UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of mammary tumours on hematopoiesis and dendritic cell development Sio, Alexander 2013

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

Item Metadata

Download

Media
24-ubc_2013_fall_sio_alexander.pdf [ 5.62MB ]
Metadata
JSON: 24-1.0073993.json
JSON-LD: 24-1.0073993-ld.json
RDF/XML (Pretty): 24-1.0073993-rdf.xml
RDF/JSON: 24-1.0073993-rdf.json
Turtle: 24-1.0073993-turtle.txt
N-Triples: 24-1.0073993-rdf-ntriples.txt
Original Record: 24-1.0073993-source.json
Full Text
24-1.0073993-fulltext.txt
Citation
24-1.0073993.ris

Full Text

      THE EFFECTS OF MAMMARY TUMOURS ON HEMATOPOIESIS AND DEN- DRITIC CELL DEVELOPMENT  by  Alexander Sio  B.Sc., The University of British Columbia, 2013  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in  THE FACULTY OF GRADUATE STUDIES  (MICROBIOLOGY AND IMMUNOLOGY)  THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)     July 2013  © Alexander Sio, 2013 ii  Abstract  Tumours are often associated with defects in hematopoiesis and DC function. Leukocytosis, anemia, development of immunosuppressive cells such as MDSCs and tolerogenic DCs are all part of tumour development. In breast cancer, although these hematopoietic and DC de- fects may serve as prognostic indicators, little is known about their origin. Our studies show that mammary tumours affect the hematopoietic system leading to myeloproliferative-like disease characterized by neutrophilia, anemia, and defects in the HSPC compartment. These defects were associated with changes in global epigenetic regulation, and specifically in the upregulation and histone methylation status of Hoxa genes, most notably Hoxa9, a gene that critically controls HSPC differentiation. These changes in gene regulation and HSPC defects were found to be driven by tumour-secreted G-CSF. Additionally, our research shows that the generation of immunosuppressive DCs in mammary tumour bearing mice was associated with impaired DC differentiation, where DCs that develop in the presence of mammary tu- mours and their factors acquire immunosuppressive features. G-CSF played an important role in the suppression of DC development. Tumour derived factors induced FLT3L and GM- CSF DCs to expression Arginase-I, an enzyme known to suppress T cell proliferation and induce tolerance. Our data revealed the expression of Arg1 in these DCs was associated with enabling histone modifications in the Arg1 locus. Targeting of G-CSF may alleviate tumour- induced symptoms of anemia, leukocytosis, and DC defects, resulting in better patient sur- vival and preventing DC immunosuppression.   iii  Preface Contributions: Section 3 was based on work conducted at UBC, Harder Lab, by Dr. K Harder, A Sio, M Chehal, K Tsai, Dr. D Krebs, and XL Fan. In Section 3.1, M Chehal and I conducted the ma- jority of the experiments. I was responsible for analysis of the HSPC compartment, erythro- poiesis, myelopoiesis in the BM compartment, analysis of the functional significance of mammary tumours on the hematopoietic system, analysis of the FLT3L BM differentiation model, sample preparation for the epigenetic studies, analysis of the role of G-CSF in mam- mary tumour secreted factors and design of the experiments and data analysis. M Chehal was responsible for establishing the mammary tumour models, analysis of the HSPC compart- ment, erythropoiesis and myelopoiesis in the spleen of tumour mice, analysis of epigenetic regulation, experimental design and data analysis. K Tsai conceptualized and performed the experiments regarding histone methylation of hematopoietic regulatory genes, and Dr. D Krebs did the studies on the histone methylation status associated with hematopoietic regula- tory genes. Dr. K Harder conceptualized the entire project, designed the experiments, and provided feedback. In Section 3.2, M Chehal conducted the majority of experiments, while I contributed to a significant portion of the other experiments. I performed experiments to un- derstand the role of mammary tumours on FLT3L DC development, analyzed DC develop- ment in the BM of mammary tumour mice, analyzed the immunosuppressive effects of DCs on T cell expansion, investigated the role of G-CSF in DC development, designed the exper- iments and analyzed the data. M Chehal conceptualized the role of mammary tumour secret- ed factors on DC function and development, performed experiments to understand the role of mammary tumours on DC development in vitro and in vivo, investigated the role of G-CSF in DC development, designed the experiments and analyzed the data. XL Fan conceptualized, designed and performed the experiments to investigate the MT-CM suppression DC mediat- ed effector cell function. Dr. D Krebs performed the experiments on protein analysis of im- munosuppressive enzymes expressed by DCs under the influence of MT-CM. Dr. K Harder conceptualized the entire project, designed the experiments, and provided feedback. Section 4 was based on research and discussion written by Dr. K Harder, M Chehal, and I. Lastly, I contributed to the writing, figure making and editing of manuscripts based on Section 3.1 and 3.2, entitled “Dysregulated hematopoiesis caused by mammary cancer is associated with iv  changes and Hox gene expression in hematopoietic cells”, and “Impaired DC development and functionality in mammary tumour-bearing mice”, respectively.  Ethics approval: Animal experiments were performed in accordance with Canadian Council for Animal Care and the University of British Columbia Animal Care Committee guidelines. Protocol A09- 0814. v  Table of contents  Abstract .................................................................................................................................... ii Preface ..................................................................................................................................... iii Table of contents ..................................................................................................................... v List of tables.......................................................................................................................... viii List of figures .......................................................................................................................... ix List of abbreviations ............................................................................................................. xii Acknowledgements ............................................................................................................... xv Dedication ............................................................................................................................. xvi Section 1 Introduction ............................................................................................................ 1 1.1 Cancer and hematopoiesis ..................................................................................................... 1 1.1.1 Hematologic abnormalities in neoplastic disease ............................................................. 1 1.1.1.1 Anemia in neoplasia ................................................................................................. 1 1.1.1.2 Leukocytosis in neoplasia ........................................................................................ 2 1.1.2 Disruption of hematopoiesis in tumour-bearing hosts ...................................................... 3 1.1.2.1 Hox gene regulation of hematopoiesis ..................................................................... 3 1.1.2.2 The role of histone methylation in the regulation of Hox gene expression .............. 5 1.1.3 G-CSF and its role in hematopoiesis ................................................................................ 8 1.1.3.1 The role of G-CSF in cancer .................................................................................... 9 1.2 Cancer and the DC lineage .................................................................................................. 10 1.2.1 The DC lineage ............................................................................................................... 10 1.2.1.1 DC subsets ............................................................................................................. 11 1.2.1.2 DC function ............................................................................................................ 11 1.2.1.3 DC development .................................................................................................... 13 1.2.2 The role of DCs in neoplasia .......................................................................................... 16 1.2.2.1 DC localization, numbers and phenotype in neoplasia .......................................... 16 1.2.2.2 DC developmental defects in neoplasia ................................................................. 17 1.3 Research objectives ............................................................................................................. 18 vi  Section 2 Materials and methods ......................................................................................... 20 2.1 Tumour cells. ...................................................................................................................... 20 2.2 General reagents. ................................................................................................................. 20 2.3 Flow cytometry antibodies (Abs). ....................................................................................... 20 2.4 Competitive reconstitution and CFU-S12 assays. ................................................................ 21 2.5 qPCR. .................................................................................................................................. 21 2.6 Chromatin immunoprecipitation (ChIP). ............................................................................ 22 2.7 Methylcellulose assays. ....................................................................................................... 22 2.8 Western blot. ....................................................................................................................... 23 2.9 Mice. ................................................................................................................................... 23 2.10 In vitro BM analysis and DC culture. ................................................................................. 23 2.11 Tumour injection and tissue processing. ............................................................................. 23 2.12 Flow cytometry and FACS. ................................................................................................ 23 2.13 Menin inhibitor treatment of BM cultures. ......................................................................... 24 2.14 DC functional assays ........................................................................................................... 24 2.15 Statistical analysis. .............................................................................................................. 24 Section 3 Mammary cancer-induced abnormalities in hematopoiesis and dendritic cell development ........................................................................................................................... 25 3.1 Mammary cancer induces epigenetic changes in the hematopoietic system and defects in myelopoiesis and erythropoiesis ...................................................................................................... 25 3.1.1 Peripheral blood parameters of mammary tumour bearing mice .................................... 25 3.1.2 Leukemoid reaction in mammary tumour bearing mice ................................................. 26 3.1.3 Impaired erythropoiesis in mammary tumour bearing mice ........................................... 28 3.1.4 Defective hematopoietic stem/ progenitor cell numbers and localization in mammary tumour bearing mice .................................................................................................................... 31 3.1.5 Defects in bone marrow HSPC numbers and localization are not due to metastasis of MT-1 cells .................................................................................................................................... 34 3.1.6 Functional defects in the hematopoietic stem/ progenitor cell compartment in mammary tumour bearing mice .................................................................................................................... 34 3.1.7 In vitro differentiation of hematopoietic stem/ progenitor cells is impaired by mammary tumour secreted factors ................................................................................................................ 36 3.1.8 Regulation of Hox gene expression and histone modifications at the Hoxa9 locus by mammary tumour secreted factors ............................................................................................... 42 vii  3.1.9 The contribution of G-CSF in mammary tumour secreted factor suppression of hematopoietic stem/ progenitor cell differentiation ..................................................................... 46 3.1.10 Role of global H3K4me3 in mammary tumour secreted factor-mediated suppression of HSPC defects ........................................................................................................................... 46 3.2 Mammary cancer induces changes in DC development, function and phenotype .............. 48 3.2.1 Suppression of DC development and maturation by mammary tumours ....................... 48 3.2.2 The role of mammary tumour secreted factors in suppressing DC development in vitro – FLT3L 52 3.2.3 The role of mammary tumour secreted factors in suppressing DC development in vitro – GM-CSF ...................................................................................................................................... 58 3.2.4 Mammary tumour derived G-CSF is responsible for DC developmental impairment ... 62 3.2.5 MT-1 conditioned media and G-CSF mediate impairment of DC functional activity ... 64 3.2.6 Upregulation of immunosuppressive enzymes and markers by MT-1 conditioned media treatment of FLT3L and GM-CSF DCs is accompanied by changes in histone modifications ... 70 3.2.7 Mammary tumour conditioned media expands functional DC progenitors that can differentiate into pDCs and cDCs ................................................................................................ 75 Section 4 Concluding comments .......................................................................................... 77 4.1 Conclusion and discussion .................................................................................................. 77 4.2 Summary of implications and applications ......................................................................... 84 References .............................................................................................................................. 86 Appendix ................................................................................................................................ 96 Appendix A -- Primer sequences ..................................................................................................... 96 A.1 Sub-appendix – mRNA qPCR primers ........................................................................... 96 A.2 Sub-appendix – ChIP qPCR primers .............................................................................. 96 viii  List of tables  Table 3.1    Hematologic analyzes of MT-1 bearing mice ...................................................... 26  ix  List of figures  Figure 1.1    Hematopoietic stem cell differentiation. .............................................................. 4 Figure 1.2    Regulation of gene transcription by H3K4 and H3K27 methylation. .................. 6 Figure 1.3    DC differentiation. .............................................................................................. 15 Figure 3.1    Mammary tumours induce changes in granulocyte and monocyte numbers in mice. ........................................................................................................................................ 27 Figure 3.2    Mammary tumour development is associated with impaired BM erythropoiesis and heightened splenic erythropoiesis. ................................................................................... 29 Figure 3.3    Mammary tumour development is associated with reduced numbers of BM erythrocytes and increased numbers of splenic erythrocytes. ................................................. 30 Figure 3.4    Melanoma tumour development is not associated with impaired BM erythropoiesis. ......................................................................................................................... 31 Figure 3.5    Perturbations in hematopoietic stem/progenitor cell frequency in mammary tumour-bearing mice. .............................................................................................................. 32 Figure 3.6    Perturbations in hematopoietic stem/progenitor cell numbers and location in mammary tumour-bearing mice. ............................................................................................. 33 Figure 3.7    Melanoma tumour development does not disrupt the BM hematopoietic stem/progenitor compartment. ................................................................................................ 34 Figure 3.8    Mammary tumour growth is associated with significant functional changes in the stem/progenitor compartments in BM and spleen. ........................................................... 36 Figure 3.9    Mammary tumour conditioned media expands hematopoietic progenitors in in vitro BM cultures supplemented with FLT3L or GM-CSF. ................................................... 38 Figure 3.10    Mammary tumour conditioned media expands hematopoietic progenitors in in vitro BM cultures supplemented with FLT3L, while G-CSF neutralization suppresses mammary tumour conditioned media expansion of hematopoietic progenitors. .................... 39 Figure 3.11    Mammary tumour conditioned media expands hematopoietic progenitors in BM cultures supplemented with GM-CSF. ............................................................................ 40 Figure 3.12    Identification of cytokines produced by MT-1 cells. ....................................... 41 Figure 3.13    Mammary tumour growth is associated with enhanced Hoxa gene expression and changes in histone methylation. ....................................................................................... 43 x  Figure 3.14    Mammary tumour conditioned media treatment of bone marrow cultures is associated with enabling changes in histone methylation at the Hoxa9 locus. ....................... 45 Figure 3.15    G-CSF enhances Hoxa9 expression and CFC numbers, while Menin inhibitors reverse mammary tumour conditioned media induced expansion of LSKs, LKs and CFCs in in vitro BM cultures. ............................................................................................................... 47 Figure 3.16    DC progenitor gating scheme. .......................................................................... 48 Figure 3.17    Perturbations in DC progenitor numbers and location in mammary tumour bearing mice. ........................................................................................................................... 49 Figure 3.18    Perturbations in DC frequency in mammary tumour bearing mice. ................ 51 Figure 3.19    Perturbations in DC numbers in mammary tumour bearing mice. ................... 52 Figure 3.20    Mammary tumour conditioned media impairs DC differentiation in in vitro FLT3L DC cultures. ................................................................................................................ 53 Figure 3.21    Mammary tumour conditioned media impairs DC expression of MHCII and co- stimulatory receptors in in vitro FLT3L DC cultures. ............................................................ 55 Figure 3.22    Mammary tumour conditioned media enhances DC progenitor generation and impairs essential DC transcription factor mRNA expression in in vitro FLT3L DC cultures. ................................................................................................................................................. 57 Figure 3.23    Mammary tumour conditioned media impairs DC differentiation in in vitro GM-CSF DC cultures. ............................................................................................................ 59 Figure 3.24    Mammary tumour conditioned media impairs DC differentiation in in vitro GM-CSF DC cultures at different concentrations and treated for different periods of time. . 60 Figure 3.25    Mammary tumour conditioned media impairs cytokine secretion in in vitro GM-CSF DC cultures. ............................................................................................................ 61 Figure 3.26    Mammary tumour conditioned media derived G-CSF impairs DC development in in vitro FLT3L and GM-CSF DC cultures. ........................................................................ 63 Figure 3.27    Mammary tumours impair DC activation of NK and T cells. .......................... 65 Figure 3.28    Mammary tumours impair antigen specific DC activation of T cells and activates DC suppression of T cell proliferation..................................................................... 68 Figure 3.29    Mammary tumour conditioned media induces immunosuppressive protein expression in in vitro FLT3L and GM-CSF DC cultures. ...................................................... 71 xi  Figure 3.30    Mammary tumour conditioned media derived G-CSF contributes to Arg1 expression, while Ezh2, Hoxa9 and Arg1 are expressed at differential levels in CD11c +  and CD11c -  fractions in in vitro FLT3L and GM-CSF DC cultures. ............................................ 73 Figure 3.31    Mammary tumour conditioned media treatment of FLT3L DC cultures produce functional DC progenitors that differentiate into pDCs and cDCs. ........................................ 76  xii  List of abbreviations Ab antibody AML acute myeloid leukemia AngII Angiotensin II Arg1 Arginase I BM bone marrow cDC conventional dendritic cell CDP common dendritic cell progenitor CFSE carboxyfluorescein succinimidyl ester CFU colony forming unit CFU-G colony forming unit - granulocyte CFU-GM colony forming unit – granulocyte macrophage CFU-M colony forming unit - macrophage ChIP chromatin immunoprecipitation CLP common lymphoid progenitor CMP common myeloid progenitor DAPI 4',6-diamidino-2-phenylindole DC dendritic cell Fizz1 found in inflammatory zone 1 G-CSFR granulocyte colony stimulating factor receptor GMP granulocyte macrophage progenitor GRA granulocyte GrB granzyme B H&E hematoxylin & eosin H3K27 histone 3 lysine 27 H3K4 histone 3 lysine 4 HCT hematocrit HGB hemoglobin HSC hematopoietic stem cell HSPC hematopoietic stem/ progenitor cell LPS lipopolysaccharide xiii  LT-HSC long term hematopoietic stem cell LYM lymphocyte MDP monocyte dendritic cell progenitor MDS myelodysplastic syndrome MDSC myeloid derived suppressor cell MEP megakaryocyte erythrocyte progenitor MFP mammary fat pad MI Menin inhibitor MON monocyte MT mammary tumour MT-1-CM mammary tumour 1 conditioned media MT-CM mammary tumour conditioned media NK cell natural killer cell OVA ovalbumin PAMP pathogen associated molecular pattern PAS protein A sepharose PcG Polycomb Group pDC plasmacytoid dendritic cell PI propidium iodide PLT platelet PRC2 Polycomb Repressive Complex 2 PreDC dendritic cell precursor PRR pattern recognition receptor RBC red blood cell SQ subcutaneous ST-HSC short term hematopoietic stem cell T-CM tumour conditioned media Th T helper Treg regulatory T cell TrxG Trithorax Group VEGFa vascular endothelial growth factor a xiv  WBC white blood cell xv  Acknowledgements  I would like to thank…  Dr. K Harder, for your lessons on life and science.  Dr. P Johnson and Dr. N Abraham for your guidance and patience in helping me see my project come to fruition.  M Chehal, Ph.D. candidate, for enlarging my vision of science and providing suc- cinct, coherent and accurate answers to my endlessly long, convoluted, and often silly ques- tions.  E Gagnon, M.Sc., for sharing in my struggles and travails.  M Roberts and XL Fan, Ph.D. candidates, and Dr. D Krebs, for your humor and your love of science.  My family for the support they have given me in this time of protracted studentship.  God for giving me the strength to continue my journey. xvi  Dedication  To my family.   1  Section 1 Introduction  1.1 Cancer and hematopoiesis  1.1.1 Hematologic abnormalities in neoplastic disease Hematopoietic system abnormalities are commonly observed in cancer patients. In particular, anemia and proliferation of immature myeloid cells have been reported to be risk factors in cancer patients (Qiu et al., 2010a; Wilcox, 2010). Although anemia has been recognized in cancer patients for several decades, it is not known whether the induction of anemia is im- portant in the development of cancer, or whether it is a side-effect of the presence of cancer in patients. Similarly, the expansion of white blood cells of the myeloid lineage (leukemoid reaction) has long been associated with neoplastic disease. In recent years, this expansion of leukocytes has been linked to higher mortality rates and poor prognosis in malignancy, and has been described as a mechanism by which tumours suppress immunosurveillance, rejec- tion of malignant tissues and hamper immunotherapeutic efforts (Wilcox, 2010).  Despite the diverse functions of red blood cells and leukocytes of the myeloid lineage, these cell types originate from a common hematopoietic progenitor, the common myeloid progeni- tor (CMP). Due to the critical associations between leukocytosis or anemia with cancer pa- tient prognosis, and their common lineage, it is not unreasonable to speculate that the cancer- associated changes in myelopoiesis and erythropoiesis may be linked by a common pathway. Therefore, it is critical to understand the mechanism of how tumour cells influence hemato- poietic cell differentiation to determine the etiology of these hematopoietic aberrations. The elucidation of these mechanisms may lead to the discovery of therapeutic targets that func- tion to block cancer-induced aberrant myelopoiesis and erythropoiesis. 1.1.1.1 Anemia in neoplasia The quality of life of anemic cancer patients is rated lower than that of non-anemic patients (Leonard et al., 2005). Notably, pre-existing anemia in patients that have not undergone treatment is a phenotype that is negatively associated with the survival rate of breast, among other, cancer patients (Gislason and Nou, 1985; Metindir and Bilir Dilek, 2009; Qiu et al., 2  2010a; Qiu et al., 2010b). In one study, overall anemia among breast cancer patients was found to be higher than a control group with benign diseases (10.1% to 6.0%), while breast cancer patients with anemia had a lower 3-year survival rate compared to those without ane- mia (84.0% vs 92.2%) (Qiu et al., 2010a). In addition, anemia was shown to be an independ- ent prognostic factor in breast cancer patients (Qiu et al., 2010a). In a separate study, 44% of pre-treatment breast cancer patients were anemic, revealing that the rate of anemia among different populations of breast cancer patients vary widely, and that a high occurrence of anemia among breast cancer patients may be more evident in certain groups of patients com- pared to others (Tas et al., 2002). Nevertheless, anemia is a prevalent phenomenon among breast cancer patients, with possible prognostic value in determining patient survival rates. 1.1.1.2 Leukocytosis in neoplasia The cancer associated expansion of leukocytes, and in particular a subset of cells broadly termed myeloid derived suppressor cells (MDSCs) from the myeloid lineage, is associated with poor prognoses in cancer patients (Diaz-Montero et al., 2009; Gabitass et al., 2011; Qiu et al., 2010a). A large body of literature has shown that MDSCs are important for the surviv- al of tumours. Multiple studies have demonstrated the ability of MDSCs to suppress activa- tion of helper and cytotoxic T cells, to secrete factors that can suppress antigen presenting cells, and to directly prevent T cell activation through mechanisms such as the production of Arginase-I (Arg1) (Gabrilovich, 2004; Gabrilovich et al., 2012). MDSC infiltration of tu- mours also increases angiogenesis of tumour vasculature, thereby aiding tumour growth and establishment of a pre-metastatic niche (Shojaei et al., 2008; Wilcox, 2010). The expansion of these MDSCs has been linked to several immunologically relevant cytokines secreted by tumour cells, including IL-6, G-CSF, M-CSF and GM-CSF (Gabrilovich and Nagaraj, 2009). Results from recent studies indicate that the ability of certain chemotherapeutic treatments, such as gemcitabine and 5-fluorouracil, to contain tumour growth may be due to their ability to deplete or prevent MDSC expansion (Suzuki et al., 2005; Vincent et al., 2010; Wilcox, 2010). Due to the importance of MDSCs in mediating tumour survival by suppressing vari- ous functions of the immune system, these results provide a new link between chemothera- peutic suppression of tumour growth and immunological surveillance. 3  1.1.2 Disruption of hematopoiesis in tumour-bearing hosts Hematopoiesis involves the differentiation of bone marrow (BM) hematopoietic stem and progenitor cells (HSPCs) into more committed lineages of blood cells (Fig. 1.1). This process is governed by DNA-binding transcription factors and epigenetic modifications of chromatin that regulate cell fate decisions, maintenance of stem cell identity and self-renewal capacity (Cedar and Bergman, 2011). Recent studies have focused on studying the effects of tumour- development on leukocyte expansion, and, in turn, how these leukocytes, such as MDSCs, affect tumour growth. However, little is known about how tumours affect leukocyte devel- opment, the stage at which hematopoiesis is affected, and what types of cytokines are most relevant in tumour-induced hematopoietic abnormalities. Tumour-induced hematopoietic changes and their relationship with tumour growth and development need to be elucidated to determine their importance in cancer related symptoms, such as leukocytosis and anemia. 1.1.2.1 Hox gene regulation of hematopoiesis Key proteins involved in the process of hematopoietic differentiation and maintenance of hematopoietic stem cells (HSCs) include the Hox genes, and in particular, genes in the Hoxa cluster such as Hoxa9 (Argiropoulos and Humphries, 2007).  Loss-of-function mutations in Hoxa9 lead to reduced stem cell repopulating and self-renewal capacity, whereas gain-of- function mutations in Hoxa9 lead to increased stem cell repopulating capacity with concur- rent inability to differentiate into more restricted cell types (Argiropoulos and Humphries, 2007; Thorsteinsdottir et al., 2002). Hoxa9 gain-of-function in the HSC compartment has al- so been shown to lead to the development of myelodysplastic syndrome (MDS) or acute my- eloid leukemia (AML). MDS or AML involve the replacement of normal BM cells with rap- idly dividing, undifferentiated cells of the myeloid lineage, and often occur together with anemia and impaired erythropoiesis (Argiropoulos and Humphries, 2007; Tefferi and Vardiman, 2009). These changes in patients with MDS or AML are reminiscent of the leuko- cyte expansion and anemia observed in solid tumour patients.  4    Figure 1.1    Hematopoietic stem cell differentiation. LT-HSC, long term hematopoietic stem cell; ST-HSC, short term hematopoietic stem cell; MPP, multi-potent progenitor; CLP, common lymphoid progenitor; CMP, common myeloid progenitor; MEP, megakaryocyte erythrocyte progenitor; GMP, granulocyte macrophage progenitor; MDP, monocyte dendritic cell progenitor; cDC, conventional dendritic cell; pDC, plasmacytoid dendritic cell.  5  Although no studies have shown an association between the development of solid tumours and the dysregulation of HSPCs through aberrant expression of Hoxa genes, it is tempting to speculate that the phenotypic similarities between patients with MDS/AML and patients with solid tumours, including the expansion of leukocytes and induction of anemia, may be linked through the Hoxa gene cluster. This link may provide the rationale for use of therapies di- rected at MDS/AML as an adjuvant therapy to alleviate symptoms observed in solid tumour patients and increase patient survival. 1.1.2.2 The role of histone methylation in the regulation of Hox gene expression HSPC differentiation is governed by epigenetic regulation through the Polycomb Group (PcG) and Trithorax Group (TrxG) complexes, which modulate chromatin remodelling through progressive histone methylation (Fig. 1.2) (Zardo et al., 2008). The components SUZ12, Eed and EZH2 make up the epigenetic silencing Polycomb Repressive Complex 2 (PRC2). PRC2 is part of the PcG complexes and is important in the determination of cell fate in hematopoietic progenitor cells through its ability to tri-methylate histone 3 lysine 27 (H3K27), which silences gene transcription (Zardo et al., 2008). In opposition to the methyl- transferase activity of PRC2, histone demethylases such as JMJD3 have been shown to be important in Hox gene regulation, and their activity is associated with stem cell differentia- tion and rapid reduction in H3K27 methylation (Agger et al., 2007). In another mechanism indirectly opposing the activity of PRC2, the TrxG complexes such as MLL1-5 are important in the regulation of early hematopoietic progenitor cell fate through the tri-methylation of histone 3 lysine 4 (H3K4), which opens chromatin for permissive gene transcription (Zardo et al., 2008). The Menin protein, encoded by Men1, interacts with MLL for leukemic trans- formation of hematopoietic progenitors, and it has an essential role in hematopoiesis, where- by loss of Menin results in reduced fitness and repopulating capacity of HSCs under hemato- poietic stress (Maillard et al., 2009). 6   Figure 1.2    Regulation of gene transcription by H3K4 and H3K27 methylation. Tri-methylation of Histone 3 at lysine position 4 by MLL and other core components WDR5, RbBPS and Ash2L promotes gene transcription (left), while di- or tri-methylation of Histone 3 at lysine position 27 by EZH2, EED and SUZ12 prohibits gene transcription (right). Adapted from Wen H, Schaller MA, Dou Y, Hogaboam CM, Kunkel SL. Dendritic cells at the interface of innate and acquired immunity: the role for epigenetic changes. J Leukoc Biol. 2008;83:439-46. 7   Hoxa9 expression is regulated by PcG and TrxG complexes, linking epigenetic regulation of Hoxa9 with HSPC homeostasis and leukemic transformation (Cao et al., 2002; Cao and Zhang, 2004; Yan et al., 2006). In a mouse model where HSCs were sensitized to genetic mutations that change HSC function (c-Mpl background), the loss of the individual compo- nents of PRC2 – EZH2, SUZ12 and Eed – resulted in increased HSC activity, suggesting that PRC2 regulates HSC homeostasis through its restriction of HSC activity (Majewski et al., 2008; Majewski et al., 2010). An analysis of the genes affected by loss of SUZ12 showed that Hoxa9 was upregulated in HSC/ progenitors, linking the upregulation of Hoxa9 due to loss of PRC2 activity with increased HSC activity (Majewski et al., 2010). Loss of Eed in one study led to myeloproliferative defects in mice, resulting in an increase in primitive mye- loid progenitors, whereas overexpression of Ezh2 in a retroviral transfection model in HSCs reinforced long-term repopulating capacity (Kamminga et al., 2006; Lessard et al., 1999). Therefore, it is clear that PRC2 restricts HSC activity through its ability to repress the tran- scription of genes important in regulating hematopoiesis, such as Hoxa9.  Conversely, translocations of Mll with various fusion partners increased H3K4me3 in the Hoxa9 locus, which led to increased Hoxa9 expression and maintenance of cellular prolifera- tion and stem cell identity in leukemic cells (Chen et al., 2006; Faber et al., 2009; Hess, 2004; Horton and Huntly, 2012). Additionally, MLL is necessary for HSC development and the transition from HSC to multipotent progenitors due to its role in activating Hox gene transcription, indicating its importance in HSC biology (Ernst et al., 2004a; Ernst et al., 2004b). Thus, the activity of MLL opposes that of the PRC2 complex by tri-methylating H3K4me3 and allowing active gene transcription, especially in the Hox gene locus, thereby maintaining stem cell identity and HSC differentiation potential. Menin is also important in leukemic transformation and maintenance of HSC activity through its interactions with MLL, since the Menin/MLL interaction is required for the increased expression of Hoxa9 (Chen et al., 2006; Yokoyama et al., 2005). Deletion of Men1 in conditional Men1 knockout mice re- vealed reduced Hoxa9 expression and hematopoietic progenitor colony formation, while pe- ripheral blood leukocytes were also reduced (Chen et al., 2006).  In leukemic cells, disruption of the Menin/MLL interaction reduces Hoxa9 expression and proliferation in MLL fusion 8  protein-transformed bone marrow cells, in addition to inducing cellular differentiation, a sign of loss of leukemic activity (Grembecka et al., 2012). In summary, the opposing effects of PRC2 and Menin/MLL appear to counteract the actions of one another, and aberrant expres- sion of each component of these complexes leads to hematologic abnormalities through their modulation of Hoxa9 expression.  Due to the central role of Hoxa genes in hematopoiesis and leukemogenesis, and the capacity of tumours to induce leukocytosis and anemia in cancer patients, we hypothesize that dysreg- ulation of the Hoxa gene cluster through enabling histone methylations could be important in the development of these hematologic changes in cancer patients.  1.1.3 G-CSF and its role in hematopoiesis Discovered in the 1980’s, G-CSF is a hematopoietic cytokine that has a variety of functions important in hematopoiesis, including mobilization of hematopoietic stem cells and mainte- nance of the pool of neutrophils under steady-state and stress conditions (Lieschke et al., 1994; Liu et al., 1996; Panopoulos and Watowich, 2008). Individuals with neutropenia, either through congenital or acquired disorders, are highly immunodeficient and susceptible to bac- terial infections (Zeidler et al., 2003). G-CSF and G-CSFR knockout mice have a reduced neutrophil count and are much less able to fight off infection from Listeria monocytogenes (Lieschke et al., 1994; Zhan et al., 1998), Pseudomonas aeruginosa (Gregory et al., 2007), and Candida albicans (Basu et al., 2000). In addition to its role in regulating granulopoiesis, G-CSF directly increases neutrophil phagocytosis, mobilization and survival, thereby con- tributing to the functional aspects of neutrophilic granulocytes (Eyles et al., 2006).  Another way in which G-CSF controls hematopoiesis is its ability to induce hematopoietic stem cell mobilization. Injecting G-CSF alone, or G-CSF in combination with FLT3L, anoth- er cytokine important in controlling hematopoietic stem cell maintenance as well as dendritic cell (DC) differentiation, into healthy individuals or primates can induce the expansion and mobilization of HSPCs into the peripheral blood (Brasel et al., 1997; Papayannopoulou et al., 1997). The administration of G-CSF as a method of stem cell mobilization has been exploit- 9  ed as a means of collecting high levels of HSPCs from the peripheral blood of donors for stem cell transplantation in humans.  Studies on the role of G-CSF in HSC homeostasis are rare (Alexander, 2000). However, due to the broad expression of the G-CSFR on HSCs, it has been hypothesized that HSC homeo- stasis is controlled by the level of G-CSF (Liu et al., 2000). In addition, the repopulation ca- pacity of HSCs from G-CSFR -/-  mice are 5 times less efficient than that of wild type mice in irradiated recipients, indicating that G-CSF may be important in regulating physiological HSC function (Liu et al., 2000). 1.1.3.1 The role of G-CSF in cancer Tumours are known to hi-jack the immune system of the tumour-bearing host for immune- evasion and tumour promoting angiogenesis (Gabrilovich and Nagaraj, 2009; Gabrilovich et al., 2012). Due to the importance of G-CSF in hematopoiesis, it is not surprising that certain tumour types have been shown to upregulate the production of G-CSF, especially breast tu- mours (Lawicki et al., 2007; Lawicki et al., 2009). In particular, breast cancer patients have increased levels of G-CSF in tumour tissues compared to surrounding normal tissues (Park et al., 2011), in tumour tissues compared to tissues from control individuals (Chavey et al., 2007), and in the serum compared to control individuals (Park et al., 2011). Indeed, higher levels of G-CSF are associated with a poorer prognosis, increased tumour grade, and level of lymph node infiltration in breast cancer patients (Dehqanzada et al., 2007; Lawicki et al., 2007; Lawicki et al., 2009). Additionally, high levels of G-CSF administered to mice have been shown to impair BM erythropoiesis (de Haan et al., 1994; de Haan et al., 1992; Nijhof et al., 1994). Due the association between patient anemia and poor prognosis in many types of cancer, including breast cancer, and G-CSF’s ability to impair erythropoiesis, tumour pro- duction of G-CSF may affect patient outcome by impairing erythropoiesis, thus contributing to patient anemia.  Several studies have pointed to G-CSF’s ability to induce MDSCs as a means of suppressing the tumour-bearing host’s immune system and immunosurveillance, leading to enhanced tu- mour growth and survival. Thus, induction of MDSCs by tumour produced G-CSF provides 10  a possible mechanistic link between increased levels of G-CSF and increased risk of death. A study using the 4T1 mammary tumour model of breast cancer showed that these tumours, as well as metastases in the liver and lung, were highly infiltrated by MDSCs, implicating their involvement in tumour development and metastases (DuPre et al., 2007). These tumours also constitutively expressed g-csf, gm-csf and m-csf mRNA transcripts, and high levels of G-CSF were detected in the serum of 4T1 tumour-bearing mice (DuPre and Hunter, 2007). In anoth- er study, depletion of G-CSF using an anti-G-CSF antibody reduced the load of MDSCs and tumour growth, while delivery of exogenous G-CSF reduced tumour responsiveness to anti- VEGF treatment (Shojaei et al., 2009). These studies clearly show that G-CSF expands MDSCs (leukemoid reaction) in certain tumour types, which have a direct effect on aiding tumour development. In this way, G-CSF may also contribute to negative breast cancer pa- tient outcome.  In addition to its possible role in affecting survival of cancer patients and the expansion of MDSCs, G-CSF may also be pivotal in leukemic transformation (Beekman and Touw, 2010). Recent studies on the use of G-CSF as an adjuvant therapy to boost white blood cell produc- tion in breast cancer patients with severe neutropenia have shown a correlation between G- CSF administration and increased risk of developing MDS/AML (Citron et al., 2003; Cole and Strair, 2010; Hershman et al., 2007). Although the use of G-CSF increases the overall survival of breast cancer patients with severe neutropenia, the increased risk of leukemic transformation is evident.  These observations and reports support the idea that G-CSF, whether given as an adjuvant therapy or produced by tumours, affect hematopoiesis, myelopoiesis, erythropoiesis and leu- kemic transformation of myeloid cells resulting in reduced patient survival.  1.2 Cancer and the DC lineage  1.2.1 The DC lineage In 1973, Steinman and Cohn discovered cells in the peripheral lymphoid organs of mice that had never been observed before, and named these cells dendritic cells (DCs) for their den- 11  drite-like appendages (Steinman and Cohn, 1973). Subsequent analyzes on DCs showed that they are a population of cells that have a much higher capacity for induction of the mixed leukocyte reaction than any other known cell type at the time, and thus were powerful induc- ers of naïve T cell responses (Steinman and Witmer, 1978). DCs have now been shown to consist of multiple subtypes with specialized functions that can orchestrate innate and adap- tive immunity in ways not seen with any other type of cell. 1.2.1.1 DC subsets Although different organs and lymphoid tissues harbor different types of DCs, the splenic DC subsets harbor functionally equivalent subtypes to those that are present in other lym- phoid tissues. In the spleen, DCs are subdivided into three broad categories based on cell sur- face marker expression and function. These subsets include the CD8 + , CD4 +  and plasmacy- toid DCs (pDCs). It is well established that CD8 +  DCs have a high capacity for cross- presentation of antigen on MHCI that is necessary for the induction of CD8 +  T cell respons- es, whereas CD4 +  DCs can phagocytosis and present exogenous antigens which is important in determining the type of CD4 +  T cell responses (Coquerelle and Moser, 2010; Shortman and Heath, 2010). On the other hand, the pDC subset has been shown to be important in anti- viral responses (Coquerelle and Moser, 2010). Lastly, there are subsets of DCs that only arise after the occurrence of an insult as in the case of L. monocytogenes infections, and are thus termed inflammatory DCs (Coquerelle and Moser, 2010). These DCs arise from monocytic precursors. The specialized functions of each DC subset allows for a coordinated and specifi- cally catered response to an immunological insult, thereby orchestrating the optimal response from effector cells. 1.2.1.2 DC function In order to activate appropriate effector functions, DCs must detect the presence of foreign or non-self antigens prior to initiation of an immune response. Danger signals that belong to tumours, pathogens, and other non-self entities signal through pattern recognition receptors (PRRs) to activate inflammatory responses. DCs that encounter these molecular patterns be- come activated, resulting in 1. migration of DCs from peripheral tissues to secondary lym- phoid organs, 2. induction of the antigen presentation program, 3. reduction in phagocytosis 12  and processing of extracellular antigens, 4. upregulation of co-stimulatory molecules, and 5. secretion of pro-inflammatory cytokines (Banchereau et al., 2000; Coquerelle and Moser, 2010). This activation empowers the DC to potently stimulate T cell responses by engaging T cell receptors with MHCI- and MHCII-bound peptides and co-stimulatory molecules includ- ing CD80, CD86 and CD40L, as well as determining T cell fate through secretion of cyto- kines such as IL-12, TNF-α, IL-6 and IL-10 (Steinman and Idoyaga, 2010). Additionally, ac- tivated DCs are important in priming natural killer cells (NK cells) due to their secretion of IL-2 and trans-presentation of IL-15 through the IL-15Rα, as well as their production of type I interferons (Lucas et al., 2007).  Each DC subset has specialized functions based on their unique capacities to process antigen and detect danger signals. CD8 +  DCs’ ability to phagocytose and cross-present extracellular antigen on MHCI allows for the induction of CD8 +  T cell responses against foreign peptides from intracellular pathogens and tumour cells, while its superior ability to secrete IL-12 is thought to skew CD4 +  T cells towards a Th1 phenotype (Kronin et al., 2001; Pooley et al., 2001; Pulendran, 2005; Schnorrer et al., 2006; Shortman and Heath, 2010). Conversely, CD4 +  DCs can activate and skew CD4 +  T cells towards Th2 responses, due to their ability to efficiently present peptide antigen on MHCII and secrete IL-10 (Pooley et al., 2001; Pulendran, 2005). For pDCs, activation of anti-viral responses and recruitment of other im- mune cells occurs through their secretion of type I interferons (Jegalian et al., 2009). On the other hand, inflammatory DCs express iNOS and produce TNF-α, which have direct roles in killing bacteria, recruiting neutrophils and mediating inflammation (Serbina et al., 2003). Therefore, DC subsets are thought to be effective and specific in their responses to immuno- logical stimuli by their pre-determined capabilities. In this manner, the numbers and balance of DCs may also influence the type of immune response evoked (Reis e Sousa, 2006).  In addition to the activation of immunity, research in the past decade has shown that DCs are powerful inducers of tolerance through their expression of immunosuppressive enzymes and ability to activate regulatory T cells (Tregs). The expression of the immunosuppressive en- zyme Arg1 by DCs can metabolize and deplete L-arginine from the extracellular environ- ment, resulting in the suppression of T cell activation/ proliferation (Munder, 2009). On the 13  other hand, induction of Tregs by DCs can occur through the presentation of antigen without the concomitant expression of activating co-stimulatory molecules, as in the case of admin- istration of sterile antigen without a danger signal (Maldonado and von Andrian, 2010). Ad- ditional expression of tolerogenic receptors such as PD-L1 may mediate or enhance Treg in- duction by tolerogenic DCs (Francisco et al., 2009). In turn, Tregs suppress T helper (Th) cell functions in an antigen-specific manner, through the expression of immunosuppressive cytokines such as IL-10 and TGF-β, killing effector T cells by secretion of cytolytic en- zymes, disruption of T cell metabolism, and inducing tolerance in DCs in a positive feedback loop (Vignali et al., 2008). 1.2.1.3 DC development Due to the prominent role of DCs in immune responses, much recent research has been fo- cused on understanding DC ontogeny. In mice, DC development begins in the BM (Fig. 1.3). The LT-HSC at the top of the hierarchy gives rise to cells of the myeloid lineage. The mono- cyte dendritic cell progenitor (MDP) arises from the LT-HSC through a number of interme- diate progenitors, and the MDP is the precursor that produces monocytes and DCs (Liu et al., 2009). The common dendritic cell progenitor (CDP), derived from the MDP, is the main pro- genitor responsible for the production of most DC subsets at steady-state (Naik et al., 2007; Onai et al., 2007). These subsets include conventional DCs (including CD8 +  and CD4 +  DCs) through an intermediate precursor called the preDC, and pDCs (Fig. 1.3).  The successful development and differentiation of DC subsets is critically dependent on the expression of appropriate transcription factors that dictate lineage commitment. IRF-4, IRF- 8, and E2-2 are transcription factors with important roles in the development of specific DC subsets. IRF-4 has been shown to be critical in the development of the CD4 +  DC subset. CD4 +  DCs have a high expression of IRF-4 compared to CD8 +  DCs, and loss-of-function mutations in mice for IRF-4 have a significant impact on the numbers of CD4 +  DCs whereas CD8 +  DCs are unaffected (Tamura et al., 2005). Similarly, CD8 +  DC development critically depends on the expression of IRF-8. IRF-8 is expressed in CD8 +  DCs but not CD4 +  DCs, and a deficiency in IRF-8 in hematopoietic cells result in impaired CD8 +  DC differentiation in the lymphoid organs of mice (Aliberti et al., 2003; Schiavoni et al., 2002). Lastly, pDCs are de- 14  pendent on the expression of IRF-4, IRF-8 and E2-2, and knockout of IRF-8 and E2-2 in mice shows the most severe phenotype in losing almost all pDCs, while IRF-4 knockout mice show modestly reduced numbers of pDCs (Cisse et al., 2008; Geissmann et al., 2010; Schiavoni et al., 2002; Steinman and Idoyaga, 2010; Tsujimura et al., 2003). The importance of the transcription factors IRF-4, IRF-8 and E2-2 in determining DC differentiation and maintenance of steady-state DCs make them valuable targets for studying DC development. 15   Figure 1.3    DC differentiation. LT-HSC, long term hematopoietic stem cell; MDP, monocyte dendritic cell progenitor; CDP, common dendritic cell progenitor; pDC, plasmacytoid dendritic cell; preDC, dendritic cell precursor. 16  1.2.2 The role of DCs in neoplasia DCs play a multifaceted and critical role in immunity. The process by which DCs recognize, acquire, process and present antigens to lymphocytes to destroy pre-malignant cells is known as immunosurveillance, and is now thought to prevent most neoplasia from turning into full- blown cancerous tumours (Dunn et al., 2002; Dunn et al., 2004; Schreiber et al., 2011). It is therefore not unexpected that tumours either co-opt DCs to mediate immunosuppression and/or suppress the activation of DCs to escape immune recognition (Gabrilovich, 2004). This can occur through two mechanisms: reducing the numbers of mature DCs (suppressing immune activation) and skewing the balance of DCs to increase those with an imma- ture/inhibitory phenotype (inducing immunosuppression). 1.2.2.1 DC localization, numbers and phenotype in neoplasia DCs are often physically associated with solid tumours. In particular for breast cancer, one study has shown that over 90% of breast cancer patients had tumours that were penetrated by DCs of an immature phenotype, whereas 60% of patients had mature dendritic cells that were sequestered in the peritumoural areas (Bell et al., 1999). Sequestration of mature DCs from tumours may function to prevent DCs from inducing a robust T cell response by preventing the acquisition of tumour antigen.  In addition to the high density of immature DCs associat- ed with tumour tissues, another study has shown that the numbers of mature DCs in the pe- ripheral blood of breast cancer patients was severely reduced while immature DCs became more abundant, suggesting a systemic dysfunction in DC differentiation (Pinzon-Charry et al., 2005). This increase in immature DCs, which were shown to be functionally poor stimu- lators of T cell proliferation and IFN-γ production, was associated with breast cancer metas- tasis (Pinzon-Charry et al., 2005). Other studies have revealed that the overall level of pe- ripheral blood DCs in breast, head and neck, and lung cancer patients was much lower com- pared to healthy controls (Almand et al., 2000; Della Bella et al., 2003). These breast cancer- induced abnormalities in DC localization and maturation are important mechanisms through which tumours prevent immune cell activation.  Aside from aberrations in localization and numbers, the phenotype of tumour-associated DCs is generally immunosuppressive or Th2 prone. These DCs cannot induce a robust T cell re- 17  sponse against tumour cells, and may even induce immunosuppression themselves. Experi- ments using a rat colon carcinoma tumour model showed that DCs that had infiltrated the tumours express little CD80/CD86, and were poor stimulators of allogeneic T cell prolifera- tion (Chaux et al., 1997). Syngeneic melanoma tumours in murine hosts directly converted DCs into tolerogenic cells that secreted TGF-β, which induced the polarization of CD4+ T cells into Tregs (Ghiringhelli et al., 2005). DCs derived from the peripheral blood of breast cancer patients were poor stimulators of the mixed leukocyte reaction and autologous T cell proliferation, compared to DCs from healthy individuals (Satthaporn et al., 2004). Other studies demonstrated that tumours tended to induce a Th2-prone response in a DC-dependent manner, which promoted cancer progression (Ochi et al., 2012; Olkhanud et al., 2011; Pedroza-Gonzalez et al., 2011). In the 4T1 breast cancer model, tumour-infiltrating CD4 +  T cells that secreted Th2 cytokines increased primary tumour progression, and this tumour pro- gression was enhanced by the co-transfer of BM-derived DCs (Olkhanud et al., 2011). In an- other model of breast cancer using human-derived cancer cells, DCs responded to tumour- derived TSLP by inducing the expression of OX40L that mediated the development of Th2 cells (Pedroza-Gonzalez et al., 2011). Neutralization of either TSLP or OX40L reduced the production of the Th2 cytokines IL-4 and IL-13, while suppressing tumour development in vivo (Pedroza-Gonzalez et al., 2011). Therefore, sequestration of DCs away from tumour an- tigens, systemic depletion of DCs, and induction of immunosuppressive/Th2 prone DCs may all contribute to tumour-induced immune escape. 1.2.2.2 DC developmental defects in neoplasia Although the ability of DCs to mount an anti-tumour response and the numbers of DCs is clearly affected by tumour burdens in both mouse models and human cancer patients, little is known about the changes in DC progenitors in cancer. Evidence suggests that DC develop- ment may be impaired and/or inappropriately shifted towards an immunosuppressive pheno- type, although no clear link has been established between tumour-induced DC developmental changes and their effects on DC functions (Gabrilovich, 2004). The recent body of work on dissecting the DC developmental pathway has paved the way for studying tumour-induced changes in DC development. One study demonstrated that tumour infiltrating preDCs were not functionally different compared to splenic preDCs in melanoma tumour bearing mice. 18  Adoptively transferred preDCs from control mice into tumour bearing mice differentiated into cDCs in the spleen and tumour to a similar degree (Diao et al., 2010). There was no dif- ference between tumour cDCs and splenic cDCs’ ability to induce OT-II cell proliferation, upregulate co-stimulatory molecules, or induce a mixed leukocyte reaction, suggesting that there were no functional differences between splenic and tumour preDC-derived cDCs (Diao et al., 2010). It is worth noting that the comparisons in preDC differentiation and cDC func- tion were made between cDCs from the spleen and tumour of the same tumour-bearing mouse. Systemic defects in DC development induced by tumours may not be detected in this system, in contrast to a system testing preDC and cDC function between tumour and control mice. In addition, since only preDC differentiation was assessed in this study, the effects of tumour development on more primitive DC progenitor development upstream of preDCs may not have been revealed. Analysis of mice with larger tumour sizes may also reveal a more severe impairment in DC development. Diao et al. only analyzed mice carrying small (<0.5cm) tumour burdens (Diao et al., 2010). Therefore, many questions remain in regards to whether and how DC development is negatively impacted by tumour burdens, and the conse- quences of these changes in promoting tumor development.  1.3 Research objectives Hypothesis 1: Mammary tumours induce epigenetic dysregulation of the hematopoietic sys- tem leading to anemia, aberrant myelopoiesis and abnormalities in HSPC numbers and local- ization. Objectives for hypothesis 1: Aim A: To describe the myeloid, erythroid and other hematopoietic abnormalities in mamma- ry tumour bearing mice Aim B: To investigate mammary tumour-induced impairment of hematopoiesis by thorough analysis of hematopoietic stem/progenitor function, localization and subset composition Aim C: To investigate tumour-induced changes in hematopoietic differentiation using an in vitro system of hematopoietic differentiation Aim D: To determine the factor(s) in mammary tumour conditioned media responsible for dysregulation of hematopoiesis in vivo and in vitro 19  Aim E: To determine the factor(s) responsible for mammary tumour-mediated impairment of hematopoietic differentiation Aim F: To describe the mammary tumour mediated changes in the epigenetic regulation of hematopoietic transcription factors in hematopoietic stem/ progenitor cells Aim G: To test the role of mammary tumour conditioned media induced elevation of activat- ing histone modifications in impairing hematopoietic cell differentiation  Hypothesis 2: Mammary tumour-derived factors modify DC development to preferentially give rise to DCs that potentiate immunosuppression/ suppress immune activation Objectives for hypothesis 2: Aim A: To characterize mammary tumour induced changes in DC development and matura- tion Aim B: To determine the effects of mammary tumour secreted factors on DC development Aim C: To investigate the mammary tumour derived factors responsible for suppression of DC development Aim D: To describe the functional impact of mammary tumours on DC function Aim E: To determine the relationship between mammary tumour induction of immunosup- pressive enzymes in DCs and corresponding histone modifications 20  Section 2 Materials and methods 2.1 Tumour cells. NOP cell lines were derived from spontaneous tumours in mice expressing neu linked to the OVA OTI/OTII peptide sequences and a dominant-negative p53 transgene (Wall et al., 2007; Yang et al., 2009).  NOP cells were maintained in NOP medium (RPMI 1640 (Invitrogen) supplemented with 10% heat inactivated (HI)-FBS, 100U/mL penicillin-G, 100g/mL strep- tomycin, 2mM glutaMAX, 10µM 2-mercaptoethanol and insulin/transferrin/selenium (Lon- za)). NOP12, 18, 21, 23 cell lines are referred to as mammary tumour (MT) -1, 2, 3, 4, re- spectively. CMT93, 4T1, and B16 cell lines were maintained in DMEM with 10% HI-FBS. Tumour-conditioned media (T-CM) was produced by culturing cells at 5×10 6 cells/25mL for 4 days. Cellular debris was removed by centrifugation and T-CM was stored at -80°C.  2.2 General reagents. GM-CSF and FLT3L were produced and standardized in our laboratory.  Menin inhibitors (MI-2/MI-3) and control inhibitor (MI-nc) have been previously described (Grembecka et al., 2012).  Methocult GF 3534 was purchased from StemCell Technologies. Recombinant cyto- kines used in in vitro assays were purchased from Peprotech, eBioscience and R&D Systems, including IL-3, IL-6, IL-10, VEGF, G-CSF, M-CSF, GM-CSF, TPO, TSLP and SCF. Neu- tralizing antibodies against G-CSF (9B4CSF) and IL-6 (MP5-20F3) and isotype control Abs were purchased from eBioscience. The Proteome Profiler Mouse Cytokine Array Kit, Panel A, used to determine MT-1 cytokine levels was purchased from R&D Systems.  2.3 Flow cytometry antibodies (Abs). Antibodies were from eBioscience, BD biosciences, Beckman Coulter or produced in-house (Biomedical Research Centre): CD11c (N418), CD11b (M1/70), Gr-1 (RB6-8C5), CD115 (AFS98), MHC class II (I-A/I-E), CD19 (eBio1D3), DX5 (DX5), Ter119 (TER119), CD71 (R17217), c-kit (2B8), Flt3 (A2F10), CD45.1 (A20), CD45.2 (104), Sca-1 (D7), FcγRII/III (2.4G2), CD34 (RAM34), CD24 (M1/69), SIRPα (P84), CD8α (53-6.7), CD4 (RM4-5), CD40 (HM40-3), CD80 (16-10A1), CD86 (GL-1), PD-L1 (M1H5), Ly6c (AL-21), B220 (RA3-6B2), CD44 (IM7), Thy1.2 (53-2.1), NK1.1 (PK136), IL7Rα (A7R34).  The lineage cocktail for HSC analysis contained Abs recognizing Gr-1, CD3, CD11b, CD19, Ter119, 21  DX5, MHCII, CD11c, ± FceR1. The lineage cocktail for DC progenitor analysis contained Abs recognizing Gr-1, CD3, CD11b, CD19, Ter119, DX5, MHCII, IL7Rα.  2.4 Competitive reconstitution and CFU-S12 assays. Recipient congenic BoyJ (CD45.1 + ) mice were subjected to lethal irradiation (650rads, 2 doses, 4hr apart) and injected IV with 5×10 6  BM cells or splenocytes from control or tumour bearing mice (CD45.2 + ) the following day. For reconstitution with in vitro cultured cells, BM cells from CD45.2 +  UBC-GFP mice were cultured in vitro as described for 8-10 days in the presence of FLT3L and MT-CM.  FACS sorted lineage - Sca-1 + c-kit +  (LSK) cells, or unfrac- tionated cells from in vitro BM cultures, were injected into sublethally irradiated RAG1 -/-  (500rads, 1 dose) or lethally irradiated BoyJ (650rads, 2 doses, 4hr apart) mice. Competitive repopulation experiments used 5×10 6 CD45.1 + CD45.2 +  BM carrier cells.  Donor cells were distinguished by expression of CD45.2 and/or GFP expression. For peripheral blood analysis to assess hematopoietic reconstitution, RBCs were lysed and blood was analyzed by flow cy- tometry at the indicated times. For the CFU-S12 assay, 5×10 5  BM cells or splenocytes from control or tumour bearing mice were injected IV into lethally irradiated (900rads, 1 dose) re- cipient C57BL/6 mice. Recipients were euthanized 12 days post-transfer and spleens were collected and fixed in Carnoy’s solution (60% ethanol, 30% chloroform and 10% glacial ace- tic acid) for colony enumeration.  2.5 qPCR. Total RNA was isolated using the RNeasy kit (Qiagen) or TRI Reagent (Invitrogen). Reverse transcription was performed using the iScript cDNA Synthesis Kit (BioRad).  Blood collect- ed from control and tumour bearing mice via cardiac punctures was lysed in TRI Reagent (Invitrogen) after red cell lysis. qPCR was performed using the SsoFast EvaGreen real-time PCR mix (BioRad) on a Bio-Rad CFX96 instrument, and analyzed using CFX Manager software. Target genes were normalized to GAPDH or Rps29. For statistical analyses, error bars indicate SEM within a sample. For primers, see appendix A1.    22  2.6 Chromatin immunoprecipitation (ChIP). 10 7  cells were suspended in 5mL medium and cross-linked by adding formaldehyde 37% to 1% final for 10min at room temperature.  Glycine was added to a final concentration of 125mM for 5min to stop the cross-linking reaction.  Cells were rinsed and lysed at 4×10 7  cells/mL in SDS lysis buffer (1% SDS, 10mM EDTA, 50mM Tris pH 8.1, protease inhibitor cocktail (Roche)).  Lysate was diluted with 3 volumes of ChIP dilution buffer (0.01% SDS, 1.1% Triton-X 100, 1.2mM EDTA, 16.7mM Tris-HCl, 167mM NaCl, protease inhibitor cocktail (Roche)), incubated on ice 15min, transferred to 15mL polystyrene tubes (BD Bio- sciences) and sonicated 10-15min using a Bioruptor (Diagenode), high setting, 30sec on 30sec off pulse.  To prepare “Input” DNA, 50µL sonicated lysate was mixed with 350µL elu- tion buffer (1% SDS, 100mM NaHCO3) containing 25µg proteinase K (Invitrogen) and incu- bated at 65ºC for 4-5hr to reverse cross-links. DNA was purified using the phenol chloroform method and DNA concentration was determined.  For ChIP, lysates were further diluted 2- fold with ChIP dilution buffer.  Lysates containing 12.5µg DNA were precleared on 40µL protein A Sepharose (PAS) beads.  Lysates were transferred to new tubes and incubated with precipitating Abs overnight at 4ºC with rocking.  Precipitates were collected for 1.5hr with 40µL PAS beads coated with sheared salmon sperm DNA (75ng/µL beads) and BSA (0.1mg/µL beads).  Precipitates were washed 3X with low salt wash buffer (20mM Tris HCl pH 8, 2mM EDTA, 1% Triton X-100, 0.1% SDS, 150mM NaCl, protease inhibitor cocktail) and 1X with 1mL high salt buffer (20mM Tris HCl pH 8, 2mM EDTA, 1% Triton X-100, 0.1% SDS, 500mM NaCl, protease inhibitor cocktail).  DNA was eluted in 400µL elution buffer by incubating at 30ºC for 15min.  Eluate was transferred to new tubes and incubated with 25µg proteinase K at 65ºC for 4-5hr. DNA was purified using the phenol chloroform method and analyzed by qPCR. Antibodies used: H3K4me3 (Cell Signalling, #9571), H3K27me3 (Millipore, #07-449). For primers, see appendix A2.  2.7 Methylcellulose assays. The methylcellulose assay was carried out according to manufacturer’s instructions (Stemcell Technologies). Briefly, cell samples were resuspended in 2% HI-HBS, IMDM, and a 1:10v/v ratio of cells to Methocult were mixed and plated in 3cm plates. CFCs were typed and enu- merated 12 days later. 23   2.8 Western blot. Abs used: H3K27me3 (Millipore, #07-449), H3K4me3 (Cell Signaling #9751S), histone H3 (Cell Signalling, #9715), Arg1 (BD Biosciences, #610709).  2.9 Mice. MMTVneu OTI/OTII  mice were previously described (Wall et al., 2007; Yang et al., 2009). Mice were C57BL/6 genetic background (>10 generations). Animal experiments were per- formed in accordance with Canadian Council for Animal Care and the University of British Columbia Animal Care Committee guidelines. Mice were age/sex-matched and analyzed be- tween 9-14 wks of age.  2.10 In vitro BM analysis and DC culture. BM was cultured in NOP medium without insulin/transferrin/selenium in the presence of cy- tokines at 1×10 6 cells/mL for 3-8 days. For GM-CSF experiments, BM was cultured at 2×10 5 cells/mL unless otherwise indicated. For T-CM treatments, BM cultures were supple- mented with 25% T-CM unless otherwise indicated.  2.11 Tumour injection and tissue processing. Mice were injected with 1-5×10 6 MT cells subcutaneously (SQ) in the left flank or mammary fat-pad (MFP), or 7.5×10 5  B16 cells (SQ).  Spleens, femurs and blood were analyzed once tumours reached 1cm 3 . Blood (cardiac punctures) was analyzed on the scil Vet abc hematol- ogy analyzer (scil animal care company) or treated with ammonium chloride before flow cy- tometry.  2.12 Flow cytometry and FACS. Cells were incubated with 2.4G2 mAb (Fc block) and stained with fluorochrome-labelled antibodies (Abs). Data was acquired on an LSRII (BD Biosciences) using FACS Diva soft- ware and analyzed with FlowJo software (TreeStar).  Dead cells were excluded based on propidium iodide (PI) or 4',6-diamidino-2-phenylindole (DAPI) uptake, red cells were ex- cluded by lysis and size. 24   2.13 Menin inhibitor treatment of BM cultures. BM cells were treated with Menin inhibitors, MI-2/MI-3, on day 1 of culture and analyzed on day 3 for CFCs.  Alternatively, inhibitors were added on day 3 and cells were analyzed on day 8 by flow cytometry/qPCR.  2.14 DC functional assays FLT3L or GM-CSF DCs were cultured for 8-10 days. For phagocytosis assays, GM-CSF DCs were co-cultured with doxorubicin treated CFSE (carboxyfluorescein succinimidyl es- ter) labelled tumour cells (B16-OVA or MT-1) for 16 hr. Cells were analyzed by flow cy- tometry. For DC:T cell co-culture experiments, GM-CSF or FLT3L DCs (control or 25% MT-1-CM treated) were incubated with SIINFEKL peptide, ovalbumin (OVA) protein, or doxorubicin-treated B16-OVA or MT-1 cells for 16 hr, then co-cultured with OTI, OTII or B3Z cells for 2-3 days. Supernatant was collected and analyzed for cytokine content by ELI- SA. To measure T cell proliferation in DC:T cell co-culture experiments, OTI or OTII cells were labelled with CFSE prior to co-culture with DCs, and proliferation was measured by CFSE dilution. For T cell suppression assays, GM-CSF or FLT3L DCs (control, 25% MT-1- CM treated, or 20ng/mL G-CSF treated) were co-cultured with OTI or OTII splenocytes in the presence of OVA protein for 2-3 days. T cell proliferation was measured by flow cytome- try analysis of CFSE dilution. For NK cell functional assays, mice were injected with LPS for 16 hr, and splenocytes were extracted and analyzed by flow cytometry for IFN- and granzyme B (GrB) expression. For cytokine expression, GM-CSF DCs were stimulated with LPS for 16 hr, supernatant was collected and analyzed for cytokine content by ELISA.  2.15 Statistical analysis. Statistical comparisons were performed using unpaired t-test. Error bars and ± symbols rep- resent SEM. *, P < 0.05, **, P < 0.01, ***, P < 0.001, ****, P < 0.0001. 25  Section 3 Mammary cancer-induced abnormalities in hematopoiesis and dendritic cell development  3.1 Mammary cancer induces epigenetic changes in the hematopoietic system and defects in myelopoiesis and erythropoiesis  3.1.1  Peripheral blood parameters of mammary tumour bearing mice Due to the presence of hematologic abnormalities in cancer patients, the blood compartment of mammary tumour bearing mice was analyzed for evidence of anemia and leukocytosis. Automated analysis of hematologic parameters was performed on the peripheral blood of MT-1 bearing mice on RAG1 -/-  (immunodeficient) or MMTVneu (immunosufficient) back- grounds. MT-1 tumours induced a ~10 fold expansion of granulocytes and significant reduc- tion in red blood cells and platelets compared to control mice (Table 3.1). Therefore, leuko- cytosis (abnormal expansion of white blood cells), anemia and thrombocytopenia (abnormal reduction in platelets) were evident in MT-1 bearing mice, similar to phenotypes observed in cancer patients. The changes in multiple mature hematopoietic lineages suggested that MT-1 tumours were impairing the functions of cells in the primitive hematopoietic stem/progenitor compartment that give rise to red blood cells, platelets and leukocytes. Notably, lymphocyte numbers remained normal in MT-1 bearing mice on the MMTVneu background, suggesting that lymphocyte differentiation was not affected by the presence of the MT-1 tumour.  26  Table 3.1    Hematologic analyzes of MT-1 bearing mice  RAG1 -/-  MMTVneu  control MT-1 control MT-1 WBC, 10 3 /µL 3.7 ± 0.5 19.8 ± 2.6**** 7.9 ± 0.9 30.8 ± 6.8* RBC, 10 6 /µL 9.9 ± 0.1 8.3 ± 0.4*** 10.3 ± 0.1 9.2 ± 0.4* HGB, g/dL 14.9 ± 0.1 12.6 ± 0.5*** 15.0 ± 0.2 13.5 ± 0.5* HCT, % 46.4 ± 0.6 39.1 ± 1.6*** 46.6 ± 1.1 41.7 ± 1.7 PLT, 10 3 /µL) 1103 ± 59 884 ± 57* 1119 ± 30 866 ± 46** LYM, 10 3 /µL 0.8 ± 0.2 0.7 ± 0.1 4.7 ± 0.6 5.5 ± 0.8 MON, 10 3 /µL 0.3 ± 0.0 0.7 ± 0.1*** 0.8 ± 0.1 1.5 ± 0.2* GRA, 10 3 /µL 2.5 ± 0.3 18.3 ± 2.5**** 2.4 ± 0.2 23.9 ± 6.1*  RAG1 -/- : Control n=13, MT-1 n=12; MMTVneu: Control n=4, MT-1 n=4. Blood was ob- tained from cardiac puncture and analyzed on a scil Vet abc hematology analyzer, scil an- imal care company, Gurnee, IL. *, P < .05; **, P < .01; **, P < .001; ****, P < .0001. WBC indicates white blood cell; RBC, red blood cell; HGB, hemoglobin; HCT, hemato- crit; PLT, platelet; LYM, lymphocyte; MON, monocyte; GRA, granulocyte.  3.1.2 Leukemoid reaction in mammary tumour bearing mice To fully characterize the changes observed in the leukocyte compartment of MT-1 bearing mice, flow cytometry analysis of the myeloid compartment, including monocytes (CD11b + CD115 + ) and granulocytes (CD11b + Gr-1 + ) was performed. Analyzes on the spleen, BM and blood compartments revealed that MT-1 bearing mice showed an expansion of granulocytes in all three organs (Fig. 3.1A). Notably, in a comparison between control, MT-1 or B16 (melanoma) tumour bearing mice, the elevation of the CD11b + Gr-1 +  granulocytic population was only observed in the MT-1 bearing mice (Fig. 3.1A). Total BM cellularity in MT-1 bearing mice was reduced along with CD11b + CD115 +  monocytes (Fig. 3.1B), while total splenocytes, splenic monocytes and granulocytes were expanded. This data shows that MT-1 induced BM hypocellularity and splenic hypercellularity in tumour bearing mice, with the overall expansion of the CD11b + Gr-1 +  granulocyte population. This expansion of the CD11b + Gr-1 +  population correlated with the leukocytosis detected by automated hematologic analyzes. 27   Figure 3.1    Mammary tumours induce changes in granulocyte and monocyte numbers in mice. (A) Frequencies of CD11b + Gr-1 hi  and CD11b + Gr-1 int  cells in BM, blood and spleen of control, MT-1 and B16 tumour-bearing mice. Mouse genetic backgrounds include MMTVneu (MT-1), C57BL/6 (B16) for immunosuf- ficient models, and RAG1 -/-  (MT-1 and B16) for immunodeficient models. (B) Absolute numbers of unfraction- ated cells, monocytes (gated as CD115 hi CD11b + Gr-1 low/neg ) and granulocytes (gated as CD115 low/neg CD11b + Gr- 1 hi ) in spleen and BM of control and MT-1 bearing mice. Data are representative of a minimum of three inde- pendent experiments, n=3. 28  3.1.3 Impaired erythropoiesis in mammary tumour bearing mice The occurrence of anemia in MT-1 bearing mice warranted further examination of the BM and splenic erythroid compartment to determine whether the reduction in RBCs was due to impaired erythropoiesis. Macroscopic and histological analysis (H&E) of femurs and BM sections of MT-1 bearing mice showed pale bones and a reduction of red blood cells, respec- tively (Fig. 3.2A). Normal BM cells were replaced with cells that resembled granulocytes and cells with a blast-like morphology (Fig. 3.2A). The BM erythrocyte compartment in MT- 1 bearing mice showed a reduction in the frequencies and absolute numbers of CD71 + Ter119 +  and CD71 - Ter119 +  immature erythroid cells, whereas no changes in the abso- lute numbers of CD71 + Ter119 -  cells was observed, suggesting impaired erythrocyte produc- tion (Fig. 3.2B,3.3A,B). In contrast, CD71 + Ter119 + , CD71 - Ter119 +  and CD71 + Ter119 -  splenic subsets of erythrocyte progenitors were significantly higher in MT-1 bearing mice compared to controls (Fig. 3.2C,3.3C). This data indicates that MT-1 growth increased eryth- ropoiesis the spleen, and impaired erythropoiesis in the BM. Interestingly, elevated splenic erythropoiesis did not fully compensate for decreased BM RBC production, as demonstrated by the anemia in MT-1 bearing mice (Table 3.1). In comparison, B16 tumours did not induce changes in erythroid progenitors in the BM, although there was a variable increase in erythroid progenitors in the spleen of RAG1 -/-  B16 tumour bearing mice (Fig. 3.3C, 3.4). 29   Figure 3.2    Mammary tumour development is associated with impaired BM erythropoiesis and height- ened splenic erythropoiesis. (A) Femurs from control and MT-1 bearing mice (left), H&E stained BM sections imaged at ×10 (middle) and ×40 (right) original magnification. (B) Frequencies of CD71 + Ter119 - , CD71 + Ter119 +  and CD71 - Ter119 +  erythroid populations in the BM of control and MT-1 bearing mice. SQ, subcutaneous; MFP, mammary fat-pad. (C) Frequencies of splenic CD71 + Ter119 - , CD71 + Ter119 +  and CD71 - Ter119 +  erythroid populations in control and MT-1 bearing RAG1 -/-  and MMTVneu mice. Data are representative of three independent experiments, n=3-4. 30   Figure 3.3    Mammary tumour development is associated with reduced numbers of BM erythrocytes and increased numbers of splenic erythrocytes. (A) Absolute numbers of erythroid populations in the BM of RAG1 -/-  mice, or (B) MMTVneu mice.  (C) Abso- lute numbers of CD71 + Ter119 - , CD71 + Ter119 +  and CD71 - Ter119 +  erythroid populations in the spleens of con- trol and MT-1 (MMTV, RAG1 -/- ), or control and B16 tumour-bearing (C57BL/6, RAG1 -/- ) mice. SQ, subcuta- neous; MFP, mammary fat-pad. Data are representative of three independent experiments, n=3-4. 31   Figure 3.4    Melanoma tumour development is not associated with impaired BM erythropoiesis. Absolute numbers of CD71 + Ter119 - , CD71 + Ter119 +  and CD71 - Ter119 +  erythroid populations in the BM of C57BL/6 or RAG1 -/-  mice with B16 tumours.  Data are representative of three independent experiments, n=3.  3.1.4 Defective hematopoietic stem/ progenitor cell numbers and localization in mammary tumour bearing mice The changes in hematopoiesis observed in MT-1 bearing mice led us to examine the HSPC compartment in the BM and spleen by flow cytometry to determine whether the changes in erythropoiesis and myelopoiesis originated in the stem/progenitor cell compartments. In or- der to study HSPCs, an antibody cocktail to identify differentiated cells was used (lineage: CD3 + CD19 + CD11b + CD11c + DX5 + Gr-1 + Ter119 + MHCII +FcεRI+). LSKs (lineage-Sca-1+c-kit+) were increased in the BM of MT-1 bearing mice, whereas LKs (lineage - Sca-1 - c-kit + ) were reduced on RAG1 -/-  or MMTVneu backgrounds (Fig. 3.5, 3.6A). Further dissection of the subpopulations of BM HSPCs showed that the increase in LSKs was due to increases in short-term hematopoietic stem cells (ST-HSCs; LSK, CD34 + Flt3 - ) and multi-potent progeni- tors (MPP; LSK, CD34 + Flt3 + ), whereas long-term hematopoietic stem cells (LT-HSC; LSK, CD34 - Flt3 - ) were unchanged or reduced, depending on the strain of mice analyzed (Fig. 3.5, 3.6A). BM LKs were reduced overall due to a reduction in lineage -  cells, and this loss of LKs was mainly due to a reduction in the numbers of megakaryocyte erythrocyte progenitors (MEPs; LK, FcγRII/III-CD34lo) (Fig. 3.6A). The numbers of granulocyte macrophage pro- genitors (GMP; LK, FcγRII/IIIhiCD34+) were unchanged (RAG1-/-) or reduced (MMTVneu), while common myeloid progenitors (CMP; LK, FcγRII/IIIintCD34+) were unchanged (Fig. 3.6A). In contrast, total splenocytes and all splenic HSPC subsets were increased in MT-1 bearing mice (Fig. 3.6B), suggesting that hematopoiesis had been shifted from the BM to the spleen. These results may also indicate that the spleen was compensating for the production of RBCs through extramedullary hematopoiesis, since there was a large increase in the num- 32  bers of MEPs in the spleen of MT-1 bearing mice (Fig. 3.6B), corresponding to the increase in more differentiated erythroblast populations (Fig. 3.2C, 3.3C). A comparison of B16 mel- anoma tumour bearing mice showed that these hematopoietic abnormalities in MT-1 bearing mice may be specific to mammary tumours, as B16 tumour mice did not display similar changes in BM HSPCs (Fig. 3.7).   Figure 3.5    Perturbations in hematopoietic stem/progenitor cell frequency in mammary tumour-bearing mice. Gating scheme for classification of HSPCs. BM from control and MT-1 or B16 tumour-bearing RAG1 -/-  mice was stained with antibodies recognizing lineage-specific antigens (lineage - ), as well as Sca-1, c-kit, CD34, FcγRII/III, and Flt3, and analyzed by flow cytometry. Lineage-DAPI- cells were analyzed. 33    Figure 3.6    Perturbations in hematopoietic stem/progenitor cell numbers and location in mammary tu- mour-bearing mice. (A) BM total cell numbers and numbers of the HSPC subsets; LSK (lineage - Sca-1 + c-kit + ), LK (lineage - Sca-1 - c- kit + ), LT-HSC (LSK, Flt3 - CD34 - ), ST-HSC (LSK, Flt3 + CD34 - ), MPP (LSK, Flt3 + CD34 + ), CMP (LK, FcγRII/IIIintCD34+), GMP (LK, FcγRII/IIIhiCD34+), and MEP (LK, FcγRII/IIIloCD34-). (B) Splenic total cell numbers and stem/progenitors were identified and enumerated as in A.  Data are representative of a minimum of three independent experiments, n=3. 34   Figure 3.7    Melanoma tumour development does not disrupt the BM hematopoietic stem/progenitor compartment. Total BM cellularity and absolute numbers of the following stem/progenitor cell subsets in B16 tumour-bearing mice: LSK (lineage - Sca-1 + c-kit + ), LK (lineage - Sca-1 - c-kit + ), LT-HSC (LSK, Flt3 - CD34 - ), ST-HSC (LSK, Flt3 - CD34 + ), MPP (LSK, Flt3 + CD34 +), CMP (LK, FcγRII/IIIintCD34+), GMP (LK, FcγRII/IIIhiCD34+), and MEP (LK, FcγRII/IIIloCD34-). Data are representative of two independent experiments, n=3.  3.1.5 Defects in bone marrow HSPC numbers and localization are not due to metas- tasis of MT-1 cells The changes in HSPCs, myeloid and RBC numbers in the BM of MT-1 bearing mice sug- gested the possibility that metastasis of MT-1 cells to the BM had occurred, and MT-1 cells were directly interacting with the BM through the establishment of a metastatic niche. How- ever, due to the lack of observed tumour cells in the H&E BM sections of MT-1 bearing mice, and the absence of the neu transgene (uniquely expressed in the MT-1 cell line) upon PCR amplification of BM cells extracted from MT-1 bearing mice, we concluded that the observed abnormalities in the BM of tumour-bearing mice were not due to metastasis of tu- mour cells to the BM, but to tumour-secreted factors acting on sites distant to the tumour (Fig. 3.2A, data not shown).  3.1.6 Functional defects in the hematopoietic stem/ progenitor cell compartment in mammary tumour bearing mice In order to determine whether mammary tumours impacted HSPC function in tumour bearing mice, multiple assays were employed to analyze the colony forming potential and reconstitu- 35  tion capabilities of the HSPCs. Splenocytes or BM cells from MT-1 bearing mice were ana- lyzed by the methylcellulose colony forming assay to assess myeloid progenitor colony form- ing units (CFU-G, granulocytic colony; CFU-M, macrophage colony; CFU-GM, granulo- cyte-macrophage colony). Myeloid progenitors were largely unchanged in the BM of MT-1 bearing mice, with the exception of a <2 fold increase in CFU-M (Fig. 3.8A). Conversely, splenic CFU’s were significantly elevated in MT-1 bearing mice (Fig. 3.8A). In contrast, the CFU-S12 assay, which identifies primitive progenitors able to form colonies on the spleens of irradiated recipients, showed that MT-1 bearing mice had a reduction in primitive progenitors in the BM correlating with the reduction in LKs (Fig. 3.6A, 3.8B), but an increase of these progenitors in the spleen correlating with the overall increase in LKs (Fig. 3.6B, 3.8B). Le- thal reconstitution assays were performed using 1:1 ratios of CD45.2 +  BM or splenocytes from control or MT-1 bearing mice to CD45.1 +  control BM or splenocytes, respectively. Our results revealed no differences in the reconstitution capability of BM harvested from MT-1 bearing mice compared to BM from control mice at 3, 8, and 20 weeks post-transfer (Fig. 3.8C, data not shown). As expected, splenocytes from control mice were unable to reconsti- tute the hematopoietic system long term (Fig. 3.8C). In contrast, splenocytes from MT-1 bearing mice were much more capable of lethal reconstitution at 3, 8 and 20 weeks, correlat- ing with the increased LT-HSC numbers in the spleen of MT-1 bearing mice (Fig. 3.6B, 3.8C, data not shown). 36   Figure 3.8    Mammary tumour growth is associated with significant functional changes in the stem/progenitor compartments in BM and spleen. (A) Total numbers of colony-forming cells (CFU-G/M or CFU-GM) in spleen (left) and BM (right) of control and MT-1 bearing MMTVneu mice were determined by methylcellulose assay. (B) Numbers of primitive multi- potent colony forming cells in the BM and spleen of control and MT-1 bearing RAG1 -/- mice were determined by CFU-S12 assay. (C) CD45.2 +  splenocytes or BM cells from MMTVneu control or MT-1 bearing mice were transplanted into lethally-irradiated CD45.1 +  recipients at a 1:1 ratio together with CD45.1 +  carrier cells. En- graftment efficiency was determined by assessing the relative frequency of CD45.2 +  versus CD45.1 +  cells in the blood at 3 and 8 weeks by flow cytometry. Percentages of CD45.2 +  nucleated PI -  cells are shown. Data repre- sent a minimum of two independent experiments, n=3-5.  3.1.7 In vitro differentiation of hematopoietic stem/ progenitor cells is impaired by mammary tumour secreted factors In order to investigate the mechanism behind MT-1 induced hematopoietic abnormalities, an in vitro BM culture system using hematopoietic cytokines was developed to determine the effects of MT-1 conditioned media (MT-1-CM) on progenitor content. Culture of BM with FLT3L or GM-CSF, but not other cytokines, in combination with MT-1-CM after 9 days ex- panded total cellularity (Fig. 3.9A, 3.10A). LSKs were also increased in FLT3L cultures at day 9, while the increase in LSKs in GM-CSF cultures peaked at day 3 (Fig. 3.9A, 3.10A, 37  3.11). An assessment of an assortment of tumour conditioned media (T-CM) showed that 5 different MT cell lines increased total cellularity and LSK content when cultured with BM supplemented with FLT3L, while B16 melanoma and CMT93 colon cancer T-CM were without effect (Fig. 3.9B).  To investigate possible cytokine constituents in MT-1-CM that were able to synergize with FLT3L to increase progenitor content, BM supplemented with FLT3L was treated with dif- ferent cytokines known to be important in tumour physiology, including IL-3, IL-6, TGF-β, VEGF, IL-10, M-CSF, G-CSF, GM-CSF, TPO and TSLP (DeNardo and Coussens, 2007; Korkaya et al., 2011). Out of the cytokines tested, only IL-6 and G-CSF treatment of BM cul- tures supplemented with FLT3L increased total cellularity and LSKs to the extent observed with MT-1-CM (Fig. 3.9C). A cytokine array performed on MT-1-CM revealed an abun- dance of G-CSF, but not IL-6, indicating that G-CSF (and possibly other chemo- kines/cytokines such as CXCL1, M-CSF, MIP-2, MIG, TNFα and CXCL10) may be respon- sible for or contribute to the increase in progenitors in vitro (Fig. 3.12).  The myeloid progenitor content of MT-1-CM treated FLT3L BM cultures was assessed to determine their functional activity. Methylcellulose assays of day 9 FLT3L BM cultures treated with MT-1-CM revealed a large enrichment/expansion of CFU-G/M/GM compared to untreated FLT3L BM control cultures, or an equivalent number of freshly harvested BM cells (Fig. 3.9D). Methylcellulose assays of day 3 GM-CSF BM cultures treated with MT-1-CM showed a similar, though less dramatic, enrichment of CFU’s (Fig. 3.9D). An assessment of progenitor content by flow cytometry revealed that MT-1-CM treatment of FLT3L BM cul- tures showed an increase in total cells and numbers of LSKs, LKs, CMPs and GMPs that was evident by day 3 (Fig. 3.10A). There was also an increase in myeloid cells, as indicated by the increase in CD11b + CD115 hi  cells and a transient (day 3) increase in CD11b + Gr-1 hi  cells (Fig. 3.10A). GM-CSF BM cultures supplemented with MT-1-CM showed a similar trend, although most progenitors including LSKs, LT-HSCs, ST-HSCs, and GMPs peaked at day 3, while CD11b + CD115 hi  and CD11b + Gr-1 hi  cells did not change compared to control GM-CSF BM cultures (Fig. 3.11). 38   Figure 3.9    Mammary tumour conditioned media expands hematopoietic progenitors in in vitro BM cul- tures supplemented with FLT3L or GM-CSF. (A) Total cells (left) and LSKs (right) in BM cultured in media containing the indicated cytokines, at a plated density of 1×10 6 cells/mL, supplemented ± MT-1-CM (designated as MT-1). (B) BM was cultured in the pres- ence of FLT3L supplemented with conditioned media from the indicated tumour cell lines, and total cellularity and LSK numbers were determined. (C) As in A, except that BM was cultured with FLT3L in combination with the indicated cytokines or MT-1-CM. (D) Total CFCs in FLT3L or GM-CSF BM cultures treated ± MT-1-CM. BM cells were plated in methylcellulose after 8 days (FLT3L) or 3 days (GM-CSF) of culture, and colonies were scored 12 days later. Freshly isolated BM was also plated to compare relative CFCs in the cultures and equivalent numbers of plated uncultured BM. Data are representative of a minimum of three independent exper- iments. 39   Figure 3.10    Mammary tumour conditioned media expands hematopoietic progenitors in in vitro BM cultures supplemented with FLT3L, while G-CSF neutralization suppresses mammary tumour condi- tioned media expansion of hematopoietic progenitors. (A) Total cells and absolute numbers of HSPCs (LSK, LK, CMP, GMP) in FLT3L BM cultures treated with MT-1-CM over a period of 9 days. Numbers of CD11b + CD115 hi  and CD11b + Gr-1 hi  cells in the cultures are also shown. (B) Total cells and numbers of LSKs and LKs in FLT3L BM cultures supplemented with a reduced dose (1.6%) of MT-1-CM, treated with G-CSF or IL-6 neutralizing Abs (or isotype control Abs, iso). (C) Numbers of CFCs were enumerated in control and G-CSF or IL-6 neutralizing Ab treated FLT3L BM cultures treated ± 1.6% MT-1-CM. BM cells were plated in methylcellulose after 8 days of culture, and colonies were scored 12 days later. Data are representative of a minimum of three independent experiments. 40   Figure 3.11    Mammary tumour conditioned media expands hematopoietic progenitors in BM cultures supplemented with GM-CSF. Total cells and absolute numbers of HPSCs and myeloid cells in GM-CSF BM cultures treated with MT-1-CM over a period of 9 days. Data are representative of two independent experiments. 41   Figure 3.12    Identification of cytokines produced by MT-1 cells. A mouse cytokine array containing a panel of 40 different cytokines/chemokines was used to detect cyto- kine/chemokine expression by MT-1 cells. Top panel, control medium; bottom panel, MT-1-CM.  For this ex- periment, all medium contained 0.5% serum.  Duplicate spots at the upper and lower left, and upper right, are positive controls. Factors showing differential expression are boxed. The position of IL-6 on the array is also shown although no expression was detected.  Secretion of G-CSF by tumour cells has been associated with the expansion of leukocytes in tumour bearing hosts, while G-CSF administration alone or with FLT3L has been shown to mobilize HSPCs from the BM as well as impair erythropoiesis (Brasel et al., 1997; de Haan et al., 1994; de Haan et al., 1992; Gabrilovich, 2004; Nijhof et al., 1994). Therefore, to un- derstand the contribution of G-CSF in the ability of MT-1-CM to increase progenitor content, antibody neutralization of G-CSF was carried out. Neutralizing antibodies against G-CSF and IL-6 were added to in vitro FLT3L BM cultures supplemented with MT-1-CM. Neutralizing G-CSF, but not IL6, reduced culture cellularity, LKs and LSKs in in vitro FLT3L BM cul- tures (Fig. 3.10B). Assessment of progenitor content in FLT3L BM cultures supplemented with MT-1-CM by the methylcellulose assay revealed that total CFUs were also reduced up- 42  on G-CSF neutralization, showing that MT-1-CM derived G-CSF is necessary for the in- crease in total cellularity and progenitor content in FLT3L BM cultures (Fig. 3.10C).  3.1.8 Regulation of Hox gene expression and histone modifications at the Hoxa9 locus by mammary tumour secreted factors The HSPC abnormalities detected in MT-1 bearing mice, the increase in progenitor content in FLT3L and GM-CSF BM cultures treated with MT-1-CM, and the associated leukocytosis and anemia in MT-1 bearing mice, warranted further analysis into the genetic regulation of hematopoiesis and Hoxa gene expression. qPCR analysis of the mRNA expression of Hoxa locus genes in FLT3L and GM-CSF BM cultures supplemented with MT-1-CM showed an increase in Hoxa7, Hoxa9 and Hoxa10 mRNA levels (Fig. 3.13A). A corresponding reduc- tion in Ezh2 message, the catalytic component of the PRC2 epigenetic silencing complex, suggested that MT-1-CM might be regulating the expression levels of Hoxa genes through histone modifications at the Hoxa locus. mRNA expression levels of Jmjd3, which counter- acts PRC2 by demethylating H3K27me3, and Mll1, an H3K4 methyl-transferase, did not dif- fer between control or MT-CM treated cells (Fig. 3.13A, data not shown). Importantly, qPCR analysis of sorted lineage -  BM progenitors from MT-1 bearing mice revealed a similar in- crease in Hoxa9 and reduction in Ezh2, thereby providing a mechanistic link between the changes in HPSC composition and function in MT-1 bearing mice and MT-1-CM treated FLT3L/ GM-CSF BM cultures (Fig. 3.13B).   43   Figure 3.13    Mammary tumour growth is associated with enhanced Hoxa gene expression and changes in histone methylation. (A) Expression of Ezh2, Jmjd3, Hoxa7, Hoxa9 and Hoxa10 mRNA in FLT3L and GM-CSF BM cultures treated ± MT-1-CM. (B) Expression of Hoxa9 and Ezh2 mRNA in FACS sorted lineage -  BM cells from control and MT-1 bearing mice. (C) Expression of Hoxa9 and Ezh2 mRNA in FLT3L BM cultures treated with a reduced dose of MT-1-CM (1.6%) together with isotype control Abs (iso) or neutralizing Abs against G-CSF. (D) Ly- sates from GM-CSF BM cells treated with an increasing dose of MT-1-CM were analyzed by Western blot for H3K27me3, histone H3 and β-actin (loading control). Data represents a minimum of three independent experi- ments. 44   To determine how G-CSF contributes to the MT-1-CM-mediated induction of Hoxa9 expres- sion in BM cultures, G-CSF was neutralized with anti-G-CSF antibodies, and the level of Hoxa9 and Ezh2 expression was assessed through qPCR. Neutralization of G-CSF from FLT3L BM cultures supplemented with MT-1-CM showed that Ezh2 levels could be partial- ly restored, while Hoxa9 expression was reduced to control levels (Fig. 3.13C). The dysregu- lation of Ezh2 mRNA levels suggested that MT-1-CM modified the histone methylation sta- tus of FLT3L and GM-CSF BM culture in a genome-wide manner. Western blot analysis of the levels of global H3K27me3 showed that it was reduced in a dose-dependent manner when GM-CSF BM cultures were treated with an increasing dose of MT-1-CM from 6% to 25% (Fig. 3.13D).  To further understand the mechanism regulating Hoxa9 expression, we used ChIP to assess the levels of H3K4me3 and H3K27me3 associated with the Hoxa9 locus (Fig. 3.14A). MT-1- CM treatment of FLT3L or GM-CSF BM cultures showed reduced H3K27me3 (inhibitory) modifications of the Hoxa9 locus (amplicon A), and elevated H3K4me3 (activating) modifi- cations of the Hoxa9 locus (amplicon B-E) (Fig. 3.14B). These results correlate with the in- creased expression of Hoxa9 induced by MT-1-CM treatment of FLT3L and GM-CSF BM cultures, indicating that MT-1-CM may regulate Hoxa9 expression by enriching enabling his- tone modifications in the Hoxa9 locus (Fig. 3.13A, 3.14)  45    Figure 3.14    Mammary tumour conditioned media treatment of bone marrow cultures is associated with enabling changes in histone methylation at the Hoxa9 locus. (A) Depiction of the location of amplicons in the Hoxa9 locus used for ChIP analysis. (B) Repressive (H3K27me3-amplicon A) and activating (H3K4me3-amplicons B-E) histone modifications were investigated by ChIP of chromatin isolated from FLT3L or GM-CSF BM cultured ± MT-1-CM. Chromatin was immunopre- cipitated using anti-H3K27me3, anti-H3K4me3 or IgG control antibodies and amplicons A-E were analyzed by qPCR. Data represents a minimum of three independent experiments. 46   3.1.9 The contribution of G-CSF in mammary tumour secreted factor suppression of hematopoietic stem/ progenitor cell differentiation Neutralization of G-CSF in MT-1-CM reduced progenitor content and restored Hoxa9 ex- pression to levels comparable to control BM cells. Experiments were performed to better un- derstand how G-CSF contributes to MT-1-CM mediated dysregulation of progenitor expan- sion, and whether G-CSF alone could induce the expression of Hoxa9, and increase progeni- tor content. FLT3L and GM-CSF BM cultures were supplemented with G-CSF and analyzed by qPCR for Hoxa9 expression and methylcellulose assay for the expansion of myeloid pro- genitors. Addition of G-CSF to FLT3L and GM-CSF BM cultures upregulated Hoxa9, and increased total CFCs to levels similar to MT-1-CM in a dose-dependent manner similar to MT-1-CM (Fig. 3.15A,B). Therefore, G-CSF in MT-1-CM mediated dysregulation of hema- topoietic cell differentiation in vitro.  3.1.10 Role of global H3K4me3 in mammary tumour secreted factor-mediated sup- pression of HSPC defects Menin and MLL are part of the TrxG complex that catalyzes the tri-methylation of H3K4, which is an epigenetic activating mark for permissive gene expression. We employed the re- cently created Menin inhibitors to prevent the interaction between Menin and MLL, which were shown to inhibit Hoxa9 expression and prevent leukemic transformation (Grembecka et al., 2012). Menin inhibitor treatment (MI-2 and MI-3) of FLT3L BM cultures supplemented with MT-1-CM reduced total cellularity, LSKs and LKs compared to cells treated with the control drug, MI-nc (Fig. 3.15C-E). Assessment of myeloid progenitors by enumerating CFCs through the methylcellulose assay indicated a corresponding reduction in progenitor content induced by MI-2 and MI-3 in day 3 GM-CSF BM cultures supplemented with MT-1- CM (Fig. 3.15F). These results link the expansion of progenitor content by MT-1-CM with the increase in H3K4me3 activating marks in the Hoxa9 locus. Additionally, reduced LSKs, LKs and CFCs in the methylcellulose assay by Menin inhibitors in MT-1-CM treated FLT3L and GM-CSF BM cultures resembled neutralization of G-CSF in MT-1-CM. Taken together, these results link G-CSF with the enrichment of H3K4me3 activating marks in BM progeni- tor cells, culminating in enhanced Hoxa9 expression and increased progenitor content. 47   Figure 3.15    G-CSF enhances Hoxa9 expression and CFC numbers, while Menin inhibitors reverse mammary tumour conditioned media induced expansion of LSKs, LKs and CFCs in in vitro BM cultures. (A-B) Expression of Hoxa9 mRNA in FLT3L and GM-CSF BM cultures treated with MT-1-CM or G-CSF (1, 5 or 20ng/mL), and total numbers of CFCs from FLT3L and GM-CSF BM cultures treated with MT-1-CM or 20ng/mL G-CSF. (C-E) Total cells and absolute numbers of LSKs and LKs in BM cultures supplemented with FLT3L and MT-1-CM, and treated with Menin inhibitors MI-2 or MI-3 or control drug (MI-nc) (3.1µM, 6.3µM, or 12.5µM). (F) Total CFCs in BM cultures supplemented with GM-CSF and MT-1-CM and treated with 3.1µM or 6.3µM of control drug (MI-nc) or Menin inhibitors (MI-2 or MI-3). Data represent a minimum of three independent experiments. 48   3.2 Mammary cancer induces changes in DC development, function and phenotype  3.2.1 Suppression of DC development and maturation by mammary tumours Tumour-bearing hosts have skewed numbers and subsets of DCs. We set out to determine whether these defects were associated with aberrant DC development. Using lineage specific markers that excluded differentiated cell types (lineage: CD3 + CD19 + CD11b +  DX5 + Gr- 1 + Ter119 + MHCII +IL7Rα+), we analyzed the BM and splenic compartments of MT-1 bearing mice for MDPs (lineage - CD11c - c-kit hi CD115 + Flt3 + ), CDPs (lineage - CD11c - c- kit int CD115 + Flt3 + ) and preDCs (lineage - CD11c + c-kit lo CD115 + Flt3 + ) by flow cytometry (Fig. 1.3, 3.16). This was performed with control and tumour-bearing immunodeficient (RAG1 -/- ) and immunosufficient (MMTVneu) background mice. We found changes in DC progenitor numbers in MT-1 bearing mice, with a marked reduction in preDCs in the BM, and an in- crease of MDP, CDP and preDC progenitor subsets in the spleen (Fig. 3.17). This phenome- non was observed in both RAG1 -/-  and MMTVneu genetic backgrounds. A significant reduc- tion in BM MDPs and CDPs was observed on the RAG1 -/-  background only, which suggested that the absence of lymphocytes contributed to the tumour-induced detrimental effects on DC development in MT-1 bearing mice (Fig. 3.17). Our data shows that the MT-1 tumour can impair the DC developmental arm of hematopoiesis in the BM, while increasing total DC progenitors in the spleen.   Figure 3.16    DC progenitor gating scheme. Gating scheme for classification of DC progenitors. BM from control mice were stained with antibodies recog- nizing lineage-specific antigens, MHCII, CD11c, c-kit, CD115, and Flt3, and analyzed by flow cytometry. Lin- eage - PI -  cells were analyzed. MDP, monocyte dendritic cell progenitor; CDP, common dendritic cell progenitor; preDC, DC precursor. 49    Figure 3.17    Perturbations in DC progenitor numbers and location in mammary tumour bearing mice. BM and splenic numbers of DC progenitor subsets in control or MT-1 bearing mice; MDP (lineage - CD11c - c- kit hi CD115 + Flt3 + ), CDP (lineage - CD11c - c-kit int CD115 + Flt3 + ) and preDC (lineage - CD11c + c-kit lo CD115 + Flt3 + ). Data are representative of a minimum of two independent experiments, n=3.  Since pDCs fully differentiate in the BM prior to extravasation and migration into the blood (Liu et al., 2007; Naik et al., 2007; Onai et al., 2007), we first investigated the correlation be- tween reduced BM DC progenitors and pDCs using flow cytometry. We found a reduction in the frequency and numbers of pDCs (CD11c + CD45RA + ) in the BM of MT-1 bearing mice, regardless of the site of tumour development (SQ or MFP) (Fig. 3.18A, 3.19). The reduced numbers of pDCs corresponded to the reduced numbers of CDPs from which pDCs are de- rived (Liu et al., 2009; Naik et al., 2007; Onai et al., 2007) (Fig. 1.3, 3.17). The observed en- hancement in splenic DC progenitors of MT-1 bearing mice suggested that the presence of the tumour increased peripheral DC development. Surprisingly, we found a reduction in the splenic frequencies of total CD11c + MHCII +  cDCs and pDCs in MT-1 bearing mice (Fig. 3.18B). A corresponding reduction in the absolute numbers of CD4 +  DCs (CD11c + CD4 + ) and pDCs was clear, whereas CD8 +  DCs (CD11c + CD8 + ) were reduced but did not reach signifi- cance (Fig. 3.19). A 3 and 5-fold increase in total splenic cellularity (splenomegaly) in MMTVneu and RAG1 -/-  MT-1 bearing mice offset the reduced frequencies of the CD8 +  DCs 50  (Fig. 3.1B). We also found a high percentage of CD11c + MHCII +  cells that did not express CD4 or CD8 in the tumour of MT-1 bearing mice (Fig. 3.18B). PreDCs not expressing CD4 or CD8 are precursors to CD4 +  and CD8 +  DCs, therefore, the tumour infiltrating CD11c + MHC +  cells that were CD4 - CD8 -  may represent DCs that were not fully mature (Naik et al., 2006). Alternatively, these CD11c + MHCII +  cells may represent CD4 - CD8 -  DCs which have been shown to be similar in function to CD4 + CD8 -  DCs, and thus may constitute DCs that have been skewed towards inducing a Th2 response (Proietto et al., 2004; Sathe and Wu, 2011).  51   Figure 3.18    Perturbations in DC frequency in mammary tumour bearing mice. (A) Frequencies of CD11c + CD45RA +  cells in the BM of control and MT-1 tumour bearing mice. (B) Frequen- cies of CD11c + MHCII + , CD11c + CD4 + , CD11c + CD8 + , CD11c + CD45RA +  cells in the spleen and tumour of con- trol and MT-1 bearing mice. SQ, subcutaneous; MFP, mammary fat pad. Data was acquired on the RAG1 -/-  background. Data are representative of a minimum of three independent experiments, n=3.  52    Figure 3.19    Perturbations in DC numbers in mammary tumour bearing mice. Absolute numbers of pDCs in the BM, and CD11c + MHCII + , CD11c +  CD8 + , CD11c +  CD4 +  and pDCs in the spleen of control and MT-1 bearing mice on RAG1 -/-  background. Data are representative of a minimum of three independent experiments, n=3.  Our results indicate that the presence of mammary tumours impaired DC development in the BM, which resulted in a shift of DC progenitors from the BM to the spleen, ultimately cul- minating in reduced mature DC subset frequency (pDCs, CD8 +  DCs, CD4 +  DCs) and num- bers (CD4 +  DCs, pDCs). The lower numbers of DCs in the spleen of MT-1 bearing mice compared to control mice revealed that DC development in the spleen was also compro- mised, despite the increase in splenic DC progenitors. Therefore, our evidence suggests that mammary tumours suppress the immune system by reducing the total numbers of DCs via suppression of DC progenitor differentiation.  3.2.2 The role of mammary tumour secreted factors in suppressing DC development in vitro – FLT3L To elucidate the molecular mechanism behind mammary tumour-mediated impairment of DC development, we used an in vitro method of DC culture using FLT3L, a cytokine that induc- es the differentiation of CD8 + -like, CD4 + -like and pDCs from BM cells that resemble in vivo splenic DCs (Naik et al., 2005; Naik et al., 2007; Onai et al., 2007). Our previous results showed that soluble, mammary tumour-derived factors increased in vitro HSPC progenitor content (Fig. 3.9). This led us to treat FLT3L DC cultures with T-CM from a variety of tu- mour cells to investigate the effects of tumour derived factors on DC development. We sup- plemented FLT3L DC cultures with T-CM from multiple mammary tumour cell lines includ- 53  ing MT-1-4 and 4T1, as well as T-CM from B16 melanoma cells and CMT93 colon carci- noma cells, and analyzed DC subsets using flow cytometry (Fig. 3.20A, data not shown). In- terestingly, all mammary tumour derived T-CM impaired the differentiation of pDCs and cDCs including CD8 + -like (CD45RA - CD11c + CD24 +SIRPα-) and CD4+-like (CD45RA- CD11c + CD24 - SIRPα+) DCs into a CD24intSIRPαint/hi subset of cells, whereas B16 and CMT93 T-CM had no effect (Fig. 3.20A).  Figure 3.20    Mammary tumour conditioned media impairs DC differentiation in in vitro FLT3L DC cultures. (A) Frequencies of pDCs (CD11c + CD45RA + ), cDCs (CD11c + CD45RA - ), CD8 + -like (CD11c + CD45RA - CD24 +SIRPα-), CD4+-like (CD11c+CD45RA-CD24-SIRPα+) DCs from FLT3L DC cultures treated with T-CM 54  from MT-1, 4T1, B16, and CMT93 cell lines. (B) Frequencies of pDCs, cDCs, CD8 + -like, CD4 + -like DCs from FLT3L DC cultures treated with MT-1-CM for different periods of time (day 0-9, 3-9, 6-9, 8-9). Data are repre- sentative of three independent experiments.  The suppression of DC differentiation by mammary tumour conditioned media prompted fur- ther investigation of the effects of MT-1-CM on FLT3L DCs at different stages of develop- ment. Addition of MT-1-CM from day 0-9 or day 3-9 of FLT3L DC cultures led to impaired DC differentiation, whereas addition of MT-1-CM from day 6-9 and 8-9 had no effect (Fig. 3.20B), which suggested that MT-1-CM predominantly impacted the differentiation of early DC progenitors into mature DCs. Additionally, DC activation was impaired by early treat- ment with MT-1-CM in the form of reduced baseline and lipopolysaccharide (LPS) -induced expression of antigen presenting and co-stimulatory molecules including MHCII, CD80, CD86 and CD40 in the cDC population of FLT3L DC cultures (Fig. 3.21A,B). Notably, MT- 1-CM treated FLT3L DCs were still responsive to TLR stimulation, as these cells upregulat- ed PD-L1 upon LPS stimulation (Fig. 3.21B). 55    Figure 3.21    Mammary tumour conditioned media impairs DC expression of MHCII and co-stimulatory receptors in in vitro FLT3L DC cultures. (A) Proportion of CD11c + CD45RA -  cells expressing low, intermediate and high levels of MHCII in FLT3L DC cultures treated with MT-1-CM for different periods of time (day 0-9, 3-9, 6-9, 8-9). Cells were stimulated with (+) or without (-) LPS. Numbers below graphs represent frequencies of cells in each sector. (B) Frequency of cells expressing CD80, CD86, CD40, PD-L1, and CD8 in FLT3L DC cultures treated with MT-1-CM for dif- ferent periods of time (day 0-9, 3-9, 6-9, 8-9).  Cells were stimulated with (black bars) or without (white bars) LPS. Data are representative of two experiments. 56   DC progenitors in FLT3L DC cultures treated with MT-1-CM were analyzed by flow cytom- etry to determine whether MT-1-CM impaired DC progenitor production. Total cellularity was increased in FLT3L DC cultures treated with MT-1-CM, mimicking the splenomegaly observed in MT-1 bearing mice (Fig. 3.22A). Analysis of DC progenitor subsets showed a sustained increase in MDPs and CDPs over 9 days by MT-1-CM treatment of FLT3L DCs (Fig. 3.22A). Surprisingly, total preDCs were either equal or lower in MT-1-CM treated DC cultures compared to untreated controls (Fig. 3.22A). These results showed that in vitro MT- 1-CM treated FLT3L DC cultures mimicked a similar block in preDC development in the BM of MT-1 bearing mice (Fig. 3.17). This suggests that the impairment in DC differentia- tion of FLT3L DC cultures treated with MT-1-CM was due to impaired transition from CDPs to pDCs and preDCs.  To investigate the molecular nature of impaired DC development by MT-1-CM, we meas- ured the mRNA levels of transcription factors necessary for DC development by qPCR. There was a strong inhibition of DC transcription factor expression in FLT3L DCs treated with MT-1-CM from day 0-9, 3-9 and 6-9 for Irf4 and E2-2, whereas Irf8 mRNA expression was reduced after all treatment periods (Fig. 3.22B). These results indicate that MT-1-CM suppressed the expression of transcription factors necessary for the development of the CD8 + , CD4 +  and pDC subsets, which correlates with the impaired differentiation of mature DC subsets in FLT3L cultures.  Our results show that MT-1-CM treatment of FLT3L DC cultures led to a defect in DC dif- ferentiation, impairment in DC development and accumulation of MDPs and CDPs, and a reduction in the expression of essential transcription factors required for the differentiation of the CD8 + , CD4 +  and pDC subsets. 57   Figure 3.22    Mammary tumour conditioned media enhances DC progenitor generation and impairs es- sential DC transcription factor mRNA expression in in vitro FLT3L DC cultures. (A) Absolute numbers of total cells, MDPs, CDPs, and preDCs in FLT3L DC cultures treated with MT-1-CM over a period of 9 days. (B) Expression of transcription factors Irf4, Irf8 and E2-2 in FLT3L DC cultures treated with MT-1-CM over different periods of time (day 0-9, 3-9, 6-9, 8-9). Expression was analyzed by qPCR. Data are representative of two independent experiments. 58  3.2.3 The role of mammary tumour secreted factors in suppressing DC development in vitro – GM-CSF GM-CSF is a classical cytokine used to culture in vitro DCs (Inaba et al., 1992). We tested the effects of T-CM on GM-CSF DC cultures. Similar to FLT3L DCs, flow cytometry analy- sis revealed that T-CM from mammary tumour cell lines (MT-1-4, 4T1) impaired expression of MHCII in GM-CSF DC cultures at baseline and after LPS stimulation, whereas T-CM from B16 and CMT93 cell lines had no effect (Fig. 3.23A,B). Further analysis of the effects of MT-1-CM on MHCII expression revealed that increasing the dose of MT-1-CM (6.25% to 25%) inhibited MHCII, CD80, CD86 and CD40 expression (Fig. 3.24A, data not shown). Additionally, treatment of GM-CSF DC cultures with MT-1-CM at earlier time points (day 0-9, 3-9, 6-9) led to a greater reduction in MHCII expression at baseline and post-LPS stimu- lation compared to later time points (day 8-9) (Fig. 3.24B).  59   Figure 3.23    Mammary tumour conditioned media impairs DC differentiation in in vitro GM-CSF DC cultures. (A) Frequencies of CD11c + MHCII +  DCs in GM-CSF DC cultures treated with T-CM from MT-1, CMT-93, 4T1, and B16 cell lines. Cells were stimulated ± LPS. (B) Frequencies of CD11c + MHCII +  DCs in GM-CSF DC cultures treated with T-CM from MT-1, 2, 3, 4. Cells were stimulated ± LPS. Data are representative of a mini- mum of two independent experiments.   60   Figure 3.24    Mammary tumour conditioned media impairs DC differentiation in in vitro GM-CSF DC cultures at different concentrations and treated for different periods of time. (A) Frequencies of CD11c + MHCII +  DCs in GM-CSF DC cultures treated with MT-1-CM at concentrations of 6.25%, 12.5% and 25%. Cells were stimulated ± LPS. (B) Frequencies of CD11c + MHCII +  DCs in GM-CSF DC cultures treated with MT-1-CM for different periods of time (day 0-9, 3-9, 6-9, 8-9). Cells were stimulated ± LPS. Data are representative of a minimum of two independent experiments. 61   Pathogen associated molecular pattern (PAMP) stimulation of PRRs, such as the TLRs, on DCs can induce the production of important cytokines for T cell activation and polarization, including IL-12, IL-10, IL-6 and TNFα (Steinman and Idoyaga, 2010). T-CM from MT-1 to 4 cell lines impaired LPS-induced IL-12 production in GM-CSF DC cultures compared to LPS-stimulated control cultures without T-CM, while IL-10 production was increased by T- CM from MT-1 and 2, but not MT-3 and 4, after LPS stimulation (Fig. 3.25). No changes in IL-6 and TNFα production were detected in MT-CM treated GM-CSF DCs (data not shown). The higher expression in IL-10 and lower expression in IL-12 induced by 2 of 4 MT-CM re- vealed that GM-CSF DCs were skewed towards an immunosuppressive or Th2 prone pheno- type by MT-CM.   Figure 3.25    Mammary tumour conditioned media impairs cytokine secretion in in vitro GM-CSF DC cultures. Expression of IL-12 and IL-10 in GM-CSF DC cultures treated with T-CM from MT-1, 2, 3, 4. Cytokine levels were measured by ELISA. Cells were stimulated ± LPS. Data are representative of two independent experi- ments. 62   In summary, T-CM from mammary tumours prevents DC maturation, activation and inflam- matory cytokine production, even following PAMP stimulation. This suppression of DC ac- tivation by T-CM from mammary tumours suggests that mammary tumours possess an en- hanced ability to impair host immunity compared to the other tumour types tested.  3.2.4 Mammary tumour derived G-CSF is responsible for DC developmental im- pairment Our studies on the hematopoietic changes induced by mammary tumours showed that these changes were predominantly driven by tumour-secreted G-CSF. Mammary tumour-derived G-CSF impaired hematopoietic stem cell differentiation and maintenance and increased pro- genitor content. In relation to DCs, G-CSF treatment of healthy individuals skews DC differ- entiation towards a DC2 phenotype, with subsequent induction of Th2 skewed immune re- sponses (Klangsinsirikul and Russell, 2002). However, the effects of G-CSF on DC devel- opment have not been investigated. To determine whether mammary tumour derived G-CSF blocked FLT3L DC differentiation, antibody-mediated neutralization of G-CSF was per- formed. A reduced dose of MT-1-CM (1.6%) was used to treat FLT3L DC cultures to facili- tate complete G-CSF neutralization upon addition of anti-G-CSF neutralizing antibody. G- CSF neutralization partially restored the frequency of pDCs and cDCs in FLT3L DC cultures compared to cultures treated with isotype control antibodies (Fig. 3.26A). Neutralization of G-CSF in MT-1-CM also partially restored MHCII expression on FLT3L DC cultures (Fig. 3.26A). Therefore, MT-1 derived G-CSF is responsible for some of the effects in the im- pairment of DC differentiation in FLT3L cultures, as well as the reduction in PAMP-induced DC maturation/activation. 63    Figure 3.26    Mammary tumour conditioned media derived G-CSF impairs DC development in in vitro FLT3L and GM-CSF DC cultures. (A) Frequencies of pDCs and cDCs (left) and expression levels of MHCII in cDCs (right) from FLT3L DC cul- tures treated with 1.6% MT-1-CM for 9 days. Cultures were treated with neutralizing G-CSF Abs (G-CSF) or isotype control Abs (iso). (B) Frequencies of pDCs and cDCs (left) and expression levels of MHCII in cDCs (right) from FLT3L DC cultures treated with MT-1-CM or G-CSF (1, 5, or 20 ng/mL) for 9 days. (C) Frequen- cies of CD11c + MHCII +  cells (left) and expression levels of MHCII in CD11c +  cells (right) from GM-CSF DC cultures treated with MT-1-CM or G-CSF (1, 5, 20 ng/mL) for 9 days. (D) Numbers of MDPs, CDPs, and preDCs from FLT3L DC cultures treated with MT-1-CM or G-CSF (1, 5, 20 ng/mL) for 9 days. Data are repre- sentative of two independent experiments. 64   Next, we examined whether addition of G-CSF alone was sufficient to impair DC differentia- tion. G-CSF was added to FLT3L or GM-CSF DC cultures, and cultures were analyzed for pDCs and cDCs or CD11c + MHCII +  populations, respectively, by flow cytometry. A partial reduction in pDCs and cDCs in FLT3L cultures, and CD11c + MHCII +  cells in GM-CSF cul- tures, was observed with an increasing dose of G-CSF, similar to MT-1-CM treated cells, although the reduction in pDCs in FLT3L cultures was less pronounced (Fig. 3.26B,C). Cor- respondingly, FLT3L cDCs and GM-CSF CD11c +  cells showed reduced MHCII expression after treatment with G-CSF, reflecting similar phenotypic changes observed with MT-1-CM treatment (Fig. 3.26B,C). Further analysis of FLT3L DC cultures treated with G-CSF showed that similar to MT-1-CM, the DC progenitors MDPs and CDPs accumulated, whereas preDCs from G-CSF treated samples were similar to control and MT-1-CM treated samples (Fig. 3.26D). Therefore, MT-1-CM derived G-CSF likely contributes to suppression of DC development, resulting in the accumulation of MDPs and CDPs and reduced frequencies of mature DCs.  3.2.5 MT-1 conditioned media and G-CSF mediate impairment of DC functional ac- tivity DCs are critical in activating both the adaptive and innate immune systems to respond to ne- oplasia (Dunn et al., 2002; Dunn et al., 2004; DuPage et al., 2012; Quezada et al., 2011; Schreiber et al., 2011). Due to the dramatic impairment of DC development and differentia- tion mediated by MT-1-CM, we tested the effects of MT-1-CM on DC functionality.  NK cells require priming by activated DCs for their effector functions (Kobayashi et al., 2005; Lucas et al., 2007).  To investigate whether mammary tumours impacted the ability of DCs to prime NK cells, splenic NK cells from LPS-stimulated control and MT-1 bearing mice were assessed for IFN- and GrB production by intracellular flow cytometry.  The pres- ence of MT-1 resulted in defective NK cell activity, reducing their ability to produce IFN-γ and GrB after LPS stimulation (Fig. 3.27A,B). Defective DC development and activation likely contributed to impaired NK cell activation, since DC priming of NK cells has been shown to be critical in eliciting appropriate NK cell effector functions (Lucas et al., 2007). 65    Figure 3.27    Mammary tumours impair DC activation of NK and T cells. (A) Frequencies of splenic NK1.1 + IFN-γ+ NK cells in control and MT-1 bearing mice stimulated ± LPS. Cells were gated on NK1.1 +  populations. (B) Frequencies of granzyme B +  (GrB + ) NK cells in control and MT-1 bear- ing mice stimulated ± LPS. Cells were gated on NK1.1 +  populations. (C). Frequencies of CFSE + CD11c +  cells in GM-CSF DC cultures treated ± MT-1-CM co-cultured with CFSE labelled, doxorubicin treated B16-OVA or MT-1 cells. (D) Frequencies of CFSE -  cells in GM-CSF DC cultures treated ± MT-1-CM incubated with doxo- rubicin treated B16-OVA or MT-1 cells, then co-cultured with CFSE labelled OT-I cells. (E) Expression of IFN-γ in OTI cells (left) or IL-2 in B3Z cells (right) co-cultured with GM-CSF DCs treated ± MT-1-CM, after incubation with doxorubicin treated B16-OVA or MT-1 cells. IFN-γ and IL-2 levels were measured by ELISA. Data are representative of two independent experiments. 66   The adaptive arm of the immune system is important in the rejection of neoplasia. As dis- cussed above, immature DCs are efficient at phagocytosis while mature DCs downregulate phagocytic activity. To investigate MT-1-CM’s effect on DCs’ ability to phagocytose anti- gens, GM-CSF DCs treated with or without MT-1-CM were co-cultured with doxorubicin- treated, CFSE-labelled MT-1 or B16 tumour cells. Upon phagocytosis of dying tumour cells labelled with CFSE, DCs become CFSE + . Frequencies of CFSE + CD11c +  DCs treated with MT-1-CM were increased 1.3 fold and 2.5 fold when co-cultured with dying, CFSE-labelled B16 or MT-1 cells, respectively, compared to controls that were not treated with MT-1-CM (Fig. 3.27C). Thus, MT-1-CM increases the phagocytic ability of GM-CSF DCs, consistent with the idea that DCs are immature in the presence of MT-1-CM.  To determine whether MT-1-CM impaired DCs priming of T cells, we investigated whether MT-1-CM treated GM-CSF DCs could induce T cell proliferation as measured via CFSE di- lution. Control and MT-1-CM treated GM-CSF DCs were co-cultured with doxorubicin- treated, OVA protein expressing B16 cells, MT-1 cells (which express a fusion protein of the activated rat neu oncogene, the OTI CD8 +  epitope and OTII CD4 +  epitope), or media alone. These DCs were then purified and further co-cultured with CFSE-labelled OTI cells to de- termine their ability to induce antigen-specific T cell proliferation. GM-CSF DC cultures treated with MT-1-CM induced lower CFSE dilution from OTI cells, which indicated that they were impaired in mediating OTI cell proliferation (Fig. 3.27D). Therefore, either DC cross-presentation activity, co-stimulatory molecule expression or cytokine production was suboptimal. To test whether GM-CSF DC cross-presentation was suppressed by MT-1-CM, OTI or B3Z cells, a MHC class I (H-2b)-restricted OVA-specific hybridoma, were co- cultured with control or MT-1-CM treated GM-CSF DCs pulsed with doxorubicin-treated B16-OVA or MT-1 cells. Activation of OTI and B3Z cells was measured by production of IFN-γ and IL-2, respectively. GM-CSF DCs treated with MT-1-CM were less able to induce IFN-γ and IL-2 production from OTI and B3Z cells, respectively, which showed that MT-1- CM may have prevented cross-presentation of tumour antigens by GM-CSF DCs (Fig. 3.27E).  67  CD4 +  and CD8 +  T cell activation is required for an optimal immune response against tu- mours. Presentation of antigenic peptides from whole OVA protein on MHCI and MHCII requires phagocytosis and antigen processing in DCs, whereas addition of exogenous SI- INFEKL peptide can be directly loaded onto surface MHCI without processing. When MT-1- CM treated GM-CSF DCs were pulsed with either OVA protein or SIINFEKL peptide, lower OTI cell proliferation was induced compared to control GM-CSF DCs pulsed with OVA or SIINFEKL (Fig. 3.28A). These results indicated that even without impaired antigen pro- cessing, the capacity of GM-CSF DCs to induce CD8 +  T cell proliferation was reduced by MT-1-CM (Fig. 3.28A). Similar results were obtained with OVA-pulsed, MT-1-CM treated GM-CSF DCs co-cultured with OTII cells (Fig. 3.28A). Lastly, we assessed whether the abil- ity of DCs to activate T cell cytokine production was impaired by mammary tumour secreted factors. This was done by detecting IFNγ and IL-2 production by OTI and B3Z cells, respec- tively, after co-culture with MT-1-CM treated GM-CSF DCs pulsed with OVA protein or SIINFEKL peptide. We found that regardless of the cell types (OTI or B3Z) or antigens test- ed (OVA or SIINFEKL), MT-1-CM treatment of GM-CSF DCs induced lower levels of cy- tokine production compared to non-treated GM-CSF DCs (Fig. 3.28B). This data demon- strates that DC activation of T cells was impaired by MT-1-CM treatment.   68   Figure 3.28    Mammary tumours impair antigen specific DC activation of T cells and activates DC sup- pression of T cell proliferation. (A) Frequencies of CFSE -  cells in GM-CSF DC cultures treated ± MT-1-CM incubated with OVA protein or SIINFEKL peptide, then co-cultured with CFSE labelled OTI or OTII cells. (B) Expression of IFN-γ in OT-I cells (left) or IL-2 in B3Z cells (right) co-cultured with GM-CSF DCs treated ± MT-1-CM, incubated with OVA protein or SIINFEKL peptide. IFN-γ and IL2 levels were measured by ELISA. (C) Frequencies of CFSE- cells in GM-CSF DC cultures treated ± MT-1-CM co-cultured with CFSE labelled OTII splenocytes in the presence of OVA protein. Cells were gated on Thy1.1 + CD4 +  populations. (D) Frequencies of CFSE -  cells in GM-CSF DC cultures treated ± MT-1-CM or G-CSF (20 ng/mL) co-cultured with CFSE labelled OTI or OTII splenocytes in the presence of OVA protein. Cells were gated on Thy1.1 + CD8 +  populations for OTI spleno- cytes, and Thy1.1 + CD4 +  populations for OTII splenocytes.  69   DCs cultured in MT-1-CM showed an immunosuppressive phenotype including reduced ex- pression of MHCII, co-stimulatory molecules and increased expression of PD-L1. These ob- servations led us to hypothesize that MT-1-CM treated DCs may, in addition to being poor activators of T cells, suppress antigen-specific endogenous DC activation of T cells. We test- ed this hypothesis by co-culturing CFSE-labelled OTI or OTII splenocytes with FLT3L or GM-CSF DCs treated with and without MT-1-CM and OVA protein. OTI and OTII spleno- cytes contain T cells with TCRs specific to processed OVA peptides bound to MHCI (SI- INFEKL, OVA257-264) and MHCII (OVA323-339), respectively. In addition, they also contain antigen presenting cells such as DCs, which, upon addition of OVA protein, induce T cell proliferation. CD8 +  T cells were detected by gating on CD8 + Thy1.2 +  cells, while CD4 +  T cells were detected by gating on CD4 + Thy1.2 +  cells. MT-1-CM treated FLT3L DCs showed suppression of both CD8 +  OTI and CD4 +  OTII cell proliferation at a DC:splenocyte ratio of 1:2 and 1:1, respectively, while control FLT3L DCs did not suppress T cell proliferation (Fig. 3.28C, data not shown). GM-CSF DCs treated with MT-1-CM suppressed OTII prolif- eration at a ratio of 1:8 but not OTI cell proliferation (Fig. 3.28C, data not shown). Therefore, MT-1-CM treatment of FLT3L and GM-CSF DCs resulted in suppressed OTII proliferation, whereas OTI proliferation was suppressed by MT-1-CM treated FLT3L DCs only.  To further investigate the contribution of mammary tumour derived G-CSF in the induction of DC mediated suppression of T cell proliferation, G-CSF treated FLT3L and GM-CSF DCs were compared to MT-1-CM treated DCs to determine whether they suppressed OTI and OTII cell proliferation. G-CSF treated, MT-1-CM treated, or control non-treated GM-CSF/ FLT3L DCs were co-cultured with control or OVA-treated CFSE-labelled OTI and OTII splenocytes. G-CSF treated FLT3L DCs suppressed CD8 +  OTI and CD4 +  OTII proliferation to a similar extent to MT-1-CM treated cells at a DC:splenocyte ratio of 1:2 and 1:8. Similar- ly, G-CSF treated GM-CSF DCs suppressed OTII proliferation at 1:4. OTI T cell prolifera- tion was unaffected by MT-1-CM or G-CSF treated GM-CSF DCs (Fig. 3.28D). These re- sults showed that tumour secreted G-CSF may contribute significantly to the ability of MT-1- CM to suppress DC activation of T cells.  70  Collectively, our data revealed a striking impairment in DC development and maturation by MT-1-CM that correlated with aberrant DC activation of effector cell function.  3.2.6 Upregulation of immunosuppressive enzymes and markers by MT-1 condi- tioned media treatment of FLT3L and GM-CSF DCs is accompanied by changes in his- tone modifications Tumours may induce immunosuppressive DCs that aid tumour progression by inhibiting the immune system. These DCs express immunosuppressive enzymes and markers including Arg1, found in inflammatory zone 1 (Fizz1), vascular endothelial growth factor a (VEGFa) and Ym1. To further characterize the T cell proliferation suppressive capacity of MT-1-CM treated DCs, we measured the mRNA expression of these markers by qPCR. Arg1, Fizz1, Vegfa and Ym1 were upregulated in GM-CSF DCs treated with MT-1-CM compared to un- treated controls (Fig. 3.29A). Recent discoveries of macrophage gene regulation showed that M2 markers are controlled by JMJD3, a catalytic enzyme that demethylates H3K27, resulting in permissive gene expression (Ishii et al., 2009; Satoh et al., 2010). We performed ChIP analysis of the Arg1, Ym1 and Fizz1 locus to determine the levels of H3K27me3 associated with GM-CSF DCs treated with MT-1-CM, and found a reduction in the H3K27me3 repres- sive mark associated with the Arg1, Ym1 and Fizz1 loci in accordance with their increased mRNA expression (Fig. 3.29B).   71   Figure 3.29    Mammary tumour conditioned media induces immunosuppressive protein expression in in vitro FLT3L and GM-CSF DC cultures. (A) Expression of Fizz1, Arg1, Vegfa and Ym1 mRNA in GM-CSF DC cultures treated ± MT-1-CM. (B) Re- pressive H3K27me3 histone modifications were investigated by ChIP of chromatin isolated from GM-CSF DC cultures treated ± MT-1-CM. Chromatin was immunoprecipitated using anti-H3K27me3 or IgG control Abs and amplicons were analyzed by qPCR. (C) As in A, except with FLT3L DC cultures. (D) Lysates from GM-CSF and FLT3L DC cultures treated with an increasing dose of MT-1-CM were analyzed by Western blot for Arg1 and total Stat3 (T-Stat3, loading control). Data represents a minimum of three independent experiments. 72   Surprisingly, analyzes of FLT3L DCs treated with MT-1-CM only showed an increase in Arg1 mRNA, whereas expression of Fizz1, Vegfa and Ym1 were downregulated with MT-1- CM treatment compared to control cells (Fig. 3.29C). Western blot analysis of GM-CSF and FLT3L DCs treated with MT-1-CM revealed that treatment of cells with MT-1-CM led to a dose dependent increase in Arg1 protein expression, which strongly correlated with the in- creased Arg1 mRNA expression detected by qPCR (Fig. 3.29A,C,D).  To investigate the contribution of MT-1-CM derived G-CSF to DC polarization, we assessed whether G-CSF could also induce Arg1 expression in FLT3L and GM-CSF DCs. Our results revealed that an increasing dose of G-CSF increased the levels of Arg1 mRNA in FLT3L and GM-CSF DCs, but not to the same extent as MT-1-CM (Fig. 3.30A). Neutralization of G- CSF in MT-1-CM with an anti-G-CSF antibody led to reduced expression of Arg1 compared to isotype control in FLT3L DC cultures (Fig. 3.30B). This data showed that MT-1-CM de- rived G-CSF contributed to Arg1 expression in DCs.  73   Figure 3.30    Mammary tumour conditioned media derived G-CSF contributes to Arg1 expression, while Ezh2, Hoxa9 and Arg1 are expressed at differential levels in CD11c +  and CD11c -  fractions in in vitro FLT3L and GM-CSF DC cultures. (A) Expression of Arg1 in FLT3L and GM-CSF DC cultures treated with MT-1-CM or G-CSF (1, 5, 20 ng/mL). (B) Expression of Arg1 in FLT3L DC cultures treated with 1.6% MT-1-CM and G-CSF neutralizing or isotype control Abs (iso). (C) Expression of Ezh2, Hoxa9 and Arg1 in CD11c -  and CD11c +  fractions from sorted FLT3L and GM-CSF DC cultures treated ± MT-1-CM. For all experiments, Arg1, Hoxa9 and Ezh2 expression was measured by qPCR. Data are representative of a minimum of two independent experiments. 74   BM cells cultured in GM-CSF and FLT3L give rise to DC precursors that are CD11c -  prior to differentiating into CD11c +  DCs. The large population of CD11c -  cells in FLT3L and GM- CSF cultures treated with MT-1-CM prompted an investigation of whether the CD11c -  DC precursors were a major contributor to the expression of immunosuppressive markers. CD11c +  and CD11c -  fractions from GM-CSF and FLT3L DCs treated with or without MT-1- CM were sorted and analyzed for their expression of Arg1 by qPCR (Fig. 3.30C). We found that both the CD11c +  and CD11c -  fractions from GM-CSF and FLT3L DCs treated with MT- 1-CM had higher expression of Arg1 compared to the equivalent fractions from control cells, although the CD11c +  fraction showed the highest expression of Arg1.  Hematopoietic differentiation is controlled by genes in the Hoxa cluster, while Ezh2, part of PRC2, represses gene expression by methylating H3K27. As discussed in section 1, Hoxa9 and Ezh2 control hematopoietic progenitor activity, therefore we wanted to determine wheth- er the regulation of Hoxa9 and Ezh2 by MT-1-CM was restricted to progenitors or mature DCs in FLT3L and GM-CSF DC cultures. Sorted CD11c +  and CD11c -  fractions from FLT3L and GM-CSF DC cultures treated with MT-1-CM were analyzed for Hoxa9 and Ezh2 expres- sion. We found that Ezh2 expression was reduced by MT-1-CM treatment of FLT3L and GM-CSF DC cultures in CD11c +  and CD11c -  cell fractions, while Hoxa9 expression was el- evated most significantly in the CD11c +  fraction of FLT3L and GM-CSF DC cultures but not the CD11c -  fraction (Fig. 3.30C). These results suggested that MT-1-CM mediated reduction in Ezh2 may be important for controlling gene expression in both mature DCs and DC pro- genitors, possibly through reducing repressive histone modifications in Arg1 and other loci. Conversely, Hoxa9 is upregulated in the CD11c- and CD11c +  fractions of MT-1-CM treated FLT3L DC cultures, whereas only the CD11c +  fraction of MT-1-CM treated GM-CSF DC cultures showed upregulated Hoxa9 expression. Therefore, Hoxa9 expression was correlated with MT-1-CM treated mature DCs (GM-CSF and FLT3L DC cultures), and DC progenitors (FLT3L DC cultures).   75  3.2.7 Mammary tumour conditioned media expands functional DC progenitors that can differentiate into pDCs and cDCs MDP, CDP and preDCs differentiate into pDCs, CD8 + -like and CD4 + -like DCs in the pres- ence of FLT3L in vitro (Naik et al., 2007; Onai et al., 2007). Due to the expansion of DC progenitors induced by MT-1-CM in FLT3L DC cultures, we investigated whether these DC progenitors could give rise to mature DCs. FLT3L DC cultures treated with MT-1-CM for 8 days, were rinsed with fresh media to remove MT-CM, and cultured in either control media or media containing 25% MT-1-CM (MT-1 media) for a further 7 days for a total of 15 days. We found that by removing MT-1-CM, day 8 FLT3L DC cultures treated with MT-1-CM differentiated into pDCs, cDCs (CD8 + -like and CD4 + -like DCs), whereas FLT3L DC cultures that were maintained in MT-1-CM did not differentiate to a significant degree (Fig. 3.31A, B). These results indicated that treating DC cultures with MT-1-CM increased functional DC progenitors that retained the ability to differentiate into mature DC subsets when MT-1-CM was removed, whereas cultures that were continually treated with MT-1-CM did not differen- tiate into mature DCs. Thus, MT-1-CM blocked DC differentiation and did not simply delay differentiation of DC progenitors into pDC and cDCs.  In summary, the induction of immunosuppressive genes and markers, most notably Arg1, was at least partially controlled by MT-1-CM derived G-CSF. Our studies on histone methyl- ation and Ezh2 expression in DCs treated with MT-1-CM revealed an association between epigenetic regulation and DC immunosuppressive gene expression mediated by mammary tumour secreted factors. Additionally, we showed that MT-1-CM induced immunosuppres- sive functionality in DCs. The expansion of functional DC progenitors indicates that MT-1- CM’s effect on DC development was reversible, making DC progenitors a possible target for therapeutic intervention. 76    Figure 3.31    Mammary tumour conditioned media treatment of FLT3L DC cultures produce functional DC progenitors that differentiate into pDCs and cDCs. FLT3L DC cultures were treated with MT-1-CM for 8 days then transferred to either 0% MT-1-CM (cont me- dia) or 25% MT-1-CM (MT-1 media) for 7 days (day 15). Cells were analyzed by flow cytometry. (A) Fre- quencies of pDCs and cDCs. (B) Frequencies of CD8 + -like and CD4 + -like DCs. Cells were gated on cDCs from A. Data represent three independent experiments.   77  Section 4 Concluding comments 4.1 Conclusion and discussion Perturbations in myelopoiesis are commonly observed in neoplastic diseases and may be crit- ically important in disease aetiology. Tumours harness elements of both the innate and adap- tive immune systems to facilitate their own growth and metastasis. Many cancers induce tu- mour-promoting macrophages, MDSCs and/or immature DCs through perturbations of im- mune cell development and differentiation. It is now clear that understanding these changes are an important part of developing ways to manipulate the immune system to initiate or re- invigorate anti-tumour immune responses (Bayne et al., 2012; Pylayeva-Gupta et al., 2012). In the case of breast cancer, the immune status of the tumour may predict treatment outcome, thus the overall state of the immune system may serve as both a guide for treatment options and as an indicator of disease severity/progression (DeNardo et al., 2011; West et al., 2011).  Here I've shown that mammary tumours affect the hematopoietic system leading to myeloproliferative-like disease characterized by neutrophilia, anemia, and defects in the HSPC compartment. These defects were associated with changes in global epigenetic regula- tion, and specifically in the upregulation and histone methylation status of Hoxa genes, most notably Hoxa9, a gene that critically controls HSPC differentiation. These changes in gene regulation and HSPC defects were found to be driven by tumour-secreted G-CSF.  Since tumour-induced changes in hematopoiesis were observed using four different MT- lines, we established that these defects in hematopoiesis are likely common to many mamma- ry tumours. Another important point is that these hematopoietic defects and myeloid/ anemic abnormalities occurred in both immunosufficient and immunodeficient mouse backgrounds. This observation indicates that the lymphoid compartment does not contribute significantly to the establishment of anemia, leukocytosis and HSPC abnormalities in tumour bearing mice.  We did not observe changes in the numbers of peripheral blood lymphocytes in mammary tumour bearing immuno-sufficient mice, showing that the presence of mammary tumours does not impair lymphocyte development. Studies in ovarian cancer have shown that as dis- 78  ease severity increases, the levels of lymphocytes in peripheral blood decline (Milne et al., 2012), while patients with lung, breast, and other tumour types show a reduced peripheral lymphocyte count compared to healthy individuals (Kaszubowski et al., 1980). This is in contrast to our results showing no changes in lymphocyte counts in the blood of mammary tumour bearing mice, which may be explained by the short period of tumour progression as- sociated with xenograph murine tumour models. Cancers develop over periods of years in humans, as opposed to weeks in xenograph mouse models, which may impact the degree to which lymphocyte development is affected.  In order to determine whether tumour secreted factors were the cause of the observed hema- topoietic defects, we established an in vitro system of hematopoietic cell differentiation using MT-CM to investigate the regulation of HSPCs by mammary tumours. We found that MT- CM synergized with FLT3L and GM-CSF to enhance BM proliferation and progenitor cell expansion. Importantly, the ability of MT-CM to synergize with FLT3L or GM-CSF to cause this expansion was shared with MT-CM from every MT cell line analyzed. Melanoma and colon cancer cells lacked this activity in vitro and melanoma did not significantly perturb myelopoiesis and hematopoiesis in vivo. Our data further indicates that different tumour types interact with the hematopoietic system in unique ways, and that disruption of hemato- poiesis is a common characteristic of mammary tumours.  In our assessment of HSPC function in tumour bearing mice, HSPC competitive reconstitu- tion experiments in chimeric mice showed that the long term reconstitution capacity and LT- HSC numbers of BM from mammary tumour bearing mice was intact. However, restricted primitive BM progenitors from the CFU-S12 assay were reduced by mammary tumours. Con- versely, spleens of MT-bearing mice contained large numbers of LSK/LKs that could recon- stitute the myeloid and lymphoid compartments of lethally-irradiated recipients, and con- tained abundant primitive CFU-S12 colony-forming progenitors, as well as more restricted myeloid progenitors. This data clearly shows that the extramedullary hematopoiesis induced by mammary tumours leads to the production of functional HSPCs and their progeny in the spleen. Our in vitro assessment of BM cell differentiation using cells propagated in MT-CM and FLT3L showed that these cells expressed LSK/LK markers and included high myeloid 79  CFC numbers. However, they could not reconstitute lethally or sub-lethally irradiated mice. Notably, actively cycling HSCs and HSCs isolated from inflammatory environments may be functionally defective (Passegue et al., 2005). In summary, we showed that mammary tumour secreted factors play a critical role in increasing HSPC colony forming potential and recon- stitution capacity in vivo and in vitro.  The enhanced HSPC activity in the spleen suggested that tumour development might induce proliferation and mobilization of the stem cell compartment leading to the seeding of the spleen. A very recent study of murine lung adenocarcinomas showed that tumour-produced angiotensin II (AngII) increased extramedullary hematopoiesis in the spleen and migration of HSCs from the BM to the spleen (Cortez-Retamozo et al., 2012; Cortez-Retamozo et al., 2013). These HSPCs from tumour bearing mice preferentially differentiated into tumour as- sociated neutrophils and macrophages and enhanced tumour progression (Cortez-Retamozo et al., 2012; Cortez-Retamozo et al., 2013). The increase in splenic HSCs in this lung adeno- carcinoma model reflects similar changes observed in our mammary tumour model. Of note, the BM niche of these lung adenocarcinoma mice was not changed by the presence of the tumour, whereas the presence of the mammary tumour in our tumour bearing mice increased LSKs in the BM dramatically. Our results indicate that G-CSF was responsible for mammary tumour-induced changes in HSPC functionality, whereas AngII was responsible for extrame- dullary hematopoiesis in the lung cancer model (Cortez-Retamozo et al., 2013). The differ- ences in hematopoiesis between our studies may be the result of the functional convergence (leukocytosis) of different mechanisms (G-CSF vs AngII) used by different tumours to in- duce immune suppression. If leukocytosis can be induced through different mechanisms by tumours, a focus on the development of therapies targeting the effector cells of leukocytosis (depletion of leukocytes), and not the signalling pathways that lead to leukocytosis (depletion of G-CSF or AngII), would be more broadly effective in preventing immunosuppression.  Disruption of the BM niche and loss of MEPs and erythroblasts may have contributed to the enhanced splenic hematopoiesis as a compensatory mechanism for RBC output. Through our in vitro studies, we established that tumour secreted G-CSF was involved in the changes ob- served in BM HSPC functionality including increased cellular proliferation and progenitor 80  activity. Although G-CSF injection into mice has been shown to impair erythropoiesis, it does not induce anemia as measured by RBCs in the peripheral blood. This seemingly con- tradictory phenomenon is explained by increased splenic erythropoiesis/ compensatory splen- ic output of RBCs (de Haan et al., 1994; de Haan et al., 1992; Nijhof et al., 1994). In con- trast, our results show that mammary tumour bearing mice do indeed exhibit anemia even in the presence of increased erythrocyte production in the spleen, which suggests that aside from G-CSF, other tumour secreted factors or some other unknown variable may contribute to the anemia of mammary tumour bearing mice. Future experiments focused on determining which tumour secreted factors contribute to the overall anemic phenotype in mammary tu- mour bearing mice could help elucidate the cause of anemia. These results may also lead to multiple or synergistic strategies to ameliorate anemia. Cancer patients with pre-existing anemia before therapy exhibit reduced survival following treatment for breast, lung, prostate, and other cancers, therefore anemia may be an independent prognostic indicator or an indica- tor of disease progression (Caro et al., 2001; Qiu et al., 2010a). Therapies directed towards neutralizing tumour-induced anemia may be one novel strategy by which patient survival could be increased through the simple means of neutralizing the factors responsible, includ- ing G-CSF.  Tumours can express cytokines and growth factors important for tumour growth, survival or spread, e.g. the 4T1 MT line expresses GM-CSF, G-CSF and M-CSF associated with a leu- kemoid reaction (DuPre and Hunter, 2007). Attenuation of tumour-derived G-CSF reduces 4T1 tumour growth and MDSC accumulation (Waight et al., 2011). G-CSF can also syner- gize with FLT3L to mobilize hematopoietic stem cells (Brasel et al., 1997). In addition, G- CSF is overexpressed in human breast carcinomas compared to adjacent normal tissue (in 17 of 23 patients) (Park et al., 2011) or control individuals (Chavey et al., 2007), and in breast cancer patient serum compared to controls (Park et al., 2011). There is a clear association be- tween increased G-CSF levels and breast cancer clinical grade and lymph node infiltration, further supporting a role for G-CSF in tumour progression and disease severity (Dehqanzada et al., 2007; Lawicki et al., 2007; Lawicki et al., 2009). Therefore, in addition to its possible role in causing anemia in breast cancer patients, it may also be a critical factor in myelopro- 81  liferation resulting in MDSC expansion and mobilization of stem cells into the spleen, two phenotypes we observed in our mammary tumour models.  The phenotypic similarities between MT-bearing mice and mice with myeloproliferative dis- ease or MDS (i.e. a leukemoid reaction and HSPC defects) suggested the possibility that mammary tumours modulate key hematopoietic regulatory genes. Ezh2 loss-of-function mu- tations are associated with myeloid disorders (Ernst et al., 2010; Nikoloski et al., 2010) and Hox genes are critical regulators of hematopoietic stem cell renewal, HSPC expansion (Thorsteinsdottir et al., 2002), and differentiation repression (Bach et al., 2010). HOXA9 down-regulation allows hematopoietic cell differentiation while its overexpression leads to acute myeloid leukemia (Sauvageau and Sauvageau, 2010). Our study shows that mammary tumours were associated with enhanced Hoxa9 and depressed Ezh2 expression in HSPCs. Histone methylation patterns associated with the Hoxa9 locus favoured permissive gene ex- pression, correlating with the enhanced gene expression of Hoxa9 in HSPCs exposed to mammary tumour secreted factors. These results indicate that mammary tumours may be mediating HSPC defects through enabling epigenetic modifications in the form of increased Hoxa9 locus association with H3K4me3 and reduced association with H3K27me3, ultimately resulting in leukocytosis and anemia in mammary tumour bearing mice.  To investigate whether changes in H3K4me3 were required for MT-CM activity, we em- ployed Menin inhibitors that attenuate H3K4 methylation and suppress HOXA9 expression in transformed leukemic cells (Grembecka et al., 2012). These inhibitors reduced culture cel- lularity and dramatically reduced culture progenitor cell content, linking H3K4me3 regula- tion of the Hoxa9 locus with the induction of hematopoietic stem/progenitor cell expansion by tumours and G-CSF. Our results demonstrate that as the development of cancer drugs, such as the Menin inhibitors, that modify the cellular epigenetic landscape advances, they may also have applications in regulating non-neoplastic cells. Thus, these drugs may be used to indirectly suppress tumour progression and survival by regulating tumour-promoting cells of the hematopoietic system such as MDSCs (Grembecka et al., 2012).  82  Abnormalities in DC function often occur alongside aberrant hematopoietic differentiation (Gabrilovich, 2004; Gabrilovich et al., 2012). It has been hypothesized that DC differentia- tion and development may be impaired by tumours to induce a tolerogenic and/or Th2 prone response that mediates immunosuppression and facilitate tumour growth. Our research shows that DCs that develop in the presence of mammary tumours and their factors acquire tolero- genic features. These DCs are immature and highly phagocytic, express low levels of co- stimulatory molecules and Th1 cytokines, express high levels of immunosuppressive en- zymes, induce poor T cell proliferation, produce immunosuppressive cytokines and suppress endogenous DC activation of T cells. As the field of immunotherapy struggles to overcome the immunosuppressive effects of the tumour bearing host’s immune system, understanding the changes induced by tumours and the mechanisms underlying these changes will lead to the development of better strategies to reverse these changes (Gabrilovich et al., 2012).  Our research shows that the generation of tolerogenic DCs in mammary tumour bearing mice was associated with impaired DC differentiation, as measured by the reduced expression of crucial DC transcription factors including Irf4, Irf8, and E2-2. Analysis of DC progenitors in mammary tumour bearing mice showed that DC development was impaired in the BM as re- vealed by the reduced numbers DC progenitors, and a corresponding accumulation of DC progenitors in the spleen. Surprisingly, the accumulation of these splenic DC progenitors in tumour mice did not lead to an increase in the total numbers of splenic pDCs or cDCs, and in fact led to reductions in CD4 +  DCs and pDCs. Using an in vitro system of DC development, we showed that FLT3L BM cultures treated with MT-CM accumulated MDPs and CDPs, but not preDCs, indicating a possible block in DC differentiation between the CDP and preDC stage. This was corroborated by the lack of pDCs that developed from this system, as pDCs, like preDCs, arise from the CDP (Fig. 1.3). This is the first evidence showing that DC devel- opment is impaired by mammary tumours in a tumour bearing host at the level of DC-lineage restricted progenitors. Past research on the effects of tumour growth on preDCs and DC dif- ferentiation showed no differences in preDC differentiation into cDCs, nor cDC functions in terms of T cell activation, between the spleen and tumour (Diao et al., 2010). As previously discussed, due to the small size of the tumours, the comparison between preDCs and cDCs derived from the tumour and spleen of the same tumour mice without comparing control 83  mice, the limited analysis of DC progenitors (preDCs only), and evidence from other studies showing functional defects in cDCs from tumour mice that were not observed in this study, indicated a need for more comprehensive analyzes of DC development in tumour mice. Our studies show that tumour growth had a much more profound effect on DC development, which correlated with impairment in DC activation of cellular immunity.  In our search for the factors responsible for mammary tumour-mediated DC developmental impairment, we found that, much like the impairment in hematopoiesis by mammary tu- mours, G-CSF played an important role in the suppression of DC development. Neutraliza- tion of G-CSF in MT-CM restored normal DC development and partially restored MHCII expression, whereas the addition of exogenous G-CSF alone to FLT3L DC cultures resulted in the partial suppression of DC development and differentiation. G-CSF has been shown to induce a Th2 immune response upon injection into mice or humans, which may lead to toler- ance against foreign antigens (Arpinati et al., 2000). Injection of G-CSF into healthy human volunteers increased peripheral blood DCs that preferentially induced Th2 responses from allogeneic T cells, which predominantly produced IL-4 and IL-10 instead of IFN-γ (Arpinati et al., 2000). Additionally, peripheral blood stem cells collected from G-CSF treated patients showed a higher Th2 response compared to BM aspirates from untreated patients (Klangsinsirikul and Russell, 2002). Much like the effects of MT-CM, these studies indicate that G-CSF steers DCs towards a DC2-induced Th2 response. To our knowledge, this is the first study showing that G-CSF has a direct role in suppressing DC progenitor differentiation.  MT-CM treatment of FLT3L and GM-CSF DCs induced the expression of Arg1, an enzyme known to suppress T cell proliferation and induce tolerance. Our data revealed that the ex- pression of Arg1 in MT-CM treated DCs was associated with enabling histone modifications in the Arg1 locus, specifically in the reduced association of the Arg1 locus with H3K27me3. Other immunosuppressive factors, including Fizz1, Ym1 and Vegfa, were only induced by MT-CM in GM-CSF, but not FLT3L, DC cultures, suggesting that either GM-CSF DCs were more sensitive to MT-CM induction of an immunosuppressive phenotype, or, alternatively, other suppressive factors with the exception of Arg1 were dispensable for the induction of T cell tolerance by MT-1-CM treated DCs. 84   No study to date has directly shown a relationship between tumour induction of DC devel- opmental impairment and the generation of DCs that are tolerogenic or tumour promoting. However, a study on an ovarian cancer model showed that tumour infiltrating DCs from ad- vanced tumours had high Arginase activity, low MHCII and CD40 expression, and high PD- L1 expression compared to DCs from early tumours, observations that we also noted in the DCs in our mammary tumour model (Scarlett et al., 2012). It was proposed that as the tu- mours advanced, the subset of DCs that infiltrated the tumours changed from an im- munostimulatory type to one that promoted tumour progression. This was concluded based on the delay of tumour growth upon depletion of DCs in mice bearing late stage ovarian can- cer (Scarlett et al., 2012). The switch from cancer suppressing to cancer promoting DCs, as the tumour progressed, suggested that DC generation in ovarian cancer mice was being modi- fied by the tumours. This supports our hypothesis that tumours induce the generation of DCs that favour tumour growth through tumour induced modifications of the DC developmental pathway. The depletion of factors that are critical to DC development, and hence the source of DCs, including FLT3L and GM-CSF, could be a powerful supplementary method of en- hancing tumour immunotherapy. Future experiments comparing the differentiation of DC progenitors in mice bearing small and large mammary tumours could help clarify whether the size of the tumour influences DC differentiation.  4.2 Summary of implications and applications We have established that tumour induced defects in hematopoiesis are likely common to many mammary tumours. Thus, understanding the tumour-immune system interactions be- tween different tumour types may be an important step in designing or improving therapeutic regimens. Additionally, tumour mediated hematopoietic disruption may result from a variety of factors, such as G-CSF and AngII. This may have implications on drug and therapy de- sign, arguing for a focus on therapies directed at disrupting the end stage effector cell in tu- mour induced immunosuppression, such as the MDSC.  In our studies on hematopoietic disruption in mammary tumour bearing mice, we noted that the presence of the tumour led to an impairment in erythropoiesis. The association between 85  anemia, tumour progression and patient survival are now clear, thus, therapies directed to- wards preventing anemia may not only improve the patient quality of life, but also increase patient survival. The importance of tumour-derived G-CSF in hematopoietic disruption in mammary tumour mice indicates that one route of preventing anemia may be to reduce circu- lating G-CSF. Additional studies on mouse survival using a modified MT-1 cell line with knockdown of G-CSF may clarify the survival benefits of G-CSF neutralization.  Mammary tumour growth had a profound effect on DC development, which correlated with impairment in DC activation of cellular immunity. Considering the importance of DCs in promoting tumour progression, understanding and manipulating DC development to enforce anti-tumour immunity could be a crucial strategy to improve tumour immunotherapy. As the size of tumours increase, DCs tend to become more skewed towards tolerance and pro- tumour (Scarlett et al., 2012). Therefore, one strategy may be to prevent the production of host derived DCs altogether by depriving factors necessary for DC development, such as FLT3L and GM-CSF. Combining this strategy with the depletion of factors that skew DC development towards tolerance, such as G-CSF and AngII, may turn the tide against tumour- induced immunosuppression, thus promoting tumour eradication. 86  References Agger, K., Cloos, P.A., Christensen, J., Pasini, D., Rose, S., Rappsilber, J., Issaeva, I., Canaani, E., Salcini, A.E., and Helin, K. (2007). UTX and JMJD3 are histone H3K27 demethylases involved in HOX gene regulation and development. Nature 449, 731-734. Alexander, W. (2000). Knockout Mouse Models of Cytokine Action in Hematopoietic Stem Cell Regulation. In Madam Curie Bioscience Database (Austin, Landes Bioscience). Aliberti, J., Schulz, O., Pennington, D.J., Tsujimura, H., Reis e Sousa, C., Ozato, K., and Sher, A. (2003). Essential role for ICSBP in the in vivo development of murine CD8alpha + dendritic cells. Blood 101, 305-310. 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. Clin Cancer Res 6, 1755-1766. Argiropoulos, B., and Humphries, R.K. (2007). Hox genes in hematopoiesis and leukemogenesis. Oncogene 26, 6766-6776. Arpinati, M., Green, C.L., Heimfeld, S., Heuser, J.E., and Anasetti, C. (2000). Granulocyte- colony stimulating factor mobilizes T helper 2-inducing dendritic cells. Blood 95, 2484- 2490. Bach, C., Buhl, S., Mueller, D., Garcia-Cuellar, M.P., Maethner, E., and Slany, R.K. (2010). Leukemogenic transformation by HOXA cluster genes. Blood 115, 2910-2918. Banchereau, J., Briere, F., Caux, C., Davoust, J., Lebecque, S., Liu, Y.J., Pulendran, B., and Palucka, K. (2000). Immunobiology of dendritic cells. Annu Rev Immunol 18, 767-811. Basu, S., Hodgson, G., Zhang, H.H., Katz, M., Quilici, C., and Dunn, A.R. (2000). "Emergency" granulopoiesis in G-CSF-deficient mice in response to Candida albicans infection. Blood 95, 3725-3733. Bayne, L.J., Beatty, G.L., Jhala, N., Clark, C.E., Rhim, A.D., Stanger, B.Z., and Vonderheide, R.H. (2012). Tumor-derived granulocyte-macrophage colony-stimulating factor regulates myeloid inflammation and T cell immunity in pancreatic cancer. Cancer cell 21, 822-835. Beekman, R., and Touw, I.P. (2010). G-CSF and its receptor in myeloid malignancy. Blood 115, 5131-5136. Bell, D., Chomarat, P., Broyles, D., Netto, G., Harb, G.M., Lebecque, S., Valladeau, J., Davoust, J., Palucka, K.A., and Banchereau, J. (1999). In breast carcinoma tissue, immature dendritic cells reside within the tumor, whereas mature dendritic cells are located in peritumoral areas. J Exp Med 190, 1417-1426. Brasel, K., McKenna, H.J., Charrier, K., Morrissey, P.J., Williams, D.E., and Lyman, S.D. (1997). Flt3 ligand synergizes with granulocyte-macrophage colony-stimulating factor or granulocyte colony-stimulating factor to mobilize hematopoietic progenitor cells into the peripheral blood of mice. Blood 90, 3781-3788. Cao, R., Wang, L., Wang, H., Xia, L., Erdjument-Bromage, H., Tempst, P., Jones, R.S., and Zhang, Y. (2002). Role of histone H3 lysine 27 methylation in Polycomb-group silencing. Science 298, 1039-1043. Cao, R., and Zhang, Y. (2004). SUZ12 is required for both the histone methyltransferase activity and the silencing function of the EED-EZH2 complex. Mol Cell 15, 57-67. 87  Caro, J.J., Salas, M., Ward, A., and Goss, G. (2001). Anemia as an independent prognostic factor for survival in patients with cancer: a systemic, quantitative review. Cancer 91, 2214- 2221. Cedar, H., and Bergman, Y. (2011). Epigenetics of haematopoietic cell development. Nature reviews. Immunology 11, 478-488. Chaux, P., Favre, N., Martin, M., and Martin, F. (1997). Tumor-infiltrating dendritic cells are defective in their antigen-presenting function and inducible B7 expression in rats. International journal of cancer. Journal international du cancer 72, 619-624. Chavey, C., Bibeau, F., Gourgou-Bourgade, S., Burlinchon, S., Boissiere, F., Laune, D., Roques, S., and Lazennec, G. (2007). Oestrogen receptor negative breast cancers exhibit high cytokine content. Breast cancer research : BCR 9, R15. Chen, Y.X., Yan, J., Keeshan, K., Tubbs, A.T., Wang, H., Silva, A., Brown, E.J., Hess, J.L., Pear, W.S., and Hua, X. (2006). The tumor suppressor menin regulates hematopoiesis and myeloid transformation by influencing Hox gene expression. Proceedings of the National Academy of Sciences of the United States of America 103, 1018-1023. Cisse, B., Caton, M.L., Lehner, M., Maeda, T., Scheu, S., Locksley, R., Holmberg, D., Zweier, C., den Hollander, N.S., Kant, S.G., et al. (2008). Transcription factor E2-2 is an essential and specific regulator of plasmacytoid dendritic cell development. Cell 135, 37-48. Citron, M.L., Berry, D.A., Cirrincione, C., Hudis, C., Winer, E.P., Gradishar, W.J., Davidson, N.E., Martino, S., Livingston, R., Ingle, J.N., et al. (2003). Randomized trial of dose-dense versus conventionally scheduled and sequential versus concurrent combination chemotherapy as postoperative adjuvant treatment of node-positive primary breast cancer: first report of Intergroup Trial C9741/Cancer and Leukemia Group B Trial 9741. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 21, 1431- 1439. Cole, M., and Strair, R. (2010). Acute myelogenous leukemia and myelodysplasia secondary to breast cancer treatment: case studies and literature review. Am J Med Sci 339, 36-40. Coquerelle, C., and Moser, M. (2010). DC subsets in positive and negative regulation of immunity. Immunological reviews 234, 317-334. Cortez-Retamozo, V., Etzrodt, M., Newton, A., Rauch, P.J., Chudnovskiy, A., Berger, C., Ryan, R.J., Iwamoto, Y., Marinelli, B., Gorbatov, R., et al. (2012). Origins of tumor- associated macrophages and neutrophils. Proceedings of the National Academy of Sciences of the United States of America 109, 2491-2496. Cortez-Retamozo, V., Etzrodt, M., Newton, A., Ryan, R., Pucci, F., Sio, S.W., Kuswanto, W., Rauch, P.J., Chudnovskiy, A., Iwamoto, Y., et al. (2013). Angiotensin II Drives the Production of Tumor-Promoting Macrophages. Immunity. de Haan, G., Engel, C., Dontje, B., Nijhof, W., and Loeffler, M. (1994). Mutual inhibition of murine erythropoiesis and granulopoiesis during combined erythropoietin, granulocyte colony-stimulating factor, and stem cell factor administration: in vivo interactions and dose- response surfaces. Blood 84, 4157-4163. de Haan, G., Loeffler, M., and Nijhof, W. (1992). Long-term recombinant human granulocyte colony-stimulating factor (rhG-CSF) treatment severely depresses murine marrow erythropoiesis without causing an anemia. Experimental hematology 20, 600-604. Dehqanzada, Z.A., Storrer, C.E., Hueman, M.T., Foley, R.J., Harris, K.A., Jama, Y.H., Shriver, C.D., Ponniah, S., and Peoples, G.E. (2007). Assessing serum cytokine profiles in 88  breast cancer patients receiving a HER2/neu vaccine using Luminex technology. Oncol Rep 17, 687-694. Della Bella, S., Gennaro, M., Vaccari, M., Ferraris, C., Nicola, S., Riva, A., Clerici, M., Greco, M., and Villa, M.L. (2003). Altered maturation of peripheral blood dendritic cells in patients with breast cancer. Br J Cancer 89, 1463-1472. DeNardo, D.G., Brennan, D.J., Rexhepaj, E., Ruffell, B., Shiao, S.L., Madden, S.F., Gallagher, W.M., Wadhwani, N., Keil, S.D., Junaid, S.A., et al. (2011). Leukocyte complexity predicts breast cancer survival and functionally regulates response to chemotherapy. Cancer discovery 1, 54-67. DeNardo, D.G., and Coussens, L.M. (2007). Inflammation and breast cancer. Balancing immune response: crosstalk between adaptive and innate immune cells during breast cancer progression. Breast cancer research : BCR 9, 212. Diao, J., Zhao, J., Winter, E., and Cattral, M.S. (2010). Recruitment and differentiation of conventional dendritic cell precursors in tumors. J Immunol 184, 1261-1267. Diaz-Montero, C.M., Salem, M.L., Nishimura, M.I., Garrett-Mayer, E., Cole, D.J., and Montero, A.J. (2009). Increased circulating myeloid-derived suppressor cells correlate with clinical cancer stage, metastatic tumor burden, and doxorubicin-cyclophosphamide chemotherapy. Cancer Immunol Immunother 58, 49-59. 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. Dunn, G.P., Old, L.J., and Schreiber, R.D. (2004). The three Es of cancer immunoediting. Annu Rev Immunol 22, 329-360. DuPage, M., Mazumdar, C., Schmidt, L.M., Cheung, A.F., and Jacks, T. (2012). Expression of tumour-specific antigens underlies cancer immunoediting. Nature 482, 405-409. DuPre, S.A., and Hunter, K.W., Jr. (2007). Murine mammary carcinoma 4T1 induces a leukemoid reaction with splenomegaly: association with tumor-derived growth factors. Experimental and molecular pathology 82, 12-24. DuPre, S.A., Redelman, D., and Hunter, K.W., Jr. (2007). The mouse mammary carcinoma 4T1: characterization of the cellular landscape of primary tumours and metastatic tumour foci. International journal of experimental pathology 88, 351-360. Ernst, P., Fisher, J.K., Avery, W., Wade, S., Foy, D., and Korsmeyer, S.J. (2004a). Definitive hematopoiesis requires the mixed-lineage leukemia gene. Dev Cell 6, 437-443. Ernst, P., Mabon, M., Davidson, A.J., Zon, L.I., and Korsmeyer, S.J. (2004b). An Mll- dependent Hox program drives hematopoietic progenitor expansion. Current biology : CB 14, 2063-2069. Ernst, T., Chase, A.J., Score, J., Hidalgo-Curtis, C.E., Bryant, C., Jones, A.V., Waghorn, K., Zoi, K., Ross, F.M., Reiter, A., et al. (2010). Inactivating mutations of the histone methyltransferase gene EZH2 in myeloid disorders. Nature genetics 42, 722-726. Eyles, J.L., Roberts, A.W., Metcalf, D., and Wicks, I.P. (2006). Granulocyte colony- stimulating factor and neutrophils--forgotten mediators of inflammatory disease. Nat Clin Pract Rheumatol 2, 500-510. Faber, J., Krivtsov, A.V., Stubbs, M.C., Wright, R., Davis, T.N., van den Heuvel-Eibrink, M., Zwaan, C.M., Kung, A.L., and Armstrong, S.A. (2009). HOXA9 is required for survival in human MLL-rearranged acute leukemias. Blood 113, 2375-2385. 89  Francisco, L.M., Salinas, V.H., Brown, K.E., Vanguri, V.K., Freeman, G.J., Kuchroo, V.K., and Sharpe, A.H. (2009). PD-L1 regulates the development, maintenance, and function of induced regulatory T cells. J Exp Med 206, 3015-3029. Gabitass, R.F., Annels, N.E., Stocken, D.D., Pandha, H.A., and Middleton, G.W. (2011). Elevated myeloid-derived suppressor cells in pancreatic, esophageal and gastric cancer are an independent prognostic factor and are associated with significant elevation of the Th2 cytokine interleukin-13. Cancer Immunol Immunother 60, 1419-1430. Gabrilovich, D. (2004). Mechanisms and functional significance of tumour-induced dendritic-cell defects. Nature reviews. Immunology 4, 941-952. Gabrilovich, D.I., and Nagaraj, S. (2009). Myeloid-derived suppressor cells as regulators of the immune system. Nature reviews. Immunology 9, 162-174. Gabrilovich, D.I., Ostrand-Rosenberg, S., and Bronte, V. (2012). Coordinated regulation of myeloid cells by tumours. Nature reviews. Immunology 12, 253-268. 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. Ghiringhelli, F., Puig, P.E., Roux, S., Parcellier, A., Schmitt, E., Solary, E., Kroemer, G., Martin, F., Chauffert, B., and Zitvogel, L. (2005). Tumor cells convert immature myeloid dendritic cells into TGF-beta-secreting cells inducing CD4+CD25+ regulatory T cell proliferation. J Exp Med 202, 919-929. Gislason, T., and Nou, E. (1985). Sedimentation rate, leucocytes, platelet count and haemoglobin in bronchial carcinoma: an epidemiological study. European journal of respiratory diseases 66, 141-146. Gregory, A.D., Hogue, L.A., Ferkol, T.W., and Link, D.C. (2007). Regulation of systemic and local neutrophil responses by G-CSF during pulmonary Pseudomonas aeruginosa infection. Blood 109, 3235-3243. Grembecka, J., He, S., Shi, A., Purohit, T., Muntean, A.G., Sorenson, R.J., Showalter, H.D., Murai, M.J., Belcher, A.M., Hartley, T., et al. (2012). Menin-MLL inhibitors reverse oncogenic activity of MLL fusion proteins in leukemia. Nature chemical biology 8, 277-284. Hershman, D., Neugut, A.I., Jacobson, J.S., Wang, J., Tsai, W.Y., McBride, R., Bennett, C.L., and Grann, V.R. (2007). Acute myeloid leukemia or myelodysplastic syndrome following use of granulocyte colony-stimulating factors during breast cancer adjuvant chemotherapy. J Natl Cancer Inst 99, 196-205. Hess, J.L. (2004). MLL: a histone methyltransferase disrupted in leukemia. Trends Mol Med 10, 500-507. Horton, S.J., and Huntly, B.J. (2012). Recent advances in acute myeloid leukemia stem cell biology. Haematologica 97, 966-974. Inaba, K., Inaba, M., Romani, N., Aya, H., Deguchi, M., Ikehara, S., Muramatsu, S., and Steinman, R.M. (1992). Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with granulocyte/macrophage colony-stimulating factor. J Exp Med 176, 1693-1702. Ishii, M., Wen, H., Corsa, C.A., Liu, T., Coelho, A.L., Allen, R.M., Carson, W.F.t., Cavassani, K.A., Li, X., Lukacs, N.W., et al. (2009). Epigenetic regulation of the alternatively activated macrophage phenotype. Blood 114, 3244-3254. Jegalian, A.G., Facchetti, F., and Jaffe, E.S. (2009). Plasmacytoid dendritic cells: physiologic roles and pathologic states. Adv Anat Pathol 16, 392-404. 90  Kamminga, L.M., Bystrykh, L.V., de Boer, A., Houwer, S., Douma, J., Weersing, E., Dontje, B., and de Haan, G. (2006). The Polycomb group gene Ezh2 prevents hematopoietic stem cell exhaustion. Blood 107, 2170-2179. Kaszubowski, P.A., Husby, G., Tung, K.S., and Williams, R.C., Jr. (1980). T-lymphocyte subpopulations in peripheral blood and tissues of cancer patients. Cancer research 40, 4648- 4657. Klangsinsirikul, P., and Russell, N.H. (2002). Peripheral blood stem cell harvests from G- CSF-stimulated donors contain a skewed Th2 CD4 phenotype and a predominance of type 2 dendritic cells. Experimental hematology 30, 495-501. Kobayashi, H., Dubois, S., Sato, N., Sabzevari, H., Sakai, Y., Waldmann, T.A., and Tagaya, Y. (2005). Role of trans-cellular IL-15 presentation in the activation of NK cell-mediated killing, which leads to enhanced tumor immunosurveillance. Blood 105, 721-727. Korkaya, H., Liu, S., and Wicha, M.S. (2011). Breast cancer stem cells, cytokine networks, and the tumor microenvironment. The Journal of clinical investigation 121, 3804-3809. Kronin, V., Fitzmaurice, C.J., Caminschi, I., Shortman, K., Jackson, D.C., and Brown, L.E. (2001). Differential effect of CD8(+) and CD8(-) dendritic cells in the stimulation of secondary CD4(+) T cells. Int Immunol 13, 465-473. Lawicki, S., Czygier, M., Omyla, J., Bedkowska, E., and Szmitkowski, M. (2007). [The plasma levels of granulocyte-colony stimulating factor and granulocyte-macrophage colony stimulating factor in breast cancer patients]. Pol Merkur Lekarski 23, 259-263. Lawicki, S., Czygier, M., Wojtukiewicz, M., and Szmitkowski, M. (2009). [The plasma levels and diagnostic utility of granulocyte-colony stimulating factor (G-CSF) and granulocyte-macrophage - colony stimulating factor (GM-CSF) in patients with I and II stage of breast cancer]. Przegl Lek 66, 365-369. Leonard, R.C., Untch, M., and Von Koch, F. (2005). Management of anaemia in patients with breast cancer: role of epoetin. Annals of oncology : official journal of the European Society for Medical Oncology / ESMO 16, 817-824. Lessard, J., Schumacher, A., Thorsteinsdottir, U., van Lohuizen, M., Magnuson, T., and Sauvageau, G. (1999). Functional antagonism of the Polycomb-Group genes eed and Bmi1 in hemopoietic cell proliferation. Genes & development 13, 2691-2703. Lieschke, G.J., Grail, D., Hodgson, G., Metcalf, D., Stanley, E., Cheers, C., Fowler, K.J., Basu, S., Zhan, Y.F., and Dunn, A.R. (1994). Mice lacking granulocyte colony-stimulating factor have chronic neutropenia, granulocyte and macrophage progenitor cell deficiency, and impaired neutrophil mobilization. Blood 84, 1737-1746. Liu, F., Poursine-Laurent, J., and Link, D.C. (2000). Expression of the G-CSF receptor on hematopoietic progenitor cells is not required for their mobilization by G-CSF. Blood 95, 3025-3031. Liu, F., Wu, H.Y., Wesselschmidt, R., Kornaga, T., and Link, D.C. (1996). Impaired production and increased apoptosis of neutrophils in granulocyte colony-stimulating factor receptor-deficient mice. Immunity 5, 491-501. Liu, K., Victora, G.D., Schwickert, T.A., Guermonprez, P., Meredith, M.M., Yao, K., Chu, F.F., Randolph, G.J., Rudensky, A.Y., and Nussenzweig, M. (2009). In vivo analysis of dendritic cell development and homeostasis. Science 324, 392-397. Liu, K., Waskow, C., Liu, X., Yao, K., Hoh, J., and Nussenzweig, M. (2007). Origin of dendritic cells in peripheral lymphoid organs of mice. Nat Immunol 8, 578-583. 91  Lucas, M., Schachterle, W., Oberle, K., Aichele, P., and Diefenbach, A. (2007). Dendritic cells prime natural killer cells by trans-presenting interleukin 15. Immunity 26, 503-517. Maillard, I., Chen, Y.X., Friedman, A., Yang, Y., Tubbs, A.T., Shestova, O., Pear, W.S., and Hua, X. (2009). Menin regulates the function of hematopoietic stem cells and lymphoid progenitors. Blood 113, 1661-1669. Majewski, I.J., Blewitt, M.E., de Graaf, C.A., McManus, E.J., Bahlo, M., Hilton, A.A., Hyland, C.D., Smyth, G.K., Corbin, J.E., Metcalf, D., et al. (2008). Polycomb repressive complex 2 (PRC2) restricts hematopoietic stem cell activity. PLoS biology 6, e93. Majewski, I.J., Ritchie, M.E., Phipson, B., Corbin, J., Pakusch, M., Ebert, A., Busslinger, M., Koseki, H., Hu, Y., Smyth, G.K., et al. (2010). Opposing roles of polycomb repressive complexes in hematopoietic stem and progenitor cells. Blood 116, 731-739. Maldonado, R.A., and von Andrian, U.H. (2010). How tolerogenic dendritic cells induce regulatory T cells. Adv Immunol 108, 111-165. Metindir, J., and Bilir Dilek, G. (2009). Preoperative hemoglobin and platelet count and poor prognostic factors in patients with endometrial carcinoma. Journal of cancer research and clinical oncology 135, 125-129. Milne, K., Alexander, C., Webb, J.R., Sun, W., Dillon, K., Kalloger, S.E., Gilks, C.B., Clarke, B., Kobel, M., and Nelson, B.H. (2012). Absolute lymphocyte count is associated with survival in ovarian cancer independent of tumor-infiltrating lymphocytes. Journal of translational medicine 10, 33. Munder, M. (2009). Arginase: an emerging key player in the mammalian immune system. Br J Pharmacol 158, 638-651. Naik, S.H., Metcalf, D., van Nieuwenhuijze, A., Wicks, I., Wu, L., O'Keeffe, M., and Shortman, K. (2006). Intrasplenic steady-state dendritic cell precursors that are distinct from monocytes. Nat Immunol 7, 663-671. Naik, S.H., Proietto, A.I., Wilson, N.S., Dakic, A., Schnorrer, P., Fuchsberger, M., Lahoud, M.H., O'Keeffe, M., Shao, Q.X., Chen, W.F., et al. (2005). Cutting edge: generation of splenic CD8+ and CD8- dendritic cell equivalents in Fms-like tyrosine kinase 3 ligand bone marrow cultures. J Immunol 174, 6592-6597. Naik, S.H., Sathe, P., Park, H.Y., Metcalf, D., Proietto, A.I., Dakic, A., Carotta, S., O'Keeffe, M., Bahlo, M., Papenfuss, A., et al. (2007). Development of plasmacytoid and conventional dendritic cell subtypes from single precursor cells derived in vitro and in vivo. Nat Immunol 8, 1217-1226. Nijhof, W., De Haan, G., Dontje, B., and Loeffler, M. (1994). Effects of G-CSF on erythropoiesis. Ann N Y Acad Sci 718, 312-324; discussion 324-315. Nikoloski, G., Langemeijer, S.M., Kuiper, R.P., Knops, R., Massop, M., Tonnissen, E.R., van der Heijden, A., Scheele, T.N., Vandenberghe, P., de Witte, T., et al. (2010). Somatic mutations of the histone methyltransferase gene EZH2 in myelodysplastic syndromes. Nature genetics 42, 665-667. Ochi, A., Nguyen, A.H., Bedrosian, A.S., Mushlin, H.M., Zarbakhsh, S., Barilla, R., Zambirinis, C.P., Fallon, N.C., Rehman, A., Pylayeva-Gupta, Y., et al. (2012). MyD88 inhibition amplifies dendritic cell capacity to promote pancreatic carcinogenesis via Th2 cells. J Exp Med 209, 1671-1687. Olkhanud, P.B., Rochman, Y., Bodogai, M., Malchinkhuu, E., Wejksza, K., Xu, M., Gress, R.E., Hesdorffer, C., Leonard, W.J., and Biragyn, A. (2011). Thymic stromal lymphopoietin is a key mediator of breast cancer progression. J Immunol 186, 5656-5662. 92  Onai, N., Obata-Onai, A., Schmid, M.A., Ohteki, T., Jarrossay, D., and Manz, M.G. (2007). Identification of clonogenic common Flt3+M-CSFR+ plasmacytoid and conventional dendritic cell progenitors in mouse bone marrow. Nat Immunol 8, 1207-1216. Panopoulos, A.D., and Watowich, S.S. (2008). Granulocyte colony-stimulating factor: molecular mechanisms of action during steady state and 'emergency' hematopoiesis. Cytokine 42, 277-288. Papayannopoulou, T., Nakamoto, B., Andrews, R.G., Lyman, S.D., and Lee, M.Y. (1997). In vivo effects of Flt3/Flk2 ligand on mobilization of hematopoietic progenitors in primates and potent synergistic enhancement with granulocyte colony-stimulating factor. Blood 90, 620- 629. Park, S., Kim, E.S., Noh, D.Y., Hwang, K.T., and Moon, A. (2011). H-Ras-specific upregulation of granulocyte colony-stimulating factor promotes human breast cell invasion via matrix metalloproteinase-2. Cytokine 55, 126-133. Passegue, E., Wagers, A.J., Giuriato, S., Anderson, W.C., and Weissman, I.L. (2005). Global analysis of proliferation and cell cycle gene expression in the regulation of hematopoietic stem and progenitor cell fates. J Exp Med 202, 1599-1611. Pedroza-Gonzalez, A., Xu, K., Wu, T.C., Aspord, C., Tindle, S., Marches, F., Gallegos, M., Burton, E.C., Savino, D., Hori, T., et al. (2011). Thymic stromal lymphopoietin fosters human breast tumor growth by promoting type 2 inflammation. J Exp Med 208, 479-490. Pinzon-Charry, A., Ho, C.S., Laherty, R., Maxwell, T., Walker, D., Gardiner, R.A., O'Connor, L., Pyke, C., Schmidt, C., Furnival, C., and Lopez, J.A. (2005). A population of HLA-DR+ immature cells accumulates in the blood dendritic cell compartment of patients with different types of cancer. Neoplasia 7, 1112-1122. Pooley, J.L., Heath, W.R., and Shortman, K. (2001). Cutting edge: intravenous soluble antigen is presented to CD4 T cells by CD8- dendritic cells, but cross-presented to CD8 T cells by CD8+ dendritic cells. J Immunol 166, 5327-5330. Proietto, A.I., O'Keeffe, M., Gartlan, K., Wright, M.D., Shortman, K., Wu, L., and Lahoud, M.H. (2004). Differential production of inflammatory chemokines by murine dendritic cell subsets. Immunobiology 209, 163-172. Pulendran, B. (2005). Variegation of the immune response with dendritic cells and pathogen recognition receptors. J Immunol 174, 2457-2465. Pylayeva-Gupta, Y., Lee, K.E., Hajdu, C.H., Miller, G., and Bar-Sagi, D. (2012). Oncogenic Kras-Induced GM-CSF Production Promotes the Development of Pancreatic Neoplasia. Cancer cell 21, 836-847. Qiu, M.Z., Xu, R.H., Ruan, D.Y., Li, Z.H., Luo, H.Y., Teng, K.Y., Wang, Z.Q., Li, Y.H., and Jiang, W.Q. (2010a). Incidence of anemia, leukocytosis, and thrombocytosis in patients with solid tumors in China. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine 31, 633-641. Qiu, M.Z., Yuan, Z.Y., Luo, H.Y., Ruan, D.Y., Wang, Z.Q., Wang, F.H., Li, Y.H., and Xu, R.H. (2010b). Impact of pretreatment hematologic profile on survival of colorectal cancer patients. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine 31, 255-260. Quezada, S.A., Peggs, K.S., Simpson, T.R., and Allison, J.P. (2011). Shifting the equilibrium in cancer immunoediting: from tumor tolerance to eradication. Immunological reviews 241, 104-118. 93  Reis e Sousa, C. (2006). Dendritic cells in a mature age. Nature reviews. Immunology 6, 476-483. Sathe, P., and Wu, L. (2011). The network of cytokines, receptors and transcription factors governing the development of dendritic cell subsets. Protein Cell 2, 620-630. Satoh, T., Takeuchi, O., Vandenbon, A., Yasuda, K., Tanaka, Y., Kumagai, Y., Miyake, T., Matsushita, K., Okazaki, T., Saitoh, T., et al. (2010). The Jmjd3-Irf4 axis regulates M2 macrophage polarization and host responses against helminth infection. Nat Immunol 11, 936-944. Satthaporn, S., Robins, A., Vassanasiri, W., El-Sheemy, M., Jibril, J.A., Clark, D., Valerio, D., and Eremin, O. (2004). Dendritic cells are dysfunctional in patients with operable breast cancer. Cancer Immunol Immunother 53, 510-518. Sauvageau, M., and Sauvageau, G. (2010). Polycomb group proteins: multi-faceted regulators of somatic stem cells and cancer. Cell Stem Cell 7, 299-313. Scarlett, U.K., Rutkowski, M.R., Rauwerdink, A.M., Fields, J., Escovar-Fadul, X., Baird, J., Cubillos-Ruiz, J.R., Jacobs, A.C., Gonzalez, J.L., Weaver, J., et al. (2012). Ovarian cancer progression is controlled by phenotypic changes in dendritic cells. J Exp Med 209, 495-506. Schiavoni, G., Mattei, F., Sestili, P., Borghi, P., Venditti, M., Morse, H.C., 3rd, Belardelli, F., and Gabriele, L. (2002). ICSBP is essential for the development of mouse type I interferon- producing cells and for the generation and activation of CD8alpha(+) dendritic cells. J Exp Med 196, 1415-1425. Schnorrer, P., Behrens, G.M., Wilson, N.S., Pooley, J.L., Smith, C.M., El-Sukkari, D., Davey, G., Kupresanin, F., Li, M., Maraskovsky, E., et al. (2006). The dominant role of CD8+ dendritic cells in cross-presentation is not dictated by antigen capture. Proceedings of the National Academy of Sciences of the United States of America 103, 10729-10734. Schreiber, R.D., Old, L.J., and Smyth, M.J. (2011). Cancer Immunoediting: Integrating Immunity's Roles in Cancer Suppression and Promotion. Science 331, 1565-1570. Serbina, N.V., Salazar-Mather, T.P., Biron, C.A., Kuziel, W.A., and Pamer, E.G. (2003). TNF/iNOS-producing dendritic cells mediate innate immune defense against bacterial infection. Immunity 19, 59-70. Shojaei, F., Wu, X., Qu, X., Kowanetz, M., Yu, L., Tan, M., Meng, Y.G., and Ferrara, N. (2009). G-CSF-initiated myeloid cell mobilization and angiogenesis mediate tumor refractoriness to anti-VEGF therapy in mouse models. Proceedings of the National Academy of Sciences of the United States of America 106, 6742-6747. Shojaei, F., Zhong, C., Wu, X., Yu, L., and Ferrara, N. (2008). Role of myeloid cells in tumor angiogenesis and growth. Trends Cell Biol 18, 372-378. Shortman, K., and Heath, W.R. (2010). The CD8+ dendritic cell subset. Immunological reviews 234, 18-31. Steinman, R.M., and Cohn, Z.A. (1973). Identification of a novel cell type in peripheral lymphoid organs of mice. I. Morphology, quantitation, tissue distribution. J Exp Med 137, 1142-1162. Steinman, R.M., and Idoyaga, J. (2010). Features of the dendritic cell lineage. Immunological reviews 234, 5-17. Steinman, R.M., and Witmer, M.D. (1978). Lymphoid dendritic cells are potent stimulators of the primary mixed leukocyte reaction in mice. Proceedings of the National Academy of Sciences of the United States of America 75, 5132-5136. 94  Suzuki, E., Kapoor, V., Jassar, A.S., Kaiser, L.R., and Albelda, S.M. (2005). Gemcitabine selectively eliminates splenic Gr-1+/CD11b+ myeloid suppressor cells in tumor-bearing animals and enhances antitumor immune activity. Clin Cancer Res 11, 6713-6721. Tamura, T., Tailor, P., Yamaoka, K., Kong, H.J., Tsujimura, H., O'Shea, J.J., Singh, H., and Ozato, K. (2005). IFN regulatory factor-4 and -8 govern dendritic cell subset development and their functional diversity. J Immunol 174, 2573-2581. Tas, F., Eralp, Y., Basaran, M., Sakar, B., Alici, S., Argon, A., Bulutlar, G., Camlica, H., Aydiner, A., and Topuz, E. (2002). Anemia in oncology practice: relation to diseases and their therapies. American journal of clinical oncology 25, 371-379. Tefferi, A., and Vardiman, J.W. (2009). Myelodysplastic syndromes. N Engl J Med 361, 1872-1885. Thorsteinsdottir, U., Mamo, A., Kroon, E., Jerome, L., Bijl, J., Lawrence, H.J., Humphries, K., and Sauvageau, G. (2002). Overexpression of the myeloid leukemia-associated Hoxa9 gene in bone marrow cells induces stem cell expansion. Blood 99, 121-129. Tsujimura, H., Tamura, T., and Ozato, K. (2003). Cutting edge: IFN consensus sequence binding protein/IFN regulatory factor 8 drives the development of type I IFN-producing plasmacytoid dendritic cells. J Immunol 170, 1131-1135. Vignali, D.A., Collison, L.W., and Workman, C.J. (2008). How regulatory T cells work. Nature reviews. Immunology 8, 523-532. Vincent, J., Mignot, G., Chalmin, F., Ladoire, S., Bruchard, M., Chevriaux, A., Martin, F., Apetoh, L., Rebe, C., and Ghiringhelli, F. (2010). 5-Fluorouracil selectively kills tumor- associated myeloid-derived suppressor cells resulting in enhanced T cell-dependent antitumor immunity. Cancer research 70, 3052-3061. Waight, J.D., Hu, Q., Miller, A., Liu, S., and Abrams, S.I. (2011). Tumor-derived G-CSF facilitates neoplastic growth through a granulocytic myeloid-derived suppressor cell- dependent mechanism. PloS one 6, e27690. 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 research 67, 6442-6450. West, N.R., Milne, K., Truong, P.T., Macpherson, N., Nelson, B.H., and Watson, P.H. (2011). Tumor-infiltrating lymphocytes predict response to anthracycline-based chemotherapy in estrogen receptor-negative breast cancer. Breast cancer research : BCR 13, R126. Wilcox, R.A. (2010). Cancer-associated myeloproliferation: old association, new therapeutic target. Mayo Clinic proceedings. Mayo Clinic 85, 656-663. Yan, J., Chen, Y.X., Desmond, A., Silva, A., Yang, Y., Wang, H., and Hua, X. (2006). Cdx4 and menin co-regulate Hoxa9 expression in hematopoietic cells. PloS one 1, e47. Yang, T., Martin, M.L., Nielsen, J.S., Milne, K., Wall, E.M., Lin, W., Watson, P.H., and Nelson, B.H. (2009). Mammary tumors with diverse immunological phenotypes show differing sensitivity to adoptively transferred CD8+ T cells lacking the Cbl-b gene. Cancer Immunol Immunother 58, 1865-1875. Yokoyama, A., Somervaille, T., and Cleary, M.L. (2005). The menin tumor suppressor protein is an essential oncogenic cofactor for MLL-associated leukemogenesis. Blood 106, 196a-196a. 95  Zardo, G., Cimino, G., and Nervi, C. (2008). Epigenetic plasticity of chromatin in embryonic and hematopoietic stem/progenitor cells: therapeutic potential of cell reprogramming. Leukemia 22, 1503-1518. Zeidler, C., Schwinzer, B., and Welte, K. (2003). Congenital neutropenias. Rev Clin Exp Hematol 7, 72-83. Zhan, Y., Lieschke, G.J., Grail, D., Dunn, A.R., and Cheers, C. (1998). Essential roles for granulocyte-macrophage colony-stimulating factor (GM-CSF) and G-CSF in the sustained hematopoietic response of Listeria monocytogenes-infected mice. Blood 91, 863-869.    96  Appendix Appendix A  -- Primer sequences A.1 Sub-appendix – mRNA qPCR primers Primer 5’  3’ sequence Hoxa9 forward:  AAA CAA TGC CGA GAA TGA GAG GGG Hoxa9 reverse: AAA CTC CTT CTC CAG TTC CAG CGT Ezh2 forward: TCA CGT GTG GAG CTG CTG ACC Ezh2 reverse: AGC CTG CCA CAT CAG ACG GTG IRF-4 forward: GCT GCA TAT CTG CCT GTA TTA CCG IRF-4 reverse: GTG GTA ACG TGT TCA GGT AAC TCG TAG IRF-8 forward: ACG AGG TTA CGC TGT GCT CTG IRF-8 reverse: CAC GCC CAG CTT GCA TTT E2-2 forward: AGA CCA AGC TCC TGA TTC TC E2-2 reverse: AGG CTC TGA GGA CAC CTT CT GAPDH forward: CAT TTG CAG TGG CAA AGT GGA G GAPDH reverse: GTC TCG CTC CTG GAA GAT GGT G Rps29 forward: ACG GTC TGA TCC GCA AAT AC Rps29 reverse: CAT GAT CGG TTC CAC TTG GT A.2 Sub-appendix – ChIP qPCR primers Primer 3’  5’ sequence Hoxa9 amplicon A forward: TCC ACC TTT CTC TCG ACA G Hoxa9 amplicon A reverse: CCT GGC ACG ATT AGG CCT T Hoxa9 amplicon B forward: CAT TAA CAA CAG TGG CTG TGG CCT Hoxa9 amplicon B reverse: AAT TAA CCC GGG AGG AAC ACT GGA Hoxa9 amplicon C forward: AAC GAA TCT GTT GGT CGC TCC TGA Hoxa9 amplicon C reverse: AGC AGC CAA ATC GCA TTC TCA CTC Hoxa9 amplicon D forward: TGC ACA ACT GTT GAT GAC TGG CTG Hoxa9 amplicon D reverse: GCA GAA TGA TCT GCC ACA CGA AGA Hoxa9 amplicon E forward: AGC AGT GGA AAG TCC ACC CTT TCA Hoxa9 amplicon E reverse: ACC ACC CTA CTT CCA AAG GCT TCT 

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:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0073993/manifest

Comment

Related Items