6 A, left). Interestingly, whereas the weights were stable in both groups after 8 to 9 weeks on the diets, they were consistently lower in the 15% CHO group (Fig 6.6 B), which is not unexpected, given that long-term low CHO diets reduce body mass (Hession et al., 2009). Also, plasma Ins levels, taken at death, were significantly lower in the 15% CHO group (Fig 6.6 C). Importantly, as shown in Figure 6.6 D, at 1 year of age almost half the mice on 5058 had developed tumors compared with none in the mice on the 15% CHO diet. Furthermore, 70% (7 of 10) of mice on 5058 developed tumors during their life span, with only 1 reaching normal life expectancy, whereas less than 30% (3 of 11) of the mice on the 15% CHO diet developed tumors, with more than half reaching or exceeding normal life expectancy. Of note, in the 5 mice on the 15% CHO diet that exceeded normal life spans, only 1 had kidneys that showed above-normal levels of protein in the urine (data not shown). These long-term mouse studies suggest that this 15% high amylose CHO, 58% protein, 26% fat diet is both safe and efficacious.  145  Figure 6.6 The 15% CHO diet reduces the incidence of tumors in a spontaneous mouse model of breast cancer. A) BG measurements at 9 weeks after diet switch of 10 (5058) and 11 (15% CHO) female NOP mice. B) Body weight of these same mice over time. C) Plasma insulin of NOP mice at death. D) Survival curve and tumor incidence versus time (months) of NOP mice on 5058 versus 15% CHO diet. Dots indicate tumor events. Except for the survival curve, all results are given as mean ± SEM. Reprinted with the permission of Cancer Research and the American Association for Cancer Research (Ho et al., 2011). * = p ≤ 0.05 comparing the low CHO and 5058 groups. ‡ = p ≤ 0.05 in a log-rank test for significant differences between the survival curves.  146  6.2.7  Our Low CHO Diets Do Not Increase Ketosis  After an initial depletion of glycogen stores in the liver, very low CHO diets have been reported to induce ketogenesis, which is a process in which fatty acids are broken down into ketone bodies (Masko et al., 2010; Jornayvaz et al., 2010). These ketone bodies circulate in the blood stream to provide energy to the body when BG is low and have also been reported to have an anti-tumor effect (Magee et al., 1979). To test if our low CHO diets were triggering the formation of ketone bodies and that it was actually these ketone bodies that were responsible for the slower tumor growth we were observing, we analyzed the levels of -hydroxybutyrate in the plasma samples of low CHO diet-fed mice. We found the plasma levels of hydroxybutyrate were not higher in mice from our short-term tumor studies fed the 10% (Fig 6.7 A) CHO diet compared to 5058-fed mice. Furthermore, in our longterm cancer-incidence study, there was, again, no difference between the plasma βhydroxybutyrate levels of 15% CHO diet-fed and 5058-fed mice (Fig 6.7 B).  Figure 6.7 Plasma -hydroxybutyrate levels in low CHO-fed mice are the same as those in 5058-fed mice. A) Plasma -hydroxybutyrate levels in tumor-bearing mice on the 10% CHO diet or 5058 diet. Results are pooled from three independent studies. B) Plasma hydroxybutyrate levels from the NOP mice on the 15% CHO diet or 5058 diet (Fig 6.6), collected at time of death. Reprinted with the permission of Cancer Research and the American Association for Cancer Research (Ho et al., 2011). 6.3  Discussion To exploit the fact that cancer cells rely more heavily on glycolysis than normal  cells, we designed low CHO, high protein diets to see if we could limit BG and tumor  147  growth. In designing our diets, we wanted to avoid NCKDs because of the difficulty in achieving long-term compliance with no CHO diets in potential future human studies (Masko et al., 2010) and because Masko and colleagues (2010) recently reported that a 10% or 20% CHO diet slows tumor growth as effectively as NCKDs. Following early studies with 8% CHO diets, using 10% and 15% CHO, high protein diets in which 70% of the CHO was in the form of amylose, we found that, compared with a Western diet, they were indeed capable of reducing BG, Ins, and lactate levels and, importantly, in slowing the growth of implanted murine and human tumors, with little or no effects on mouse weight.  6.3.1  Diet-Induced Biochemical Changes Associated with Reduced  Cancer Growth We assessed the effects of our low CHO diets in both murine tumor-bearing immunocompetent mice and human tumor-bearing immunocompromised mice, because immune status has been shown to influence tumor growth (Kroemer and Pouyssgeur, 2008), but found that low CHO diets slowed tumor growth to a similar extent without any difference in tumor-associated immune cell composition between the low CHO and 5058 groups (data not shown). Like others, we have shown that the reduction of Ins and BG correlated with reduced tumor growth and cancer incidence (Braun et al., 2011; Yuhara et al., 2011; Yang et al., 2004; Hemkens et al., 2009; Kim et al., 2011). There are, however, other mechanisms demonstrated by other groups that do not seem to play a role in our models. Of note, Venkateswaran and colleagues (2008) recently found that CHO reduction (from 45% to 10%) slowed the growth of LNCaP xenografts and attributed this to reduced Ins-like growth factor-I (IGF-I) levels. Interestingly, we detected no changes in IGF-I levels in mice on our low CHO diets, unless there was CR (e.g., with our 8% CHO diet; data not shown). Our findings suggest that although IGF-I reduction may be a relevant mechanism in some models, low CHO diets may also slow tumor growth in an IGF-I–independent manner. Ketosis is another biochemical change induced by low-CHO diets, and it has been reported that ketosis can occur when dietary CHO is reduced to 10% - 20%, a  148  level close to the diets we used (Masko et al., 2010). It was, therefore, surprising that we did not see elevated ketone bodies in our low-CHO fed mice. Indeed, tumorbearing mice on our 10% diet as well as NOP mice on our 15% CHO diet for many months had β-hydroxybutyrate levels not statistically different from that of mice fed a Western diet, and substantially less than those reported (∼15 mg/dL) for mice on NCKDs (Masko et al., 2010; Jornayvaz et al., 2010). While our results seem to be at odds with these previous reports, they are consistent with very recent studies showing that ketosis requires high dietary fat (Bielohuby et al., 2011). This also suggests that ketosis does not contribute to the slower tumor growth we observe with our low CHO diets.  6.3.2  Low CHO Diets as Complementary Treatments  We also found that our low CHO diets were additive with the tumor suppressive effects of CCI-779 and Celebrex. Related to this, while it has been shown that COX2 is overexpressed in many human cancers, and that Celebrex may be beneficial in preventing/slowing colon, breast (Basu et al., 2004), and prostate cancers (Brown et al., 2006; DuBois, 2006) by blocking both omega-6 fatty acid-induced inflammation (Brown et al., 2006) and tumor-induced angiogenesis (Wang and DuBois, 2004), high-dose Celebrex has cardiovascular side effects (Menter et al., 2010). As our low CHO diets show additive effects with Celebrex, it might allow for a lower, safer dose of Celebrex, without loss of therapeutic efficacy (McKellar and Singh, 2009).  6.3.3  Blood Glucose Regulation in Humans  Although our work strongly suggests that cancer can be treated and/or prevented by limiting BG, some caution must be exercised in extrapolating our results to humans. This is because, while fasting BG levels have been shown to be significantly reduced in cancer patients on a low CHO diet (Fearon et al., 1988), they may not be reduced as much as in mice (Brinkworth et al., 2009). On the contrary, substantial postprandial reductions in BG have been reported in humans on low CHO diets (Behall and Hallfrisch, 2002; Fearon et al., 1988; Moisey et al., 2008; Gannon and Nuttall, 2004).  149  Given that postprandial BG in humans is elevated for up to 2 hours after a meal and we typically eat 3 or more meals a day, it is very likely that a low CHO diet will significantly reduce the daily area-under-the-curve BG exposure. In keeping with this, it has been reported that low GI meals greatly reduce the BG area-under-thecurve compared with high GI meals in humans (Moisey et al., 2008) and that reducing the CHO content of meals in mild diabetics from 55% to 20% reduces the BG area-under-the-curve by 36%, which is similar to what we see with our mice (Gannon and Nuttall, 2004). Also, our low CHO studies with human HCT-116 cells in Rag2M mice and low CHO studies with other human tumors (Otto et al., 2008) suggest that there are no inherent differences between human and mouse cancer cells in their response to BG levels. Consistent with the notion that reducing BG in humans can be beneficial, there is a wealth of epidemiologic evidence showing a clear association between BG and/or Ins levels (which are determined by BG levels) and the incidence of human cancers (Jee et al., 2005; Stolzenberg-Solomon et al., 2005; Lajous et al., 2005; Keku et al., 2005; Gnagnarella et al., 2008; Mulholland et al., 2008; Mulholland et al., 2009). Thus, although our studies were conducted, out of necessity, with mice, the fact that human BG can be significantly reduced with low CHO diets and the association of many cancers with high BG levels suggest that our findings are very likely relevant to human cancers as well, particularly in cancers that have been associated with higher baseline BG and/or Ins levels, such as pancreatic (Jee et al., 2005; Stolzenberg-Solomon et al., 2005), breast (Lajous et al., 2005), colorectal (Keku et al., 2005), endometrial (Gnagnarella et al., 2008; Mulholland et al., 2008), and esophageal cancers (Mulholland et al., 2009). In addition to these cancers, a low CHO diet may also be beneficial in earlystage prostate cancer, even though it is not typically detectable by PET (Jadvar et al., 2003). This is because the metastases of these tumors kill the patients and, given the pivotal role of lactate in promoting metastasis (Gatenby et al., 2006), our low CHO diets could significantly reduce metastasis by reducing tumor-associated lactate levels.  150  6.3.4  Dietary Macronutrient Proportions  In terms of macronutrient composition, even though high protein has been shown to promote satiety (Anderson and Moore, 2004)—thus reducing obesity, BG, and Ins levels—and enhance both antitumor immunity, through amino acid supplementation, and life span (Evoy et al., 1998; Srivastava et al., 2010; D'Antona et al., 2010), we were concerned, based on the literature (Martin et al., 2005; Brenner et al., 1982; Food and Nutrition Board, 2005), that high protein levels might cause kidney damage. More recent data, however, suggest that this may only occur in individuals with existing chronic kidney disease (Martin et al., 2005; Higashiyama et al., 2010) and that in normal people, the increase in glomerular filtration rate and kidney cellularity that occur with long-term high protein consumption may be a normal response (Martin et al., 2005). Consistent with this, we found that while the 5 long-lived NOP mice on our 15% CHO diet had larger than normal kidneys (data not shown), only 1 had elevated urinary albumin. Moreover, because they lived beyond the normal life span of C57BL/6 mice on a Western diet, we can infer that the overall health of the mice was not adversely affected. In humans, most epidemiological studies examining high protein diets and cancer progression have been confounded by not taking into account protein source, fat content, and red meat consumption. This is important because high fat diets increase cancer risk (Zhang et al., 1999) and plant protein seems to decrease whereas animal protein increases cancer mortality (Fung et al., 2010). Interestingly, colonic cancer-inducing damage caused by red meats may be avoided with high amylose, low CHO diets (Toden et al., 2006). These studies suggest that macronutrient sources and combinations are very important and that testing them through highly controlled studies, such as those achieved with mice, represents a powerful approach to this question. Our study, herein, shows that a high amylose containing low CHO, high protein diet reduces BG, Ins, and glycolysis, slows tumor growth, reduces tumor incidence, and works additively with existing therapies without weight loss or kidney failure. Such a diet, therefore, has the potential of being both a novel cancer prophylactic and treatment, warranting further investigation of its applicability in the clinic, especially in combination with existing therapies.  151  7. Summary and Conclusions Uncontrolled proliferation has long been the gold standard for characterizing cancer cells. If left unchecked, these cells have the potential to proliferate indefinitely, ultimately giving rise to tumors, solid or liquid, which are often fatal to the host (Richard and Palai, 2010; Polaskis, 2007). As such, traditional cancer chemotherapeutic agents have aimed to exploit this characteristic by damaging DNA to selectively push actively proliferating cells towards apoptosis (Zhou et al., 2003; Dai and Grant, 2011; Ijichi et al. 2008). Because cancer cells are not the only cells in the host actively proliferating, chemotherapy also kills normal, rapidly proliferating cells, such as hemopoietic cells and the cells of the gut lining and hair follicle; this detrimental side-effect, causing a host of clinical symptoms (e.g., vomiting, hairloss), limits the tolerable dose that can be used in the clinic, lest a patient dies from the cure before the disease (Yuan et al., 2008; Ajani, 2008). Despite all of its disadvantages, chemotherapy regimens are still commonplace, even if therapeutic efficacy and patient response rates are not always ideal (Yeo et al. 2005; Yuan et al. 2008; Vermorken et al., 2008). Because direct cytotoxic therapies have reached a therapeutic plateau, much research is being carried out to elucidate other aspects of cancer biology to find specific, safe, and effective therapeutic targets.  7.1  Targeted Therapy and the Tumor Microenvironment Contemporary cancer therapy research is currently focusing on cancer-specific  mechanisms in order to minimize side effects, and so basic research into cancer cell biology, in both humans and animal models, has become the foundation upon which targeted therapies are being developed. As a result, a number of paradigm changes have occurred because of recent insights into critical proteins which cancers depend on for survival. For example, imatinib therapy, which involves a small molecule, Gleevec, binding to the ATP binding pocket of BCR-ABL, has shown great success in preventing the growth of chronic myelogenous leukemia cells (Breccia and Alimena, 2011). As well, our understanding of the role of mTOR in promoting the growth of many tumors that possess activated oncogenes (e.g., PI3K, Akt) and/or inactivated tumor-suppressors (e.g. PTEN) has led to a class of clinically effective 152  rapamycin analogs, which include temsirolimus and everolimus (Law, 2005; Eisen et al., 2011). Recently, the scope of research has extended beyond the biology of cancer cells themselves and into the support network of non-malignant cells and factors that comprise the tumor microenvironment. This has led to the identification of targetable host-cancer cell interactions, including those involving the immune system (Prehn and Prehn, 2008; Prehn and Main, 1957). A major hurdle in cancer immunotherapy is host immune cell-mediated immunosuppression, which dramatically reduces the efficacy of therapies such as cancer vaccines and adopted T cell transfer therapy that kill cancer cells (Schreiber et al., 2010; Yanelli and Wroblewski, 2004; Vasievich and Huang, 2011; Goldstein et al. 2012). Thus, attempts to reduce the immunosuppressive properties of immunosuppressive cells such as MDSCs, M2 Ms, Tregs, have become a very active area in immunotherapeutic research (Sinha et al., 2005a; Rodriguez et al., 2005; Goldstein et al., 2012). In particular, based on the hypothesis that tumor cells skew the host immune cell phenotype to help them survive and proliferate, there is a trend in research to reverse the programming of pro-tumor immune cells (e.g., M2 Ms) through methods such as TLR-agonist therapy or the inhibition of autocrine or paracrine immunosuppressive factors (e.g., IL-10) (Allavena and Mantovani, 2012). One such immunosuppressive mechanism is Arg1, which is an arginine catabolic enzyme expressed in MDSCs and murine M2 Ms, because it both enhances cancer growth through polyamine production and represses T H-1 mediated T cell immunity through L-arginine-depletion-induced T cell anergy (Minois et al., 2011; Mills, 2001; Bansal and Ochoa 2003; Munder, 2009). Moreover, M2 Ms are particularly interesting because they are present in many tumors of both human and mouse origin, and they suppress T H-1 immunity through the production of cytokines (e.g., IL-10, TGF-) and promote metastasis through the expression of MMPs (Rauh et al., 2005; Tjiu et al., 2009; Kurahara et al., 2011; Shirabe et al., 2011). Another aspect of the tumor microenvironment that is currently of interest is tumor metabolism, and, specifically, the Warburg effect. Although Warburg 153  discovered in the 1920's that cancer cells take up more glucose than normal tissue and preferentially break it down via glycolysis rather than OXPHOS, even in the presence of abundant O2, it wasn't until recently that this phenomenon was revisited as a possible therapeutic target (Pedersen, 2007). In particular, the use of NCKDs to treat prostate cancer is being tested on animal models as a way to limit IGF-1, a known prostate cancer promoting factor (Venkateswaran et al., 2007). Also, aiming at the induction of ketosis, NCKDs are being tested against brain tumors, since normal brain tissue, but not malignant tissue, can derive ATP from ketone bodies (Seyfried et al., 2011). Although there are no major studies looking at the impact of CHO on cancer progression in humans, there have been epidemiological surveys focusing on CHO consumption and mortality, including cancer mortality (Fung et al., 2011). In terms of glycolysis, its terminal metabolite, lactate, is becoming a center of tumor microenvironment research with the discovery that it, and pyruvate, act as energy shuttles between oxygenized and hypoxic tumor cells and thus are lynchpins for the co-dependent survival of these 2 types of tumor cells (Sonveaux et al., 2008; Feron, 2009).  7.2  Reversing Pro-Tumor M2 Ms In Chapter 3, we investigated immunosuppressive immune cells, focusing on  M2 Ms, and asked whether or not Ms polarized to a pro-tumor M2 phenotype could be phenotypically reversed to an anti-tumor M1 phenotype. Since aberrant M2-polarization is thought to promote intracellular microbial and viral infections as well as tumor growth, there is great value in assessing if M polarization is a permanent state or context dependent. Using IL-4 stimulated mouse BMMs as a model of a polarized M2 M, we demonstrated that even after IL-4 polarization for 3 or 6 days, and after the upregulation of Arg1, which is synonymous with the murine M2a M phenotype (Stein et al., 1992; Mosser and Edwards, 2008; Mantovani et al., 2002; Martinez et al., 2008), cytokine withdrawal was able to reduce the severity of this phenotype. Importantly, it was possible to M1-polarize formerly M2-polarized BMMs, which is a proof-of-principle for future research on reversing the M2 phenotype to treat disease. We also demonstrated that this phenotype reversal goes 154  beyond protein expression, but extends to polarized M functions, such as cytokine production and T cell stimulation. Our results demonstrate the feasibility of M reprogramming from one functional extreme to another and underscore the context dependence of M function that has been previously reported (Stout and Suttles, 2005). Our results also demonstrated interesting similarities between DCs and Ms. While DCs are well-known to require maturation by TLR ligands or cytokines for optimal Ag presentation to T cells (Wilson and O'Neill, 2003), we showed that BMMs stimulated with IL-4 stimulated T cell proliferation to a greater degree than unstimulated Ms. This is, perhaps, not surprising, since IL-4 has been shown to induce MHC-II, which is required for Ag presentation. This is, nevertheless, very similar to work showing cytokine-induced maturation of DCs augments Ag presentation. Also, Yao and colleagues (2005) have shown that IL-4 represses IL-10 in DCs to increase IL-12 production. While Siamon Gordon's group (Varin et al., 2010) has shown that in the presence of Neisseria meningitidis, IL-4 augments IL12p70 production in Ms, our results indicate that, like in DCs, an IL-4-induced increase in IL-12 production in Ms might be a general phenomenon not dependent on the presence of an infection. Although this is well beyond the scope of this thesis, the parallels in function and surface marker expression (i.e., CD11c) between DCs and IL-4-stimulated Ms shown in our results seem to blur the line between DCs and Ms.  7.3  Manipulating M2 M Generation in vivo Based on our 3 previous publications (Rauh et al., 2005; Kuroda et al., 2009;  Kuroda et al., 2011), we wanted to identify in Chapters 4 and 5 the elements responsible for the M2-skewing of Ms that take place in the SHIP-/- mouse. We also wanted to reconcile the 2 disparate mechanisms of M2-skewing (i.e. TGF- in MP and IL-4 from basophils) presented in our previous publications. Another aim of these chapters, having established that M2 Ms can be reprogrammed under the  155  right conditions, was to identify a treatment target from our in vitro assays to use in vivo to reduce the M2-skewing of Ms within the SHIP-/- mouse. We showed previously that SHIP-/-, but not SHIP+/+ BMM cultures supplemented with MP during M differentiation generated Arg1+ M2 Ms in a PI3K- and TGF--dependent manner (Rauh et al., 2005). In a subsequent publication, we demonstrated that IL-3-stimulated basophil or basophil progenitors (i.e., FcR+ c-kit- cells) in lin- SHIP-/-, but not SHIP+/+ BM, M2 skewed Ms during M differentiation because they secreted substantially more IL-4 in vitro than their SHIP+/+ counterparts, due to increased PI3K activity in the absence of SHIP (Kuroda et al., 2009). In Chapter 4, we successfully demonstrated a link between these 2 mechanisms by showing that MP-induced SHIP-/- M2-skewing was also dependent on the secretion of IL-4 by DX5+ basophil and basophil progenitors. We extended these findings by showing that IgG within MP can sensitize both mature Ms and M progenitors to respond more robustly to IL-4, such that a normally insufficient level of IL-4 was able to skew Ms to an M2 phenotype. Intriguingly, although we have not been able to detect IL-4 in either SHIP+/+ or SHIP/-  MP (unpublished data), a recent paper showed that there was a slight increase in  IL-4 in SHIP-/- MP. The level reported, however (i.e., ~15 pg/mL) (Maxwell et al., 2011) would not be sufficient to M2-skew Ms in the absence of IgG. Thus, our results complement theirs to explain why we see M2-skewing within the SHIP-/mouse. In Chapter 5, we identified the COX-2 pathway as a novel positive regulator of SHIP-/- M2-skewing both in vitro and in vivo, and our data suggest that it increases M2 skewing by enhancing the sensitivity of both mature Ms and Mprogenitors to IL-4. Tying in the role of TGF-, we also demonstrated that TGF- signaling was necessary for IL-4-induced Arg1 expression in MP-sensitized BMMs. Other groups have shown that TGF- induces the expression of COX-2 in both human and murine epithelial cells, and we showed, herein, that TGF- also induces COX-2 expression in mouse BMMs (Saha et al., 1999; Rodriguez-Barbero et al., 2006). Furthermore, the connection between COX-2 and IL-4 induction of Arg1 suggests that TGF- in  156  MP may be mediating its IL-4 sensitization effects through the induction of COX-2. A unified model of SHIP-/- M2-skewing is presented in Figure 7.1.  157  Figure 7.1 Model of M2-skewing in maturing Ms in vivo and in vitro. -/Proposed model of M2-skewing, in vivo and in vitro of SHIP Ms and M progenitors. Solid lines represent the changing status of progenitors or Ms after the influence exerted by factors added or secreted by cells represented by dashed lines. Black lines represent interactions discovered through BMENV assays and proposed as a model of the in vivo interactions. Gray lines represent external manipulations, which affect skewing. * = shown only in mature Ms 158  7.4  Manipulating the Metabolic Environment In a departure from reversing pro-tumor aspects of the immune system, we  tackled the tumor metabolic environment, specifically to limit a tumor‘s glucose supply by lowering dietary CHO. Our work showed, for the first time, that tumor growth could be reduced by lowering dietary CHO and replacing these calories lost with protein, without caloric restriction or weight loss, in our short term studies. In contrast to prior studies by other groups, which have concentrated on NCKDs or very high fat, low CHO diets (Masko et al., 2010; Venkateswaran et al., 2007), we showed that diets with 10% to 15% low glycemic index, amylose-based CHO, and moderate amounts of fat (~25%) can effectively slow tumor growth without ketosis. Also, in contrast to these previous studies, which focused on prostate cancer, we showed the effectiveness of our low CHO, high protein diets in mouse models of both head and neck and lung carcinomas in immunocompetent mice, as well as in human colorectal xenographs in immunocompromised mice. Furthermore, and again in contrast with earlier studies (Venkateswaran et al., 2007), we found that decreases in BG, lactate, and Ins correlated with slower cancer progression, without a drop in IGF-1. This suggested that low CHO diets could be slowing tumor progression by simply limiting BG and glycolysis, and that low CHO diets as a cancer treatment could be applicable beyond cancers that depended on high IGF-1 levels (e.g., prostate cancers). In terms of low CHO diets and cancer treatment, we demonstrated that they can work additively with anti-cancer agents (e.g. Temsirolimus, Celebrex) to enhance the therapeutic efficacy of these drugs. This suggests that low CHO diets may potentially be a valuable supplementary treatment to standard cancer therapies in a clinical setting, since toxicity is a limiting factor in treatment dosage, even for targeted therapies (Yuan et al., 2008; Ajani, 2008; Law, 2005; Eisen et al., 2011). Encouragingly, we also showed in a lifetime study that HER2/Neu-expressing mice, which develop spontaneous breast cancer, have a substantially lower cancer incidence when put on our low CHO diet. This demonstrates that low CHO diets may also be useful as a prophylactic.  159  7.5  Limitations of This Thesis Although  we  successfully  demonstrated  the  plasticity  of  the  tumor  microenvironment in a mouse model, there are some limitations to this research. Firstly, in regards to the M studies, we selected Arg1+, IL-4-induced Ms as our model M2 M, specifically focusing on immune regulatory and L-arginine metabolic functions. While this is how the M1-M2 dichotomy was originally described, it is now apparent that other M2 M subsets, which are IL-4-independent, have a significant impact on immunosuppression in cancer and cancer progression as a whole (Mosser and Edwards, 2008; Martinez et al., 2008). For example, while IL-4-induced M2a Ms support tumor growth and subvert immunity by producing less proinflammatory cytokines and by skewing L-arginine use away from the production of tumoricidal NO towards pro-tumor polyamines, M2b Ms, which are induced by immune complexes and TLR signaling, are implicated in VEGF production and tumor angiogenesis (Martinez, et al., 2008; Barbara-Guillem et al., 2002). A third type of M2 M, M2c, which is induced by M deactivating factors (e.g., TGF-, IL10, PGE2) and mediates its effects via IL-10 has been implicated in the progression of intestinal polyps in a genetically predisposed mouse model (Nakanishi et al., 2011). Because of the functional differences amongst these subsets, our results, especially those on the reversibility of the M2-phenotype may not be applicable to all M2 M subsets. Based on the results from Chapters 4 and 5, we present a unified model of M M2-skewing (Fig. 7.2). While we demonstrated a requirement for COX-2 activity for MP-induced M2 skewing and showed that TGF- upregulated COX-2 protein, there are still many questions left unanswered. First, we have yet to show definitively that PGE2 is the biologically relevant molecule in vivo, since COX-2 is responsible for the synthesis of many other inflammatory factors (Simmons et al., 2004). Also, it will be interesting to determine the source of COX-2 activity (i.e., basophils or Ms). Furthermore, because the function COX-2 depends on its intracellular location (Yamashita et al., 2007), it would also be interesting to examine its distribution within COX-2+ cells, particularly after TGF- stimulation.  160  Figure 7.2 Unified Model of M2-skewing in vivo. Proposed model of M2-skewing, in vivo, in the presence of constitutive, low levels of Hi + IL-4 from growth factor-(e.g. tumor-derived growth factors)-stimulated PI3K DX5 cells (e.g., basophils), and the possible roles of IgG, TGF-, and COX-2 in M2skewing, and the theoretical impact of inhibitors to these pathways on M2-skewing. With regard to our low CHO diet studies, we felt it was important to keep our diets isocaloric, lest CR slows tumor growth and confound our results. Thus, we had to raise either fat or protein content to compensate for lowering the CHO. Because of the association between high fat and cancer risk, and the documented immune benefits of amino acid supplementation, we chose to raise protein levels (Zhang et al., 1999; Evoy et al., 1998; Srivastava et al., 2010; D'Antona et al., 2010). In so doing, however, we may have boosted immune function within the tumor 161  microenvironment. Thus, it is important to keep in mind that our low CHO diets may be conferring an immune benefit (e.g., boosting T cell responses by offsetting amino acid depletion). While this is certainly beneficial as a treatment, it makes our interpretation of dietary CHO reduction on tumor growth slightly more difficult. However, given that our low CHO diets are effective even in T cell deficient RAG2M mice, and given the success of high fat, low CHO diets used by other groups at slowing tumor growth without increasing dietary protein (Venkateswaran et al., 2007; Masko et al., 2010), it is likely the reduction of dietary CHO, BG, and Ins are the most important factors in tumor growth reduction in our studies.  7.5.1  In Translation  While we focused on Arg1 as a marker for M2 Ms because of its established reliability as such in the mouse system (Mosser and Edwards, 2008; Allavena and Mantovani, 2012), human Ms exposed to the canonical M2-skewing cytokines, IL-4 and IL-13 do not express Arg1 in vitro (Raes et al., 2005). Also, iNOS expression and NO production in human Ms are also controversial, since LPS + IFN- do not induce iNOS or NO from human Ms in vitro (Schneemann et al., 1993). However, it has also been shown that human Ms do, in fact, produce biologically relevant NO in some contexts, and Arg1 expression can be elicited from human Ms with costimulants that increase cyclic adenosine monophosphate, which intriguingly enough is a secondary messenger downstream of PGE2 signaling (Erdely et al., 2006; Voudoukis et al., 1997). Under homeostatic conditions, human granulocytes express Arg1 in their granules, and in metastatic renal cell carcinoma (RCC) patients, Arg1 + MDSCs accumulate in their peripheral blood (Ochoa et al., 2007a; Munder et al., 2006). Furthermore, MDSC Arg1 is dependent on COX-2 activity and associated with Larginine depletion-mediated T cell anergy and reduced IFN- in RCC patients (Ochoa et al., 2007a). Importantly, human granulocyte release of Arg1 has a similar capacity to induce T cell anergy as mouse M Arg1 (Ochoa et al., 2007a; Munder et al., 2006). Intriguingly, it seems that Arg1 induction in human granulocytes occurs during maturation, since isolated mature granulocyte Arg1 expression cannot be 162  modulated with pro- or anti-inflammatory cytokines in vitro (Munder et al., 2005). In our work, herein, we've uncovered a similar mode of in vivo Arg1 induction in the SHIP-/- mouse that is COX-2 dependent in maturing Ms. Nonetheless, the differences between mice and humans caution us against the direct translation of the findings in this thesis. However, since the iNOS-Arg1 immunosuppression axis is very relevant in the human system, albeit in MDSCs and granulocytes instead of Ms, we feel that our focus on Arg1 was warranted. In our short-term low CHO studies, we used primary tumor size as a readout for cancer progression, and because of humane end point requirements, mice had to be euthanized when tumors reached 1000 - 1500 mm3. While this is a standard measure of cancer progression, it may not be reflective of cancer mortality, since metastasis is the primary cause of cancer death (Weigelt et al., 2005). Therefore, to fully assess the clinical potential of low CHO diets, we must keep in mind that our experiments did not test the impact of low CHO diets on metastasis.  7.6  Future Directions The work in this thesis demonstrates that both the cellular and non-cellular  elements of the tumor microenvironment can be manipulated, and in the case of our low CHO studies, may be an effective approach to cancer treatment. The M skewing studies also underscore the power of rational drug design, or in this case, rational therapy design, since we started with an in vivo phenomenon, then identified critical elements in vitro, and chose an appropriate intervention to reverse the M2 Mskewing in SHIP-/- mice in vivo. Similarly, we designed low CHO diets based on the Warburg effect to reduce the tumor fuel supply, and found that this approach was, indeed, effective. We showed that DX5+ cells, which we hypothesize are basophil and basophil progenitors, are part of a network of interactions involving COX-2, TGF-, and IgG that are responsible for skewing Ms within SHIP-/- mice to an M2 phenotype. While we demonstrated that basophil depletion in the inducible SHIP -/- mouse after SHIP1 deletion was ineffective at ameliorating this M2-skewing, perhaps we can better  163  assess the impact of these cells by inducing SHIP deletion after basophil depletion to remove the confounding factor of lingering M2 Ms. It is interesting to note that our results regarding M2-skewing within the SHIP-/mouse suggested that all of the 3 M2 Msubsets were involved, i.e., M2a (IL-4), M2b (IgG) and M2c (COX-2 and TGF-). This, in turn, suggests that M2 Ms skewed in vivo may be a combination of the three subsets. Thus, to investigate, in vitro, an M2 M that is reflective of the in vivo situation, we may need to test M2 Ms generated with a combined stimulus. As our data, at best, only hint at a link amongst COX-2, IL-4, IgG, and TGF- for in vivo M2 induction, as assessed by ex vivo PMs and in vivo mimicking BMENV assays, it would be interesting to elucidate the relationship amongst these elements with in vivo M2 Ms to see if rather than discreet subsets, they are a result of combinatorial stimulation. This will give us a better understanding of a 'true' M2 M. Furthermore, it would be interesting to know whether or not the same M2 induction circuit is involved in the M2-skewing of TAMs. In terms of our low CHO studies, we are keenly interested in combining the manipulation of cellular (i.e., immune cells) and non-cellular (i.e., glucose metabolism) elements of the tumor microenvironment. Specifically, we want to assess the impact of combining low CHO diets with immunotherapy. We have shown, herein, that Celebrex collaborates with our low CHO diets to slow tumor growth. Since we show that Celebrex is effective at reducing the M2-skewing in the SHIP-/- mouse, it would be interesting to see if Celebrex impacts the TAM M2 phenotype and if this plays a role in reducing tumor growth. Related to this, it has been shown that M transcription of M2-associated genes can be reduced and M1associated genes increased with Celebrex treatment in a mouse intestinal polyp model (Nakanishi et al., 2011). Also, we feel that while both glucose limitation and M2 M phenotype reversal may slow the growth of tumors, they may not actively destroy tumor cells. It is, therefore, of great interest to us to see if our low CHO diets can boost the therapeutic efficacy of an immune activating agent, such as the TLR1/TLR2 ligand, bacterial lipoprotein (BLP), which has been shown to induce tumor regression in a mouse model (Zhang et al., 2011).  164  The tumor microenvironment encompasses a plethora of interacting factors and cells, all of which directly and indirectly influence the growth and survival of tumor cells. We investigated two aspects of this environment. We feel that continuing  with  a  similar  multi-faceted  approach  to  targeting  the  tumor  microenvironment will lead to therapies that will someday be able to manage or even cure cancer.  165  References Ajani, J.A. (2008). Optimizing docetaxel chemotherapy in patients with cancer of the gastric and gastroesophageal junction: evolution of the docetaxel, cisplatin, and 5-fluorouracil regimen. Cancer 113, 945-955. Akerstrom, B., and Bjorck, L. (1986). A physicochemical study of protein G, a molecule with unique immunoglobulin G-binding properties. J Biol Chem 261, 10240-10247. Albina, J.E., and Reichner, J.S. (2003). Oxygen and the regulation of gene expression in wounds. Wound Repair Regen 11, 445-451. Aliaga, J.C., Deschenes, C., Beaulieu, J.F., Calvo, E.L., and Rivard, N. (1999). Requirement of the MAP kinase cascade for cell cycle progression and differentiation of human intestinal cells. Am J Physiol 277, G631-641. Allavena, P., and Mantovani, A. (2012). Immunology in the clinic review series; focus on cancer: tumour-associated macrophages: undisputed stars of the inflammatory tumour microenvironment. Clin Exp Immunol 167, 195-205. Almand, B., Resser, J.R., Lindman, B., Nadaf, S., Clark, J.I., Kwon, E.D., Carbone, D.P., and Gabrilovich, D.I. (2000). Clinical significance of defective dendritic cell differentiation in cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 6, 1755-1766. Alshaker, H.A., and Matalka, K.Z. (2011). IFN-gamma, IL-17 and TGF-beta involvement in shaping the tumor microenvironment: The significance of modulating such cytokines in treating malignant solid tumors. Cancer Cell Int 11, 33. Anderson, G.H., and Moore, S.E. (2004). Dietary proteins in the regulation of food intake and body weight in humans. J Nutr 134, 974S-979S. Apostolopoulos, V., Pietersz, G.A., Gordon, S., Martinez-Pomares, L., and McKenzie, I.F. (2000). Aldehyde-mannan antigen complexes target the MHC class I antigen-presentation pathway. European journal of immunology 30, 1714-1723. Ashcroft, G.S. (1999). Bidirectional regulation of macrophage function by TGF-beta. Microbes and infection / Institut Pasteur 1, 1275-1282. Ashiru, O., Boutet, P., Fernandez-Messina, L., Aguera-Gonzalez, S., Skepper, J.N., Vales-Gomez, M., and Reyburn, H.T. (2010). Natural killer cell cytotoxicity is suppressed by exposure to the human NKG2D ligand MICA*008 that is shed by tumor cells in exosomes. Cancer Res 70, 481-489.  166  Auffray, C., Fogg, D., Garfa, M., Elain, G., Join-Lambert, O., Kayal, S., Sarnacki, S., Cumano, A., Lauvau, G., and Geissmann, F. (2007). Monitoring of blood vessels and tissues by a population of monocytes with patrolling behavior. Science 317, 666-670. Bansal, V., and Ochoa, J.B. (2003). Arginine availability, arginase, and the immune response. Curr Opin Clin Nutr Metab Care 6, 223-228. Barbera-Guillem, E., Nyhus, J.K., Wolford, C.C., Friece, C.R., and Sampsel, J.W. (2002). Vascular endothelial growth factor secretion by tumor-infiltrating macrophages essentially supports tumor angiogenesis, and IgG immune complexes potentiate the process. Cancer Res 62, 7042-7049. Barnden, M.J., Allison, J., Heath, W.R., and Carbone, F.R. (1998). Defective TCR expression in transgenic mice constructed using cDNA-based alpha- and beta-chain genes under the control of heterologous regulatory elements. Immunol Cell Biol 76, 34-40. Basu, G.D., Pathangey, L.B., Tinder, T.L., Lagioia, M., Gendler, S.J., and Mukherjee, P. (2004). Cyclooxygenase-2 inhibitor induces apoptosis in breast cancer cells in an in vivo model of spontaneous metastatic breast cancer. Mol Cancer Res 2, 632-642. Bauer, D.E., Harris, M.H., Plas, D.R., Lum, J.J., Hammerman, P.S., Rathmell, J.C., Riley, J.L., and Thompson, C.B. (2004). Cytokine stimulation of aerobic glycolysis in hematopoietic cells exceeds proliferative demand. Faseb J 18, 1303-1305. Behall, K.M., and Hallfrisch, J. (2002). Plasma glucose and insulin reduction after consumption of breads varying in amylose content. Eur J Clin Nutr 56, 913920. Beissert, S., Hosoi, J., Grabbe, S., Asahina, A., and Granstein, R.D. (1995). IL-10 inhibits tumor antigen presentation by epidermal antigen-presenting cells. J Immunol 154, 1280-1286. Berg, J.M.T., J.L.; Stryer, L., ed. (2002). Biochemistry 5th Edition, 5th Edition edn (New York: W H Freeman). Bernacki, R.J., Oberman, E.J., Seweryniak, K.E., Atwood, A., Bergeron, R.J., and Porter, C.W. (1995). Preclinical antitumor efficacy of the polyamine analogue N1, N11-diethylnorspermine administered by multiple injection or continuous infusion. Clin Cancer Res 1, 847-857. Bielohuby, M., Menhofer, D., Kirchner, H., Stoehr, B.J., Muller, T.D., Stock, P., Hempel, M., Stemmer, K., Pfluger, P.T., Kienzle, E., et al. (2010). Induction of ketosis in rats fed low-carbohydrate, high fat diets depends on the relative abundance of dietary fat and protein. Am J Physiol Endocrinol Metab. 167  Bierie, B., and Moses, H.L. (2010). Transforming growth factor beta (TGF-beta) and inflammation in cancer. Cytokine Growth Factor Rev 21, 49-59. Biswas, S.K., Gangi, L., Paul, S., Schioppa, T., Saccani, A., Sironi, M., Bottazzi, B., Doni, A., Vincenzo, B., Pasqualini, F., et al. (2006). A distinct and unique transcriptional program expressed by tumor-associated macrophages (defective NF-kappaB and enhanced IRF-3/STAT1 activation). Blood 107, 2112-2122. Bogdan, C., Vodovotz, Y., Paik, J., Xie, Q.W., and Nathan, C. (1994). Mechanism of suppression of nitric oxide synthase expression by interleukin-4 in primary mouse macrophages. Journal of leukocyte biology 55, 227-233. Bottazzi, B., Polentarutti, N., Acero, R., Balsari, A., Boraschi, D., Ghezzi, P., Salmona, M., and Mantovani, A. (1983). Regulation of the macrophage content of neoplasms by chemoattractants. Science 220, 210-212. Bragoszewski, P., Kupryjanczyk, J., Bartnik, E., Rachinger, A., and Ostrowski, J. (2008). Limited clinical relevance of mitochondrial DNA mutation and gene expression analyses in ovarian cancer. BMC Cancer 8, 292. Brand-Miller, J., and Buyken, A.E. (2011). The glycemic index issue. Curr Opin Lipidol. Braun, S., Bitton-Worms, K., and Leroith, D. (2011). The Link between the Metabolic Syndrome and Cancer. Int J Biol Sci 7, 1003-1015. Breccia, M., and Alimena, G. (2011). The current role of high-dose imatinib in chronic myeloid leukemia patients, newly diagnosed or resistant to standard dose. Expert Opin Pharmacother 12, 2075-2087. Brenner, B.M., Meyer, T.W., and Hostetter, T.H. (1982). Dietary protein intake and the progressive nature of kidney disease: the role of hemodynamically mediated glomerular injury in the pathogenesis of progressive glomerular sclerosis in aging, renal ablation, and intrinsic renal disease. The New England journal of medicine 307, 652-659. Brinkworth, G.D., Noakes, M., Buckley, J.D., Keogh, J.B., and Clifton, P.M. (2009). Long-term effects of a very-low-carbohydrate weight loss diet compared with an isocaloric low-fat diet after 12 mo. Am J Clin Nutr 90, 23-32. Brito, C., Naviliat, M., Tiscornia, A.C., Vuillier, F., Gualco, G., Dighiero, G., Radi, R., and Cayota, A.M. (1999). Peroxynitrite inhibits T lymphocyte activation and proliferation by promoting impairment of tyrosine phosphorylation and peroxynitrite-driven apoptotic death. J Immunol 162, 3356-3366.  168  Brown, M.D., Hart, C.A., Gazi, E., Bagley, S., and Clarke, N.W. (2006). Promotion of prostatic metastatic migration towards human bone marrow stoma by Omega 6 and its inhibition by Omega 3 PUFAs. British journal of cancer 94, 842-853. Buhtoiarov, I.N., Sondel, P.M., Eickhoff, J.C., and Rakhmilevich, A.L. (2007). Macrophages are essential for antitumour effects against weakly immunogenic murine tumours induced by class B CpGoligodeoxynucleotides. Immunology 120, 412-423. Buttgereit, F., and Brand, M.D. (1995). A hierarchy of ATP-consuming processes in mammalian cells. Biochem J 312 ( Pt 1), 163-167. Bygdeman, M. (2003). Pharmacokinetics of prostaglandins. Best Pract Res Clin Obstet Gynaecol 17, 707-716. Chang, C.I., Liao, J.C., and Kuo, L. (1998). Arginase modulates nitric oxide production in activated macrophages. Am J Physiol 274, H342-348. Charles, N., Watford, W.T., Ramos, H.L., Hellman, L., Oettgen, H.C., Gomez, G., Ryan, J.J., O'Shea, J.J., and Rivera, J. (2009). Lyn kinase controls basophil GATA-3 transcription factor expression and induction of Th2 cell differentiation. Immunity 30, 533-543. Conti, P., Kempuraj, D., Kandere, K., Di Gioacchino, M., Barbacane, R.C., Castellani, M.L., Felaco, M., Boucher, W., Letourneau, R., and Theoharides, T.C. (2003). IL-10, an inflammatory/inhibitory cytokine, but not always. Immunol Lett 86, 123-129. Cordain, L., Eaton, S.B., Sebastian, A., Mann, N., Lindeberg, S., Watkins, B.A., O'Keefe, J.H., and Brand-Miller, J. (2005). Origins and evolution of the Western diet: health implications for the 21st century. Am J Clin Nutr 81, 341354. Cosper, P.F., and Leinwand, L.A. (2011). Cancer causes cardiac atrophy and autophagy in a sexually dimorphic manner. Cancer Res 71, 1710-1720. Czeh, M., Loddenkemper, C., Shalapour, S., Schon, C., Robine, S., Goldscheid, E., Stein, H., Schuler, T., Willimsky, G., and Blankenstein, T. (2010). The immune response to sporadic colorectal cancer in a novel mouse model. Oncogene 29, 6591-6602. D'Antona, G., Ragni, M., Cardile, A., Tedesco, L., Dossena, M., Bruttini, F., Caliaro, F., Corsetti, G., Bottinelli, R., Carruba, M.O., et al. (2010). Branched-chain amino acid supplementation promotes survival and supports cardiac and skeletal muscle mitochondrial biogenesis in middle-aged mice. Cell metabolism 12, 362-372.  169  Dai, Y., and Grant, S. (2011). Methods to Study Cancer Therapeutic Drugs That Target Cell Cycle Checkpoints. In Methods in Molecular Biology, W.X. Li, ed. (Springer Science+Business Media), pp. 257-304. Damen, J.E., Liu, L., Ware, M.D., Ermolaeva, M., Majerus, P.W., and Krystal, G. (1998). Multiple forms of the SH2-containing inositol phosphatase, SHIP, are generated by C-terminal truncation. Blood 92, 1199-1205. Dang, C.V., Hamaker, M., Sun, P., Le, A., and Gao, P. (2011). Therapeutic targeting of cancer cell metabolism. J Mol Med (Berl) 89, 205-212. De Boniface, J., Poschke, I., Mao, Y., and Kiessling, R. (2011). Tumor-dependent down-regulation of the zeta-chain in T-cells is detectable in early breast cancer and correlates with immune cell function. Int J Cancer. Dhanapal, R., Saraswathi, T., and Govind, R.N. (2011). Cancer cachexia. J Oral Maxillofac Pathol 15, 257-260. Diaz-Ruiz, R., Uribe-Carvajal, S., Devin, A., and Rigoulet, M. (2009). Tumor cell energy metabolism and its common features with yeast metabolism. Biochim Biophys Acta 1796, 252-265. Ding, L., Linsley, P.S., Huang, L.Y., Germain, R.N., and Shevach, E.M. (1993). IL-10 inhibits macrophage costimulatory activity by selectively inhibiting the upregulation of B7 expression. J Immunol 151, 1224-1234. Doyle, A.G., Herbein, G., Montaner, L.J., Minty, A.J., Caput, D., Ferrara, P., and Gordon, S. (1994). Interleukin-13 alters the activation state of murine macrophages in vitro: comparison with interleukin-4 and interferon-gamma. European journal of immunology 24, 1441-1445. DuBois, R.N. (2006). Cyclooxygenase-2 selective inhibitors and prostate cancer: what is the clinical benefit? Journal of clinical oncology : official journal of the American Society of Clinical Oncology 24, 2691-2693. Dudkina, N.V., Kouril, R., Peters, K., Braun, H.P., and Boekema, E.J. (2010). Structure and function of mitochondrial supercomplexes. Biochim Biophys Acta 1797, 664-670. Dunn, G.P., Bruce, A.T., Ikeda, H., Old, L.J., and Schreiber, R.D. (2002). Cancer immunoediting: from immunosurveillance to tumor escape. Nat Immunol 3, 991-998. Edwards, J.P., and Emens, L.A. (2010). The multikinase inhibitor sorafenib reverses the suppression of IL-12 and enhancement of IL-10 by PGE in murine macrophages. Int Immunopharmacol 10, 1220-1228.  170  Edwards, J.P., Zhang, X., Frauwirth, K.A., and Mosser, D.M. (2006). Biochemical and functional characterization of three activated macrophage populations. Journal of leukocyte biology 80, 1298-1307. Eisen, T., Sternberg, C.N., Robert, C., Mulders, P., Pyle, L., Zbinden, S., Izzedine, H., and Escudier, B. (2012). Targeted Therapies for Renal Cell Carcinoma: Review of Adverse Event Management Strategies. J Natl Cancer Inst. Erdely, A., Kepka-Lenhart, D., Clark, M., Zeidler-Erdely, P., Poljakovic, M., Calhoun, W.J., and Morris, S.M., Jr. (2006). Inhibition of phosphodiesterase 4 amplifies cytokine-dependent induction of arginase in macrophages. Am J Physiol Lung Cell Mol Physiol 290, L534-539. Eruslanov, E., Daurkin, I., Ortiz, J., Vieweg, J., and Kusmartsev, S. (2010). Pivotal Advance: Tumor-mediated induction of myeloid-derived suppressor cells and M2-polarized macrophages by altering intracellular PGE catabolism in myeloid cells. Journal of leukocyte biology 88, 839-848. Evoy, D., Lieberman, M.D., Fahey, T.J., 3rd, and Daly, J.M. (1998). Immunonutrition: the role of arginine. Nutrition 14, 611-617. Faget, J., Biota, C., Bachelot, T., Gobert, M., Treilleux, I., Goutagny, N., Durand, I., Leon-Goddard, S., Blay, J.Y., Caux, C., and Menetrier-Caux, C. (2011). Early detection of tumor cells by innate immune cells leads to Treg recruitment through CCL22 production by tumor cells. Cancer Res. Fantin, V.R., St-Pierre, J., and Leder, P. (2006). Attenuation of LDH-A expression uncovers a link between glycolysis, mitochondrial physiology, and tumor maintenance. Cancer Cell 9, 425-434. Fearon, K.C., Borland, W., Preston, T., Tisdale, M.J., Shenkin, A., and Calman, K.C. (1988). Cancer cachexia: influence of systemic ketosis on substrate levels and nitrogen metabolism. Am J Clin Nutr 47, 42-48. Feron, O. (2009). Pyruvate into lactate and back: from the Warburg effect to symbiotic energy fuel exchange in cancer cells. Radiother Oncol 92, 329-333. Fichtner-Feigl, S., Terabe, M., Kitani, A., Young, C.A., Fuss, I., Geissler, E.K., Schlitt, H.J., Berzofsky, J.A., and Strober, W. (2008). Restoration of tumor immunosurveillance via targeting of interleukin-13 receptor-alpha 2. Cancer Res 68, 3467-3475. Fico, A., Paglialunga, F., Cigliano, L., Abrescia, P., Verde, P., Martini, G., Iaccarino, I., and Filosa, S. (2004). Glucose-6-phosphate dehydrogenase plays a crucial role in protection from redox-stress-induced apoptosis. Cell Death Differ 11, 823-831.  171  Fiorentino, D.F., Zlotnik, A., Vieira, P., Mosmann, T.R., Howard, M., Moore, K.W., and O'Garra, A. (1991). IL-10 acts on the antigen-presenting cell to inhibit cytokine production by Th1 cells. J Immunol 146, 3444-3451. Fischer-Posovszky, P., Wang, Q.A., Asterholm, I.W., Rutkowski, J.M., and Scherer, P.E. (2011). Targeted deletion of adipocytes by apoptosis leads to adipose tissue recruitment of alternatively activated M2 macrophages. Endocrinology 152, 3074-3081. Fischer, K., Hoffmann, P., Voelkl, S., Meidenbauer, N., Ammer, J., Edinger, M., Gottfried, E., Schwarz, S., Rothe, G., Hoves, S., et al. (2007). Inhibitory effect of tumor cell-derived lactic acid on human T cells. Blood 109, 3812-3819. Food and Nutrition Board, I.o.M., ed. (2005). Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein and amino acids (macronutrients) (Washtington, DC: The National Academies Press). Fox, C.J., Hammerman, P.S., and Thompson, C.B. (2005). Fuel feeds function: energy metabolism and the T-cell response. Nat Rev Immunol 5, 844-852. Franke, T.F., Kaplan, D.R., and Cantley, L.C. (1997). PI3K: downstream AKTion blocks apoptosis. Cell 88, 435-437. Freedland, S.J., Mavropoulos, J., Wang, A., Darshan, M., Demark-Wahnefried, W., Aronson, W.J., Cohen, P., Hwang, D., Peterson, B., Fields, T., et al. (2008). Carbohydrate restriction, prostate cancer growth, and the insulin-like growth factor axis. Prostate 68, 11-19. Fung, T.T., van Dam, R.M., Hankinson, S.E., Stampfer, M., Willett, W.C., and Hu, F.B. (2010). Low-carbohydrate diets and all-cause and cause-specific mortality: two cohort studies. Annals of internal medicine 153, 289-298. Gallardo-Soler, A., Gomez-Nieto, C., Campo, M.L., Marathe, C., Tontonoz, P., Castrillo, A., and Corraliza, I. (2008). Arginase I induction by modified lipoproteins in macrophages: a peroxisome proliferator-activated receptorgamma/delta-mediated effect that links lipid metabolism and immunity. Molecular endocrinology 22, 1394-1402. Gambhir, S.S. (2002). Molecular imaging of cancer with positron emission tomography. Nat Rev Cancer 2, 683-693. Ganapathy, V., Thangaraju, M., and Prasad, P.D. (2009). Nutrient transporters in cancer: relevance to Warburg hypothesis and beyond. Pharmacol Ther 121, 29-40. Gannon, M.C., and Nuttall, F.Q. (2004). Effect of a high-protein, low-carbohydrate diet on blood glucose control in people with type 2 diabetes. Diabetes 53, 2375-2382. 172  Gatenby, R.A., Gawlinski, E.T., Gmitro, A.F., Kaylor, B., and Gillies, R.J. (2006). Acid-mediated tumor invasion: a multidisciplinary study. Cancer Res 66, 5216-5223. Gatenby, R.A., and Gillies, R.J. (2004). Why do cancers have high aerobic glycolysis? Nat Rev Cancer 4, 891-899. Geissmann, F., Jung, S., and Littman, D.R. (2003). Blood monocytes consist of two principal subsets with distinct migratory properties. Immunity 19, 71-82. Geissmann, F., Manz, M.G., Jung, S., Sieweke, M.H., Merad, M., and Ley, K. (2010). Development of monocytes, macrophages, and dendritic cells. Science 327, 656-661. Ghosn, E.E., Cassado, A.A., Govoni, G.R., Fukuhara, T., Yang, Y., Monack, D.M., Bortoluci, K.R., Almeida, S.R., and Herzenberg, L.A. (2010). Two physically, functionally, and developmentally distinct peritoneal macrophage subsets. Proc Natl Acad Sci U S A 107, 2568-2573. Gillies, R.J., Raghunand, N., Karczmar, G.S., and Bhujwalla, Z.M. (2002). MRI of the tumor microenvironment. J Magn Reson Imaging 16, 430-450. Gimborn, K., Lessmann, E., Kuppig, S., Krystal, G., and Huber, M. (2005). SHIP down-regulates FcepsilonR1-induced degranulation at supraoptimal IgE or antigen levels. J Immunol 174, 507-516. Gnagnarella, P., Gandini, S., La Vecchia, C., and Maisonneuve, P. (2008). Glycemic index, glycemic load, and cancer risk: a meta-analysis. Am J Clin Nutr 87, 1793-1801. Godsland, I.F. (2010). Insulin resistance and hyperinsulinaemia in the development and progression of cancer. Clin Sci (Lond) 118, 315-332. Gogebakan, O., Kohl, A., Osterhoff, M.A., van Baak, M.A., Jebb, S.A., Papadaki, A., Martinez, J.A., Handjieva-Darlenska, T., Hlavaty, P., Weickert, M.O., et al. (2011). Effects of Weight Loss and Long-Term Weight Maintenance With Diets Varying in Protein and Glycemic Index on Cardiovascular Risk Factors: The Diet, Obesity, and Genes (DiOGenes) Study: A Randomized, Controlled Trial. Circulation 124, 2829-2838. Goldstein, M.J., Kohrt, H.E., Houot, R., Varghese, B., Lin, J.T., Swanson, E., and Levy, R. (2012). Adoptive Cell Therapy for Lymphoma with CD4 T Cells Depleted of CD137 Expressing Regulatory T Cells. Cancer Res. Gordon, S. (1999). Macrophage-restricted molecules: role in differentiation and activation. Immunol Lett 65, 5-8.  173  Gordon, S., and Martinez, F.O. (2010). Alternative activation of macrophages: mechanism and functions. Immunity 32, 593-604. Gorelik, E., Edwards, R.P., Feng, X., Marrangoni, A.M., Grandis, J.R., Drenning, S.D., Velikokhatnaya, L., Kwon, J.A., and Lokshin, A.E. (2004). IL-12 receptor-mediated upregulation of FasL in human ovarian carcinoma cells. Int J Cancer 112, 620-627. Gottfried, E., Kunz-Schughart, L.A., Ebner, S., Mueller-Klieser, W., Hoves, S., Andreesen, R., Mackensen, A., and Kreutz, M. (2006). Tumor-derived lactic acid modulates dendritic cell activation and antigen expression. Blood 107, 2013-2021. Gough, P.J., and Gordon, S. (2000). The role of scavenger receptors in the innate immune system. Microbes and infection / Institut Pasteur 2, 305-311. Gregoire, C., Chasson, L., Luci, C., Tomasello, E., Geissmann, F., Vivier, E., and Walzer, T. (2007). The trafficking of natural killer cells. Immunol Rev 220, 169-182. Guarente, L. (2008). Mitochondria--a nexus for aging, calorie restriction, and sirtuins? Cell 132, 171-176. Halestrap, A.P., and Meredith, D. (2004). The SLC16 gene family-from monocarboxylate transporters (MCTs) to aromatic amino acid transporters and beyond. Pflugers Arch 447, 619-628. Hamilton, M.J., Antignano, F., von Rossum, A., Boucher, J.L., Bennewith, K.L., and Krystal, G. (2010). TLR agonists that induce IFN-beta abrogate resident macrophage suppression of T cells. J Immunol 185, 4545-4553. Han, W.G., van der Voort, E.I., el Bannoudi, H., Louis-Plence, P., Huizinga, T.W., and Toes, R.E. (2010). DX5(+)CD4(+) T cells modulate cytokine production by CD4(+) T cells towards IL-10 via the production of IL-4. European journal of immunology 40, 2731-2740. Hanahan, D., and Weinberg, R.A. (2000). The hallmarks of cancer. Cell 100, 57-70. Hanahan, D., and Weinberg, R.A. (2011). Hallmarks of cancer: the next generation. Cell 144, 646-674. Hatakeyama, H., and Kanzaki, M. (2011). Molecular basis of insulin-responsive GLUT4 trafficking systems revealed by single molecule imaging. Traffic. Helgason, C.D., Damen, J.E., Rosten, P., Grewal, R., Sorensen, P., Chappel, S.M., Borowski, A., Jirik, F., Krystal, G., and Humphries, R.K. (1998). Targeted disruption of SHIP leads to hemopoietic perturbations, lung pathology, and a shortened life span. Genes Dev 12, 1610-1620. 174  Hemkens, L.G., Grouven, U., Bender, R., Gunster, C., Gutschmidt, S., Selke, G.W., and Sawicki, P.T. (2009). Risk of malignancies in patients with diabetes treated with human insulin or insulin analogues: a cohort study. Diabetologia 52, 1732-1744. Henquin, J.C., Nenquin, M., Ravier, M.A., and Szollosi, A. (2009). Shortcomings of current models of glucose-induced insulin secretion. Diabetes Obes Metab 11 Suppl 4, 168-179. Hession, M., Rolland, C., Kulkarni, U., Wise, A., and Broom, J. (2009). Systematic review of randomized controlled trials of low-carbohydrate vs. low-fat/lowcalorie diets in the management of obesity and its comorbidities. Obes Rev 10, 36-50. Higashiyama, A., Watanabe, M., Kokubo, Y., Ono, Y., Okayama, A., and Okamura, T. (2010). Relationships between protein intake and renal function in a Japanese general population: NIPPON DATA90. Journal of epidemiology / Japan Epidemiological Association 20 Suppl 3, S537-543. Hirano, M., Davis, R.S., Fine, W.D., Nakamura, S., Shimizu, K., Yagi, H., Kato, K., Stephan, R.P., and Cooper, M.D. (2007). IgEb immune complexes activate macrophages through FcgammaRIV binding. Nat Immunol 8, 762-771. Ho, V.W., Leung, K., Hsu, A., Luk, B., Lai, J., Shen, S.Y., Minchinton, A.I., Waterhouse, D., Bally, M.B., Lin, W., et al. (2011). A low carbohydrate, high protein diet slows tumor growth and prevents cancer initiation. Cancer Res 71, 4484-4493. Holness, M.J., and Sugden, M.C. (2003). Regulation of pyruvate dehydrogenase complex activity by reversible phosphorylation. Biochem Soc Trans 31, 11431151. Hoption Cann, S.A., van Netten, J.P., van Netten, C., and Glover, D.W. (2002). Spontaneous regression: a hidden treasure buried in time. Med Hypotheses 58, 115-119. Huber, M., Helgason, C.D., Scheid, M.P., Duronio, V., Humphries, R.K., and Krystal, G. (1998). Targeted disruption of SHIP leads to Steel factor-induced degranulation of mast cells. The EMBO journal 17, 7311-7319. Hursting, S.D., Smith, S.M., Lashinger, L.M., Harvey, A.E., and Perkins, S.N. (2010). Calories and carcinogenesis: lessons learned from 30 years of calorie restriction research. Carcinogenesis 31, 83-89. Ichim,  C.V. (2005). Revisiting immunosurveillance and immunostimulation: Implications for cancer immunotherapy. J Transl Med 3, 8.  175  Ijichi, K., Adachi, M., Hasegawa, Y., Ogawa, T., Nakamura, H., Kudoh, A., Yasui, Y., Murakami, S., and Ishizaki, K. (2008). Pretreatment with 5-FU enhances cisplatin cytotoxicity in head and neck squamous cell carcinoma cells. Cancer Chemother Pharmacol 62, 745-752. Ingersoll, M.A., Spanbroek, R., Lottaz, C., Gautier, E.L., Frankenberger, M., Hoffmann, R., Lang, R., Haniffa, M., Collin, M., Tacke, F., et al. (2010). Comparison of gene expression profiles between human and mouse monocyte subsets. Blood 115, e10-19. Jadvar, H., Pinski, J.K., and Conti, P.S. (2003). FDG PET in suspected recurrent and metastatic prostate cancer. Oncology reports 10, 1485-1488. Jee, S.H., Ohrr, H., Sull, J.W., Yun, J.E., Ji, M., and Samet, J.M. (2005). Fasting serum glucose level and cancer risk in Korean men and women. Jama 293, 194-202. Jenkins, S.J., Ruckerl, D., Cook, P.C., Jones, L.H., Finkelman, F.D., van Rooijen, N., MacDonald, A.S., and Allen, J.E. (2011). Local macrophage proliferation, rather than recruitment from the blood, is a signature of TH2 inflammation. Science 332, 1284-1288. Jornayvaz, F.R., Jurczak, M.J., Lee, H.Y., Birkenfeld, A.L., Frederick, D.W., Zhang, D., Zhang, X.M., Samuel, V.T., and Shulman, G.I. (2010). A high-fat, ketogenic diet causes hepatic insulin resistance in mice, despite increasing energy expenditure and preventing weight gain. Am J Physiol Endocrinol Metab 299, E808-815. Josson, S., Matsuoka, Y., Chung, L.W., Zhau, H.E., and Wang, R. (2010). Tumorstroma co-evolution in prostate cancer progression and metastasis. Semin Cell Dev Biol 21, 26-32. Kandalaft, L.E., Motz, G.T., Duraiswamy, J., and Coukos, G. (2011). Tumor immune surveillance and ovarian cancer: lessons on immune mediated tumor rejection or tolerance. Cancer Metastasis Rev 30, 141-151. Kantoff, P.W., Higano, C.S., Shore, N.D., Berger, E.R., Small, E.J., Penson, D.F., Redfern, C.H., Ferrari, A.C., Dreicer, R., Sims, R.B., et al. (2010a). Sipuleucel-T immunotherapy for castration-resistant prostate cancer. N Engl J Med 363, 411-422.  176  Kantoff, P.W., Schuetz, T.J., Blumenstein, B.A., Glode, L.M., Bilhartz, D.L., Wyand, M., Manson, K., Panicali, D.L., Laus, R., Schlom, J., et al. (2010b). Overall survival analysis of a phase II randomized controlled trial of a Poxviral-based PSA-targeted immunotherapy in metastatic castration-resistant prostate cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 28, 1099-1105.Karnoub, A.E., and Weinberg, R.A. (2006). Chemokine networks and breast cancer metastasis. Breast disease 26, 7585. Kawai, T., and Akira, S. (2011). Toll-like receptors and their crosstalk with other innate receptors in infection and immunity. Immunity 34, 637-650. Keku, T.O., Lund, P.K., Galanko, J., Simmons, J.G., Woosley, J.T., and Sandler, R.S. (2005). Insulin resistance, apoptosis, and colorectal adenoma risk. Cancer Epidemiol Biomarkers Prev 14, 2076-2081. Keller, R., Geiges, M., and Keist, R. (1990). L-arginine-dependent reactive nitrogen intermediates as mediators of tumor cell killing by activated macrophages. Cancer Res 50, 1421-1425. Khalid, S., Hwang, D., Babichev, Y., Kolli, R., Altamentova, S., Koren, S., Goodwin, P.J., Ennis, M., Pollak, M., Sonenberg, N., and Fantus, I.G. (2010). Evidence for a tumor promoting effect of high-fat diet independent of insulin resistance in HER2/Neu mammary carcinogenesis. Breast Cancer Res Treat 122, 647659. Kilinc, M.O., Aulakh, K.S., Nair, R.E., Jones, S.A., Alard, P., Kosiewicz, M.M., and Egilmez, N.K. (2006). Reversing tumor immune suppression with intratumoral IL-12: activation of tumor-associated T effector/memory cells, induction of T suppressor apoptosis, and infiltration of CD8+ T effectors. J Immunol 177, 6962-6973. Kim, J.S., Kim, E.S., Liu, D., Lee, J.J., Solis, L., Behrens, C., Lippman, S.M., Hong, W.K., Wistuba, II, and Lee, H.Y. (2011). Prognostic impact of insulin receptor expression on survival of patients with nonsmall cell lung cancer. Cancer. Kim, J.W., Tchernyshyov, I., Semenza, G.L., and Dang, C.V. (2006). HIF-1-mediated expression of pyruvate dehydrogenase kinase: a metabolic switch required for cellular adaptation to hypoxia. Cell Metab 3, 177-185. Kleinbongard, P., Rassaf, T., Dejam, A., Kerber, S., and Kelm, M. (2002). Griess method for nitrite measurement of aqueous and protein-containing samples. Methods Enzymol 359, 158-168. Kobukai, S., Kremers, G.J., Cobb, J.G., Baheza, R., Xie, J., Kuley, A., Zhu, M., and Pham, W. (2011). Induction of antitumor immunity by dendritic cells loaded with membrane-translocating mucin 1 Peptide antigen. Transl Oncol 4, 1-8.  177  Kortylewski, M., Xin, H., Kujawski, M., Lee, H., Liu, Y., Harris, T., Drake, C., Pardoll, D., and Yu, H. (2009). Regulation of the IL-23 and IL-12 balance by Stat3 signaling in the tumor microenvironment. Cancer Cell 15, 114-123. Kretschmann, K.L., Eyob, H., Buys, S.S., and Welm, A.L. (2010). The macrophage stimulating protein/Ron pathway as a potential therapeutic target to impede multiple mechanisms involved in breast cancer progression. Curr Drug Targets 11, 1157-1168. Kroemer, G., and Pouyssegur, J. (2008). Tumor cell metabolism: cancer's Achilles' heel. Cancer Cell 13, 472-482. Kronenberg, M., and Gapin, L. (2002). The unconventional lifestyle of NKT cells. Nat Rev Immunol 2, 557-568. Kulp, S.K., Yang, Y.T., Hung, C.C., Chen, K.F., Lai, J.P., Tseng, P.H., Fowble, J.W., Ward, P.J., and Chen, C.S. (2004). 3-phosphoinositide-dependent protein kinase-1/Akt signaling represents a major cyclooxygenase-2-independent target for celecoxib in prostate cancer cells. Cancer Res 64, 1444-1451. Kurahara, H., Shinchi, H., Mataki, Y., Maemura, K., Noma, H., Kubo, F., Sakoda, M., Ueno, S., Natsugoe, S., and Takao, S. (2011). Significance of M2-polarized tumor-associated macrophage in pancreatic cancer. The Journal of surgical research 167, e211-219. Kuroda, E., Antignano, F., Ho, V.W., Hughes, M.R., Ruschmann, J., Lam, V., Kawakami, T., Kerr, W.G., McNagny, K.M., Sly, L.M., and Krystal, G. (2011). SHIP represses Th2 skewing by inhibiting IL-4 production from basophils. J Immunol 186, 323-332. Kuroda, E., Ho, V., Ruschmann, J., Antignano, F., Hamilton, M., Rauh, M.J., Antov, A., Flavell, R.A., Sly, L.M., and Krystal, G. (2009). SHIP represses the generation of IL-3-induced M2 macrophages by inhibiting IL-4 production from basophils. J Immunol 183, 3652-3660. Lajous, M., Willett, W., Lazcano-Ponce, E., Sanchez-Zamorano, L.M., HernandezAvila, M., and Romieu, I. (2005). Glycemic load, glycemic index, and the risk of breast cancer among Mexican women. Cancer causes & control : CCC 16, 1165-1169. Law, B.K. (2005). Rapamycin: an anti-cancer immunosuppressant? Crit Rev Oncol Hematol 56, 47-60. Lehrer, S.B. (1976). Isolation and immunochemical properties of mouse IgE. Immunochemistry 13, 837-843.  178  Lin, E.Y., Li, J.F., Bricard, G., Wang, W., Deng, Y., Sellers, R., Porcelli, S.A., and Pollard, J.W. (2007). Vascular endothelial growth factor restores delayed tumor progression in tumors depleted of macrophages. Molecular oncology 1, 288-302. Lin, H.H., Stacey, M., Stein-Streilein, J., and Gordon, S. (2010). F4/80: the macrophage-specific adhesion-GPCR and its role in immunoregulation. Adv Exp Med Biol 706, 149-156. Lloyd, C.M., Phillips, A.R., Cooper, G.J., and Dunbar, P.R. (2008). Three-colour fluorescence immunohistochemistry reveals the diversity of cells staining for macrophage markers in murine spleen and liver. J Immunol Methods 334, 7081. Lumeng, C.N., Bodzin, J.L., and Saltiel, A.R. (2007). Obesity induces a phenotypic switch in adipose tissue macrophage polarization. J Clin Invest 117, 175-184. Magee, B.A., Potezny, N., Rofe, A.M., and Conyers, R.A. (1979). The inhibition of malignant cell growth by ketone bodies. Aust J Exp Biol Med Sci 57, 529-539. Makitie, T., Summanen, P., Tarkkanen, A., and Kivela, T. (2001). Tumor-infiltrating macrophages (CD68(+) cells) and prognosis in malignant uveal melanoma. Invest Ophthalmol Vis Sci 42, 1414-1421. Mantovani, A., Sica, A., Sozzani, S., Allavena, P., Vecchi, A., and Locati, M. (2004). The chemokine system in diverse forms of macrophage activation and polarization. Trends Immunol 25, 677-686. Mantovani, A., Sozzani, S., Locati, M., Allavena, P., and Sica, A. (2002). Macrophage polarization: tumor-associated macrophages as a paradigm for polarized M2 mononuclear phagocytes. Trends Immunol 23, 549-555. Margolin, K., Ernstoff, M.S., Hamid, O., Lawrence, D., McDermott, D., Puzanov, I., Wolchok, J.D., Clark, J.I., Sznol, M., Logan, T.F., et al. (2012). Ipilimumab in patients with melanoma and brain metastases: an open-label, phase 2 trial. The lancet oncology. Martin, W.F., Armstrong, L.E., and Rodriguez, N.R. (2005). Dietary protein intake and renal function. Nutrition & metabolism 2, 25. Martinez-Zaguilan, R., Seftor, E.A., Seftor, R.E., Chu, Y.W., Gillies, R.J., and Hendrix, M.J. (1996). Acidic pH enhances the invasive behavior of human melanoma cells. Clin Exp Metastasis 14, 176-186. Martinez, F.O., Helming, L., and Gordon, S. (2009). Alternative activation of macrophages: an immunologic functional perspective. Annu Rev Immunol 27, 451-483.  179  Martinez, F.O., Sica, A., Mantovani, A., and Locati, M. (2008). Macrophage activation and polarization. Front Biosci 13, 453-461. Masko, E.M., Thomas, J.A., 2nd, Antonelli, J.A., Lloyd, J.C., Phillips, T.E., Poulton, S.H., Dewhirst, M.W., Pizzo, S.V., and Freedland, S.J. (2010). LowCarbohydrate Diets and Prostate Cancer: How Low Is "Low Enough"? Cancer Prev Res (Phila Pa). Massi, D., Marconi, C., Franchi, A., Bianchini, F., Paglierani, M., Ketabchi, S., Miracco, C., Santucci, M., and Calorini, L. (2007). Arginine metabolism in tumor-associated macrophages in cutaneous malignant melanoma: evidence from human and experimental tumors. Hum Pathol 38, 1516-1525. Mathupala, S.P., Ko, Y.H., and Pedersen, P.L. (2010). The pivotal roles of mitochondria in cancer: Warburg and beyond and encouraging prospects for effective therapies. Biochim Biophys Acta 1797, 1225-1230. Matsuda, M., and Mori, T. (1996). Effect of estrogen on hyperprolactinemia-induced glucose intolerance in SHN mice. Proc Soc Exp Biol Med 212, 243-247. Maxwell, M.J., Duan, M., Armes, J.E., Anderson, G.P., Tarlinton, D.M., and Hibbs, M.L. (2011). Genetic segregation of inflammatory lung disease and autoimmune disease severity in SHIP-1-/- mice. J Immunol 186, 7164-7175. Mbeunkui, F., and Johann, D.J., Jr. (2009). Cancer and the tumor microenvironment: a review of an essential relationship. Cancer Chemother Pharmacol 63, 571582. McKellar, G., and Singh, G. (2009). Celecoxib in arthritis: relative risk management profile and implications for patients. Ther Clin Risk Manag 5, 889-896. Melnicoff, M.J., Horan, P.K., Breslin, E.W., and Morahan, P.S. (1988). Maintenance of peritoneal macrophages in the steady state. Journal of leukocyte biology 44, 367-375. Menter, D.G., Schilsky, R.L., and DuBois, R.N. (2010). Cyclooxygenase-2 and cancer treatment: understanding the risk should be worth the reward. Clinical cancer research : an official journal of the American Association for Cancer Research 16, 1384-1390. Mills, C.D. (2001). Macrophage arginine metabolism to ornithine/urea or nitric oxide/citrulline: a life or death issue. Crit Rev Immunol 21, 399-425. Min, B., Prout, M., Hu-Li, J., Zhu, J., Jankovic, D., Morgan, E.S., Urban, J.F., Jr., Dvorak, A.M., Finkelman, F.D., LeGros, G., and Paul, W.E. (2004). Basophils produce IL-4 and accumulate in tissues after infection with a Th2-inducing parasite. J Exp Med 200, 507-517.  180  Minchinton, A.I., and Tannock, I.F. (2006). Drug penetration in solid tumours. Nat Rev Cancer 6, 583-592. Minois, N., Carmona-Gutierrez, D., and Madeo, F. (2011). Polyamines in aging and disease. Aging (Albany NY). Moisey, L.L., Kacker, S., Bickerton, A.C., Robinson, L.E., and Graham, T.E. (2008). Caffeinated coffee consumption impairs blood glucose homeostasis in response to high and low glycemic index meals in healthy men. Am J Clin Nutr 87, 1254-1261. Monteiro, J.P., Benjamin, A., Costa, E.S., Barcinski, M.A., and Bonomo, A. (2005). Normal hematopoiesis is maintained by activated bone marrow CD4+ T cells. Blood 105, 1484-1491. Moon, K.A., Kim, S.Y., Kim, T.B., Yun, E.S., Park, C.S., Cho, Y.S., Moon, H.B., and Lee, K.Y. (2007). Allergen-induced CD11b+ CD11c(int) CCR3+ macrophages in the lung promote eosinophilic airway inflammation in a mouse asthma model. Int Immunol 19, 1371-1381. Morrison, A.C., and Correll, P.H. (2002). Activation of the stem cell-derived tyrosine kinase/RON receptor tyrosine kinase by macrophage-stimulating protein results in the induction of arginase activity in murine peritoneal macrophages. J Immunol 168, 853-860. Mosser, D.M., and Edwards, J.P. (2008). Exploring the full spectrum of macrophage activation. Nat Rev Immunol 8, 958-969. Mulholland, H.G., Cantwell, M.M., Anderson, L.A., Johnston, B.T., Watson, R.G., Murphy, S.J., Ferguson, H.R., McGuigan, J., Reynolds, J.V., Comber, H., and Murray, L.J. (2009). Glycemic index, carbohydrate and fiber intakes and risk of reflux esophagitis, Barrett's esophagus, and esophageal adenocarcinoma. Cancer causes & control : CCC 20, 279-288. Mulholland, H.G., Murray, L.J., Cardwell, C.R., and Cantwell, M.M. (2008). Dietary glycaemic index, glycaemic load and endometrial and ovarian cancer risk: a systematic review and meta-analysis. British journal of cancer 99, 434-441. Munder, M. (2009). Arginase: an emerging key player in the mammalian immune system. British journal of pharmacology 158, 638-651. Munder, M., Mollinedo, F., Calafat, J., Canchado, J., Gil-Lamaignere, C., Fuentes, J.M., Luckner, C., Doschko, G., Soler, G., Eichmann, K., et al. (2005). Arginase I is constitutively expressed in human granulocytes and participates in fungicidal activity. Blood 105, 2549-2556.  181  Munder, M., Schneider, H., Luckner, C., Giese, T., Langhans, C.D., Fuentes, J.M., Kropf, P., Mueller, I., Kolb, A., Modolell, M., and Ho, A.D. (2006). Suppression of T-cell functions by human granulocyte arginase. Blood 108, 1627-1634. Mukhopadhyay, S., Peiser, L., and Gordon, S. (2004). Activation of murine macrophages by Neisseria meningitidis and IFN-gamma in vitro: distinct roles of class A scavenger and Toll-like pattern recognition receptors in selective modulation of surface phenotype. Journal of leukocyte biology 76, 577-584. Nagashio, R., Sato, Y., Jiang, S.X., Ryuge, S., Kodera, Y., Maeda, T., and Nakajima, T. (2008). Detection of tumor-specific autoantibodies in sera of patients with lung cancer. Lung Cancer 62, 364-373. Nakanishi, Y., Nakatsuji, M., Seno, H., Ishizu, S., Akitake-Kawano, R., Kanda, K., Ueo, T., Komekado, H., Kawada, M., Minami, M., and Chiba, T. (2011). COX2 inhibition alters the phenotype of tumor-associated macrophages from M2 to M1 in ApcMin/+ mouse polyps. Carcinogenesis 32, 1333-1339. Nebeling, L.C., Miraldi, F., Shurin, S.B., and Lerner, E. (1995). Effects of a ketogenic diet on tumor metabolism and nutritional status in pediatric oncology patients: two case reports. J Am Coll Nutr 14, 202-208. Negus, R.P., Stamp, G.W., Relf, M.G., Burke, F., Malik, S.T., Bernasconi, S., Allavena, P., Sozzani, S., Mantovani, A., and Balkwill, F.R. (1995). The detection and localization of monocyte chemoattractant protein-1 (MCP-1) in human ovarian cancer. J Clin Invest 95, 2391-2396. Nelson, D.L.C., M.M., ed. (2000). Lehninger Principles of Biochemistry Third Edition (New York: Worth Publishers). Niese, K.A., Chiaramonte, M.G., Ellies, L.G., Rothenberg, M.E., and Zimmermann, N. (2010). The cationic amino acid transporter 2 is induced in inflammatory lung models and regulates lung fibrosis. Respir Res 11, 87. Nimmerjahn, F., and Ravetch, J.V. (2007). Fc-receptors as regulators of immunity. Adv Immunol 96, 179-204. Novelli, F., D'Elios, M.M., Bernabei, P., Ozmen, L., Rigamonti, L., Almerigogna, F., Forni, G., and Del Prete, G. (1997). Expression and role in apoptosis of the alpha- and beta-chains of the IFN-gamma receptor on human Th1 and Th2 clones. J Immunol 159, 206-213. Ochoa, A.C., Zea, A.H., Hernandez, C., and Rodriguez, P.C. (2007a). Arginase, prostaglandins, and myeloid-derived suppressor cells in renal cell carcinoma. Clin Cancer Res 13, 721s-726s.  182  Ochoa, A.C., Zea, A.H., Hernandez, C., and Rodriguez, P.C. (2007b). Arginase, prostaglandins, and myeloid-derived suppressor cells in renal cell carcinoma. Clinical cancer research : an official journal of the American Association for Cancer Research 13, 721s-726s. Odegaard, J.I., Ricardo-Gonzalez, R.R., Goforth, M.H., Morel, C.R., Subramanian, V., Mukundan, L., Red Eagle, A., Vats, D., Brombacher, F., Ferrante, A.W., and Chawla, A. (2007). Macrophage-specific PPARgamma controls alternative activation and improves insulin resistance. Nature 447, 1116-1120. Osborne, C.K., Bolan, G., Monaco, M.E., and Lippman, M.E. (1976). Hormone responsive human breast cancer in long-term tissue culture: effect of insulin. Proc Natl Acad Sci U S A 73, 4536-4540. Otto, C., Kaemmerer, U., Illert, B., Muehling, B., Pfetzer, N., Wittig, R., Voelker, H.U., Thiede, A., and Coy, J.F. (2008). Growth of human gastric cancer cells in nude mice is delayed by a ketogenic diet supplemented with omega-3 fatty acids and medium-chain triglycerides. BMC Cancer 8, 122. Paludan, S.R. (1998). Interleukin-4 and interferon-gamma: the quintessence of a mutual antagonistic relationship. Scandinavian journal of immunology 48, 459-468. Parillo, M., Licenziati, M.R., Vacca, M., De Marco, D., and Iannuzzi, A. (2011). Metabolic Changes after a Hypocaloric, Low-Glycemic Index Diet in Obese Children. J Endocrinol Invest. Pedersen, P.L. (2007). The cancer cell's "power plants" as promising therapeutic targets: an overview. Journal of bioenergetics and biomembranes 39, 1-12. Pijl, H. (2011). Obesity: evolution of a symptom of affluence. Neth J Med 69, 159166. Pinheiro, C., Longatto-Filho, A., Scapulatempo, C., Ferreira, L., Martins, S., Pellerin, L., Rodrigues, M., Alves, V.A., Schmitt, F., and Baltazar, F. (2008). Increased expression of monocarboxylate transporters 1, 2, and 4 in colorectal carcinomas. Virchows Arch 452, 139-146. Pilcher, J.B., Tsang, V.C., Zhou, W., Black, C.M., and Sidman, C. (1991). Optimization of binding capacity and specificity of protein G on various solid matrices for immunoglobulins. J Immunol Methods 136, 279-286. Pixley, F.J., and Stanley, E.R. (2004). CSF-1 regulation of the wandering macrophage: complexity in action. Trends Cell Biol 14, 628-638. Platt, N., Suzuki, H., Kurihara, Y., Kodama, T., and Gordon, S. (1996). Role for the class A macrophage scavenger receptor in the phagocytosis of apoptotic thymocytes in vitro. Proc Natl Acad Sci U S A 93, 12456-12460. 183  Polakis, P. (2007). The many ways of Wnt in cancer. Curr Opin Genet Dev 17, 4551. Pollak, M. (2008). Insulin and insulin-like growth factor signalling in neoplasia. Nat Rev Cancer 8, 915-928. Popovic, P.J., Zeh, H.J., 3rd, and Ochoa, J.B. (2007). Arginine and immunity. J Nutr 137, 1681S-1686S. Prehn, R.T., and Main, J.M. (1957). Immunity to methylcholanthrene-induced sarcomas. J Natl Cancer Inst 18, 769-778. Prehn, R.T., and Prehn, L.M. (2008). The flip side of immune surveillance: immune dependency. Immunol Rev 222, 341-356. Qian, B.Z., Li, J., Zhang, H., Kitamura, T., Zhang, J., Campion, L.R., Kaiser, E.A., Snyder, L.A., and Pollard, J.W. (2011). CCL2 recruits inflammatory monocytes to facilitate breast-tumour metastasis. Nature 475, 222-225. Raes, G., Brys, L., Dahal, B.K., Brandt, J., Grooten, J., Brombacher, F., Vanham, G., Noel, W., Bogaert, P., Boonefaes, T., et al. (2005). Macrophage galactosetype C-type lectins as novel markers for alternatively activated macrophages elicited by parasitic infections and allergic airway inflammation. J Leukoc Biol 77, 321-327. Rahim, S.S., Khan, N., Boddupalli, C.S., Hasnain, S.E., and Mukhopadhyay, S. (2005). Interleukin-10 (IL-10) mediated suppression of IL-12 production in RAW 264.7 cells also involves c-rel transcription factor. Immunology 114, 313-321. Rauh, M.J., Ho, V., Pereira, C., Sham, A., Sly, L.M., Lam, V., Huxham, L., Minchinton, A.I., Mui, A., and Krystal, G. (2005). SHIP represses the generation of alternatively activated macrophages. Immunity 23, 361-374. Ravetch, J.V., and Bolland, S. (2001). IgG Fc receptors. Annu Rev Immunol 19, 275290. Richard, V., and Pillai, M.R. (2010). The stem cell code in oral epithelial tumorigenesis: 'the cancer stem cell shift hypothesis'. Biochim Biophys Acta 1806, 146-162. Rivard, N., Boucher, M.J., Asselin, C., and L'Allemain, G. (1999). MAP kinase cascade is required for p27 downregulation and S phase entry in fibroblasts and epithelial cells. Am J Physiol 277, C652-664.  184  Rodriguez-Barbero, A., Dorado, F., Velasco, S., Pandiella, A., Banas, B., and LopezNovoa, J.M. (2006). TGF-beta1 induces COX-2 expression and PGE2 synthesis through MAPK and PI3K pathways in human mesangial cells. Kidney international 70, 901-909. Rodriguez, P.C., Hernandez, C.P., Quiceno, D., Dubinett, S.M., Zabaleta, J., Ochoa, J.B., Gilbert, J., and Ochoa, A.C. (2005). Arginase I in myeloid suppressor cells is induced by COX-2 in lung carcinoma. J Exp Med 202, 931-939. Rojas, A., Figueroa, H., and Morales, E. (2010). Fueling inflammation at tumor microenvironment: the role of multiligand/RAGE axis. Carcinogenesis 31, 334-341. Rowley, D.A., and Stach, R.M. (1998). B lymphocytes secreting IgG linked to latent transforming growth factor-beta prevent primary cytolytic T lymphocyte responses. Int Immunol 10, 355-363. Ruffell, B., Au, A., Rugo, H.S., Esserman, L.J., Hwang, E.S., and Coussens, L.M. (2011). Breast Cancer Special Feature: Leukocyte composition of human breast cancer. Proc Natl Acad Sci U S A. Ruggeri, L., Capanni, M., Urbani, E., Perruccio, K., Shlomchik, W.D., Tosti, A., Posati, S., Rogaia, D., Frassoni, F., Aversa, F., et al. (2002). Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants. Science 295, 2097-2100. Ruschmann, J., Ho, V., Antignano, F., Kuroda, E., Lam, V., Ibaraki, M., Snyder, K., Kim, C., Flavell, R.A., Kawakami, T., et al. (2010). Tyrosine phosphorylation of SHIP promotes its proteasomal degradation. Exp Hematol 38, 392-402, 402 e391. Rutschman, R., Lang, R., Hesse, M., Ihle, J.N., Wynn, T.A., and Murray, P.J. (2001). Cutting edge: Stat6-dependent substrate depletion regulates nitric oxide production. J Immunol 166, 2173-2177. Ryan, E.P., Bushnell, T.P., Friedman, A.E., Rahman, I., and Phipps, R.P. (2008). Cyclooxygenase-2 independent effects of cyclooxygenase-2 inhibitors on oxidative stress and intracellular glutathione content in normal and malignant human B-cells. Cancer immunology, immunotherapy : CII 57, 347-358. Sadhu, C., Ting, H.J., Lipsky, B., Hensley, K., Garcia-Martinez, L.F., Simon, S.I., and Staunton, D.E. (2007). CD11c/CD18: novel ligands and a role in delayed-type hypersensitivity. J Leukoc Biol 81, 1395-1403. Saha, D., Datta, P.K., Sheng, H., Morrow, J.D., Wada, M., Moses, H.L., and Beauchamp, R.D. (1999). Synergistic induction of cyclooxygenase-2 by transforming growth factor-beta1 and epidermal growth factor inhibits apoptosis in epithelial cells. Neoplasia 1, 508-517. 185  Saikali, Z., Setya, H., Singh, G., and Persad, S. (2008). Role of IGF-1/IGF-1R in regulation of invasion in DU145 prostate cancer cells. Cancer Cell Int 8, 10. Sakai, K., Suzuki, H., Oda, H., Akaike, T., Azuma, Y., Murakami, T., Sugi, K., Ito, T., Ichinose, H., Koyasu, S., and Shirai, M. (2006). Phosphoinositide 3-kinase in nitric oxide synthesis in macrophage: critical dimerization of inducible nitricoxide synthase. J Biol Chem 281, 17736-17742. Sakata, D., Yao, C., and Narumiya, S. (2010). immunoactivator. J Pharmacol Sci 112, 1-5.  Prostaglandin  E2,  an  Sambrook, J., and Russell, D.W., eds. (2001). Molecular Cloning A Laboratory Manual, 3rd edn (New York: Cold Spring Harbor Laboratory Press). Sanchez, C.J., Le Treut, T., Boehrer, A., Knoblauch, B., Imbert, J., Olive, D., and Costello, R.T. (2011). Natural killer cells and malignant haemopathies: a model for the interaction of cancer with innate immunity. Cancer Immunol Immunother 60, 1-13. Sant'Anna, O.A., Mouton, D., Ibanez, O.M., Bouthillier, Y., Mevel, J.C., Reis, M.H., and Biozzi, G. (1985). Basal immunoglobulin serum concentration and isotype distribution in relation to the polygenic control of antibody responsiveness in mice. Immunogenetics 22, 131-139. Satoh, T., Takeuchi, O., Vandenbon, A., Yasuda, K., Tanaka, Y., Kumagai, Y., Miyake, T., Matsushita, K., Okazaki, T., Saitoh, T., et al. (2010). The Jmjd3Irf4 axis regulates M2 macrophage polarization and host responses against helminth infection. Nat Immunol 11, 936-944. Schneemann, M., Schoedon, G., Hofer, S., Blau, N., Guerrero, L., and Schaffner, A. (1993). Nitric oxide synthase is not a constituent of the antimicrobial armature of human mononuclear phagocytes. J Infect Dis 167, 1358-1363. Schreiber, T.H., Raez, L., Rosenblatt, J.D., and Podack, E.R. (2010). Tumor immunogenicity and responsiveness to cancer vaccine therapy: the state of the art. Semin Immunol 22, 105-112. Schupp, M., and Lazar, M.A. (2010). Endogenous ligands for nuclear receptors: digging deeper. J Biol Chem 285, 40409-40415. Serganova, I., Rizwan, A., Ni, X., Thakur, S.B., Vider, J., Russell, J., Blasberg, R., and Koutcher, J.A. (2011). Metabolic Imaging: A link between Lactate Dehydrogenase A, Lactate and Tumor Phenotype. Clinical cancer research : an official journal of the American Association for Cancer Research. Seyfried, T.N., Marsh, J., Shelton, L.M., Huysentruyt, L.C., and Mukherjee, P. (2011). Is the restricted ketogenic diet a viable alternative to the standard of care for managing malignant brain cancer? Epilepsy Res. 186  Sharda, D.R., Yu, S., Ray, M., Squadrito, M.L., De Palma, M., Wynn, T.A., Morris, S.M., Jr., and Hankey, P.A. (2011). Regulation of macrophage arginase expression and tumor growth by the Ron receptor tyrosine kinase. J Immunol 187, 2181-2192. Shaw, R.J. (2006). Glucose metabolism and cancer. Current opinion in cell biology 18, 598-608. Sheng, H., Shao, J., Dixon, D.A., Williams, C.S., Prescott, S.M., DuBois, R.N., and Beauchamp, R.D. (2000). Transforming growth factor-beta1 enhances Haras-induced expression of cyclooxygenase-2 in intestinal epithelial cells via stabilization of mRNA. J Biol Chem 275, 6628-6635. Shim, H., Dolde, C., Lewis, B.C., Wu, C.S., Dang, G., Jungmann, R.A., DallaFavera, R., and Dang, C.V. (1997). c-Myc transactivation of LDH-A: implications for tumor metabolism and growth. Proc Natl Acad Sci U S A 94, 6658-6663. Shirabe, K., Mano, Y., Muto, J., Matono, R., Motomura, T., Toshima, T., Takeishi, K., Uchiyama, H., Yoshizumi, T., Taketomi, A., et al. (2012). Role of tumorassociated macrophages in the progression of hepatocellular carcinoma. Surg Today 42, 1-7. Sica, A., Allavena, P., and Mantovani, A. (2008a). Cancer related inflammation: the macrophage connection. Cancer letters 267, 204-215. Sica, A., Larghi, P., Mancino, A., Rubino, L., Porta, C., Totaro, M.G., Rimoldi, M., Biswas, S.K., Allavena, P., and Mantovani, A. (2008b). Macrophage polarization in tumour progression. Semin Cancer Biol 18, 349-355. Siddle, K. (2011). Signalling by insulin and IGF receptors: supporting acts and new players. J Mol Endocrinol 47, R1-10. Simmons, D.L., Botting, R.M., and Hla, T. (2004). Cyclooxygenase isozymes: the biology of prostaglandin synthesis and inhibition. Pharmacol Rev 56, 387-437. Singer, K., Gottfried, E., Kreutz, M., and Mackensen, A. (2011). Suppression of Tcell responses by tumor metabolites. Cancer Immunol Immunother 60, 425431. Sinha, P., Clements, V.K., Miller, S., and Ostrand-Rosenberg, S. (2005a). Tumor immunity: a balancing act between T cell activation, macrophage activation and tumor-induced immune suppression. Cancer Immunol Immunother 54, 1137-1142. Sinha, P., Clements, V.K., and Ostrand-Rosenberg, S. (2005b). Reduction of myeloid-derived suppressor cells and induction of M1 macrophages facilitate the rejection of established metastatic disease. J Immunol 174, 636-645. 187  Sitkovsky, M., and Lukashev, D. (2005). Regulation of immune cells by local-tissue oxygen tension: HIF1 alpha and adenosine receptors. Nat Rev Immunol 5, 712-721. Sonveaux, P., Vegran, F., Schroeder, T., Wergin, M.C., Verrax, J., Rabbani, Z.N., De Saedeleer, C.J., Kennedy, K.M., Diepart, C., Jordan, B.F., et al. (2008). Targeting lactate-fueled respiration selectively kills hypoxic tumor cells in mice. J Clin Invest 118, 3930-3942. Srivastava, M.K., Sinha, P., Clements, V.K., Rodriguez, P., and Ostrand-Rosenberg, S. (2010). Myeloid-derived suppressor cells inhibit T-cell activation by depleting cystine and cysteine. Cancer Res 70, 68-77. Stach, R.M., and Rowley, D.A. (1993). A first or dominant immunization. II. Induced immunoglobulin carries transforming growth factor beta and suppresses cytolytic T cell responses to unrelated alloantigens. J Exp Med 178, 841-852. Stefater, J.A., 3rd, Ren, S., Lang, R.A., and Duffield, J.S. (2011). Metchnikoff's policemen: macrophages in development, homeostasis and regeneration. Trends Mol Med. Steidl, C., Lee, T., Shah, S.P., Farinha, P., Han, G., Nayar, T., Delaney, A., Jones, S.J., Iqbal, J., Weisenburger, D.D., et al. (2010). Tumor-associated macrophages and survival in classic Hodgkin's lymphoma. N Engl J Med 362, 875-885. Stein, M., Keshav, S., Harris, N., and Gordon, S. (1992). Interleukin 4 potently enhances murine macrophage mannose receptor activity: a marker of alternative immunologic macrophage activation. J Exp Med 176, 287-292. Stocks, T., Rapp, K., Bjorge, T., Manjer, J., Ulmer, H., Selmer, R., Lukanova, A., Johansen, D., Concin, H., Tretli, S., et al. (2009). Blood glucose and risk of incident and fatal cancer in the metabolic syndrome and cancer project (mecan): analysis of six prospective cohorts. PLoS Med 6, e1000201. Stolzenberg-Solomon, R.Z., Graubard, B.I., Chari, S., Limburg, P., Taylor, P.R., Virtamo, J., and Albanes, D. (2005). Insulin, glucose, insulin resistance, and pancreatic cancer in male smokers. Jama 294, 2872-2878. Stout, R.D., and Suttles, J. (2004). Functional plasticity of macrophages: reversible adaptation to changing microenvironments. Journal of leukocyte biology 76, 509-513. Stout, R.D., and Suttles, J. (2005). Immunosenescence and macrophage functional plasticity: dysregulation of macrophage function by age-associated microenvironmental changes. Immunol Rev 205, 60-71.  188  Stout, R.D., Jiang, C., Matta, B., Tietzel, I., Watkins, S.K., and Suttles, J. (2005). Macrophages sequentially change their functional phenotype in response to changes in microenvironmental influences. J Immunol 175, 342-349. Sugihara, A.Q., Rolle, C.E., and Lesniak, M.S. (2009). Regulatory T cells actively infiltrate metastatic brain tumors. Int J Oncol 34, 1533-1540. Taha, C., and Klip, A. (1999). The insulin signaling pathway. J Membr Biol 169, 1-12. Taha, C., Mitsumoto, Y., Liu, Z., Skolnik, E.Y., and Klip, A. (1995). The insulindependent biosynthesis of GLUT1 and GLUT3 glucose transporters in L6 muscle cells is mediated by distinct pathways. Roles of p21ras and pp70 S6 kinase. J Biol Chem 270, 24678-24681. Takahashi, K. (2001). Development and Differentiation of Macrophages and Related Cells: Historical Review and Current Concepts. Journal of Clinical and Experimental Hematopathology 41, 1-33. Takeda, K., and Akira, S. (2005). Toll-like receptors in innate immunity. Int Immunol 17, 1-14. Takizawa, F., Adamczewski, M., and Kinet, J.P. (1992). Identification of the low affinity receptor for immunoglobulin E on mouse mast cells and macrophages as Fc gamma RII and Fc gamma RIII. J Exp Med 176, 469-475. Tanaka, M., Miyazaki, H., Takeda, Y., and Takeo, S. (1993). Detection of serum cytokine levels in experimental cancer cachexia of colon 26 adenocarcinomabearing mice. Cancer letters 72, 65-70. Thomas, J.A., and Badini, M. (2011). The role of innate immunity in spontaneous regression of cancer. Indian J Cancer 48, 246-251. Tjiu, J.W., Chen, J.S., Shun, C.T., Lin, S.J., Liao, Y.H., Chu, C.Y., Tsai, T.F., Chiu, H.C., Dai, Y.S., Inoue, H., et al. (2009). Tumor-associated macrophageinduced invasion and angiogenesis of human basal cell carcinoma cells by cyclooxygenase-2 induction. J Invest Dermatol 129, 1016-1025. Toden, S., Bird, A.R., Topping, D.L., and Conlon, M.A. (2006). Resistant starch prevents colonic DNA damage induced by high dietary cooked red meat or casein in rats. Cancer Biol Ther 5, 267-272. Trinchieri, G., and Sher, A. (2007). Cooperation of Toll-like receptor signals in innate immune defence. Nat Rev Immunol 7, 179-190. Triozzi, P.L., Aldrich, W., and Ponnazhagan, S. (2010). Regulation of the activity of an adeno-associated virus vector cancer vaccine administered with synthetic Toll-like receptor agonists. Vaccine 28, 7837-7843.  189  Turner, G.A. (1979). Increased release of tumour cells by collagenase at acid pH: a possible mechanism for metastasis. Experientia 35, 1657-1658. van den Broek, M.F., Kagi, D., Zinkernagel, R.M., and Hengartner, H. (1995). Perforin dependence of natural killer cell-mediated tumor control in vivo. Eur J Immunol 25, 3514-3516. Vander Heiden, M.G., Cantley, L.C., and Thompson, C.B. (2009). Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science 324, 1029-1033. VanSaun, M.N., Lee, I.K., Washington, M.K., Matrisian, L., and Gorden, D.L. (2009). High fat diet induced hepatic steatosis establishes a permissive microenvironment for colorectal metastases and promotes primary dysplasia in a murine model. The American journal of pathology 175, 355-364. Varin, A., Mukhopadhyay, S., Herbein, G., and Gordon, S. (2010). Alternative activation of macrophages by IL-4 impairs phagocytosis of pathogens but potentiates microbial-induced signalling and cytokine secretion. Blood 115, 353-362. Vasievich, E.A., and Huang, L. (2011). The suppressive tumor microenvironment: a challenge in cancer immunotherapy. Mol Pharm 8, 635-641. Venkateswaran, V., Haddad, A.Q., Fleshner, N.E., Fan, R., Sugar, L.M., Nam, R., Klotz, L.H., and Pollak, M. (2007). Association of diet-induced hyperinsulinemia with accelerated growth of prostate cancer (LNCaP) xenografts. J Natl Cancer Inst 99, 1793-1800. Vermorken, J.B., Mesia, R., Rivera, F., Remenar, E., Kawecki, A., Rottey, S., Erfan, J., Zabolotnyy, D., Kienzer, H.R., Cupissol, D., et al. (2008). Platinum-based chemotherapy plus cetuximab in head and neck cancer. N Engl J Med 359, 1116-1127. Vesely, M.D., Kershaw, M.H., Schreiber, R.D., and Smyth, M.J. (2011). Natural innate and adaptive immunity to cancer. Annu Rev Immunol 29, 235-271. Vessiere, F., Georlette, M., Warnier, G., Leclerc, J.C., Van Pel, A., and Boon, T. (1982). Immunogenic variants obtained by mutagenesis of mouse Lewis lung carcinoma. Recognition of variant-specific antigens by cytolytic T lymphocytes. Eur J Cancer Clin Oncol 18, 867-874. Voehringer, D., Shinkai, K., and Locksley, R.M. (2004). Type 2 immunity reflects orchestrated recruitment of cells committed to IL-4 production. Immunity 20, 267-277.  190  Vouldoukis, I., Becherel, P.A., Riveros-Moreno, V., Arock, M., da Silva, O., Debre, P., Mazier, D., and Mossalayi, M.D. (1997). Interleukin-10 and interleukin-4 inhibit intracellular killing of Leishmania infantum and Leishmania major by human macrophages by decreasing nitric oxide generation. Eur J Immunol 27, 860-865. Wall, E.M., Milne, K., Martin, M.L., Watson, P.H., Theiss, P., and Nelson, B.H. (2007). Spontaneous mammary tumors differ widely in their inherent sensitivity to adoptively transferred T cells. Cancer Res 67, 6442-6450. Wang, D., and DuBois, R.N. (2004). Cyclooxygenase 2-derived prostaglandin E2 regulates the angiogenic switch. Proc Natl Acad Sci U S A 101, 415-416. Wang, J.W., Howson, J.M., Ghansah, T., Desponts, C., Ninos, J.M., May, S.L., Nguyen, K.H., Toyama-Sorimachi, N., and Kerr, W.G. (2002). Influence of SHIP on the NK repertoire and allogeneic bone marrow transplantation. Science 295, 2094-2097. Wang, R., Zhang, J., Chen, S., Lu, M., Luo, X., Yao, S., Liu, S., Qin, Y., and Chen, H. (2011). Tumor-associated macrophages provide a suitable microenvironment for non-small lung cancer invasion and progression. Lung Cancer. Warburg, O. (1956). On the origin of cancer cells. Science 123, 309-314. Warburg, O., Wind, F., and Negelein, E. (1927). THE METABOLISM OF TUMORS IN THE BODY. J Gen Physiol 8, 519-530. Watzer, B., Zehbe, R., Halstenberg, S., James Kirkpatrick, C., and Brochhausen, C. (2009). Stability of prostaglandin E(2) (PGE (2)) embedded in poly-D,L: lactide-co-glycolide microspheres: a pre-conditioning approach for tissue engineering applications. J Mater Sci Mater Med 20, 1357-1365. Weigelt, B., Peterse, J.L., and van 't Veer, L.J. (2005). Breast cancer metastasis: markers and models. Nat Rev Cancer 5, 591-602. Wheeler, M.L., and Pi-Sunyer, F.X. (2008). Carbohydrate issues: type and amount. J Am Diet Assoc 108, S34-39. Whiteside, T.L. (2010). Inhibiting the inhibitors: evaluating agents targeting cancer immunosuppression. Expert Opin Biol Ther 10, 1019-1035. Williams, J.A., and Shacter, E. (1997). Regulation of macrophage cytokine production by prostaglandin E2. Distinct roles of cyclooxygenase-1 and -2. J Biol Chem 272, 25693-25699. Wilson, H.L., and O'Neill, H.C. (2003). Murine dendritic cell development: difficulties associated with subset analysis. Immunol Cell Biol 81, 239-246. 191  Wolfs, I.M., Donners, M.M., and de Winther, M.P. (2011). Differentiation factors and cytokines in the atherosclerotic plaque micro-environment as a trigger for macrophage polarisation. Thromb Haemost 106. Xu, M., Mizoguchi, I., Morishima, N., Chiba, Y., Mizuguchi, J., and Yoshimoto, T. (2010). Regulation of antitumor immune responses by the IL-12 family cytokines, IL-12, IL-23, and IL-27. Clin Dev Immunol 2010. Yamashita, M., Tsuji, S., Nishiyama, A., Myrvik, Q.N., Henriksen, R.A., and Shibata, Y. (2007). Differential subcellular localization of COX-2 in macrophages phagocytosing heat-killed Mycobacterium bovis BCG. Am J Physiol Cell Physiol 293, C184-190. Yang, Y.X., Hennessy, S., and Lewis, J.D. (2004). Insulin therapy and colorectal cancer risk among type 2 diabetes mellitus patients. Gastroenterology 127, 1044-1050. Yannelli, J.R., and Wroblewski, J.M. (2004). On the road to a tumor cell vaccine: 20 years of cellular immunotherapy. Vaccine 23, 97-113. Yao, Y., Li, W., Kaplan, M.H., and Chang, C.H. (2005). Interleukin (IL)-4 inhibits IL10 to promote IL-12 production by dendritic cells. J Exp Med 201, 1899-1903. Yeo, W., Mok, T.S., Zee, B., Leung, T.W., Lai, P.B., Lau, W.Y., Koh, J., Mo, F.K., Yu, S.C., Chan, A.T., et al. (2005). A randomized phase III study of doxorubicin versus cisplatin/interferon alpha-2b/doxorubicin/fluorouracil (PIAF) combination chemotherapy for unresectable hepatocellular carcinoma. J Natl Cancer Inst 97, 1532-1538. Yin, L.L., Chung, C.M., Chen, J., Fok, K.L., Ng, C.P., Jia, R.R., Ren, X., Zhou, J., Zhang, T., Zhao, X.H., et al. (2009). A suppressor of multiple extracellular matrix-degrading proteases and cancer metastasis. J Cell Mol Med 13, 40344041. Yona, S., and Jung, S. (2010). Monocytes: subsets, origins, fates and functions. Curr Opin Hematol 17, 53-59. Yuan, J.N., Chao, Y., Lee, W.P., Li, C.P., Lee, R.C., Chang, F.Y., Yen, S.H., Lee, S.D., and Whang-Peng, J. (2008). Chemotherapy with etoposide, doxorubicin, cisplatin, 5-fluorouracil, and leucovorin for patients with advanced hepatocellular carcinoma. Med Oncol 25, 201-206. Yuhara, H., Steinmaus, C., Cohen, S.E., Corley, D.A., Tei, Y., and Buffler, P.A. (2011). Is Diabetes Mellitus an Independent Risk Factor for Colon Cancer and Rectal Cancer? Am J Gastroenterol.  192  Zenaro, E., Donini, M., and Dusi, S. (2009). Induction of Th1/Th17 immune response by Mycobacterium tuberculosis: role of dectin-1, Mannose Receptor, and DCSIGN. Journal of leukocyte biology 86, 1393-1401. Zhang, S., Hunter, D.J., Rosner, B.A., Colditz, G.A., Fuchs, C.S., Speizer, F.E., and Willett, W.C. (1999). Dietary fat and protein in relation to risk of non-Hodgkin's lymphoma among women. J Natl Cancer Inst 91, 1751-1758. Zhang, X., Goncalves, R., and Mosser, D.M. (2008). The isolation and characterization of murine macrophages. Curr Protoc Immunol Chapter 14, Unit 14 11. Zhang, Y., Liu, S., Liu, J., Zhang, T., Shen, Q., Yu, Y., and Cao, X. (2009). Immune complex/Ig negatively regulate TLR4-triggered inflammatory response in macrophages through Fc gamma RIIb-dependent PGE2 production. J Immunol 182, 554-562. Zhang, Y., Luo, F., Cai, Y., Liu, N., Wang, L., Xu, D., and Chu, Y. (2011). TLR1/TLR2 agonist induces tumor regression by reciprocal modulation of effector and regulatory T cells. J Immunol 186, 1963-1969. Zhou, B.B., Anderson, H.J., and Roberge, M. (2003). Targeting DNA checkpoint kinases in cancer therapy. Cancer Biol Ther 2, S16-22. Zhuang, L., Scolyer, R.A., Murali, R., McCarthy, S.W., Zhang, X.D., Thompson, J.F., and Hersey, P. (2010). Lactate dehydrogenase 5 expression in melanoma increases with disease progression and is associated with expression of BclXL and Mcl-1, but not Bcl-2 proteins. Mod Pathol 23, 45-53.  193  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0072765/manifest

Comment

Related Items