Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Hyaluronan binding and CD44 in regulating hematopoiesis and CD8 T cell response Lee-Sayer, Sally 2017

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

Item Metadata

Download

Media
24-ubc_2017_september_lee-sayer_sally.pdf [ 3.01MB ]
Metadata
JSON: 24-1.0348315.json
JSON-LD: 24-1.0348315-ld.json
RDF/XML (Pretty): 24-1.0348315-rdf.xml
RDF/JSON: 24-1.0348315-rdf.json
Turtle: 24-1.0348315-turtle.txt
N-Triples: 24-1.0348315-rdf-ntriples.txt
Original Record: 24-1.0348315-source.json
Full Text
24-1.0348315-fulltext.txt
Citation
24-1.0348315.ris

Full Text

   HYALURONAN BINDING AND CD44 IN REGULATING HEMATOPOIESIS AND CD8 T CELL RESPONSE  by  Sally Lee-Sayer  B.Sc., The University of British Columbia, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Microbiology and Immunology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  June 2017  © Sally Lee-Sayer, 2017ii  Abstract CD44 is a ubiquitously expressed transmembrane glycoprotein. Through CD44, hematopoietic cells can be induced to bind hyaluronan, a component of the extracellular matrix, at specific developmental or functional stages, leading to the hypothesis that the interaction between CD44 and hyaluronan is important in regulating development and function of immune cells. I demonstrated that chondroitin sulfation of CD44 differentially regulates hyaluronan binding by classically and alternatively activated bone marrow-derived macrophages. T cells also increase hyaluronan binding upon activation, and chondroitin sulfate is likely involved in this regulation, since inhibitors of glycosylation, actin rearrangement, or sialylation had no effect on hyaluronan binding by activated T cells.  By using competitive cell transfer models, I found CD44 expression and hyaluronan binding were involved in regulating CD8 memory T cell formation and hematopoietic reconstitution. CD44-/- OT-I CD8 T cells formed significantly more memory cells than CD44+/+ OT-I CD8 T cells in the lymphoid organs, despite forming similar effector cell numbers, after intravenous infection with ovalbumin-expressing Listeria monocytogenes. While in competition, hyaluronan-binding CD8 effector T cells had increased pAkt expression and glucose uptake, both of which negatively regulate memory potential. Furthermore, hyaluronan-binding CD8 effector T cells showed increased death. Overall, this work implicates CD44 and hyaluronan binding as negative regulators of survival through contraction and memory formation by high affinity OT-I CD8 T cells. To investigate the role of hyaluronan binding in hematopoiesis, I transduced CD44-/- bone marrow cells with CD44 point mutants with increased or abolished hyaluronan binding (GOF or iii  LOF, respectively), and then transferred these cells into lethally irradiated hosts in competition with wild type cells. GOF cells out-competed wild type cells, which in turn out-competed LOF cells, in the reconstitution of all myeloid and most lymphoid populations, suggesting a competitive advantage in early hematopoiesis. Within the bone marrow stem and progenitor cells, GOF cells out-competed wild type cells, which in turn out-competed LOF cells, suggesting a role of the CD44-hyaluornan interaction in positively regulating reconstitution by hematopoietic stem and progenitor cells. Overall, this study identified novel roles for CD44 and HA binding in regulating CD8 T memory, and hematopoietic stem and progenitor cells.  iv  Lay Summary All blood cells originate from stem cells in the bone marrow through a process called hematopoiesis. Blood cells that protect the organism from disease are called immune cells, which include T cells. This thesis aims to understand how interaction with the environment influences the function of hematopoiesis and T cells. Specifically, the ability to interact with hyaluronan is investigated. Hyaluronan is a carbohydrate polymer that is a major non-cellular component of various tissues. In bone marrow transplant following irradiation-induced bone marrow injury, cells with increased ability to bind hyaluronan were better able to replenish blood cells. In T cells, hyaluronan binding cells were strongly activated and had reduced formation of memory cells, which are cells that confer effective protection against re-infection by the same pathogen. This work has highlighted the importance of hyaluronan in regulating the development and function of blood cells.   v  Preface I have designed and performed all the experiments and analyses presented in this thesis, with the following exceptions:  Manisha Dosanjh helped with making single cell suspensions from spleens and peripheral lymph nodes for flow analysis in select experiments  Past lab members, including Jesse Cooper and Leslie Sanderson, were also involved in the cloning of CD44 mutants and reporter genes into the retrovirus constructs used in Chapter 5  Anita Dahiya and Meghan Dougan, undergraduate students under my mentorship, performed the experiments and analyses for Fig. 4.1  Meghan Dougan also performed the experiment and analyses for Fig. 5.10A and Fig. 5.11 under my supervision  Appendix A was performed in collaboration with Dr. Chris Maxwell’s lab. Jenny Zhao, a student from the Maxwell Lab, performed the shRNA knockdown of RHAMM in cells shown in Fig. A. 2B-D. She also performed the immunoblot for Fig. A. 2B.  Data related to macrophages presented in chapter 3 have been published (Ruffell et al., 2011). Version of chapters 4 and 5 are in preparation for manuscript submissions.  Some data from the BM competitive transfer experiment in Chapter 5 are being included in another manuscript, which is has been submitted. I contributed to the following publications as part of my graduate program  Ruffell, B. Poon, G. F. T., Lee, S. S. M., Brown, K. L., Tjew, S-L, Cooper, J and Johnson, P. (2011), Differential use of chondroitin sulfate to regulate hyaluronan binding vi  by receptor CD44 in inflammatory and interleukin 4-activated macrophages. J. Biol. Chem., 286: 19179-19190. I was responsible for figures 5B, 6 and 7.  Lee-Sayer, S. S. M., Dong, Y., Arif, A. A., Olsson, M., Brown, K. L., & Johnson, P. (2015). The where, when, how, and why of hyaluronan binding by immune cells. Front. Immunol., 6, 150. I contributed to a third of the material of the review.  Maeshima, N., Poon, G. F. T., Dosanjh, M., Felberg, J., Lee, S. S. M., Cross, J. L., Birkenhead, D. and Johnson, P. (2011), Hyaluronan binding identifies the most proliferative activated and memory T cells. Eur. J. Immunol., 41: 1108–1119. I helped with tissue processing for in vivo experiments.  Dong, Y., Poon, G. F.T., Arif, A. A., Lee-Sayer, S. S. M., Dosanjh, M., and Johnson, P. (2017), The survival of fetal and bone marrow monocyte derived alveolar macrophages is promoted by CD44 and its interaction with hyaluronan. Mucosal Immunol. (In review.) I performed the experiment for one figure in collaboration with Dr. Grace F. T. Poon.  Animal studies were conducted in accordance with protocols approved by the University of British Columbia Animal Care Committee and Canadian Council of Animal Care. The following are project titles and their corresponding certificate numbers applicable to this thesis:   CD44 and CD45 breeding protocol number: A08-0936; A13-0015   Function of CD44 and hyaluronan - adaptive immunity: A11-0316; A15-0208  vii  Table of Contents Abstract  ................................................................................................................................... ii Lay Summary ............................................................................................................................... iv Preface  ....................................................................................................................................v Table of Contents ........................................................................................................................ vii List of Tables .............................................................................................................................. xiii List of Figures ............................................................................................................................. xiv List of Symbols .......................................................................................................................... xvii List of Abbreviations ............................................................................................................... xviii Acknowledgements ................................................................................................................... xxii Chapter 1: Introduction ................................................................................................................1  1.1 Overview of the immune system .................................................................................... 1  1.2 Hyaluronan and CD44 ................................................................................................... 2 1.2.1 Alternate HA Receptors .............................................................................................. 5 1.2.2 Regulation of HA ........................................................................................................ 8 1.2.3 CD44 and the immune system .................................................................................... 9  1.3 Hematopoiesis ............................................................................................................... 10 1.3.1 Identifying HSPC ...................................................................................................... 11 1.3.2 Identifying HSPC by flow cytometry ....................................................................... 13 1.3.3 HSPC localization ..................................................................................................... 14 1.3.4 BM niche for HSPC .................................................................................................. 17 1.3.4.1 Factors for HSC localization, maintenance ...................................................... 18 1.3.4.2 Factors for HSPC proliferation and differentiation .......................................... 21  viii  1.3.5 HSPC niche and HSPC function ............................................................................... 21 1.3.6 CD44 and HA in hematopoiesis ............................................................................... 22  1.4 CD8 T lymphocytes ....................................................................................................... 23 1.4.1 CD8 T Cell activation ............................................................................................... 23 1.4.2 CD8 effector T cells .................................................................................................. 25 1.4.3 Memory cell formation ............................................................................................. 26 1.4.3.1 Cytokines .......................................................................................................... 26 1.4.3.2 Metabolism ....................................................................................................... 28 1.4.4 Memory and subsequent responses ........................................................................... 32 1.4.5 Models of CD8 T memory cell formation ................................................................ 32  1.5 CD44 and HA in CD8 T cell functions ........................................................................ 34  1.6 Molecular regulation of CD44-mediated HA binding ............................................... 37  1.7 Thesis rationale and hypotheses .................................................................................. 38 Chapter 2: Materials and Methods ............................................................................................40  2.1 Mice ................................................................................................................................ 40  2.2 Antibodies and reagents ............................................................................................... 40  2.3 Buffers, solutions and growth media .......................................................................... 41  2.4 Cell isolation .................................................................................................................. 42  2.5 Retroviral transduction ................................................................................................ 43  2.6 Generation of BM-derived macrophages ................................................................... 43  2.7 CD8 T cell isolation and activation in vitro ................................................................ 44  2.8 BM reconstitution ......................................................................................................... 45  2.9 Adoptive transfer and systemic Listeria monocytogenes infection ........................... 45  ix   2.10 Flow cytometry and cell sorting .................................................................................. 46  2.11 RNA isolation, reverse transcription and quantitative PCR .................................... 47  2.12 Statistical analysis ......................................................................................................... 47 Chapter 3: Molecular Mechanism of HA Binding by Immune Cells ......................................48  3.1 Introduction and rationale ........................................................................................... 48  3.2 Results ............................................................................................................................ 50 3.2.1 HA binding is increased in classically activated and alternatively activated BMDM . ……………………………………………………………………………………50 3.2.2 Classically and alternatively activated BMDM differentially regulate HA binding through chondroitin sulfate ...................................................................................... 52 3.2.3 Classically and alternatively activated BMDM differentially regulate expression of CD44 v10 ................................................................................................................. 54 3.2.4 Actin polymerization is not required for induction of HA binding by BMDM ........ 56 3.2.5 Sialic acid can negatively regulate CD44-mediated HA binding in BMDM ........... 56 3.2.6 Inhibition of GAG addition with xyloside has no effect on HA binding by splenic CD4 and CD8 T cells ............................................................................................... 60 3.2.7 Actin polymerization is not required for induction of HA binding by splenic T cells ................................................................................................................................. 62 3.2.8 Sialylation of CD44 negatively regulates HA binding by splenic T cells ................ 62 3.2.9 O-link glycosylation does not negatively regulate HA binding by splenic T cells .. 65  3.3 Discussion ...................................................................................................................... 67 Chapter 4: CD44 and HA Binding Reduce Memory Potential of CD8 Effector T Cells ......70  4.1 Introduction and rationale ........................................................................................... 70  x   4.2 Results ............................................................................................................................ 71 4.2.1 The magnitude and extent of induced FL-HA binding in activated CD8 T cells correlate positively with the strength of the TCR signal during activation ............. 71 4.2.2 HA binding and CD44 expression have no direct effect on the generation of effector CD8 T cells .............................................................................................................. 75 4.2.3 CD44-/- OT-I CD8 T cells have a competitive advantage in memory formation ..... 75 4.2.4 The increased CD44-/- to CD44+/+ ratios in OT-I CD8 T cells are sustained in secondary responses ................................................................................................ 80 4.2.5 CD44-/- OT-I CD8 T cells are not more responsive to IL-7 or IL-15 ....................... 80 4.2.6 HA-binding WT OT-I CD8 T cells are more susceptible to death ........................... 82 4.2.7 HA-binding OT-I CD8 T cells exhibit an increased ability for glucose uptake ....... 86 4.2.8 HA-binding OT-I CD8 T cells have increased expression of pAkt .......................... 87 4.2.9 The HA-binding CD8 effector population is selected against during contraction.... 90  4.3 Discussion ...................................................................................................................... 92 Chapter 5: CD44-Mediated HA Binding Provides a Competitive Advantage in Bone Marrow Engraftment and Reconstitution ............................................................96  5.1 Introduction and rationale ........................................................................................... 96  5.2 Results ............................................................................................................................ 97 5.2.1 Hematopoietic cells exhibit different HA binding profiles in steady state ............... 97 5.2.2 CD44 is dispensable for early hematopoietic development .................................... 101 5.2.3 HA binding positively impacts reconstitution in competitive BM transfer ............ 105 5.2.4 HA binding positively impacts reconstitution of the myeloid and lymphoid populations in competitive BM transfer ................................................................ 108  xi  5.2.5 HA binding positively impacts reconstitution of the HSPC populations in competitive BM transfer ........................................................................................ 110 5.2.6 HSPC increase their HA binding when induced to proliferate ............................... 113 5.2.7 HA increases cell number of proliferating HSC and MPP ..................................... 116  5.3 Discussion .................................................................................................................... 118 Chapter 6: Summary and Future Directions ..........................................................................122  6.1 Regulation of HA binding in macrophages and T cells ........................................... 122  6.2 The role of CD44 and HA binding in CD8 T cells ................................................... 123 6.2.1 HA-binding CD44 negatively regulates CD8 T memory formation ...................... 123 6.2.2 Future directions for studying the role of CD44 and HA binding in CD8 memory T cells formation ....................................................................................................... 126  6.3 The role of CD44 and HA binding in hematopoiesis ............................................... 127 6.3.1 HA binding is exhibited by certain hematopoietic populations in homeostasis ..... 127 6.3.2 HA binding positively regulates BM reconstitution through enhancement of proliferation ........................................................................................................... 128 6.3.3 Future directions for studying the role of CD44 and HA binding in BM HSPC .... 131  6.4 Concluding remarks ................................................................................................... 132 References  ................................................................................................................................133 Appendix A HA Binding by the Invasive Ductal Carcinoma Cell Line MDA-MB-231 is Mediated by CD44, and not by RHAMM ......................................................... 158 A.1 Introduction and Rationale ...................................................................................... 158 A.2 Methods................................................................................................................... 159 A.3 Results ..................................................................................................................... 161  xii  A.3.1 HA-binding by breast tumour line MDA-MB-231 is mediated by CD44 .. 161 A.3.2 RHAMM is dispensable for HA-binding by breast tumour line MDA-MB-231 ............................................................................................................. 161 A.4 Discussion ............................................................................................................... 164   xiii  List of Tables Table 1.1 HA binding ability of hematopoietic cells ...................................................................... 6  xiv  List of Figures Figure 1.1 CD44 structure............................................................................................................... 4 Figure 1.2 Schematic of domain and module organization of HA-binding proteins ...................... 7 Figure 1.4 Haematopoietic lineage tree ........................................................................................ 12 Figure 1.5 Identification of hematopoietic stem and progenitor cells in adult bone marrow ....... 15 Figure 1.6 Morphological construction of bone marrow .............................................................. 16 Figure 1.7 Key cell types involved in regulating HSC maintenance ............................................ 19 Figure 1.8 Overall reactions in glycolysis, pyruvate oxidation and the TCA cycle ..................... 29 Figure 1.9 Oxidative phosphorylation .......................................................................................... 30 Figure 3.1 Classically and alternatively activated BMDM induce HA binding through different mechanisms ................................................................................................................ 51 Figure 3.2 GAG addition affects HA binding of activated BMDM ............................................. 53 Figure 3.3 Relative mRNA expression of CD44s and CD44 v10 in BMDM............................... 55 Figure 3.4 Inhibiting actin polymerization has no effect on HA binding in activated BMDM .... 57 Figure 3.5 Neuraminidase treatment slightly increases HA binding in activated BMDM ........... 59 Figure 3.6 Inhibition of GAG addition or actin polymerization has no effect on HA binding by naïve and activated T cells ......................................................................................... 61 Figure 3.7 Neuraminidase treatment increases HA binding in naïve and activated T cells, but inhibition of neuraminidase has no effect on HA binding ............................................................ 63 Figure 3.8 Inhibition of O-linked glycosylation has no effect on HA binding by splenic T cells 66 Figure 4.1 TCR peptide affinity and avidity correlate positively with degree of HA binding by activated OT-I CD8 T cells ........................................................................................ 73  xv  Figure 4.2 CD44-/- CD8 T cells have no cell intrinsic defect in cell numbers or IFNγ production during the effector response ....................................................................................... 76 Figure 4.3 CD44-/- OT-I CD8 T cells have a competitive advantage in memory cell formation . 79 Figure 4.4 The CD44-/- OT-I CD8 T cell advantage over CD44+/+ OT-I CD8 T cells is maintained in secondary effector response ................................................................................... 81 Figure 4.5 CD44-/- OT-I CD8 T cells do not have an increased response to the common-γ cytokines IL-7 or IL-15 .............................................................................................. 83 Figure 4.6 CD44-/- OT-I CD8 T cells are less susceptible to apoptosis ........................................ 85 Figure 4.7 HA-binding WT OT-I CD8 T cells have increased 2-NBDG uptake and increased pAkt expression.......................................................................................................... 88 Figure 4.8 The HA-binding CD8 effector population is selected against during contraction ...... 91 Figure 5.1 Gating strategies .......................................................................................................... 98 Figure 5.2 CD44 expression and HA-binding by mature immune populations and HSPC from WT and CD44-/- mice ................................................................................................. 99 Figure 5.3 Cell numbers of total BMC, and BM HSPC populations in WT and CD44-/- mice .. 103 Figure 5.4 Competitive BM reconstitution by WT and CD44-/- BMC into irradiated congenic WT mice .......................................................................................................................... 104 Figure 5.5 Retroviral transduction of CD44 constructs into CD44-/- BMC ................................ 106 Figure 5.6 Competitive BM reconstitution by tv-WT and tLOF BMC or tv-WT and tGOF BMC transferred into irradiated congenic WT mice ......................................................... 107 Figure 5.7 Competitive reconstitution of the myeloid compartment by tv-WT and tLOF BMC or tv-WT and tGOF BMC transferred into irradiated congenic WT mice ................... 109  xvi  Figure 5.8 Competitive reconstitution of the lymphoid compartment by tv-WT and tLOF BMC or tv-WT and tGOF BMC transferred into irradiated congenic WT mice ............... 111 Figure 5.9 Competitive reconstitution of BM HSPC by tv-WT and tLOF BMC or tv-WT and tGOF BMC transferred into irradiated congenic WT mice ..................................... 112 Figure 5.10 HA-binding by BM HSPC is induced by proliferation ........................................... 114 Figure 5.11 Exogenous HA increases cell number of proliferating HSPC................................. 117 Figure 6.1 Model for the role of CD44 and HA in CD8 T cell response .................................... 125 Figure 6.2 Model for the role of CD44 and HA in BM HSPC ................................................... 130 Figure A.1 FL-HA binding by MDA-MB-231 tumour cells is mediated by CD44 ................... 162 Figure A.2 FL-HA binding by MDA-MB-231 is not mediated by RHAMM ............................ 163   xvii  List of Symbols α alpha β beta γ gamma µ micro   xviii  List of Abbreviations 2-NBDG 2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)-2-deoxyglucose 4-MU  4- methylumbelliferone AICD  activation-induced cell death APC  antigen-presenting cell(s) APC-HA HA conjugated to Alexa Flour® 647 ATP  adenosine triphosphate  BM  bone marrow BMC   bone marrow cells BMDC bone marrow-derived dendritic cells BMDM bone marrow-derived macrophages BRC AbLab Biomedical Research Centre Antibody Lab CAR  CXCL12-abundent reticular CCR  CC chemokine receptor CD  cluster of differentiation CD44s  standard CD44 isoform CFU  colony forming units CLP  common lymphoid progenitors CMP  common myeloid progenitors CoA  coenzyme A Con-A  concanavalin-A CXCL  chemokine C-X-C motif ligand  CXCR  chemokine C-X-C motif receptor  xix  DAMP  danger-associated molecular patterns DAPI  4',6-diamidino-2-phenylindole DC  dendritic cell(s) EAE  experimental autoimmune encephalomyelitis ECM  extracellular matrix EDTA  ethylenediaminetetraacetic acid eomes  eomesodermin  FADH2 flavin adenine dinucleotide (hydroquinone form) FCS  fetal calf serum FL-HA fluorescein-labelled HA GAG  glycosaminoglycan Glut1  glucose transport 1 GFP   green florescent protein  GOF  gain of function, specifically CD44 with S183A point mutation GTP  guanosine triphosphate HA  hyaluronan HMW  high molecular weight HSC  hematopoietic stem cell(s) HSPC  hematopoietic stem and progenitor cell(s) IL  interleukin IL-15Rα IL-15 receptor α IFN  interferon  KLRG1 killer cell lectin-like receptor subfamily G member 1  xx  LepR   leptin receptor  LM  Listeria monocytogenes LM-OVA Listeria monocytogenes expressing recombinant ovalbumin LN  lymph nodes LOF  loss of function, specifically CD44 with R43A point mutation LPS  lipopolysaccharide LSK  lineage- Sca-1+ c-Kit+, aka KLS LYVE1 lymphatic vessel endothelial hyaluronan receptor 1 MHC  major histocompatibility complex  MPEC  memory-precursor effector cells MPP  multipotent progenitos MSPC  mesenchymal stem cells mTOR  mammalian target of rapamycin mTORC1 mTOR complex 1 NADH  nicotinamide adenine dinucleotide (reduced form) NeuAc2en  2-Deoxy-2,3-dehydro-N-acetylneuraminic acid  NK  natural killer NKT  natural killer T OXPHOS oxidative phosphorylation PAMP  pathogen-associated molecular patterns PDGFR-α platelet-derived growth factor receptor-α PI3K  phosphoinositide 3 kinase pLN  peripheral lymph nodes  xxi  PMA  phorbol 12-myristate 13-acetate pNAcGal  phenyl-N-diacetyl-α-D-galactosaminide  RBC  red blood cell(s) RFP  red fluorescent protein RHAMM hyaluronan-mediated motility receptor SCF  stem cell factor SLAM  signalling lymphocyte activation molecule SLEC  short-lived effector cells SPkls  side population KLS TCA  tricarboxylic acid Tcm  central memory T cell(s) Tem  effector memory T cell(s) tGOF CD44-/- bone marrow cells transduced with an GOF-CD44 containing vector tLOF CD44-/- bone marrow cells transduced with an LOF-CD44 containing vector TSG-6  TNF-stimulated gene 6 protein tv-WT wild type bone marrow cells transduced with an empty vector  UBC  University of British Columbia WT  wild type xyloside p-nitrophenyl β-D-xylopyranoside     xxii  Acknowledgements I would like to thank my supervisor Dr. Pauline Johnson for her support and patience through all the years of hardship and small breakthroughs. Pauline has been an excellent mentor, and her diligence as a scientist and her genuine care for her students have left lasting impressions on me. I am constantly in awe of her limitless curiosity about science and her enthusiasm about data.  I also thank Dr. Ninan Abraham, Dr. Ken Harder and Dr. Kelly McNagny, for their support, expertise and constructive critiques as members of my graduate committee members. I also offer enduring gratitude to all past and present lab members in the Johnson Lab. Their mere presence made long experiment days enjoyable. Thank you for all the times I needed to discuss data and ideas, or to vent about the unpleasantness of the day. I would like to especially thank Manisha Dosanjh, for all her help with my experiments and for keeping the lab running week after week. I am forever indebted to my family, especially my parents, Yu-Wen and Jenny, for all the years of unconditional support. Thank you for understanding when I was always late for dinner. Lastly I would like to offer my deepest gratitude to my husband, Clayton, who can now talk about CD44 and hyaluronan. Thank you for putting up with my unconventional and unpredictable schedule, and thank you for doing laundry and cooking delicious meals. I held the Canada Graduate Scholarship Masters (CGS-M) in 2010-2011, and then the Postgraduate Scholarship – PhD (PGS-P) in 2011-2014 from the Natural Sciences and Engineering Research Council of Canada. This work was initially supported by funding from the  xxiii  Natural Sciences and Engineering Research Council of Canada and later by funding from Canadian Institutes of Health Research.     1   Chapter 1: Introduction 1.1 Overview of the immune system The immune system consists of a complex array of functionally interconnected organs, cells and molecules, and is primarily tasked with protecting the body against infections. Defects in any aspect of the immune system can lead to increased susceptibility to infectious diseases and cancer. On the other hand, an immune response is inherently damaging to the host, and the ability to limit and resolve the immune response is also important, as defects in immune regulation contribute to development of allergies and autoimmune diseases. The immune system can be divided into two branches – innate and adaptive. The innate immune cells are the first line of immune defense, and respond to danger- and pathogen-associated molecular patterns (DAMP and PAMP, respectively), which activates them. Additional innate immune cells are then recruited to the site of damage or infection. Sometimes the innate immune system alone is able to resolve the infection and repair the tissue. If not, the adaptive immune system is activated and contributes to pathogen clearance.  The adaptive system consists of T and B lymphocytes, which are able to recognize specific antigens through their respective antigen receptors. The antigen specificity of individual T and B lymphocytes arise from somatic recombination (and somatic hypermutation in B cells), which results in an infinite number of possible targets. T and B lymphocytes are an essential component of the immune system, as defects in their development leads to severe immunodeficiencies (Bosma and Carroll, 1991; Mombaerts et al., 1992). Furthermore, T and B lymphocytes have the ability to respond more quickly to the same antigen during subsequent re-encounters, a phenomenon termed “memory,” and is the basis for vaccine function.    2   1.2 Hyaluronan and CD44 Hyaluronan (HA) is a component of the extracellular matrix (ECM), and is found all over the body. Most of the HA in the body is found in the skin and skeleton, but it is also present in the lung, spleen, lymph nodes (LN) and, to a less extent, the liver, while blood has the lowest concentration of HA (Fraser et al., 1997). HA is a glycosaminoglycan (GAG) composed of alternating β1-3- and β1-4-linked D-glucuronic acid & D-N-acetylglucosamine.  The major cell receptor for HA is CD44, a ubiquitously expressed type I transmembrane cell surface glycoprotein (Ponta et al., 2003). CD44 is encoded by a single gene, and contains 20 exons, 10 standard exons and 10 variable exons (Screaton et al., 1992) (Fig. 1.1A). CD44 contains an extracellular domain, a single pass transmembrane domain, and a cytoplasmic tail. The extracellular domain includes a distal globular HA-binding domain and a proximal stem region. The most commonly expressed form of CD44 contains only the standard exons and is referred to as standard CD44 (CD44s). Additional variable exons (V1-10) can be inserted into the stem region after exon 5 by alternative splicing (Ponta et al., 2003) (Fig 1.1A, B). While hematopoietic cells predominantly express CD44s, other cell types can express other isoforms. For example CD44v6 is expressed by various cancer cells (reviewed in (Heider et al., 2004)), and keratinocytes express CD44v3-10 and CD44v1-10 (Bloor et al., 2001; Bourguignon and Bikle, 2015).  CD44 can be post-translationally modified by O- and N-glycosylation, and the inclusion of  additional variable exons also introduces additional glycosylation sites (Fig. 1.1B). Glycosylation can both positively and negative regulate HA-binding by CD44 (Dasgupta et al.,    3   A   CD44s  Hematopoietic cells CD44v6  Cancer cells CD44v3-10  Keratinocytes CD44v1-10  Keratinocytes  B   C     4   Figure 1.1 CD44 structure (A) Genomic organization of CD44, and alternative transcript splicing of CD44. (B) Schematic of CD44 structure and post-translational modifications. Adapted from (Ponta et al., 2003). (C) Structure of HA-binding domain of mouse CD44. HA 8-mer is shown in yellow, and residues involved in HA-binding are shown in blue. PDB: 2JCR from (Banerji et al., 2007).    5   1996; English et al., 1998; Lesley et al., 1995; Skelton et al., 1998). Chondroitin sulfate can be added to a serine residue in the stem region, and this inhibits HA-binding by CD44 (Ruffell and Johnson, 2005).  CD44 binds HA through mostly hydrogen bonds, in a distinct groove in the HA-binding domain (Banerji et al., 2007) (Fig. 1.1C). The complexity in the regulation of CD44-mediated HA-binding by immune cells increases the difficulty in deciphering the molecular mechanism of induced HA binding in the immune system. Different immune cell types and subsets seem to employ different mechanisms in regulating HA-binding by CD44, and CD44 expression levels and HA binding capacity alter at specific developmental and functional stages of the immune cell type (Table 1.1).  1.2.1 Alternate HA Receptors CD44 belongs to the Link module superfamily of proteins. The Link module is found in tandem pairs in cartilage-specific proteoglycans, such as aggrecan versican, neurocan and brevican. These proteoglycans from large complexes with HA and provide structural integrity of several tissues. Other members of this superfamily, including CD44, lymphatic vessel endothelial hyaluronan receptor 1 (LYVE1), TNF-stimulated gene 6 protein (TSG-6), and HA receptor for endocytosis (HARE, aka stabilin-2) only contain a single Link module (Day and Prestwich, 2002) (Fig. 1.2).  Other than CD44, LYVE1 and HARE are the only other known members of the Link module superfamily to bind HA and contain a transmembrane domain, and are thus the only other   6   Table 1.1 HA binding ability of hematopoietic cells Adapted from (Lee-Sayer et al., 2015). Cell Type Stimulation Type of HA binding HA Receptor Refs Monocyte (human) TNFα, LPS, IL-1, IFN-γ,  Inducible CD44 (Brown et al., 2001; Levesque and Haynes, 1996, 1997) Alveolar macrophages (human, rodents) None Constitutive CD44 (Culty et al., 1992; Culty et al., 1994; Katoh et al., 2001; Teder and Heldin, 1997; Underhill et al., 1993) Peritoneal macrophage (mouse) LPS with IFNγ, or IL-4 Inducible CD44 (Ruffell et al., 2011) Bone marrow derived macrophages (mouse) LPS with IFNγ, TNFα, or IL-4 Inducible CD44 (Ruffell et al., 2011) Monocyte-derived DC (human) CD40L expressing fibroblasts Inducible CD44  (Termeer et al., 2001) B cells (human, mouse) PMA, IL-5, LPS Inducible, a subset binds CD44 (Hathcock et al., 1993; Katoh et al., 1995; Kryworuchko et al., 1999; Murakami et al., 1990) T cells (mouse) PMA/ionomycin, CD3 antibodies, specific antigen or superantigen Inducible, often a subset binds CD44 (DeGrendele et al., 1997; Lesley et al., 1994; Maeshima et al., 2011) CD4+ CD25+ T regulatory cells (human & mouse) CD3 +/- CD28 activation Inducible, a subset binds CD44 (Bollyky et al., 2007; Firan et al., 2006) Neutrophil (mouse) LPS induced liver Inflammation in vivo Binding to SHAP modified HA CD44, not RHAMM (McDonald et al., 2008) NK cells (mouse) IL-2, IL15 Inducible, a subset binds CD44 (Sague et al., 2004) Platelets (mouse) None Constitutive CD44 (Koshiishi et al., 1994)     7    Figure 1.2 Schematic of domain and module organization of HA-binding proteins With the exception of RHAMM, all proteins shown contain the Link module. CD44, LYVE1 and HARE contain a transmembrane domain. Neurocan, brevican, aggrecan and versican include a Ig module and a large GAG attachment domain. HARE and RHAMM both include BX7B motifs. CCP, complement control protein; CUB, complement C1r/C1s, Uegf, Bmp1; EGF, epidermal growth factor; GAG, glycosaminoglycan. Adapted from (Day and Prestwich, 2002; Politz et al., 2002).   8   known HA receptors in the superfamily (Day and Prestwich, 2002; Politz et al., 2002) (Fig. 1.2). LYVE-1 has 41% sequence identity to CD44 and binds to both soluble and immobilized HA (Banerji et al., 1999). LYVE1 is exclusively expressed in lymphatic endothelial cells, and may be involved in supporting CD44-mediated rolling on the endothelium, and trafficking of HA to the LN for degradation (Jackson, 2003). HARE is expressed by sinusoidal endothelial cells of the liver, spleen and LN of rats and humans, and mediate HA uptake by these cells in the liver (Zhou et al., 2003; Zhou et al., 2000). HARE-deficient mice have increased serum levels of HA, suggesting that HARE is a key scavenger of HA (Hirose et al., 2012).  1.2.2 Regulation of HA The bones, liver, spleen and lymph nodes, as well as skin, are sites of considerable HA metabolism (Fraser et al., 1997). Unlike other GAGs, HA is synthesized at the plasma membrane, rather than the Golgi apparatus (Fraser et al., 1997). In mammals, three functional HA synthases have been identified, and are termed HAS1, HAS2 and HAS3 (Weigel, 2015). The three HA synthases are not functionally redundant and differ in their relative kinetics of HA synthesis and the average molecular mass of the resulting products (Toole, 2004). HAS2 is the most biologically significant of the three, as deletion of Has2 results in fetal death mid-gestation, while deletion of both Has1 and Has3 does not cause embryonic lethality (Goncharova et al., 2012). Synthesis of HA can be inhibited with 4-methylumbelliferone (4-MU), which can inhibit HA synthesis through two mechanisms. 4-MU can competitively inhibit UDP-glucuronosyltransferase, an enzyme involved in HA synthesis; as well, 4-MU can bind to UDP-glucuronic acid, a substrate of polysaccharide synthesis, rendering it biochemically inactive (Nagy et al., 2015). Administration of 4-MU reduces disease symptoms in murine models of    9   multiple sclerosis, arthritis, and superantigen-induced lung inflammation (McKallip et al., 2013; Mueller et al., 2014; Yoshioka et al., 2013), while it exacerbates murine atherosclerosis (Nagy et al., 2010), suggesting that HA plays different roles in different disease models. HA of different molecular weights appear to have varying effects on cells (Petrey and de la Motte, 2014). Large HA molecules can be degraded specifically by a family of enzymes known as hyaluronidases. At least six hyaluronidases have been identified in humans and mice (Csoka et al., 2001). In homeostasis, the various HA synthases and hyaluronidases are responsible for the constant turnover of HA, where HA exists in the high molecular weight (HMW) form of 1 x 106 Da or more. During inflammation, smaller HA molecules are produced, and HA also can be degraded into smaller fragments of 20-250 KDa by hyaluronidases and reactive oxygen species (Noble, 2002).   1.2.3 CD44 and the immune system Despite the ubiquitous nature of CD44 expression, CD44-/- mice have normal development and no dramatic defects in hematopoiesis (Schmits et al., 1997). The differences between wild type (WT) and CD44-/- mice were mostly only observed during immune responses. In a model of concanavalin-A -induced hepatitis, CD44-/- mice are more susceptible to disease, and injection with heat-killed Corynebacterium parvum induces significantly larger granulomas in the livers of CD44-/- mice than in WT (Schmits et al., 1997). CD44-/- mice also fail to recover from bleomycin-induced sterile lung inflammation due to unremitting inflammation, leading to increased HA and impaired clearance of apoptotic neutrophils (Teder et al., 2002). While these    10   studies suggest an anti-inflammatory role for CD44 in an immune response, others suggest the opposite. CD44-/- mice are less susceptible to inflammation in models of experimental autoimmune encephalomyelitis (EAE) and Dermatophagoides farina-induced airway hyper-responsiveness (Guan et al., 2011; Katoh et al., 2011). CD44-/- mice are also more resistant to collagen-induced arthritis (Stoop et al., 2001). Furthermore, removal of CD44v7 and CD44v6 through administration of specific antibodies or genomic deletion is curative for colitis induced by 2,4,6-trinitrobenzenesulfonic acid (Wittig et al., 1998; Wittig et al., 2000). The heterogeneity in the consequences of CD44 deletion in different immune responses is likely due the different types of immune response and different effector cells being affected, and highlights the complexity of the role of CD44 in immune regulation.   1.3 Hematopoiesis In adult mice, the hematopoietic stem cells (HSC) give rise to all blood cells, including erythrocytes (red blood cells, RBC), platelets and all immune cells, through the process of hematopoiesis, which primarily occurs in the bone marrow (BM) (Orkin and Zon, 2008). The HSC can self-renew and differentiate into the more proliferative multipotent progenitors (MPP) (Morrison et al., 1997). The BM hematopoietic stem and progenitor cell (HSPC) population, which includes the HSC and MPP, is very heterogeneous and the boundary between HSC and MPP based on the ability for self-renewal is blurry (Eaves, 2015; Oguro et al., 2013). The HSPC population likely consists of a series of cells on a continuum of reducing self-renewal potential, with long-term HSC and MPP on opposite ends (Orkin and Zon, 2008).     11    MPP give rise to the common lymphoid and myeloid progenitors (CLP and CMP, respectively). The CLP eventually give rise to B cells, T cells, natural killer (NK) cells, NKT cells, innate lymphoid cells and some dendritic cells (DC) (Kondo et al., 1997; Montecino-Rodriguez et al., 2001; Vivier et al., 2016), while the CMP give rise to platelets, erythrocytes, monocytes, macrophage, granulocytes (neutrophils, eosinophils, basophils and mast cells), and DC (Akashi et al., 2000) (Fig. 1.3).   1.3.1 Identifying HSPC The idea of HSC initially arose from the observation that transplantation of adult bone marrow cells (BMC) can rescue mice from lethal irradiation, through the replenishing of depleted lymphoid and myeloid populations (Ford et al., 1956; Jacobson et al., 1951; Lorenz et al., 1952). The frequency of HSC in the BM were estimated based on the minimal number of BMC required to protect mice against radiation, a procedure now known as the limiting dilution assay (McCulloch and Till, 1960).  The earliest method of quantifying HSC involved transferring cells intravenously into a host followed by counting the colony-forming units (CFU) in the spleen (Eaves, 2015); however, these spleen- colonizing cells were later found to be only capable of short-term reconstitution and are thus no true HSC (Jones et al., 1990; Ploemacher and Brons, 1989). Various in vitro and in vivo methods have since been developed to assess self-renewal and multipotency of putative HSC, and these assays have been used to verify putative HSC, enabling development of the use of specific surface antigens in the identification of HSC (Eaves, 2015; Perry and Li, 2010).    12    Figure 1.3 Haematopoietic lineage tree The HSCP population includes HSC and MPP. HSC gives rise to MPP, which can in turn give rise to the CMP and CLP. CLP eventually gives rise to the lymphoid populations, including, T cells, B cells, NKT cells, and ILC. The CMP gives rise to the MEP and GMP. MEP eventually generates RBC and platelets, whereas GMP give rise to monocytes, macrophages and granulocytes, including neutrophils, eosinophils, basophils and mast cells. Dendritic cells can arise from both CMP and CLP. HSC, haematopoietic stem cells; MPP, multipotent progenitors; CLP, common lymphoid progenitors; CMP, common myeloid progenitors; MEP, megakaryocyte and erythrocyte progenitors; GMP, granulocyte and macrophage progenitors; NKT; natural killer T; ILC, innate lymphoid cells; DC; dendritic cells. Adapted from (Passegué et al., 2003).   13   The BMC that are only capable of short-term reconstitution also had little or no self-renewal capability, and were termed the MPP (Christensen and Weissman, 2001; Morrison and Weissman, 1994; Yang et al., 2005). These cells can generate all hematopoietic cell types. However, without replenishment MPP can only support hematopoiesis short-term, due to their reduced self-renewal capability (Morrison and Weissman, 1994). MPP arise from HSC (Morrison et al., 1997).   1.3.2 Identifying HSPC by flow cytometry HSC comprise a small percentage of total BM: it is estimated that 0.01% of BM are HSC. HSC are found within the lineage- Sca1+ c-Kit+ (cluster of differentiation (CD) 117) (LSK, also KLS) population (Challen et al., 2009; Okada et al., 1991; Spangrude et al., 1988). Lineage surface markers are used to exclude more mature cells in the BM, such as T and B cells, NK and NKT cells, granulocytes, monocytes, macrophages and DC.  The LSK population includes the most primitive progenitors in the hematopoietic system, including HSC and MPP (Okada et al., 1992).  Additional surface markers can be used to identify HSC from within the LSK population (Fig. 1.4). Unlike their human counterpart, mouse HSC are CD34-/lo, and a single mouse CD34-/lo LSK bone marrow cells (BMC), is able to reconstitute a recipient animal long-term (Osawa et al., 1996). Similarly, Flt3- LSK BMC also confer long-term reconstitution, whereas the Flt3+ LSK BMC only reconstitute short-term and were MPP (Adolfsson et al., 2005; Christensen and Weissman, 2001). Interestingly, HSC are found to have increased ability for drug efflux, resulting in reduced staining by the DNA-binding dye Hoechst 33342. Typical analysis of Hoechst staining displays    14   Hoechst signal in two separate channels, where the HSC populations are in a streak in the side, termed the “side population or SP.” Although on its own, the side population is enriched for HSC, LSK staining is often used in conjunction, and the population is referred to as SPkls or SParKLS (Challen et al., 2009; Goodell et al., 1996; Goodell et al., 1997). More recently, the signalling lymphocyte activation molecule (SLAM) family of receptors, including CD150, CD48 and CD244, are also being used for identification of specific populations within the LSK population. CD150 is expressed by cells with increased reconstitution capacity and thus mark the HSC, whereas MPP are CD150-CD48-CD224+ and restricted progenitors are CD150- CD48+ CD244+ (Fig. 1.4) (Kiel et al., 2005; Yilmaz et al., 2006).   1.3.3 HSPC localization BMC with different relative distances to the center of the femur bone exhibit differences in CFU and proliferation, leading to the idea that HSC and MPP reside in specific and distinct locations within the BM (Lord et al., 1975). HSC transplanted into irradiated hosts localize closely to the endosteal bone marrow surface (Jiang et al., 2009; Xie et al., 2009), and majority of HSC reside closely to the endosteal surface (Kunisaki et al., 2013; Nombela-Arrieta et al., 2013).  The bone endosteum is well vascularized (Fig. 1.5), and HSC identified by the SLAM receptors CD150 and CD48 are enriched near blood vessels and the endosteal region (Kiel et al., 2005). Furthermore, quiescent Ki67- or BrdU- HSC are associated with small arterioles in the endosteum, and move to a peri-sinusoidal niche when induced to proliferate (Kunisaki et al., 2013).     15     Figure 1.4 Identification of hematopoietic stem and progenitor cells in adult bone marrow The loss and gain of cell surface molecules is used to delineate sub-populations of haematopoietic stem cells (HSC) and multipotent progenitors (MPPs). Weissman et al (Adolfsson et al., 2005; Christensen and Weissman, 2001) subdivides the Lineage markers- , Sca-1+, c-Kit+ (LSK) population of bone marrow cells on the basis of expression of Flt3 and Thy1.1. Jacobsen et al (Yang et al., 2005) makes use of CD34 and Flt3 expression. Signalling lymphocyte activation molecule (CD150, CD48, CD229 and CD244) markers are used to subdivide LSK cells in the model proposed by Morrison et al (Kiel et al., 2005; Yilmaz et al., 2006). Flt3, fms-like tyrosine kinase; LTHSC, long-term reconstituting haematopoietic stem cells; ST-HSC, short-term reconstituting haematopoietic stem cells. Adapted from (Brown et al., 2015).   16    Figure 1.5 Morphological construction of bone marrow The bone marrow is contained in the central medullary cavity of bone. The main blood source to the bone marrow is provided by the nutrient artery. The nutrient artery crosses the cortex through the nutrient canal into the medullary cavity, where it divides into ascending and descending arteries, from which radial arteries arise. The radial arteries enter the cortex through the endosteum, which lines the medullary cavity, and become cortical capillaries. Blood from these capillaries can then mix with blood from periosteal capillaries (not shown) and endosteal capillaries. The cortical capillaries enter the medullary vascular sinuses, which form a dense network through the medullary cavity. The sinuses eventually collect and enter the central sinus from which the blood leaves the bone marrow. The medullary vascular sinuses are lined with endothelial cells and surrounded by adventitial reticular cells. Haematopoiesis occurs in the extravascular spaces between the sinuses. Adapted from (Nagasawa, 2006).    17   The close association of HSC to the vasculature is confounded by other studies that suggest HSC reside in hypoxic conditions, evidenced by their high retention of the hypoxia marker pimodizole and HIF-1α expression in situ (Nombela-Arrieta et al., 2013; Parmar et al., 2007). Limited oxygen consumption limits cell cycling, thus allowing HSC to maintain quiescence, and HSC are found to utilize aerobic glycolysis, instead of oxidative phosphorylation (OXPHOS) (Simsek et al., 2010). Furthermore, HSC are found to be equally hypoxic regardless of their proximity to the vasculature structures (Nombela-Arrieta et al., 2013), indicating that the maintenance of hypoxia is in part due to metabolic regulations within the HSC. Alternatively, HSC may not have full access to contents of the blood, despite their proximity to the vasculature. Perfusion of mice with dye functionally measures relative blood flow experienced by the HSC, and reveals that HSC with the greatest serial reconstitution capability experience negligible blood flow (Parmar et al., 2007; Winkler et al., 2010).  The composition and localization of the MPP niche is not well defined. Since MPP eventually gives rise to progenitors, immature and mature cells that exit the BM, it is conceivable that MPP are located closely to the vasculature (Trumpp et al., 2010).   1.3.4 BM niche for HSPC Like all stem cells, HSC have the properties of self-renewal and multipotency. Together, these two properties can be referred to as “stemness.” Proliferation can lead to loss of stemness, as the more proliferative MPP are as potent as HSC in reconstitution but have reduced ability to self-renew. Furthermore, the Hayflick limit postulated that each cell can divide a finite number of    18   times in its lifetime (Allsopp and Weissman, 2002). HSC are thus normally quiescent, and the quiescence is actively maintained during homeostasis by the HSC niche. The HSC niche is composed of a complex array of various factors, cell types and structures (Fig. 1.6), and the exact relationship between many of these components are yet to be elucidated. Furthermore, multiple types of HSC niches may exist in the BM, and HSC may move between these niches depending on their need for quiescence, proliferation or differentiation (reviewed (Trumpp et al., 2010)).  The niche composition and location of the MPP niche is less well defined. Since MPP arise from proliferating HSC, they are likely found in very similar environments, and MPP likely co-localize with factors that support cycling of HSC (Trumpp et al., 2010).  1.3.4.1 Factors for HSC localization, maintenance  The chemokine C-X-C motif ligand (CXCL) 12 (aka stromal cell-derived factor, or SDF-1) calcium ions, stem cell factor (SCF, aka Steel factor), angiopoietin-1, thrombopoietin, osteopontin and their receptors are shown to be important for maintaining HSC through maintaining HSC localization and/ or quiescence.  HSC express the CXCL12 receptor (CXCR 4), and specifically migrate towards CXCL12 but not towards other chemokines (Wright et al., 2002). Conditional deletion of CXCL12 or CXCR 4 in adult mice reduces HSC numbers and reconstitution by transferred BMC, and BM HSC colonization is impaired in mice deficient in CXCL12 (Ara et al., 2003; Sugiyama et al., 2006; Zou et al., 1998).     19    Figure 1.6 Key cell types involved in regulating HSC maintenance Various cell types have been implicated for their roles in promoting HSC maintenance, including perivascular stromal cells, endothelial cells (ECs), macrophages and CAR cells. Regulators of HSC maintenance include CXCL12, SCF, and angiopoietin 1 (ANGPT1) and signalling pathways including Notch and Wnt. Recent studies have suggested that osteoblasts are dispensable for HSC maintenance but may have a role in regulating lymphoid progenitor cells. Certain cell types have been shown to negatively affect HSC maintenance, including adipocytes, which are increasingly present after chemotherapy and radiation. Regional localization of HSCs during quiescence and after activation revealed that quiescent HSCs associate with arterioles ensheathed with NG2+ pericytes but after activation relocate near the Lepr-expressing perisinusoidal area. Adapted from (Mendelson and Frenette, 2014).   20   Calcium ions also appear to be important for HSC localization, as deletion of its sensing receptor, calcium-sensing receptor, or CaR,  results in reduced BMC and HSC numbers and increased HSPC in circulation and in the spleen (Adams et al., 2006).  Mice homozygous for the SCF receptor c-Kit mutations exhibit mild to severe developmental defects, where some mutations result in neonatal mortality (Reith et al., 1990; Russell and Lawson, 1959). Viable c-Kit mutation and mutations in SCF both result in reduction in HSC numbers and HSC self-renewal (Barker, 1994; Thorén et al., 2008). SCF and c-Kit promote HSC survival through Bcl2 (Thorén et al., 2008).  Angiopoietin-1 and its receptor Tie2 promote HSC quiescence. Tie2 expression is enriched in the SPkls population, and combined deficiency in Tie1 and Tie2 results in defects in postnatal BM hematopoiesis (Arai et al., 2004; Puri and Bernstein, 2003). Furthermore, treatment of BMC with angiopoietin-1 or ectopic expression of angiopoietin-1 in BMC enhances their reconstitution ability (Arai et al., 2004). Thrombopoietin and its receptor c-Mpl are required for adult hematopoeisis, and deletions of either thrombopoietin or c-Mpl results in reduction in HSC numbers in adult mice (Kimura et al., 1998; Qian et al., 2007). BMC deficient in c-Mpl are less able to reconstitute irradiated hosts, while BM recipients deficient in thrombopoietin require more BMC for reconstitution (Fox et al., 2002; Kimura et al., 1998). Thrombopoietin promotes HSC quiescence by inhibiting their entry into cell cycle (Qian et al., 2007; Yoshihara et al., 2007).    21   Osteopontin also restricts HSC proliferation, as osteopontin-deficient mice have increased HSC numbers, and exogenous osteopontin suppresses HSC proliferation in vitro (Nilsson et al., 2005; Stier et al., 2005).   1.3.4.2 Factors for HSPC proliferation and differentiation Factors that directly promote HSC proliferation and differentiation are not as well characterized, since HSC can be induced to proliferate indirectly in several different scenarios of hematopoietic stress (King and Goodell, 2011). When in combination with SCF or thrombopoietin, interleukin (IL) -3 and IL-6 promote the survival, proliferation and differentiation of HSC (Seita and Weissman, 2010). Systemic bacterial infection also induces proliferation of HSPC, and this is possibly mediated through IFNγ (Baldridge et al., 2010; MacNamara et al., 2011). However, studies involving IFNγ-deficient mice yield conflicting results regarding the role of IFNγ in inducing HSC proliferation (Baldridge et al., 2010; de Bruin et al., 2013). Notch and Wnt signalling pathways have also been implicated in regulating HSC proliferation by several studies, but the exact nature and the extent of this regulation is unclear (reviewed in (Mendelson and Frenette, 2014)).  1.3.5 HSPC niche and HSPC function Long-term HSC are dormant and are estimated to divide approximately 5 times in a mouse’s life (Wilson et al., 2008). Computational modeling reveals that HSC can exit and re-enter dormancy, allowing them to enter cell cycle in response to increased need for hematopoietic cells without exhausting themselves (Wilson et al., 2008). It is speculated that this exit from and re-entry into dormancy is in part dictated by the type of niche the HSC occupies (Trumpp et al., 2010).     22   In this model, quiescent HSC normally reside in close proximity to cells that produce factors promoting HSC dormancy during homeostasis. Periodically, few select HSC exit the dormant niche to proliferate and differentiate into MPP. This may be mediated by asymmetric division, if the newly generated daughter HSC is pushed out of the fully occupied niche (Ting et al., 2012). During injury or infection, in response to drastic loss of hematopoietic cells, HSC can become activated and exit dormancy to expand and differentiate in order to replenish lost cells. In addition, hematopoietic progenitors can exit the BM and differentiate in the spleen and liver. After the peripheral hematopoietic cells are replenished, the signal for increased hematopoiesis also decreases, and activated HSC either differentiate or return to the dormant niche and become quiescent. (King and Goodell, 2011; Mendelson and Frenette, 2014; Trumpp et al., 2010)  1.3.6 CD44 and HA in hematopoiesis  The BM ECM is primarily comprised of collagen, fibronectin, and laminin, whose roles in BM are well understood (reviewed in (Klamer and Voermans, 2014)). The bone is a major site of HA synthesis, and a quarter of the total HA in the body can be found in the skeletons (Reed et al., 1988). However, as there is no clear structural role for HA in the bone, the role of HA in the BM has not been extensively investigated. Both osteoblasts and osteoclasts, which reside at the bone-BM junction, can synthesize HA (Bastow et al., 2007). HA deposition is concentrated at the endosteum and marrow sinusoidal, but not arteriolar endothelium, the postulated locations of quiescent HSC (Avigdor et al., 2004; Genever and Dickson, 1996).  BM HA appears to play a role in maintenance of BMC, as injection with exogenous HA improved cell number recovery post 5-fluorouracil-mediated cell depletion (Matrosova et al.,    23   2004). In Has1/Has3 double knockout mice with mesenchyme-specific prx-cre Has2 deletion, the absence of HA in the BM environment reduces the ability for BM stromal cells to support and maintain HSC, thereby directly demonstrating the importance of HA in influencing hematopoiesis (Goncharova et al., 2012). The interaction between CD44 and HA also aided in HSC localization in the BM. HA-binding lineage- BMC are more able to reconstitute lethally irradiated mice (Nilsson et al., 2003), and incubation with an anti-CD44 monoclonal antibody or soluble HA prior to BM transfer reduces homing and long-term reconstitution of BMC (Avigdor et al., 2004).   1.4 CD8 T lymphocytes T cells are an important part of the adaptive immunity, as they are involved in antigen-specific cell-mediated immunity.  T cell development occurs in the thymus. Most thymocytes are selected for their expression of the T cell receptor (TCR) αβ heterodimer, and then lose the expression of either CD4 or CD8, thus resulting in two populations of naïve T cells – CD4 T cells and CD8 T cells. Activated CD8 T cells differentiate and are referred to as cytotoxic T lymphocytes or killer T cells, and directly kill infected target cells presenting specific peptide antigen.  1.4.1 CD8 T Cell activation  Each individual naïve T cell exiting the thymus expresses a distinct TCR. T cells are activated when they encounter their cognate antigen in the correct context, and three signals are required to initiate a T cell response. The interaction between the TCR and MHC is referred to as “signal 1.” CD4 and CD8 T cells recognize and bind to their cognate antigen in the context of major    24   histocompatibility complex (MHC) class II and class I, respectively. Naïve T cells circulate between lymphatic organs, and interact with the antigen-presenting cells (APC) in these organs in search of their cognate peptide. These interactions can be visualized, and are short (Mempel et al., 2004; Stoll et al., 2002). When a T cell finds a DC that presents its cognate antigen in the context of the correct MHC, the T cell stops, and establishes a longer contact with the DC (Mempel et al., 2004; Stoll et al., 2002). The interaction between T cells and APC via the co-stimulatory molecules is referred to as “signal 2,” and is also required for T cell activation. In the absence of signal 2, T cells that encounter “signal 1” become anergic, and will fail to respond to its cognate antigen (Dure and Macian, 2009). CD28 is the most important co-stimulatory molecule on T cells; upon interaction with CD80 and CD86 on the APC, CD28 signals through the phosphoinositide 3 kinase (PI3K)/ Akt pathway and promotes cell proliferation (Frauwirth et al., 2002; Jacobs et al., 2008; Rathmell et al., 2003).  “Signal 3” is comprised of cytokines that shape the T cell response. In particular, IL-12 is required for CD8 T cell differentiation and effector function. In the absence of IL-12, activated CD8 T cells exhibit reduced cytolytic activity, despite having undergone similar proliferation (Curtsinger et al., 2005; Curtsinger et al., 2003). IL-12 promotes CD8 effector T cell differentiation, as IL-12 signalling enhances the expression of the T-box transcription factor T-bet and inhibits eomesodermin expression (Takemoto et al., 2006). Type I interferons (IFN), including IFNα and IFNβ, have been shown to be important for CD8 effector functions, such as cytolytic activity and IFNγ production (Curtsinger et al., 2005). Injection of IFNα enhances expansion of antigen-specific CD8 T cells, and type I IFN receptor deficient CD8 T cells exhibit    25   diminished cell numbers due to reduced survival, while proliferation is unaffected by type I IFN signalling (Kolumam et al., 2005; Le Bon et al., 2006; Thompson et al., 2006).   1.4.2 CD8 effector T cells CD8 T cells can directly kill infected target cells by one of two ways: release of perforin and granzymes, and Fas-mediated killing (Barry and Bleackley, 2002).  Both mechanisms are specifically targeted and occur when activated CD8 T cells form an immune synapse upon recognizing its cognate antigen presented by MHC I on the target cell. Perforin and granzymes are stored in cytotoxic granules in effector CD8 T cells. They are released into the immune synapse formed with the target cell. Perforin inserts into the cell membrane of target cells and oligomerizes in the cell membrane to form a pore, allowing granzymes to passively diffuse into the target cell (Chowdhury and Lieberman, 2008). Effector T cells can also trigger apoptosis in target cells via Fas-Fas ligand interaction, triggering apoptosis of the cell (Barry and Bleackley, 2002).   Effector CD8 T cells also secrete various cytokines, including IL-2, IFNγ, TNFα and MIP1β.  Activated CD8 T cells rely on autocrine and/or paracrine IL-2 signaling for clonal expansion (Malek and Castro, 2010). IL-2 also promote the production of effector proteins, such as perforin and granzyme B, as well as IFNγ (Pipkin et al., 2010). IFNγ and TNFα is major cytokines produced by CD8 T cells, and have various pro-inflammatory effects, including activation of macrophages (Mosser and Edwards, 2008). Cells that produce multiple cytokines are referred to as being poly-functional, and CD8 T cells with higher antigen sensitivity are more likely to    26   become poly-functional, which in turn, allows them to be more suppressive (Almeida et al., 2009).  Based on the expressions of CD127 and KLRG1, effector T cells can be divided into CD127- KLRG1+ short-lived effector cells (SLEC) and CD127+KLRG1- memory-phenotyped effector cells (MPEC) (Joshi et al., 2007). Despite having similar expressions of IL-12, Gramzyme B, and CD62L, MPEC are intrinsically more skewed towards memory formation than SLEC (Sarkar et al., 2008). Reduced expression of T-bet leads to increased MEPC and memory formation (Joshi et al., 2007), indicating that T-bet drives the terminal differentiation of effector CD8 T cells. T-bet expression is induced downstream of TCR and IFNγ signalling (Afkarian et al., 2002; Lighvani et al., 2001), but can also be incrementally increased with subsequent exposure to IL-12 (Takemoto et al., 2006).  1.4.3 Memory cell formation After clearance of the antigen and resolution of inflammation, 90 to 95 % of the effector T cells die, leaving a small population to persist as memory cells. Memory T cells confer improved protection upon re-encountering the same pathogen, due to their quicker response, leading to faster pathogen clearance and reduced immunopathology. Together with memory B cells, they form immunological memory and are the basis for vaccine function.  1.4.3.1 Cytokines The common γ cytokines IL-7 and IL-15 are important to the formation and maintenance of memory CD8 T cells. IL-7 is constitutively produced by fibroblastic reticular cells in the T cell    27   zones of LN and by stromal cells in the BM (Link et al., 2007; Mazzucchelli et al., 2009). IL-7 regulates homeostasis of T cells (Maraskovsky et al., 1996; Tan et al., 2001), and its heterodimeric receptor includes CD127 (aka IL-7 receptor α) and CD132 (the common γ chain). Naïve and memory T cells express high levels of CD127 (Alves et al., 2008; Paiardini et al., 2005; Xue et al., 2002). After activation, CD8 T cells up-regulate expression of CD25, the receptor subunit specific for IL-2, another member of the common γ cytokine family, and IL-2 signalling downregulates expression of CD127 (Xue et al., 2002). A subset of CD8 effector T cells expresses CD127 and is more likely to form memory cells after the contraction phase (Joshi et al., 2007; Kaech et al., 2003). CD127 expression is required for memory formation; however ectopic expression of CD127 does not prevent death during the contraction phase (Hand et al., 2007; Haring et al., 2008).  IL-15 promotes the proliferation and survival of CD8 memory cells; in the absence of IL-15 signalling, CD8 T memory cell formation and maintenance is impaired (Schluns et al., 2002). The effects of IL-15 seem to be concentration-dependent, promoting proliferation at high concentration and survival at low (Berard et al., 2003). The IL-15 receptor is a heterotrimer of the IL-15 receptor α subunit (IL-15Rα), CD122 (IL-2 receptor β subunit) and CD132, where the latter two can also be part of the IL-2 receptor heterotrimer. CD122 and CD132 are necessary for the IL-15 mediated proliferation and survival (Berard et al., 2003). IL-15Rα expression is low on naïve T cells, and is up-regulated on effector and memory T cells (Schluns et al., 2002). While IL-15Rα was not essential for IL-15 mediated proliferation or survival on CD8 T cells, the receptor subunit increases the sensitivity of cells to low concentrations of IL-15 (Berard et al., 2003; Burkett et al., 2003). IL-15Rα can complex with IL-15, and this complex induces greater    28   proliferation than IL-15 alone when presented to CD8 T cells in trans by neighbouring cells (Dubois et al., 2002; Rubinstein et al., 2006; Stoklasek et al., 2006).   1.4.3.2 Metabolism Adenosine triphosphate (ATP) is the main energy currency of cells, and it is generated through two main pathways: glycolysis and oxidative phosphorylation (OXPHOS). Mammalian glycolysis occurs in the cytosol through the Embden–Meyerhof–Parnas pathway, and converts glucose to pyruvate, generating two ATP molecules from one glucose molecule. Additional energy potential is released from pyruvate through oxidation, primary by the tricarboxylic acid cycle (TCA or Kreb) cycle. To enter the TCA cycle, pyruvate is oxidized and conjugated to acetyl-coenzyme A (CoA). The TCA cycle occurs in the mitochondrial matrix, and interconverts between several biosynthetic precursors and completely oxidizes remaining carbons into carbon dioxide, and generates one guanosine triphosphate (GTP) per pyruvate molecule. GTP can be readily converted into ATP by the nucleoside-diphosphate kinase (Fig. 1.7). The most efficient ATP synthesis occurs through OXPHOS, which occurs through a series of redox reactions in the inner mitochondrial membrane and transports protons from the matrix into the inter-membrane space, thereby generating a proton gradient, which is then collapsed by the ATP synthase. The ATP synthase utilizes the proton motive force to generate ATP (Fig. 1.8). By fully oxidizing 2 pyruvate molecules, OXPHOS can generate more than 30 ATP molecules. The process of OXPHOS requires oxygen, reduced nicotinamide adenine dinucleotide (NADH), and the hydroquinone form of flavin adenine dinucleotide (FADH2). A single glucose molecule    29    Figure 1.7 Overall reactions in glycolysis, pyruvate oxidation and the TCA cycle Glycolysis converts one glucose to two pyruvate and generates 2 ATP and 2 NADH. Through pyruvate oxidation, a single carbon in each pyruvate is oxidized to carbon dioxide, and the remaining acetyl group is conjugated to Co A, forming acetyl-CoA, and generating one NADH per pyruvate. The acetyl group is then able to enter the TCA cycle, where the remaining carbons eventually become fully oxidized to carbon dioxide, generating one GTP, three NADH, 1 FADH2 per acetyl-CoA. GTP is readily converted to ATP.    30    Figure 1.8 Oxidative phosphorylation The complexes of OXPHOS are embedded in the inner mitochondrial membrane. NADH and FADH2 are generated from the TCA cycle and donate electrons to complexes I and II, respectively. The electrons are then sequentially transferred to ubiquinone (coenzyme Q or CoQ, Q), complex III, cytochrome C (Cyt C) and then complex IV, which donates an electron to oxygen to generate water. The flow of electrons generates energy, which enables complexes I, III and IV to transport protons into inter-membrane space from the matrix. The resulting proton gradient creates a mitochondrial membrane potential. The ATP synthase then couples the passive transport of protons back into the matrix with charging an inorganic phosphate (Pi) onto an adenosine diphosphate (ADP), creating an ATP. An anti-porter then releases ATP from the mitochondria in exchange for cytosolic ADP. Adapted from (Yusoff et al., 2015).   31   generates two NADH through glycolysis, and six NADH and two FADH2 through the TCA cycle (Fig. 1.8), making the latter that main driver of OXPHOS. Activated T cells divide rapidly, and this rapid division requires two things: increased ATP production, and production of anabolic substrates for generating macromolecules. Upon activation, T cells up-regulate the surface expression of glucose transporter 1 (Glut1) (Wieman et al., 2007). The regulation of Glut1 is multi-faceted: CD28 mediates the trafficking of Glut1 from vesicles to the cell surface, and IL-2 increases the expression of Glut1 and other glycolytic enzymes through activation of PI3K/ Akt (Frauwirth et al., 2002; Jacobs et al., 2008; Wieman et al., 2007). Effector T cells have increased glycolytic output, and depend on glucose for survival and proliferation, even in the presence of an alternative carbon source, such as glutamine (Greiner et al., 1994; Jacobs et al., 2008). Glucose is also essential for CD8 effector T cell functions, as glucose deprivation results in reduced IFNγ transcript expression, reduced granzyme and perforin production and reduced cytolytic activity (Cham et al., 2008; Cham and Gajewski, 2005). During the contraction phase of the T cell response, IL-2 and other cytokines become limiting. This results in inactivation of the PI3K/Akt pathway, which causes internalization of Glut1 and amino acid transporters, thereby limiting the amount of available substrates for ATP generation and anabolism (Edinger and Thompson, 2002; Finlay et al., 2012; Wieman et al., 2007). Compared to effector T cells, memory T cells exhibit increased OXPHOS and reduced glycolysis, and inhibition of OXPHOS has no effect on effector generation but results in reduced memory pool (O’Sullivan et al., 2014; van der Windt et al., 2012). The increase in OXPHOS is mediated by mitochondrial fusion, and deletion of the fusion protein Opa1 reduces memory    32   generation (Buck et al., 2016). CD8 memory T cells require extracellular glucose to generate fatty acids, which are then oxidized through OXPHOS to generate ATP. The ATP is then used for fatty acid synthesis. This cycle of fatty acid synthesis and oxidation yields no significant net gain of ATP (O’Sullivan et al., 2014).  1.4.4 Memory and subsequent responses Memory CD8 T cells can be divided into two subsets based on their expression of CC chemokine receptor (CCR) 7 and CD62L (L-selectin), both of which mediate homing to LN. Central memory T cells (Tcm) express high CCR7 and CD62L, while effector memory T cells (Tem) express low levels of both (Sallusto et al., 1999). The expression of CCR7 and CD62L enable Tcm trafficking into lymphoid organs, whereas Tem is enriched at the effector site (Jameson and Hamilton, 2012). During subsequent responses, Tem primarily produce effector cytokines, while Tcm readily make IL-2 and expand (Sallusto et al., 2004).   1.4.5 Models of CD8 T memory cell formation A single antigen-specific naïve CD8 T cell can give rise to different effector and memory subsets upon activation, suggesting that T cell fate is shaped by the T cell response and is not innate to individual naïve CD8 T cells (Stemberger et al., 2007). Several models of CD8 T cell fate determination exist (Ahmed et al., 2009; Kaech and Cui, 2012). The textbook view that memory cells arise from effector cells, where a clone that dominates the effector population also dominates the memory pool, is contradicted by recent studies (reviewed in (Ahmed et al., 2009; Kaech and Cui, 2012)).    33   Paired CD8 T daughter cells are shown to have asymmetric segregation of various markers, where the cell closer to the APC (the proximal daughter) display characteristics with effector function and activation, such as increased expression of IFNγ, granzyme B, CD25, CD44 and CD69 and reduced expression of CD62L (Chang et al., 2007). When these two daughter populations were separately transferred into secondary hosts, both were equally able to protect against immediate infection but only the distal daughter cell population was able to confer protection upon secondary challenge 30 days later (Chang et al., 2007). This study suggests that the fates of CD8 T cells are determined based on their proximity to the APC during their first division. The difference in their fates is facilitated by the asymmetric partitioning of proteasomes, which leads to differential expression of T-bet due to different levels in T-bet degradation (Chang et al., 2011). The mTOR complex 1 (mTORC1) is also asymmetrically segregated during the first division, resulting in the proximal daughter cell population with increased glycolysis and expression of effector molecules, while the distal daughter cell population has increased lipid metabolism, expression of anti-apoptotic molecules and increased memory potential (Pollizzi et al., 2016). Single cell gene expression analysis using machine learning algorithms further supports this model of bifurcated differentiation, where a single activated CD8 T cell gives rise to both effector and memory populations (Arsenio et al., 2014). However, the computational analysis and the conclusion from this study have been called into question (Flossdorf et al., 2015). While the evidence in support of the asymmetric division model arises primarily from a single research group, other recent studies from various groups argue against this model. Utilizing single cell tracking and transfer, individual naïve CD8 T cells are found to possess different    34   differentiation potentials and give rise to different ratios of effector and memory pools (Buchholz et al., 2013; Gerlach et al., 2013; Plumlee et al., 2013). Furthermore, the effector clones that have proliferated the most have reduced representation in the memory population, suggesting that differentiation of effector CD8 T cells results in reduced memory potential (Buchholz et al., 2013; Gerlach et al., 2013). Furthermore, IL-12 and T-bet, which are required for differentiation of effector CD8 cells, promote formation of SLEC (Joshi et al., 2007; Kaech and Cui, 2012; Lazarevic et al., 2013). Dampening of inflammation by clearing the infection with antibiotics or antivirals also results in reduction of the SLEC population and accelerated memory CD8 T cell development, despite exhibiting similar expansion (Badovinac and Harty, 2007; Badovinac et al., 2004; Fousteri et al., 2011). CD8 T cells that are activated later in infection also have increased longevity in the memory phase (D'Souza and Hedrick, 2006; Fousteri et al., 2011). The memory potential of individual CD8 effector cell appears to be negatively regulated by the amount of combined activation signals they experience, where increased differentiation and proliferation lead to loss of memory potential. Similarly, the strength of activation signal of CD8 T cells is also suggested to have similar effects on regulating CD8 T cell fate, although no direct evidence exists (Kaech and Cui, 2012). According to this model, CD8 T cells follow a linear path of differentiation, where memory precursor cells give rise to terminally differentiated effector cells, a progression supported by transcriptional profiling of naïve, effector and memory CD8 T cells (Best et al., 2013; Holmes et al., 2005; Kaech et al., 2002).  1.5 CD44 and HA in CD8 T cell functions Unlike other immune cells which develop in the bone marrow, T cell development occurs in the thymus through the process of thymopoiesis, where thymocytes undergo massive proliferation    35   and then selection to give rise to mature naïve T cells, which are CD44lo. CD44 expression is increased in memory phenotype T cells, which arise from homeostatic proliferation (Yamada et al., 2001). Several studies have investigated the role of HA binding at various stages of the T cell response. Exposure to HA during activation in vitro increases proliferation in peripheral human T cells (Galandrini et al., 1994). Furthermore, since DC actively synthesize and express HA on the cell surface, it is suggested that HA-binding by T cells mediates the T-APC interaction during T cell activation (Mahaffey and Mummert, 2007; Mummert et al., 2002). Despite these findings, HA-binding by T cells is unlikely to participate in T cell activation in physiological conditions, as CD44 expression in T cells is only up-regulated after activation and HA-binding by T cells is not detectable until at least 24 hour post TCR-stimulation (Maeshima et al., 2011). Earlier studies investigating the role of HA-binding utilize mostly human peripheral blood and murine T lymphocyte cell lines, focusing on the effects of HA on polarization, adhesion, rolling, and migration. Chemokine stimulation or contact with human umbilical vein endothelial cells polarizes T cells, and recruits CD44 to the uropod of the cell (del Pozo et al., 1995; Rosenman et al., 1993). Moreover, stimulation with CD44-specific monoclonal antibodies causes T cell polarization though protein kinase C (Fanning et al., 2005). T cells also bind to endothelial cells or fibroblasts through the interaction between CD44 and HA, and this interaction is reduced upon treatment with hyaluronidases or HA-blocking CD44-specific antibodies (Evanko et al., 2012; Murakami et al., 1996). CD44 also mediates rolling of T lymphocyte cell lines on endothelial cells and HA-coated surfaces under shear flow conditions (Clark et al., 1996; DeGrendele et al., 1996; Mohamadzadeh et al., 1998; Nandi et al., 2000). Lastly, T cell lines are better able to migrate through a three dimensional lattice that is embedded with HMW HA    36   (Fanning et al., 2005), and there is a trend for CD44-deficient T cell lines to migrate better in murine hosts (Nandi et al., 2004). Since cell lines and in vitro system were used in earlier studies mentioned above for examining the function of CD44-mediated HA-binding, the physiological relevance of such findings is unclear. To address this issue, more recent studies make use of ex vivo human or murine T cells or in vivo murine disease models. Such studies confirm the role of CD44 and HA-binding in migration. For example, addition of HA increases ex vivo human peripheral T cell migration through collagen (Evanko et al., 2012), and CD44 mediates rolling of activated murine CD4 T cells under flow conditions (Bonder et al., 2006). In a model of atherosclerosis, CD44-deficient T cells are found in reduced numbers in lesions and have reduced mRNA expression of the chemokines receptor ccr5 (Sjöberg et al., 2009). In response to IL-15, endothelial cells up-regulate HA synthesis and facilitate the migration of superantigen-stimulated LN T cells into the peritoneum in a CD44-dependent manner. Also, CD44-deficient leukocytes, including T cells, are less able to migrate to the skin in a model of atopic dermatitis (Gonda et al., 2005). Another mechanism for restricting the activated effector T cell population is through induced cell death of the T cells. Repeated stimulation of T cells leads to activation-induced cell death (AICD). Several studies have shown that CD44 is involved in mediating AICD. CD44-deficient T cells exhibit increased resistance to AICD in vitro and in vivo (Chen et al., 2001; Katoh et al., 2003). Expression of specific CD44 variant exons seems to affect AICD differently;  Th1 CD4 T cells deficient in CD44v7 and CD44v10 have reduced Bcl2 expression and are more apoptotic in the murine EAE model (Ghazi-Visser et al., 2013), while the Jurkat T cell line expressing cd44v2-10 are more protected from Fas-mediated apoptosis (Mielgo et al., 2005). Furthermore,    37   treatment with anti-CD44 antibody that targets the HA-binding pocket increases surface FasL expression of human peripheral T cells, and the addition of hyaluronidase decreases AICD by the T cells (Nakano et al., 2007). Addition of HMW HA also enhances AICD in the Jurkat T cell line and murine splenic T cells (Ruffell and Johnson, 2008). CD44-deficient CD4 T cells have a reduced ability to form memory cells post clearance of an influenza infection, independent of the Fas-FasL pathway (Baaten et al., 2010). These studies suggest that the interaction between HA and CD44 positively regulates AICD. However, other studies suggest also that CD44 and its variant exon 7-containing isoform confer resistance to apoptosis in T cells (Marhaba et al., 2006; Wittig et al., 2000).  1.6 Molecular regulation of CD44-mediated HA binding  The Cd44 locus contains 20 exons – 10 invariant exons and 10 variant exons (exons v1-11 in human, and exons v1-10 in mice). The most common form of CD44, termed standard or hematopoietic CD44 (CD44s or CD44H, respectively), consists solely of the invariant exons and is the main isoform expressed by hematopoietic cells (Ponta et al., 2003). The protein includes an amino terminal HA-binding domain, a variable stem region, a single-pass transmembrane domain and carboxyl terminal cytoplasmic tail with multiple phosphorylation sites. The stem region is extensively glycosylated and the inclusion of variable exons through alternative mRNA splicing can introduce additional glycosylation sites (Ponta et al., 2003). CD44 can also be decorated with heparan sulfate and chondroitin sulphate in the HA-binding domain and the stem region (Ponta et al., 2003). All of these potential post-translational modifications make CD44 highly variable in both molecular weight and HA-binding capacity.     38   1.7 Thesis rationale and hypotheses This study aims to determine if CD44 and HA play a role in the development of specific hematopoietic populations, and in CD8 effector T cell decision between death and memory during T cell contraction. I hypothesize that CD44 expression and HA binding are tightly regulated in the hematopoietic system, where HA binding is induced at specific stages to enable cell-HA interaction. The first aim of my project is to determine the mechanism of HA induction, focusing on the modifications of CD44. I hypothesize that different immune cell types utilize similar mechanisms for regulating HA binding. For this aim, I will be looking at macrophages and T cells. The second aim is to determine if the expression of CD44 affects CD8 T cell memory formation. CD44-mediated HA binding is found on the most actively proliferating CD8 effector T cells (Maeshima et al., 2011). As proliferation and differentiation reduce the memory potential of CD8 effector T cells (Buchholz et al., 2013; Gerlach et al., 2013), I hypothesize that CD44-expressing, HA-binding effector CD8 T cells to be under-represented in the memory population. I further hypothesize that HA-binding CD8 T cells are more likely to die during the contraction phase of the T cell response, since HA is shown to induce AICD in activated T cells (Ruffell and Johnson, 2008). The third aim of this study is to determine if HA binding plays a role in the development of specific immune populations. I will first characterize HA binding by various immune populations, and then perform competitive BM transfer experiments to assess if the reconstitution of these immune populations is affected by the deletion of CD44 or altered HA    39   binding in the donor BMC. I hypothesize that HA binding is specifically up-regulated during hematopoietic development to enable cell-niche interaction, and the ability to bind HA would confer a competitive advantage to the transferred BMC.   40   Chapter 2: Materials and Methods  2.1 Mice C57BL/6 mice (CD45.2), B6.SJL-Ptprca Pep3 mice congenic for CD45.1, and OT-I mice heterozygous for the transgenic TCR specific for the SIINFEKL peptide, OVA257–264, in the context of H2-Kb on a C57BL/6 background were obtained from Jackson Laboratories. OT-I mice on a CD45.1/CD45.2 background were generated by a crossing with OT-I to B6.SJL-Ptprca Pep3. The OT-I transgene was maintained heterozygous by crossing animals heterozygous for the transgene with non-transgenic animals; offspring were then genotyped by PCR. CD44-/- mice (Schmits et al., 1997) were backcrossed onto C57BL/6 for nine generations, and then crossed with OT-I to generate CD44-/- OT-I transgenic mice. All mice were bred and maintained in specific pathogen-free conditions at the Centre for Disease Modeling or Westbrook Animal Facility at the University of British Columbia (UBC). Animal experiments were performed in accordance with the protocols approved by the University Animal Care Committee and Canadian Council of Animal Care guidelines. Mice were used between 6 and 9 weeks of age at the start of experiments and were matched for sex and age for comparative studies.  2.2 Antibodies and reagents The following monoclonal antibodies specific for mouse antigens were used for flow cytometry: CD4 (GK1.5), CD8α (53-6.7), CD11b (M1/70), CD11c (N418), CD19 (1D3), CD45RB/B220 (RA3 6B2), CD45.1 (A20), CD45.2 (104), CD49b (DX5), CD62L (MEL-14), CD122 (DNT15Ra), CD127 (A7R34), CD150 (mSHAD150), F4/80 (BM8), IFNγ (XMG1.2), Ly6G/ Gr1 (RB6-8C5), KLRG1 (2F1), T-bet (eBio4B10), MHC II (M5/114.15.2), NK1.1 (PK136),    41   pAkt (SDRNR), Sca1 (D7), TCRβ (H57-597), TCR Vα2 (B20.1), TCR Vβ5.1/5.2 (MR9-4) γδTCR (eBioGL3) and Ter119 (TER-119) conjugated to biotin, PE, PE-Cy5, PE-Cy5.5, PE-Cy7, PerCP-Cy5.5, allophycocyanin or Alexa Fluor® 700, Alexa Fluor® 405 (eBioscience); CD11b (M1/70), CD45RB/B220 (RA3 6B2), CD45.2 (104), CD117/c-Kit (2B8) or siglec F (E50-2440) conjugated to PE, PE0CF594, PECy7 or allophycocyanin (BD Biosciences); CD44 (IM7.8) conjugated to Alexa Fluor® 647 or Pacific Blue (Biomedical Research Centre Antibody Lab [BRC AbLab] at UBC).  Streptavidin conjugated to PECy5, PECy7 or Allophycocyanin-eFluor® 780 (eBioscience) were used to detect labelling by biotinylated antibodies. 4',6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich) or the LIVE/DEAD® Fixable Aqua Dead Cell Stain Kit (Molecular Probes) were used to stain dead cells. Rooster comb HA was conjugated to fluorescein (FL-HA) according to (de Belder and Wik, 1975) and or HA conjugated to Alexa Flour® 647 (APC-HA) (BRC AbLab) were used to detect HA binding. Tissue culture grade high glucose DMEM, RPMI, fetal calf serum (FCS), 1M HEPES buffer, 100X non-essential amino acids, 55 mM 2-mercaptoethanol and penicillin-streptomycin were purchased from Thermo Fisher Scientific. Recombinant mouse cytokines SCF, L IL-3 and 10 IL-6 were purchased from Cedarlane (Burlington, NC). NaCl, KCl, Na2PO4, KH2PO4, NH4Cl, Tris, HCl, ethylenediaminetetraacetic acid (EDTA), sodium pyruvate, L-glutamine were purchased from Sigma-Aldrich.  2.3 Buffers, solutions and growth media PBS 1.37M NaCl, 27 mM KCl, 108 mM Na2PO4, 14.7 mM KH2PO4     42   RBC lysis buffer 0.84% NH4Cl in 10 mM Tris-HCl, pH 7.25  flow analysis buffer 1% FCS, 2mM EDTA  in PBS  complete DMEM high glucose DMEM  with 15% FCS, and 50 U/ ml penicillin/ streptomycin, 1 mM sodium pyruvate, and 2 mM L-glutamine 3/6/SCF complete DMEM with 100 ng/ mL SCF, 6 ng/ mL IL-3 and 10 ng/ mL IL-6  complete RPMI RPMI-1640 medium with 10% FCS, 50 U/ ml penicillin/ streptomycin, 10 mM HEPES, 1X non-essential amino acids, 55 µM β-mercaptoethanol, 1 mM sodium pyruvate, and 2 mM L-glutamine   2.4 Cell isolation  Single cell suspensions from peripheral LN (pLN), spleen and thymus were generated by passing the organs through a 70 µm filter membrane. If analysis of the myeloid population was required, spleens and pLN were first minced and incubated with 1 mg/ mL collagenase IV (Worthington) for 20 min, prior to passing through the membrane. Blood was collected from cardiac puncture of recently euthanized mice; BM cells were flushed with flow analysis buffer from the tibia and femur of mice using a 26 ½ gauge needle, and then passed through the needle to make single cell suspensions. Liver lymphocytes were isolated from perfused liver using Lympholyte®-M (Cedarlane), following the manufacturer’s protocol. Cell pellets from these organs were resuspended in 5 mL RBC lysis buffer for 5 min at room temperature. Equal volume of flow analysis buffer, complete DMEM or complete RPMI, was then added to dilute the RBC lysis buffer.     43   2.5 Retroviral transduction CD44 constructs with bicistronically expressed GFP or red fluorescent protein (RFP) reporter genes, and the retrovirus packaging cell lines were created as in (Ruffell et al., 2011). Packaging cell line E+GP86 were cultured in complete DMEM (Markowitz et al., 1988; Markowitz et al., 1990).  BM donor mice were injected intravenously with 25 mg/ mL of 5-fluorouracil (Sigma-Aldrich) dissolved in 1X PBS. Four days later, bones were extracted from euthanized mice and BMC were isolated and then cultured in 3/6/SCF medium at 1-1.5x106 cells/ mL for two days. Cells were then co-cultured for two additional days in fresh 3/6/SCF medium with packaging cells that were treated with 10 µg/ mL of mitomycin C (Sigma-Aldrich) a day prior.  2.6 Generation of BM-derived macrophages BM-derived macrophages (BMDM) were generated from BM isolated from the tibia and femurs of mice. After removal of red blood cells by lysis as described above, BMC were plated to 1-2 x 107 cells in a 10 cm2 Petri dish in complete DMEM with 5-10 % of L929 cell-conditioned medium as a source of macrophage colony-stimulating factor. Four days later, cells were re-plated at a 1:4 dilution and stimulated with 10 ng/ mL mouse recombinant IL-4 (eBioscience), or with 10 ng/ml mouse recombinant IFNγ (eBioscience) and 100 ng/mL ultrapure lipopolysaccharide (LPS) (InvivoGen) for 2 days to generate alternatively and classically activated BMDM respectively.      44   2.7 CD8 T cell isolation and activation in vitro CD8 T cells were isolated as described previously (Maeshima et al., 2011). Briefly, after RBC lysis, single cell suspensions of splenocytes were labelled with biotinylated antibodies against CD4 (GK1.5), CD11b (M1/70), CD11c (N418), CD45RB/B220 (RA3 6B2) and Ter119 (TER119) (UBC AbLab), and then with anti-biotin microbeads (Miltenyi Biotec), all at concentrations previously determined by titration. The unlabelled fractions were collected from the MACS LS columns (Miltenyi Biotec) per manufacturer’s protocol, and purified cells typically contained 95% CD8+  TCRβ+ cells. For cell division analysis, total splenocytes or isolated CD8 T cells were labelled with 1 μM CFSE (Molecular Probe) at 107 cells/ mL in serum-free medium at 37oC for 13 mins, and then washed with serum-containing medium warmed to 37oC. Cells were cultured in complete RPMI. Isolated CD8 T cells were activated with either 2.5 ng/ ml phorbol 12-myristate 13-acetate (PMA) (Sigma-Aldrich) and 0.5 µg/ ml ionomycin (EMD Millipore), 1 µg/ ml αCD3ε (145-2C11; ATCC) and 5 µg/ ml αCD28 (clone 37.51; BD Biosciences), or with BM-derived dendritic cells (BMDC). BMDC were generated by culturing BMC with granulocyte-macrophage colony-stimulating factor as described previously (Poon et al., 2015). CD11c+ cells were isolated 7 days later with anti-CD11c microbeads (Miltenyi Biotec), and 5 x 104 purified BMDC were plated with 100 ng/ mL LPS (InvivoGen) and 10, or 1 pg/ mL SIINFEKL (Sigma-Aldrich), SIIQFEKL or SIITFEKL (both were generously provided by Dr. Tang, UBC) overnight in 200 µL in each well of a 96-well plate. After washing peptides and LPS off the BMDC, isolated CD8 T cells were plated atop stimulated and loaded BMDC at a 10-to-1 ratio in 200 µL. To assess responsiveness to IL-7 and IL-15 ex vivo, equal mixtures of    45   WT and CD44-/- OT-I CD8 T cells were transferred into a congenic hos, which was infected intravenously with Listeria monocytogenes expressing recombinant ovalbumin (LM-OVA) (Pope et al., 2001). Bulk BMC and splenocytes were isolated 20 days post infection, CFSE-labelled, and then cultured with 10 ng/ mL IL-7 and/ or 10 ng/ mL IL-15 for 5 days. Transferred T cells were identified from the bulk cells based on expression of CD45.1 and CD45.2 congenic markers, CD8 and TCR Vα2.  2.8 BM reconstitution  Cells transduced with retrovirus were sorted based on positive expression of the GFP or RFP reporter and CD44 expression. Bone marrow recipient mice were irradiated with two doses of 6.5 Gy of gamma radiation four hours apart. One day later, irradiated mice received at least 2 x 105 BMC intravenously.  2.9 Adoptive transfer and systemic Listeria monocytogenes infection  For competitive adoptive transfer experiments, each host mouse (CD45.1) received 2.5 x 104 OT-I CD8 T cells (CD45.2 or CD45.1/CD45.2) and the same number of CD44-/- OT-I CD8 T cells (CD45.2); for non-competitive adoptive transfer experiments, each host mouse received 5 x 104 of either OT-I CD8 T cells or CD44-/- OT-I CD8 T cells (Maeshima et al., 2011). Each mouse was infected a day later with 30,000 CFU of LM-OVA intravenously (Maeshima et al., 2011; Pope et al., 2001). Bacteria were grown to mid-log phase in brain heart infusion broth, and then diluted in PBS, and the CFU was confirmed by plating diluted bacterial culture on brain heart infusion agar plates. For secondary challenge, mice were re-infected intravenously with 30,000 CFU of LM-OVA 30 days after the primary infection. For in vivo IL-15 stimulation, each    46   mouse received 1 μg of recombinant mouse IL-15 (Peprotech) intravenously on day 31, and was euthanized 7 days later. For glucose uptake analysis, each mouse received 100 μg of 2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)-2-deoxyglucose (2-NBDG) intravenously, and were then euthanized 15 min post injection (Blagih et al., 2015).  2.10 Flow cytometry and cell sorting Cells were incubated with 2.4G2 tissue culture supernatant containing FcγRIII/FcγRII-specific monoclonal antibody for 20 min on ice, and then for 20 min with a mixture of monoclonal antibodies conjugated to fluorophores. Dead cells were labelled by 25 ng/ ml DAPI or by LIVE/DEAD® Fixable Dead Cell Stain per manufacturer’s protocol. To measure IFNγ production, up to 2 x 106 cells from the spleen or pLN were isolated from infected mice and then stimulated with 100, 10 or 1 pg/ mL SIINFEKL and 5 µg/ ml Penicillium brefeldianum brefeldin A (Sigma-Aldrich) for 4 hours. For intracellular of IFNγ and pAkt, after surface labelling, cells were fixed, permeabilized and labelled using the Foxp3/Transcription Factor Staining Buffer Set (Affymetrix), per manufacturer’s protocol. Caspase 3/7 activity was detected using FLICA® 660 Caspase 3/ 7 Assay Kit, far-red fluorescence (ImmunoChemistry Technologies), following the manufacturer’s protocol. Labelled cells were analyzed on an LSRII or Canto flow cytometer (BD Biosciences) using FACSDiva acquisition software (BD Biosciences). Data analyses were performed using FlowJo (Tree Star). For sorting, labelled cells were suspended in flow analysis buffer and sorted based on fluorescence on the BD Influx or BD Aria by the UBCFlow Facility.     47   2.11 RNA isolation, reverse transcription and quantitative PCR Total RNA was extracted from 2 x 106 48-h stimulated BMDM using TRIzol reagent (Invitrogen) and reverse-transcribed using the iScript cDNA Synthesis kit (Bio-Rad). Quantitative mRNA expression was analyzed by real-time PCR (Bio-Rad CFX384), with SsoFast EvaGreen (Bio-Rad). CD44s and CD44v10 were amplified using the common forward primer 5′-ACCATCGAGAAGAGCACC-3′ and the reverse primers 5′-TCATAGGACCAGAAGTTGTGG-3′ and 5′-GTCTCGATCTCCTGGTAAGG-3′, respectively. GAPDH served as the endogenous reference gene, and normalized gene expression to GAPDH was calculated by the CFX384 Touch™ Real-Time PCR Detection System (Bio-Rad).  2.12 Statistical analysis Statistical analyses were performed using Graphpad Prism (version 6) and Microsoft Excel (2010). For competition experiments, p-values were calculated using paired Student’s t-test. Unpaired Student’s t-test for homogeneous variances was used for other data sets. Statistical differences were considered significant when p < 0.05.   48   Chapter 3: Molecular Mechanism of HA Binding by Immune Cells  3.1 Introduction and rationale  Although CD44 is ubiquitously expressed by all immune cells, not all immune cells bind HA. HA binding is tightly regulated and different populations exhibit different HA binding capabilities (reviewed in (Lee-Sayer et al., 2015)). HA binding can be induced in activated macrophages and T cells (Lesley et al., 1994; Maeshima et al., 2011; Ruffell et al., 2011). To date, various factors have been identified to affect HA binding, but the physiological mechanism that regulates the up-regulation of HA binding is unclear. In this study, I examined the mechanism that regulates HA binding in classically and alternatively activated macrophages, and also aimed to determine if similar mechanisms occur in T cells.  The interaction between CD44 and HA can be regulated through various mechanisms. The addition of chondroitin sulfate on a serine in the stem region of CD44 (S180 in humans and S183 in mice) inhibits HA binding (Ruffell and Johnson, 2005), and chondroitin sulfate decoration of CD44 is shown to negatively regulate HA binding in monocytes, Jurkat T cells, HEK 293 cells and L fibroblasts (Levesque and Haynes, 1999; Ruffell and Johnson, 2005, 2008).  Other factors have also been shown to regulate HA binding by CD44. Through the interaction with the ERM proteins, CD44 can associate with the cytoskeleton, and the inhibition of cytoskeletal functions reduces HA binding in human myeloid cell lines, peripheral blood monocytes, and Jurkat T cells (Brown et al., 2005; Liu et al., 1996; Tsukita et al., 1994). Other post-translational modifications can also regulate HA binding by CD44. Sialylation can negatively regulate HA binding, as removal of sialic acid increases HA binding by CD44 in    49   monocytes and Th2 CD4 T cells (Faller and Guvench, 2014; Katoh et al., 2010; Katoh et al., 1999). In contrast, N- and O-linked glycosylation can both positively and negatively regulate HA binding (Dasgupta et al., 1996; English et al., 1998; Lesley et al., 1995; Skelton et al., 1998). The inclusion of additional exons can also alter HA binding, through the addition of post-translational modification sites (Bennett et al., 1995; Dasgupta et al., 1996). Some of these mechanisms were also investigated in this study. Given that chondroitin sulfate has been shown to regulate HA binding in multiple cells types, it is hypothesized to be utilized in HA binding regulation by all cells in some way. In our lab, classically activated macrophages (M1), generated by stimulation with TNFα, or with IFNγ with LPS, are found to reduce chondroitin sulfation on CD44, which is achieved by down regulation of the sulfation enzymes. Alternatively activated macrophages (M2), generated by stimulation with IL-4, may use a different mechanism to induce HA binding, which occurs despite an increase in chondroitin sulfation of CD44. My contribution to this body of work focused on phenotypic characterization of classically and alternatively activated macrophages, and the role of chondroitin sulfate in regulating their HA binding upon stimulation.  The second part of this chapter examines the mechanism by which T cells regulate HA binding. Upon activation, naïve T cells up-regulate CD44 expression, and transiently increase HA binding  (Maeshima et al., 2011). However, the physiological mechanism by which HA binding is induced remains unclear. Transduction of Jurkat T cells with CD44 harbouring a point mutation at the site of  chondroitin sulfate addition results in constitutive HA binding (Ruffell and Johnson, 2008), suggesting a role for chondroitin sulfation in the negative regulation of HA binding in T cells. However, it is unclear if chondroitin sulfate regulates HA binding in primary    50   T cells. Alternatively, HA binding by all T cells might be regulated by sialylation, as ex vivo reactivated Th2 T cells are shown to induce HA binding through de-sialylation of CD44 through the up-regulation of an endogenous neuraminidase (Katoh et al., 2010).  Understanding the mechanism by which cells regulate HA binding can help in understanding the situations, both spatially and temporally, in which cells bind HA. Furthermore, this understanding might aid in deciphering the function of CD44 and HA binding in immune cells.  3.2 Results 3.2.1 HA binding is increased in classically activated and alternatively activated BMDM  Analysis of surface markers revealed that classically and alternatively activated BMDM can be distinguished phenotypically by their surface expression of F4/80 and CD11c. Classically activated BMDM displayed increased expression of F4/80 and reduced expression of CD11c, when compared to alternatively activated BMDM (Fig. 3.1A). While subsequent stimulation with IL-4 had little effect on F4/80 or CD11c expression, subsequent LPS stimulation increased F4/80 and reduced CD11c expression in alternatively activated BMDM (Fig. 3.1A), consistent with the ability for LPS to convert alternatively activated to classically activated macrophages (Brown et al., 2011). Polarization of BMDM into either classically or alternatively activated macrophages with IFNγ plus LPS or with IL-4, respectively, significantly increased both CD44 expression and FL-HA binding (Fig. 3.1B, C). While subsequent IL-4 treatment had minimal effect on classically activated BMDM, subsequent LPS stimulation further enhanced both CD44 expression and FL-    51    Figure 3.1 Classically and alternatively activated BMDM induce HA binding through different mechanisms BMDM derived from WT mice were stimulated with either IFN-γ plus LPS (classically activated macrophages, or M1) or IL-4 (alternatively activated macrophages, or M2), or were left unstimulated (-) for 2 days, or were re-stimulated with LPS or IL-4 on the second day (LPS or L, and IL-4 or 4, respectively). Expressions based on the mean fluorescence intensities were calculated to be relative to the unstimulated control. (A) Relative expression of F4/80 and CD11c. (B) Mean fluorescence intensity of FL-HA binding. (C) Mean fluorescence intensity of CD44 expression. Representative data (A) or means and SE (B-C) from six independent experiments with six biological replicates shown with significance indicated as: * p < 0.05, ** p < 0.01, *** p < 0.001.   52   HA binding by alternatively activated BMDM (Fig. 3.1B, C). This data suggest that classically and alternatively activated macrophages induce HA binding via different mechanisms, where the signal for classical activation can further augment HA binding by alternatively activated macrophages, while the signal for alternative activation does not augment HA binding by classically activated macrophages.  3.2.2 Classically and alternatively activated BMDM differentially regulate HA binding through chondroitin sulfate Next I investigated if classically activated and alternatively activated macrophages utilized different mechanisms for inducing HA binding upon activation. The role of chondroitin sulfate was examined, since chondroitin sulfate is shown to negatively regulate HA binding and is removed upon activation (Delcommenne et al., 2002; Ruffell and Johnson, 2005). Alternatively activated BMDM exhibited increased chondroitin sulfation on CD44 (Ruffell et al., 2011). The presence of chondroitin sulfate may explain why alternatively activated BMDM exhibited reduced FL-HA binding compared to classically activated BMDM, despite having higher CD44 expression (Fig. 3.1B, C).  To delineate the role of chondroitin sulfation in HA binding by BMDM, cells were treated with p-nitrophenyl β-D-xylopyranoside (xyloside), a competitive inhibitor of GAG addition. Xyloside treatment enhanced FL-HA binding by alternatively activated BMDM, and also by TNFα-polarized BMDM to a lesser extent (Fig. 3.2). In contrast, xyloside had little effect on FL-HA binding by classically activated BMDM or LPS-stimulated alternatively activated BMDM (Fig. 3.2). This is consistent with the hypothesis that LPS stimulation induces HA binding through the    53    Figure 3.2 GAG addition affects HA binding of activated BMDM BMDM derived from WT mice were stimulated with either TNFα, IFN-γ plus LPS (classical activation), or IL-4 (alternative activation), or were left unstimulated for 2 days, or were re-stimulated with LPS or IL-4 on the second day, in the presence or absence of 2 mM β-D-xyloside for the entire duration of stimulation. Histograms of FL-HA binding. Histogram representative of two independent experiments shown.   54   removal of chondroitin sulfate, and suggests that LPS can further enhance HA binding in alternatively activated macrophages through this removal of chondroitin sulfate.  3.2.3 Classically and alternatively activated BMDM differentially regulate expression of CD44 v10 LPS was determined to induce HA binding through the removal of chondroitin sulfate, and can further increase HA binding in alternatively activated macrophages. However it was unclear how HA binding was induced initially in macrophages upon alternative activation. As such, other mechanisms were explored.  The expression of variable exons can also alter HA-binding (Jackson et al., 1995; Liao et al., 1993). Thus, the presence of variable exons was examined using quantitative PCR. Consistent with surface protein expression levels determined by flow cytometry, alternatively activated BMDM expressed significantly higher level of CD44s transcript than unstimulated or classically activated BMDM (Fig. 3.3A). Primers were designed to potentially detect the expression of all v10-containing CD44 isoforms. However, only a single PCR product corresponding to CD44v10 was detected. The expression of exon v10 was detected in unstimulated BMDM, and its expression level was significantly reduced in classically activated BMDM and significantly increased in alternatively activated BMDM as compared to unstimulated BMDM (Fig. 3.3B). Overall, different polarizing conditions differentially regulated CD44 v10 exon expression at the transcript level. However, it remained to be determined if protein expression followed the same trend.    55    Figure 3.3 Relative mRNA expression of CD44s and CD44 v10 in BMDM BMDM derived from WT mice and either unstimulated (M0), stimulated with IFN-γ plus LPS (M1), or IL-4 (M2) for 2 days. Transcript expression levels were measured by quantitative PCR and normalized to GAPDH expression. Expressions were calculated to be relative to the unstimulated (M0) control. (A) Relative expression of CD44. (B) Relative expression of CD44 v10. Means and SE from three independent experiments shown with significance indicated as: * p < 0.05, ** p < 0.01, *** p < 0.001.     56   Since CD44 expression was higher in alternatively activated BMDM than the classically activated BMDM, the increased CD44 density on the cell surface may somehow allow alternative activated macrophages to partially negate the effects of chondroitin sulfate decoration. Furthermore, expression of the variant exon v10 of CD44 may have contributed to the initial induction of HA binding in alternatively activated macrophages, as its expression was higher in alternatively activated BMDM than classically activated BMDM.   3.2.4 Actin polymerization is not required for induction of HA binding by BMDM Another possible mechanism for inducing HA binding is through altering the clustering pattern of CD44 on the cell surface. Cytokine stimulation may therefore alter CD44 clustering through rearrangement of the cytoskeleton, thereby increasing its avidity to HA. To investigate if actin rearrangement influenced HA binding in macrophages, unstimulated, classically activated and alternatively activated BMDM, and BMDM activated with LPS plus IL-4 were treated with cytochalasin D, an inhibitor of actin rearrangement. Treatment with cytochalasin D slightly decreased CD44 expression and FL-HA binding by all BMDM tested (Fig.3.4). However the reduction in CD44 expression and FL-HA binding after treatment with cytochalasin D was not significant when compared to either the untreated or DMSO vehicle controls. Therefore HA binding in BMDM did not require actin polymerization in BMDM.  3.2.5 Sialic acid can negatively regulate CD44-mediated HA binding in BMDM Earlier, I had investigated the effect of chondroitin sulfation of CD44 on HA binding in BMDM, and removal of chondroitin sulfate had no effect on the induction of HA binding in alternatively     57    Figure 3.4 Inhibiting actin polymerization has no effect on HA binding in activated BMDM  BMDM derived from WT mice and were either unstimulated (-), or stimulated with IFN-γ plus LPS or IL-4 for 48 hr, or with IL-4 for 24 hr followed by addition of LPS for another 24 hr, in the presence or absence of  20 µg/ mL cytochalasin D or DMSO control for 1 hr prior to flow analysis. (A) Histograms of CD44 expression and FL-HA binding. Plots representative of two biological replicates. (B) Relative CD44 expression (top) and FL-HA binding. Means and SE from three biological replicates.    58   activated BMDM. Other forms of post-translational modification can also alter the affinity of CD44 to HA.  To determine if sialic acid decoration regulated the induction of HA binding in macrophages, classically and alternatively activated BMDM were treated with neuraminidase to remove sialic acid on cell surface proteins. Neuraminidase treatment slightly reduced cell surface CD44 expression in alternatively activated BMDM, and slightly increased FL-HA binding by both classically and alternatively activated BMDM (Fig. 3.5). However more experimental repeats are required to determine if the effect is statistically significant. Furthermore the ability for neuraminidase treatment to increase HA binding by cells would suggest the presence of sialic acid on the cells, and thus argue against the removal of sialic acid as the mechanism by which macrophages induce HA binding upon activating stimulation. Overall, chondroitin sulfation on CD44 limited HA binding in alternatively activated BMDM, and prevention of chondroitin sulfate addition with xyloside treatment during stimulation increased HA binding by alternatively activated BMDM. LPS was able to further increase HA binding in alternatively activated BMDM. Therefore, LPS triggered removal of chondroitin sulfate on CD44, thereby increasing HA binding, and this was likely the mechanism by which LPS-stimulated BMDM increased their HA binding. HA binding by alternatively activated BMDM was not through the removal of chondroitin sulfate, and could be further increased with subsequent LPS treatment. The induction of HA binding in BMDM was likely not regulated by actin rearrangement or sialylation, but the expression of CD44 v10 in alternatively activated macrophages may play a role in inducing HA binding.      59    Figure 3.5 Neuraminidase treatment slightly increases HA binding in activated BMDM BMDM derived from WT mice were stimulated with either IFN-γ plus LPS or with IL-4 for 48 hr, in the presence or absence of 0.1 U/ mL neuraminidase for 1.25 hr prior to flow analysis. (A) Histograms of CD44 expression and FL-HA binding. Unstained control, untreated cells and neuraminidase-1 treated cells shown. Plots representative of two biological replicates shown. (B) Relative CD44 expression and FL-HA binding. Means and SE from two biological replicates shown.    60   3.2.6 Inhibition of GAG addition with xyloside has no effect on HA binding by splenic CD4 and CD8 T cells As part of my project focused on the function of T cells, I was also interested in understanding the mechanism by which activated T cells induced HA binding. In Jurkat T cells, the S180A point mutation, which removes the site of chondroitin sulfation on CD44, increases HA binding (Ruffell and Johnson, 2008), suggesting that chondroitin sulfation may also regulate HA binding in primary T cells.  To investigate the role of chondroitin sulfation on CD44 in regulating HA binding, total T cells were isolated from the spleens of naïve mice, and were treated with xyloside, or were activated ex vivo in the presence or absence of xyloside. Xyloside-treated naïve T cells did not increase FL-HA binding as compared to the control cells (Fig. 3.6A). Similarly, xyloside also did not increase the FL-HA binding by activated T cells (Fig. 3.6A).  Unfortunately, these findings did not rule out the role of chondroitin sulfate in regulating HA binding by activated T cells. Xyloside inhibits the addition of chondroitin sulfate, and a lack of difference after xyloside treatment in naïve and activated cells therefore suggests that either CD44 on naïve cells is decorated with chondroitin sulfate, which is then removed upon activation, or that chondroitin sulfate is not present on CD44 expressed by both naïve and activated T cells. The data therefore cannot affirm or deny the role of chondroitin sulfate in regulating HA binding by T cells.       61    Figure 3.6 Inhibition of GAG addition or actin polymerization has no effect on HA binding by naïve and activated T cells (A) Histograms of FL-HA binding splenic T cells with or without xyloside treatment. Naïve splenic T cells were treated with 2 mM xyloside for 1hr. Splenic T cells activated with with 1 µg/ mL α-CD3 monoclonal antibody and mitomycin C-treated splenocytes were grown with or without 2 mM xyloside. Representative plot from two independent experiments with one biological replicate total. (B) Histograms of FL-HA binding of splenic T cells with or without cytochalasin D treatment. Splenic T cells were activated with 1 µg/ mL α-CD3 monoclonal antibody and mitomycin C-treated splenocytes or treated with or 20 or 50 µg/ mL cytochalasin D or DMSO control for 1 hr prior to flow analysis. Representative plot from two independent experiments with one biological replicate total.   62   3.2.7 Actin polymerization is not required for induction of HA binding by splenic T cells   Other alternate mechanisms of inducing HA binding were also explored. Actin and microtubule functions are shown to be important for inducing HA binding in Jurkat T cells (Liu et al., 1996), suggesting that cytoskeleton rearrangement could be involved in inducing HA binding. Furthermore, TCR engagement triggers actin rearrangement to form the immune synapse (Kumari et al., 2014), and thus rearrangement of actin might also alter the CD44 clustering pattern, thereby inducing HA binding.   To investigate this, total T cells were isolated from the spleens of naïve mice and were then activated ex vivo in the presence or absence of cytochalasin D, which inhibited actin polymerization. Cytochalasin D did not reduce FL-HA binding in activated T cells, at 20 or 50 µm/ mL (Fig. 3.6B), suggesting that actin rearrangement is dispensable for HA binding in activated T cells.   3.2.8 Sialylation of CD44 negatively regulates HA binding by splenic T cells In addition to chondroitin sulfate, sialic acid is another key putative post-translational modification on CD44 that might regulate HA binding in T cells. Splenic T cells were isolated from naïve mice and were treated with neuraminidase-1 to remove sialic acid. Neuraminidase-1 treatment significantly increased HA binding T cells from naïve mice from less than 2% to approximately 10% in both CD4 and CD8 T cells (Fig. 3.7 A, B). The induction of HA binding was observed only in the CD44hi memory phenotype T cells. Furthermore, neuraminidase-1 treatment also significantly increased HA binding by ex vivo activated CD4 and CD8 T cells    63    Figure 3.7 Neuraminidase treatment increases HA binding in naïve and activated T cells, but inhibition of neuraminidase has no effect on HA binding Naïve splenic T cells or splenic T cells activated with 1 µg/ mL α-CD3 monoclonal antibody and mitomycin C-treated splenocytes were treated with 0.1 U/ mL neuraminidase-1 1.25 hr prior to flow analysis, or with 4 mM of NeuAc2en 16 hr prior to flow analysis. (A) Flow plot of naïve T CD4 and CD8 cells from CD44-/- and WT mice with out without neuraminidase treatment. (B) Percent FL-HA binding CD4 and CD8 T cells with or without neuraminidase-1 treatment. Means and SD from three independent experiments with eight biological replicates total shown with significance indicated as: * p < 0.05, ** p < 0.01, *** p < 0.001.  (C) Percent FL-HA binding CD4 and CD8 T cells with or without NeuAc2en treatment. Means and SD from one    64   independent experiment with four biological replicates each shown with significance indicated as: * p < 0.05, ** p < 0.01, *** p < 0.001   65   (Fig. 3.7B), indicating that CD44 on both memory phenotype and activated T cells were sialylated and that sialylation inhibited HA binding. This data also suggest that naïve and activated T cells were sialylated and that removal of sialic acid did not occur upon activation. To determine if removal of sialic acid induced HA binding in activated T cells, splenic T cells were activated ex vivo in the presence or absence of the neuraminidase inhibitor, 2-Deoxy-2,3-dehydro-N-acetylneuraminic acid (NeuAc2en). Treatment with NeuAc2en did not reduce HA binding by activated CD4 or CD8 T cells (Fig. 3.7C), but it was unclear if the inhibitor did indeed reduce endogenous neuraminidase activity. Increased HA binding by neuraminidase treatment of activated T cells, suggest that while sialic acid negative regulates HA binding, sialic acid on CD44 is not removed in activated T cells. Furthermore, the lack of effect by NeuAc2en supports the lack of sialic acid removal of CD44 in activated T cells. These experiments suggest that sialylation of CD44 on activated T cells negatively regulates HA binding, but this is likely not a physiological mechanism by which HA binding is induced in activated T cells.  3.2.9 O-link glycosylation does not negatively regulate HA binding by splenic T cells Since sialic acid likely does not regulate HA binding in activated T cells, another post-translational modification was considered. O-linked glycosylation is shown to inhibit HA binding in colon carcinoma cells, and removal of O-linked glycosylation enhances HA binding (Dasgupta et al., 1996). Activated T cells may remove O-linked glycosylation to induce HA binding.  Splenic CD4 and CD8 T cells were isolated from naïve mice and were then activated ex vivo in the presence or absence of the O-linked glycosylation inhibitor phenyl-N-diacetyl-α-D-     66    Figure 3.8 Inhibition of O-linked glycosylation has no effect on HA binding by splenic T cells Percent FL-HA binding CD4 and CD8 T cells with or without pNAcGal treatment. Splenic T cells activated with 1 µg/ mL α-CD3 monoclonal antibody and mitomycin C-treated splenocytes were treated with 20 mM of pNAcGal 16 hr prior to flow analysis. Means and SD from three independent experiments with seven total biological replicates shown with significance indicated as: * p < 0.05, ** p < 0.01, *** p < 0.001.    67   galactosaminide (pNAcGal), to investigate if O-linked glycosylation played a role in regulating HA binding in activated T cells. Treatment with pNAcGal had no effect on HA binding by splenic CD4 or CD8 T cells, suggesting that O-linked glycosylation was not involved in regulating T cell HA binding (Fig. 3.8), although the efficacy of the inhibitor was not verified.  Based on my findings, activated T cells did not induce HA binding through actin rearrangement, removal of sialic acid or O-linked glycosylation, and removal of chondroitin sulfate was not conclusively ruled out as a mechanism for inducing HA binding in activated T cells. As such, future experiments should focus on investigating the role of chondroitin sulfation of CD44 in regulating HA binding in T cells.  3.3 Discussion In this study, classical and alternative activation of BMDM both up-regulated CD44 expression, and induced HA binding. Furthermore, classically activated and alternatively activated BMDM were found to regulate HA binding through different mechanisms. As part of the same study, experiments conducted by others have found that classical activation of BMDC with either TNFα or IFNγ with LPS reduces chondroitin sulfate on CD44, while alternative activation of BMDC increases chondroitin sulfation on CD44 (Ruffell et al., 2011). Together with my data, these findings indicate that LPS induces HA binding through the removal of chondroitin sulfate, while stimulation with IL-4 induces HA binding through a different mechanism.  There are two possible ways by which chondroitin sulfate might inhibit the interaction between CD44 and HA. Chondroitin sulfate may cause steric hindrance by competing for the HA binding pocket on CD44, or chondroitin sulfate might stabilize the non-binding conformation of CD44    68   (Ruffell and Johnson, 2009). The inhibition of HA binding by chondroitin sulfate is unlikely to be due to direct binding of chondroitin sulfate to the HA binding pocket, as CD44s has a lower affinity for chondroitin sulfate than for HA (Aruffo et al., 1990). Crystal structure of the HA-binding domain of CD44 identified conformational changes between the HA-binding and non-binding forms of CD44 (Banerji et al., 2007), and the negatively charged chondroitin sulfate might be able to stabilize the non-binding CD44 conformation, through its interaction with a positively charged region of CD44. Since alternatively activated BMDM was heavily decorated with chondroitin sulfate, the induction of HA binding must be through a different mechanism. Alternatively activated BMDM expressed higher transcript level of CD44v10. The v10 exon contains a Bx7B motif, which recognizes chondroitin sulfate and is required for CD44-mediated cell-cell interaction (Hayes et al., 2002). Furthermore, v10-containing CD44 can interact with chondroitin sulfate expressed on other CD44 molecules (Chiu et al., 1999). Therefore, it is possible that the v10 exon interacts with chondroitin sulfate on an adjacent CD44 molecule, and thereby diverting chondroitin sulfate away from stabilizing the non-HA binding conformation of CD44 mediated by intramolecular interaction. This proposed mechanism would explain how alternatively activated BMDM were able to bind HA, despite heavy chondroitin sulfation on CD44. For splenic T cells, the induction of HA binding was not regulated through sialylation, O-linked glycosylation, or actin rearrangement, as their respective inhibitors had no effect on HA binding. Given that chondroitin sulfate regulates HA binding in macrophages, monocytes and Jurkat T cells (Levesque and Haynes, 1999; Ruffell and Johnson, 2008; Ruffell et al., 2011), chondroitin sulfate was hypothesized to also regulate HA binding by T cells. However, its role in regulating    69   HA binding by activated T cells could not be conclusively affirmed or denied, as treatment with xyloside had no effect on HA binding by naïve or activated T cells. The caveat of the experiment is that the inhibitor used would is expected to eliminate factors the negatively regulate HA binding, the same factors hypothesized to be removed upon T cell activation to enable HA binding. Thus, the lack of effect by the inhibitors is difficult to interpret.    70   Chapter 4: CD44 and HA Binding Reduce Memory Potential of CD8 Effector T Cells  4.1 Introduction and rationale Memory cells provide protection during re-exposure to infectious agents and are the basis for vaccine efficacy. Upon infection, CD8 T cells undergo massive expansion followed by the contraction phase, where majority of effector CD8 T cells die, leaving a small population of memory cells. The mechanism by which an effector CD8 T cell decides between death and survival during the contraction phase remains unclear, and several models have been proposed (Kaech and Cui, 2012). Factors such as γc family cytokines, transcription factors and CD4 T cell help are known to affect the formation of CD8 memory T cells (Bevan, 2004; Kaech and Cui, 2012; Rochman et al., 2009), while the role of the ECM is less well understood. One way that effector CD8 T cells can interact with the ECM is through the glycoprotein CD44. CD8 T cells up-regulate CD44 expression after activation (Maeshima et al., 2011), and CD44 expression remains high on antigen-experienced effector and memory T cells (Lesley et al., 2000). Activated CD8 T cells transiently increase CD44-mediated binding to FL-HA in vitro and in vivo (Maeshima et al., 2011), and the interaction between CD44 and HA can mediate T cell rolling and extravasation to inflammatory sites (DeGrendele et al., 1997; Siegelman et al., 1999). Thus the up-regulation of CD44 facilitates the interaction of activated T cells with its environment through binding to HA.  HA induces AICD in Jurkat T cells and in ex vivo re-stimulated mouse splenic T cells (Ruffell and Johnson, 2008). Furthermore, transduction of a CD44 mutant with increased HA binding    71   into Jurkat T cells increases their cell death (Ruffell and Johnson, 2008). While some mouse studies also show a pro-apoptotic role of CD44 in T cells (McKallip et al., 2002; Nakano et al., 2007), other studies suggest that CD44 and its variant exon 7-containing isoform confer resistance to apoptosis in T cells (Marhaba et al., 2006; Wittig et al., 2000). This apparent disparity in the role of CD44 in regulating T cell death and survival might depend on the state of the T cell, as TCR signals can lead to survival, proliferation or apoptosis. In a mouse model of influenza infection, CD44-/- CD4 Th1 cells were more apoptotic and failed to form memory cells (Baaten et al., 2010).  However, it remains to be determined if the same holds true for other CD4 T helper cell populations and for CD8 effector T cells.  In this study, I made use of CD8 T cells expressing the transgenic OT-I TCR, which has high affinity for the SIINFEKL sequence of OVA (Hogquist et al., 1994). To investigate the role of CD44 on CD8 effector and memory T cells, CD44+/+ and CD44-/- CD8 T cells were adoptively transferred in competition into WT hosts, which were then infected intravenously with LM-OVA. The relationship between HA binding and TCR peptide affinity and avidity were also investigated by utilizing variants of the SIINFEKL peptide at varying amounts in vitro.  4.2 Results 4.2.1 The magnitude and extent of induced FL-HA binding in activated CD8 T cells correlate positively with the strength of the TCR signal during activation Despite the high CD44 expression, binding to FL-HA by the activated T cells is heterogeneous, and the HA-binding cells are the most actively proliferating (Maeshima et al., 2011), providing a link between HA binding and proliferation in activated T cells. TCR signal strength is also    72   shown to influence the degree of proliferation by activated T cells (Denton et al., 2011). The link between TCR signal strength and HA binding was thus investigated. OT-I CD8 T cells were activated with LPS-stimulated, peptide-loaded BMDC in vitro. Peptide affinity was varied by using the WT SIINFEKL (N4) peptide or altered peptide ligands SIIQFEKL (Q4) and SIITFEKL (T4), which have reduced affinity for the OT-I TCR (Daniels et al., 2006; Turner et al., 2008), and peptide avidity was varied by loading 100, 10 or 1 pg/ mL of N4 peptide onto BMDC.  OT-I CD8 T cells activated with BMDC loaded with 100 pg/ mL of the N4 peptide, the condition with the highest peptide affinity and avidity, resulted in the most proliferation as evidenced by their increased CFSE dilution and cell number (Fig. 4.1A, B). Cells activated with 100 pg/ mL N4 peptide also exhibited the highest CD44 expression and FL-HA binding, both of which were reduced in cells activated by BMDC loaded with peptides with reduced affinity at 100 pg/ mL (Fig. 4.1C). CD44 expression and FL-HA binding were also progressively reduced in cells activated with BMDC loaded with progressively reduced amounts of the N4 peptide (Fig. 4.1C).  Maximal induction of CD44 expression and FL-HA binding were observed in cells that have undergone two divisions (Fig. 4.1A, E). After the first division, OT-I CD8 T cells uniformly up-regulated CD44 expression, and after the second division, a portion of cells exhibited even higher CD44 expression, resulting in bifurcated CD44 expression (Fig. 4.1A). The bifurcation of CD44 expression was less prominent in CD8 T cells activated with reduced peptide affinity or avidity, as the population with the further increased CD44 expression was reduced, resulting in reduced overall CD44 expression (Fig. 4.1A-C). Similarly, the induction of FL-HA binding was observed in cells that have divided twice (Fig. 4.1A). The bifurcated CD44 expression and FL- HA binding were gradually reduced in cells that had divided more times, as the population with   73    Figure 4.1 TCR peptide affinity and avidity correlate positively with degree of HA binding by activated OT-I CD8 T cells Splenic OT-I CD8 T cells were CFSE-labelled and then activated by BMDC, which were stimulated with 100 ng/ ml LPS and loaded with 100, 10 or 1 pg/ ml of SIINFEKL (N4), SIIQFEKL (Q4) or SIITFEKL (T4) peptide a day prior to T cell activation. (A) Flow cytometry plots showing CD44 expression or FL-HA binding versus CFSE dilution of cells 3 days post activation. (B) T cell counts 3 days post activation. (C and D) CD44 expression and FL-HA binding of cells activated with 100 pg/ ml of N4, Q4 or T4 peptides (C), or with 100, 10 or 1 pg/ mL of N4 peptide (D). (E) CD44 expression and FL-HA binding of cells activated with 100 pg/    74   mL N4 peptide on day 3 after each cell division as assessed by CFSE dilution. Means and SD from two independent experiments with six biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 calculated by two-tailed Student’s t-test.   75   increased CD44 expression and bound FL-HA reduced in size (Fig. 4.1A, E). Peak CD44 expression and FL-HA binding were observed on cells that have undergone two divisions (Fig. 4.1A, E).  4.2.2 HA binding and CD44 expression have no direct effect on the generation of effector CD8 T cells  Since high affinity and avidity TCR engagement induced a strong proliferative response and high, albeit transient, FL-HA binding in CD8 T cells in vitro, HA engagement was hypothesized to be important for T cell expansion. CD44-/- CD8 effector T cells might therefore be defective in clonal expansion. To test this, 5 x 104 of either CD44+/+ or CD44-/- OT-I CD8 T cells were adoptively transferred into congenic WT hosts, which were intravenously infected with LM-OVA a day later. At the peak of the response, 7 days post infection, slightly higher numbers of CD44-/- OT-I CD8 T cells recovered from the pLN and spleen, although the difference was not statistically significant (Fig. 4.2A). There was also no significant difference in the percentages of IFNγ-producing cells between CD44+/+ and CD44-/- OT-I CD8 T cells (Fig. 4.2B), suggesting that that CD44 is dispensable for the generation of effector CD8 T cells.  4.2.3 CD44-/- OT-I CD8 T cells have a competitive advantage in memory formation Since the role of the HA-binding form of CD44 might be to sequester cells to an HA-rich niche, the effect of CD44 deletion may only be revealed when cells are in competition. To investigate this, 5 x 104 of an equal mixture of CD45.1/ 45.2 CD44+/+ OT-I CD8 T cells and CD45.2 CD44-/- OT-I CD8 T cells were adoptively transferred into CD45.1 WT hosts, followed by intravenous infection with LM-OVA a day later. OT-I CD8 T cells were identified as CD8+ Vα2+ CD45.2+    76     Figure 4.2 CD44-/- CD8 T cells have no cell intrinsic defect in cell numbers or IFNγ production during the effector response Recipient CD45.1 WT mice received 5 x 104 of either CD44+/+ or CD44-/- CD45.2 OT-I CD8 T cells, and were infected intravenously with 30,000 CFU of LM-OVA a day later. (A) Number of transferred CD8 T cells 7 days post infection in the pLN and spleen. (B) Percent of IFNγ-expressing cells within the transferred CD44+/+ or CD44-/- OT-I CD8 T cell populations 7 days post infection. Means and SD from two independent experiments with six total biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 calculated by two-tailed Student’s t-test.         77   (Fig. 4.3A). No major differences between CD44+/+ and CD44-/- OT-I CD8 T cell numbers in the pLN, spleen and BM were observed during the effector phase (Fig. 4.3B). When the percentages of CD44+/+ and CD44-/- T cells were compared, the CD44+/+ T cells had a significant advantage, during the effector phase, on days 5 and 7 post infection, in the liver and spleen, but not the pLN or BM (Fig. 4.3B). However, control experiments, where CD45.1/ CD45.2 CD44+/+ OT-I CD8 T cells were co-transferred with CD45.2 CD44+/+ OT-I CD8 T cells in equal proportions, showed that the difference observed in the spleen was in fact due to allelic effects of CD45 (Fig.4.3C), as suggested by (Nolte et al., 2004; Waterstrat et al., 2010), and was thus not an effect of CD44 expression. Based on the control experiment, there was a small but significant CD44-dependent competitive advantage of CD44+/+ CD8 T cells in the liver, where HA is known to be abundantly expressed after infection (Wang and Kubes, 2016). Somewhat unexpectedly, the slight CD44+/+ advantage in the liver was not sustained. By day 10 post infection, after the contraction phase of the response, significantly higher numbers of  CD44-/- CD8 T cells were observed in the spleen, pLN and BM, but not in the liver (Fig. 4.3B). This CD44-/- OTI CD8 T cell advantage was maintained at day 20 and day 30, at a ratio of approximately 3-to-2, in the BM and pLN (Fig. 4.3B), and was not due to allelic effects of CD45 (Fig. 4.3D). Increased numbers of CD44-/- were also observed on day 30 post infection, when the competing CD44+/+ and CD44-/- OT-I CD8 T cells were both on the CD45.2 background (Fig. 4.3E), further supporting that the competitive advantage was not due to CD45 allelic effects. Although there was a similar trend, no significant increase in the number of memory CD44-/- OT-I CD8 T cells was observed as compared to CD44+/+ OT-I CD8 T cells when not in competition (Fig. 4.3F),     78       79   Figure 4.3 CD44-/- OT-I CD8 T cells have a competitive advantage in memory cell formation (A-D) Recipient CD45.1 WT mice received 5 x 104 of equal mixture of CD45.1/ CD45.2 CD44+/+ or CD45.2 CD44-/- OT-I CD8 T cells, and were infected intravenously with 30,000 CFU of LM-OVA a day later. (A) Gating strategy for identifying adoptively transferred OT-I CD8 T cells. (B) Flow cytometry plots of CD45.1 versus CD45.2 of total adoptively transferred OT-I CD8 T cells 30 days post infection (left), total number (centre) and percent (right) of transferred CD44+/+ and CD44-/- OT-I CD8 T cells within the total transferred OT-I CD8 T cell population post infection. (C-D) Recipient CD45.1 WT mice received equal mixtures of 5 x 104 of CD44+/+ and CD44+/+ or CD44+/+ and CD44-/- OT-I CD8 T cells and were infected intravenously with 30,000 CFU of LM-OVA a day later. Ratios of CD45.1/ CD45.2 OT-I CD8 T cells over CD45.2 OT-I CD8 T cells on days 5 (C) and 20 (D) post infection. (E) Number of transferred CD8 T cells 30 days post infection. Recipient CD45.1 WT mice received 5 x 104 of equal mixture of CD45.2 CD44+/+ or CD45.2 CD44-/- OT-I CD8 T cells, and were infected intravenously with 30,000 CFU of LM-OVA a day later. CD44 expression was used to differentiation between the two populations. (F) Number of transferred CD8 T cells 30 days post infection. Recipient CD45.1 WT mice received 5 x 104 of either of CD45.2 CD44+/+ or CD45.2 CD44-/- OT-I CD8 T cells, and were infected intravenously with 30,000 CFU of LM-OVA a day later. Means and SD from at least two independent experiments with at least six total biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 calculated by two-tailed Student’s t-test.   80   demonstrating an enhanced effect when the cells were in competition.   4.2.4 The increased CD44-/- to CD44+/+ ratios in OT-I CD8 T cells are sustained in secondary responses To determine if this competitive advantage of CD44-/- CD8 memory T cells was maintained during subsequent T cell responses, mice were re-challenged intravenously with LM-OVA 30 days after the primary infection. Significantly higher numbers of effector CD44-/- OT-I CD8 T were recovered from the spleen and BM on days 2 and 5 post secondary infection (Fig. 4.4A). Higher numbers of CD44-/- OT-I CD8 T were also recovered from the pLN, but the difference was not statistically significant (Fig. 4.4A). CD44-/- OT-I CD8 T cells out-competed CD44+/+ OT-I CD8 T cells by ratios of approximately 60-to-40 at days 1, 2 and 5 post secondary infection in the pLN and BM, where the difference was significant (Fig. 4.4B). Similar trends were observed in the spleen but were not significant; the difference was even less apparent in the liver (Fig. 4.4B). The competition experiments demonstrate a competitive disadvantage for cells expressing CD44 in memory formation.   4.2.5 CD44-/- OT-I CD8 T cells are not more responsive to IL-7 or IL-15 Alternatively, the loss of CD44 may confer an advantage through IL-7 and/or  IL-15 signalling. IL-7 and IL-15 are important cytokines for memory CD8 T cell formation, and the limited availability of these cytokines restricts the size of the memory pool (Rochman et al., 2009; Schluns and Lefrancois, 2003). Since the competitive advantage of CD44-/- CD8 T cells over CD44+/+ CD8 T cells was observed after the contraction phase, CD44-/- cells might possess an enhanced ability to compete for or respond to IL-7 and/ or IL-15. To investigate this, expression    81    Figure 4.4 The CD44-/- OT-I CD8 T cell advantage over CD44+/+ OT-I CD8 T cells is maintained in secondary effector response Recipient CD45.1 WT mice received 5 x 104 of equal mixture of CD45.1/ 45.2 CD44+/+ and CD45.2 CD44-/- OT-I CD8 T cells, and were infected intravenously with 30,000 CFU of LM-OVA a day later. Mice were re-challenged with 30,000 CFU of LM-OVA 30 days post-primary infection. (A) Total number of transferred CD44+/+ and CD44-/- OT-I CD8 T cells post-secondary infection. (B) Percent of transferred CD44+/+ and CD44-/- cells within the total transferred OT-I CD8 T cell population post-secondary infection. Means and SD from two independent experiments with six total biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 as calculated by two-tailed Student’s t-test.   82   levels of CD127 and CD122, the respective IL-7 and IL-15 receptor subunits, were compared. When in competition, CD44+/+ and CD44-/- OT-I CD8 T cells had similar expression levels of CD127 and CD122 (Fig. 4.5A).  To examine the proliferative responses to IL-7 and IL-15, BMC and splenocytes were isolated 20 days post infection, from recipient mice that received equal propotions of CD44+/+ and CD44-/- OT-I CD8 T cells followed by intravenous LM-OVA infection. The bulk cells were CFSE-labelled, and then stimulated ex vivo with recombinant mouse IL-7 and/ or IL-15. Transferred OT-I CD8 T cells were identified based on CD8, TCR Vα2, CD45.1 and CD45.2 expression similar to the analysis of competition between adoptively transferred CD45.1/ 45.2 and CD45.2 OT-I CD8 T cells (Fig. 4.3A). CD44+/+ and CD44-/- proliferated to similar extent in response to IL-7 and/ or IL-15 ex vivo (Fig. 4.5B), indicating that their responsiveness to these cytokines was comparable. This suggests that the competitive advantage of CD44-/- T cells in memory formation was not due to increased responsiveness to IL-7 or IL-15.  4.2.6 HA-binding WT OT-I CD8 T cells are more susceptible to death The competitive advantage of CD44-/- OT-I CD8 T cells in memory formation was therefore likely due to CD44 and its interaction with HA promoting more effective population contraction of the CD44+/+ CD8 T cells population. Since majority of effector T cells die through apoptosis (Krammer et al., 2007), differences in susceptibility to death might lead to differences in the size of the resulting memory population. To test this, cell death was measured by staining with cell viability dyes. When in competition, CD44+/+ OT-I CD8 T cells had a reduced percentage of live cells compared to CD44-/- OT-I CD8 T cells on days 7 and 10 post infection (Fig. 4.6A). To further investigate whether the presence of CD44 affected apoptosis during the contraction   83    Figure 4.5 CD44-/- OT-I CD8 T cells do not have an increased response to the common-γ cytokines IL-7 or IL-15 (A) Mean fluorescence intensity of CD127 and CD122 expression obtained from flow analysis of CD44+/+ and CD44-/- OT-I CD8 T cells 20 days post infection. Recipient CD45.1 WT mice received 5 x 104 of equal mixture of CD45.1/ 45.2 CD44+/+ and CD45.2 CD44-/- OT-I CD8 T cells, and were infected intravenously with 30,000 CFU of LM-OVA a day later. Means and SD from at least two independent experiments with six total biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 calculated by two-tailed Student’s t-test. (B) CFSE dilution of transferred CD44+/+ and CD44-/- OT-I CD8 T. Recipient CD45.1 WT mice received 5 x 104 of equal mixture of CD45.1/ 45.2 CD44+/+ and CD45.2 CD44-/- OT-I CD8 T cells, and were infected intravenously with 30,000 CFU of LM-OVA a day later. Bulk BMC and splenocytes were isolated 20 days post infection, CFSE-labelled and cultured with 10 ng/ ml IL-7 and or 10 ng/ ml IL-15 ex vivo for 5 days. Representative of two independent experiments with six biological replicates is shown.    84   phase, caspases-3 and-7 activities were measured in T cells after LM-OVA infection by utilizing the FLICA reagent, a fluorescent reagent that binds to active caspase-3 and -7. On day 10 post infection, cells were isolated from the pLN, spleen and BM from recipients that received equal mixture of CD44+/+ and CD44-/- OT-I CD8 T, and were then incubated with the FLICA reagent. Compared to CD44-/- OT-I CD8 T cells, CD44+/+ OT-I CD8 T cells from the same recipient mice had a higher percent of FLICA+ cells, and this difference was significant in the spleen and BM (Fig. 4.6B). A similar trend was observed in the pLN, but the difference is not significant due to larger biological variations (Fig. 4.6B).  To determine if this difference in apoptosis between CD44+/+ and CD44-/- OT-I CD8 T cells was due to the ability to bind HA, CD44+/+ cells were divided into HA-binding and non-binding populations on the basis of their binding to FL-HA. The percentages of live cells within the HA- binding CD44+/+ population were significantly reduced compared to the non-binding CD44+/+ or CD44-/- OTI CD8 T cells on day 10 (Fig. 4.6C). Similarly, on day 10 post infection, HA-binding CD44+/+ had significantly higher percentages of FLICA+ cells than non-binding CD44+/+ or CD44-/- OT-I CD8 T cells (Fig. 4.6D).  This indicates that CD44+/+ CD8 T cells are more susceptible to apoptosis than CD44-/- CD8 T cells. Furthermore, the HA-binding population are more apoptotic and more susceptible to death than the non-binding population. This finding suggests that an interaction between CD44 on the effector T cells and HA within the lymphoid organs may augment the signal that induces apoptosis and contraction of OT-I CD8 T cells.    85    Figure 4.6 CD44-/- OT-I CD8 T cells are less susceptible to apoptosis Recipient CD45.1 WT mice received 5 x 104 of equal mixture of CD45.1/ 45.2 CD44+/+ and CD45.2 CD44-/- OT-I CD8 T cells, and were infected intravenously with 30,000 CFU of LM-OVA a day later. Dead cells were identified by positive staining with DAPI or LIVE/DEAD® Fixable Viability Dye – Aqua. (A) Percent live cells within the transferred CD44+/+ and CD44-/- OT-I CD8 T cell populations 7 and 10 days post infection. (B) Percent FLICA+ within the transferred CD44+/+ and CD44-/- OT-I CD8 T cell populations 10 days post infection. (C) Percent live cells within the transferred HA-binding CD44+/+, non-binding CD44+/+ and CD44-/- OT-I CD8 T cell populations 10 days post infection. (D) Percent FLICA+ within the transferred HA-binding CD44+/+, non-binding CD44+/+ and CD44-/- OT-I CD8 T cell populations 10 days post infection. Means and SD from at least two independent experiments with at least six biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 calculated by two-tailed Student’s t-test.   86   4.2.7 HA-binding OT-I CD8 T cells exhibit an increased ability for glucose uptake CD8 effector proliferation is linked to reduced memory potential (Cui et al., 2009; Kaech et al., 2002; Kinjyo et al., 2015), suggesting a common pathway might account for the enhanced proliferative state of effector CD8 T cells and their increased susceptibly to death during contraction. The two could be linked through cellular metabolism, since glycolysis sustains effector T cell proliferation (Greiner et al., 1994; Jacobs et al., 2008), but also inhibits the formation of memory T cells, which requires a switch from glycolysis to OXPHOS (O’Sullivan et al., 2014; van der Windt et al., 2012). Based on this, HA-binding CD8 effector cells, being more actively proliferating but also more susceptible to death during contraction, are expected to have increased glycolysis.  To investigate this, recipient mice that received equal proportions of CD45.1/ 45.2 CD44+/+ and CD45.2 CD44-/- followed by intravenous LM-OVA infection, were injected with the fluorescent glucose analogue 2-NBDG intravenously 15 min prior to euthanasia (Blagih et al., 2015).  2-NBDG is non-metabolizable, and its uptake is used to estimate intracellular glycolytic flux (TeSlaa and Teitell, 2014). In the pLN and BM, a significantly higher percentage of HA-binding CD44+/+ cells were labelled by 2-NBDG, compared to non-binding CD44+/+ and CD44-/- OT-I CD8 T cells from the same recipient mice at the peak of the T cell response on day 7 (Fig. 4.7A, B). Furthermore, a portion of the 2-NBDG+ cells had increased 2-NBDG labelling, and between 40-50% of the HA-binding CD44+/+ OT-I CD8 T cells were 2-NBDGhi in the pLN and BM, while only approximately 5% of non-binding CD44+/+ and CD44-/- OT-I CD8 T cells were 2-NBDGhi (Fig. 4.7C). Similar trends were also observed 10 days post infection, after the contraction phase (Fig. 4.7B, C), suggesting an inability to turn off glycolysis in HA-binding    87   CD8 T cells during memory formation. Larger and significant differences in 2-NBDG uptake were observed in the pLN and BM, but not spleen, consistent with the fact that the difference between CD44+/+ and CD44-/- OT-I CD8 T cells in competition was more prominent in the pLN and BM.  4.2.8 HA-binding OT-I CD8 T cells have increased expression of pAkt In CD8 T cells, PI3K/ Akt up-regulates glycolysis upon T cell activation and sustains glycolysis in effector CD8 T cells (Frauwirth et al., 2002; Jacobs et al., 2008; Wieman et al., 2007), and the switch from glycolysis to OXPHOS in memory formation is mediated through the inactivation of PI3K/ Akt during the contraction phase (Edinger and Thompson, 2002; Finlay et al., 2012; Wieman et al., 2007). The inactivation of Akt is important for CD8 memory cell formation, as sustained pAkt signal reduces CD8 T cell memory numbers without effecting effector cell numbers (Kim et al., 2012). Based on this, HA-binding CD44+/+ CD8 T cells, which exhibited higher glucose uptake, were expected to have increased pAkt expression. Levels of pAkt were measured in competitively transferred CD45.1/ 45.2 CD44+/+ and CD45.2 CD44-/- OT-I CD8 T cells, 7 days post LM-OVA infection, after ex vivo restimulation with SIINFEKL for 4 hr. When compared to non-binding CD44+/+ and CD44-/- OT-I CD8 T cells from the same hosts, HA-binding CD44+/+ OT-I CD8 effector T cells exhibited a slight increase in pAkt expression, and the difference was significant in cells isolated from the BM (Fig. 4.7D, E). Similar trends were also observed in the pLN, although the difference was not significant (Fig. 4.7 E). This finding is consistent with our hypothesis that CD44-mediated HA binding enhances an apoptotic signal and reduces memory fate programming, by augmenting Akt activation.    88    Figure 4.7 HA-binding WT OT-I CD8 T cells have increased 2-NBDG uptake and increased pAkt expression    89   Recipient CD45.1 WT mice received 5 x 104 of equal mixture of CD45.1/ 45.2 CD44+/+ and CD45.2 CD44-/- OT-I CD8 T cells, and were infected intravenously with 30,000 CFU of LM-OVA a day later. (A) Flow cytometry plot of 2-NBDG over CD8 of PBS control, HA-binding and non-binding CD44+/+, and CD44-/- OT-I CD8 T cells in the pLN 7 days post infection. Mice received 100 μg of 2-NBDG intravenously 15 min prior to euthanasia. (B) Percent 2-NBDG+ of transferred HA-binding and non-binding CD44+/+, and CD44-/- OT-I CD8 T cells 7 days post infection. (C) Percent 2-NBDGhi of transferred HA-binding and non-binding CD44+/+, and CD44-/- OT-I CD8 T cells 7 days post infection. Means and SD from two independent experiments with six total biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 as calculated by two-tailed Student’s t-test. (D) Flow cytometry plot of pAkt expression of HA-binding and non-binding CD44+/+ OT-I CD8 T cells in the BM, pLN and spleen 7 days post infection. Staining by the isotype antibody is shown as a negative control. (E) Mean fluorescence intensity of pAkt expression of HA-binding CD44+/+ and CD44+/+, and CD44-/- OT-I CD8 T cells 7 days post infection. Means and SD from two independent experiments with eight total biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 as calculated by two-tailed Student’s t-test.   90   4.2.9 The HA-binding CD8 effector population is selected against during contraction As HA-binding CD8 T cells were found to be more prone to apoptosis during contraction, they were expected to be selected against during memory formation. To investigate this, the size of the HA-binding population was measured at different time points in LM-OVA infected mice. In competitive adoptive transfer experiments with CD44+/+ and CD44-/- OT-I CD8 T cells, 20-40% of the CD44+/+ OT-I CD8 T cells bound FL-HA on day 5, and the percent decreased to 5-10% by day 20 (Fig. 4.8A). The profile of percent HA-binding cells over the infection course supported the idea that HA-binding on CD8 effector T cells is selected against.  However, the reduction in percent HA-binding could also be due to loss of HA-binding. To determine if the HA-binding population was indeed selected against, HA-binding and non-binding CD8 T cells were sorted from in vitro activated CD45.1/ 45.2 or CD45.2 WT splenic CD8 T cells, and were then competitively transferred into CD45.1 WT mice that had been infected with LM-OVA 5 days prior. Based on expression of the CD45 alleles, significantly higher percentages of the initially non-binding CD8 T cell population was recovered from the pLN and spleen 11 days post infection, after the contraction phase (Fig. 4.8 B, C). The competitive advantage of non-binding CD8 T cells over HA-binding cells was independent of the allelic effects of CD45, as the non-binding population out-competed the HA-binding population regardless of the CD45.1/ 45.2 or CD45.2 background (Fig. 4.8B). On average, the non-binding population out-competed the HA-binding population with a ratio of 3-to-1 (Fig. 4.8C). At the time of analysis, 6 days post adoptive cell transfer, little HA-binding was detected in cells that used to bind FL-HA (Fig. 4.8D), suggesting that the loss of HA binding enabled these cells to survive the contraction phase. These findings suggest that the interaction between    91    Figure 4.8 The HA-binding CD8 effector population is selected against during contraction (A) Percent HA binding by transferred CD44+/+ OT-I CD8 T cells, as assessed by binding to FL-HA. Recipient CD45.1 WT mice received 5 x 104 of equal mixture of CD45.1/ 45.2 CD44+/+ and CD45.2 CD44-/- OT-I CD8 T cells, and were infected intravenously with 30,000 CFU of LM-OVA a day later. Means and SD from at least two independent experiments with at least six total biological replicates shown. (B-D) CD8 T cells were isolated from CD45.1/ 45.2 and CD45.2 WT mice and activated with PMA and ionomycin for 2 days. Cells were then sorted into HA-binding and non-binding populations based on binding to FL-HA, and then 5 – 8 x 105 cells of equal mixture of CD45.2 HA-binding and CD45.1/ 45.2 non-binding, or CD45.1/ 45.2 HA-binding and CD45.2 non-binding CD8 T cells were adoptively transferred into CD45.1 WT recipients, which had been intravenous infected with LM-OVA 5 days prior. Mice were euthanized for analysis 6 days after T cell transfer. (B) Flow cytometry plots of CD45.1 versus CD45.2 for CD45.2 HA-binding versus CD45.1/ 45.2 non-binding (left) and CD45.1/45.2 HA-binding versus CD45.2 non-binding competitions (right). (C) Percent of transferred cells derived from HA-binding and non-binding CD8 T cells. (D) HA binding of transferred cells as assessed by binding to FL-HA. Means and SD from at two independent experiments with at ten total biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 as calculated by two-tailed Student’s t-test.    92   CD8 effector cells and HA through CD44 increases the susceptibility of these cells to apoptosis and thereby reduces their memory potential, and HA-binding CD8 T cells might be able to increase resistance to cell death by disengaging from the HA-rich niche through the loss of HA-binding.   4.3 Discussion Here, by using a murine model for systemic bacterial infection and OT-I CD8 T cells, I found that the absence of CD44 enhanced memory CD8 T cell formation. The extent of HA binding by the activated CD8 T cells was found to correlate positively to the strength of TCR stimulation, and HA-binding CD8 effector T cells exhibited increased pAkt expression, increased glucose uptake, and increased susceptibility to cell death. These findings suggest a model whereby the interaction between CD44 and HA promotes the loss of memory potential in CD8 effector T cells and provides a means of limiting the lifespan of strongly activated effector CD8 T cells.  CD44-mediated cell-HA interaction depends on the ability for the cells to bind HA. HA binding by OT-I CD8 effector T cells was transiently increased upon activation and was heterogeneous, where only a subset of the population became bound FL-HA, and the highest amount of FL-HA binding was observed on cells activated with the highest peptide affinity and avidity in vitro. This suggests that the most strongly activated CD8 effector T cells are most likely to bind HA. HA-binding effector T cells are previously found to be more actively proliferating (Maeshima et al., 2011). In this study, the HA-binding effector CD8 T cells were also found to exhibit increased glucose uptake, suggesting that they have increased glycolytic output, which would support their proliferative state. The HA-binding effector CD8 T cells also had increased    93   expression of pAkt, a positive regulator of glycolysis (Lunt and Vander Heiden, 2011). Stimulation of tumour cells with antibodies against the HA binding site of CD44 increased Akt activation, suggesting that CD44 may signal to Akt upon engaging HA. This signalling event may be though Lck, as it is associated with CD44 and cross-linking CD44 activated Lck in T lymphoma cells and human peripheral T lymphocytes (Li et al., 2001; Taher et al., 1996; Wong et al., 2008).  This study finds that HA-binding effector CD8 T cells had reduced memory potential, and that high TCR peptide affinity and avidity resulted in more HA-binding cells. These findings leads to the inference that T cells that have received the strongest TCR signal during activation are less likely to become memory cells. Our data therefore supports the model of decreasing potential of memory differentiation, where the weakly activated effector T cells are the predominant contributors to the memory pool, and the strongly activated effector T cells become terminally differentiated and fail to survival through the contraction phase (Kaech and Cui, 2012; Restifo, 2014). Transcriptomic profiling of CD8 T cells at different stages of an immune response find that memory CD8 T cells are the transcriptional intermediate between naïve and effector CD8 T cells (Best et al., 2013; Holmes et al., 2005; Kaech et al., 2002), supporting the idea that CD8 T cells lose memory potential to become effector cells. Our data suggest that through the interaction with HA, CD44 plays a role in limiting the lifespan strongly activated effector CD8 T cells, thereby providing a mechanism for CD8 T cell fate decision. An alternative model of CD8 memory formation is the idea that activated CD8 T cells divide asymmetrically during the first division, producing two daughter cells with distinct fates. The proximal daughter cell gives rise to terminally differentiated effector cells, while the distal    94   daughter cell gives rise to memory cells (Chang et al., 2007). In that study, a bifurcated pattern of CD44 expression was observed after the first division, suggesting the CD44 is asymmetrically partitioned (Chang et al., 2007). This asymmetric partitioning of CD44 upon activation could therefore explain why only a subset of activated CD8 T cells bind HA (Maeshima et al., 2011). However, our lab has not been able to replicate this bifurcated CD44 expression after the first division with in vitro or in vivo activated CD8 T cells. Instead I found evidence for bifurcated CD44 expression after the second division. Since I also found that HA-binding effector CD8 T cells are less likely to form memory cells, it is possible that the CD44hi population observed after the second division is more likely to be selected against during the contraction phase. Therefore, the two populations with distinct CD44 expression levels may ultimately have different fates. However, it is still unclear if the bifurcated CD44 expression is due to asymmetric partition of CD44 or stochastic variations of CD44 expression.  In another study, using OT-II transgenic CD4 T cells, the expression of CD44 is shown to confer a competitive advantage in the generation of CD4 memory T cells after a Th1 response to influenza infection (Baaten et al., 2010). The contradictory effects of CD44 expression in memory formation of CD4 and CD8 T cells may be due to differences in the role of Akt and glycolysis on memory formation. In CD8 T cells, sustained pAkt signalling and glycolysis negatively regulates memory formation (Kim et al., 2012; Sukumar et al., 2013). In contrast, CD4 memory survival requires sustained glucose uptake mediated through Akt (Maekawa et al., 2015). Another potential factor that might contribute to the discrepancy is the affinity of the transgenic TCR used: while the OT-I TCR has high affinity to SIINFEKL, the OT-II TCR has low affinity to its peptide. Furthermore, it is unknown if the OT-II CD4 effector T cells bound    95   HA, and if HA-binding marked CD4 effector T cells with a different cell fate. Furthermore, in cells that express a high affinity TCR and therefore experience strong TCR signalling, but not in cells expressing a low affinity TCR, the augmentation of Akt activation upon binding to HA might be enough to push cells over the threshold for reduced lifespan. Here, I have shown a role of CD44 and HA binding in the negative regulation of OT-I CD8 T memory cell formation. HA binding and CD44 expression in activated OT-I CD8 T cells were also shown to positively correlate with the TCR peptide affinity and avidity during activation. My findings therefore support the decreasing potential model of memory differentiation, and provide a mechanism by which the CD8 T cell fate can be determined.     96   Chapter 5: CD44-Mediated HA Binding Provides a Competitive Advantage in Bone Marrow Engraftment and Reconstitution  5.1 Introduction and rationale In the hematopoietic compartment, expression of the glycoprotein CD44 and the binding to its ligand, HA, is tightly regulated (reviewed in (Lee-Sayer et al., 2015)). The interaction between HA and CD44 is implicated in regulating HSPC functions by several studies. For example, BM stromal cells from mice deficient in all three HAS have reduced ability to maintain HSC numbers and reduced reconstitution capability (Goncharova et al., 2012), and exogenous HA aids in bone marrow cell recovery from chemical-induced bone marrow ablation (Matrosova et al., 2004).  However, little is known about the role of CD44 and HA in the development of mature immune populations. The lack of major developmental defects in CD44-/- mice suggests that CD44 is dispensable for hematopoietic development. However the role of CD44 might be to retain specific cells to specific HA-rich niche, and therefore the effect of CD44 deletion or altered HA binding might not be observed unless cells are competing for HA. Furthermore, there is evidence that the deletion of CD44 could be compensated for. For example, CD44-/- splenic T cells exhibited increased binding to fibronectin (Maeshima, 2010), ICAM-1 is shown to compensate for CD44 as the co-receptor for the tyrosine kinases c-Met (Olaku et al., 2011), and the reported alternate HA receptor known as receptor for hyaluronan-mediated motility (RHAMM) is also suggested to compensate for CD44 deletion for cell migration in a model of arthritis (Nedvetzki et al., 2004).     97   The goal of this study is to determine if CD44 and HA binding have a role in specific stages of immune cell development. Given that the effects of HA binding may not be observed unless under competition, I investigated the effect of HA binding in immune development by competing HA-binding and non-binding BMC. To achieve this, WT and CD44-/- BMC were competitively transferred. Furthermore, I made use of CD44 mutants that harbour either increased or abolished HA binding (gain and loss of function, or GOF and LOF, respectively). LOF-CD44 harbours the point mutation R43A, since R43 on mouse CD44 is identified as a key residue in HA binding in alanine-scanning mutagenesis and structural studies of CD44 (Peach et al., 1993; Takeda et al., 2003). GOF-CD44 harbours the point mutation S183A, the site of chondroitin sulfate addition on mouse CD44, which is shown to negatively regulate HA binding by CD44 (Ruffell and Johnson, 2005).   5.2 Results 5.2.1 Hematopoietic cells exhibit different HA binding profiles in steady state While CD44 expression level and HA binding status for some immune cell populations, such as human monocytes and several macrophage populations, have been well characterized (Lee-Sayer et al., 2015), characterization of other populations is lacking. For example, little is known about the HA binding status of eosinophils and NKT cells. Here, binding to FL-HA was used to assess HA binding by various immune cell populations, which were identified based on gating strategies outlined in Fig. 5.1. After lysis of RBC, cells from the BM, spleen and blood of WT mice could be divided into a CD44hi and a CD44lo population (Fig. 5.2A). To assess the percent of HA binding cells, cells from CD44-/- mice were used as negative controls, and less than 10% of isolated BMC, splenocytes and blood cells were found to bind FL-HA (Fig. 5.2A).      98    Figure 5.1 Gating strategies Debris, dead and doublet cells were first excluded based on SSC, FSC-A, FSC-H and viability dye staining. (A) Gating strategy for lymphoid cells. αβT, NKT and T gating was based on expression of NK1.1 and TCRβ. NK1.1- TCRβ- B220+ cells in the periphery were designated as B cells. (B) Gating strategy for myeloid cells. Neutrophils (SSCmid Gr1hi), monocytes (SSClo Gr1mid) and eosinophils (SSChi Gr1mid) were found within the CD11c- population. SSChi Gr1mid eosinophils were verified as Siglec-F+. (C) Gating strategy for HSCP. HSPC were within the lineage- (lineage included CD11b, CD11b, CD19, B220, CD49b, TCRβ, γδTCR, MHC II, Ter119 and Gr1) population in the bone marrow. CD150+ HSC and CD150- MPP were within the LSK (lineage- Sca-1+ c-Kit+) population. GMP were c-Kithi Sca-/lo cells within the CD16/32+ population, and CLP were c-Kitmid Sca-1mid cells within the CD127+ population.    99    Figure 5.2 CD44 expression and HA-binding by mature immune populations and HSPC from WT and CD44-/- mice HA-binding was assessed by binding to FL-HA. HA-binding cells were identified by gates drawn based on the FL-HA staining of CD44-/- cells, which do not bind HA. Percent HA binding    100   of representative samples are shown within the gates. (A) Flow cytometry plots of FL-HA binding versus CD44 expression by WT or CD44-/- BMC, splenocytes and blood cells (left), and average percent FL-HA binding by WT BMC, splenocytes and blood cells (right). (B) Flow cytometry plots of FL-HA binding versus CD44 expression by neutrophils, eosinophils and monocytes in WT or CD44-/- blood (left), and average percent FL-HA binding by neutrophils, eosinophils and monocytes of WT mice (right). (C) Flow cytometry plots of HA binding versus CD44 expression by αβT, NKT, NK and B (total B220+, which includes B cell progenitors in the BM) cells in WT or CD44-/- spleens (left), and average percent HA binding by αβT, NKT, NK and B cells of WT mice (right). (D) Flow cytometry plots of HA binding versus CD44 expression by LSK, MPP, CLP, MP and HSC of the WT BM (top), and their average percent FL-HA binding (bottom). Means and SD from at least two independent experiments with at least six biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 as calculated by two-tailed Student’s t-test.    101   Monocytes and neutrophils in the BM, spleen and blood expressed high CD44, but less than 10% bound FL-HA (Fig. 5.2B). In contrast, eosinophils, which also expressed high CD44, exhibited 20-60% FL-HA binding (Fig. 5.2B). In agreement with published findings, αβT and B cells from naïve mice expressed low CD44 and less than 10% bound FL-HA (Fig. 5.2C). NK and NKT cells have high CD44 expression, and 20-40% of NK and NKT cells from naïve mice bound FL-HA (Fig. 5.2C).  The LSK population expressed high CD44 and a small subset of the population bound FL-HA (Fig.2.2D). Interestingly, while similar percentages of cells within the MPP, CLP and GMP populations bound FL-HA at around 25%, significantly reduced percentage of HSC bound FL-HA at around 7% (Fig.5.2D). Furthermore, the percentages of FL-HA binding cells by the MPP population were always higher than percentages of FL-HA binding cells by the HSC population of the same mouse, despite the large biological variation in amount of FL-HA binding observed (Fig. 5.2D).  5.2.2 CD44 is dispensable for early hematopoietic development The observed induction of HA binding in specific hematopoietic compartments suggests that CD44 and HA binding might have important developmental roles in the HA-binding populations. The inability to bind HA may therefore be detrimental to hematopoietic development, and CD44-/- mice are thus expected to have developmental defects in their hematopoietic compartment, especially in cells that bind HA in homeostasis. However, no major developmental defects have been reported for CD44-/- mice. Furthermore, CD44-/- were found to have similar BM cellularity as WT (Fig. 5.3A). Number of lineage-, LSK, MPP and HSC    102   populations were also not different between WT and CD44-/- mice (Fig. 5.3B), suggesting that BM hematopoiesis in CD44-/- mice is normal.  To further investigate the role of CD44, an equal mixture of total BMC from WT and CD44-/- mice were competitively transferred into lethally irradiated hosts, to compare the reconstitution of different immune compartments by WT and CD44-/- BMC. After 7 weeks, donor-derived cells reconstituted over 95% of the BM, thymus, spleen, pLN and blood (Fig. 5.4A). A slight but significant competitive advantage was observed for WT reconstitution of the thymus and mature T cell populations (Fig. 5.4B, C). WT cells also had a competitive advantage in reconstituting NK cells in the spleen and pLN, although the differences were smaller than that for T cells (Fig. 5.4C). No significant difference between WT and CD44-/- derived cells was observed for B cells, monocytes, neutrophils, F4/80+ and CD11c+ DC (Fig. 5.4C, D). A slight competitive advantage for WT reconstitution of eosinophils was observed, but only in the spleen (Fig. 5.4C).  WT and CD44-/- BMC were able to reconstitute the BM lineage- and LKS populations similarly, as no competitive difference between WT and CD44-/- was observed in their reconstitution of the BM lineage- and LSK populations (Fig. 5.4E).  The lack of competitive difference in BM HSPC reconstitution by WT and CD44-/- BMC suggest that the slight competitive advantage of WT BMC-derived cells in thymocytes, T cells, NK cells and splenic eosinophils occurred further downstream in hematopoietic development.     103    Figure 5.3 Cell numbers of total BMC, and BM HSPC populations in WT and CD44-/- mice Single cell suspensions were made from the BM from the tibia and femur of one leg of each mouse. (A) Number of total BMC, lineage- and LSK populations in one leg. (B) Number of MPP and HSC in one leg. Means and SD from at two independent experiments with at least six biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 as calculated by two-tailed Student’s t-test.    104    Figure 5.4 Competitive BM reconstitution by WT and CD44-/- BMC into irradiated congenic WT mice Lethally irradiated CD45.1 WT mice were reconstituted with 8 x 106 of equal mixture of CD45.1/ 45.2 WT and CD45.2 CD44-/- BM for 7 weeks. (A) Percent reconstitution by total CD45.1+ donor-derived cells. (B) Percent of CD45.1/ 45.2 WT and CD45.2 CD44-/- cells within the donor-derived populations from each organ. (C) Percent of CD45.1/ 45.2 WT and CD45.2 CD44-/- cells within the donor-derived αβT, B and NK cell populations from each organ. (C) Percent of CD45.1/ 45.2 WT and CD45.2 CD44-/- cells within the donor-derived monocyte, neutrophil and eosinophil populations from each organ. (D) Percent of CD45.1/ 45.2 WT and CD45.2 CD44-/- cells within the donor-derived F4/80+ macrophages and CD11c+ DC populations from each organ. (E) Percent of CD45.1/ 45.2 WT and CD45.2 CD44-/- cells within the donor-derived lineage- and LSK populations from each organ. Means and SD from two independent experiments with eight biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 as calculated by two-tailed Student’s t-test.    105   5.2.3 HA binding positively impacts reconstitution in competitive BM transfer As CD44-deletion may be compensated for by other molecules, the effect of HA binding was analyzed in the context of CD44 expression. To this end, LOF-CD44 or GOF-CD44 mutants were utilized. These constructs also contained either a GFP or RFP bicistronic reporter, and were transduced into CD44-/- bone marrow cells using a retrovirus vector. WT BMC were transduced with the empty vector (tv-WT), and CD44-/- BMC were transduced with either the LOF-CD44 construct (tLOF) or the GOF-CD44 construct (tGOF). GFP and RFP reporter expression correlated with CD44 expression (Fig. 5.5A). HA binding in transduced cells corresponded with the CD44 expressed. Less than 10% of tv-WT bound FL-HA, whereas tLOF did not bind FL-HA, t GOF had increased HA binding of approximately 50% (Fig. 5.5B). CD44 expression levels on transduced cells were similar to that of WT BMC transduced with the empty vector (Fig. 5.5C).  After sorting for transduced cells, tv-WT were mixed in equal mixture tLOF or tGOF, and were transferred into congenic recipient mice that were lethally irradiated. Competition between tv-WT and tLOF was performed with GFP-tv-WT against RFP-tLOF, as well as with RFP-tv-WT against GFP-tLOF; competition between tv-WT and tGOF was performed with RFP-tv-WT against GFP-tLOF (Fig. 5.6A). After 7 or 11 weeks, the donor-derived cells from the tv-WT/ tLOF and the tv-WT/ tGOF BM transfer reconstituted 70-95% and 45-80%, respectively, of the BM, thymus, spleen, pLN and blood (Fig. 5.6B). Percent reconstitution was highest in the thymus, at around 80-95%, and lowest in the blood, at around 45-70% (Fig. 5.6B).     106    Figure 5.5 Retroviral transduction of CD44 constructs into CD44-/- BMC  BMC were isolated from mice that had received 5-flourouracil 4 days prior, and were then cultured with 100 ng/ mL SCF, 6 ng/ mL IL-3 and 10 ng/ mL IL-6 for 4 days. (A) Flow cytometry plots of GFP or RFP expression versus CD44 expression. CD44-/- BMC were transduced with GOF-CD44 or LOF-CD44 constructs or with an empty vector. (B) Flow cytometry plots of FL-HA binding versus CD44 expression of WT and CD44-/- BMC transduced with an empty vector, or CD44-/- BMC transduced with the LOF-CD44 or GOF-CD44 constructs. (C) CD44 expression of WT BMC transduced with an empty vector, or CD44-/- BMC transduced with LOF-CD44 or GOF-CD44 constructs after sorting based on reporter expression.    107    Figure 5.6 Competitive BM reconstitution by tv-WT and tLOF BMC or tv-WT and tGOF BMC transferred into irradiated congenic WT mice  Lethally irradiated CD45.1 WT mice were reconstituted with 2 x 105 of equal mixture of CD45.2 tv-WT and tLOF or tGOF for 7 or 11 weeks. (A) Flow cytometry plots of GFP expression versus RFP expression of the CD45.2+ donor-derived cells. GFP+ tv-WT versus RFP+ tLOF (top), and RFP+ tv-WT versus GFP+ tLOF (middle), and GFP+ tv-WT versus RFP+ tGOF (bottom). (B) Percent of tv-WT and tLOF (left) or tv-WT and tGOF (right) within the donor-derived populations from each organ. (C) Percent of reconstitution by total CD45.2+ donor-derived cells. Means and SD from two independent experiments with eight biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 as calculated by two-tailed Student’s t-test.    108   Percent reconstitution by the tv-WT- versus tLOF- or tGOF-derived cells was inferred based on GFP and RFP reporter expression after excluding host cells in the analysis (Fig. 5.6A). In the tissues examined – BM, thymus, spleen, pLN and blood – tv-WT-derived cells reconstituted between 70-95%, when in competition with tLOF (Fig.5.6A, C). In contrast, tv-WT only reconstituted 10-15% of the tissues examined, when in competition with tGOF (Fig. 5.6A, C). The differences in the two competitions were significant for all tissues examined (Fig. 5.6C). Thus, tGOF outcompeted tv-WT, which in turn outcompeted tLOF, presumably at an earlier stage of hematopoiesis, as the BM was over 90% tGOF, when in competition with tv-WT, and was over 90% tv-WT, when in competition with tLOF (Fig. 5.6C).  5.2.4 HA binding positively impacts reconstitution of the myeloid and lymphoid populations in competitive BM transfer Similar trends were observed when individual myeloid populations were examined. When in competition with tLOF, tv-WT reconstituted over 75% of the donor-derived monocytes, neutrophils BM, spleen and blood, and F4/80+, CD11c+ and CD11c+ MHCIIhi cells in the BM and spleen, and the differences were significant for all populations, except for eosinophils, due to larger biological variations (Fig. 5.7A). In contrast, when tv-WT were in competition with tGOF, over 85% of the donor-derived monocytes, neutrophils BM, spleen and blood, and F4/80+, CD11c+ and CD11c+ MHCIIhi cells in the BM and spleen, and the differences were significant for all populations (Fig. 5.7B).  Competitions between tv-WT and tLOF, and between tv-WT and tGOF were also examined for the lymphoid populations, by analysing T, NK, NKT and B cell populations in the BM, spleen,   109    Figure 5.7 Competitive reconstitution of the myeloid compartment by tv-WT and tLOF BMC or tv-WT and tGOF BMC transferred into irradiated congenic WT mice Lethally irradiated CD45.1 WT mice were reconstituted with 2 x 105 of equal mixture of CD45.2 tv-WT and tLOF or tGOF for 7 or 11 weeks. Percent of tv-WT and tLOF (A) or of tv-WT and tGOF (B) within in the donor-derived neutrophils, eosinophils, monocytes, F4/80+ macrophages, total CD11c+ DC and CD11c+MHCIIhi mature DC populations. Means and SD from two independent experiments with eight biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 as calculated by two-tailed Student’s t-test.    110   pLN and blood. When in competition with tLOF, tv-WT also reconstituted over half of the populations examined. However, the difference was smaller in these populations, as tv-WT only reconstituted 55-75% of the T and NK cell populations, and 65-85% of the NKT and B cell populations, and the differences in the T and NK cell populations were mostly not significant (Fig. 5.8A). In contrast, tGOF significantly out-competed tv-WT in all the lymphoid populations examined, and less than 18% of these populations were reconstituted by tv-WT (Fig. 5.8B). The lack of large difference between tv-WT and tLOF was likely due to the fact that CD44 expression is down-regulated during development of these lymphoid populations, thereby abolishing the difference between tv-WT and tLOF in terms of HA binding.  In the myeloid and lymphoid populations examined, tv-WT out-competed tLOF, and was in turn out-competed by tGOF, suggesting that HA binding either positively affected the development of all immune populations, or it conferred a competitive advantage on BM HSPC shortly after BM transfer and the observed differences in reconstitution between tv-WT and tLOF and between tv-WT and tGOF were reflective of the competitive difference upstream.   5.2.5 HA binding positively impacts reconstitution of the HSPC populations in competitive BM transfer To investigate if altered HA binding capability conferred competitive advantage or disadvantage in reconstitution of HSPC, percent tv-WT and tLOF, or tv-WT and tGOF were quantified for the BM donor-derived lineage-, LKS, CD150+ MPP and CD150- HSC populations. Similar trends in competition were also observed in the BM HSPC populations. When in competition with tLOF, tv-WT significantly out-competed tLOF, and reconstituted over 80% of the BM HSPC     111    Figure 5.8 Competitive reconstitution of the lymphoid compartment by tv-WT and tLOF BMC or tv-WT and tGOF BMC transferred into irradiated congenic WT mice Lethally irradiated CD45.1 WT mice were reconstituted with 2 x 105 of equal mixture of CD45.2 tv-WT and tLOF or tGOF for 7 or 11 weeks. Percent of tv-WT and tLOF (A) or of tv-WT and tGOF (B) within the donor-derived αβT, NKT, NK and B (B220+CD19+IgM+ in the BM; B220+ in the spleen and blood) populations. Means and SD from two independent experiments with eight biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 as calculated by two-tailed Student’s t-test.    112    Figure 5.9 Competitive reconstitution of BM HSPC by tv-WT and tLOF BMC or tv-WT and tGOF BMC transferred into irradiated congenic WT mice (A) Lethally irradiated CD45.1 WT mice were reconstituted with 2 x 105 of equal mixture of CD45.2 tv-WT and tLOF or tGOF for 7 or 11 weeks. Percent of tv-WT and tLOF (left) or of tv-WT and tGOF (right) of donor-derived cells within lineage-, LSK, MPP and HSC populations. Means and SD from two independent experiments with at least six biological replicates are shown. *p<0.05, **p<0.01, ***p<0.001 as calculated by two-tailed Student’s t-test.     113   populations (Fig. 5.9A). In contrast, tv-WT was significantly out-competed by tGOF, and reconstituted less than 15% of the BM HSPC (Fig.5.9B). These competition studies revealed a role for HA binding in reconstituting the BM HSPC in BM transfer, where the increased ability to bind HA conferred a competitive advantage to the BMC. However, only 5 and 25% of HSC and MPP bound FL-HA in homeostasis, potentially calling the effects of HA binding into question.   5.2.6 HSPC increase their HA binding when induced to proliferate Despite the low percent of FL-HA binding in HSPC during homeostasis, it is possible that HSPC increase HA binding in response to environmental changes. A possible signal that can increase HA binding is proliferative signals. MPP actively turnover while HSC are quiescent, and a higher percentage of MPP was found to bind FL-HA than by HSC, suggesting that proliferative signals may increase HA binding in HSPC. To test this, lineage-depleted BMC were cultured with either SCF alone, or a mixture of SCF, IL-3 and IL-6, to induce their proliferation, and percentages of cells that bound FL-HA were quantified 3 days later. While ex vivo cytokine stimulation had little effect on percent HA binding by the LSK and MPP populations, percent HA binding by the HSC was significantly increased from around 7% to above 70% (Fig. 5.10A).  To investigate if induction of proliferation also increased HA binding by HSPC in vivo, mice were intravenously infected with Listeria monocytogenes (LM), as systemic bacterial infections are shown to induce proliferation of HSPC in the BM (Baldridge et al., 2010; MacNamara et al., 2011). Total BMC numbers were transiently reduced on day 3 in infected animals, and then returned to normal by day 5 (Fig. 5.10B). Similar trends in cell numbers were also observed for    114                  Figure 5.10 HA-binding by BM HSPC is induced by proliferation (A) Percent FL-HA binding by LSK, MPP and HSC before and after ex vivo culture in 100 ng/ mL SCF or in 100 100 ng/ mL SCF, 6 ng/ mL IL-3 and 10 ng/ mL IL-6 for 3 days. HA binding was assessed by binding to FL-HA. (B-E) BMC were isolated from the femur and tibia of uninfected control mice, or mice that were intravenously infected with 30,000 CFU of LM 3, 5 or 7 days prior. HA binding was assessed by binding to FL-HA. (A) Number of total (left), MPP (middle) or HSC (right) in the BM of the tibia and femur from both legs of each mouse. (C) Flow cytometry plots of FL-HA binding versus CD44 expression by MPP (left) and HSC (right) in control and infected animals 5 days post infection. (D) Percent of FL-HA binding by MPP (left) and HSC (right) in control and infected animals 3, 5 or 7 days post infection. Means and SD    115   from at least two independent experiments with at least six biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 as calculated by two-tailed Student’s t-test.    116   the MPP and HSC populations, but the changes were not significant due to large biological variations (Fig. 5.10B). The transient reduction in HSPC cell numbers on day 3 was likely due the loss of these cells to differentiation in response to the increased needs for immune cells during an infection. Therefore, HSPC likely experienced increased proliferation around days 3 to day 5, when the HSPC niche was not fully occupied. On day 5 post infection, a distinct FL-HA binding population was observed in BM MPP and HSC populations (Fig. 5.10C), and the percentage of FL-HA binding MPP and HSC was significantly higher in the infected animals 3 days post infection, where it increased from 18 to 50% for MPP and from 7 to 50% for HSC (Fig.5.10D). On day 5 post infection, the difference between infected and control mice in percentage of FL-HA binding by the HSC population was also significant, and was increased from 10 to 30% (Fig. 5.10D). Percentage of FL-HA binding in MPP also increased from in the infected mice 5 days post infection, but the difference was not significant (Fig. 5.10D). By day 7 post infection, percentage of FL-HA binding in the infected animal was reduced to control levels (Fig.5.10D).  5.2.7 HA increases cell number of proliferating HSC and MPP To test if the interaction with HA would then have an effect on HA-binding HSPC, lineage-depleted BMC were cultured with SCF, in the presence or absence of exogenous HA ex vivo. After 3 days, BMC cultured with exogenous HA had significantly higher numbers of LSK and MPP as compared to BMC cultured without HA (Fig. 5.11). A slight increase in cell number was also observed in HSC cultured in the presence of exogenous HA, but this difference was not statically significant (Fig. 5.11).     117    Figure 5.11 Exogenous HA increases cell number of proliferating HSPC  Lineage-depleted BMC were cultured in 100 ng/ mL SCF with or without 5 or 50 µg/ 0.3 cm2 of HA. Cells numbers were quantified 3 days later. Means and SD from one representative experiment o two with three biological replicates shown. *p<0.05, **p<0.01, ***p<0.001 as calculated by two-tailed Student’s t-test.    118   Overall, I found that HA binding was induced in HSPC by signals that induced their proliferation, and HA could then in turn further increase the cell number of proliferating HSPC, presumably through either enhanced proliferation or survival. This effect of HA on proliferating HSPC may also account for the competitive advantage of HA-binding HSPC in hematopoietic reconstitution following BM transfer into irradiated mice.  5.3 Discussion In this study, CD44 expression and HA binding in various myeloid, lymphoid and HSPC populations were characterized, and NK cells NKT cells, eosinophils and the LSK population exhibited a distinct HA-binding population. HA binding by resting NKT and eosinophils has not been previously characterized, and the finding that a subset of these cells binds HA is novel and surprising, given that few immune cell types bind HA when resting (Lee-Sayer et al., 2015). Contrary to findings from another study (Sague et al., 2004), NK cells were also found to bind HA, and this discrepancy may be to do variations in the sensitivity in the reagents used in detecting HA binding.  Induction of HA binding has been observed upon activation of various immune cell types, including T, B and NK cells, macrophages, monocytes and monocyte-derived DC (reviewed in (Lee-Sayer et al., 2015)). These activation signals also induce proliferation of the cells in many cases, and in activated T cells, the HA-binding T cells are the most actively proliferating (Maeshima et al., 2011), suggesting a link between induction of HA binding and the proliferative state of the cell. Here, HSPC also increased HA binding in response to proliferative signals. In the BM during homeostasis, HSC exhibited significantly reduced HA binding as compared to    119   MPP, CLP and GMP, which had similar percentages of HA-binding cells to each other. Since HSC are quiescent, while MPP, CLP and GMP actively turnover (Akashi et al., 2000; Cheshier et al., 1999; Kondo et al., 1997), our data indicated that HA binding is up-regulated as HSPC become more proliferative. Furthermore, HSPC can be induced to proliferate in vitro by culturing with cytokines, or in vivo with systemic bacterial infection (Baldridge et al., 2010; MacNamara et al., 2011), both of which also increased percent of HA-binding HSPC.  To determine the role of HA binding in hematopoietic development, WT and CD44-/- BMC were competitively transferred into irradiated mice, and CD44 was found to be dispensable for hematopoietic development. Since CD44 deletion may have been compensated for (Maeshima, 2010; Nedvetzki et al., 2004; Olaku et al., 2011), the role of HA binding was studied in the context of CD44 expression. When in competition, tGOF out-competed tv-WT, which in turn out-competed tLOF, in myeloid and most lymphoid populations, as well as the BM HSPC population, suggesting that the ability to bind HA conferred a competitive advantage in the HSPC population in early hematopoiesis.  My findings suggest that the interaction with HA promotes the proliferation or survival of proliferating HSPC. The PI3K/ Akt pathway promotes both proliferation and survival of cells. In HSPC, Akt positively regulates cell cycling: double deletion of Akt1 and Akt2 reduces HSC long term reconstitution due to excessive quiescence (Juntilla et al., 2010), and constitutive pAkt signalling depletes HSC through excessive cycling (Kharas et al., 2010). CD44 ligation with an antibody induces PI3K/ Akt signalling in tumour cells (Klingbeil et al., 2009), and the interaction between CD44 and HA promotes survival of tumour cells (Ghatak et al., 2002). Proliferation and apoptosis are linked and enhanced cell cycling is associated with increased apoptosis (reviewed    120   (Alenzi, 2004)). Therefore, it is possible that HA binding is induced in HSPC in cell cycle to increase their survival.  In the BM, HA is deposited near the endosteum, and is therefore theorized to interact with HSC (Avigdor et al., 2004). Furthermore, BM stromal cells incapable of synthesizing HA have reduced abilities to support HSC numbers and function (Goncharova et al., 2012). However, we found that in homeostasis, HSC exhibited little binding to FL-HA. One possibility is that, in the absence of HA, the proliferating HSC have reduced survival and their death vacates the HSC niche, causing other HSC to enter cell cycle. Over time, the HSC pool loses stem-ness from unnecessary turnover and is gradually reduced in size.  Irradiation also causes hematopoietic stress on HSPC, by ablating mature hematopoietic cells and their progenitors, and BMC transferred into irradiated hosts proliferate to repopulate the hematopoietic compartment (Zhong et al., 2002). In this study, cells with increased HA binding ability at the time of BM transfer repopulated the over half of the HSPC, as well as the mature immune populations examined. The interaction with HA at the time of BM transfer possibly enhanced survival of the proliferative cells with enhanced HA binding in this scenario.  Despite the suggested effects of HA binding on proliferating  HSPC, CD44-/- mice have no major developmental defects (Schmits et al., 1997), and were also shown to have similar numbers of HSPC as compared to WT in this study. Several studies have suggested that function of CD44 may be compensated for by other molecules in CD44-/- cells (Maeshima, 2010; Nedvetzki et al., 2004; Olaku et al., 2011). In neonates, CD44-/- mice have reduced number of BM HSPC as compared to WT, but this difference disappears by day 7 post birth (Cao et al., 2016), further    121   supporting the existence of compensatory mechanism in CD44-/- mice, which likely takes effect shortly after birth. Alternatively, the interaction between HA and CD44 might provide a competitive advantage through enhanced BM homing or migration of the transferred BMC to the BM. Indeed, CD44 and HA have been implicated in positively regulating BMC migration to the BM of untreated mice (Avigdor et al., 2004). However, irradiation is shown to damage the BM vasculature (Cao et al., 2011), and this likely allows cells in the blood to easily flow into the BM regardless of HA binding. Since I found exogenous HA to have a direct effect on proliferating HSPC in vitro, BM homing alone is unlikely to account for the difference observed in competition. To definitely rule out effects due to homing, the competition experiment should be repeated where the cells are transferred directly into the BM through intra-femoral injection. In this study, I have shown that it is possible to alter the competitiveness of BMC in reconstitution by manipulating the affinity of CD44 to HA. Furthermore, this competitive advantage might be due to increased survival in the proliferating HSPC, downstream of the interaction between CD44 and HA. My findings suggest that the interaction between the HSPC and environmental HA is dynamically regulated, and this enables HSPC to better respond to environmental changes, such as hematopoietic stress from blood cell ablation or infections.    122   Chapter 6: Summary and Future Directions  6.1 Regulation of HA binding in macrophages and T cells  Chondroitin sulfate plays a role in regulating HA binding by both classically and alternatively activated macrophages. I found that classically activated macrophages induce HA binding through the up-regulation of CD44 and the removal of chondroitin sulfation on CD44, while alternatively activated macrophages are able to induce HA binding despite increased sulfation on CD44. Stimulation of LPS reduced chondroitin sulfation on CD44, and was able to further increase HA binding by alternatively activated macrophages.  Chondroitin sulfate is also the most likely candidate for regulating HA binding in T cells, since I ruled out other possible mechanisms including sialylation, O-linked glycosylation and actin rearrangement. This hypothesis is supported by the fact that Jurkat T cells expressing GOF-CD44 have higher HA binding than cells expressing WT CD44, suggesting that the removal of chondroitin sulfate on CD44 is sufficient to induce HA binding in T cells (Ruffell and Johnson, 2008).  To determine if chondroitin regulates HA binding in T cells, future experiments should first determine if chondroitin sulfate is present on CD44 in naïve T cells. Chondroitin sulfate can be detected by immunoblotting CD44 that is immunoprecipitated from T cell lysates (Ruffell and Johnson, 2005; Ruffell et al., 2011). I expect CD44 expressed by naïve T cells to be decorated with chondroitin sulfate, which is reduced in activated T cells. Furthermore, if chondroitin sulfate on CD44 prevents HA binding by T cells, expression of sufficient levels of CD44 that harbours the S183A mutation (S180A in humans) should be sufficient to induce HA binding in    123   naïve mouse T cells. To negate the need to activate T cells to enable retroviral transduction, lentivirus can be used for gene delivery into naïve T cells (Frecha et al., 2008). Lastly, it is also unclear if increased CD44 expression level alone is sufficient to induce HA binding by T cells. To test this, the WT CD44 construct can be transduced into naïve T cells to induce up-regulation of CD44 expression without activation.  6.2 The role of CD44 and HA binding in CD8 T cells  6.2.1 HA-binding CD44 negatively regulates CD8 T memory formation  In this study, CD44 expression and HA binding by OTI CD8 T cells were shown to confer a competitive disadvantage in memory formation in competitive situations. The CD44+/+ OTI CD8 effector T cells, in particular the HA-binding fraction, were more prone to apoptosis than the non-binding CD44+/+ OTI CD8 or CD44-/- OT-I CD8 T cells. Furthermore, the degree of HA binding induced by antigen presentation and TCR activation of naïve OTI CD8 positively correlated with the peptide avidity and affinity, suggesting that T cells receiving a stronger TCR signal during activation are less likely to form memory cells.  My data therefore support the model of decreasing memory potential for CD8 T cell memory formation (Kaech and Cui, 2012; Restifo, 2014). According to this model, CD8 T cells that have received the most combined activation signals, become the most proliferative effector T cells, but also lose memory potential and have reduced survival through the contraction phase. Weakly activated cells, in contrast, proliferate to a lesser degree and have greater memory potential, making them a minor constituent of the effector population but a major contributor to the memory population. This model is supported by studies that find shortened infection or delayed    124   exposure to antigen increases the memory formation of CD8 T cells (Badovinac and Harty, 2007; Badovinac et al., 2004; D'Souza and Hedrick, 2006; Fousteri et al., 2011). CD8 T cells activated with reduced TCR affinity have reduced effector expansion and shortened effector expansion phase (Zehn et al., 2009). Furthermore, exposure to the “signal 3” cytokine IL-12 after activation can enhance proliferation and loss of memory potential of effector CD8 T cells (Cui et al., 2009), suggesting that the more combined activation signals an effector T cell receives the less likely it will become a memory cell. In other words, effector cells that are precursors to memory cells can give rise to terminally differentiated effector cells, a model supported by several transcriptomic analyses of naïve, memory and effector CD8 T cells (Best et al., 2013; Holmes et al., 2005; Kaech et al., 2002). My findings suggest a mechanism by which this model can occur. Strong TCR signal induced HA binding mediated by CD44 (Fig. 6.1A). CD44 associates with Lck and anti-CD44 antibody ligation activate Akt (Klingbeil et al., 2009; Lefebvre et al., 2010; Taher et al., 1996; Wong et al., 2008), and I found that HA-binding CD8 effector T cells had a slight but significant increase in pAkt expression at the peak of the T cell response. This suggests that CD44 can signal to Akt upon HA engagement. The augmented Akt signal then positively regulates glycolysis, which in turn positively regulates proliferation and negatively regulates memory potential (Fig. 6.1B). The strongly activated effector CD8 T cells then are more prone to apoptosis during the contraction phase (Fig. 6.1C), leaving the weakly activated effector CD8 T cells as the major contributor to memory cells (Fig. 6.1D).       125    Figure 6.1 Model for the role of CD44 and HA in CD8 T cell response (A) Strong TCR signal induces HA binding in activated CD8 T cells, enabling these cells to bind to environmental HA. (B) Upon HA binding, CD44 signals to PI3/ Akt through Lck. This promotes glycolysis and proliferation of effector cells. (C) The more proliferative effector cells also have reduced memory potential. (D) This results in a relative small contribution to the memory pool by the strongly activated cells, relative to cells that had received a weaker TCR signal during activation.   126   6.2.2 Future directions for studying the role of CD44 and HA binding in CD8 memory T cells formation In this study, I found that HA-binding CD8 effector T cells are selected against in memory formation. However, I also found that CD8 effector T cells lose HA binding during the T cell response. While this loss of HA binding highlights the selective pressure against HA-binding CD8 T cells, it also causes difficulty in studying the effect of HA binding. Instead of relying on FL-HA staining to track HA-binding cells, another approach is to disable the loss of HA binding in effector CD8 T cells. Based on the finding that expression of GOF-CD44 is sufficient to increase HA binding in Jurkat T cells, I have hypothesized earlier that CD8 T cells regulate HA binding through chondroitin sulfation. This can be tested by transduction of the GOF-CD44 construct into naïve and activated T cells. If my predictions were true, these transduced T cells can then be used in in vivo studies. Based on my findings, I would predict that the GOF-CD44 expressing OT-I CD8 T cells to be competitively disadvantaged in memory formation against OTI CD8 T cells expressing WT CD44. It would also be interesting to see if the inability to lose HA binding would be sufficient to negatively impact memory formation in a non-competitive situation, where the CD8 effector T cells cannot be released from the HA-rich niche through the loss of HA binding. Since HA binding positively correlates with the strength of the TCR signal during activation, and weakly activated CD8 T cells bind little HA, I would predict that the competitive advantage of CD44-/- CD8 T cells would be lost if the T cells were not strongly activated. The OT-I transgenic TCR used in my study has a high affinity to its peptide, and very low concentration (in the range of pg/ mL) of the peptide is required for activation of OT-I CD8 T cells. The effect of TCR    127   affinity on memory formation by CD44+/+ and CD44-/- CD8 T cells can be tested in vivo, by infecting mice with LM-OVA, where the OVA protein contains mutations in the SIINFEKL sequence that reduced the affinity of the peptide to the OT-I TCR (Zehn et al., 2009). Infection with these LM-OVA strains with reduced TCR affinity would result in a curtailed CD8 T cell response, with reduced number of effector cells and early peaks for the response (Zehn et al., 2009). Therefore, I predict that there will be little or no HA-binding CD8 effector T cells. Based on my model, I further predict that the competitive advantage of the CD44-/- OT-I CD8 T cells will be lost in these infections. To further investigate the role of CD44-HA interaction in memory cell formation, I could remove HA from the environment. HA is synthesized by three HAS, and deletion of all three enzymes is embryonically lethal. However, a Has2flox/flox; Has1−/−; Has3−/−  (HAS triple knockout) mouse can be crossed with inducible Cre-expressing or lymphoid organ specific Cre-expressing mice (Goncharova et al., 2012). Since the increased susceptibility to apoptosis during contraction and the reduced memory potential in HA-binding CD8 effector T cells are likely due to the cells’ interaction with HA, I predict that the competitive advantage of CD44-/- CD8 T cells against CD44+/+ CD8 T cells, or HA-binding CD8 T cells against non-binding CD8 T cells would be lost in the HAS triple knockout hosts.   6.3 The role of CD44 and HA binding in hematopoiesis 6.3.1 HA binding is exhibited by certain hematopoietic populations in homeostasis To identify where HA binding and CD44 are potentially important in hematopoietic development, I first characterized HA binding in major lymphoid and myeloid populations in the    128   BM, spleen and blood, by staining cells with FL-HA. In agreement with published observations (reviewed in (Lee-Sayer et al., 2015)), the majority of T and B cells expressed low CD44 and bound little FL-HA, while neutrophils and monocytes bound little FL-HA despite high CD44 expression. FL-HA binding by NKT cells and eosinophils has not been directly characterized, and here, I found that both cell types exhibited high CD44 expression and included a distinct HA-binding population. While HA is suggested to enhance survival of eosinophils in homeostasis and inflammation (Ohkawara et al., 2000), the role for HA binding in NKT cells is yet to be elucidated. I also found that the LSK population included a distinct FL-HA binding population. After separating HSC from MPP, based on CD150 expression (Kiel et al., 2005; Yilmaz et al., 2006), a significantly higher percent of MPP bound FL-HA than HSC, which exhibited little FL-HA binding. The percentage of FL-HA binding cells in the CLP and GMP were similar to that in the MPP. The data suggest that HSC bind little HA during homeostasis, and HA binding is induced during the transition between HSC into MPP, and is then maintained in CLP and GMP.   6.3.2 HA binding positively regulates BM reconstitution through enhancement of proliferation Since MPP enter cell cycle more frequency than HSC, and are thus more proliferative (Cheshier et al., 1999), the induction of HA binding when HSC transition into MPP might be linked to the increased proliferation. Here, I found that the induction of proliferation did indeed increase HA binding in HSPC, and exogenous HA was able to further increase cell numbers of proliferation HSPC in vitro, suggesting that HA enhances the survival or proliferation of HSPC. In tumour    129   cells, the interaction between CD44 and HA is implicated in enhancing survival (Ghatak et al., 2002). It is possible that proliferating HSPC induce HA binding as a means to counter the increased apoptosis associated with cell cycling (Alenzi, 2004). The increased survival of cycling HSPC would also explain why I found BMC with higher HA binding ability to significantly out-compete BMC with lower HA binding ability in reconstituting BM HSPC.  Based on my findings, I have proposed the following model. In homeostasis, dormant HSC do not bind HA (Fig. 6.2A), and are retained in the dormant HSC niche by various factors, including CXCL12 and SCF (Trumpp et al., 2010). In response to an increased need of MPP to replenish the mature hematopoietic populations, HSC become activated and enter cell cycle. The activated HSC also begin to gain HA binding, allowing them to interact with HA, which promotes cell survival through PI3K/ Akt (Fig.6.2B, C). The increased HA binding and proliferation persist as the activated HSC differentiate into MPP, which exhibit further enhanced HA binding and proliferation. The MPP may also cover themselves with HA, to retain the HA-CD44 signal while they move away from regions rich in HA in the BM (Fig. 6.2D).  In an irradiated mouse, transferred BMC enter the BM via the vasculature. The transferred BMC, including HSC and MPP, become activated and proliferates in response to the ablation of hematopoietic cells (Mendelson and Frenette, 2014; Trumpp et al., 2010). HSPC also increase HA binding in response to increased cell cycling (Fig. 6.2E). Activated HSPC with increased HA binding localize to HA-rich proliferative niches in the BM, and the interaction between CD44 and HA enhances their survival. After the hematopoietic cells are replenished, the proliferative signals are removed,  HSPC return to the normal rate of turnover, and activated HSC are returned to dormancy (Wilson et al., 2008).    130    Figure 6.2 Model for the role of CD44 and HA in BM HSPC (A) HSC reside closely to the endosteal surface and vasculature, and HA deposition is most concentrated on endosteal surfaces. Dormant HSC do not bind HA. Activated HSC can also return to dormancy, so the two states inter-convert. (B) As HSC become activated and transition into MPP, they become more proliferative and also increase HA binding, and this allows them to move to a proliferative niche that is richer in HA deposition. Proliferating HSC move to regions that are higher in HA density, and may even cover the cell surface with HA to retain the HA-CD44 signal. (C) Upon binding to HA, CD44 signals to PI3K/ Akt, thereby enhancing survival. (D) As the activated HSC become MPP, they retain the increased proliferation and HA binding. (E) In BMC transplant into irradiated hosts, HSC enter the BM through the vasculature, and proliferate in response to the hematopoietic stress from ablation of hematopoietic cells. The HSC become activated and increase proliferation in the proliferative niche. Upon binding to HA, CD44 signalling enhances survival of  HSC. When homeostasis is restored, activated HSC return to dormancy.  131  6.3.3 Future directions for studying the role of CD44 and HA binding in BM HSPC In this study, I found that increased ability to bind HA confers a competitive advantage on BMC in BM transfer, and that the addition of HA further enhanced the increase in cell number of HSPC cultured with cytokines in vitro. This suggests that the interaction between CD44 and HA directly enhances proliferation or survival. To directly ascertain that the interaction between CD44 and HA enhances survival, but not proliferation, BMC can be labelled with CFSE and then cultured in vitro. Other methods of quantifying proliferation including staining for Ki67 expression and BrdU incorporation into cells can also be used (Nakamura-Ishizu et al., 2014). Cell death, or reduced survival, can be quantified using cell viability dyes and Annexin-V staining. Also, changes in survival signalling can be inferred from expression levels of pro- and anti-apoptotic genes. Furthermore, a lack of difference in proliferation, coupled with a difference in cell number, would suggest a difference in survival. The effect of exogenous HA on BMC with HA binding capability and no HA binding capability also can be compared in vitro. Since CD44-/- cells are likely affected by compensatory mechanisms (Maeshima, 2010; Nedvetzki et al., 2004; Olaku et al., 2011), tLOF and tGOF can be used to repeat these in vitro experiment. I predict that the tLOF to be unaffected by the addition of exogenous HA, whereas the enhancement of increased cell number would be even more pronounced with tGOF.  To confirm the role of environmental HA on HSPC reconstitution, BM HA can be removed by using the Prx1-Cre; Has2flox/flox; Has1−/−; Has3−/− triple knockout mice (Goncharova et al., 2012). Prx1 is preferentially expressed in BM MSPC, which gives rise to mature BM stromal populations (Greenbaum et al., 2013; Morikawa et al., 2009). Each triple knockout mouse would also receive 100 μl of 3 mM of the HA synthesis inhibitor 4-MU intravenously at the time of  132  irradiation, and the three additional intravenous injections of 4-MU thereafter, at 6 hour interval, to further reduce levels of endogenous HA (Goncharova et al., 2012). I predict that the competitive difference in reconstitution between tv-WT and tGOF and between tv-WT and tLOF to be lost when BMC are transferred into these HA-depleted mice.   6.4 Concluding remarks In this study, I have identified novel roles of CD44 and HA binding in regulating hematopoiesis and CD8 T cell memory formation. HA binding by BMC conferred a competitive advantage to the BMC in reconstituting the HSPC populations, and also the lymphoid and myeloid populations subsequently. In a model of systemic bacterial infection, CD44 expressing and HA binding CD8 effector T cells were selected against in memory formation when in competition. I also found that high affinity and avidity TCR peptides led to increased HA binding by activated CD8 T cells in vivo. My findings thus provide support and a potential mechanism for the decreasing memory potential model of CD8 memory T cell formation. The consequences of CD44 expression and altered HA binding were observed in competition, suggesting that cells compete for environmental HA or HA-rich niche. Furthermore, my findings indicate that the interaction between CD44 and HA regulate aspects of metabolism, survival and death of the HSPC and CD8 effector T cells, but the molecular regulations of these signalling events downstream of CD44 still need to be defined.   133  References Adams, G.B., Chabner, K.T., Alley, I.R., Olson, D.P., Szczepiorkowski, Z.M., Poznansky, M.C., Kos, C.H., Pollak, M.R., Brown, E.M., and Scadden, D.T. (2006). Stem cell engraftment at the endosteal niche is specified by the calcium-sensing receptor. Nature 439, 599-603. Adolfsson, J., Månsson, R., Buza-Vidas, N., Hultquist, A., Liuba, K., Jensen, C.T., Bryder, D., Yang, L., Borge, O.-J., Thoren, L.A.M., et al. (2005). Identification of Flt3+ lympho-myeloid stem cells lacking erythro-megakaryocytic potential. Cell 121, 295-306. Afkarian, M., Sedy, J.R., Yang, J., Jacobson, N.G., Cereb, N., Yang, S.Y., Murphy, T.L., and Murphy, K.M. (2002). T-bet is a STAT1-induced regulator of IL-12R expression in naive CD4+ T cells. Nat Immunol 3, 549-557. Ahmed, R., Bevan, M.J., Reiner, S.L., and Fearon, D.T. (2009). The precursors of memory: models and controversies. Nat Rev Immunol 9, 662-668. Akashi, K., Traver, D., Miyamoto, T., and Weissman, I.L. (2000). A clonogenic common myeloid progenitor that gives rise to all myeloid lineages. Nature 404, 193-197. Alenzi, F.Q. (2004). Links between apoptosis, proliferation and the cell cycle. Br J Biomed Sci 61, 99-102. Allsopp, R.C., and Weissman, I.L. (2002). Replicative senescence of hematopoietic stem cells during serial transplantation: does telomere shortening play a role? Oncogene 21, 3270-3273. Almeida, J.R., Sauce, D., Price, D.A., Papagno, L., Shin, S.Y., Moris, A., Larsen, M., Pancino, G., Douek, D.C., Autran, B., et al. (2009). Antigen sensitivity is a major determinant of CD8+ T-cell polyfunctionality and HIV-suppressive activity. Blood 113, 6351-6360. Alves, N.L., van Leeuwen, E.M.M., Derks, I.A.M., and van Lier, R.A.W. (2008). Differential regulation of human IL-7 receptor α expression by IL-7 and TCR signaling. J Immunol 180, 5201-5210. Ara, T., Tokoyoda, K., Sugiyama, T., Egawa, T., Kawabata, K., and Nagasawa, T. (2003). Long-term hematopoietic stem cells require stromal cell-derived factor-1 for colonizing bone marrow during ontogeny. Immunity 19, 257-267. Arai, F., Hirao, A., Ohmura, M., Sato, H., Matsuoka, S., Takubo, K., Ito, K., Koh, G.Y., and Suda, T. (2004). Tie2/angiopoietin-1 signaling regulates hematopoietic stem cell quiescence in the bone marrow niche. Cell 118, 149-161. Arsenio, J., Kakaradov, B., Metz, P.J., Kim, S.H., Yeo, G.W., and Chang, J.T. (2014). Early specification of CD8+ T lymphocyte fates during adaptive immunity revealed by single-cell gene-expression analyses. Nat Immunol 15, 365-372.  134  Aruffo, A., Stamenkovic, I., Melnick, M., Underhill, C.B., and Seed, B. (1990). CD44 is the principal cell surface receptor for hyaluronate. Cell 61, 1303-1313. Avigdor, A., Goichberg, P., Shivtiel, S., Dar, A., Peled, A., Samira, S., Kollet, O., Hershkoviz, R., Alon, R., Hardan, I., et al. (2004). CD44 and hyaluronic acid cooperate with SDF-1 in the trafficking of human CD34+ stem/progenitor cells to bone marrow. Blood 103, 2981-2989. Baaten, B.J.G., Li, C.-R., Deiro, M.F., Lin, M.M., Linton, P.J., and Bradley, L.M. (2010). CD44 regulates survival and memory development in Th1 cells. Immunity 32, 104-115. Badovinac, V.P., and Harty, J.T. (2007). Manipulating the rate of memory CD8+ T cell generation after acute infection. J Immunol 179, 53-63. Badovinac, V.P., Porter, B.B., and Harty, J.T. (2004). CD8+ T cell contraction is controlled by early inflammation. Nat Immunol 5, 809-817. Baldridge, M.T., King, K.Y., Boles, N.C., Weksberg, D.C., and Goodell, M.A. (2010). Quiescent hematopoietic stem cells are activated by IFNγ in response to chronic infection. Nature 465, 793-797. Banerji, S., Ni, J., Wang, S.-X., Clasper, S., Su, J., Tammi, R., Jones, M., and Jackson, D.G. (1999). LYVE-1, a new homologue of the CD44 glycoprotein, is a lymph-specific receptor for hyaluronan. J Cell Biol 144, 789-801. Banerji, S., Wright, A.J., Noble, M., Mahoney, D.J., Campbell, I.D., Day, A.J., and Jackson, D.G. (2007). Structures of the Cd44-hyaluronan complex provide insight into a fundamental carbohydrate-protein interaction. Nat Struct Mol Biol 14, 234-239. Barker, J. (1994). Sl/Sld hematopoietic progenitors are deficient in situ. Exp Hematol 22, 174-177. Barry, M., and Bleackley, R.C. (2002). Cytotoxic T lymphocytes: all roads lead to death. Nat Rev Immunol 2, 401-409. Bastow, E.R., Byers, S., Golub, S.B., Clarkin, C.E., Pitsillides, A.A., and J, F.A. (2007). Hyaluronan synthesis and degradation in cartilage and bone. Cell Mol Life Sci 65. Bennett, K.L., Modrell, B., Greenfield, B., Bartolazzi, A., Stamenkovic, I., Peach, R., Jackson, D.G., Spring, F., and Aruffo, A. (1995). Regulation of CD44 binding to hyaluronan by glycosylation of variably spliced exons. J Cell Biol 131, 1623-1633. Berard, M., Brandt, K., Paus, S.B., and Tough, D.F. (2003). IL-15 promotes the survival of naive and memory phenotype CD8+ T cells. J Immunol 170, 5018-5026. Best, J.A., Blair, D.A., Knell, J., Yang, E., Mayya, V., Doedens, A., Dustin, M.L., and Goldrath, A.W. (2013). Transcriptional insights into the CD8+ T cell response to infection and memory T cell formation. Nat Immunol 14, 404-412.  135  Bevan, M.J. (2004). Helping the CD8+ T-cell response. Nat Rev Immunol 4, 595-602. Blagih, J., Coulombe, F., Vincent, Emma E., Dupuy, F., Galicia-Vázquez, G., Yurchenko, E., Raissi, Thomas C., van der Windt, Gerritje J.W., Viollet, B., Pearce, Erika L., et al. (2015). The energy sensor AMPK regulates T cell metabolic adaptation and effector responses in vivo. Immunity 42, 41-54. Bloor, B.K., Jelvagharan, M., White, K.N., and Odell, E.W. (2001). Characterization of CD44 splicing patterns in normal keratinocytes, dysplastic and squamous carcinoma cell lines. Int J Oncol 18, 1053-1059. Bollyky, P.L., Lord, J.D., Masewicz, S.A., Evanko, S.P., Buckner, J.H., Wight, T.N., and Nepom, G.T. (2007). Cutting Edge: High molecular weight hyaluronan promotes the suppressive effects of CD4+CD25+ regulatory T cells. J Immunol 179, 744-747. Bonder, C.S., Clark, S.R., Norman, M.U., Johnson, P., and Kubes, P. (2006). Use of CD44 by CD4+ Th1 and Th2 lymphocytes to roll and adhere. Blood 107, 4798-4806. Bosma, M.J., and Carroll, A.M. (1991). The SCID mouse mutant: definition, characterization, and potential uses. Ann Rev Immunol 9, 323-350. Bourguignon, L.Y.W., and Bikle, D. (2015). Selective hyaluronan–CD44 signaling promotes miRNA-21 expression and interacts with vitamin D function during cutaneous squamous cell carcinomas progression following UV irradiation. Front Immunol 6, 224. Brown, G., Mooney, C.J., Alberti-Servera, L., Muenchow, L.v., Toellner, K.-M., Ceredig, R., and Rolink, A. (2015). Versatility of stem and progenitor cells and the instructive actions of cytokines on hematopoiesis. Crit Rev Clin Lab Sci 52, 168-179. Brown, K.L., Birkenhead, D., Lai, J.C.Y., Li, L., Li, R., and Johnson, P. (2005). Regulation of hyaluronan binding by F-actin and colocalization of CD44 and phosphorylated ezrin/radixin/moesin (ERM) proteins in myeloid cells. Expt Cell Res 303, 400-414. Brown, K.L., Maiti, A., and Johnson, P. (2001). Role of sulfation in CD44-mediated hyaluronan binding induced by inflammatory mediators in human CD14+ peripheral blood monocytes. J. Immunol. 167, 5367-5374. Brown, K.L., Poon, G.F.T., Birkenhead, D., Pena, O.M., Falsafi, R., Dahlgren, C., Karlsson, A., Bylund, J., Hancock, R.E.W., and Johnson, P. (2011). Host defense peptide LL-37 selectively reduces proinflammatory macrophage responses. J Immunol 186, 5497-5505. Buchholz, V.R., Flossdorf, M., Hensel, I., Kretschmer, L., Weissbrich, B., Gräf, P., Verschoor, A., Schiemann, M., Höfer, T., and Busch, D.H. (2013). Disparate individual fates compose robust CD8+ T cell immunity. Science 340, 630-635.  136  Buck, Michael D., O’Sullivan, D., Klein Geltink, Ramon I., Curtis, Jonathan D., Chang, C.-H., Sanin, David E., Qiu, J., Kretz, O., Braas, D., van der Windt, Gerritje J.W., et al. (2016). Mitochondrial dynamics controls T cell fate through metabolic programming. Cell. Burkett, P.R., Koka, R., Chien, M., Chai, S., Chan, F., Ma, A., and Boone, D.L. (2003). IL-15Rα expression on CD8(+) T cells is dispensable for T cell memory. Proc Natl Acad Sci U S A 100, 4724-4729. Cao, H., Heazlewood, S.Y., Williams, B., Cardozo, D., Nigro, J., Oteiza, A., and Nilsson, S.K. (2016). The role of CD44 in fetal and adult hematopoietic stem cell regulation. Haematologica 101, 26-37. Cao, X., Wu, X., Frassica, D., Yu, B., Pang, L., Xian, L., Wan, M., Lei, W., Armour, M., Tryggestad, E., et al. (2011). Irradiation induces bone injury by damaging bone marrow microenvironment for stem cells. Proc Natl Acad Sci U S A 108, 1609-1614. Challen, G.A., Boles, N., Lin, K.K., and Goodell, M.A. (2009). Mouse hematopoietic stem cell identification and analysis. Cytometry A 75, 14-24. Cham, C.M., Driessens, G., O'Keefe, J.P., and Gajewski, T.F. (2008). Glucose deprivation inhibits multiple key gene expression events and effector functions in CD8(+) T cells. Eur J Immunol 38, 2438-2450. Cham, C.M., and Gajewski, T.F. (2005). Glucose availability regulates IFN-γ production and p70S6 kinase activation in CD8+ effector T cells. J Immunol 174, 4670-4677. Chang, J.T., Ciocca, M.L., Kinjyo, I., Palanivel, V.R., McClurkin, C.E., DeJong, C.S., Mooney, E.C., Kim, J.S., Steinel, N.C., Oliaro, J., et al. (2011). Asymmetric proteasome segregation as a mechanism for unequal partitioning of the transcription factor T-bet during T lymphocyte division. Immunity 34, 492-504. Chang, J.T., Palanivel, V.R., Kinjyo, I., Schambach, F., Intlekofer, A.M., Banerjee, A., Longworth, S.A., Vinup, K.E., Mrass, P., Oliaro, J., et al. (2007). Asymmetric T lymphocyte division in the initiation of adaptive immune responses. Science 315, 1687-1691. Chen, D., McKallip, R.J., Zeytun, A., Do, Y., Lombard, C., Robertson, J.L., Mak, T.W., Nagarkatti, P.S., and Nagarkatti, M. (2001). CD44-deficient mice exhibit enhanced hepatitis after concanavalin A injection: evidence for involvement of CD44 in activation-induced cell death. J Immunol 166, 5889-5897. Chen, H., Mohan, P., Jiang, J., Nemirovsky, O., He, D., Fleisch, M.C., Niederacher, D., Pilarski, L.M., Lim, C.J., and Maxwell, C.A. (2014). Spatial regulation of Aurora A activity during mitotic spindle assembly requires RHAMM to correctly localize TPX2. Cell Cycle 13, 2248-2261.  137  Cheshier, S.H., Morrison, S.J., Liao, X., and Weissman, I.L. (1999). In vivo proliferation and cell cycle kinetics of long-term self-renewing hematopoietic stem cells. Proc Natl Acad Sci U S A 96, 3120-3125. Chiu, R.K., Droll, A., Dougherty, S.T., Carpenito, C., Cooper, D.L., and Dougherty, G.J. (1999). Alternatively spliced CD44 isoforms containing exon v10 promote cellular adhesion through the recognition of chondroitin sulfate-modified CD44. Exp Cell Res 248, 314-321. Chowdhury, D., and Lieberman, J. (2008). Death by a thousand cuts: granzyme pathways of programmed cell death. Ann Rev Immunol 26, 389-420. Christensen, J.L., and Weissman, I.L. (2001). Flk-2 is a marker in hematopoietic stem cell differentiation: A simple method to isolate long-term stem cells. Proc Natl Acad Sci U S A 98, 14541-14546. Clark, R.A., Alon, R., and Springer, T.A. (1996). CD44 and hyaluronan-dependent rolling interactions of lymphocytes on tonsillar stroma. J Cell Biol 134, 1075-1087. Csoka, A.B., Frost, G.I., and Stern, R. (2001). The six hyaluronidase-like genes in the human and mouse genomes. Matrix Biol 20, 499-508. Cui, W., Joshi, N.S., Jiang, A., and Kaech, S.M. (2009). Effects of signal 3 during CD8 T cell priming: bystander production of IL-12 enhances effector T cell expansion but promotes terminal differentiation. Vaccine 27, 2177-2187. Culty, M., Nguyen, H.A., and Underhill, C.B. (1992). The hyaluronan receptor (CD44) participates in the uptake and degradation of hyaluronan. J Cell Biol 116, 1055-1062. Culty, M., O'Mara, T.E., Underhill, C.B., Yeager, H., Jr., and Swartz, R.P. (1994). Hyaluronan receptor (CD44) expression and function in human peripheral blood monocytes and alveolar macrophages. J. Leukoc. Biol. 56, 605-611. Curtsinger, J.M., Lins, D.C., Johnson, C.M., and Mescher, M.F. (2005). Signal 3 tolerant CD8 T cells degranulate in response to antigen but lack granzyme B to mediate cytolysis. J Immunol 175, 4392-4399. Curtsinger, J.M., Lins, D.C., and Mescher, M.F. (2003). Signal 3 determines tolerance versus full activation of naive CD8 T cells: dissociating proliferation and development of effector function. J Exp Med 197, 1141-1151. Day, A.J. (2001). Understanding hyaluronan-binding interactions. Glycoforum. Day, A.J., and Prestwich, G.D. (2002). Hyaluronan-binding proteins: tying up the giant. J Biol Chem 277, 4585-4588. D'Souza, W.N., and Hedrick, S.M. (2006). Cutting edge: latecomer CD8 T cells are imprinted with a unique differentiation program. J Immunol 177, 777-781.  138  Daniels, M.A., Teixeiro, E., Gill, J., Hausmann, B., Roubaty, D., Holmberg, K., Werlen, G., Hollander, G.A., Gascoigne, N.R.J., and Palmer, E. (2006). Thymic selection threshold defined by compartmentalization of Ras/MAPK signalling. Nature 444, 724-729. Dasgupta, A., Takahashi, K., Cutler, M., and Tanabe, K.K. (1996). O-linked glycosylation modifies CD44 adhesion to hyaluronate in colon carcinoma cells. Biochem Biophys Res Commun 227, 110-117. de Bruin, A.M., Demirel, Ö., Hooibrink, B., Brandts, C.H., and Nolte, M.A. (2013). Interferon-γ impairs proliferation of hematopoietic stem cells in mice. Blood 121, 3578-3585. DeGrendele, H.C., Estess, P., Picker, L.J., and Siegelman, M.H. (1996). CD44 and its ligand hyaluronate mediate rolling under physiologic flow: a novel lymphocyte-endothelial cell primary adhesion pathway. J Exp Med 183, 1119-1130. DeGrendele, H.C., Estess, P., and Siegelman, M.H. (1997). Requirement for CD44 in activated T cell extravasation into an inflammatory site. Science 278, 672-675. del Pozo, M.A., Sánchez-Mateos, P., Nieto, M., and Sánchez-Madrid, F. (1995). Chemokines regulate cellular polarization and adhesion receptor redistribution during lymphocyte interaction with endothelium and extracellular matrix. Involvement of cAMP signaling pathway. J Cell Biol 131, 495-508. Delcommenne, M., Kannagi, R., and Johnson, P. (2002). TNF-α increases the carbohydrate sulfation of CD44: induction of 6-sulfo N-acetyl lactosamine on N- and O-linked glycans. Glycobiol 12, 613-622. Denton, A.E., Wesselingh, R., Gras, S., Guillonneau, C., Olson, M.R., Mintern, J.D., Zeng, W., Jackson, D.C., Rossjohn, J., Hodgkin, P.D., et al. (2011). Affinity thresholds for naive CD8+ CTL activation by peptides and engineered influenza A viruses. J Immunol 187, 5733-5744. Dubois, S., Mariner, J., Waldmann, T.A., and Tagaya, Y. (2002). IL-15Rα recycles and presents IL-15 In trans to neighboring cells. Immunity 17, 537-547. Dure, M., and Macian, F. (2009). IL-2 signaling prevents T cell anergy by inhibiting the expression of anergy inducing genes. Mol Immunol 46, 999-1006. Eaves, C.J. (2015). Hematopoietic stem cells: concepts, definitions, and the new reality. Blood 125, 2605-2613. Edinger, A.L., and Thompson, C.B. (2002). Akt maintains cell size and survival by increasing mTOR-dependent nutrient uptake. Mol Biol Cell 13, 2276-2288. English, N.M., Lesley, J.F., and Hyman, R. (1998). Site-specific de-N-glycosylation of CD44 can activate hyaluronan binding, and CD44 activation states show distinct threshold densities for hyaluronan binding. Cancer Research 58, 3736-3742.  139  Evanko, S.P., Potter-Perigo, S., Bollyky, P.L., Nepom, G.T., and Wight, T.N. (2012). Hyaluronan and versican in the control of human T-lymphocyte adhesion and migration. Matrix Biology 31, 90-100. Faller, C.E., and Guvench, O. (2014). Terminal sialic acids on CD44 N-glycans can block hyaluronan binding by forming competing intramolecular contacts with arginine sidechains. Proteins 82, 3079-3089. Fanning, A., Volkov, Y., Freeley, M., Kelleher, D., and Long, A. (2005). CD44 cross-linking induces protein kinase C-regulated migration of human T lymphocytes. Int Immunol 17, 449-458. Finlay, D.K., Rosenzweig, E., Sinclair, L.V., Feijoo-Carnero, C., Hukelmann, J.L., Rolf, J., Panteleyev, A.A., Okkenhaug, K., and Cantrell, D.A. (2012). PDK1 regulation of mTOR and hypoxia-inducible factor 1 integrate metabolism and migration of CD8+ T cells. J Exp Med 209, 2441-2453. Firan, M., Dhillon, S., Estess, P., and Siegelman, M.H. (2006). Suppressor activity and potency among regulatory T cells is discriminated by functionally active CD44. Blood 107, 619-627. Flossdorf, M., Rossler, J., Buchholz, V.R., Busch, D.H., and Hofer, T. (2015). CD8+ T cell diversification by asymmetric cell division. Nat Immunol 16, 891-893. Ford, C.E., Hamerton, J.L., Barnes, D.W., and Loutit, J.F. (1956). Cytological identification of radiation-chimaeras. Nature 177, 452-454. Fousteri, G., Dave, A., Juedes, A., Juntti, T., Morin, B., Togher, L., Farber, D.L., and Herrath, M.v. (2011). Increased memory conversion of naïve CD8 T cells activated during late phases of acute virus infection due to decreased cumulative antigen exposure. PLos ONE 6, e14502. Fox, N., Priestley, G., Papayannopoulou, T., and Kaushansky, K. (2002). Thrombopoietin expands hematopoietic stem cells after transplantation. J Clin Invest 110, 389-394. Fraser, J.R.E., Laurent, T.C., and Laurent, U.B.G. (1997). Hyaluronan: its nature, distribution, functions and turnover. J Intern Med 242, 27-33. Frauwirth, K.A., Riley, J.L., Harris, M.H., Parry, R.V., Rathmell, J.C., Plas, D.R., Elstrom, R.L., June, C.H., and Thompson, C.B. (2002). The CD28 signaling pathway regulates glucose metabolism. Immunity 16, 769-777. Frecha, C., Costa, C., Nègre, D., Gauthier, E., Russell, S.J., Cosset, F.-L., and Verhoeyen, E. (2008). Stable transduction of quiescent T cells without induction of cycle progression by a novel lentiviral vector pseudotyped with measles virus glycoproteins. Blood 112, 4843-4852. Galandrini, R., Galluzzo, E., Albi, N., Grossi, C.E., and Velardi, A. (1994). Hyaluronate is costimulatory for human T cell effector functions and binds to CD44 on activated T cells. J Immunol 153, 21-31.  140  Genever, P.G., and Dickson, I.R. (1996). Influence of vitamin D status on hyaluronan localization in bone. Bone 18, 429-435. Gerlach, C., Rohr, J.C., Perié, L., van Rooij, N., van Heijst, J.W.J., Velds, A., Urbanus, J., Naik, S.H., Jacobs, H., Beltman, J.B., et al. (2013). Heterogeneous differentiation patterns of individual CD8+ T cells. Science 340, 635-639. Ghatak, S., Misra, S., and Toole, B.P. (2002). Hyaluronan oligosaccharides inhibit anchorage-independent growth of tumor cells by suppressing the phosphoinositide 3-kinase/Akt cell survival pathway. J Biol Chem 277, 38013-38020. Ghazi-Visser, L., Laman, J.D., Nagel, S., van Meurs, M., van Riel, D., Tzankov, A., Frank, S., Adams, H., Wolk, K., Terracciano, L., et al. (2013). CD44 variant isoforms control experimental autoimmune encephalomyelitis by affecting the lifespan of the pathogenic T cells. FASEB J 27, 3683-3701. Goncharova, V., Serobyan, N., Iizuka, S., Schraufstatter, I., de Ridder, A., Povaliy, T., Wacker, V., Itano, N., Kimata, K., Orlovskaja, I.A., et al. (2012). Hyaluronan expressed by the hematopoietic microenvironment is required for bone narrow hematopoiesis. J Biol Chem 287, 25419-25433. Gonda, A., Gál, I., Szántó, S., Sarraj, B., Glant, T.T., Hunyadi, J., and Mikecz, K. (2005). CD44, but not l-selectin, is critically involved in leucocyte migration into the skin in a murine model of allergic dermatitis. Exp Dermatol 14, 700-708. Goodell, M.A., Brose, K., Paradis, G., Connor, A., and Mulligan , R. (1996). Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J Exp Med 183, 1797-1806. Goodell, M.A., Rosenzweig, M., Kim, H., Marks, D.F., DeMaria, M., Paradis, G., Grupp, S.A., Sieff, C.A., Mulligan, R.C., and Johnson, R.P. (1997). Dye efflux studies suggest that hematopoietic stem cells expressing low or undetectable levels of CD34 antigen exist in multiple species. Nat Med 3, 1337-1345. Greenbaum, A., Hsu, Y.-M.S., Day, R.B., Schuettpelz, L.G., Christopher, M.J., Borgerding, J.N., Nagasawa, T., and Link, D.C. (2013). CXCL12 production by early mesenchymal progenitors is required for hematopoietic stem cell maintenance. Nature 495, 227-230. Greiner, E.F., Guppy, M., and Brand, K. (1994). Glucose is essential for proliferation and the glycolytic enzyme induction that provokes a transition to glycolytic energy production. J Biol Chem 269, 31484-31490. Guan, H., Nagarkatti, P.S., and Nagarkatti, M. (2011). CD44 reciprocally regulates the differentiation of encephalitogenic Th1/Th17 and Th2/Treg cells through epigenetic modulation involving DNA methylation of cytokine gene promoters thereby controlling the development of experimental autoimmune encephalomyelitis. J Immunol 186, 6955-6964.  141  Hand, T.W., Morre, M., and Kaech, S.M. (2007). Expression of IL-7 receptor α is necessary but not sufficient for the formation of memory CD8 T cells during viral infection. Proc Natl Acad Sci USA 104, 11730-11735. Haring, J.S., Jing, X., Bollenbacher-Reilley, J., Xue, H.-H., Leonard, W.J., and Harty, J.T. (2008). Constitutive expression of IL-7 receptor α does not support increased expansion or prevent contraction of antigen-specific CD4 or CD8 T cells following Listeria monocytogenes infection. J Immunol 180, 2855-2862. Hathcock, K.S., Hirano, H., Murakami, S., and Hodes, R.J. (1993). CD44 expression on activated B cells. Differential capacity for CD44-dependent binding to hyaluronic acid. J Immunol 151, 6712-6722. Hayes, G.M., Chiu, R., Carpenito, C., Dougherty, S.T., and Dougherty, G.J. (2002). Identification of sequence motifs responsible for the adhesive interaction between exon v10-containing CD44 isoforms. J Biol Chem 277, 50529-50534. Heider, K.-H., Kuthan, H., Stehle, G., and Munzert, G. (2004). CD44v6: a target for antibody-based cancer therapy. Cancer Immunol. Immunother. 53, 567-579. Hirose, Y., Saijou, E., Sugano, Y., Takeshita, F., Nishimura, S., Nonaka, H., Chen, Y.-R., Sekine, K., Kido, T., Nakamura, T., et al. (2012). Inhibition of Stabilin-2 elevates circulating hyaluronic acid levels and prevents tumor metastasis. Proc Natl Acad Sci U S A 109, 4263-4268. Hogquist, K.A., Jameson, S.C., Heath, W.R., Howard, J.L., Bevan, M.J., and Carbone, F.R. (1994). T cell receptor antagonist peptides induce positive selection. Cell 76, 17-27. Holmes, S., He, M., Xu, T., and Lee, P.P. (2005). Memory T cells have gene expression patterns intermediate between naïve and effector. Proc Natl Acad Sci U S A 102, 5519-5523. Jackson, D.G. (2003). The lymphatics revisited: new perspectives from the hyaluronan receptor LYVE-1. Trends Cardiovas Med 13, 1-7. Jackson, D.C., Bell, J.I., Dickinson, R., Timans, J., Shields, J., and Whittle, N. (1995). Proteoglycan forms of the lymphocyte homing receptor CD44 are alternatively spliced variants containing the v3 exon. J Cell Biol 128, 673-685. Jacobs, S.R., Herman, C.E., MacIver, N.J., Wofford, J.A., Wieman, H.L., Hammen, J.J., and Rathmell, J.C. (2008). Glucose uptake is limiting in T cell activation and requires CD28-mediated Akt-dependent and independent pathways. J Immunol 180, 4476-4486. Jacobson, L.O., Simmons, E.L., Marks, E.K., and Eldredge, J.H. (1951). Recovery from radiation injury. Science 113, 510-511. Jameson, S., and Hamilton, S. (2012). CD8 T cell memory: it takes all kinds. Front Immunol 3.  142  Jiang, Y., Bonig, H., Ulyanova, T., Chang, K., and Papayannopoulou, T. (2009). On the adaptation of endosteal stem cell niche function in response to stress. Blood 114, 3773-3782. Jiang, J., Mohan, P., and Maxwell, C.A. (2013). The cytoskeletal protein RHAMM and ERK1/2 activity maintain the pluripotency of murine embryonic stem cells. PLoS ONE 8, e73548. Jones, R.J., Wagner, J.E., Celano, P., Zicha, M.S., and Sharkis, S.J. (1990). Separation of pluripotent haematopoietic stem cells from spleen colony-forming cells. Nature 347, 188-189. Joshi, N.S., Cui, W., Chandele, A., Lee, H.K., Urso, D.R., Hagman, J., Gapin, L., and Kaech, S.M. (2007). Inflammation directs memory precursor and short-lived effector CD8+ T cell fates via the graded expression of T-bet transcription factor. Immunity 27, 281-295. Juntilla, M.M., Patil, V.D., Calamito, M., Joshi, R.P., Birnbaum, M.J., and Koretzky, G.A. (2010). AKT1 and AKT2 maintain hematopoietic stem cell function by regulating reactive oxygen species. Blood 115, 4030-4038. Kaech, S.M., and Cui, W. (2012). Transcriptional control of effector and memory CD8+ T cell differentiation. Nat Rev Immunol 12, 749-761. Kaech, S.M., Hemby, S., Kersh, E., and Ahmed, R. (2002). Molecular and functional profiling of memory CD8 T cell differentiation. Cell 111, 837-851. Kaech, S.M., Tan, J.T., Wherry, E.J., Konieczny, B.T., Surh, C.D., and Ahmed, R. (2003). Selective expression of the interleukin 7 receptor identifies effector CD8 T cells that give rise to long-lived memory cells. Katoh, S., Matsubara, Y., Taniguchi, H., Fukushima, K., Mukae, H.,  Kadota, J., Matsukura, S., and Kohno, S. (2001). Characterization of CD44 expressed on alveolar macrophages in patients with diffuse panbronchiolitis. Clin Exp Immunol 126, 545-550. Katoh, S., Kaminuma, O., Hiroi, T., Mori, A., Ohtomo, T., Maeda, S., Shimizu, H., Obase, Y., and Oka, M. (2011). CD44 is critical for airway accumulation of antigen-specific Th2, but not Th1, cells induced by antigen challenge in mice. Eur J Immunol 41, 3198-3207. Katoh, S., Maeda, S., Fukuoka, H., Wada, T., Moriya, S., Mori, A., Yamaguchi, K., Senda, S., and Miyagi, T. (2010). A crucial role of sialidase Neu1 in hyaluronan receptor function of CD44 in T helper type 2-mediated airway inflammation of murine acute asthmatic model. Clin Exp Immunol 161, 233-241. Katoh, S., Matsumoto, N., Kawakita, K., Tominaga, A., Kincade, P.W., and Matsukura, S. (2003). A role for CD44 in an antigen-induced murine model of pulmonary eosinophilia. J Clin Invest 111, 1563-1570. Katoh, S., Miyagi, T., Taniguchi, H., Matsubara, Y.-i., Kadota, J.-i., Tominaga, A., Kincade, P.W., Matsukura, S., and Kohno, S. (1999). Cutting edge: an inducible sialidase regulates the hyaluronic acid binding ability of CD44-bearing human monocytes. J Immunol 162, 5058-5061.  143  Katoh, S., Zheng, Z., Oritani, K., Shimozato, T., and Kincade, P.W. (1995). Glycosylation of CD44 negatively regulates its recognition of hyaluronan. J. Exp. Med. 182, 419-429.Nat Immunol 4, 1191-1198. Kharas, M.G., Okabe, R., Ganis, J.J., Gozo, M., Khandan, T., Paktinat, M., Gilliland, D.G., and Gritsman, K. (2010). Constitutively active AKT depletes hematopoietic stem cells and induces leukemia in mice. Blood 115, 1406-1415. Kiel, M., Yilmaz, O., Iwashita, T., Yilmaz, O., Terhorst, C., and Morrison, S. (2005). SLAM family receptors distinguish hematopoietic stem and progenitor cells and reveal endothelial niches for stem cells. Cell 121, 1109-1121. Kim, E.H., Sullivan, J.A., Plisch, E.H., Tejera, M.M., Jatzek, A., Choi, K.Y., and Suresh, M. (2012). Signal integration by Akt regulates CD8 T cell effector and memory differentiation. J Immunol 188, 4305-4314. Kimura, S., Roberts, A.W., Metcalf, D., and Alexander, W.S. (1998). Hematopoietic stem cell deficiencies in mice lacking c-Mpl, the receptor for thrombopoietin. Proc Natl Acad Sci U S A 95, 1195-1200. King, K.Y., and Goodell, M.A. (2011). Inflammatory modulation of HSCs: viewing the HSC as a foundation for the immune response. Nat Rev Immunol 11, 685-692. Kinjyo, I., Qin, J., Tan, S.-Y., Wellard, C.J., Mrass, P., Ritchie, W., Doi, A., Cavanagh, L.L., Tomura, M., Sakaue-Sawano, A., et al. (2015). Real-time tracking of cell cycle progression during CD8(+) effector and memory T-cell differentiation. Nat Commun 6, 6301. Klamer, S., and Voermans, C. (2014). The role of novel and known extracellular matrix and adhesion molecules in the homeostatic and regenerative bone marrow microenvironment. Cell Adh Migr 8, 563-577. Klingbeil, P., Marhaba, R., Jung, T., Kirmse, R., Ludwig, T., and Zöller, M. (2009). CD44 variant isoforms promote metastasis formation by a tumor cell-matrix cross-talk that supports adhesion and apoptosis resistance. Mol Cancer Res 7, 168-179. Kolumam, G.A., Thomas, S., Thompson, L.J., Sprent, J., and Murali-Krishna, K. (2005). Type I interferons act directly on CD8 T cells to allow clonal expansion and memory formation in response to viral infection. J Exp Med 202, 637-650. Kondo, M., Weissman, I.L., and Akashi, K. (1997). Identification of clonogenic common lymphoid progenitors in mouse mone marrow. Cell 91, 661-672. Koshiishi, I., Shizari, M., and Underhill, C.B. (1994). CD44 can mediate the adhesion of platelets to hyaluronan. Blood 84, 390-396. Krammer, P.H., Arnold, R., and Lavrik, I.N. (2007). Life and death in peripheral T cells. Nat Rev Immunol 7, 532-542.  144  Kryworuchko, M., Diaz-Mitoma, F., and Kumar, A. (1999). Interferon-gamma inhibits CD44-hyaluronan interactions in normal human B lymphocytes. Exp Cell Res 250, 241-252. Kumari, S., Curado, S., Mayya, V., and Dustin, M.L. (2014). T cell antigen receptor activation and actin cytoskeleton remodeling. Biochem Biophys Acta 1838, 10.1016/j.bbamem.2013.1005.1004. Kunisaki, Y., Bruns, I., Scheiermann, C., Ahmed, J., Pinho, S., Zhang, D., Mizoguchi, T., Wei, Q., Lucas, D., Ito, K., et al. (2013). Arteriolar niches maintain haematopoietic stem cell quiescence. Nature 502, 637-643. Lazarevic, V., Glimcher, L.H., and Lord, G.M. (2013). T-bet: a bridge between innate and adaptive immunity. Nat Rev Immunol 13, 777-789. Le Bon, A., Durand, V., Kamphuis, E., Thompson, C., Bulfone-Paus, S., Rossmann, C., Kalinke, U., and Tough, D.F. (2006). Direct stimulation of T cells by type I IFN enhances the CD8+ T cell response during cross-priming. J Immunol 176, 4682-4689. Lee-Sayer, S.S.M., Dong, Y., Arif, A.A., Olsson, M., Brown, K.L., and Johnson, P. (2015). The where, when, how, and why of hyaluronan binding by immune cells. Front Immunol 6, 150. Lefebvre, D.C., Lai, J.C.Y., Maeshima, N., Ford, J.L., Wong, A.S.L., Cross, J.L., and Johnson, P. (2010). CD44 interacts directly with Lck in a zinc-dependent manner. Mol Immunol 47, 1882-1889. Lesley, J., English, N., Charles, C., and Hyman, R. (2000). The role of the CD44 cytoplasmic and transmembrane domains in constitutive and inducible hyaluronan binding. Eur J Immunol 30, 245-253. Lesley, J., English, N., Perschl, A., Gregoroff, J., and Hyman, R. (1995). Variant cell lines selected for alterations in the function of the hyaluronan receptor CD44 show differences in glycosylation. J Exp Med 182, 431-437. Lesley, J., Howes, N., Perschl, A., and Hyman, R. (1994). Hyaluronan binding function of CD44 is transiently activated on T cells during an in vivo immune response. J Exp Med 180, 383-387. Levesque, M.C., and Haynes, B.F. (1996). In vitro culture of human peripheral blood monocytes induces hyaluronan binding and up-regulates monocyte variant CD44 isoform expression. J. Immunol. 156, 1557-1565. Levesque, M.C., and Haynes, B.F. (1997). Cytokine induction of the ability of human monocyte CD44 to bind hyaluronan is mediated primarily by TNF-alpha and is inhibited by IL-4 and IL-13. J Immunol 159, 6184-6194. Levesque, M.C., and Haynes, B.F. (1999). TNFα and IL-4 regulation of hyaluronan binding to monocyte CD44 involves posttranslational modification of CD44. Cell Immunol 193, 209-218.  145  Li, R., Wong, N., Jabali, M.D., and Johnson, P. (2001). CD44-initiated cell spreading induces Pyk2 phosphorylation, is mediated by Src family kinases, and is negatively regulated by CD45. J Biol Chem 276, 28767-28773. Liao, H.X., Levesque, M.C., Patton, K., Bergamo, B., Jones, D., Moody, M.A., Telen, M.J., and Haynes, B.F. (1993). Regulation of human CD44H and CD44E isoform binding to hyaluronan by phorbol myristate acetate and anti-CD44 monoclonal and polyclonal antibodies. J Immunol 151, 6490-6499. Lighvani, A.A., Frucht, D.M., Jankovic, D., Yamane, H., Aliberti, J., Hissong, B.D., Nguyen, B.V., Gadina, M., Sher, A., Paul, W.E., and O'Shea, J.J. (2001). T-bet is rapidly induced by interferon-γ in lymphoid and myeloid cells. Proc Natl Acad Sci U S A 98, 15137-15142. Link, A., Vogt, T.K., Favre, S., Britschgi, M.R., Acha-Orbea, H., Hinz, B., Cyster, J.G., and Luther, S.A. (2007). Fibroblastic reticular cells in lymph nodes regulate the homeostasis of naive T cells. Nat Immunol 8, 1255-1265. Liu, D., Zhang, D., Mori, H., and Sy, M.-S. (1996). Binding of CD44 to hyaluronic acid can be induced by multiple signals and requires the CD44 cytoplasmic domain. Cell Immunol 174, 73-83. Lord, B., Testa, N., and Hendry, J. (1975). The relative spatial distributions of CFUs and CFUc in the normal mouse femur. Blood 46, 65-72. Lorenz, E., Congdon, C., and Uphoff, D. (1952). Modification of acute irradiation injury in mice and guinea-pigs by bone marrow injections. Radiol 58, 863-877. Lunt, S.Y., and Vander Heiden, M.G. (2011). Aerobic glycolysis: meeting the metabolic requirements of cell proliferation. Annu Rev Cell Dev Biol 27, 441-464. MacNamara, K., Jones, M., Martin, O., and Winslow, G. (2011). Transient activation of hematopoietic stem and progenitor cells by IFNγ during acute bacterial infection. PLoS ONE 6, e28669. Maekawa, Y., Ishifune, C., Tsukumo, S.-i., Hozumi, K., Yagita, H., and Yasutomo, K. (2015). Notch controls the survival of memory CD4+ T cells by regulating glucose uptake. Nat Med 21, 55-61. Maeshima, N. (2010). Hyaluronan binding by CD44 on T cells. In Microbiology and Immunology (Vancouver, University of British Columbia). Maeshima, N., Poon, G.F.T., Dosanjh, M., Felberg, J., Lee, S.S.M., Cross, J.L., Birkenhead, D., and Johnson, P. (2011). Hyaluronan binding identifies the most proliferative activated and memory T cells. Eur J Immunol 41, 1108-1119. Mahaffey, C.L., and Mummert, M.E. (2007). Hyaluronan synthesis is required for IL-2-mediated T cell proliferation. J Immunol 179, 8191-8199.  146  Malek, T.R., and Castro, I. (2010). Interleukin-2 receptor signaling: at the interface between tolerance and immunity. Immunity 33, 153-165. Maraskovsky, E., Teepe, M., Morrissey, P.J., Braddy, S., Miller, R.E., Lynch, D.H., and Peschon, J.J. (1996). Impaired survival and proliferation in IL-7 receptor-deficient peripheral T cells. J Immunol 157, 5315-5323. Marhaba, R., Freyschmidt-Paul, P., and Zöller, M. (2006). In vivo CD44-CD49d complex formation in autoimmune disease has consequences on T cell activation and apoptosis resistance. Eur J Immunol 36, 3017-3032. Markowitz, D., Goff, S., and Bank, A. (1988). A safe packaging line for gene transfer: separating viral genes on two different plasmids. J Virol 62, 1120-1124. Markowitz, D., Hesdorffer, C., Ward, M., Goff, S., and Bank, A. (1990). Retroviral gene transfer using safe and efficient packaging cell lines. Ann N Y Acad Sci 612, 407-414. Matrosova, V.Y., Orlovskaya, I.A., Serobyan, N., and Khaldoyanidi, S.K. (2004). Hyaluronic acid facilitates the recovery of hematopoiesis following 5-fluorouracil administration. Stem Cell 22, 544-555. Maxwell, C.A., Benítez, J., Gómez-Baldó, L., Osorio, A., Bonifaci, N., Fernández-Ramires, R., Costes, S.V., Guinó, E., Chen, H., Evans, G.J.R., et al. (2011). Interplay between BRCA1 and RHAMM regulates epithelial apicobasal polarization and may influence risk of breast cancer. PLoS Biol 9, e1001199. Maxwell, C.A., Keats, J.J., Belch, A.R., Pilarski, L.M., and Reiman, T. (2005). Receptor for hyaluronan-mediated motility correlates with centrosome abnormalities in multiple myeloma and maintains mitotic integrity. Cancer Res 65, 850-860. Maxwell, C.A., Keats, J.J., Crainie, M., Sun, X., Yen, T., Shibuya, E., Hendzel, M., Chan, G., and Pilarski, L.M. (2003). RHAMM Is a centrosomal protein that interacts with dynein and maintains spindle pole stability. Mol Biol Cell 14, 2262-2276. Mazzucchelli, R.I., Warming, S., Lawrence, S.M., Ishii, M., Abshari, M., Washington, A.V., Feigenbaum, L., Warner, A.C., Sims, D.J., Li, W.Q., et al. (2009). Visualization and identification of IL-7 producing cells in reporter mice. PLoS ONE 4, e7637. McCulloch, E.A., and Till, J.E. (1960). The radiation sensitivity of normal mouse bone marrow cells, determined by quantitative marrow transplantation into irradiated mice. Radiat Res 13, 115-125.  McDonald, B., McAvoy, E.F., Lam, F., Gill, V., de la Motte, C., Savani, R.C., and Kubes, P. (2008). Interaction of CD44 and hyaluronan is the dominant mechanism for neutrophil sequestration in inflamed liver sinusoids. J Exp Med 205, 915-927.  147  McKallip, R.J., Do, Y., Fisher, M.T., Robertson, J.L., Nagarkatti, P.S., and Nagarkatti, M. (2002). Role of CD44 in activation‐induced cell death: CD44‐deficient mice exhibit enhanced T cell response to conventional and superantigens.  Int Immunol  14, 1015-1026. McKallip, R.J., Hagele, H.F., and Uchakina, O.N. (2013). Treatment with the hyaluronic acid synthesis inhibitor 4-methylumbelliferone suppresses SEB-Induced lung inflammation. Toxins 5, 1814-1826. Mempel, T.R., Henrickson, S.E., and von Andrian, U.H. (2004). T-cell priming by dendriticcells in lymph nodes occurs in three distinct phases. Nature 427, 154-159. Mendelson, A., and Frenette, P.S. (2014). Hematopoietic stem cell niche maintenance during homeostasis and regeneration. Nat Med 20, 833-846. Mielgo, A., van Driel, M., Bloem, A., Landmann, L., and Günthert, U. (2005). A novel antiapoptotic mechanism based on interference of Fas signaling by CD44 variant isoforms. Cell Death Differ 13, 465-477. Mohamadzadeh, M., DeGrendele, H., Arizpe, H., Estess, P., and Siegelman, M. (1998). Proinflammatory stimuli regulate endothelial hyaluronan expression and CD44/HA-dependent primary adhesion. J Clin Invest 101, 97-108. Mombaerts, P., Iacomini, J., Johnson, R.S., Herrup, K., Tonegawa, S., and Papaioannou, V.E. (1992). RAG-1-deficient mice have no mature B and T lymphocytes. Cell 68, 869-877. Montecino-Rodriguez, E., Leathers, H., and Dorshkind, K. (2001). Bipotential B-macrophage progenitors are present in adult bone marrow. Nat Immunol 2. Morikawa, S., Mabuchi, Y., Kubota, Y., Nagai, Y., Niibe, K., Hiratsu, E., Suzuki, S., Miyauchi-Hara, C., Nagoshi, N., Sunabori, T., et al. (2009). Prospective identification, isolation, and systemic transplantation of multipotent mesenchymal stem cells in murine bone marrow. J Exp Med 206, 2483-2496. Morrison, S.J., Wandycz, A.M., Hemmati, H.D., Wright, D.E., and Weissman, I.L. (1997). Identification of a lineage of multipotent hematopoietic progenitors. Development 124, 1929-1939. Morrison, S.J., and Weissman, I.L. (1994). The long-term repopulating subset of hematopoietic stem cells is deterministic and isolatable by phenotype. Immunity 1, 661-673. Mosser, D.M., and Edwards, J.P. (2008). Exploring the full spectrum of macrophage activation. Nat Rev Immunol 8, 958-969. Mueller, A.M., Yoon, B.H., and Sadiq, S.A. (2014). Inhibition of hyaluronan synthesis protects against central nervous system (CNS) autoimmunity and increases CXCL12 expression in the inflamed CNS. J Biol Chem 289, 22888-22899.  148  Mummert, M.E., Mummert, D., Edelbaum, D., Hui, F., Matsue, H., and Takashima, A. (2002). Synthesis and surface expression of hyaluronan by dendritic cells and its potential role in antigen presentation. J Immunol 169, 4322-4331. Murakami, S., Miyake, K., June, C.H., Kincade, P.W., and Hodes, R.J. (1990). IL-5 induces a Pgp-1 (CD44) bright B cell subpopulation that is highly enriched in proliferative and Ig secretory activity and binds to hyaluronate. J. Immunol. 145, 3618-3627. Murakami, S., Saho, T., Asari, A., Hino, E., Kasai, D., Shimabukuro, Y., and Okada, H. (1996). CD44-hyaluronate interaction participates in the adherence of T-lymphocytes to gingival fibroblasts. J Dent Res 75, 1545-1552. Nagasawa, T. (2006). Microenvironmental niches in the bone marrow required for B-cell development. Nat Rev Immunol 6, 107-116. Nagy, N., Freudenberger, T., Melchior-Becker, A., Röck, K., ter Braak, M., Jastrow, H., Kinzig, M., Lucke, S., Suvorava, T., Kojda, G., et al. (2010). A Novel Mechanism for the Inhibition of Hyaluronan Biosynthesis by 4-Methylumbelliferone. J Biol Chem 122, 2313-2322. Nagy, N., Kuipers, H.F., Frymoyer, A.R., Ishak, H.D., Bollyky, J.B., Wight, T.N., and Bollyky, P.L. (2015). 4-methylumbelliferone treatment and hyaluronan Inhibition as a therapeutic strategy in inflammation, autoimmunity, and cancer. Front Immunol 6, 123. Nakamura-Ishizu, A., Takizawa, H., and Suda, T. (2014). The analysis, roles and regulation of quiescence in hematopoietic stem cells. Development 141, 4656-4666. Nakano, K., Saito, K., Mine, S., Matsushita, S., and Tanaka, Y. (2007). Engagement of CD44 up-regulates Fas Ligand expression on T cells leading to activation-induced cell death. Apoptosis 12, 45-54. Nandi, A., Estess, P., and Siegelman, M. (2004). Bimolecular complex between rolling and firm adhesion receptors required for cell arrest. Immunity 20, 455-465. Nandi, A., Estess, P., and Siegelman, M.H. (2000). Hyaluronan anchoring and regulation on the surface of vascular endothelial cells is mediated through the functionally active form of CD44. J Biol Chem 275, 14939-14948. Nedvetzki, S., Gonen, E., Assayag, N., Reich, R., Williams, R.O., Thurmond, R.L., Huang, J.-F., Neudecker, B.A., Wang, F.-S., Turley, E.A., and Naor, D. (2004). RHAMM, a receptor for hyaluronan-mediated motility, compensates for CD44 in inflamed CD44-knockout mice: A different interpretation of redundancy. Proc Natl Acad Sci U S A 101, 18081-18086. Nilsson, S.K., Haylock, D.N., Johnston, H.M., Occhiodoro, T., Brown, T.J., and Simmons, P.J. (2003). Hyaluronan is synthesized by primitive hemopoietic cells, participates in their lodgment at the endosteum following transplantation, and is involved in the regulation of their proliferation and differentiation in vitro. Blood 101, 856-862.  149  Nilsson, S.K., Johnston, H.M., Whitty, G.A., Williams, B., Webb, R.J., Denhardt, D.T., Bertoncello, I., Bendall, L.J., Simmons, P.J., and Haylock, D.N. (2005). Osteopontin, a key component of the hematopoietic stem cell niche and regulator of primitive hematopoietic progenitor cells. Blood 106, 1232-1239. Noble, P.W. (2002). Hyaluronan and its catabolic products in tissue injury and repair. crotty 21, 25-29. Nolte, M.A., Kraal, G., and Mebius, R.E. (2004). Effects of fluorescent and nonfluorescent tracing methods on lymphocyte migration in vivo. Cytometry  A 61A, 35-44. Nombela-Arrieta, C., Pivarnik, G., Winkel, B., Canty, K.J., Harley, B., Mahoney, J.E., Park, S.-Y., Lu, J., Protopopov, A., and Silberstein, L.E. (2013). Quantitative imaging of hematopoietic stem and progenitor cell localization and hypoxic status in the bone marrow microenvironment. Nat Cell Biol 15, 533-543. O’Sullivan, D., van der Windt, G.J.W., Ching-Cheng Huang, S., Curtis, J.D., Chang, C.-H., Buck, M.D., Qiu, J., Smith, A.M., Lam, W.Y., DiPlato, L.M., et al. (2014). Memory CD8(+) T cells use cell intrinsic lipolysis to support the metabolic programming necessary for development. Immunity 41, 75-88. Oguro, H., Ding, L., and Morrison, Sean J. (2013). SLAM family markers resolve functionally distinct subpopulations of hematopoietic stem cells and multipotent progenitors. Cell Stem Cell 13, 102-116. Ohkawara, Y., Tamura, G., Iwasaki, T., Tanaka, A., Kikuchi, T., and Shirato, K. (2000). Activation and transforming growth factor-beta production in eosinophils by hyaluronan. Am J Respir Cell Mol Biol 23, 444-451. Okada, S., Nakauchi, H., Nagayoshi, K., Nishikawa, S., Miura, Y., and Suda, T. (1992). In vivo and in vitro stem cell function of c-kit- and Sca-1-positive murine hematopoietic cells. Blood 80, 3044-3050. Okada, S., Nakauchi, H., Nagayoshi, K., Nishikawa, S., Nishikawa, S., Miura, Y., and Suda, T. (1991). Enrichment and characterization of murine hematopoietic stem cells that express c-kit molecule. Blood 78, 1706-1712. Olaku, V., Matzke, A., Mitchell, C., Hasenauer, S., Sakkaravarthi, A., Pace, G., Ponta, H., and Orian-Rousseau, V. (2011). c-Met recruits ICAM-1 as a coreceptor to compensate for the loss of CD44 in Cd44 null mice. Mol Biol Cell 22, 2777-2786. Orkin, S.H., and Zon, L.I. (2008). Hematopoiesis: an evolving paradigm for stem cell biology. Cell 132, 631-644. Osawa, M., Hanada, K.-i., Hamada, H., and Nakauchi, H. (1996). Long-term lymphohematopoietic reconstitution by a single CD34-low/ negative hematopoietic stem cell. Science 273, 242-245.  150  Paiardini, M., Cervasi, B., Albrecht, H., Muthukumar, A., Dunham, R., Gordon, S., Radziewicz, H., Piedimonte, G., Magnani, M., Montroni, M., et al. (2005). Loss of CD127 expression defines an expansion of effector CD8+ T cells in HIV-infected individuals. J Immunol 174, 2900-2909. Parmar, K., Mauch, P., Vergilio, J.-A., Sackstein, R., and Down, J.D. (2007). Distribution of hematopoietic stem cells in the bone marrow according to regional hypoxia. Proc Natl Acad Sci U S A 104, 5431-5436. Passegué, E., Jamieson, C.H.M., Ailles, L.E., and Weissman, I.L. (2003). Normal and leukemic hematopoiesis: Are leukemias a stem cell disorder or a reacquisition of stem cell characteristics? Proc Natl Acad Sci U S A 100, 11842-11849. Peach, R.J., Hollenbaugh, D., Stamenkovic, I., and Aruffo, A. (1993). Identification of hyaluronic acid binding sites in the extracellular domain of CD44. J Cell Biol 122, 257-264. Perry, J.M., and Li, L. (2010). Functional assays for hematopoietic stem cell self-renewal. In Cellular Programming and Reprogramming: Methods and Protocols, S. Ding, ed. (Totowa, NJ: Humana Press), pp. 45-54. Petrey, A.C., and de la Motte, C.A. (2014). Hyaluronan, a crucial regulator of inflammation. Front Immunol 5, 101. Pipkin, M.E., Sacks, J.A., Cruz-Guilloty, F., Lichtenheld, M.G., Bevan, M.J., and Rao, A. (2010). Interleukin-2 and inflammation induce distinct transcriptional programs that promote the differentiation of effector cytolytic T cells. Immunity 32, 79-90. Ploemacher, R.E., and Brons, R.H. (1989). Separation of CFU-S from primitive cells responsible for reconstitution of the bone marrow hemopoietic stem cell compartment following irradiation: evidence for a pre-CFU-S cell. Exp Hematol 17, 263-266. Plumlee, C.R., Sheridan, B.S., Cicek, B.B., and Lefrançois, L. (2013). Environmental cues dictate the fate of individual CD8(+) T cells responding to infection. Immunity 39, 10.1016/j.immuni.2013.1007.1014. Politz, O., Gratchev, A., McCourt, P.A.G., Schledzewski, K., Guillot, P., Johansson, S., Svineng, G., Franke, P., Kannicht, C., Kzhyshkowska, J., et al. (2002). Stabilin-1 and -2 constitute a novel family of fasciclin-like hyaluronan receptor homologues. Biochem. J. 362, 155-164. Pollizzi, K.N., Sun, I.-H., Patel, C.H., Lo, Y.-C., Oh, M.-H., Waickman, A.T., Tam, A.J., Blosser, R.L., Wen, J., Delgoffe, G.M., and Powell, J.D. (2016). Asymmetric inheritance of mTORC1 kinase activity during division dictates CD8+ T cell differentiation. Nat Immunol 17, 704-711. Ponta, H., Sherman, L., and Herrlich, P.A. (2003). CD44: From adhesion molecules to signalling regulators. Nat Rev Mol Cell Biol 4, 33-45.  151  Pope, C., Kim, S.-K., Marzo, A., Williams, K., Jiang, J., Shen, H., and Lefrançois, L. (2001). Organ-specific regulation of the CD8 T cell response to Listeria monocytogenes infection. J Immunol 166, 3402-3409. Puri, M.C., and Bernstein, A. (2003). Requirement for the TIE family of receptor tyrosine kinases in adult but not fetal hematopoiesis. Proc Natl Acad Sci U S A 100, 12753-12758. Qian, H., Buza-Vidas, N., Hyland, C.D., Jensen, C.T., Antonchuk, J., Månsson, R., Thoren, L.A., Ekblom, M., Alexander, W.S., and Jacobsen, S.E.W. (2007). Critical role of thrombopoietin in maintaining adult quiescent hematopoietic stem cells. Cell Stem Cell 1, 671-684. Rathmell, J.C., Elstrom, R.L., Cinalli, R.M., and Thompson, C.B. (2003). Activated Akt promotes increased resting T cell size, CD28-independent T cell growth, and development of autoimmunity and lymphoma. Eur J Immunol 33, 2223-2232. Reed, R.K., Lilja, K., and Laurent, T.C. (1988). Hyaluronan in the rat with special reference to the skin. Acta Physiol Scand 134, 405-411. Reith, A.D., Rottepel, R., Giddens, E., Brady, C., Forrester, L., and Bernstein, A. (1990). W mutant mice with mild or severe developmental defects contain distinct point mutations in the kinase domain of the c-k t receptor. Genes Dev 4, 390-400. Restifo, N.P. (2014). Big bang theory of stem-like T cells confirmed. Blood 124, 476-477. Rochman, Y., Spolski, R., and Leonard, W.J. (2009). New insights into the regulation of T cells by γc family cytokines. Nat Rev Immunol 9, 480-490. Rosenman, S.J., Ganji, A.A., Tedder, T.F., and Gallatin, W.M. (1993). Syn-capping of human T lymphocyte adhesion/activation molecules and their redistribution during interaction with endothelial cells. J Leukoc Biol 53, 1-10. Rubinstein, M.P., Kovar, M., Purton, J.F., Cho, J.-H., Boyman, O., Surh, C.D., and Sprent, J. (2006). Converting IL-15 to a superagonist by binding to soluble IL-15Rα. Proc Natl Acad Sci U S A 103, 9166-9171. Ruffell, B., and Johnson, P. (2005). Chondroitin sulfate addition to CD44H negatively regulates hyaluronan binding. Biochem Biophys Res Commun 334, 306-312. Ruffell, B., and Johnson, P. (2008). Hyaluronan induces cell death in activated T cells through CD44. J Immunol 181, 7044-7054. Ruffell, B., and Johnson, P. (2009). The regulation and function of hyaluronan binding by CD44 in the immune system. Glycoforum. Ruffell, B., Poon, G.F.T., Lee, S.S.M., Brown, K.L., Tjew, S.-L., Cooper, J., and Johnson, P. (2011). Differential use of chondroitin sulfate to regulate hyaluronan binding by receptor CD44 in inflammatory and interleukin 4-activated macrophages. J Biol Chem 286, 19179-19190.  152  Russell, E.S., and Lawson, F.A. (1959). Selection and inbreeding for longevity of a lethal type. J Hered 50, 19-25. Sague, S.L., Tato, C., Puré, E., and Hunter, C.A. (2004). The regulation and activation of CD44 by natural killer (NK) cells and its role in the production of IFN-gamma. J Interferon Cytokine Res 24, 301-309. Sallusto, F., Geginat, J., and Lanzavecchia, A. (2004). Central memory and effector memory T cell subsets: function, generation, and maintenance. Ann Rev Immunol 22, 45-63. Sallusto, F., Lenig, D., Forster, R., Lipp, M., and Lanzavecchia, A. (1999). Two subsets of memory T lymphocytes with distinct homing potentials and effector functions. Nature 401, 708-712. Sarkar, S., Kalia, V., Haining, W.N., Konieczny, B.T., Subramaniam, S., and Ahmed, R. (2008). Functional and genomic profiling of effector CD8 T cell subsets with distinct memory fates. J Exp Med 205, 625-640. Schluns, K.S., and Lefrancois, L. (2003). Cytokine control of memory T-cell development and survival. Nat Rev Immunol 3, 269-279. Schluns, K.S., Williams, K., Ma, A., Zheng, X.X., and Lefrançois, L. (2002). Cutting edge: requirement for IL-15 in the generation of primary and memory antigen-specific CD8 T cells. J Immunol 168, 4827-4831. Schmits, R., Filmus, J., Gerwin, N., Senaldi, G., Kiefer, F., Kundig, T., Wakeham, A., Shahinian, A., Catzavelos, C., Rak, J., et al. (1997). CD44 regulates hematopoietic progenitor distribution, granuloma formation, and tumorigenicity. Blood 90, 2217-2233. Screaton, G.R., Bell, M.V., Jackson, D.G., Cornelis, F.B., Gerth, U., and Bell, J.I. (1992). Genomic structure of DNA encoding the lymphocyte homing receptor CD44 reveals at least 12 alternatively spliced exons. Proc Natl Acad Sci U S A 89, 12160-12164. Seita, J., and Weissman, I.L. (2010). Hematopoietic stem cell: self-renewal versus differentiation. Wiley Interdiscip Rev Syst Biol Med 2, 640-653. Siegelman, M.H., DeGrendele, H.C., and Estess, P. (1999). Activation and interaction of CD44 and hyaluronan in immunological systems. J Leuk Biol 66, 315-321. Simsek, T., Kocabas, F., Zheng, J., DeBerardinis, R.J., Mahmoud, A.I., Olson, E.N., Schneider, J.W., Zhang, C.C., and Sadek, H.A. (2010). The distinct metabolic profile of hematopoietic stem cells reflects their location in a hypoxic niche. Cell Stem Cell 7, 380-390. Sjöberg, S., Eriksson, E.E., Tivesten, Å., Carlsson, A., Klasson, A., Levin, M., Borén, J., and Krettek, A. (2009). CD44-deficiency on hematopoietic cells limits T-cell number but does not protect against atherogenesis in LDL receptor-deficient mice. Atherosclerosis 206, 369-374.  153  Skelton, T.P., Zeng, C., Nocks, A., and Stamenkovic, I. (1998). Glycosylation provides both stimulatory and inhibitory effects on cell surface and soluble CD44 binding to hyaluronan. J Biol Chem 140, 431-446. Spangrude, G., Heimfeld, S., and Weissman, I.L. (1988). Purification and characterization of mouse hematopoietic stem cells. Science 241, 58-62. Stemberger, C., Huster, K.M., Koffler, M., Anderl, F., Schiemann, M., Wagner, H., and Busch, D.H. (2007). A single naive CD8+ T cell precursor can develop into diverse effector and memory subsets. Immunity 27, 985-997. Stier, S., Ko, Y., Forkert, R., Lutz, C., Neuhaus, T., Grünewald, E., Cheng, T., Dombkowski, D., Calvi, L.M., Rittling, S.R., and Scadden, D.T. (2005). Osteopontin is a hematopoietic stem cell niche component that negatively regulates stem cell pool size. J Exp Med 201, 1781-1791. Stoklasek, T.A., Schluns, K.S., and Lefrançois, L. (2006). Combined IL-15/IL-15Rα immunotherapy maximizes IL-15 activity in vivo. J Immunol 177, 6072-6080. Stoll, S., Delon, J., Brotz, T.M., and Germain, R.N. (2002). Dynamic imaging of T cell-dendritic cell interactions in lymph nodes. Science 296, 1873-1876. Stoop, R., Katani, H., McNeish, J.D., Otterness, I.G., and Mikecz, K. (2001). Increased resistance to collagen-induced arthritis in CD44-deficient DBA/1 mice. Arthritis Rheum 44, 2922-2931. Sugiyama, T., Kohara, H., Noda, M., and Nagasawa, T. (2006). Maintenance of the hematopoietic stem cell pool by CXCL12-CXCR4 chemokine signaling in bone marrow stromal cell niches. Immunity 25, 977-988. Sukumar, M., Liu, J., Ji, Y., Subramanian, M., Crompton, J.G., Yu, Z., Roychoudhuri, R., Palmer, D.C., Muranski, P., Karoly, E.D., et al. (2013). Inhibiting glycolytic metabolism enhances CD8(+) T cell memory and antitumor function. The Journal of Clinical Investigation 123, 4479-4488. Taher, T.E.I., Smit, L., Griffioen, A.W., Schilder-Tol, E.J.M., Borst, J., and Pals, S.T. (1996). Signaling through CD44 is mediated by tyrosine kinases: association with p56lck IN T lymphocytes J Biol Chem 271, 2863-2867. Takeda, M., Terasawa, H., Sakakura, M., Yamaguchi, Y., Kajiwara, M., Kawashima, H., Miyasaka, M., and Shimada, I. (2003). Hyaluronan recognition mode of CD44 revealed by cross-saturation and chemical shift perturbation experiments. J Biol Chem 278, 43550-43555.  Takemoto, N., Intlekofer, A.M., Northrup, J.T., Wherry, E.J., and Reiner, S.L. (2006). Cutting edge: IL-12 inversely regulates T-bet and eomesodermin expression during pathogen-induced CD8+ T cell differentiation. J Immunol 177, 7515-7519.  154  Tan, J.T., Dudl, E., LeRoy, E., Murray, R., Sprent, J., Weinberg, K.I., and Surh, C.D. (2001). IL-7 is critical for homeostatic proliferation and survival of naïve T cells. Proc Natl Acad Sci U S A 98, 8732-8737. Teder, P., Vandivier, R.W., Jiang, D., Liang, J., Cohn, L., Puré, E., Henson, P.M., and Noble, P.W. (2002). Resolution of lung inflammation by CD44. Science 296, 155-158. Termeer, C., Johannsen, H., Braun, T., Renkl, A., Ahrens, T., Denfeld, R.W., Lappin, M.B., Weiss, J.M., and Simon, J.C. (2001). The role of CD44 during CD40 ligand-induced dendritic cell clustering and maturation. J Leukoc Biol 70, 715-722. TeSlaa, T., and Teitell, M.A. (2014). Techniques to monitor glycolysis. Methods Enzymol 542, 91-114. Thompson, L.J., Kolumam, G.A., Thomas, S., and Murali-Krishna, K. (2006). Innate inflammatory signals induced by various pathogens differentially dictate the IFN-I dependence of CD8 T cells for clonal expansion and memory formation. J Immunol 177, 1746-1754. Thorén, L.A., Liuba, K., Bryder, D., Nygren, J.M., Jensen, C.T., Qian, H., Antonchuk, J., and Jacobsen, S.-E.W. (2008). Kit regulates maintenance of quiescent hematopoietic stem cells. J Immunol 180, 2045-2053. Ting, S.B., Deneault, E., Hope, K., Cellot, S., Chagraoui, J., Mayotte, N., Dorn, J.F., Laverdure, J.-P., Harvey, M., Hawkins, E.D., et al. (2012). Asymmetric segregation and self-renewal of hematopoietic stem and progenitor cells with endocytic Ap2a2. Blood 119, 2510-2522. Toole, B.P. (2004). Hyaluronan: from extracellular glue to pericellular cue. Nat Rev Cancer 4, 528+. Trumpp, A., Essers, M., and Wilson, A. (2010). Awakening dormant haematopoietic stem cells. Nat Rev Immunol 10, 201-209. Tsukita, S., Oishi, K., Sato, N., Sagara, J., Kawai, A., and Tsukita, S. (1994). ERM family members as molecular linkers between the cell surface glycoprotein CD44 and actin-based cytoskeletons. J Cell Biol 126, 391-401. Turley, E.A. (1982). Purification of a hyaluronate-binding protein fraction that modifies cell social behavior. Biochem Biophys Res Commun 108, 1016-1024. Turner, M.J., Jellison, E.R., Lingenheld, E.G., Puddington, L., and Lefrançois, L. (2008). Avidity maturation of memory CD8 T cells is limited by self-antigen expression. J Exp Med 205, 1859-1868. Underhill, C.B., Nguyen, H.A., Shizari, M., and Culty, M. (1993). CD44 Positive Macrophages Take Up Hyaluronan During Lung Development. Dev Biol 155, 324-336.  155  van der Windt, Gerritje J.W., Everts, B., Chang, C.-H., Curtis, Jonathan D., Freitas, Tori C., Amiel, E., Pearce, Edward J., and Pearce, Erika L. (2012). Mitochondrial respiratory capacity is a critical regulator of CD8+ T cell memory development. Immunity 36, 68-78. Veiseh, M., Kwon, D.H., Borowsky, A.D., Tolg, C., Leong, H.S., Lewis, J.D., Turley, E.A., and Bissell, M.J. (2014). Cellular heterogeneity profiling by hyaluronan probes reveals an invasive but slow-growing breast tumor subset. Proc Natl Acad Sci USA 111, E1731-E1739. Vivier, E., van de Pavert, S.A., Cooper, M.D., and Belz, G.T. (2016). The evolution of innate lymphoid cells. Nat Immunol 17, 790-794. Wang, J., and Kubes, P. (2016). A reservoir of mature cavity macrophages that can rapidly invade visceral organs to affect tissue repair. Cell 165, 668-678. Wang, Z., Wu, Y., Wang, H., Zhang, Y., Mei, L., Fang, X., Zhang, X., Zhang, F., Chen, H., Liu, Y., et al. (2014). Interplay of mevalonate and Hippo pathways regulates RHAMM transcription via YAP to modulate breast cancer cell motility. Proc Natl Acad Sci U S A 111, E89-E98. Waterstrat, A., Liang, Y., Swiderski, C.F., Shelton, B.J., and Van Zant, G. (2010). Congenic interval of CD45/Ly-5 congenic mice contains multiple genes that may influence hematopoietic stem cell engraftment. Blood 115, 408-417. Weigel, P.H. (2015). Hyaluronan synthase: the mechanism of initiation at the reducing end and a pendulum model for polysaccharide translocation to the cell exterior. Int J Cell Biol. Wieman, H.L., Wofford, J.A., and Rathmell, J.C. (2007). Cytokine stimulation promotes glucose uptake via phosphatidylinositol-3 kinase/Akt regulation of Glut1 activity and trafficking. Mol Biol Cell 18, 1437-1446. Wilson, A., Laurenti, E., Oser, G., van der Wath, R.C., Blanco-Bose, W., Jaworski, M., Offner, S., Dunant, C.F., Eshkind, L., Bockamp, E., et al. (2008). Hematopoietic stem cells reversibly switch from dormancy to self-renewal during homeostasis and repair. Cell 135, 1118-1129. Winkler, I.G., Barbier, V., Wadley, R., Zannettino, A.C.W., Williams, S., and Lévesque, J.-P. (2010). Positioning of bone marrow hematopoietic and stromal cells relative to blood flow in vivo: serially reconstituting hematopoietic stem cells reside in distinct nonperfused niches. Blood 116, 375-385. Wittig, B., Schwärzler, C., Föhr, N., Günthert, U., and Zöller, M. (1998). Cutting edge: curative treatment of an experimentally induced colitis by a CD44 variant V7-specific antibody. J Immunol 161, 1069-1073. Wittig, B.M., Johansson, B., Zöller, M., Schwärzler, C., and Günthert, U. (2000). Abrogation of experimental colitis correlates with increased apoptosis in mice deficient for Cd44 variant exon 7 (Cd44v7). J Exp Med 191, 2053-2064.  156  Wong, N.K.Y., Lai, J.C.Y., Birkenhead, D., Shaw, A.S., and Johnson, P. (2008). CD45 down-regulates Lck-mediated CD44 signaling and modulates actin rearrangement in T cells. J Immunol 181, 7033-7043. Wright, D.E., Bowman, E.P., Wagers, A.J., Butcher, E.C., and Weissman, I.L. (2002). Hematopoietic stem cells are uniquely selective in their migratory response to chemokines. J Exp Med 195, 1145-1154. Xie, Y., Yin, T., Wiegraebe, W., He, X.C., Miller, D., Stark, D., Perko, K., Alexander, R., Schwartz, J., Grindley, J.C., et al. (2009). Detection of functional haematopoietic stem cell niche using real-time imaging. Nature 457, 97-101. Xue, H.-H., Kovanen, P.E., Pise-Masison, C.A., Berg, M., Radovich, M.F., Brady, J.N., and Leonard, W.J. (2002). IL-2 negatively regulates IL-7 receptor α chain expression in activated T lymphocytes. Proc Natl Acad Sci U S A 99, 13759-13764. Yamada, H., Matsuzaki, G., Chen, Q., Iwamoto, Y., and Nomoto, K. (2001). Reevaluation of the origin of CD44(high) "memory phenotype" CD8 T cells: comparison between memory CD8 T cells and thymus-independent CD8 T cells. Eur J Immunol 31, 1917-1926. Yang, L., Bryder, D., Adolfsson, J., Nygren, J., Månsson, R., Sigvardsson, M., and Jacobsen, S.E.W. (2005). Identification of Lin– Sca1+kit+CD34+Flt3– short-term hematopoietic stem cells capable of rapidly reconstituting and rescuing myeloablated transplant recipients. Blood 105, 2717-2723. Yilmaz, Ö.H., Kiel, M.J., and Morrison, S.J. (2006). SLAM family markers are conserved among hematopoietic stem cells from old and reconstituted mice and markedly increase their purity. Blood 107, 924-930. Yoshihara, H., Arai, F., Hosokawa, K., Hagiwara, T., Takubo, K., Nakamura, Y., Gomei, Y., Iwasaki, H., Matsuoka, S., Miyamoto, K., et al. (2007). Thrombopoietin/MPL signaling regulates hematopoietic stem cell quiescence and interaction with the osteoblastic niche. Cell Stem Cell 1, 685-697. Yoshioka, Y., Kozawa, E., Urakawa, H., Arai, E., Futamura, N., Zhuo, L., Kimata, K., Ishiguro, N., and Nishida, Y. (2013). Suppression of hyaluronan synthesis alleviates inflammatory responses in murine arthritis and in human rheumatoid synovial fibroblasts. Arthritis Rheum 65, 1160-1170. Yusoff, A.A.M., Ahmad, F., Idris, Z., Hasnan, J., and Abdullah, J.M. (2015). Understanding mitochondrial DNA in brain tumorigenesis. In Molecular Considerations and Evolving Surgical Management Issues in the Treatment of Patients with a Brain Tumor (Rijeka: InTech), p. Ch. 0. Zehn, D., Lee, S.Y., and Bevan, M.J. (2009). Complete but curtailed T cell response to very low affinity antigen. Nature 458, 211-214.  157  Zhong, J.F., Zhan, Y., Anderson, W.F., and Zhao, Y. (2002). Murine hematopoietic stem cell distribution and proliferation in ablated and nonablated bone marrow transplantation. Blood 100, 3521-3526. Zhou, B., McGary, C.T., Weigel, J.A., Saxena, A., and Weigel, P.H. (2003). Purification and molecular identification of the human hyaluronan receptor for endocytosis. Glycobiol 13, 339-349. Zhou, B., Weigel, J.A., Fauss, L., and Weigel, P.H. (2000). Identification of the hyaluronan receptor for endocytosis (HARE). J. Biol. Chem. 275, 37733-37741. Zou, Y.-R., Kottmann, A.H., Kuroda, M., Taniuchi, I., and Littman, D.R. (1998). Function of the chemokine receptor CXCR4 in haematopoiesis and in cerebellar development. Nature 393, 595-599.    158  Appendix A  HA Binding by the Invasive Ductal Carcinoma Cell Line MDA-MB-231 is Mediated by CD44, and not by RHAMM  A.1 Introduction and Rationale HA-binding by the human breast tumour cell line MDA-MB-231 has been reported to correlate with metastasis (Veiseh et al., 2014). Veiseh et al found that, unlike other breast cancer cell lines, MDA-MB-231 cells has heterogeneous HA binding, which follows a bimodal pattern. When sorted and transferred into mice, the HA-binding fraction forms a significantly higher number of metastatic foci than the non-binding fraction. These cells have high CD44 expression and bound FL-HA. The authors also suggest that RHAMM (hyaluronan-mediated motility receptor) is expressed on the surface of the MDA-MB-231 cells, but this is not supported by the data shown, where there is little above-background staining by anti-RHAMM antibody.  In collaboration with Dr. Christopher Maxwell at the British Columbia Children’s Hospital Research Institute, I aimed to evaluate the contribution of CD44 and RHAMM to HA binding by MDA-MB-231 tumour cells. RHAMM was first described as a protein that increases association to HA when added to cells, and altered cell motility (Turley, 1982), leading to its name. However, RHAMM has little similarity in protein sequence domain organization to other well-known HA-binding proteins, such as CD44, TSG-6 and LYVE1, and does not contain the Link module found in other HA-binding proteins (Day and Prestwich, 2002) (Fig. 1.2). More specifically, RHAMM lacks the HA-binding LINK domain shared by other HA-binding proteins. The RHAMM protein also lacks a membrane spanning domain and HMMR, the gene that encodes RHAMM, lacks a signal  159  sequence (Day, 2001), suggesting that RHAMM is normally an intracellular protein. It was thus surprising from the biochemical and protein structural standpoints that RHAMM was described as a membrane protein that binds HA for uptake by Veiseh et al.  RHAMM has been implicated in cancer migration and metastasis (Wang et al., 2014). More recently, studies have shown that RHAMM is localised intracellularly in the cytoplasm of breast tumour cells and mouse embryonic stem cells and is involved in regulating mitosis and mitotic spindles, through its association with microtubules (Chen et al., 2014; Jiang et al., 2013; Maxwell et al., 2005; Maxwell et al., 2003). RHAMM is localized to the site of microtubule assembly and centrosomes in breast tumour cells and mouse embryonic stem cells (Chen et al., 2014; Jiang et al., 2013). RHAMM interacts with microtubules and centrosome though its amino and carboxyl termini, respectively (Maxwell et al., 2005), and co- immunoprecipitates with dynein (Maxwell et al., 2003). Over-expression of RHAMM results in defects in the mitosis, while inhibition of RHAMM functions disrupts spindle integrity after spindle assembly (Chen et al., 2014; Maxwell et al., 2005; Maxwell et al., 2003). Furthermore, RHAMM is expressed intracellularly in mouse embryonic stem cells, but not on the cell surface (Jiang et al., 2013).  The goal of this study is to determine the role of CD44 and RHAMM in HA binding by MDA-MB-231 cells.  A.2 Methods Cell culture  MDA-MB-231 cells were obtained from the Maxwell lab and were cultured in RPMI-1640 medium with 2.05 mM L-glutamine (HyClone). Knockdown of RHAMM using shRNA was  160  performed by Dr. Christopher Maxwell’s lab from British Columbia Children’s Hospital Research Institute (Maxwell et al., 2011). Briefly, RHAMM-specific shRNA constructs (5′-CGTCTCCTCTATGAAGAACTA-3′ and 5′-GCCAACTCAAATCGGAAGTAT-3′) were purchased (Sigma-Aldrich), and were independently validated by the manufacturer for reduction in mRNA levels. The plasmids for lentiviral packaging, envelope and control non-hairpin (NHP) control were purchased from Addgene. Lentiviral particles were produced following Addgene protocol. Transduced cells were selected with 0.5μg/ml puromycin (GIBCO) and maintained with 0.3μg/ml puromycin. Flow analysis The following monoclonal antibodies specific for human antigens were used for flow cytometry:  CD44 (IM7.8) conjugated to Pacific Blue (Biomedical Research Centre Antibody Bab [BRC AbLab] at UBC); RHAMM (EPR4054, recognizes the N-terminus) (Abcam); and HA blocking CD44 (Hermes-1) tissue culture supernatant generated in-house. Propidium iodide (Sigma-Aldrich, St. Louis, MO) was used to stain dead cells. Rooster comb HA was conjugated to Fluorescein (FL-HA) according to (de Belder and Wik, 1975).   2 x 105 cells were suspended in FACS buffer (PBS, 0.5% BSA and 2mM EDTA). Cells incubated with fluorescently labelled monoclonal antibodies and FL-HA for 20 min for surface staining. Cells were then stained with 5 µm/ mL propidium iodide to exclude dead cells. For pre-treatment with hyaluronidase, cells were incubated with 20 U/ mL of hyaluronidase from Streptomyces hyaluronlyticus (Calbiochem) at 37oC for 30 minutes, followed by 10 washes with FACS buffer. For HA blocking with Hermes-1 antibody, cells were incubated with 1/ 30 dilution  161  of Hermes-1 containing tissue culture supernatant on ice for 30 minutes. To assess intracellular RHAMM expression, cells were washed with PBS, pelleted, and the pellet was loosened. 1mL of ice-cold methanol was added dropwise to the cells and then kept on ice for 20 min. Permeabilized cells were pelleted and resuspended in PBS with 10% FBS and anti-RHAMM antibody. Labelled cells were analyzed on an LSRII cytometer (BD Biosciences) using FACSDiva acquisition software (BD Biosciences). Data analyses were performed using FlowJo (Treestar, Ashland, OR).   A.3 Results A.3.1 HA-binding by breast tumour line MDA-MB-231 is mediated by CD44 MDA-MB-231 exhibited homogeneity in high FL-HA binding and high CD44 expression (Fig. A.1A-C). The level of FL-HA binding by the entire population could be further increased by pre-treatment with hyaluronidase, suggesting that cells were covered with pericellular HA (Fig. A.1A-C). FL-HA binding was completely abolished by pre-incubating the cells with an anti-CD44 antibody against the HA-binding region of CD44 (Fig.A.1D), indicating that HA binding is mediated by CD44.  A.3.2 RHAMM is dispensable for HA-binding by breast tumour line MDA-MB-231 RHAMM expression was observed intracellularly, but not on the cell surface of MDA-MB-231 (Fig. A.2A). Since RHAMM was not detectable on the surface, it could not be blocked with incubation with the anti-RHAMM antibody. To further evaluate the role of RHAMM in HA binding, we knocked down RHAMM by transducing the cells with one of two RHAMM-specific shRNA. RHAMM-specific shRNA (RH1 and RH2) reduced RHAMM expression (Fig. A.2B,   162   Figure A.6.1 FL-HA binding by MDA-MB-231 tumour cells is mediated by CD44  (A) FL-HA and CD44 expression of unstained controls cells (-), stained cells (control) and cells previously treated with hyaluronidase (H’ase). (B) CD44 expression of unstained control (-) and stained (CD44) cells. (C) FL-HA binding by unstained control cells (-), stained cells (control) and cells previously treated with hyaluronidase (H’ase). (D) FL-HA and CD44 expression of unstained cells (left, gray shaded), control cells (control), and cells previously blocked with anti-CD44 monoclonal antibody (αCD44 mAb). Plots are representative of at least two independent experiments.  163       Figure A.6.2 FL-HA binding by MDA-MB-231 is not mediated by RHAMM (A) Surface and intracellular RHAMM expression of cells stained with secondary alone (2o alone), polyclonal anti-RHAMM rabbit sera (αRHAMM + 2o), and control rabbit sera (poly sera + 2o). (B-D) Cells were unstained as controls or transduced with the non-hairpin control (NHP) or RHAMM-specific shRNA 1 and 2 (RH1 and RH2). (B) Immunoblot for RHAMM and GAPDH expression. (C) FL-HA binding versus intracellular RHAMM expression. (D) Histogram of FL-HA binding by cells. Plots are representative of at least two independent experiments.  164  C), but did not reduce HA binding by MDA- MB-231 cells (Fig. A.2D). In fact, knocking down of RHAMM expression slightly increased HA binding by MDA-MB-231 cells (Fig. A.2D).  A.4 Discussion In this study, MDA-MB-231 human breast tumour cells were found to constitutively bind HA as a homogenous population, rather than a heterogeneous population as described in (Veiseh et al., 2014). Also, the cells were associated with pericellular HA, as digestion with H’ase further increased FL-HA binding by these cells. More importantly, RHAMM expression was not detected on the cell surface, but was detected intracellularly and thus supports the recent papers describing it as a microtubule binding protein (Chen et al., 2014; Jiang et al., 2013; Maxwell et al., 2005; Maxwell et al., 2003). Furthermore, RHAMM is dispensable for surface HA binding by MDA-MB-231 and its knock down did not reduce surface HA binding by these cells. Thus, in these cells, CD44, not RHAMM mediates surface HA binding by these cells. 

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.24.1-0348315/manifest

Comment

Related Items