Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Formation of functional selectin ligands on activated T cells and thymic progenitors : the role of core… Merzaban, Jasmeen Sayed 2005

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

Item Metadata


831-ubc_2005-105372.pdf [ 28.91MB ]
JSON: 831-1.0099816.json
JSON-LD: 831-1.0099816-ld.json
RDF/XML (Pretty): 831-1.0099816-rdf.xml
RDF/JSON: 831-1.0099816-rdf.json
Turtle: 831-1.0099816-turtle.txt
N-Triples: 831-1.0099816-rdf-ntriples.txt
Original Record: 831-1.0099816-source.json
Full Text

Full Text

FORMATION OF FUNCTIONAL SELECTIN LIGANDS ON ACTIVATED T CELLS AND THYMIC PROGENITORS: THE ROLE OF CORE 2 $1,6-N-ACETYLGLUCOSAMINYLTRANSFERASES IN THE CONTROL OF LYMPHOCYTE TRAFFICKING AND THYMIC PROGENITOR HOMING by JASMEEN SAYED MERZABAN B.Sc, The University of British Columbia, 2000 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Experimental Medicine) THE UNIVERSITY OF BRITISH COLUMBIA August 2005 © Jasmeen Merzaban, 2005 ABSTRACT The core 2 p l^-A'-acetylglucosaminyltransferase (C2GlcNAcT) family of enzymes (C2GlcNAcT-I,-II,-III) synthesize branched O-glycans. A significant body of work highlights the importance of C2GlcNAcT-I in controlling selectin-ligand-mediated cell trafficking while little is known about the role of the two other C2GlcNAcT isoenzymes. 1) The first objective of this thesis is to determine in vitro and in vivo T cell stimulation conditions that guide P-selectin ligand expression in absence of C2GlcNAcT-I. Mitogen stimulation of splenocytes maintained under very-high-density culture conditions uncovers C2GlcNAcT-I-independent P-selectin ligand formation in CD8 + T cells, but not CD4 + T cells. CD8 + T cells of C2GlcNAcT-In u 1 1 mice also roll under shear flow on immobilized P-selectin in a PSGL-l(P-selectin-glycoprotein-ligand- l)-specific manner. Using RT-PCR analysis, we identify C2GlcNAcT-III as the likely source of core 2 activity. Up regulation of P-selectin ligand in C2GlcNAcT-Fu 1 1 CD8 + T cells correlates with higher core 2 enzyme activity, as measured by a standard enzymatic assay and cell-surface binding of the core 2-sensitive mAb 1B11. This reveals the well-established C2GlcNAcT-I substrates - CD43 and CD45 - as additional physiological targets of C2GlcNAcT-III. Adoptive transfer of C2GlcNAcT-In u 1 1 T cells from mice transgenic for the male antigen (HY) T cell receptor shows that C2GlcNAcT-I-independent P-selectin ligand formation occurs on CD8 + T cells under in vivo stimulation conditions. In consequence, C2GlcNAcT-III emerges as a contributor to P-selectin ligand formation that may co-operate with C2GlcNAcT-I to control CD8 + T cell trafficking. 2) A second objective of this thesis is to explore whether C2GlcNAcT-I plays a role in controlling early T cell progenitor migration to the thymus. PSGL-1 is expressed by these thymic progenitors. In order for PSGL-1 to be recognized by P-selectin which is constitutively expressed at low levels on the thymic endothelium, it must be modified by C2GlcNAcT-I. Using mice with deficiencies in PSGL-1, C2GlcNAcT-I and P-selectin, we employ parabiosis and competitive repopulation to show that C2GlcNAcT-I modification of PSGL-1 expressed on thymic progenitors is a functionally imperative component of the thymic homing process. TABLE OF CONTENTS Abstract ii Table of Contents iv List of Tables ix List of Figures x List of Abbreviations xiii Acknowledgments xvii CoAuthorship Statement xix CHAPTER 1 INTRODUCTION 1 1.1 Overview 1 1.2 Leukocyte trafficking 3 1.3 Glycans in biology 12 1.3.1 N-glycans 13 1.3.2 Glycolipids 17 1.3.3 Glycophosphatidylinositol Linkages 18 1.3.4 Glycosaminoglycans 19 1.3.5 Hyaluronan 20 1.3.6 O-glycans 21 1.3.7 Focus 31 1.4 Decorated ligands attract selectin counter-receptors 31 1.4.1 Leukocyte-selectin 32 L-selectin 32 iv Ligands for L-selectin -.35 1.4.2 Endothelial cell-selectin 39 E-selectin 39 Ligands for E-selectin 40 1.4.3 Platelet (& endothelial) cell-selectin 45 P-selectin 45 Ligands for P-selectin 46 1.4.4 Selectin knockout mice 51 Single knockout mice 51 Double knockout mice 52 Triple knockout mice 54 1.5 Glycosyltransferases involved in selectin ligand biosynthesis 56 1.5.1 Glycosyltransferases involved in formation of L-selectin ligands 57 1.5.2 Glycosyltransferases involved in the formation of P-selectin ligands 63 1.5.3 Glycosyltransferases involved in the formation of E-selectin ligands 72 1.6 A closer look at Core O glycan branching 76 1.6.1 CD43 and the discovery of C2GlcNAcT 76 1.6.2 Regulation of C2GlcNAcT activity 84 1.6.3 Regulation of P-selectin ligand formation 87 1.7 Thymic progenitor homing 90 1.7.1 Origin of T cell progenitors 90 1.7.2 Thymic environment and T cell development 96 1.7.3 Thymic progenitor migration from the blood to the cortico-medullary junction of the thymus 102 1.8 Thesis objectives 104 1.8.1 C2GlcNAcTs and their role in the formation of functional selectin ligands in Activated T cells 104 1.8.2 C2GlcNAcT-I and Thymic Progenitor Homing 105 CHAPTER 2 Materials and Methods 107 2.1 Mice 107 2.1.1 Mice used for in vitro and in vivo T cell stimulations 107 2.1.2 Mice used for thymic homing experiments 107 2.2 Media used and Antibodies used for experiments 108 2.3 Cell Isolation and T cell Cultures 109 2.4 Flow cytometric analysis 109 2.4.1 P-selectin-hIG and E-selectin-hIG labeling of activated cells 109 2.4.2 Procedure used to harvest and stain tissues in thymic homing experiments 110 2.5 Assays 111 2.5.1 Core 2 and core 4 enzymatic assays 111 2.5.2 CD43 Western blots 111 2.5.3 In vitro flow chamber 112 2.6 Adoptive Transfer models and Parabiosis 113 2.6.1 HY adoptive transfer model 113 2.6.2 Competitive Repopulation 114 2.6.3 Parabiosis 115 2.7 Real-time RT-PCR 115 CHAPTER 3 P-selectin binds CD8+ T cells in the absence of C2GlcNAcT-1 119 3.1 Introduction 119 vi 3.2 In vitro results 120 3.2.1 P-selectin binds PSGL-1 on C2GlcNAcT-In u 1 1 Con A-activated Splenocytes 120 3.2.2 Increased Core 2 O-glycan branching on CD43 and CD45 correlates with increased P-selectin ligand formation 124 3.2.3 C2GlcNAcT-I RNA and C2GlcNAcT-III RNA are expressed in activated splenocytes 130 Identity between C2GlcNAcT isoenzymes 130 C2GlcNAcT isoenzyme RNA expression in activated T cells - low-density versus high-density cultures 132 3.2.4 P-selectin ligand formation in activated C2GlcNAcT-In u 1 1 cells is restricted to CD8 + T cells 134 3.2.5 Induction of core 2 activity by a secreted/soluble factor within high-density culture supernatant 137 3.2.6 IL-2 and IL-4 differentially regulate the P-selectin ligand formation 138 3.2.7 Activated C2GlcNAcT-In u 1 1 cells roll on immobilized P-selectin in a PSGL-1 -dependent manner 140 3.3 In vivo T cell receptor signalling induces Core 2 O-glycan-branched modifications independent of C2GlcNAcT-1 142 3.3.1 In vivo activation of C2GlcNAcT-f u 1 1 CD8 + T cells induces P-selectin ligand... 142 3.3.2 CD45 is modified upon in vivo activation of CD8 + T cells 144 3.4 Discussion 146 CHAPTER 4 E-selectin binds activated CD8+ T cells in absence of C2GlcNAcT-1 153 4.1 Introduction 153 4.2 E-selectin binds activated T cells in the absence of C2GlcNAcT-1 154 vii 4.3 Activated C2GlcNAcT-In u" cells roll on immobilized E-selectin in a PSGL-1-independent C2GlcNAcT-III-independent manner 155 4.4 Activated CD4 + and CD8 + T cells bind E-selectin-hIG differently 156 4.5 Discussion 158 CHAPTER 5 Role of selectins in progenitor homing to the thymus 161 5.1 Introduction 161 5.2 P-selectin ligand is expressed on lymphoid progenitors 161 5.3 Parabiosis 163 5.3.1 Wild-type Thyl.2 parabiosed to wild-type Thyl.l 165 5.3.2 C2GlcNAcT-In u 1 1 Thyl .2 parabiosed with wild-type Thyl .1 166 5.3.3 PSGL-l n u 1 1 Thyl.2 parabiosed with wild-type Thyl.l 169 5.3.4 P-selectinnu11 Thyl.2 parabiosed with wild-type Thyl.l 170 5.4 Competitive Repopulation Studies 171 5.4.1 Competitions into irradiated wild-type and P-selectinnu" recipients 171 5.4.2 Competitions into non- irradiated IL-7Rn u 1 1 recipients 173 5.5 Discussion 176 CHAPTER 6 Summary and Future Directions 187 References 189 Appendices 223 Appendix I: Core 2 branching of PSGL-1 is observed on double -negative T-regs?....223 Appendix II: Lack of CD44 does not appear to inhibit thymic progenitor homing 226 List of Publications 230 L I S T O F T A B L E S Table 2-1: Mice used in experiments represented in chapters indicated 108 Table 5-1: Thymic progenitor populations and their phenotypes ; 163 ix LIST O F F I G U R E S Figure 1-1: Migratory routes of T cells 2 Figure 1-2: T cell trafficking and the players involved 4 Figure 1-3: Major core O-glycan structures in biology 24 Figure 1-4: Capping structures for glycans of selectin ligands 26 Figure 1-5: Selectins and their ligands 34 Figure 1-6: Comparison of human and mouse domains of PSGL-1 49 Figure 1-7 Glycosylation on human and murine PSGL-1 50 Figure 1-8: Glycosyltransferases involved in creating common selectin ligands 58 Figure 1-9: Structure and function of C2GlcNAcTs 80 Figure 1-10: C2GlcNAcT-I modifies CD43 81 Figure 1-11: HSC blood cell commitment models: under revision 95 Figure 1-12: T cell development and the thymic microenvironments 97 Figure 3-1: In vitro culture stimulation conditions 121 Figure 3-2: P-selectin ligand formation in Con A-activated C2GlcNAcT-In u 1 1 splenocytes is cell-culture-density dependent 122 Figure 3-3: High-density (HD) cultured C2GlcNAcT-Fu 1 1 Con A blasts express significant core 2 activity 125 Figure 3-4: HD-cultured C2GlcNAcT-In u 1 1 Con A blasts do not contain core 4 activity 126 Figure 3-5: CD43 antibody recognition 127 Figure 3-6: Western blot showing that CD43 is modified by core 2 branches in the absence of C2GlcNAcT-1 128 Figure 3-7: CD43 and CD45 on C2GlcNAcT-In u 1 1 CD8 + T cells are modified by an alternate C2GlcNAcT enzyme .' 129 x Figure 3-8: Murine C2GlcNAcT-I, C2GlcNAcT-II and C2GlcNAcT-III show significant sequence identity 131 Figure 3-9: RT-PCR demonstrates expression of C2GlcNAcT-I and C2GlcNAcT-III RNA in activated CD8 + T cells 133 Figure 3-10: P-selectin ligand formation in Con A-activated C2GlcNAcT-In u 1 1 cells is restricted to CD8 + T cells 135 Figure 3-11: mRNA expression and activity of activated CD4 + and CD8 + C2GlcNAcT-I n u U T cells 136 Figure 3-12: High-density supernatant induces P-selectin ligand formation 138 Figure 3-13: Regulation of C2GlcNAcT by IL-2 and IL-4 139 Figure 3-14: Activated C2GlcNAcT-In u 1 1 CD8 + T cells roll on immobilized P-selectin-hlG. 141 Figure 3-15: In vivo activated C2GlcNAcT-In u 1 1 CD8 + T cells can form functional P-selectin ligand 143 Figure 3-16: In vivo activation of CD8 + T cells modifies CD45 in absence of C2GlcNAcT-I. 145 Figure 4-1: E-selectin binds activated CD8 + T cells in the absence of C2GlcNAcT-I and PSGL-1 155 Figure 4-2: Activated C2GlcNAcT-In u" CD8 + T cells roll on immobilized E-selectin independent of PSGL-1 156 Figure 4-3: Activated CD4 + T cells do not bind E-selectin in the absence of C2GlcNAcT-I whereas CD8 + T cells do 157 Figure 5-1: Diagram of parabiotic mouse model 165 Figure 5-2: C2GlcNAcT-I, PSGL-1 and P-selectin have a functional role in progenitor homing to the thymus 167 xi Figure 5-3: Wild-type cells have a competitive advantage over C2GlcNAcT-In u" and PSGL-l n u 1 1 cells in repopulating the thymus of wild-type recipient mice, and this competitive advantage is P-selectin dependent 172 Figure 5-4: Wild-type cells have a competitive advantage over C2GlcNAcT-I n u l 1 and PSGL-l n u " cells in repopulating the thymus of non- irradiated IL-7R n u U mice 174 Figure 5-5: T cell progenitor entry into the thymus is mediated by C2GlcNAcT-I-modified PSGL-1 and endothelial expressed P-selectin 186 Appendix Figure A - l : P-selectin binds activated wild-type but not C2GlcNAcT-In u 1 1 CD4+CD25+ T reg cells 224 Figure A-2: P-selectin binds double-negative T cells in the absence of C2GlcNAcT-1 225 Figure A-3: Lack of CD44 does not inhibit thymic progenitor homing in competitive repopulation experiments 227 Figure A-4: Molecules involved in CTP homing to the thymus 229 * ALL GRAPHICS IN FIGURES DESIGNED BY AMANDA MERZABAN xii LIST OF ABBREVIATIONS Ab antibody APC allophycocyanin Asn asparagine Asp aspartic acid BSA bovine serum albumin C2GlcNAcT Core 2 1-6-N-acetylglucosaminyltransferase CEC cortical epithelial cells CFSE Carboxy-fluorescein diacetate, succinimidyl ester CHS contact hypersensitivity CLA cutaneous lymphoid antigen CLP common lymphoid progenitor CMJ cotico-medullary junction CMP common myeloid progenitor Con A conconavalin A CR consensus repeats of complement regulatory protein CTO cell tracker orange CTP circulating thymic progenitor DC dendritic cell DC-CK1 dendritic cell chemokine 1 DMEM Dulbecco's Modified Eagle Media DN double negative DNA deoxyribonucleic acid DP double positive DTH delayed-type hypersensitivity EDTA ethylenediamine tetra-acetate EGF epidermal growth factor ELC Epstein Barr virus-induced receptor ligand chemokine ER endoplasmic reticulum ERAD endoplasmic reticulum-associated degradation ESL-1 E-selectin ligand 1 EST expressed sequence tag ETP early thymic progenitor FACS fluorescence activated cell sorting FBS fetal bovine serum FITC fluorescein isothiocyanate Flt3 FMS-like tyrosine kinase 3 Fuc fiicose FucT fucosyltransferase FX GDP-4-keto-6-deoxymannose 3,5-epimerase, 4-reductase GAG glycosaminoglycan GalNAc ^/-acetylgalactosamine GalT galactosyltransferase Glc glucose GlcNAc 7V-acetylglucosamine GlyCAM-1 glycosylation-dependent cell adhesion molecule-1 GM-CSF granulocyte and macrophage colony stimulating factor GMP granulocyte/macrophage progenitor GPI glycophosphatidylinositol HA hyaluronan HBSS Hank's buffered saline solution HCELL hematopoietic cell E/L-selectin ligand HD high density HEC high endothelial cell HEC-GlcNAc6ST high endothelial cell N-acetylglucosamine-6-O-sulfotransferase HEV high endothelial venule hIG human immunoglobulin HPRT hypoxanthine phosphoribosyltransferase HSC hematopoietic stem cell ICAM intracellular adhesion molecule IFN-y interferon y IL interleukin IP10 interferon-inducible protein 10 ISP intermediate single positive I-TAC interferon-inducible T cell a-chemoattractant JAM junctional adhesion molecule LAD leukocyte adhesion deficiency LD low density Leu leucine Le x Lewis X structure Lin lineage LMMP lymphoid primed multipotent progenitor LPS lipopolysaccharide LT-HSC long-term hematopoietic stem cell M-6-P mannose-6-phosphate mAb monoclonal antibody MAdCAM-1 mucosal vascular addressin cell adhesion molecule Man mannose x MCP monocyte chemoattractant protein MDC macrophage-derived chemokine MEC medullary epithelial cells MEC mucosae-associated epithelial chemokine MHC major histocompatbility complex MIG monokine-induced by y interferon MIP-1 macrophage inflammatory protein 1 MkEP megakaryocyte/erythroid progenitor MMP multipotent progenitors MW molecular mass NK natural killer cell O-Fuc-1 O-fucosyltransferase-1 OST oligosaccharyltransferase PBS phosphate buffered saline PE phycoerythrin PEC AM-1 platelet-endothelial-cell adhesion molecule 1 PE-Cy5 phycoerythrin-Cy5 Phe phenylalanine PKC protein kinase C PNAd peripheral node addressin ppGalNAcT polypeptide ^ -acetylgalactosamine transferase PSGL-1 P-selectin glycoprotein ligand 1 RANTES regulated on activation normal T cell expressed and secreted REST relative expression software tool RNA ribonucleic acid RPMI Roswell Park Memorial Institute (cell culture media) RT-PCR reverse transcriptase polymerase chain reaction SA sialic acid SCZ subcapsulary zone SDF-1 stromal cell-derived factor-1 SEM standard error of the mean Ser serine SLC secondary lymphoid tissue chemokine sLex sialyl Lewis X structure SP single positive SPF sterile pathogen free ST3GalT oc2,3 sialyltransferase STAT signal transducer and activation of transcription ST-HSC short-term hematopoietic stem cell TARC thymus- and activation-regulated chemokine Tel T cytotoxic 1 cell T c2 T cytotoxic 2 cell TCR T cell receptor TEC thymic epithelial cell TECK thymus-expressed chemokine tg transgenic TGF-p transforming growth factor (3 TGN trans-Golgi network TH1 T helper 1 cell TH2 T helper 2 cell Thr threonine TNF-a tumor necrosis factor a TPST tyrosylprotein sulphotransferase Treg T regulatory cells UTR untranslated region VCAM-1 vascular cell adhesion molecule 1 WT wild-type Xyl xylose Symbol and text nomenclature for representation of glycan structure: from the Nomenclature Committee Consortium for Functional Glycomics | | GalNAc (N-acetylgalactosamine) O G a ' (galactose) H GlcNAc (N-acetylglucosamine) <0> SA (sialic acid) A Fuc (fucose) ACKNOWLEDGMENTS To the people who have filled my life over the past six years, it is difficult to find words that express my gratitude. First and foremost, I am indebted to my supervisor, Ff. Ziltener, whose faith in me has nurtured my growth as a scientist and a human being. Your exuberant personality, intellect, positive attitude and patience have been constant sources of strength. You taught me not to dwell on my failures, but to "build character" and move on. I hope that one day I will be able to inspire someone as you have inspired me. I would also like to express my thanks to: D. Carlow, for your wisdom, and for teaching me the ropes, keeping me on my toes, compelling me to ask the difficult questions and delve deeply for the answers. M . Williams, for your encouragement, and always offering guidance and positive reinforcement. Both of you taught me so much, and provided the solid foundation that I will continue to build from. S. Corbel, from day 1, for being unwaveringly approachable and supportive. You made even the most stressful and hectic times comfortable and manageable - and I knew I could always count on you to strike the vein. J. Zuccolo, for your time, effort and dedication to our project. F. Rossi, for your enthusiasm and your brainpower. W. Seo, K. Gossens, for all of your support, motivation and advice. My committee, for your sound counsel and criticism. P. Johnson, for being a remarkable role model and mentor; K. McNagny, for standing by at every turn, and providing me with a wealth of knowledge; and D. Waterfield, for your time and guidance. J. Wong, for your friendship, your ears, your shoulders and the coffee breaks. A friend like you comes along only rarely. J. Nielsen, S. Chelliah, E. Drew and L. Stratton - whether we were jogging in preparation for the Sun Run, mulling over our failed experiments or celebrating the successes - you all helped me pass the many hurdles that have brought me to this point. The BRC's resident superwoman (and genuine friend) - N . Voglmaier - for making my days so much easier. Your hard work, with the help of L. Rollins, truly did alleviate the stress of endless days (and nights). If only every lab was as lucky as we are. L. So, for your skill and dedication, and L. Y i , for your capable hands - we couldn't have done any of this without the parbiotic mice. L. Wong and S. Hoium, from the Mouse House, for all of your help - your work is valued immensely. A. Johnson, from the FACS facility, for your one-of-a-kind expertise and xvii positive attitude. G. Gill, S. Chand, and M. Bulic: your smiles, wisdom, support and dedication helped get me through and are the backbone of the BRC! P. Carew, for so diligently answering all of my queries, and the rest of the Department of Experimental Medicine, including N. Wong. R. Stokes and A. Randhawa, for your support. My family, for their steadfast support, faith, prayers and love. Dad, thanks for passing on the (genetic) smarts that helped me make it this far and for encouraging me to keep going. Mom, I am everything I am because of your devotion and love. You are the most remarkable woman I know, and it is your reflection that I will always strive to emulate. To my sister Mandy, whose artistic abilities livened up the pages of this thesis, you are an inspiration and wise beyond your years. And to my sister Daliah, whose sharp eyes read every word, you have always been my best friend and utmost supporter, near or far. With every step I take, I thank God for the blessings He has brought to my life, and the courage He evokes in me that allows me to move past each obstacle. This research was made possible by generous grants from the CIHR to H. Ziltener's lab. I was also personally supported by scholarships from UGF and CIHR Transplantation. Finally, I would like to extend a special thank you to J. Schrader, and the entire BRC faculty, past and present, for nourishing the BRC's unique, open and friendly culture; any scientist would be lucky to work in such a fine establishment. I will always be proud of my roots. COAUTHORSHIP STATEMENT Chapter 3 J. Zuccolo provided the RT-PCR data showing that C2GlcNAcT-III is the mRNA expressed by activated T cells. The data in Figure 3-9 were a result of his work. M. J. Williams provided the Western blot data, which show that CD43 is modified by an alternate C2GlcNAcT enzyme in the absence of C2GlcNAcT-I. The data in Figure 3-6 were a result of his work. K. Gossens provided the RT-PCR data showing that C2GlcNAcT-III is expressed at higher levels in activated CD8 cells than CD4 cells. The data in Figure 3-11A are a result of his work. Dr. D. A. Carlow provided invaluable technical and scientific advice through the course of the work summarized in CHAPTER 3, especially in the case of the in vivo HY experimental data. Chapter 5 The work in this chapter was completed as part of a collaborative effort between Dr. Ziltener's lab (Dr. H.J. Ziltener, Dr. D.A. Carlow, K. Gossens and myself) and Dr. Rossi's lab (Dr. F.M. Rossi, Dr. S.Y. Corbel, J Duenas, L. So and L. Yi). The original idea was one that both Dr. Ziltener and Dr. Rossi developed. The parabiotic mice were created by L. Y i . The subsequent analysis of the bone marrow was completed by Dr. S. Y. Corbel and L. So, while analysis of the thymus, spleen and lymph node T cell populations (Figure 5-2) was conducted in the Ziltener lab by myself. Data that are not shown in this thesis but are referred to in the discussion were supplied by Dr. D. A. Carlow (short-term homing studies), K. Gossens (thymic niche occupancy modulates P-selectin mRNA expression), Dr. S. Y. Corbel (PSGL-1 expression on progenitor cells; reduced numbers of ETPs and increased progenitor niche availability in PSLG-1™11 thymi) and J. Duenas (P-selectin expression on thymic endothelium). Appendix II The preliminary data (Figure A-3) in this section were completed in collaboration with Dr. P. Johnson's lab. Dr. P. Johnson provided the CD44n u 1 1 mice for these experiments, along with rationale and scientific advice for these studies. xix CHAPTER 1 INTRODUCTION 1.1 Overview Inflammatory responses trigger the production of many cytokines of which TNF-a, IL-1(3 and IFN-yplay an integral role. These cytokines initiate a cascade of events, including: (i) activating and differentiating dendritic cells (DC), which then migrate from the inflammatory site to secondary lymphoid tissues, where they present antigen to naive T cells; (ii) production of chemokines by activated endothelial and parenchymal cells; and (iii) the upregulation of endothelial adhesion molecules, including selectins. Activation of naive T cells by these migrating DCs leads to the proliferation of naive T cells into T cell clones and differentiation into subsets of effector cells (Mosmann and Sad, 1996). Changes in homing receptor expression accompany this differentiation. While naive T cells recirculate between lymphoid organs, fractions of effector and memory T cells gain access to peripheral inflamed sites (von Andrian and Mackay, 2000). In contrast with naive T cells, these effector cells express receptors for inflammatory chemokines and ligands for E- and P-selectin, although they lack L-selectin expression (Ley, 2003). Selectin ligand formation is regulated through T-cell-activation-induced expression of a series of glycosyltransferases that add branches to extend O-linked carbohydrates on specific substrates. This highly orchestrated process prompts T cells to infiltrate the inflammatory site and interact with antigen-bearing parenchymal cells and other leukocytes, such as macrophages and neutrophils (TH I mediated response), or eosinophils, mast cells and basophils (TH2 mediated response). Figure 1-1 summarizes the migratory routes of T cells. 1 Figure 1-1: Migratory routes of T cells. Naive T cells, constitutively expressing L-selectin, recognize PNAds on the HEV of the lymph node, and are able to continuously home to lymph nodes from the blood. Lymph fluid transported from peripheral tissues carrying antigen-loaded dendritic cells flows through the lymph nodes. Dendritic cells stimulate antigen-specific T cells, which then proliferate by clonal expansion and differentiate into effector cells, expressing receptors that enable them to migrate to sites of inflammation. The traffic signals that direct the activated T cells to peripheral tissues are organ-specific, modulated by inflammatory mediators and are distinct for different subgroups of T cells. For example, T cells that home to the skin typically express CLA on N-glycans of PSGL-1 and are recognized by E-selectin expressed on the skin endothelium, whereas gut homing T cells TRI and TH2 or Tel and Tc2 cells express different co-receptors and respond to different chemoattractants. Other effector cells orchestrate humoral responses by contacting activated B cells in lymphoid organs. Most effector cells die after antigen is cleared, but a few antigen-experienced memory cells remain for long-term protection. Different subgroups of memory cells stand guard in lymphoid organs and patrol peripheral tissues to mount rapid responses whenever the antigen returns. 2 1.2 Leukocyte trafficking Nature has developed an extremely efficient mechanism to deliver leukocytes to the sites of action in inflammation and injury. An interesting analogy exemplifying leukocyte movement is offered by Ehrhardt et al. (Ehrhardt et al., 2004) in a recent review, which relates, "Comparable to a human being plunged into a roaring river, a leukocyte is exposed to high shears within the mainstream of the blood." In either scenario, stopping and exiting the flow becomes an extremely complex task for human or leukocyte. Nonetheless, the emigration of leukocytes from the circulation is a sophisticated and co-ordinated interplay controlled by multiple signalling and adhesion molecules, in particular selectins, chemoattractants and integrins (Butcher and Picker, 1996; Springer, 1994; Vestweber and Blanks, 1999; Worthylake and Burridge, 2001). An initial and essential event in leukocyte recruitment involves low-affinity adhesion, mediated by the selectins and their leukocyte and endothelial counter-receptors (Lowe, 2002; Vestweber and Blanks, 1999). These low-affinity adhesive interactions between selectins and their counter-receptors initially tether the leukocyte to the vessel wall and, in the context of vascular shear flow, cause the tethered leukocyte to roll along the endothelial cells that line the inside of the vessel. Selectin-dependent tethering and rolling slows down and brings the leukocyte into close physical proximity to the vessel wall. This process prompts leukocyte activation via chemokines present on glycosaminoglycans (GAGs) on the endothelium, the vessel wall and perivascular structures (Kunkel and Butcher, 2002; Rossi and Zlotnik, 2000; Sallusto et al, 2000), all of which engage leukocyte-borne chemokine receptors. Chemokine-dependent leukocyte activation leads to the enhanced expression of integrin family members and functional activation of these molecules via inside-out signal 3 Figure 1-2: T cell trafficking and the players involved. Once T cells are activated at the lymph nodes, they find their way to peripheral tissues and become tethered to endothelial cells, rolling slowly down stream. The arrow at the top of the diagram signifies the direction of laminar flow in the blood vessels. The traffic signals that direct effector cells to peripheral tissues are organ-specific. The molecules expressed on the inflamed endothelium are all crucial to the recruitment process. Selectins (for example, P-/E-selectin expressed on the endothelium and L-selectin (not shown) expressed on leukocytes) mediate tethering and rolling of the leukocytes through interactions with carbohydrate selectin ligands expressed on the circulating leukocytes (such as, P-selectin glycoprotein ligand 1, PSGL-1). Rolling leukocytes respond to chemoattractants displayed on endothelial cells because they express chemokine receptors that transmit intracellular signals through G proteins. Once they are slowed down, the chemokine receptors on the cells are able to bind to chemokines both immobilized on the endothelium and chemokines secreted into the vessel lumen. The activating signal induces instantaneous activation of (32 and ct4 integrins (LFA-1 and VLA-4). Activated integrins bind with high affinity to endothelial ligands (ICAM-1 and VCAM-1), causing the lymphocyte to arrest. The a4 integrins can mediate activation-independent rolling interactions, as well as arrest rolling leukocytes (not shown). Leukocytes, then, can migrate through the endothelial cells with the help of JAMs, PECAM-1 and CD99. Chemokines provide directional cues for leukocytes by forming gradients that migrating cells can sense, leading them to the stimulus. The migrating cells undergo a profound transformation, resulting in a redistribution of chemokine receptors, integrins and cytoskeletal proteins. 4 transduction events (Cinamon et al., 2001; Harris etal., 2000; van Kooyk and Figdor, 2000; Woodside et al., 2001). Adhesive engagements between the activated leukocyte integrins and their endothelial cell counter-receptors lead to firm adhesion of the leukocyte to the vessel wall, arrest of rolling, and, ultimately, the process of transmigration (Weber, 2003) to reach the extravascular space. This process is outlined in Figure 1-2. The Players Selectins capture flowing leukocytes and mediate tethering and rolling Selectins mediate functions unique to the vasculature, the tethering of flowing leukocytes to the vessel wall and the formation of labile adhesions with the endothelium that permit leukocytes subsequently to roll in the direction of flow. Selectins can mediate tethering of a flowing cell within milliseconds. Selectin-ligand interactions have rapid association and dissociation rate constants, where selectin-ligand bonds are formed and cause tethering of the leukocyte to the vessel wall, leading to intermittent rolling. A rapid dissociation rate ensures that even with multiple selectin-ligand bonds it will not take long before the bond formed most upstream randomly dissociates and allows the cell to roll forward a small distance until the next bond is formed (Lawrence and Springer, 1991; Springer, 1994). Selectins have an elongated structure that is believed to enhance their flexibility and ability to bind their ligands on flowing cells (Springer, 1994). The rolling process is considered absolutely essential for adhesion to occur (Lawrence and Springer, 1991) and the involvement of selectins in this process is clear. However, selectin-independent adhesion in some tissues is also documented (Kubes, 2002). Tissues like the mesentery (Mayadas et al., 1993), cremaster (Bullard et al., 1996) and skin (Hickey et al., 1999) require selectin-mediated processes to mediate rolling due to inherent vessel characteristics. Alternatively, liver (Wong et al., 1997), lung (Mizgerd, 2002) and 5 heart (Kubes, 2002) tissues have less need for selectins. Also, many studies suggest integrins actually help or replace selectins. a4 integrins have been shown to mediate the rolling of lymphocytes and eosinophils on vascular endothelial cells in vitro and in postcapillary venules in vivo (Alon et al., 1995; Berlin et al., 1995; Grabovsky et al., 2000; Patel, 1998). Monocytes also use oc4 integrins (Walter and Issekutz, 1997). On the other hand, neutrophils appear to be mostly dependent on selectins for rolling in acute inflammatory models (Alon et al , 1995; Johnston et al., 1997) but less so in chronic or systemic inflammatory models, where aA integrins play a more-pronounced role (Burns et al., 2001; Issekutz et al., 1996; Johnston et al., 1996). Reports by Issekutz et al. define VLA-4 as a predominant E-/P-selectin-independent mechanism for migration of resting T cells to skin inflammation (Issekutz and Issekutz, 2002), in a similar manner as migration to inflamed brain (Bartholdy et al., 2000; Engelhardt et al., 1997) and lung (Wolber et al., 1998) models. The balance of VLA-4 versus selectin-mediated mechanisms varies in tissue type and stimulus type, even within the same tissue; this may be related to differences in the expression of E- and P-selectins, as well as VCAM-1 resulting from the stimulus (Norman et al., 2003). Chemokines guide migration in a highly specific manner Chemokines form a complex superfamily of small (6-14 kDa) secreted proteins. The specific effects of chemokines are mediated by seven-transmembrane-spanning, G-protein-coupled receptors expressed on a number of different cell types, including lymphocytes. Aside from directing migration, these receptors activate integrin adhesiveness and stimulate degranulation, shape change, actin polymerization and respiratory burst (Springer, 1994). Chemokines are a major product of various leukocyte populations upon activation. After activation, therefore, IL-8, MIP-loc, MIP-lp, lymphotactin and RANTES are, according to studies, upregulated at either the mRNA or protein level, and chemokines are among the 6 most widely and abundantly expressed activation-inducible genes in lymphocytes and other cell types (Ward and Westwick, 1998). The complex migratory pathways of leukocytes are tightly controlled by changes in chemokine receptor expression and exposure to chemokines, in addition to the state of cell activation. Chemokines are important in activating integrin adhesiveness and directing leukocyte migration along the endothelium, as well as through it. Leukocytes, sensing a concentration difference, move in the direction of increasing chemokine concentration, followed by directional cell locomotion via cytoskeletal rearrangements and adhesive interactions with the extracellular matrix (Sanchez-Madrid and del Pozo, 1999). Chemokines presented on endothelial cells trigger integrin activation and the arrest of leukocytes that carry the corresponding receptors (Butcher et al., 1999; Campbell and Butcher, 2000). The chemokines are divided into the C, CC, CXC and C X 3 C subfamilies, defined according to NH2-terminal cysteine-motifs (Sallusto et al., 2000; Zlotnik and Yoshie, 2000). More recent classifications divide chemokines based on the cellular distribution of their chemokine receptors - a process that distinguishes inflammatory-inducible chemokines from homeostatic-constitutive chemokines (Cyster, 1999; Moser and Loetscher, 2001; Sallusto et al., 2000; von Andrian and Mackay, 2000). Inflammatory chemokines are expressed in inflamed tissues (stimulated by proinflammatory cytokines or pathogens) by resident and infiltrated cells, and are specialized for recruiting effector cells, such as activated CD8 + T cells. Homeostatic chemokines are produced in discrete microenvironments within lymphoid or non-lymphoid tissues and are involved in maintaining physiological traffic. One major function of chemokines involves controlling cell trafficking during lymphopoiesis, where a number of bone marrow and thymus-resident chemokines (TECK (CCL25), MDC(CCL22)/TARC(CCL17), SDF-1(CXCL12) and SLC(CCL21)/ELC(CCL19) 7 recognized by the chemokines receptors (CCR9, CCR4, CXCR4, and CCR7, respectively) help guide precursor cells to specific microanatomical sites for differentiation, including thymocyte development (Bleul and Boehm, 2000; Campbell et a l , 1999; Kim et al., 1998). Another function of chemokines arises in antigen recognition within secondary lymphoid organs, where chemokines present in the lymph nodes, Peyer's patches and spleen take part in cell recruitment and localization within these tissues, as well as in the differentiation of naive cells into memory and effector cells. Studies support that SLC and ELC binding to CCR7 are involved in cell recruitment and localization within these secondary lymphoid organs. CCR7 helps discriminate between secondary lymphoid and peripheral-tissue homing T cells (Sallusto et al., 2000). Naive T cells (and central-memory T cells) express CCR7 and L-selectin, allowing them to home to lymph nodes, whereas effector memory cells do not express CCR7 and, as a result, do not home to lymph nodes, instead circulating through peripheral tissue looking for antigen (Sallusto et al., 2000). DC-CK1(CCL18), MDC and TARC are also expressed in secondary lymphoid tissues presumably produced by activated mature DCs that have migrated from peripheral tissues (Figure 1-1). DC-CK1 helps attract naive T cells to the DCs so that the naive T cell (Adema et al., 1997) can sample the antigen presented by the DC, whereas MDC and TARC help to promote contact between CCR4-expressing, antigen-receptor-triggered T cells and DCs during immune activation (Tang and Cyster, 1999). These effector cells are then guided to inflammatory sites where chemokines (RANTES(CCL5), MCP-2(CCL8)/-3(CCL7)/-4(CCL13) and MEC(CCL28), eotaxins (CCL11, CCL24, CCL26), MDC/TARC, 1309(CCL1), MIP-la(CCL3)/-l(3(CCL4), I-TAC(CXCL11), MIG(CXCL9) and IP 10 (CXCL10)) have been upregulated in response to pathogen entry and tissue injury (Sallusto et al , 2000). The recruitment of effector cells is 8 specific in that TH1/TC1 and TH2/TC2 cells are differentially recruited depending on the chemokines present. For example, T H I and Tel, like effector memory cells, express CCR5 (recognizes the chemokines: RANTES, MIP-la/-lp and MCP-2) and CXCR3 (recognizes the chemokines: I-TAC, MIG and IP 10), which recognize different chemokines than TH2 and Tc2 cells, that express CCR3 (recognizes the eotaxins along with RANTES and MCP-2), CCR4 (recognizes MDC and TARC) and CCR8 (recognizes 1309) (Moser and Loetscher, 2001). Integrins mediate firm adhesion Integrins are a set of cell-surface adhesion molecules that regulate cell-cell and cell-matrix protein interactions. Expressed on circulating leukocytes, they comprise two subunits, a and p, whose association, in addition to divalent cations, is essential for ligand binding. Integrins are divided into subgroups based on their p subunit: pi-8. Those involved in lymphocyte homing and leukocyte migration are aifii (LFA-1, CD1 la), ctMp2 (Mac-1, CD1 lb), ccxp2 (CD1 lc), ct4pi (VLA-4) and c<4p7 (LPAM-1) (Springer, 1994). In addition to their function in mediating rolling in some tissues, integrins are involved in mediating lymphocyte arrest following successive, chemokine-induced signals. The control of integrin function occurs via regulatory signals that originate within the cell cytoplasm and are transmitted to the external ligand-binding domain of the receptor. Generating and transmitting these "inside-out" signals leads to increased adhesiveness of the integrin for its endothelial receptor, resulting from conformational changes that increase affinity and cytoskeletal changes (Kim et al., 2003). The overall result is firm adhesion. A recent study demonstrated that chemokine-triggered lymphocyte adhesion involves a previously unrecognized step (Shamri et a l , 2005). Shamri et al. suggest that signals from endothelial presented chemokines, not soluble chemokines, trigger an'extended LFA-1 9 conformation, which must immediately rearrange with endothelial ICAM-1 to achieve full and productive LFA-1 activation by stabilizing LFA-1 in a proper cytoskeletally anchored state (Shamri et al., 2005; Shimaoka et al., 2003). Integrin ligands are members of the Ig superfamily of cell-cell adhesion molecules, and include ICAM-1, ICAM-2, ICAM-3, VCAM-1 and MAdCAM-1. ICAM-1, -2 and -3 all bind LFA-1, and ICAM-1 can also bind Mac-1. VCAM-1 binds to VLA-4 and c<4pV MAdCAM-1 binds oufty and L-selectin. Under conditions of inflammation, the endothelial cell-surface expression of several members of this family is enhanced. Modulation of expression of these molecules occurs with some, but not all, mediators of inflammation and serves to regulate the type of leukocyte that is recruited to the site of inflamed tissue, as well as the temporal pattern of recruitment of leukocytes. For example, studies have shown that ICAM-1 and -2 act in a redundant way in the recirculation of lymphocytes through lymph nodes, but only ICAM-1 is found to be involved in migration of lymphocytes into inflamed tissue (Lehmann et al., 2003). ICAM-2 is constitutively expressed on vascular endothelium, whereas ICAM-1 and VCAM-1 expression increases in response to inflammatory cytokines and such agents as LPS (Springer, 1994; Springer, 1995). Transendothelial migration and PECAM, CD99, VE-cadherin, and JAMs As leukocytes move toward the borders of endothelial cells, they prepare themselves to squeeze between tightly apposed cells into the tissue, with minimal disruption to the vascular lining. In diapediesis, a rapid process, the leukocyte extends itself by a pseudopod across the endothelial border. To accomplish this, it disassembles its cytoskeleton on the apical surface and reassembles it on the abluminal side of the endothelium. Unique molecular mechanisms govern this step in leukocyte migration, namely: Platelet-endothelial-cell adhesion molecule-1 (PECAM-1, CD31), CD99, junctional adhesion molecule (JAM)-A, 10 JAM-B and JAM-C, and VE-cadherin. Leukocytes and the endothelium express these molecules, which are involved in homophilic interactions between a molecule on the leukocyte and the same molecule on the endothelium at the junctions. PECAM-1, a member of the Ig gene superfamily expressed on the surface of most leukocytes, concentrates at the borders of endothelial cells. Homophilic interaction of the amino-terminal portion of leukocyte PECAM-1 with endothelial cell PECAM-1 is required for diapedesis since blocking antibodies reduce transendothelial migration both in vitro and in vivo by up to 90% i (Liao et al., 1997; Muller, 2003). CD99 is a small, unique molecule that is highly O-linked, glycosylated and, like PECAM-1, functions in a homophilic manner to control transendothelial migration (Muller, 2003). Blocking antibodies to CD99 block diapedesis by more than 90% suggesting that CD99 controls a step in diapedesis that is different from PECAM-1. Blocking both CD99 and PECAM-1 abolishes diapedesis completely (Schenkel et al., 2002). Diapedesis can therefore be subdivided into at least two steps carried out sequentially by PECAM-1 and CD99. The exact role of the other endothelial adhesion molecules is still under study. VE-cadherin engages in a Ca2+-dependent homophilic adhesion and is believed to act as a "gate-keeper" to migrating cells through the adherens junction. JAM-A, expressed at epithelial tight junctions and intercellular borders of endothelial cells, is shown to engage in both homophilic interactions as well as heterophilic interactions with LFA-1. JAM-B, expressed on HEV, and JAM-C, expressed on naive T cells, while not formally implicated in leukocyte-endothelial interactions, are speculated to have a role in homing to lymph nodes via a heterophilic interaction (Muller, 2003). Other molecules, too, are implicated in transendothelial migration. The highly abundant cell-surface molecule CD43 is shown to have a role in leukocyte homing and transmigration (McEvoy et al., 1997a; McEvoy et al., 1997b; Stockton et al., 1998; Woodman et al., 1998) of T cells and neutrophils. Studies suggest that CD43 functions to 11 inhibit leukocyte-endothelial cell interactions by limiting rolling and adhesion within the microvasculature due to its highly antiadhesive properties. However, once the leukocyte is ready to transmigrate, CD43 may function to enhance emigration out of the vasculature into the tissues, by reducing leukocyte endothelial interactions (Woodman et al., 1998). Recently, CD44 expressed on neutrophils was implicated as an important mechanism for neutrophil transmigration (Khan et al , 2004). CD44 is a type I transmembrane glycoprotein expressed on a wide variety of cell types, including myeloid and endothelial cells (Johnson et al., 2000), that binds to hyaluronic acid (HA) among other ligands. Khan et al. exhibited that neutrophil adhesion and emigration depend upon CD44 expressed on both the leukocyte and the endothelium, in addition to HA, which is consistent with the sandwich model of a CD44-HA-CD44 interaction proposed previously for T cell extravasation (Mikecz et al., 1995). 1.3 Glycans in biology Glycans, or carbohydrates, are the most-abundant and structurally diverse biopolymers formed in nature comprising a major fraction of the outer cell surface. Numerous pathways and enzymatic activities involved in glycan biosynthesis generate this diversity in the secretory pathway. Glycans are attached to lipids or proteins that are believed to play an important role in mediating specific recognition or modulation of biological processes (Lowe and Marth, 2003; Varki, 1993). Glycans attach to proteins in multiple ways. There are six major classes of mammalian glycans. The two main groups include the N -glycans and the O-glycans. Other recognized glycans include glycolipids, glycosaminoglycans (GAGs), GPI-anchored glycans and hyaluronan (Lowe and Marth, 2003). In N-glycans, N-acetylglucosamine (GlcNAc) is linked to the amide group of asparagine and in O-glycans, carbohydrate attaches to hydroxyl groups of serine and threonine residues. O-glycans are further divided into subgroups depending on the nature of 12 the amino acid residue and sugar group involved in the carbohydrate-protein linkage, i) mucin-type O-glycoproteins, where N-acetylgalactosamine (GalNAc) is linked to serine or threonine (Shimizu and Shaw, 1993); ii) intracellular glycoproteins, with GlcNAc linked to serine or threonine (Hart et al., 1996); iii) glycoproteins with xylose (xyl) linked to serine or threonine found in GAGs (Fukuda, 1994); iv) fucose (Fuc)-initiated O-linked chains on epidermal- growth-factor-like repeat domains (EGF repeats) in proteins like Notch (Lowe, 2005; Wang et al., 2001a; Wang et al., 2001b); v) glucose (Glu) modified serine and threonine residue on some proteins, including EGF repeats on Notch (Shao et al., 2002); and vi) O-mannose (Man)-linked glycans in muscle tissue (Chiba et al., 1997; Strahl-Bolsinger et a l , 1999). 1.3.1 N-glycans In mature glycoproteins, N-linked glycosylation is structurally diverse varying in the number and size of branches among cell types, tissues and species (Paulson, 1989). But when they are first added in the ER to asparagine residues within the Asn-X-Ser/Thr consensus sequence on growing polypeptides, the N-glycans are homogeneous and relatively simple. Processing of the N-linked oligosaccharide moieties starts while the proteins are still in the ER, and continues after they arrive in the Golgi apparatus. The processing in the ER introduces only limited diversity that is shared with all glycoproteins but once the proteins arrive in the Golgi, structural diversity is generated. In the early secretory pathway, the N-glycans have a common role in promoting protein folding, quality control, and certain sorting events. Later in the Golgi, enzymes introduce much diversity which prepares the mature protein for more novel and diverse functions (Varki etal., 1999). 13 During the synthesis of N-linked glycans in mammalian cells, a 14-saccharide core (Glc3Man9GlcNAc2) unit is assembled as a membrane bound dolichylpyrophosphate precursor by a series of enzymes, beginning with the enzyme UDP-GlcNAc: dolichol phosphate GlcNAc-1-phosphate transferase (GlcNAc-PT). GlcNAc-PT is essential to N -glycan assembly and plays a critical role in protein folding of many membrane-associated and secreted polypeptides. Tunicamycin, a GlcNAc analogue that competitively inhibits of GlcNAc-PT function, leads to severe cellular disfunction and death, suggesting that N -glycans are required for cell viability (Lowe and Marth, 2003) due to their role in early processing and protein folding within the ER. The completed core oligosaccharide is then transferred from the dolichylpyrophosphate carrier to a growing polypeptide chain and is coupled through an N-glycosidic bond to the side chain of an asparagine residue by an oligosaccharyltransferase (OST) (Varki et al., 1999). After coupling occurs, glucosidases (Glcase I and II) quickly remove the two terminal glucose residues. Removal of the third glucose, however, is associated with proper glycoprotein folding and contributes to the ER retention time of that glycoprotein. Folding is monitored by a glucosyltransferase as well as a molecular chaperone ~ calnexin. This glucosyltransferase acts as a sensor that detects improperly folded proteins by adding a terminal glucose residue. Calnexin, a lectin, binds preferentially to this terminal glucose linked to the high-mannose moiety of N-glycans and helps retain the glycoprotein in the ER until it is properly folded. This is one of the most important functions of N-glycans (Hauri et al., 2000a; Helenius and Aebi, 2001; Paulson, 1989). Following glucose trimming of the properly folded glycoprotein, N-glycans become available for glycosidase reactions in the ER and Golgi. ER and Golgi (cis) 14 mannosidases remove mannose residues. Further along in the Golgi, the glycan chains undergo further trimming of mannoses and, in many cases, new sugars are added to produce complex glycans and some high-mannose glycans that have escaped terminal glycosylation. The structures undergo further diversification with the action of GlcNAcTs and mannosidases. GlcNAcT-I adds GlcNAc to the high-mannose structure (Man5GlcNAc2-Asn) and a-mannosidase II removes two mannose residues in the medial Golgi, resulting in a "hybrid" N-glycan (GlcNAciMan 3GlcNAc2-Asn). "Complex" N -glycans are then produced with the activity of other GlcNAcTs (GlcNAcT-II, III, IV, V and VI) that add on GlcNAc to mannose residues, as well as fiicosyltransferases that add fucose to the core GlcNAc. As N-glycans transit through the medial- and trans- Golgi, they become substrates for glycosyltransferases localized toward the end of the assembly line, such as sialyltransferases and sulfotransferases, which add increasing diversity to N -glycans. N-glycans have many functions beyond protein folding, which take place along the secretory pathway in the cell. ER-associated protein degradation Trimming of the N-linked glycans also plays a role in the sorting process leading to glycoprotein degradation in the ER. Proteins that fail to reach their native conformation in the ER are selectively eliminated by ER-associated degradation (ERAD). ERAD has a central clearance function in the cell: it gets rid of misfolded and mutant proteins, and unused subunits of oligomers, by targeting them to the cytosol for ubiquitination and degradation. When trimming N-glycans by ER-mannosidase-I is prevented, degradation of glycoproteins essentially stops, suggesting that the structure 15 formed by the removal of a single mannose residue (Glco-3MangGlcNAc2-Asn) is recognized by ERAD (Helenius and Aebi, 2001). Transport and targeting ERGIC-53 and VIP36 are homologous, mannose-specific lectins in the Golgi complex and early secrectory pathway (Arar et al., 1995; Hara-Kuge et al., 2004; Hauri et al., 2000a; Hauri et al., 2000b). They serve as cargo (correctly folded proteins) capture and transport receptors for ER-to-Golgi traffic (to the ER-Golgi intermediate compartment, or ERGIC) of glycoproteins, binding to mannose residues in N-glycans (Arar et al., 1995; Hara-Kuge et al , 2004; Hauri et al., 2000a; Hauri et al., 2000b). Mannose-6-phosphate and targeting of lysosomal hydrolases The selective targeting of lysosomal hydrolases from the trans-Golgi to endosomes and lysosomes also uses lectins. In the cis-Golgi a large, multisubunit enzyme, UDP-GlcNAc:lysosomal enzyme GlcNAc-1-phosphotransferase, recognizes lysosomal enzymes based on critically spaced lysine residues in the polypeptide (Nishikawa et al., 1999) and adds GlcNAc-phosphate to terminal or subterminal mannose residues. This reaction is followed by the removal of the GlcN Ac, leaving a mannose-6-phosphate (M-6-P) group that is recognized by M-6-P receptors in the trans-Golgi. M-6-P receptors are part of the P-type lectin family that have a role in sequestering lysosomal enzymes in the trans-Golgi by associating with one or more of the oligosaccharides, escorting them via clathrin-coated vesicles to endosomes, and returning empty to the trans-Golgi (Helenius and Aebi, 2001). 16 Other intracellular functions of N-linked glycans Resident proteins of the ER, Golgi and lysosomes are themselves often glycoproteins. Their sugar composition reflects their location; early in the pathway they carry mainly high-mannose glycans, whereas in the Golgi complex or in post-Golgi organelles they contain complex sugars. Some highly glycosylated membrane proteins in the lysosome, LAMP-1 and LAMP-2 are highly N-linked glycosylated. The glycans have been shown to protect the LAMPs against degradation by lysosomal proteases (Kundra andKornfeld, 1999). Extracellular functions of N-linked glycans It is apparent that N-linked glycans allow the ER and the Golgi to keep track of information, such as protein-folding status, time spent in the ER and their final destination (Helenius and Aebi, 2001). N-glycans also have an important role in cell-cell interactions. Hematopoiesis, immune function and inflammatory responses are frequently affected in animals lacking specific extracellular N-glycan linkages, this may reflect the essential role for N-glycans in regulating specific selectin ligand formation (in E-selectin ligands for example) as will be alluded to in later sections. 1.3.2 Glycolipids All cellular membranes contain glycerophospholipids, cholesterol and sphingolipids. Sphingolipids are based on ceramide and have either a phosphocholine head group (sphingomyelin) or a carbohydrate structure (glycosphingolipids). They make up a large portion of the plasma membrane in mammals embedded in the membrane, anchoring the extracellular glycan component of the glycolipid to the cell (Varki et al., 1999). Glycolipids are formed from either the addition of glucose to ceramide by ceramide glucsoyltransferase or the addition of galactose to ceramide by ceramide 17 galactosyltransferase. Glucosylceramide comprises the core of most glycolipids in mammals while galactosylceramide is much less abundant. These products can then be modified by numerous other glycosyltransferases to form more complex glycolipids. Glucosylceramide-based glycolipids play an essential role in development and in cellular and tissue function, while the role of galactosylceramide-based glycolipids is more subtle (Lowe and Marth, 2003). Glycolipids have an organizing role in the cell membrane. Glucosylceramides and glycophospholipid anchors have long acyl chains which allow them to extend into the phospholipid bilayer of cell membranes and associate with cholesterol to form lipid "rafts." According to the lipid raft model, the plasma membrane is composed of different lipid compositions on the two sides of the membrane (or leaflets). The outer leaflet of the plasma membrane contains microdomains of cholesterol and sphingolipid-rich rafts that are connected to cholesterol-rich domains in the inner leaflet. This arrangement is important for co-clustering of GPI-anchored proteins and signalling components (Munro, 2003). 1.3.3 Glycophosphatidylinositol Linkages A glycolipid with a significant biological role is glycophosphatidylinositol (GPI). GPI-linked proteins are anchored to the cell surface by an inositol phosphate attached to a glycan that is, in turn, linked to a phosphatidylinositol-linked fatty acyl chain residing in the plasma membrane. GPI-linked proteins are widely expressed in animal tissues. As with other glycan-containing polymers, the glycan linkages comprising the glycan moiety are elaborated by the sequential actions of specific enzymes, including glycosyltransferases. The biosynthesis of GPI anchors occurs in two major steps: 1) preassembly of donor GPI in the ER membrane; and 2) attachment of the GPI with 18 cleavage of the carboxy-terminal peptide from the newly synthesized protein (Varki et al., 1999). Attachment of the GPI anchor to the polypeptide is a posttranslational modification that requires two peptide signal sequences: i) amino-terminal signal peptide that directs the nascent chain into the ER and ii) carboxy-terminal signal that directs GPI anchor attachment (Varki et al., 1999). Some key functions of GPI-anchors include: 1) Allowing proteins an increased lateral mobility (giving adhesion molecules more freedom in order to interact with ligands); 2) Mediating the release or secretion of proteins by activating lipases (soluble and GPI-linked forms of proteins exist and GPI-linked proteins are rapidly shed from the surface of cells); 3) Targeting protein to apical surfaces (GPI-anchored proteins contain apical sorting signals); 4) Regulating endocytosis (potocytosis-capture and import of scarce extracellular molecules or ions against gradient through caveolae) or protein turnover (extending half-life of cell surface proteins by trapping them in caveolae and releasing them later); and 5) Involvement in signal transduction of receptor-mediated events (antibodies to GPI anchored proteins on T cells mimic T cell activation by inducing cell proliferation, IL-1 and IL-2 production, and other metabolic changes) (Varki et al., 1999). 1.3.4 Glycosaminoglycans Glycosaminoglycans (GAGs), or proteoglycans, contain four distinct types of sulphated, linear oligosaccharide polymers: heparan sulphate, chondroitin sulphate, 19 dermatan sulphate and karatan sulphate. GAGs are implicated in growth-factor binding, presentation and internalization, as well as in cell adhesion and the maintenance of extracellular matrix integrity. Adding xylose, by xylosyltransferases, to specific serines on proteoglycan core proteins initiates GAG synthesis. Xylosyl residues are elongated by the sequential actions of pi,4galactosyltransferase (pT,4GalT-7 or GalTI), pl,3galactosyltransferase (GalTII) and pl,3glucuronosyltransferase (GlcATI) to form the core protein linkage tetrasaccharide common to heparin, heparan sulphate, chondroitin sulphate and dermatan sulphate. Studies also report phosphorylation of this tetrasaccharide. Modification by adding (31,4-linked GalNAc initiates assembly of chondroitin sulphate and dermatan sulphate. Modification of the core tetrasaccharide by addition of al,4-linked GlcNAc initiates assembly of heparan sulphate. This is followed by alternating addition of glucuronic acid (for heparin, chondroitin sulphate, and dermatan sulphate), iduronic acid (for dermatan sulphate), GlcNAc (for heparin, keratin sulphate), Gal (for keratin sulphate) and GalNAc (for chondroitin sulphate and dermatan sulphate) residues, catalyzed by a number of other glycosyltransferases (Varki et al., 1999). 1.3.5 Hyaluronan Hyaluronan (HA) is a high-molecular-weight glycosaminoglycan containing repeating disaccharide GlcNAcp11,4-glucuronic acid-(3l,3. It is found in abundance in the skin, skeletal tissues, eye, umbilical cord and synovial fluid. HA is a ligand for CD44, a cell-adhesion molecule that also participates in signal transduction events. At least three distinct HA synthase genes in human and mouse determine HA synthesis (Hasl, 2, 3). Deleting Has-2 in mice, in contrast with Has-1 and Has-3, leads to loss of HA during embryogenesis, and is associated with embryonic lethality at E9.5 (Varki et al., 1999). 20 HA is one of the main ligands for CD44. Studies have illustrated that HA binding acts as a bridge connecting one CD44 to another CD44. HA binding to CD44 is tightly regulated in a cell-type and activation-state-specific manner. Two forms of CD44 are believed to be present: an inactive conformation state that does not bind HA and an active conformation that does. CD44 functions in embryogenesis, lymphopoiesis, lymphocyte activation, progenitor homing, angiogenesis, wound healing, and leukocyte rolling and extravasation at inflammatory sites (reviewed in (Johnson et al, 2000)). 1.3.6 O-glycans O-linked glycans are involved in a number of biological processes, including leukocyte homing (Somers et al., 2000; Yeh et al., 2001), functioning as ligands for receptors (Hooper and Gordon, 2001) and acting as signals for protein sorting (Alfalah et al., 1999). O-linked glycosylation takes many forms - and those of special interest are outlined below. Mucins Mucins are cell-surface or secreted glycoproteins with large numbers of clustered O-glycans on serine and threonine residues, that typically make up 20% to 30% of the amino acids (Perez-Vilar and Hill, 1999). These carbohydrate modifications account for up to 80% of the weight of mucins and are responsible, along with characteristic, pro line-rich peptides, for providing extended peptide conformations. Many important biological processes - protection of epithelial cell surfaces from chemicals and physical injury, immune response, cell adhesion, inflammation and tumour genesis - appear to be afforded and/or modulated by mucins or glycoproteins containing mucin-like domains (Berger, 1999; Bruckner et al., 2000; Dennis et al., 1999; Fukuda and Tsuboi, 1999; Fukuda and Vliegenthart, 1999; Rudd et al., 2001; Taylor-Papadimitriou et al., 1999; Ten 21 Hagen et al., 2001; Van den Steen et al., 1998). Protein O-glycosylation is important in development as well (Schwientek et al., 2002b; Ten Hagen and Tran, 2002). Initiation of mucin-type O-glycosylation occurs by the posttranslational addition of GalNAc to serine or threonine residues of the polypeptide chain, by the polypeptide GalNAcT (ppGalNAcT) family of enzymes, believed to be active in the Golgi (Bennett et al., 1998). Incorporation of GalNAc gives rise to the simplest known O-glycan structure. No defined amino acid consensus motif for O-glycosylation of serine and threonine has been described. Many studies performed in vitro and in vivo examine the glycosylation of specific peptide sequences to characterize the peptide specificities of ppGalNAcTs. Study results surmise that neighbouring proline, serine and threonine residues enhance the probability of O-glycosylation, but fail to deduce any generalized rules for the process (Brockhausen et al., 1996; Gerken et al., 2002a; Gerken et al., 2002b; Hanisch and Muller, 2000; Hanisch et al., 1999; Hanisch et a l , 2001). Despite the apparent simplicity of the reaction catalyzed by this family, at least 13 (ppGalNAcT-1 through ppGalNAcT-13) family members are functionally characterized in mammals, and analysis of public and private databases suggests there may be as many as 24 isoforms in humans (Ten Hagen et a l , 2003). Differences in substrate specificity, functional redundancy and expression patterns in terms of tissue specificity partially explain the need for a large family (reviewed in (Ten Hagen et al., 2003)). Studies reveal that while some degree of tissue-specific expression of these ppGalNAcT exists, based on real-time PCR studies, there is generally more than one gene expressed in each tissue (Kingsley et al., 2000; Young et al., 2003). To date, genetic ablation in mice of individual isoforms has failed to expose unique functions for isoforms ppGalNAcT-1, T-4, T-5, T-8 and T-13 (Ten Hagen et a l , 2003). However, mutation of one family member imparted a 22 lethal phenotype in Drosophila (Schwientek et al., 2002a; Ten Hagen and Tran, 2002). The overlapping substrate specificities and partially redundant expression patterns of ppGalNAcTs are likely the reason that a phenotype is not detected in gene knockouts. Further monosaccharides, such as Gal, GlcNAc, sialic acid and Fuc, when added sequentially to the GalNAcaSer/Thr, generate many structurally diverse O-linked oligosaccharides (Schachter and Brockhausen, 1989). Four major subtypes of O-glycans are delineated based on the saccharide groups attached to the GalNAc residue (Figure 1-3) (Schachter and Brockhausen, 1989). These are: 1) The core 1 structure is formed by addition of galactose to form Gal pl-3GalNAc-ocSer/Thr catalyzed by core 1 pi-3 galactosyltransferase (Core 1 p3GalT) - the major constituent of O-glyans in many cells; 2) The core 1 structure is a substrate in constructing core 2 structures, and GlcNAc to Gal pl-3GalNAc-aSer/Thr in a p 1-6 linkage is added to form Galpl-3 (GlcNAcpl-6) GalNAc-aSer/Thr. The core 2 P-l,6-A^-acetylglucosaminyltransferase (C2GlcNAcT) controls this reaction (Fukuda, 1991; Higgins et al., 1991; Piller et al., 1988); 3) Core 3 structures, restricted to mucins from specialized tissues, such as the stomach, small intestine and colon, are formed by adding of GlcNAc in a p 1 -3 linkage to GalNAc-aSer/Thr catalyzed by C3GlcNAcT (C3GlcNAcT-6). Core IGalT and C3GlcNAcT, therefore, use the same acceptor substrate; and 4) With the core 3 structure as a substrate, the core 4 structure is catalyzed through the action of C4GlcNAcT. GlcNAc is added to GlcNAc pl-3GalNAc-aSer/Thr in a pi-6 linkage to yield GlcNacpl-3 (GlcNAcpl-6) GalNAc-aSer/Thr. 23 Other modifications to the core GalNAc structure are identified. For example, adding a second GalNAc via a al-3 or a cd-6 linkage results in core 5 and core 7 structures, respectively, while adding GlcNAc in a pi-6 linkage, gives way to the core 6 structure. However, these latter modifications appear to be uncommon (Iwai et a l , 2002; Varki et al., 1999). Core 1 Core 2 Figure 1-3: Major core O-glycan structures in biology. The activities of core 1 |33Gal T and C2GlcNAcTs (I, II and III) form core 1 and core 2 O-glycan structures. Additional biosynthesis can yield O-glycans, bearing fucose and sialic -acid-containing lactosamines. C3GlcNAcTs and C4GlcNAcTs (as well as C2GlcNAcT-II) activities, meanwhile, form core 3 and core 4 structures. Further O-glycan biosynthesis can yield biantennary forms sometimes similar to those of core 2 subtypes. 24 For many glycosyltransferases involved in synthesis of O-glycan core structures, multiple family members are described and the existence of more enzymes is postulated. Tissue-specific or cell-activation-associated expression of glycosyltransferases (Varki, 1998) - as well as altered substrate specificities controlled by polypeptide domains of the substrate glycoproteins (Baenziger, 1994) - may determine biosynthesis of specific glycosyltranserases. Of the four main core O-glycan structures, the core 2 structures constitute the most predominant class. The core 3 and core 4 structures are less common and their expression is mainly associated with the digestive tract's mucin-producing tissue. Given that the core 1 and core 3 structures serve as substrates for core 2 and core 4 enzymes, respectively, core 1 and core 3 structures do not carry core 2 or core 4 modifications. Adding more monosaccharides to core 1 and core 3 structures results in mono- or bi-antennary structures. Commonly, the core 2 and the core 4 branches are elongated with one or multiple lactosamine structures (Galpl-4GlcNac). Adding lactosamine structures, and subsequently modifying them with fucose, sialic acid and sulphation, leads to the formation of sialyl Lewis X structures (sLex) that serve as ligands for selectins (Figure 1-4). 25 • GalNAc O Gal • GlcNAc • SA A Fuc Tri-fucosyl-sLe* 6, 6'-bis-sulfo-sLex 6-sulfo-sLex 6'-sulfo-sLex a3 H sLe x sLea B3 M <x3T B3 M 63 Capping Groups on L-selectin Ligands Capping Groups on E- and P-selectin Ligands Figure 1-4: Capping structures for glycans of selectin ligands. Structures of common sialyl Lewis X (sLex)-based structures present on L-selectin, E-selectin and P-selectin ligands are illustrated. Polylactosamine chains are displayed by O-linked, N -linked or lipid-linked glycoconjugates (R). L-selectin ligands express sulphated (SO4) sLex structures on core 2 extentions or core 1 extentions. E-selectin recognizes fucosylated-sLex and sLea capping that typically occurs on N-glycans as well as lipids. P-selectin recognizes core 2 O-glyans with a single, typical sLex -capping group and core 2 O-glycans extended by polylactosamine on the core 2 branch, modified by three fucosylation sites, and capped with a sLex moiety. O-Fucose The O-fucosylation pathway is defined by the linkage Fuca-Ser/Thr. O-fucosylation usually occurs on EGF-like domains of proteins within the consensus sequence: CXXGG-S/T-C. EGF repeats are small motifs with about 40 amino acids defined by six conserved cysteines forming three disulphide bonds. They are well known for their role in protein-protein interactions. Recent work in a number of laboratories demonstrates that such receptor-ligand interactions can be affected by the alterations in the O-linked glycan modifications on EGF repeats. A good example is Notch. 26 Notch family members are cell-surface receptors that direct developmental processes by interacting with Notch ligands (Kadesch, 2004). In this evolutionarily conserved pathway, Notch ligand binding to Notch's large extracellular domain catalyzes the release of an intracellular fragment that modulates the transcription of developmentally relevant target genes. Whereas there is only one Notch gene in Drosophila melanogaster, there are at least four in mammals. The ligands for Notch are Delta and Jagged. Glycans modulate the interaction between Notch and its ligands, and the corresponding signalling events that occur. In addition to N-glycans, Notch is also subject to O-fucose and O-glucose modifications (Moloney et al., 2000b). These glycans are borne by serine and threonine residues within some of the many, EGF-like repeats of the Notch extracellualar domain (Haines and Irvine, 2003). Serines and threonines bearing only the sugar fucose characterize the simplest Notch glycoforms. O-fucosyltransferase-1 (O-Fuc-1) (Haines and Irvine, 2003) catalyzes the addition of fucose to these serines and threonines, a reaction believed to occur in the ER (Luo and Haltiwanger, 2005). This initial modification by O-Fuc-1 is absolutely required for all ligand-induced signalling (Okajima et al., 2003; Wang et al., 2001a). O-Fuc-1 possesses, as per recent studies, distinct Notch chaperone activity, facilitating normal folding and trafficking of Notch out of the ER (Lowe, 2005; Okajima et al., 2005). These O-fiicosylated EGF domains are elongated by fringe family P 1,3 GlcNAc transferases, which modulate Notch's response to its ligands (Bruckner et al, 2000; Hicks et al., 2000; Moloney et al., 2000a; Moloney et a l , 2000b; Panin et al., 2002). The three fringe enzymes (Radical fringe, Lunatic fringe and Manic fringe) require O-linked fucose as substrates and form a disaccharide that is further elongated by at least one specific pi,4 GalT (Chen et al., 2001), and then by subsequent actions of other glycosyltransferases. An O-fucose tetrasacharide structure is proposed where an oc2,3-sialyltransferase adds a 27 sialic acid (SA), following the addition of galactose, to form a SAoc2,3Galf31,4GlcNAcpl,3FucaSer/Thr structure (Haines and Irvine, 2003; Moloney et al., 2000b). O-Glucose An O-P-linked glucose defines this pathway. Glucose occurs as a covalent modification of serine and threonine residues of some proteins, including EGF domains of Notch, and may be elongated by subsequent glycosylation (Harris and Spellman, 1993; Shao et al., 2002). O-glucosyltransferase activity and subsequent other transferase activity yields xyioseal,3xyloseal,3GlucosepSer/Thr trisaccharide on Notch (Shao et al., 2002). Distinct enzymatic activities catalyze each step in the synthesis of the O-glucose trisaccharide (Omichi et al., 1997; Shao et al., 2002). None of these enzymes has been cloned, and their potential influence on Notch signalling remains unknown. The proposed consensus sequence for adding O-glucose is CX-S-XPC, where the first and last cysteines represent the first and second cysteines within the characteristic conserved pattern of six cysteines of the EGF domain (Harris and Spellman, 1993; Shao et al., 2002). O-Mannose O-mannosylation of serine and threonine residues is abundantly present in mammals (Strahl-Bolsinger et al., 1999). An elongated O-mannose-linked glycan is enriched in muscle tissue and on a-dystroglycan which is a component of the dystrophin-glycoprotein-complex that is altered in human muscular dystrophies (Chiba et al., 1997). O-mannosylation of proteins is an essential process for cell growth and proliferation. After adding O-mannose, glycosyltransferases can add GlcNAc, Gal and Fuc residues, 28 creating a Lewis X (Lex) structure important for specific cell adhesion processes (Strahl-Bolsinger et al., 1999; Varki et al., 1999). O-GlcNAc Nuclear and cytoplasmic proteins are modified exclusively by O-p-GlcNAc. However some proteins traversing the ER/Golgi pathway are O-a-GlcNAc-modified. The importance of O-p-GlcNAc modification is evident when analyzing the conservation of O-GlcNAc glycosyltransferases through evolution, with greater-than-85% similarity between the primary sequence of the enzyme derived from the nematode, C. elegans, and humans (Kreppel et al., 1997). The role of O-GlcNAc on nuclear pore proteins in nuclear transport is illustrated by O-GlcNAc-specific mAb's ability to block nuclear transport. The O-GlcNAc modification has supposed involvement in initial peripheral binding of transport molecules, followed by the docking of the transported molecule at the center of the pore complex, and its translocation (Varki et al., 1999). O-GlcNAc also modifies several different chromatin proteins. RNA polymerase II transcription factors and the catalytic subunit of RNA polymerase II are O-GlcNAcylated. The catalytic subunit of RNA polymerase II contains a seven-amino-acid sequence at its carboxyl terminus, which is repeated and is essential for viability. Prior to transcriptional elongation, O-GlcNAc modifies this region, but upon elongation O-GlcNAc residues are removed and replaced with extensive phosphorylation (Comer and Hart, 2001; Kelly et al., 1993). The oncoprotein, c-Myc, is another O-GlcNAcylated nuclear protein. c-Myc is modified in its transactivation domain at a threonine residue, which is also phophorylated and is a mutation hot spot in many diseases, including AIDS-related lymphomas. This demonstrates that reciprocal glycosylation and phosphorylation at this biologically 29 significant site are decisive in mediating the functions of c-Myc, and ultimately gene transcription (Chou et al., 1995; Kamemura et al., 2002). p53, a tumour suppressor and transcription factor mutated in many human tumors, is also O-GlcNAcylated. This O-GlcNAc modification is believed to regulate the affinity of p53 for DNA by masking the basic region on p53 that binds to DNA (Chou and Hart, 2001; Shaw et al., 1996). Many cytoskeletal proteins, including vinculin, talin and synapsin, involved in reversible bridging of the cytoskeleton to the membrane or other structures, are also O-GlcNAcylated. The O-GlcNAc modification here is proposed to mediate protein-protein interactions in many cellular functions. For example, talin bridges integrins to the cytoskeleton via its interaction with vinculin through this glycan (Hart, 1997). O-GlcNAc modification of intermediate filaments, neurofilaments and microtubule-associated filaments is also documented. For example, tau, a microtubule-associated protein, is heavily O-GlcNAcylated. Tau plays a key role in organizing microtubules in axons; the hyperphosphorylation of this protein in Alzheimer's disease prompts abnormal filaments to form, leading to neuronal death. This hyperphosphorylation state is understood to result from abnormal O-GlcNAcylation of serine and threonine residues in tau (Arnold et al., 1996; Liu et al., 2004). In summary, the key features characterizing O-GlcNAcylation are: 1) O-GlcNAc occurs at sites similar to those used by many important kinases; 2) O-GlcNAc modifications and phosphorylation occur on many of the same residues, and may have opposing functions on many proteins; and 3) O-GlcNAcylation is highly dynamic, with rapid cycling in response to cellular signals or cellular stages analogous to phosphorylation (Varki et al., 1999; Wells et al., 2001). 30 1.3.7 Focus The identification of selectins as glycan-binding proteins important in regulating leukocyte trafficking and adhesion marked a milestone in the development of the glycobiology field. Selectins are a perfect example of the structure-function relationships between glycosyltransferases and their cognate carbohydrate structural products as recognition molecules. Selectin ligands are heavily O-glycosylated mucins, with N-linked glycans supporting the extended structure of these mucins. These ligands are involved in mediating the interactions with selectins. This creates an immunological incentive to study the role of this interaction in regulating the movement of circulating cells to various places in the body. 1.4 Decorated ligands attract selectin counter-receptors Leukocytes must engage several sequential adhesion pathways to leave the circulation as outlined in Figure 1-2. Initially, tethers are formed by adhesion receptors specialized to engage rapidly and with high tensile strength. The most important initiators of adhesion are the family of three selectins that function to mediate leukocyte adhesion under flow. The selectins are C-type lectins expressed exclusively by bone-marrow-derived cells and endothelial cells: L-selectin for Leukocyte; E-selectin for Endothelial cell; and P-selectin for Platelet and endothelial-cell. The selectins are type I transmembrane glycoproteins with similar structures - an amino-terminal lectin-like domain with four conserved cysteine residues, one EGF-like domain, various consensus repeats (CR) with homology to complement regulatory proteins, a single membrane spanning domain and a carboxy-terminal cytoplasmic domain - as outlined in Figure 1-5 (Springer, 1994; Vestweber and Blanks, 1999). All three selectin loci are organized in a similar manner and are closely linked for both mouse and human (Stoolman, 1989). 31 1.4.1 Leukocyte-selectin L-selectin L-selectin (CD62L, LAM-1, LECAM-1, LEU-8, LEC.CAM-1, gp90M E L, DREG-56, MEL-14, or TQ-1) is constitutively expressed by all myeloid cells, virtually all B cells and naive T cells, a subpopulation of memory T cells, N K cells, and both early and mature hematopoietic cells in the bone marrow. Unlike the other selectins, though, L-selectin is not expressed on endothelial cells (Tedder et al., 1995a). L-selectin expression is transiently increased when naive T cells are activated, but lost quickly due to endoproteolytic release to generate a soluble form (Butcher and Picker, 1996; Tedder et al., 1995a). L-selectin is critical for homing of naive lymphocytes to HEV-bearing secondary lymphoid organs (Butcher and Picker, 1996; Rosen, 2004), as observed in vitro by Stamper and Woodruff in 1976 (Stamper and Woodruff, 1976), when they found that lymphocytes adhered to cryostat-cut sections of lymph node in a highly specific manner. According to them, "Recirculating lymphocytes and not... cells which are unable to home into lymph nodes in vivo" are able to adhere. Later on a monoclonal antibody, MEL-14, generated by Gallatin et al. (Gallatin et al., 1983) was found to block lymphocyte binding to HEV both in vitro and in vivo - and soon after became known as the lymphocyte homing receptor, L-selectin. In inflammatory leukocyte trafficking, L-selectin mediates leukocyte-leukocyte interactions (Tedder et al., 1995a), such as in the invasion of neutrophils into sites of inflammation. L-selectin-dependent lymphocyte rolling on HEV requires localization of L-selectin to the tips of the microvilli characteristic of the surfaces of lymphocytes and other leukocytes, in order facilitate optimal binding to its ligands. Mapping of L-selectin 32 domains by mAbs has revealed that the NH2-terminal nine amino acids are critical for ligand binding. The EGF-like domain and the two short CRs maintain the spatial conformation and are also important for ligand binding (Jutila et al., 1992; Lasky, 1995; Rosen, 2004). Figure 1-5 outlines L-selectin along with its proposed ligands. 33 Figure 1-5: Selectins and their ligands. The three selectins along with their most common ligands are illustrated. A) E-selectin is expressed on endothelial cells and recognizes sLex structures on PSGL-1 and ESL-1, expressed by leukocytes; it also recognizes CLA-capped sLex structures expressed on N-glycans of CD44, referred to as HCELL (hematopoietic cell E- and L-selectin ligand) which are expressed on hematopoietic stem cells. P-selectin, also expressed on endothelial cells, recognizes a terminal sLex O-glycan of PSGL-1 expressed by leukocytes. B) L-selectin expressed by migrating leukocytes recognizes the PNAds: GlyCAM-1, MAdCAM-1 and CD34 expressed on HEV of secondary lymphoid organs, for instance, peripheral lymph nodes. Interactions between L-selectin and PSGL-1 occur between L-selectin expressed on one leukocyte and PSGL-1 expressed on another. L-selectin may also recognize podocalyxin and endoglycan, but these are not shown in this figure. 34 Ligands for L-selectin HEV-bome L-selectin ligands are collectively referred to as peripheral node addressins (PNAds) and include the glycoproteins GlyCAM-1 (glycosylation-dependent cell adhesion molecule-1) (Lasky et al., 1992), CD34 (Baumheter et al., 1993), MAdCAM-1 (mucosal vascular addressin cell adhesion molecule-1) (Berg et al., 1993), podocalyxin (Sassetti et a l , 1998), Sgp200 (Hemmerich et al., 1994b) and endoglycan (Sassetti et al., 2000) - all of which have mucin-like domains that carry O-glycans (Rosen, 2004). For L-selectin to recognize these ligands, they must be modified by various glycosyltransferases to give specific sialyled, fucosylated and sulphated oligosaccharides. Section 1.5.1 will cover this topic. PNAds are collectively recognized by the MECA-79 monoclonal antibody (Streeter et al., 1988). MECA-79 specifically recognizes the luminal surface of HEV in peripheral lymph nodes, and inhibits lymphocyte attachment to HEV (Clark et al , 1998; Streeter et al., 1988; von Andrian, 1996). Each proposed ligand for L-selectin is discussed below. GlyCAM-1 The best-characterized ligand is GlyCAM-1, a mucin secreted from the HEV of peripheral lymph nodes (Imai et al., 1991; Lasky, 1995; Lasky et al , 1992). It is detectable in serum and is proposed to regulate the adhesive and/or activation state of L-selectin- expressing leukocytes by interacting with L-selectin on such cells. In order to bind L-selectin, GlyCAM-1 must undergo modifications with branched core 2 O-glycans that are fucosylated, sialylated and sulphated (Hemmerich et al., 1994a; Hemmerich et al., 1994b; Leppanen et al., 2003; Scudder et al., 1994). Figure 1-4 above outlines L-selectin-capping glycans. 35 CD34 The sialomucin, CD34, is a transmembrane glycoprotein containing several mucin-like domains with several serine and threonine residues predicted to be heavily O-linked glycosylated. CD34 is broadly expressed on vascular endothelium (Baumheter et al., 1993; Puri et al., 1995), hematopoietic stem cells and progenitors (Gratama et al., 2001; Krause et al., 1996; Simmons et al., 1992), as well as by murine mast cells (Drew et al., 2002; Drew et al., 2005). CD34 functions as a ligand for L-selectin only when expressed in HEV, since the posttranslational modifications required to mediate the binding interaction only occur here, for instance, glycosyltransferases responsible for adding glycan linkages critical for binding are expressed (Puri et al., 1995; Renkonen et al., 2002). CD34n u 1 1 mice do not show any obvious abnormality in homing to lymph nodes (Cheng et al., 1996; Suzuki et al., 1996), suggesting that the HEV-expressed sialomucins are functionally redundant (Doyonnas et al., 2001; Sassetti et al., 1998; Sassetti et al., 2000). However, a recent study demonstrates that CD34, along with CD43, plays a central role in hematopoietic precursor cell bone marrow engraftment in a competitive transplantation model, independent of L-selectin. By reducing mast cell pro-adhesive interactions, CD34 and CD43 mediate migration of committed mast cell progenitors to the peritoneal cavity, suggesting that they block cell-cell interactions and prevent inappropriate adhesion to enhance the ability of mast cells to migrate to new microenvironments (Drew et al., 2005). Podocalyxin Podocalyxin is expressed at high levels by mesothelial cells, vascular endothelium (Sassetti et al., 1998), platelets and podocytes (Doyonnas et al , 2001). 36 Podocalyxin, present on HEV, binds both L-selectin and MECA-79 (Sarangapani et al., 2004). Podocalyxin knockout mice exhibit defects in kidney development and die within 24 hours of birth from anuric renal failure, because podocytes fail to form foot processes and slit diaphragms. In addition, these mice display herniation of the gut, suggesting that podocalyxin may be required for retraction of the gut from the umbilical cord during development. Hematopoietic and vascular endothelial cells develop normally in the podocalyxinnu" mice, again suggesting functional redundancy of other sialomucins (Doyonnas et al., 2001). Podocalyxin is also a marker of embryonic hematopoietic stem cells (McNagny et al., 1997), erythroid cells and adult hematopoietic stem cells in mice. Data from the McNagny lab suggests that podocalyxin is a valuable marker for purifying hematopoietic stem cells for transplantation (Doyonnas et al., 2005). MAdCAM-1 MAdCAM-1, isolated from mesenteric lymph nodes, supports adhesion of L-selectin-transfected lymphoid cells under shear flow (Berg et al., 1993). Through interactions with L-selectin and OC4P7 integrin, MAd-CAM-1 supports lymphocyte tethering and rolling (Berlin et al., 1995; Patel et al., 2002) and, in vivo, is demonstrated to support tethering and rolling on HEV of Peyer's patches (Bargatze et al., 1995; Patel et al., 2002). Therefore, it is assumed that MAdCAM-1 maintains ligand activity for both L-selectin and the OC4P7 integrin. PSGL-1 L-selectin also directly binds P-selectin glycoprotein ligand-1 (PSGL-1) on monocytes, neutrophils, lymphocytes and hematopoietic progenitor cells (Guyer et al., 1996; Spertini et al., 1996; Tu et al., 1996; Walcheck et a l , 1996). PSGL-1 is a homodimeric selectin ligand expressed on the surface of almost all circulating leukocytes (McEver and Cummings, 1997). This interaction, it is supposed, takes place through secondary tethering via neutrophil-on-neutrophil rolling, thereby enhancing leukocyte recruitment at inflammatory sites (Spertini et al., 1996; Walcheck et al., 1996). A study by Sperandio et al. finds that PSGL-1 mediates L-selectin-dependent rolling in inflamed venules (Sperandio et al., 2003). The authors deduce that, in most cases, L-selectin-dependent rolling occurs on already-adherent leukocytes - a conclusion observed in numerous in vitro (Alon et al., 1996; Bargatze et al., 1994; Walcheck et al., 1996) and in vivo (Eriksson et al., 2001; Kunkel et al., 1998) studies. In some cases, however, L-selectin and PSGL-1-dependent rolling is observed in the absence of adherent leukocytes. The authors postulate that rolling leukocytes, platelets or other cells may deposit PSGL-1 in small membrane fragments on the inflamed endothelium. Endoglycan Expressed by many different cell types - vascular endothelial cells, smooth muscle cells, hematopoietic precursor cells and leukocyte subsets - endoglycan is another member of the CD34 family of glycoproteins (Doyonnas et al., 2001; Doyonnas et a l , 2005; McNagny et al., 1997; Sassetti et al., 1998; Sassetti et al., 2000). Endoglycan's mechanism of interaction with L-selectin differs from the other CD34 family members. Fieger et al. explain that the amino terminal domain of endoglycan is very similar to that of PSGL-1 and predict that they bind selectins in an analogous fashion (Fieger et al., 2003). They underline parallels between endoglycan and PSGL-1 in its amino terminal sequence, including overlapping tyrosine sulphation sites and sLex modification of threonine residue, as well as a dependency 38 for sialylation and fucosylation. Endoglycan, like PSGL-1, also forms a disulphide-dependent homodimer. This very intriguing parallel could indicate that endoglycan, like PSGL-1, binds P-selectin and E-selectin, in addition to L-selectin. HCELL More recently, another ligand for L-selectin was identified on hematopoietic stem cells, named HCELL, (hematopoietic cell E/L-selectin ligand) (Dimitroff et al., 2000; Dimitroff et al., 2001a; Dimitroff et al., 2001b), where its expression is restricted. HCELL is a distinct glycoform of CD44 that expresses the carbohydrate epitope, cutaneous lymphoid antigen (CLA) on N-glycans. The globular NH2 domain of CD44 contains hyaluronic acid (HA)-binding motifs and several potential glycosylation sites (Delcommenne et al., 2002; Johnson et al., 2000). To date, two L-selectin ligands are identified among human bone marrow cells: HCELL and PSGL-1. HCELL has emerged as the principal L-selectin and E-selectin ligand on HSC since it binds more avidly than PSGL-1 to these selectins (Dimitroff et al., 2001a; Dimitroff et al., 2001b; Sackstein, 2004). Although CD34 binds L-selectin in HEV, the absence of proper posttranslational modifications in stem cells prevents L-selectin from recognizing it (Oxley and Sackstein, 1994; Puri et al., 1995; Renkonen et al, 2002; Sackstein et al., 1997). 1.4.2 Endothelial cell-selectin E-selectin E-selectin (CD62E, ELAM-1 or LECAM-2) is expressed by acutely inflamed endothelial cells in most organs. Unlike P-selectin, E-selectin is not constitutively present in endothelial cells, but is, rather, transcriptionally regulated by mediators, 39 including TNF-oc, IL-1 f3, IFN-yand LPS. Also unlike L-selectin, E-selectin is not expressed on myeloid and lymphoid cells. Following stimulation with these cytokines, peak expression of E-selectin occurs at four hours and then declines within 24 hours (Tedder et al., 1995a). The decline in E-selectin expression is prompted by declining transcription of the E-selectin locus, degradation of the E-selectin transcripts, and internalization and degradation of E-selectin molecules by the endothelium (Subramaniam et al., 1993). E-selectin is also expressed in non-inflamed microvessels of the skin (Keelan et al., 1994; Weninger et al., 2000) and bone marrow (Jacobsen et al., 1996; Schweitzer et al., 1996). E-selectin-mediated interactions are involved in: efficient leukocyte recruitment to an injury site; and steady-state, tissue-specific homing in cutaneous tropism of skin-homing T cells (Berg et al., 1991; Picker et al., 1990a) and HSC entry into bone marrow (Frenette et al., 1998; Mazo et al., 1998; Naiyer et al., 1999). Figure 1-5 outlines E-selectin with its proposed ligands. Ligands for E-selectin Neutrophils, monocytes, eosinophils, memory-effector T cells and NK cells express E-selectin ligands. These cell types are found in acute and chronic inflammatory sites, associated with E-selectin expression, thus implicating E-selectin in cell recruitment to such inflammatory sites. To date, two major glycoprotein ligands for E-selectin are identified: E-selectin ligand-1 (ESL-1), which binds specifically to E-selectin but not to P-selectin (Lenter et al., 1994; Levinovitz et al., 1993), and PSGL-1. Other proposed ligands are also decorated with sLex-type glycans (Figure 1-4), including L-selectin (Patel et al , 2002; Picker et al., 1991b; Zollner et al., 1997), Mac-1 (Crutchfield et al, 2000), CD44 (Dimitroff et al., 2001a; 40 Katayama et al., 2005), CD43 (Maemura and Fukuda, 1992) and some glycolipids (Burdick etal., 2001). ESL-1 ESL-1, purified from murine myeloid lineage cells, is a 150 kDa type-I transmembrane glycoprotein (Levinovitz et al., 1993) with a glutamine-rich NH2-terminal segment of 70 amino acids, followed by 16 cysteine-rich repeats, a transmembrane domain and a short cytoplasmic tail. The ectodomain contains five potential N-glycosylation sites (Steegmaier et al., 1997; Steegmaier et al , 1995). In contrast with almost all other selectin ligands that are sialomucins and require O-linked carbohydrates for selectin binding, ESL-1 binding to E-selectin depends on N -linked carbohydrates (Lenter et al , 1994; Levinovitz et al., 1993). However, as for other known ligands, binding is both sialic acid and fucose dependent (Levinovitz et al., 1993; Steegmaier et al., 1995). ESL-1 is expressed on a wide variety of cells, but only on myeloid cells is the glycoform that binds to E-selectin identified (Vestweber and Blanks, 1999). PSGL-1 The issue of whether PSGL-1 is a functional ligand for E-selectin is currently under debate. Many studies support the hypothesis that PSGL-1 is a high-efficiency ligand for E-selectin both in vitro (Asa et al., 1995; Goetz et al., 1997; Merzaban et al., 2005; Patel et al., 2002; Somers et al , 2000; Zou et al., 2005) and in vivo (Hirata et al., 2002; Hirata et al., 2000; Norman et al., 2000; Xia et al., 2002a). In vitro studies, using static or flow assays and blocking PSGL-1 mAb, identify human neutrophils binding to E-selectin in a PSGL-1-specific manner (Asa et al., 1995; Patel et al., 2002). Data published by Goetz et al. underline that E-selectin expressed on 41 CHO cells supported tethering and rolling of PSGL-1-coated microspheres (Goetz et al., 1997) and PSGL-1-expressing human neutrophils (Zou et al., 2005) under flow in vitro. A recent in vivo study demonstrates that microspheres coated with a human PSGL-1-IgG chimera attached and rolled on E-selectin in TNF-a-stimulated mouse mesenteric venules (Norman et al., 2000), supporting studies with the PSGL-l n u 1 1 mouse that conclude that PSGL-1 n u l 1 leukocytes have significantly lower rates of tethering to E-selectin compared with leukocytes from wild-type mice (Merzaban et al., 2005; Xia et al., 2002a). The role of PSGL-1 in lymphocyte adhesion to E-selectin in vivo was determined using PSGL-l n u 1 1 mice. Hirata et al. exhibit that the ability of PSGL-l n u 1 1 T H I cells to bind E-selectin declines significantly under semi-static conditions; these cells also display decreased migration into the skin when injected into P-selectinnu" mice (Hirata et al., 2002; Hirata et al., 2000). T cell binding to E-selectin is correlated with expression of HECA-452 reactivity epitopes (Borges et al., 1997b). The HECA-452 mAb is directed against a sLex epitope on PSGL-1, referred to as cutaneous lymphoid antigen (CLA) (Berg et al., 1991; Duijvestijn et al., 1988; Fuhlbrigge et al., 1997). The CLA epitope is found on a subset of skin-homing memory T cells that bind to the E-selectin expressed on the skin microvasculature (Berg et al , 1991; Picker et al., 1991a; Picker et al., 1990b). A recent study further establishes that the CLA modifications on PSGL-1 promote E-selectin binding, showing that only CLA+PSGL-1 functions as an E-selectin ligand, but both CLA+PSGL-1 and CLA" PSGL-1 are able to function as P-selectin ligands (Fuhlbrigge et al., 2002). In a similar study, Zou et al., interested in defining the interaction of neutrophil-PSGL-1-conjugated microspheres to E-selectin, determine that PSGL-1 can bind E-selectin in a non-HECA-452 manner (Zou et al , 2005). 42 Challenging the data supporting a significant role for PSGL-1 in E-selectin-dependent interactions, studies also support that PSGL-1 is not a physiological ligand for E-selectin. Pretreatment of human neutrophils with a blocking PSGL-1 mAb has no affect on the rate of tethering to E-selectin under fluid shear conditions in vitro. This supports early work with the PSGL-1 n u I 1 mouse which concludes that PSGL-1 is not required for E-selectin- mediated neutrophil rolling in TNF-a-treated cremaster venules in vivo (Yang et al., 1999), contradicting the Xia et al. study (Xia et al., 2002a). In support of Yang et al, a recent study also determines that leukocyte rolling is not altered in PSGL-1 n u l 1 mice, suggesting that an alternate E-selectin ligand, CD44, is responsible for E-selectin-mediated neutrophil extravasation (Katayama et al., 2005). The differences presented in these studies are inexorably linked to the types of experiments performed. Some microsphere experiments used recombinant PSGL-1 (Goetz et al., 1997), which may or may have not been glycosylated in an appropriate manner to bind E-selectin (Fuhlbrigge et al., 2002). Differences in expression of enzymes involved in posttranslational modifications can vary with species (mouse vs. human), cell type (neutrophil vs. lymphocyte) and maturation/activation (effector/ memory T cells) - causing differences in the ability of selectin ligands to bind E-selectin (Borges et al., 1997a; Fuhlbrigge et al., 2002; Kobzdej et al., 2002). The type of in vivo study performed can also impact the observations. With both PSGL-1-dependent and -independent binding to E-selectin described at length, a recent study set out to define the circumstances controlling this (Zanardo et al., 2004). This analysis finds that an E-selectin ligand distinct from PSGL-1 is rapidly down-regulated from circulating leukocytes during systemic inflammation, making the rolling on P-selectin and E-selectin entirely PSGL-1-dependent (Zanardo et al., 43 2004). In absence of PSGL-1, E-selectin-mediated rolling in the cremaster microcirculation following local administration of TNF-a, is partially inhibited, whereas it is abolished following systemic TNF-a administration. To determine whether the PSGL-1-independent E-selectin ligand is physiologically important, the study uses a P- and E-selectin-dependent cutaneous contact hypersensitivity model and finds that cells did roll on E-selectin in the absence of PSGL-1, and are able to support the inflammation, concluding that E-selectin mediates PSGL-1-dependent and -independent rolling. The latter can be down-regulated by systemic activation and can replace PSGL-1 to support the development of inflammation (Zanardo et al., 2004). CD44 and HCELL Flow chamber studies observe rolling of human HSCs on immobilized E-selectin, showing that CD34+ stem cells bind with a higher efficiency than CD34" cells (Greenberg et al., 2000). The proposed E-selectin ligands on human HSC are PSGL-1 and HCELL. Work from the Sackstein lab uncovers a unique role for HCELL, a glycoform of CD44, in the homing of human HSC to bone marrow (Dimitroff et al., 2001a; Sackstein, 2004). A study from the Frenette lab, meanwhile, draws more attention to CD44 as an E-selectin ligand on human and mouse myeloid cells, finding that PSGL-1 cooperates with CD44 to enable leukocyte rolling on E-selectin in vivo (Katayama et al., 2005), where CD44 binding to E-selectin depends on fucosylated, siaylated N-glycans. Several groups detail antibody-attenuated interactions between E-selectin and L-selectin (Patel et al., 2002; Picker et al., 1991b; Zollner et al., 1997), however, whether this interaction is of physiological significance is yet to be determined. 44 1.4.3 Platelet (& endothelial) cell-selectin P-selectin P-selectin (CD62P, LECAM-3, GMP-140, or PADGEM) is a highly glycosylated 140 kDa protein found constitutively expressed in secretory granules of platelets and endothelial cells. In platelets, P-selectin is found in oc-granules, but is expressed on the platelet surface after activation. In endothelial cells, P-selectin is found in Weibel-Palade bodies but, within minutes of activation (by histamine or thrombin for instance), is expressed on the surface, due to fusion of the secretory granule with the endothelial membrane (Vestweber and Blanks, 1999). P-selectin is also constitutively expressed at low levels on the endothelium of thymus (Rossi et al., 2005), lung and choroid plexus microvessels (Kivisakk et al., 2003), bone marrow microvasculature (Jacobsen et al , 1996; Schweitzer et al., 1996), in postcapillary venules in skin (Weninger et al., 2000), and on peritoneal macrophages (Tchernychev et al., 2003). Once expressed on the surface of endothelial cells, P-selectin is rapidly internalized by endocytosis (Hattori et al., 1989). P-selectin contributes to leukocyte recruitment in a variety of acute and chronic inflammatory contexts common in many diseases including atherosclerosis (Kansas, 1996; Ley, 2003). Its role in leukocyte recruitment and inflammation is a function of both acute and chronic expression of P-selectin by the endothelium and by platelet-activation-dependent P-selectin expression. The role of P-selectin in acute inflammation is well illustrated in P-selectinnu11 mice, discussed in section below. P-selectin can also contribute to chronic inflammation, presumably through transcriptional up-regulation of its expression later in the inflammatory response (Ley, 2003). Platelet-derived P-selectin also contributes to leukocyte trafficking, 45 wound healing and blood clotting. For example, activated platelets can adhere to neutrophils in a P-selectin-dependent manner and augment leukocyte and platelet recruitment to sites of vascular compromise. Figure 1-5 outlines the structure of P-selectin along with its proposed ligands. Ligands for P-selectin Leukocytes express only one glycoprotein that binds with high affinity to P-selectin - PSGL-1, a relatively minor mucin on leukocytes. PSGL-1 is a 240 kDa homodimer that, i f glycosylated appropriately, can bind all three selectins, although with different binding strengths and association kinetics (Moore, 1998). It is broadly expressed on cells of myeloid (monocytes, neutrophils), lymphoid (CD4, CD8, N K , y/8 and B cells) and dendritic cell lineages, as well as on some nonhematopoetic cells (Laszik et al., 1996; Snapp et al., 1998b). A recent study shows PSGL-1 expression by T cell progenitor cells and outlines its involvement in progenitor homing to the thymus (Rossi et al., 2005). PSGL-1 was originally identified by expression cloning from COS cells that were co-transfected with a human HL-60 cDNA library and a human a l ,3-fucosyltransferase, FucT-III (Sako et al., 1993). The mouse homologue for PSGL-1 was cloned based on similarity to the human sequence (Yang et a l , 1996), containing the predicted signal sequence and propeptide identical in size to that of human PSGL-1. PSGL-1 is a type I transmembrane mucin-like protein of 412 residues for human (Sako et a l , 1993) and 397 residues for mouse (Yang et al., 1996). The mature protein begins at residue 42 after proteolysis of a signal peptide and cleavage of a propeptide for both human and mouse forms (Leppanen et al., 2000; Sako et al., 1993; Vachino et al., 1995; Yang et al., 1996). As depicted in Figure 1-7, the mature 46 protein forms homodimers via disulphide bridges linking two 120 kDa chains close to the plasma membrane. A single cysteine residue occurs at position 320 in human and 307 in mouse, within the extracellular domain preceding the predicted transmembrane-domain-spanning region. This dimerization is important for high-affinity binding to P-selectin (Croce et al., 1998; Leppanen et al , 2000; Snapp et al., 1998a) and increases the rate of tethering to P-selectin under flow (Smith et al., 2004). There are 16 serine/threonine rich decamer repeats with the consensus sequence A-T/M-EAQTTX-P/L-A/T in the human protein, while only 10 decamer repeats with the consensus sequence ETS-Q/K-PAP-T/M-EA exist in the murine form. This repeat region is rich in serine, threonine and proline residues, suggesting that O-glycosylation of the peptide backbone is conformationally constrained by proline residues. These repeats are different in sequence and, actually, the highest homology between human and mouse PSGL-1 occurs in the transmembrane (83%) and cytoplasmic (76%) domains (Yang et al., 1996). Each subunit of human PSGL-1 contains 70 serine and threonine residues in the extracellular domain that are potential sites for O-glycosylation and three potential sites for N-glycosylation (Sako et al., 1993). The murine form also contains many extracellular serine and threonine residues, but only two potential N-glycosylation sites (Yang et al., 1996). A diagram, outlining the domains in human and mouse PSGL-1 are shown in Figure 1-6. The P-selectin binding domain of PSGL-1 occurs within the first 20 residues of the mature protein (Pouyani and Seed, 1995; Sako et al., 1995). Sulphation of tyrosine residues and O-glycosylation of a single threonine residue within the mature NH2-terminal region of PSGL-1 appear necessary for the high-affinity binding of 47 PSGL-1 to P-selectin (Leppanen et al., 2000; Liu et al., 1998; Pouyani and Seed, 1995; Sako et al., 1995; Wilkins et al., 1995). The crystal structure of P-selectin and E-selectin binding to sLex and PSGL-1 is determined and has given much insight into the molecular basis of this interaction (Somers et al., 2000). The interaction is largely electrostatic; consistent with the rapid-binding kinetics selectins are known for. PSGL-1 contains O-linked glycans forming highly extended structures that participate with the NH2-terminal-binding domain of PSGL-1 to bind P-selectin. In vitro inhibition of PSGL-1 completely abolishes neutrophil rolling on P-selectin. This is consistent with the finding that genetic deletion of PSGL-1 attenuates P-selectin-mediated rolling of leukocytes in vivo. Two laboratories have generated PSGL-1 n u l 1 mice (Xia et al., 2002a; Yang et al., 1999)! Both groups determined that these mice are viable, fertile and have modestly elevated blood leukocyte (neutrophils) counts. Leukocyte infiltration in the chemical peritonitis model was significantly delayed (Yang et al., 1999). Intravital microscopy studies in postcapillary venules of the cremaster muscle establish that leukocyte rolling significantly decreases 30 minutes after trauma, due to dependence on P-selectin for PSGL-1 early in the inflammatory response. In contrast, two hours after TNF-oc stimulation leukocytes rolling declined only modestly and depended on E-selectin (Yang et al., 1999). However Xia et al. found that significantly fewer PSGL-1 n u l 1 leukocytes rolled on E-selectin in TNF-a-treated venules of the cremaster 48 Figure 1-6: Comparison of human and mouse domains of PSGL-1. The human and mouse forms of PSGL-1 are shown outlining the important domains for each. The mature protein begins at residue 42, after cleavage of the propeptide. The murine form contains 10 decamer repeats while the human form contains either 15 or 16 repeats (purple boxes) within the extracellular domain. These repeats contain many serine and threonine residues that are O-glycosylated as well as proline residues. There is a conserved cysteine residue (arrow in figure) just before the transmembrane domain in both the mouse and human proteins that is involved in forming disulphide bonds in the mature PSGL-1 dimer structure. muscle, in which a mAb against P-selectin blocked its function, suggesting that PSGL-1 is involved in tethering of leukocytes to E-selectin (Xia et al., 2002a). These studies concur on the importance of P- selectin and PSGL-1 in the early inflammatory response, although they disagree on the importance of the interaction of E-selectin with PSGL-1. The small, O-linked, oligosaccharide-modified glycoprotein CD24 was identified as another P-selectin ligand (Aigner et al., 1997). CD24 is mainly expressed on neutrophils. Neutrophils express both PSGL-1 and CD24, with PSGL-1, 49 being the dominant ligand for P-selectin. However, in the absence of PSGL-1, CD24 can mediate rolling on P-selectin in a CD24-dependent manner (Aigner et al., 1998). M O U S C (397aa) Az ppGalNAcT B= core 1 |53GalT C= C2SlcNMcT-I ( - n i ) D= (34GalT-I E= |33&lcNAcT (?) F= ST3SalT-IV S= FucT-IV H- FucT-VH I: TPSTI /H NH 2 T-V-N-K-L-L-E-P-P-D-T"N-Y-T-Y-D-P-D-E-F-D-D-D-G-V-V-Q—f\y\S 25 17 151413 Human (4i2aa) Figure 1-7 Glycosylation on human and murine PSGL-1. The figure depicts the structure of PSGL-1 homodimer. N-terminal tyrosine sulphates (small, purple ovals) are followed by a long, extended glycoprotein backbone that is heavily glycosylated with O-glycans and N-glycans (N). Essential O-glycan modification is indicated. A stabilizing disulphide bond is located near the plasma membrane (PM), and the transmembrane domain and short cytoplasmic tail are also shown. The amino acid sequence of the mature protein (with 41-amino-acid signal sequence and propeptide-cleaved purple wavy line) is illustrated, with tyrosine sulphate residues shown (SO4) and the crucial O-glycan at T16 (T57 of full length protein) (human) T17 (T58 of full length protein) (mouse) indicated. Typical carbohydrate side chains are delineated, along with the enzymes responsible for the linkages. A purple shaded box highlights the sLe" motif critical for mediating binding of PSLG-1 to P-selectin. Diagram adapted from (Ley, 2003) 1.4.4 Selectin knockout mice Extensive research has focused on determining the exact role of each selectin in the inflammatory response. Mice deficient in L-selectin (Arbones et al., 1994), E-selectin (Labow et al., 1994) and P-selectin (Mayadas et al., 1993) have been generated and characterized in a variety of inflammatory models. Al l three selectin genes are closely linked in a gene cluster, covering approximately 300 kb on chromosome 1 (Watson et al., 1990). Studying mice with deficiencies in two or three selectins has further clarified the overlapping and unique functions of each selectin. Single knockout mice Of the single selectin knockout mice, L-selectin deficiency provides the most-dramatic phenotype. Lymphocytes from these mice fail to bind to lymph node HEV, have smaller and reduced cellularity in peripheral lymph nodes, as well as impaired recruitment in several models of inflammation (Arbones et al., 1994; Steeber et al.,-1996; Tedder et al., 1995b). Short-term homing experiments demonstrate L-selectin's involvement in lymphocyte migration to mucosal lymph nodes, Peyer's patches and spleen (Arbones et al., 1994). In a thioglycollate model of inflammation, leukocyte rolling and peritoneal emigration of neutrophils were reduced (Ley et al., 1991; Tedder et al., 1995b). In TNF-oc treated cremaster venules, initial rolling of leukocytes is P-selectin dependent, although after one hour, rolling is dependent on both P- and L-selectin (Ley and Tedder, 1995). P-selectin-deficient mice have a number of defects in leukocyte behaviour, including mild neutrophilia and absence of leukocyte rolling in mesenteric lymph node venules (Mayadas et al., 1993). These mice also exhibit protection in models of ischemia-reperfusion, lung transplantation and atherosclerosis (Connolly et al., 1997; 51 Johnson et al., 1997; Naka et al., 1997). In addition, they display a temporary delay in recruitment at early time points (one-two hours), in response to an inflammatory stimulus (Mayadas et al , 1993). E-selectin-deficient mice, by contrast, do not display any adverse phenotype in models of inflammation and contact hypersensitivity (Bullard et al., 1996; Labow et al., 1994) but they do have higher rates of mortality and bacteremia in response to Streptococcus pneumoniae (Munoz et al., 1997). Thus, mice with null mutations in individual selectins have mild-to-moderate inflammatory defects. When the functions of more than one selectin were blocked, more severe phenotypes are observed. By blocking E-selectin and P-selectin by injecting murine P-selectin Ab into E-selectinnu11 mice, Labow et al. reveal that neutrophil emigration could be inhibited. In a thioglycollate-induced peritonitis model, neutrophil accumulation at early times depends on P-selectin, while neutrophil accumulation at later time is inhibited only in absence of both E-/P-selectin (by the a-P-selectin Ab in E-selectinnu11 mice). Similarly, in a model of DTH, blocking P-selectin function in E-selectinnu11 mice eliminates edema and leukocyte accumulation in the skin (Labow et al., 1994). These results suggest that, although selectins do have some degree of overlap in function, they express some unique specificity. Double knockout mice Double and triple selectin knockout studies indicate that P-selectin is the most versatile selectin, since it alone can mediate levels of leukocyte rolling in many inflammatory models that lack either E-selectin, L-selectin, or both, suggesting that optimal leukocyte rolling depends on the cooperation of P-selectin with either of the other selectins. 52 E/P-selectin deficient (only L-selectin expression) Mice double deficient in both P- and E-selectin exhibit an especially pronounced phenotype (Bullard et al., 1996; Frenette et al., 1996). These mice have extreme leukocytosis (Frenette et a l , 1996; Robinson et al., 1999), elevated circulating inflammatory cytokines, such as IL-3 and GM-CSF levels, enlarged spleens, altered hematopoiesis, and develop spontaneous skin and mucosal infections, which eventually cost them their life (Frenette et al., 1996). These mice also have a reduced, delayed-type hypersensitivity (DTH) response to oxazolone (Staite et al., 1996), impaired wound healing (Subramaniam et al., 1997) and exhibit protective effects in atherosclerosis (Dong et al., 1998). Bullard et al. also developed P-/E-selectin double-deficient mice. They discovered that these mice have severe mucocutaneous infections, plasma cell proliferation, hypergammaglobulinemia, severe deficiencies of leukocyte rolling in cremaster venules with or without addition of TNF-a, and an absence of neutrophil emigration at four hours, in response to peritonitis (Bullard et al., 1996). Over all, these results suggest that both P- and E-selectin, beyond having overlapping functions, act together in inflammatory responses. L-/P-selectin deficient (only E-selectin) An Ab blockade study reveals that blocking L- and P-selectin in a peritonitis model reduces neutrophil recruitment (Bosse and Vestweber, 1994). In a study performed with P-selectinnu11 mice injected with L-selectin Ab (Ley et al., 1995) or with L-selectinnu11 mice injected with P-selectin Ab (Kunkel and Ley, 1996) treated with TNF-a, leukocyte rolling is substantially reduced. Jung and Ley produced double- (L/P, L/E) and triple-selectin-deficient mice through bone marrow 53 transplantation (Jung and Ley, 1999) and found that L/P-selectinnu mice convey an absence of leukocyte rolling after trauma and severely reduced rolling in inflammation, induced by TNF-a in cremaster venules. They interpret the phenotype as similar to the E/P-selectinnu" phenotype described above, suggesting that P-selectin is the most important selectin in mediating early leukocyte recruitment. These results are further confirmed with the creation of mice genetically deficient in E- and P-selectin (Robinson et al., 1999); such mice display drastic reductions in leukocyte rolling, and extravasation of neutrophils in a thioglycollate-induced peritonitis model and in a ragweed-induced peritoneal eosinophilia. L/E-selectin deficient (only P-selectin) In the same study, Jung and Ley find that L/E-selectinnuU mice exhibit rolling similar to L-selectinnu" mice, with only a moderate reduction in leukocyte rolling (Jung and Ley, 1999). This is further verified by analyzing mice genetically deficient in L/E-selectin (Robinson et al., 1999). These mice show only minor leukocytosis compared with other double-selectin-deficient mice. E-selectin and L-selectin, hence, have minor functions in rolling and adherence of leukocytes to TNF-a-activated cremaster endothelium when compared with P-selectin. Triple knockout mice Mice deficient in all three selectins show significantly impaired neutrophil recruitment, similar to mice that are double deficient in P- and E-selectin. Jung and Ley deduce that there is very little rolling observed in TNF-a-treated cremaster venules and that this rolling depends on 04 integrins (Jung and Ley, 1999). In a follow-up study, Forlow and Ley (Forlow and Ley, 2001) demonstrate that in mice 54 lacking E-/P- and L-selectin, in addition to ICAM-1, there is a 97% reduction in leukocyte rolling and a 99% reduction in neutrophil recruitment in a thioglycollate-induced model of peritonitis. However, the reduction in firm adhesion is only 63%. A substantial amount of leukocyte adhesion occurred in the absence of E-, P-, and L-selectins, as well as E/P/L-selectin/ICAM-1, despite very few rolling leukocytes. Kubes et al. expound that in order to affect recruitment, at least 90% of rolling must be blocked (Kubes et al., 1995). Analyses of mice genetically deficient in all three selectins show similar phenotypes as those described by Jung and Ley (Robinson et al , 1999). These mice develop mucocutaneous infections that eventually lead to death. Like P/E-selectinnu11 mice, they also have splenomegally that is not present in any other selectin knockout combinations. These mice, like P/L-selectinnu11 and P/E-selectin11"11 mice, have leukocytosis, resulting in part from alterations in leukocyte rolling and recruitment. P-/E-/L-selectinnu mice like P/E-selectinnu11 mice and P/L-selectinnu" mice, display drastic reductions in leukocyte rolling and in extravasation of neutrophils in thioglycollate-induced peritonitis. Selectin-independent rolling is very inefficient since the number of cells rolling is greatly reduced, the rolling velocity is rapid and the adherence of leukocytes to the endothelium is drastically impaired (Robinson etal, 1999). Studies with triple-selectin knockout mice were expected to show a much more dramatic phenotype than either the single- or double-selectin-deficient combinations, but this is not the case. These studies reviewed here confirm that P-selectin has an essential role in inflammatory responses and that E-selectin does not necessarily compensate for the loss of P-selectin, but has its own unique functions. These studies also do not support a function for L-selectin dependent recruitment in 55 inflammation and suggest that the main function of L-selectin rests in mediating homing of naive lymphocytes, as is seen in the single L-selectin knockout mouse. 1.5 Glycosyltransferases involved in selectin ligand biosynthesis Selectin ligand binding requires interaction with specific carbohydrate residues and several glycosyltransferases involved in biosynthesis of selectin ligands are identified (Figure 1-7). Modification of specific serine and threonine residues with GalNAc (a linkages), made by one of several polypeptide GalNAcT transferases (ppGalNAcTs), represents the initiating step in classical O-glycan synthesis (Figure 1-8). O-glycan chain initiation by the ppGalNacTs is followed by modification, involving distinct, branch- specific glycosyltransferases, including: at least two al,3-fucosyltransferases, FucT-VII and FucT-IV; the O-linked branching enzyme core 2 pl,6-glucosaminyltransferase-I (C2GlcNAcT-I); a pl,4-galactosyltransferase-I ((3l,4GalT-I); at least two sialyltransferases of the ST3 Gal family that add sialic acid to galactose in a 2-3 linkage, one of which is ST3Gal-IV; and sulphotransferases. Figure 1-7 outlines a composite structure of a prototypic carbohydrate selectin ligand, resulting from activity of these enzymes. In addition, tyrosine sulphotransferases must be active to produce high-affinity P-selectin binding (Ouyang et al, 1998; Somers etal., 2000). All glycosyltransferases are type II transmembrane-Golgi/ER localized enzymes with a short NH2-terminal tail in the cytosol followed by a transmembrane domain, linked to the catalytic domain in the Golgi/ER lumen by a short stem region (Figure 1-9). Each glycosyltransferase generally uses only one of several sugar nucleotide substrates (including UDP-Galactose (Gal), UDP-Glucose (Glc), UDP-N-acetylgalactosamine (GalNAc), UDV-N-acetylglucosamine (GlcNAc), GDP-fucose (Fuc), GDP-mannose (Man), UDP-xylose (Xyl), CMP-sialic-acid (SA)). Each glycosyltransferase also uses one and sometimes more glycan 56 acceptor substrates, and can make only one specific glycosidic bond. Glycan synthesis is the consequence of a series of ordered, glycosyltransferase-dependent events, characterized by glycan polymer elongation or branching (Figure 1 -8). In general, it is the repertoire of glycosyltransferases expressed by a cell that dictates the glycan structures it will make. In the context of the glycan structures that confer selectin counter-receptor activity upon a cell, the glycosyltransferase repertoire will therefore participate in the control of selectin-dependent leukocyte recruitment (Lowe, 2003). Although all three selectins recognize sLex epitope (Figure 1-7), each selectin requires fine structural details of this epitope or the peptide backbone to confer optimal binding affinities (Lowe, 2002; Varki, 1997). 1.5.1 Glycosyltransferases involved in formation of L-selectin ligands Analysis of glycans found on serine and threonine residues of L-selectin ligands (PNAds) have uncovered sulfo-sLex-capping groups on core 1, core 2, or both branches of O-glycan linkages. The enzymes involved in the key posttranslational modifications on PNAds required for L-selectin recognition were recently reviewed (Rosen, 2004). The ability of L-selectin expressed on the circulating lymphocyte to recognize carbohydrate modifications on the PNAds expressed on the HEV is due to the activity of a number of important glycosyltransferases. This section outlines those key enzymes involved in forming the 6-sulfo-sLex epitope (Figure 1-4). Refer to Figure 1-8 as a summary of the glycosyltransferases involved in the formation of this epitope. 57 Figure 1-8: Glycosyltransferases involved in creating common selectin ligands. A, Along the centre of the figure, the scheme for synthesis of core 2 O-glycans associated with formation of the functional sLex on selectin ligands (PSGL-1) is illustrated. Its synthesis begins with GalNAc modification of selected serine (S) and threonine (T) residues by one or more of at least nine distinct ppGalNAcTs. S- or T-linked GalNAc can serve as an acceptor substrate for a core 1 p3GalT and GlcNAcTs. Core 2 O-glycans are then modified by one or more (34-GalTs, and are then directly a-2,3-sialylated by ST3GalTs. The resulting a-2,3-sialylated glycans are fucosylated by FucT-IV and -VII. Enzymes implicated in the key posttranslational modifications of sulfo-sLex structures on PNAds for L-selectin binding on core 2 structures (B) and extended core 1 (C) branches are outlined. 58 C2GlcNAcTs involved in 6-sulfo-sLex epitope formation The in vivo functional relevance of core 2 O-glycans in L-selectin counter-receptor activity has been studied using C2GlcNAcT-i n u 1 1 mice. A slight reduction in the amount of L-selectin counter-receptors was observed on the peripheral node HEV of these animals (Ellies et al., 1998). However, lymphocyte homing is essentially normal in these mice, as assessed by short-term in vivo lymphocyte homing assays, and as inferred by the nearly normal number of endogenous lymphocytes in peripheral nodes, mesenteric nodes and Peyer's patches (Ellies et al., 1998). Sperandio et al, further investigated leukocyte rolling in venules of untreated and TNF-ct-treated cremaster muscles and in Peyer's patch HEV of C2GlcNAcT-In u" (Sperandio et al., 2001a). In the presence of blocking mAbs against P- and E-selectin, L-selectin-mediated leukocyte rolling was almost completely abolished in cremaster muscle venules. By contrast, leukocyte rolling in Peyer's patch HEV does not differ significantly between C2GlcNAcT-Fu I 1 and control mice. Since almost normal L-selectin ligand activity is preserved in C2GlcNAcT-In u 1 1 mice, many studies strive to identify the L-selectin binding epitope. They uncover a number of extended core 1 glycans presenting 6-sulfo sLex, to be responsible for the MECA-79 reactivity found in C2GlcNAcT-In u" mice (Figure 1-8), that is independent of core 2-branching activity (Yeh et al., 2001). The study by Yeh et al. helps determine the identity of the MECA-79 epitope as an extended core 1 structure with a 6-sulfo-sLex capping group. The studies suggest that C2GlcNAcT-I and p3GlcNAcT-III (Figure 1-8) can cooperate to produce a potent L-selectin ligand with biantennary oligosaccharides, 59 containing both core 2 branches and core 1 extensions, that exhibits a synergistic effect on binding to L-selectin (Yeh et al., 2001). P3GlcNAcTs involved in 6-sulfo-sLex epitope formation The formation of extended core 1 structures is the job of core 1 $1,3-N-acetylglucosaminyltransferase, p3GlcNAcT-3, which adds on a GlcNAc in a f3l ->3 fashion to Gal (Figure 1-8) (Mitoma et al., 2003; Yeh et al., 2001). P3G1CNACT-3 transcripts are expressed in the small intestine, colon, and placenta and are moderately expressed in the liver, kidney, pancreas and prostate. The studies also find that P3GlcNAcT-3 is expressed in MECA-79 + enriched HECs from peripheral lymph nodes (Yehetal., 2001). 6 GlcNAc6STs involved in 6-sulfo-sLex epitope formation GlcNAc-6-O-sulfotransferases (GlcNAc6STs) provide one of the most important modifications involved in defining L-selectin ligands (Fukuda et al., 2001; Rosen, 2004). Efforts to identify the enzymes that catalyze sulphation at the C-6 position of GlcNAc are many. At least six members of this family are cloned and characterized (Hemmerich and Rosen, 2000). Special attention is drawn to HEC-GlcNAc6ST (LSST, GST-3 and GlcNAc6ST-II) because of its restricted expression in HEV of lymph nodes and tonsils (Bistrup et al., 1999; Hiraoka et al., 2004; Hiraoka et al , 1999). Previous studies demonstrate that HEC-GlcNAc6ST forms 6-sulfo-sLex on both core 2 branches and MECA-79 + extended core 1 O-glycans (Figure 1-8) (Hiraoka et al., 1999). 60 HEC-GlcNAc6ST n u U mice have been generated (Hemmerich et al., 2001; Hiraoka et al , 2004). Hemmerich et al. observed loss of L-selectin binding of HEV, loss of lymphocyte binding in vitro, and reduced in vivo lymphocyte homing. MECA 79 staining of lymph node HEV is also lost (Hemmerich et al , 2001). Hiraoka et al. demonstrate partial impairment of lymphocyte homing to peripheral lymph nodes and reduction in lymphocyte counts in the peripheral lymph nodes, despite the fact that L-selectin ligands that contain 6-sulfo-sLex are reduced at HEV. By contrast, they found that mice deficient in both HEC-GlcNAc6ST and C2GlcNAcT-I exhibit significantly reduced lymphocyte homing and reduced lymphocyte counts as a result of significantly decreased 6-sulfo-sLex due to the presence of only extended core 1 modified 6-sulfo-sLex structures and the loss of expression of core 2-modified structures on HEV L-selectin ligands, relative to single knockout strains. This suggests that both HEC-GlcNAc6ST and C2GlcNAcT-I cooperate in the synthesis of biantenary 6-sulfo-sLex structures on HEV-specific L-selectin ligands required for lymphocyte homing (Hiraoka et al., 2004). Recently, the importance of these enzymes was further qualified in an elegant study in which Gauguet et al. expose distinct functions for C2GlcNAcT-I and HEC-GlcNAc6ST in controlling the "type" of lymphocyte (B cell or T cell) homing to peripheral lymph nodes. T cells which express higher levels of L-selectin than B cells are less sensitive to the loss of these enzymes than B cells; this allows these enzymes to set a threshold for the amount of L-selectin necessary for lymphocyte homing (Gauguet et al., 2004). Another interesting GlcNAc6ST is GlcNAc6ST-I (GST-2) (Uchimura et al., 2004; Uchimura et al., 1998). This enzyme is co-expressed with HEC-GlcNAc6ST in lymph node HEV, but extends to other tissues as well. It is shown in vitro to form the MECA-79 epitope, leading to L-selectin binding. Analysis of GlcNAc6ST-Inu11 mice demonstrates that GlcNAc6ST-I is responsible for some L-selectin ligand activity and 61 lymphocyte homing in vivo (Uchimura et al., 2004). GlcNAc6ST-I and HEV-GlcNAc6ST play compensatory roles in L-selectin ligand formation in peripheral lymph nodes, but GlcNAc6ST-I is the main enzyme in lymphocyte homing to Peyer's patches where HEC-GlcNAc6ST is not expressed (Uchimura et al., 2004). These studies further confirm that GlcNAc6ST-I is responsible for some of the residual lymphocyte homing found in HEC-GlcNAc6ST n u 1 1 mice (van Zante et a l , 2003). „ IK pi,4-GalTs involved in 6-sulfo-sLe* epitope formation ~r p4GalT-IV is the P1,4-GalT believed to be involved in forming functional 6-sulpho-sLex (Asano et al., 2003; Seko et al., 2003). This enzyme transfers Gal from UDP-Gal to the terminal GlcNAc of carbohydrate chains in a p 1,4-likage, to form a Galpl->4GlcNAc structure (Figure 1-8). Seko etal. find that, of the seven p4GalTs, p4GalT-IV is the only enzyme to efficiently assist in forming the 6-sulfo-sLex structure (Seko et al., 2003). They assess that this enzyme is expressed ubiquitously, with relatively high expression in kidney, placenta, prostate, stomach, thyroid, tongue, trachea and lymph node. p4GalT-IV n u 1 1 mice will further clarify the role of p4GalT-IV in formation of 6-sulfo-sLex. ^ oc3 FucTs and ST3GalTs involved in 6-sulfo-sLex eptitope formation To further identify the enzymes involved in forming the 6-sulfo-sLex-capping group of the MECA-79 epitope, fucosyltransferases and sialyltransferases are explored. Many studies conclude that FucT-VH is the key fucosyltransferase involved (Homeister et a l , 2001; Maly et a l , 1996; Smith et al., 1996). FucT-VH n u i l mice show up to a 90% reduction in lymphocyte homing to peripheral lymph nodes. The FucT-IV enzyme contributes to the residual homing found in FucT-VIInu mice. Although FucT-IVnu" mice do not show a defect in lymphocyte homing, mice doubly deficient in FucT-IV and FucT-VII possess a much stronger defect in lymphocyte homing than FucT-VII alone (Homeister et al., 2001; Smithson et al, 2001). The phenotype observed in the double-deficient mice is analogous to that of L-selectinnu11 mice (Arbones et al., 1994) evident in extreme leukocytosis, characterized by decreased neutrophil turnover, increased neutrophil production and extremely small peripheral lymph nodes. Studies exploring which oc2,3-sialyltransferases are implicated in the elaboration of L-selectin ligands are less involved. There is speculation that the combined activities of a number of sialyltransferases (ST3GalT) help form functional L-selectin ligands. This is obvious for E- and P-selectin ligands since mice genetically deficient in ST3GalT-I, ST3GalT-II, ST3GalT-III and ST3GalT-IV do not display defects in L-selectin ligand formation (Ellies et al., 2002; Rosen, 2004). 1.5.2 Glycosyltransferases involved in the formation of P-selectin ligands P-selectin ligand activity on leukocytes is assigned, primarily, to PSGL-1, based on reconstitution of P-selectin ligand binding in transfected cell lines (Li et al., 1996; Sako et al., 1993) and on analysis of PSGL-l n u 1 ' mice (Vestweber and Blanks, 1999; Xia et al., 2002a; Yang et al., 1999). It is dictated in large measure by the nature and abundance of a specific type of sLex-bearing O-glycan that decorates this glycoprotein (Martinez et al., 2005). PSGL-1 binds via its O-linked carbohydrate side chains, generated by C2GlcNAcT (Kumar et a l , 1996; Li et al., 1996; Wilkins et al., 1996). The identities of glycosyltransferases that control assembly of these glycans include: C2GlcNAcT-I, GalT-1, ST3GalT-IV, FucT-IV and FucT-VII as well as tyrosine 63 sulphotransferases. P-selectin recognition of PSGL-1 is highly specific and dependent on only a single C2GlcNAcT O-branched glycans (Thrl6 in human (Leppanen et al., 2000; Leppanen et al., 2003) and Thrl7 in mouse (Xia et al., 2003)) that display the sLex epitope. It also depends on sulphation of one or more of the two (mouse (Xia et al., 2003)) or three (human (Leppanen et al., 2000; Leppanen et al., 2003)) tyrosine residues near the amino terminus of PSGL-1 (Ouyang et al., 1998; Pouyani and Seed, 1995; Wilkins et al., 1995) (Figure 1-7) (Leppanen et al., 2000; Leppanen et al., 2003). L-selectin recognition of PSGL-1 is very similar to P-selectin, in that the O-glycosylation of Thr is required, as well as sulphation of nearby tyrosine residues (Bernimoulin et al., 2003). Figure 1-8 outlines the enzymes involved in forming the sLex capping structure on leukocytes expressing PSGL-1 recognized by P-selectin. mono-fucosylated core 2 O-glycan tri-fucosylated core 2 O-glycan on extended polylactosamine 64 C2GIcNAcTs and functional P-selectin ligands C2GlcNAcT isoenzymes (C2GlcNAcT-I, -II, -III) create the core 2 O-glycan branch (Schachter and Brockhausen, 1989) by adding GlcNAc to the Gal fj 1-3 GalNAc core 1 structure expressed on serine or threonine residues, depicted in Figure 1-7 and Figure 1-8. Capping of the core 2 branch with a sLex moiety creates the P-selectin counter receptor. Mice with deficiencies in C2GlcNAcT-II or C2GlcNAcT-III have yet to be reported. Analyses of C2GlcNAcT-In u 1 1 mice, however, show that it is the isoenzyme essential for PSGL-1 modifications on neutrophils (Ellies et al., 1998; Snapp et al., 2001; Sperandio et al , 2001b), eosinophils (Broide et al., 2002; Symon et al., 1996; Wein et al., 1995) and CD4 + T cells (Snapp et al., 2001) which is consistent with in vitro studies (Li et al , 1996; Sako et al., 1993). Despite the essential role of C2GlcNAcT-I in P-selectin ligand formation, C2GlcNAcT-I n u l 1 mice possess a relatively mild phenotype, showing only a partial reduction in selectin ligands and essentially no effect on lymphocyte homing (Ellies et al., 1998), reflecting the possible redundancy of the C2GlcNAcT isoenzymes. Whether C2GlcNAcT-II and/or C2GlcNAcT-III contribute to selectin ligand formation and cell trafficking is not fully determined. Recently, however, it was argued that C2GlcNAcT-III may be involved in PSGL-1 modifications in vitro and in vivo (Merzaban et al., 2005). The importance of these enzymes as control points in selectin ligand formation is discussed in more detail in section 1.6. 65 pMGalTs and functional P-selectin ligands Core 2 O-glycan branches are extended by the actions of |34GalT and pl,3-iV-acetylgulcosaminyltransferase, which act alternatively to form JV-acetyl lactosamine repeats on core 2 O-glycans. In collaboration with ST3GalTs and FucTs, p*4GalT synthesizes the sLex epitope. |34GalT is involved in biosynthesis of biologically important, galactose-containing oligosaccharides, such as sLe" and the 6-sulfo-sLex, discussed above. While seven pMGalT genes are isolated, their individual roles in selectin-ligand biosynthesis remain to be determined. p4GalT-Inu11 mice, which have been generated (Asano et al., 1997), exhibit growth retardation and semi-lethality before weaning because of augmented proliferation and abnormal differentiation of small intestine epithelial cells. In surviving mice, more than 80% of the core 2 O-glycans on the leukocyte membrane glycoproteins lacked galactose residues in (3-1,4 linkages. Furthermore, soluble P-selectin binding to neutrophils and monocytes of these mice are significantly reduced, indicating an impairment of selectin-ligand biosynthesis (Asano et al., 2003). These mice also exhibit blood leukocytosis. Acute and chronic inflammatory responses, including the contact hypersensitivity and delayed-type hypersensitivity responses, are suppressed in these mice, while neutrophil infiltration into inflammatory sites is largely reduced (Asano et al., 2003). These results assign to P4GalT-I a prominent role in the synthesis of O-linked glycans relevant in the formation P-selectin ligand activity in leukocytes. 66 ST3GaITs and functional P-selectin ligands oc2,3-Sialic acid linkages that form the sLex epitope are critical for P-selectin ligand recognition (Leppanen et al , 2000). At least six ST3GalTs are identified to date, and three or more of these sialyltransferases are known to sialylate the type II oligosaccharides (Gal(31,4GlcNAc) characteristic of the sLex structure in vitro: ST3Gal-III, -IV, and VI (Okajima et al , 1999; Sasaki et al., 1993). Further in vitro analysis of substrate specificity identified that ST3Gal-I, -II, and -III have preference for substrates other than the type II; ST3Gal-I and -II prefer type III oligosaccharides (Galpl,3GalNAc) (Fukuda et al, 1985; Kono et al., 1998), whereas ST3Gal-III prefers the type I oligosaccharides (Gaipi,3GlcNAc) present in sLea structures (Wen et al., 1992)(Figure 1 -4). To determine the effect of lack of these enzymes on selectin ligand activity in vivo, mice with deficiencies in ST3GalTs (ST3Gal-I, -II, -III, -IV) have been generated and studied (Ellies et al., 2002; Priatel et al., 2000). The leukocytosis phenotype typical of mice with defects in selectin ligand formation is not observed in these mice. Only ST3Gal-IV affected the synthesis of selectin ligands in neutrophils, as measured by a decrease in rolling on P-selectin in in vitro flow chamber assays, but not in TNF-a treated cremaster venules in vivo. ST3Gal-VI most closely mimics ST3Gal-IV in its substrate specificity (Kono et al., 1997; Kono et al., 1998; Okajima et al, 1999; Sasaki et al , 1993) and may account for the residual sialidase-dependent selectin ligand activity in ST3Gal-IVnu11 neutrophils. 67 These studies suggest that multiple ST3GalTs contribute to selectin ligand synthesis and indicate a high degree of functional specificity among sialyltransferases, as well as a substantial role for ST3Gal-IV in P-selectin ligand formation (Ellies et al , 2002). FucTs and functional P-selectin ligands Core 2 O-glycans with terminal oc-2,3-linked sialic acid modifications are substrates for a-l,3-fucosyltransferases that can modify one or more GlcNAc residues (Figure 1-7). A requirement for cd,3 fucosylation was first determined through analysis of cultured cells with a genetically modified fucosylation phenotype (Becker and Lowe, 2003; Lowe, 2002; Vestweber and Blanks, 1999). A requirement for fucosylation in vivo is identified, in part, through analysis of mice with genetic deficiency in the synthesis of GDP-fucose, the substrate for fucosyltransferase (FX n u" mice) (Becker et al., 2003; Smith et al., 2002). The FX protein is a ubiquitously expressed, epimerase-reductase enzyme necessary for the de novo biosynthesis of GDP-fucose from GDP-mannose (Becker and Lowe, 2003). Cellular glycoconjugates from these mice do not display fucose and these mice exhibit many abnormalities including growth retardation, infertility, neutrophilia and altered myelopoiesis, diarrhea, and partial embryonic lethality that is strain specific (Becker et al., 2003; Smith et al., 2002). In these F X n u " mice, neutrophils lack both E-and P-selectin ligand activity (Smith et al., 2002). But feeding the mice fucose reverses this phenotype by restoring GDP-fucose synthesis via a salvage pathway (Becker and 68 Lowe, 2003). In a human disease, leukocyte adhesion deficiency II syndrome (LAD II) patients suffer from recurrent infections due to defective selectin ligand biosythesis, mental retardation and skeletal abnormalities (Becker and Lowe, 1999). LAD II is caused by a mutation in a gene encoding a multispanning transmembrane protein that transports GDP-fucose from its cytosolic site of synthesis to the lumen of the Golgi (Luhn et a l , 2001). Once there, it is a substrate for many enzymes responsible for adding fucose to proteins, including O-fucose modifications of Notch, formation of sLex structures of selectin ligands and others. Like the FX n u 1 1 mice, neutrophils from these patients also do not have functional P- and E-selectin ligands, although the disease is corrected in some patients by taking fucose orally (Becker and Lowe, 2003). These studies and others help identify the importance of fucose modifications in selectin ligand synthesis. There are eight candidate FucTs involved in forming functional selectin ligands (reviewed in (Becker and Lowe, 2003; Lowe, 2003)), including six human forms and at least three murine forms cloned and confirmed for FucT activity: FucT-III, -IV, -V, -VI, -VII and -FX in human and FucT-IV, -VII and IX in mouse (Becker and Lowe, 2003). Analysis of mice with targeted mutations in fucosyltransferase genes reveals that only two of these enzymes are required to form functional selectin counter-receptor activity. These studies all support the observation that FucT-VII and FucT-IV co-operate in the synthesis of fucosylated glycans required for leukocyte P-selectin (as well as E- and L-selectin) ligand activity. In mouse, FucT-VII's contribution is dominant to that of FucT-IV. Analysis of FucT-VIInu11 mice reveals that FucT-VII is necessary for P-selectin ligand activity of neutrophils and monocytes in vitro and in vivo (Homeister et al., 2001; Maly et al., 1996; Smithson et al., 2001). FucT-VIInu11 mice have leukocytosis and their blood leukocytes do not have P- and E-selectin ligand activity, determined by flow cytometric studies and by in vitro flow chamber assays (Maly et al., 1996). These mice also have compromised neutrophil trafficking and extravasation in in vivo models of inflammation (thioglycollate-induced peritonitis model) (Maly et al., 1996), but residual recruitment is still evident as measured by flow chamber assays with immobilized P- and E-selectin (Homeister et al , 2001). This residual activity is due the presence of FucT-IV activity in the absence of FucT-VII (Homeister et al., 2001). In a contact hypersensitivity model (CHS), FucT-VIInu11 mice have a significantly reduced response (Homeister et al., 2001; Smithson et al., 2001): Studies in Lowe's lab examined the contributions of T H I and Tel cells, as well as neutrophils, to the CHS response in mice deficient in FucT-IV, FucT-VII or both enzymes. These studies illustrated that the CHS response is significantly decreased in FucT-VIInu11 mice, and is absent in mice deficient in both enzymes. Only a subtle decrease in the CHS response is assessed in FucT-IVn u U mice, suggesting that the residual response seen in the absence of FucT-VII is due to the presence of FucT-IV. Analysis of FucT-IVnu11 mice demonstrates that neutrophil recruitment to inflamed vascular beds is normal (Homeister et al., 2001; Smithson et al., 2001; Weninger et al., 2000), as is P-selectin (and E-selectin) ligand activity. However, the neutrophils do roll at higher velocities (Weninger et al., 2000) reflecting that subtle defects in selectin ligand formation occur in the absence of FucT-IV. Over all, these observations support the hypothesis that in mice FucT-VII dependent fucosylation of selectin ligands accounts for most of the selectin-dependent adhesion. 70 Tyrosylprotein sulphotransferase (TPST) and functional P-selectin ligands Protein tyrosine sulfotransferases have been purified and cloned. Tyrosine sulphation is mediated by one of two isoenzymes, TPST-1 or TPST-2 (Beisswanger et a l , 1998; Ouyang et al., 1998). These enzymes are Golgi-localized and catalyze the transfer of sulphate to tyrosine residues within acidic motifs of polypeptides. Sulphation of tyrosine residues in PSGL-1 is essential for binding to P-selectin. Ouyang et al, describe the purification of these tyrosine sulphotransferases in rat liver microsomes, based on their affinity for the N-terminal 15 amino acids of PSGL-1. They isolated human and mouse cDNAs that predict type II transmembrane proteins comprise 370 amino acids, with almost identical primary structure between human and mouse. The human cDNA encodes a fully functional N-glycosylated enzyme with an apparent molecular mass of approximately 54 kDa, expressed in mammalian cells (Beisswanger et al., 1998; Ouyang et al., 1998). Northern analysis indicated that both TPSTs are expressed in many tissues (Beisswanger et al., 1998; Ouyang et al., 1998). This group also produced TPST-1-deficient mice that were found to have reduced body mass and increased postimplantation fetal death (Ouyang et al., 2002). In order to help answer some fundamental questions concerning the involvement of TPST enzymes in ontogeny and regulation, a zebrafish model was developed (Mishiro et al., 2004). Mishiro et al. cloned, expressed and characterized the TPST enzymes present in zebrafish. In their study, they find that recombinant zebrafish TPST exhibits some properties that are similar to those previously determined for mammalian TPSTs (Beisswanger et al , 1998; Ouyang et al., 1998). Further studies concerning the ontogeny, regulation and physiological involvement of the zebrafish TPST are currently under way. 71 1.5.3 Glycosyltransferases involved in the formation of E-selectin ligands E-selectin ligand activity, unlike P-selectin, is assigned to a broader range of polypeptides that include, but are not limited to, ESL-1 and PSGL-1. As outlined above, many studies demonstrate PSGL-1 binding to E-selectin (section The structural requirements necessary for recognition of a ligand by E-selectin are different from those necessary for P- and L-selectin. Sulphation, essential for both L- and P-selectin ligands, is not required for E-selectin ligands (Berg et al., 1.991; Goetz et al., 1997; Li et al., 1996; Picker et al., 1991a; Sako et al., 1995; Snapp et al., 1998b). PSGL-1 on activated T cells - which comprise less than 5% of T cells in extracutaneous inflammatory sites, and 80 to 90% of T cells in skin inflammatory sites (Borges et al., 1997b; Fuhlbrigge et al., 1997) -is the major glycoprotein carrier for a carbohydrate epitope called cutaneous lymphoid antigen, CLA. This epitope is relevant only to lymphocyte binding to E-selectin, not P-selectin (Berg et al., 1991; Borges et al., 1997c; Tu et al., 1996). CLA is defined by reactivity with the mAb HECA-452 and is correlated to E-selectin. The HECA-452 epitope blocks lymphocyte binding to E-selectin, as well as rolling of human neutrophils in inflamed skin vessels (Berg et al., 1991; Vestweber and Blanks, 1999; von Andrian et al., 1993). HECA-452 recognizes the tetrasaccharide moieties sLex and sLea (Figure 1-4), however, the exact nature of the carbohydrate epitope that is recognized by this antibody on T cells localized in the skin is still unknown, since these cells are mostly negative for sLex and sLea (Levinovitz et al., 1993; Steegmaier et al., 1995). These studies suggest that, in addition to the sLex structure, there is another component that mediates E-selectin binding. PSGL-1 binding to E-selectin also depends on oc2,3 sialic acid and ocl,3 fucose linkages on the terminal sialofucosylations recognized by HECA-452. 72 The specifications for E-selectin binding to ESL-1 are slightly different from those for PSGL-1. The binding of ESL-1 to E-selectin is Ca2+-dependent but, unlike PSGL-1 binding to P-selectin, does not require disulphide linkages (Levinovitz et al., 1993). ESL-1 binding to E-selectin is sialic acid dependent since sialidase treatment drastically reduces binding by over 80% (Levinovitz et al., 1993). ESL-1 contains five putative N-glycans and studies show that ligand binding to E-selectin is dependent on N-glycans and not on O-glycans (Huang et al., 2000; Steegmaier et al , 1997; Steegmaier et al., 1995). Binding also depends on a(l,3)-fucosylation (Li et al., 1996; Steegmaier et al., 1995; Zollner and Vestweber, 1996). Recently CD44 expressed on neutrophils was suggested as a ligand for E-selectin. Ligand binding is mediated by sialylated, ccl,3-fucosylated N-glycans on CD44 (Katayama et al , 2005). L-selectin is also a candidate ligand for E-selectin (Green et al., 2004; Picker et al., 1991b) and again requires sLex structures for recognition. The enzymes responsible for forming these E-selectin ligand glycans on PSGL-1, ESL-1 and other E-selectin ligands will be outlined below. For the most part, they are similar to those responsible for both L-selectin and P-selectin ligand formation in that they require sLex structures. Unlike L- and P-selectin ligands, though, the sLex cap can occur on a number of different glycans, from N-glycans to glycolipids. In addition, analysis ofFX n u 1 1 mice by Lowe's group suggest that E-selectin ligands are much more sensitive to loss of fucose than P-selectin ligands, further identifying the importance of fucosylation to the formation of functional E-selectin ligands (Smith et al, 2002). 73 C2GlcNAcTs and functional E-selectin ligands As discussed above, core 2 branching on O-glycans is essential for P-selectin ligand formation; however, this is not the case for E-selectin ligands. A number of studies support that cells deficient in C2GlcNAcT-I are able to bind to E-selectin, although at slightly reduced levels compared with wild-type cells (Ellies et al., 1998; Huang et al., 2000; Merzaban et al., 2005). These studies suggest that 75% to 80% of E-selectin binding is maintained in the absence of C2GlcNAcT-I. However, there is a component of binding that is likely dependent on C2GlcNAcT-I branching on PSGL-1 and not due to branching on alternate E-selectin ligands, such as ESL-1 (Huang et al., 2000; Merzaban et al., 2005). ST3GalTs and E-selectin ligands oc2,3 sialic acid linkages are important for E-selectin ligand formation on all E-selectin ligands. ST3Gal-IV and ST3Gal-VI are implicated in E-selectin ligand formation (Ellies et al., 2002; Okajima et al, 1999). Analysis of ST3Gal-IVnu" mice reveals that ST3Gal-IV does not contribute to functional selectin ligands on PSGL-1 in vivo, but E-selectin-dependent leukocyte rolling velocity increases indicating a role for ST3Gal-IV in E-selectin ligand formation in vivo (Ellies et al., 2002). Studies have reported that E-selectin is required for slow leukocyte rolling on the vessel wall in the TNF-a-treated cremaster model of inflammation (Kunkel and Ley, 1996), suggesting that ST3Gal-IV is critical in forming E-selectin ligands necessary for reducing the rolling velocity of circulating leukocytes as they enter sites of inflammation. Analysis of these mice also implies that other sialyltransferases are involved in both E- and P-selectin ligand formation (Ellies et al., 2002). 74 FucTs and E-selectin ligand formation a2,3-rucosylation is imperative for E-selectin ligand activity (Katayama et al., 2005; Smith et al., 2002; Snapp et al., 1997; Steegmaier et al., 1995). Examination of ESL-1 and PSGL-1 in neutrophils of mice deficient in FucT-IV and FucT-VII illustrates that FucT-VII exclusively directs expression of PSGL-1 glycoforms, leading to high-affinity binding to P-selectin, whereas FucT-IV preferentially directs expression of ESL-1 glycoforms, leading to high-affinity binding to E-selectin (Martinez et al., 2005). Another group concurs with these findings, observing that E-selectin binds with normal efficiency to ESL-1 on Fuc-TVIInu11 neutrophils, but exhibits an 80% reduction in its ability to bind ESL-1 isolated from Fuc-TIVnu" neutrophils (Huang et al., 2000). The effect of FucT-rV (Weninger et al., 2000) deficiency on E-selectin mediated leukocyte interactions is remarkably similar to that reported for ST3Gal-IV (Ellies et al., 2002) above. The leukocyte rolling flux fraction and firm adhesion resulting in transmigration was normal in these mice, however, an increase in E-selectin-dependent rolling velocity is observed in both strains, suggesting that these two glycosyltransferases may collaborate to form E-selectin ligands important for slow rolling of leukocytes. These studies assign FucT-IV and FucT-VII important and specific functions in the creation of functional E-selectin ligands. 75 1.6 A closer look at Core O glycan branching 1.6.1 CD43 and the discovery of C2GlcNAcT CD43 is a highly abundant and heavily glycosylated transmembrane protein expressed on lymphohemopoietic cells. It is a representative of the mucin-type O-glycoproteins, a family of glycoproteins where GlcNAc is linked to serine or threonine residues (section 1.3.6). CD43 consists of a 229 (mouse)- or 234 (human)- amino-acid-residue extracellular domain that is heavily O-glycosylated, carrying an estimated 70 to 75 O-glycans (Fukuda, 1991). CD43 has multiple and complex functions CD43, with multiple and complex functions, is the subject of much debate. Anti-CD43 stimulation is implicated in activation of T cells (Axelsson et al., 1988; Mentzer et al., 1987; Sperling et al , 1995), monocytes (Nong et al , 1989) and neutrophils (Kuijpers et al., 1992). Several reports suggest that CD43 may play a role in regulating apoptosis of leukocytes (Bazil et al., 1996; Bazil et al., 1995; Brown et al., 1996; Dragone et al., 1995; Onami et al., 2002; Ostberg et al., 1997; Priatel et al., 2000; Todeschini et al., 2002). Upon T cell activation, CD43 is excluded from the immunological synapse by an ezrin-radixin-moesin-mediated mechanism (Allenspach et al., 2001; Delon et al., 2001; Roumier et al., 2001; Serrador et al , 1998), and is not part of the T cell receptor complex (Sperling et al., 1998). CD43 is also implicated in the control of cell-cell interactions, although its precise role in adhesion is under debate. It is proposed to have a dual function in controlling cell-cell interactions: an adhesive role, based on data that there may be CD43-specific ligands; and an anti-adhesive role, based on evidence that CD43 acts as a barrier molecule (Drew et al., 2005; Ostberg et al., 1998). CD43 is reported to bind to ICAM-1 (Rosenstein et al., 1991), galectin-1 (Baum et al., 1995) and to siglec-1 76 (van den Berg et al., 2001) but the physiological relevance for these interactions have not yet been demonstrated. In contrast, Ardman et al. (Ardman et al., 1992) demonstrate that CD43 transfected into HeLa cells results in the inhibition of adhesion and, conversely, that disruption of the CD43 gene in a T cell line results in enhanced adhesion to fibronectin and gpl20 (Manjunath et al., 1993). Another recent study from the McNagny lab supports these findings, showing that cultured CD43 n u l 1 mast cells display greatly increased homotypic aggregation compared with wild-type cells (Drew et al., 2005). Soluble CD43, thought to be shed from the surface of hemopoietic cells, occurs in human plasma at concentrations up to 10 pg per millilitre. Soluble CD43 carries hexasaccharide O-glycans that are typically located on all myeloid cells, immature B cells and activated T cells (Schmid et al., 1992). What controls CD43 shedding is unclear, but sheddases and several proteases are implicated. Bazil and Strominger show that crosslinking CD43 using anti-CD43 mAbs stimulates a serine and/or metalloprotease-catalyzed release of CD43 from both human granulocytes and lymphocytes (Bazil and Strominger, 1993; Bazil and Strominger, 1994). C-terminal analysis of soluble CD43, isolated from human plasma, uncovered a number of terminal amino acid residues: Leu 193 or Leu 209, Phe 226, and Asp 228 or Asp 232 (Schmid et al., 1992). This suggests either that several proteases mediate CD43 shedding or that endopeptidases present in plasma are responsible for this heterogeneity. The physiological role of shedding is unknown although it is suggested that the release of a heavily sialylated glycoprotein might reduce cell surface negative charges and/or act as a steric barrier facilitating cell-cell contact (Remold and Parent, 1994; Rieu et al., 1992). Loss of CD43 due to shedding may also unmask cell-surface receptors required for a specific function (for instance, in adhesion) (Bazil and Strominger, 1993; Bazil and Strominger, 1994). 77 Genetic deletion of leukocyte mucins generally results in non-lethal phenotypes evident in the lymphohemopoietic system. CD43 n u" mice have a surprisingly mild phenotype, with normal T cell ontogeny and normal hemopoiesis (Carlow et al., 2001b; Manjunath et al., 1995). Cells from CD43n u" mice have a hyperadhesive phenotype, confirming that CD43 acts as a "barrier molecule" (Drew et al., 2005; Manjunath et al., 1995; Stockton et al., 1998). Using intravital microscopy in CD43 n u 1 1 mice, Woodman and colleagues directly demonstrate the dual function of CD43 (Woodman et al., 1998). The absence of CD43 on leukocytes leads to enhanced rolling on cremaster venules, but failure to extravasate and infiltrate inflamed tissue, implicating CD43 in tissue emigration. CD43 n u" mice were originally described as having a hyperproliferative T cell phenotype (Manjunath and Ardman, 1995), however, further studies in the Ziltener lab unveil that CD43 n u 1 1 mice backcrossed onto the C57BL/6 background lose this hyper-responsiveness (Carlow et al., 2001b). This difference in phenotype is likely due to the 129 genetic background of the CD43 n u 1 1 mice originally analyzed (Carlow et al., 2001b). A very intriguing property of CD43 involves its role in mycobacteria infection of macrophages. Remold's laboratory demonstrates that efficient infection of macrophages by mycobacteria requires CD43. Macrophages derived from CD43 n u 1 1 mice do not bind M. avium, M. bovis or M. tuberculosis, while macrophages from wild-type mice do (Fratazzi et al , 2000). Interestingly, adding soluble human CD43, purified from human plasma (Schmid et al., 1992), is sufficient to restore mycobacterium infectivity of murine macrophages. Studies conducted in a collaboration between the Stokes and Ziltener labs extend these findings (Randhawa et al., 2005). Again, macrophages obtained from CD43 n u l 1 mice bind less M. tuberculosis than macrophages from wild-type mice by 40% to 60%, confirming and extending data from the Remold lab. In addition, these studies reveal that there is a gene dose effect for CD43-mediated binding of M. tuberculosis; 78 CD43 macrophages (which express half the levels of CD43 compared with CD43 cells) bind M. tuberculosis at levels intermediate between those of CD43 + / + and CD43V" macrophages (Randhawa et al., 2005). By analyzing distinct populations of macrophages (splenic, bone-marrow-derived, peritoneal and alveolar), differences in the ability of wild-type macrophages to bind the bacterium is observed and correlates with the level of CD43 expressed on each macrophage type. Although the absence of CD43 prevents entry of M. tuberculosis into macrophages, the survival and replication of M. tuberculosis is significantly enhanced within CD43 n u U macrophages; this is also apparent in vivo using CD43 n u 1 1 mice that are aerogrenically infected with M. tuberculosis. In vivo studies show that, in absence of CD43, there is increased bacterial load during both the acute and chronic stages of infection, and a more rapid onset of pathologies associated with the infection. This study establishes that CD43 is not just involved in the binding of M. tuberculosis, but also in the uptake and growth of the bacterium in both ex vivo murine macrophages as well as in vivo. C2GlcNAcTs control CD43 Glycosylation The considerable MW heterogeneity observed for CD43 is due solely to differential glycosylation of the O-linked glycans attached to the CD43 polypeptide backbone (Cyster et al , 1990). Two major forms of O-glycans are found attached to CD43, depending on the cellular source of the mucin: core 2-branched hexasaccharide or core 1 tetrasaccharide (Figure 1-9) (Fukuda, 1991). Two CD43 glycoforms with different molecular masses have been described. One glycoform, associated with neutrophils, platelets and activated T cells, has a MW of 125-130kD and carries predominantly core 2-branched hexasaccharide (Figure 1-9; structure II) O-glycans attached to the CD43 79 polypeptide backbone. The other glycoform, found on resting T cells, has a MW of 115kD and carries mainly core 1 tetrasaccharide (Figure 1-9; structure I) O-glycans Figure 1-9: Structure and function of C2GlcNAcTs Transmembrane topology and a representative catalytic scheme of a mammalian glycosyltransferase are illustrated. The glycosyltransferase used in this example is a C2GlcNAcT. The larger, carboxy-terminal (COOH) catalytic domain resides in the lumen of the Golgi apparatus, and is tethered to the Golgi membrane by a stem domain, a transmembrane segment and a short cytosolic domain. Proteolytic cleavage events can liberate catalytically active fragments (not shown). C2GlcNAcT activity results in core 2 branches by addition of GlcNAc and the liberation of GDP. Subsequent GalT and ST3GalT activity results in a hexasaccharide structure (I). When C2GlcNAcT is not present then ST3GalT activity of the core 1 substrate yields a tetrasaccharide structure (II). (Baecher-Allan et al., 1993; Carlsson and Fukuda, 1986). The switch to hexasaccharide O-glycan from tetrasaccharide O-glycan is due to the activity of C2GlcNacT-I (Barran et al., 1997; Ellies et al., 1994; Fukuda, 1991; Higgins et al., 1991; Jones et al., 1994; Piller et al., 1988) (Figure 1-9). MAb 1B11, reactive with core 2-branched hexasaccharide CD43 130kD (Jones et al., 1994), has emerged as an excellent tool in analyzing murine CD43 glycoforms and studying C 2 G l c N A c T - I enzyme activity (Barran et al., 1997; Carlow et a l , 1999; Carlow et al., 2001a; Carlow et a l , 2001b; Carlow et al., 2005; Ellies et al., 1994; Ellies et a l , 1996; Fellinger et al., 1998; Harrington et al., 2000; Jones et a l , 1994; Merzaban et al., 2005; Onami et al., 2002; Tsuboi and Fukuda, 1997). Analogous to m A b 1B11 in the murine system, mAb T305 is described to exclusively recognize the hexasaccharide glycoform of CD43 in the human system. In fact, C 2 G l c N A c T - I was cloned by screening a c D N A library for clones, inducing a glycosylation change in CD43, as measured by m A b T305 (Bierhuizen and Fukuda, 1992). Another glycosylation- sensitive CD43 antibody, m A b S7, binds to CD43 carrying core 1 branches and does not recognize CD43-carrying core 2 branches, signifying an absence of C 2 G l c N A c T - I activity (Baecher-Allan et al., 1993). Indeed, transfection o f CD43-positive cells lacking endogenous C 2 G l c N A c T activity, with C 2 G l c N A c T - I , reduces S7 and increases 1B11 reactivity (Barran et al., 1997)(Figure 1-10). — vector control transferred cells Figure 1-10: C2GlcNAcT-I modifies — C2GlcNAcT-l transferred cells CD43. — unstained cells E L 4 cells, a C2GlcNAcT-deficient cell line, were transfected with C 2 G l c N A c T - I (red line) or a control vector (blue line) and stained for CD43 glycoforms with mAbs S7 (A) and I B 11 (B) before analysis by F A C S . S7 1B11 81 Analysis of CD43 n u mice revealed a CD43-independent epitope recognized by mAb 1B11 identified as a glycoform of CD45RB exclusively expressed on resting peripheral CD8 + T cells, but not on CD4 + T cells or other tissues. 1B11 recognition of CD45RB on activated CD8 + T cells is lost due to increased core 2 branching (Carlow et al., 1999). Based on these results, 1B11 binding on T cells is used as a measure of core 2 enzyme activity as follows: 1) On T cells from wild-type (CD43+) mice, increased 1B11 binding (to CD43) indicates increased core 2 branching due to C2GlcNAcT activity (Figure 1-10); but 2) On CD8 + T cells from CD43 n u 1 1 mice, decreased 1B11 binding (to CD45RB) is a reliable indicator of increased core 2 branching due to C2GlcAcT activity. Thus, using mAb 1B11 helps to monitor the status of C2GlcNAcT activity in single cells via its reactivity with either CD43 or CD45 (Carlow et al., 1999; Carlow et al , 2001a; Merzaban et al., 2005). C2GlcNAcT-I mediated core 2 branching is recognized as a major control point in biosynthesis of branched O-glycans (Kuhns et al., 1993; Li et al., 1998; Yousefi et al., 1991). While CD43 was deduced early on to be a major substrate of this enzyme, C2GlcNAcT-I also modulates the glycosylation of CD44 (Barran et al., 1997), CD45 (Barran et al., 1997; Carlow et al., 1999), the phosphatase RPTPlcc (Barran et al., 1997) and PSGL-1 (Li et al., 1996), and it is likely that C2GlcNAcT-I alters O-glycan branching in general. 82 C2GlcNAcT Isoenzymes Core 2 O-glycan branches are essential in P selectin ligand formation (Ellies et al., 1998; Kumar et al., 1996; Sperandio et al., 2001a; Sperandio et al., 2001b). Given that both fucosylation and sialylation steps in selectin ligand synthesis are mainly dependent on pre-existing core 2 structures, C2GlcNAcT is a key control point in selectin ligand formation; study of this enzyme, therefore, is of particular interest with respect to P-selectin ligand formation and targeting effector cells to areas of inflammation (Vestweber and Blanks, 1999). C2GlcNAcT-I n u l 1 mice are viable and exhibit a surprisingly mild phenotype (Ellies et al., 1998). They have neutrophilia and show reduced binding of peripheral blood neutrophils to selectins, indicating a role of C2GlcNAcT-I in myeloid homeostasis and inflammation. Analyzing T cells from C2GlcNAcT-I n u l 1 mice suggests that alternate core 2 branching enzymes exist (Carlow et al., 1999; Ellies et a l , 1998). Indeed, biochemical analyses 20 years ago predicted the existence of multiple core 2 branching enzymes (Brockhausen et al., 1991; Brockhausen et al., 1985). Two additional core 2 isoenzymes are identified in human tissue, expressed and characterized: C2GlcNAcT-II (C2GlcNAcT-M) (Choi et al., 2004; Korekane et al , 2003; Ropp et al., 1991; Schwientek et al , 1999; Yeh et al., 1999) and C2GlcNAcT-III (C2GlcNAcT-T) (Schwientek et al., 2000). C2GlcNAcT-II was cloned by an EST-cloning strategy (Schwientek et al., 1999; Yeh et al., 1999). C2GlcNAcT-II has 48% identity with C2GlcNAcT-I and is unusual in its broad substrate specificity, catalyzing the formation of core 2 and core 4 branches (Figure 1-3). It is mainly expressed in digestive tract cells and is considered as important in forming core 2 and core 4 O-glycan structures on barrier mucins (Gendler and Spicer, 1995; Kawakubo et al., 2004; 83 Schwientek et al , 1999; Yeh et al, 1999). The third homologue of C2GlcNAcT, C2GlcNAcT-III, was identified using a tBLASTn analysis, with the coding sequence of C2GlcNAcT-II, to search human genome survey sequences. It has 42% identity with C2GlcNAcT-I and is strongly expressed in the thymus, possibly reflecting a unique role in T cell development (Schwientek et al., 2000). It is also expressed at lower levels in the small intestine, spleen and peripheral leukocytes. Unlike C2GlcNAcT-II, it has no additional core 4 activity (Schwientek et al., 2000) (Figure 1-3). C2GlcNAcT-I is ubiquitously expressed and appears to be the main enzyme responsible for core 2 branches in lymphohemopoietic cells (Ellies et al., 1998). C2GlcNAcT-I and C2GlcNAcT-III are co-expressed in thymocytes and peripheral leukocytes, while C2GlcNAcT-I and C2GlcNAcT-II are co-expressed in mucin-secreting cells. CHO cells co-transfected with human CD43, and either human C2GlcNAcT-I, human C2GlcNAcT-II or human C2GlcNAcT-III, express an epitope recognized by mAb T305, which is specific for human CD43-carrying core 2-branched O-glycans (Fox et al., 1983; Saitoh et al., 1991), suggesting that considerable redundancy exists between these enzymes. 1.6.2 Regulation of C2GlcNAcT activity Golgi organization of glycosyltransferases is important for control of glycosylation Glycosyltransferases present in the Golgi apparatus control O-glycan formation. Glycosyltransferases described so far in mammalian cells and yeast share structural motifs. They are type II transmembrane proteins with a short cytoplasmic tail (NH2-terminus), transmembrane domain, stem and catalytic domain (COOH-terminus) in the Golgi lumen (Figure 1-9) (Munro, 1998). Correct localization of glycosyltransferases within the Golgi apparatus is central in controlling oligosaccharide processing (Varki, 84 1998). Newly synthesized proteins move through the Golgi from the cis- to medial- to trans-Golgi, and finally to the trans-Golgi network (TGN), before being secreted or expressed on the cell surface. At each step, glycosyltransferases act upon these proteins and modify them with carbohydrates. Availability of acceptor substrates and nucleotide-activated sugars control the activities of the glycosyltransferases (Munro, 1998; Varki, 1998). As there is competition for the acceptor substrates, the location of glycosyltransferases within the Golgi is significant. For instance, C2GlcNAcT-I is localized to the cis to medial Golgi (Dalziel et al., 2001; Skrincosky et al., 1997; Zerfaoui et al., 2002) (but the localization of C2GlcNAcT-II and -III is not currently known). Core 2 branches added to core 1 structures on glycoproteins passing through these compartments reduce availability of core 1 structures to serve as substrates for oc2,3 sialyltransferase present in the trans-Golgi or TGN (Varki, 1998). A variety of factors are believed to influence Golgi localization, but no clear rules exist that outline which structures and residues restrict enzyme localization within the various Golgi compartments. The length of the transmembrane sequences, cytoplasmic sequences, stem sequences, overall protein domains and association with other proteins are shown in different examples as vital elements (Behnia et al., 2004; Colley, 1997; McCormick et al , 2000; Munro, 1998; Opat et al., 2000; Sasai et al., 2001; Setty et al., 2004). Additionally, studies speculate that substrate specificity is controlled by glycosyltransferases interacting with Golgi-associated proteins that may affect localization, specificity and/or activity (McCormick et al., 2000; Opat et al., 2000). C2GlcNAcT-I and C2GlcNAcT-III are co-expressed in thymocytes - a possible explanation for isoenzyme co-expression is that they differ in specificity. There is considerable homology associated with the catalytic domains (for instance, 64% between 85 mC2GlcNAcT-I and T-III), consistent with the fact that both enzymes catalyze the same core 2 branching reaction. However, homology in the cytoplasmic, transmembrane and stem domains (domains considered to be important in the control of enzyme specificity) is low (35%), indicating that the activity and/or specificity of these two enzymes may be differentially regulated. In fact, studies in our lab implicate both C2GlcNAcT-I and C2GlcNAcT-III in P-selectin ligand formation, but only C2GlcNAcT-I in E-selectin ligand formation (Merzaban et al., 2005), indicating that there may be differences in the fine specificities of these enzymes. Despite differences in the nucleotide, amino acid sequences, tissue distribution and the carbohydrate structures they form, all C2GlcNAcTs contain nine conserved cysteine residues (Beum and Cheng, 2001; Yang et al., 2003; Yen et al., 2003). An interesting recent study sets out to elucidate the structural determinants that distinguish the differences in substrate specificity among the C2GlcNAcTs by characterizing the disulphide linkages formed among the nine conserved cysteines in C2GlcNAcT-II (Singh et al , 2004). It finds that the structure formed differs from the structure proposed for mouse C2GlcNAcT-I (Coutinho et al , 2003; Yen et al., 2003), and may, as a result, account for the differences in substrate specificity. X-ray crystal structures await further description. Transcriptional and translational control of activity Many glycosyltransferases, including C2GlcNAcT-I, have multiple transcription initiation, and alternate splicing sites at the 5' UTR that are believed to control tissue specific expression of these enzymes (Falkenberg et al., 2003; Sekine et al., 1997). It is generally assumed inducing gene expression directs the activity of glycosyltransferases in activated T cells (Ley and Kansas, 2004). Furthermore, some evidence displays that N -86 glycosylation (Toki et al., 1997) and phosphorylation events control the activity of glycosyltransferases required for biosynthesizing gangliosides (Bieberich et al., 1998; Bieberich and Yu, 1999; Yu and Bieberich, 2001; Yuan, 1997). Phosphorylation events are shown to regulate glycosyltransferase activities in general (Yu and Bieberich, 2001), and to occur both on the short cytoplasmic N-terminal tail (Strous et al., 1987) and on luminal sites in the Golgi (Ma et al., 1999) catalyzed by Golgi-resident kinases. In studies over two decades past, protein kinases are suggested to regulate posttranslational control of core 2 activity (Datti and Dennis, 1993; VanderElst and Datti, 1998). More recently Chibber et al. show that PKCp2-dependent phosphorylation of human C2GlcNAcT-I, regulated by TNFa, results in higher enzyme activity, and may play a role in diabetic retinopathy (Ben-Mahmud et al., 2004; Chibber et al., 2000; Chibber etal, 2003). 1.6.3 Regulation of P-selectin ligand formation P-selectin-dependent adhesion is managed by the synthesis of sLex carbohydrates (and its derivatives, such as sulfo-sLex) expressed on PSGL-1. Currently, it is unclear whether a single "master switch" (enzyme) controls selectin ligand formation or if several enzymes are involved. Furthermore, several of the glycosyltransferases that direct key steps in selectin ligand formation are representatives of enzyme families: the C2GlcNAcT family (Merzaban et a l , 2005), the FucT family (Homeister et al., 2001; Weninger et al., 2000) and the ST3Gal family (Ellies et al , 2002). Hence, cells may use more than one enzyme to regulate key steps in selectin ligand formation. Because FucT VII and C2GlcNAcT-I are important in P-selectin ligand formation, many studies focus on these enzymes, and several cytokines are implicated in their regulation. 87 FucT-VII regulation In CD4 + T cells, both IL-12 and TGF-p (TH1 cytokines) up regulate FucT-VII enzyme expression, leading to selectin ligand formation, while IL-4 (TH2 cytokine) represses its expression (Blander et al., 1999; Lim et al., 1999; Nakayama et al., 2000; van Wely et al., 1998; Wagers and Kansas, 2000; Wagers et al., 1998; Xie et al., 1999). In CD8 + T cells, IL-12 (T c l) up regulates FucT-VII expression, which leads to effective selectin ligand formation, while IL-4 (Tc2) suppresses its expression (Cerwenka et al., 1999; Wang et al., 2000; Xie et al , 1999). C2GIcNAcT-I regulation In generating P-selectin ligands on T cells, both fucosylation and sialylation steps generally depend on pre-existing core 2 branches, giving C2GlcNAcT a crucial and limiting role in the ligand synthesis. IL-2, IL-12 and IL-15 (Tcl) are shown to promote the activity of C2GlcNAcT-I and other glycosyltransferases required for selectin ligand expression in CD8 + cells in vitro (Carlow et al., 2001a). Based on their potent impact in vitro, these cytokines were expected to modulate homing receptor formation under in vivo conditions. In order to study CD8 + T cell responses in vivo, an adoptive transfer model was used to measure the transgenic T cell response to male-specific antigen in the absence of IL-2, IL-15 and/or IL-12. Although these cytokines appear to regulate C2GlcNAcT-I in vitro, P-selectin ligand formation and proliferative responses are not impacted in the absence of these cytokines (Carlow et al., 2005). Other, not-yet-defined signals may regulate selectin ligand formation in vivo. Cytokines regulate the control of C2GlcNAcT and FucT-VII expression and, hence, STAT signalling pathways are inferred to regulate their expression. Since many signals downstream of IL-12, IL-2 and IL-4 are associated with STAT4 (Lim et al., 1999; 88 White et al., 2001), STAT5 (Quelle et al., 1995) and STAT6 (Quelle et al., 1995), respectively, these transcription factors are proposed to control C2GlcNAcT and FucT-VII expression. For instance, C2GlcNAcT-I expression in Tel cells may involve activating STAT4 and STAT5, whereas its expression in Tc2 may occur through STAT6 (Carlow etal., 2001a). 89 1.7 Thymic progenitor homing Although many early studies neglected the importance of the thymus, reducing it to a graveyard for dying lymphocytes, the thymus is now recognized for its ability to support the development of T cells from incoming, bone-marrow-derived progenitor cells (Anderson and Jenkinson, 2001; Blais et al., 2004; Guy-Grand et al., 2003; Miller, 2002; Petrie, 2002). Absence of the thymus due to a genetic defect or after a thymectomy results in a lack of peripheral T cells, yielding severe immunodeficiency (reviewed in (Miller, 2002)). Identifying the T cell progenitor that enters the thymus to fill the empty niches is under intense study, but how these cells actually enter the thymus is currently not fully understood. CHAPTER 5 of this thesis will delve more deeply into this issue. The following will discuss the identity of the circulating thymic progenitor (CTP) responsible for seeding the thymus, progenitors' development in the thymus, and the adhesion molecules responsible for driving progenitors' entry into the thymus. 1.7.1 Origin of T cell progenitors Bone-marrow-derived HSCs give rise to lymphoid, myeloid and erythroid lineages (Morrison et al., 1995). T cell development is unique from the other lineages because it takes place in the thymus, not the bone marrow (Zuniga-Pflucker, 2004). Since the thymocyte progenitors have limited, self-renewing capability, they are constantly replaced by rare (Harrison and Astle, 1997; Schwarz and Bhandoola, 2004; Yamamoto et al., 1996) circulating, bone-marrow-derived HSCs throughout adult life (Foss et al., 2001; Goldschneider et al., 1986; Matsuzaki et al., 1993; Scollay et al., 1986), occurring at an average rate of 2-3% per day (Donskoy and Goldschneider, 1992) in order to maintain the T cell pool (Donskoy and Goldschneider, 1992; Foss et al., 2001; Scollay et al., 1986). But the thymus is only periodically receptive to seeding from the blood since a 90 limited number of niches exist and their occupancy precludes further colonization until they are vacated. Once a stem cell enters it differentiates and proliferates ultimately giving rise to upwards of 30,000 single positive T cells (Foss et al., 2001; Porritt et a l , 2003; Prockop and Petrie, 2004). A number of stem cell progenitor populations have been studied in order to identify which population of bone marrow cells is responsible for seeding the thymus (Allman et al., 2003; Bhandoola et al., 2003; Martin et al., 2003). HSCs, multipotent progenitors (MPPs) and CLPs can all directly give rise to T cells when injected into the blood or directly into the thymus (Adolfsson et al., 2001; Allman et al., 2003; Christensen and Weissman, 2001; Kondo et al., 1997; Spangrude et al., 1988; Spangrude and Scollay, 1990). Identifying the thymic precursors: early thymic progenitors (ETPs) Allman et al., in a recent assessment, characterize the ETP as responsible for developing thymic-derived T cells (Allman et al., 2003). By staining thymocytes with lineage markers (CD8a/p\ TCRp, CD3e, TCRy, CD25, NK1.1, Mac-1, Gr-1 and B220), they removed double-positive (DP), single-positive (SP), intermediate-single-positive (ISP) y/8 T cells, double-negative (DN)2 (CD44+CD25+) and DN3 (CD44_CD25+) cells, NK cells, and residual myeloid cells and B cells. The thymocytes resulting within this Lin"CD44+CD25~ (DN1) population corresponded to about 0.01% of the total population of thymocytes. The c-kithl, IL-7Ra"/1°, Sca-l h l subset of these cells show T-lineage potential upon intrathymic injection. This is consistent with other studies illustrating that, within the DN1 population of thymocytes, 0.05% (Shortman and Wu, 1996) of cells that highly express c-kit (Matsuzaki et al , 1993), and do not express IL-7Ra (Laky et al., 1998) or CD24 (Porritt et al., 2004), and are CD41 0 (Wu et al., 1991) and B220+ (Martin 91 et al , 2003), can give rise to T, B and myeloid lineages (Ardavin et al., 1993; Matsuzaki etal, 1993). ETPs possess all lineage potentials known to reside within the DN1 compartment (Allman et al., 2003), however, ETPs are too numerous to be the earliest, thymus-settling progenitors, since the numbers of progenitors that seed the thymus from the blood are estimated to be much smaller (Donskoy and Goldschneider, 1992). It is, therefore, speculated that ETPs arise through expanding a small number of thymus-settling cells (Bhandoola et al , 2003). Circulating thymic progenitors (CTPs) Using a parabiotic mouse model, Donskoy and Goldschneider effectively determine that blood-borne precursors maintain thymocytopoiesis in postnatal mice (Donskoy and Goldschneider, 1992). The identity of these blood-borne precursors is still the focus of much study. Schwarz and Bhandoola (Schwarz and Bhandoola, 2004) recently published that hematopoietic progenitors that are Lin"Sca-lhlc-kithl (LSK) are normally present in the blood, capable of becoming T cells and may represent the physiological circulating thymic progenitor (CTP). By examining the blood of adult mice, they identify this rare population of Lin"c-kithl cells, containing both Sca-1" and Sca-1+ subsets, but do not come across any circulating common lymphoid progenitors (CLPs). Cells of the LSK phenotype exist in the bone marrow (Adolfsson et a l , 2001; Adolfsson et al., 2005; Christensen and Weissman, 2001; Perry et al., 2004), blood (Schwarz and Bhandoola, 2004) and thymus (ETP) (Allman et al., 2003), making the blood LSK, not the CLP population, a viable candidate for the "thymic settling" cell. This LSK population is multipotent (like the bone marrow LSK) but heterogeneous, 92 containing HSCs (Flt3") and multipotent progenitors (MPPs) (Flt3+) subsets (Schwarz and Bhandoola, 2004). These CTPs migrate from the blood across the postcapillary venules deep in the thymus into the cortical tissue close to the cortico-medullary junction (CMJ) (Lind et al , 2001; Prockop and Petrie, 2000). They migrate from the CMJ into the thymic tissue, expand and differentiate (Lind et al., 2001; Petrie, 2002). It is believed that the first stage of commitment occurs either prior to or immediately following thymic colonization, with the CTP losing its ability to repopulate erythroid lineages upon committing to lymphoid development (Kawamoto et al., 1999; Rodewald et al., 1994; Wu et al., 1991). The search for a bone marrow precursor to ETPs and CTPs All blood cell lineages are derived from a common hematopoietic stem cell (HSC). The currently accepted model (Reya et al., 2001) of blood lineage commitment is that a long-term HSC (LT-HSC) becomes a short-term HSC (ST-HSC), which then is born into a multipotent progenitor (MPP) before the first lineage commitment step of adult HSCs occurs, resulting in a separation into a common lymphoid progenitor (CLP) (Kondo et al., 1997) (results in T and B cells) or a common myeloid progenitor (CMP) (results in megakaryocytes and granulocytes/macrophages) (Akashi et al., 2000). Flt3 ligand is critical to CLPs (Adolfsson et al., 2001; Sitnicka et al., 2002). This model recently underwent a revision (Adolfsson et al , 2001; Adolfsson et al., 2005). These investigators modify the currently accepted model by suggesting that lineage commitment occurs earlier in the tree. Instead of a MPP, they propose that the ST-HSC differentiates into either a CMP or a "lymphoid-primed" (L)-MPP. This LMPP still has the potential to differentiate into a granulocyte/macrophage, but not into a megakaryocyte. Figure 1-11 outlines these models. 93 Most of the LT-HSC (CD34Tlt3~) (Adolfsson et al , 2001; Christensen and Weissman, 2001; Osawa et al., 1996), ST-HSC (CD34+Flt3-) (Osawa et al., 1996; Yang et al., 2005) and MPP activities in adult mouse bone marrow reside in the small LSK (Lin"Sca-l+c-kit+) HSC compartment (0.1% of bone marrow cells) (Ikuta and Weissman, 1992; Li and Johnson, 1995; Spangrude et a l , 1988; Weissman et al., 2001). The ST-HSC, CD34+Flt3" population gives rise to a CD34+Flt3+ (LMPP) population that results in predominantly lymphoid cells, but also gives rise to granulocyte/macrophage lineages (Adolfsson et al., 2001). These data support the revised model. ETPs' lymphoid restriction and lack of self-renewal ability suggest that they may derive from a lymphoid-restricted and non-renewing progenitor in the bone marrow: the CLP (Kondo et a l , 1997), LSK CD34+Flt3" self-renewing HSC or LSK CD34+Flt3+ multipotent, non-renewing progenitor (Adolfsson et al., 2001; Christensen and Weissman, 2001). ETPs and CLPs appear functionally similar in that they are both lymphoid-restricted, non-renewing progenitors (Allman et al , 2003; Kondo et al., 1997). However, as Allman et al. discover, unlike CLPs, ETPs possess a weak myeloid differentiation potential. ETPs also produce more DP thymocytes for longer periods of time when injected intrathymically. In addition, ETPs do not have the potential to contribute to the B cell lineage (Allman et al., 2003). Examining early progenitors in Ikarosnu11 mice (possess T cell but lack B cells), they find that these mice have ETPs but lack CLPs. From these studies, they conclude that ETPs develop via a CLP-independent pathway that depends on an early hematolymphoid progenitor population in the bone marrow, LSK CD34+Flt3+. These bone marrow progenitors have a similar surface expression profile as ETPs in terms of Sca-1, c-kit, IL-7Ra and CD27 (Allman et al., 2003). This concurs with 94 Adolfsson et al. studies (Adolfsson et al., 2001) and supports the revised model of HSC lineage commitment (Adolfsson et al., 2005). Figure 1-11: HSC blood cell commitment models: under revision. A, The illustration is of the currently accepted model for hematopoietic lineage commitment and development, hematopoietic stem cell differentiation. This model suggests that the first lineage commitment decision results in a strict separation of myelopoiesis with the CMP and lymphopoiesis with the CLP. B, Revised model, by Adolfsson et al, is based on studies that suggest pluripotent HSC upon loss of Mk and E potential develops into a LMPP that, upon loss of GM potential, generates the CLP. LT-HSC, long-term hematopoietic stem cell; ST-HSC, short-term hematopoietic stem cell; MPP, multipotent progenitor; LMPP, lymphoid primed multipotent progenitor; CLP, common lymphoid progenitor; CMP, common myeloid progenitor; GMP, granulocyte/macrophage progenitor; MkEP, megakaryocyte/erythroid progenitor; B, B cell; T, T cell. ' To further characterize this progenitor population, Perry et al. ascertain that the CD62L+ fraction of the bone marrow LSK population contains mainly T lineage progenitor activity both in vitro and in vivo, while the CD62L" fraction contains mainly B lineage potential (Perry et al., 2004). This suggests that the CD62L+ population includes a bone marrow analogue to the ETP and likely includes the population from which the thymic ETP originates (Perry et al., 2004). 1.7.2 Thymic environment and T cell development Once inside the thymus, the ETP begins a differentiation process typically characterized by changes in cell-surface phenotype, proliferation status and functionality. This differentiation occurs as the progenitor moves through the cortex in the direction of the sub-capsular zone (Lind et al., 2001). Figure 1-12 illustrates this. Thymic microenvironments The thymus contains unique types of stromal cells that provide signals for thymocyte development. Examples of extrathymic T cell lymphopoiesis also exists (Blais et al., 2004; Guy-Grand et al., 2003), but these cells do not behave as thymically derived T-cells. Oncostatin M-transgenic mice represent a unique model to address the need for thymic epithelial cells (TECs) on T cell development. These mice show that under chronic exposure to OM the lymph node can transform from a secondary lymphoid organ to a primary lymphoid organ that supports T-cell development independent of the thymus (Boileau et al., 2000; Clegg et al., 1996). This effect of OM on T cell development in lymph nodes is due to the amplification of a cryptic pathway that is operative in conditions of defective thymopoiesis and is neatly characterized in nude mice 96 Figure 1-12: T cell development and the thymic microenvironments. The thymic lobes are divided into cortical and medullary regions each characterized by exclusive sets of stromal cells which contribute to a unique microenvironment for the development and maturation of thymic precursors into mature T cells. Migration of T cell precursors from the bone marrow through the blood to the thymus is a highly regulated process. Once in the blood, the circulating thymic progenitor (CTP) must exit the blood vessel and enter the thymus at the cortico-medullary junction. The homing receptor, P-selectin, and its ligand, PSGL-1, are implicated in the regulation of CTP entry into the thymus. Once in the thymus, the early thymic progenitor (ETP) commits to the lymphoid lineage. Notch ligands and VCAM-1 expressed on cortical stromal cells, help mediate migration as well as differentiation in the early stages of thymocytes development. After a period of nine to 11 days, DN1 cells (CD25"CD44+) move outward into the cortex, progressively differentiating through DN2 (CD25 +CD44 +) and DN3 (CD25+CD44") stages before accumulating in the subcapsular region (tope of diagram). Transition to the DP stage correlates with a reversal in polarity of migration back into the cortex, but only cells that appropriately recognize self antigens (positive selection) are allowed into the medulla, where they mature. Medullary epithelial cells are involved in mediating self tolerance. Fully mature T cells (CD4 + or C D 8 + single positive; SP) are induced to leave the thymus. Migration within the thymus is likely based on chemokine gradients, but this is largely unknown. 97 (Guy-Grand et al., 2003). Blais et al, in testing the functionality of the T cells that developed from the lymph nodes of OM-transgenic mice, find that the site of development modifies T-cell quality and function, and that extrathymic T cells functionally cannot substitute for classical thymic T cells (Blais et al., 2004). These studies and others emphasize the importance of the thymus and its intricate microenvironment in the development of functional T cells. In order to understand the thymic differentiation process, it is important to understand the microenvironments present in the thymus. Each transition in development that the thymocyte undergoes depends on interactions with a complexity of thymic stromal cells which make up the thymic microenvironments, including cortical epithelial cells (CEC), medullary epithelial cells (MEC) and bone-marrow-derived M H C class 1/ M H C class II restricted DCs. As illustrated in Figure 1-12, the thymus is divided into discrete cortical and medullary areas, characterized by the presence of particular stromal cell types, as well as the thymocyte precursors at defined maturational stages. Interactions between thymocytes and thymic stromal cells are important in driving T cell commitment and maturation, ultimately giving rise to self-tolerant CD4 + and C D 8 + T cells, which emigrate from the thymus to establish the peripheral T cell pool. When T cell precursors enter at the CMJ, they are held in a complex, three-dimensional reticular network of M H C class II cortical epithelial cells (Figure 1-12). The thymic cortical epithelial cells are also the most-efficient mediators of positive selection (Anderson and Jenkinson, 2001), whereas the M H C class II + DCs at the CMJ are the most-efficient mediators of negative selection (Anderson and Jenkinson, 2001). The thymic medulla, on the other hand, provides a microenvironment for positively selected, single-positive CD4 + and C D 8 + T cells to reside for up to two weeks postselection. There, 98 the medullary epithelial cells regulate the differentiation events leading to further tolerance induction (Anderson and Jenkinson, 2001; Hanahan, 1998). These epithelial cells also regulate expansion of thymocytes in an IL-7-dependent manner (independent of TCR-MHC interactions) prior to their export to the periphery (Hare et al., 1999; Hare et al., 2000; Park etal., 2004). Progenitors move in a specific outward direction toward outer cortex only against a large flow of DP cells moving inward toward the medulla. This highly precise movement is controlled by many adhesion molecules expressed or secreted by cortical epithelial cells. Chemokines produced by epithelial cells that are found scattered throughout the cortex might be required to activate cytoskeletal reorganization and induce integrin-mediated binding to stromal ligands such as 0C4 integrins (expressed by lymphoid progenitors) and VCAM-1 (expressed by the cortical epithelial cells) (Prockop et al., 2002). This interaction enforces sustained signals for lymphoid-lineage commitment, including Notch (for example Notch-1 expressed on the developing thymocyte) signalling through the engagement of epithelial-cell-expressed delta-like ligand 1 or other Notch ligands expressed on thymic epithelial cells (Anderson and Jenkinson, 2001; Felli et al., 1999; Harman et al , 2003; Schmitt et al., 2004; Zuniga-Pflucker, 2004). T cell development Thymocyte differentiation can be followed phenotypically by the expression of cell-surface markers, including CD4, CD8, CD44, CD25, CD69 and L-selectin, as well as the status of the TCR. The CTPs recruited to the thymus first develop into CD4"CD8" double-negative (DN) cells, then become CD4 +CD8 + double-positive (DP) cells, and lastly mature into single-positive (SP) CD4 + or CD8 + T cells (Anderson and Jenkinson, 99 2001) , as outlined in Figure 1-12. The DN stage can be further subdivided on the basis of CD25 and CD44 cell-surface expression (Ceredig and Rolink, 2002; Godfrey et al., 1993). Recent work by Petrie et al. (Lind et al., 2001; Porritt et al , 2003; Prockop et al., 2002) conveys that an intricate movement through distinct thymic zones exists. As illustrated in Figure 1-12, the cortex can potentially be divided into four regions. Lymphoid progenitors enter the thymus at the CMJ and spend about nine to 11 days localized to the region closest to the medulla, where these DN1 (CD3" CD44+++CD25"c-kit+++IL-7Ra") cells undergo a proliferative expansion and differentiation. DN1 cells have multilineage potential (B, NKT, DC, as well as T cell) (Shortman and Wu, 1996) and require many cytokines to survive. They then migrate toward the outer cortex, where they differentiate into DN2 (CD3"CD44+++CD25++c-kit+ + +IL-7Ra+) cells, proliferate in response to IL-7, and begin TCRy/5 gene rearrangements (peak at day 12 or 13). Transition to the DN3 (CD3 l o wCD44+CD25+ +c-kit+IL-7Ra l o w) stage of development occurs in the next region and is marked by a decrease in proliferation and TCR|3 gene rearrangements, signifying T lineage commitment (peak at day 15) (Capone et al., 1998; Godfrey and Zlotnik, 1993; Ismaili et al., 1996; Moore and Zlotnik, 1995; Rodewald et al., 1994; Wu et al., 1996). Transition from DN (CD3 l o wCD44 l o wCD25 c-kit l o wIL-7Ra) stage to DP (CD4 l o wCD8 l o wCD3+CD44-CD25"c-kitlL-7Ra") stage occurs in the subcapsular zone and is manifested by rapid proliferation, expression of the pre-TCR as well as TCRoc gene rearrangements (day 15 or 16). The successful assembly of the pre-TCR complex provides critical survival signals and is associated for this robust proliferation to occur. The progeny of this expansion of DP cells comprises 85% to 90% of total thymocytes. Over the next three to five days, the cortical epithelial cells expressing self-MHC/peptide serves to mediate 100 positive selection of these DP thymocytes. The newly formed TCR complex both successfully binds self-MHC/peptide within an appropriate range of affinity and saves the DP thymocytes by positive selection or the DP thymoctye dies by neglect. Ultimately, cells that are positively selected must enter the medulla to become functional. After differentiation to the DP stage, the polarity of migration is reversed, and cells move inward through the cortex toward the medulla, although DP cells do not enter the medulla (Lind et al., 2001; Petrie, 2002). The final step of intrathymic migration is the movement of selected cells from the cortex into the medulla where TCRocp-dependent selection occurs (Singer, 2002). After positive selection is complete, selected cells remain in the medulla for an additional three to seven days (Shortman et al., 1990). Before they mature, the developing thymocytes must undergo negative selection to remove self-reactive cells mediated by dendritic cells, which are numerous at the CMJ as well as by macrophages which are scattered in the cortex and medulla. The remaining proliferation occurs over a two-to-four-day span that includes the DP stages of differentiation (Penit et al., 1995; Shortman et al., 1990; Tourigny et al., 1997). Immune surveillance requires the release of newly derived T cells from the thymus into the peripheral circulation. It has been shown recently that chemokines and their receptors are differentially expressed by subsets of thymic stromal cells and thymocytes, suggesting that chemokines may be involved in directing the intrathymic localization of developing thymocytes (Campbell et al , 1999; Norment and Bevan, 2000). CCR7 ligands are expressed in the thymus, with highest expression in the medulla and the SCZ. CCR7 is expressed on mature, single-positive thymocytes and on some DP thymocytes. This indicates that CCR7 and its ligands are essential for the intrathymic migration of positively selected thymocytes from the cortex to the medulla (Ueno et al., 2004). Studies also illustrate that CCR7 is essential for proper differentiation of DN thymocytes, particularly between the 101 DN1 to DN2 transition, moving thymocytes away from the medulla (Misslitz et al., 2004). In vitro studies show that thymocytes' responsiveness to chemokines - including MDC, TARC, SLC, ELC, SDF-1 and TECK - is developmental^  regulated (Campbell et al., 1999; Kim et al., 1998). Specifically, CXCR4/SDF-1 is observed at all developmental stages, CCR4/MDC-TARC at transitional stages (cortical-to-early-medullary) and, at later stages, thymocytes acquire CCR7 and L-selectin and become responsive to SLC/ELC, which help in the homing to secondary lymphoid organs (Moser and Loetscher, 2001). The ETP is categorized as part of the DN1 subset of DN T cells. There is much heterogeneity among DN1 thymocytes, and the search for the most immature subset responsible for developing T cells is quite complex. Recent work by Petrie et al. further divided the DN1 subset based on CD24 and c-kit into five additional populations (DNla= c-kif7CD24~, DNlb= c-kit7CD24l0, DNlc= c-kitlo/CD24+, DNld= c-kit7CD24+ and DNle= c-kit7CD24"). They deduce that the DNla and DNlb subsets, both of which are c-kit+ but differ in their expression of CD24, represent effective intrathymic T cell precursors. The CD24" DNla most likely represents the earliest intrathymic progenitor (Porritt et al, 2004). 1.7.3 Thymic progenitor migration from the blood to the cortico-medullary junction of the thymus CTPs that enter the thymus are required to traverse the wall of postcapillary venules at the cortico-medullary entry site. Extravasation is a complex process tightly controlled by a cascade of molecular interactions. In inflammation, low-affinity interactions established by selectins and selectin ligands promote the capture of cells from the bloodstream, leading to leukocyte tethering and rolling on endothelial cells, as 102 discussed above (Figure 1-2). Thus, by slowing down the cells, this mechanism increases the probability that they will respond to local chemokine gradients. Chemotactic signals prompt cells that express the appropriate chemokine receptor to stop and extravasate through the endothelial cell layer into the tissue, in a process further mediated by additional chemokines and integrins. The selectins are involved in mediating the trafficking and homing of circulating lymphocytes. In an analogous manner, it is suggested that E- and P-selectin constitutively expressed on bone marrow endothelial cells (Jacobsen et al., 1996; Schweitzer et al., 1996) mediate rolling of HSC on bone marrow endothelium (Frenette et al., 1998; Mazo et al., 1998) critical in HSC homing and engraftment into the bone marrow (Frenette et al., 1998; Hidalgo et al., 2002; Katayama et a l , 2003). It is well recognized that human CD34+ HSCs express PSGL-1 (Levesque et al., 1999; Sackstein and Dimitroff, 2000; Zannettino et al., 1995) and that human CD34+ HSCs bind to E-selectin expressed on human bone marrow endothelial cells (Naiyer et al., 1999; Rood et al., 1999; Schweitzer et al., 1996; Winkler et al., 2004). CD34+ HSCs also express the CD44 glycoform, HCELL, which is proposed as one of the major ligands for E-selectin on human HSCs (Sackstein, 2004). It is speculated that selectin-like receptors on CTPs may be the first step in thymic progenitor homing (Petrie, 2002). CHAPTER 5 in this thesis strives to identify the selectins involved and specifically sheds light on the importance of the interaction of PSGL-1 on the CTP with P-selectin expressed on the thymic endothelium in thymic progenitor homing and regulation. 103 1.8 Thesis objectives The overall goal of the work presented in this thesis is to determine the role of the O-glycan-branching enzymes, the C2GlcNAcTs, in selectin ligand formation controlling the migration of: (i) activated CD8 + T cells; and (ii) thymic progenitor cells. 1.8.1 C2GlcNAcTs and their role in the formation of functional selectin ligands in Activated T cells This thesis strives to define a physiological role of an alternate C2GlcNAcT isoenzyme, C2GlcNAcT-III, in forming functional P-selectin ligands in T cells both in vitro and in vivo. Much of the current literature is focused on defining the role of C2GlcNAcT-I in the synthesis of functional selectin ligands on CD4 + T cells and neutrophils. Only a limited number have focused on defining the role of C2GlcNAcT-I in CD8 + T cells. These studies found that in the absence of C2GlcNAcT-I neither CD4 + (Snapp et al., 2001), CD8 + (Carlow et al., 2001a) nor neutrophils (Ellies et al., 1998) are able to bind P-selectin suggesting that C2GlcNAcT-I is solely responsible for forming functional P-selectin ligands. Two additional core 2 isoenzymes have been identified in human tissue, expressed and characterised, C2GlcNAcT-II and C2GlcNAcT-III. C2GlcNAcT-II (Schwientek et al., 1999; Yeh et al., 1999) has 48% identity with C2GlcNAcT-I and is unusual in its broad substrate specificity catalysing the formation of core 2 and core 4 structures and is mainly expressed in cells of the digestive tract and is thought to be important in the formation of core 2 and core 4 O-glycan structures on barrier mucins. C2GlcNAcT-III (Schwientek et al., 2000) has 42% identity with C2GlcNAcT-I and is strongly expressed in the thymus and to a lesser degree in small intestine, spleen and peripheral leukocytes. This isoenzyme has no additional core 4 branching activity. C2GlcNAcT-I and C2GlcNAcT-III are co-expressed in thymocytes 104 and peripheral leukocytes, while C2GlcNAcT-I and C2GlcNAcT-II are co-expressed in mucin secreting cells. This coexpression with C2GlcNAcT-I of these isoenzymes along with their ability to all modify CD43 with core 2 branches suggests that there may be considerable redundancy between these enzymes. The role of the two more recently discovered isoenzymes in the formation of functional selectin ligands is unknown. Studies of the C2GlcNAcT-In u 1 1 mouse have shown that this mouse has a very mild phenotype showing only a slight neutrophilia (Ellies et al., 1998) further suggesting that redundancy occurs. Our interest in the role of other C2GlcNAcT isoenzymes in T cells stems from the apparently normal T cell homing phenotype in C2GlcNAcT-In u 1 1 mice. Myeloid cells constitutively express all glycosyltransferases required to form selectin ligands whereas naive T cells express PSGL-1 but do not bind P- or E-selectin. Upon activation they up-regulate specific glycosyltransferases, including C2GlcNAcT-I, that are required for P- and E-selectin ligand formation. We hypothesize that other C2GlcNAcT isoenzymes may contribute to the formation of functional selectin ligands in activated T cells. Based on the distribution of C2GlcNAcT-II and C2GlcNAcT-III in human tissues, we hypothesize that C2GlcNAcT-III may be of particular importance. 1.8.2 C2GlcNAcT-I and Thymic Progenitor Homing The current work establishes that selectin-mediated adhesion is more than just a common mechanism underlying extravasation of circulating cells in pathological conditions, such as the entry of activated CD8 + T cells into inflammatory regions, or entry of naive T cells in the lymph node; it also plays a critical role in the homeostatic homing of circulating thymic progenitors to the thymus. In a recent review by Petrie (Petrie, 2003), he hypothesized that selectin-like molecules may be involved in mediating the entry of thymic progenitors into the thymus although no basis for this had existed in 105 the literature. Using a number of well-established mouse models, we investigated the possibility that selectins and selectin ligands mediate the early steps in extravasation of thymic progenitors into the thymus. Early work in our lab, in collaboration with Dr. Fabio Rossi, found that P-selectin is expressed at constitutively low levels on the thymic endothelium while P-selectin ligand is expressed on thymic progenitors in the bone marrow, blood and thymus. This led us to hypothesize an essential role for the P-selectin-PSGL-1 interaction in controlling the entry of thymic progenitors into the thymus from the blood. 106 CHAPTER 2 MATERIALS AND METHODS 2.1 Mice Mice used for in vitro and in vivo experiments presented in Chapters three, four and five of this thesis are outlined in Table 2-1. 2.1.1 Mice used for in vitro and in vivo T cell stimulations Mice aged six to 10 weeks were used for analyses. C57BL/6 and P S G L - l n u 1 1 mice, originally from the Jackson laboratory (Bar Harbor, ME), were bred at the Biomedical Research Centre (University of British Columbia, Vancouver, B.C., Canada). C2GlcNAcT-F u l i mice were kindly supplied by Dr. J. Marth (Howard Hughes Medical Institute, University of California at San Diego, La Jolla, Calif.) and were backcrossed seven generations with C57BL/6 mice. The C2GlcNAcT-I n u 1 1 mice were mated with CD43 n u l 1 mice to produce mice that lack both C2GlcNAcT-I and CD43 genes. C2GlcNAcT-I n u U mice were also mated with P S G L - l n u 1 1 mice to produce mice that lack both C2GlcNAcT-I and PSGL-1 genes. H Y TCR-transgenic mice on the C57BL/6 background were crossed with C2GlcNAcT-I n u 1 1 mice to yield mice with a HY + C2GlcNAcT-I n u 1 1 genotype. HY + C2GlcNAcT-I n u 1 1 were crossed with H Y + C D 4 3 n u 1 1 mice to yield HY + C2GlcNAcT-I n u l7CD43 n u l lmice. 2.1.2 Mice used for thymic homing experiments C57BL/6 and congenic C57BL/6 (Thyl. l or CD45.1 (also known as Ly5.1)), IL-7R n u l l C57BL/6, PSGL-l n u l l C57BL/6 and P-selectinn u l lC57BL/6 mice were purchased from the Jackson laboratory. C2GlcNAcT-I n u 1 1 mice were obtained from Dr. J. Marth (Howard Hughes Medical Institute, UCSD) and backcrossed seven generations with C57BL/6 107 mice. To produce C2GlcNAcT-I n u l l C57BL/6 C D 4 5 1 / C D 4 5 - 2 mice, C2GlcNAcT-In u 1 1 mice were backcrossed two generations with C57BL/6 C D 4 5 1 mice. Table 2-1: Mice used in experiments represented in chapters indicated Mice Background CD45(Ly) Allele Thy Allele Experiments (chapter) Wild-type C57BL/6 CD45.2 Thy1.2 3,4&5 Wild-type C57BL76 CD45.1 Thy1.2 5 Wild-type C57BL/6 CD45.1/CD45.2 Thy1.2 5 Wild-type C57BL/6 CD45.2 Thy1.1 5 C2GlcNAcT-lnu» C57BL/6 CD45.2 Thy1.2 3,4&5 C2GlcNAcT-l™» C57BL/6 CD45.1/CD45.2 Thy1.2 5 CD43nu" C57BL/6 CD45.2 Thy1.2 3 CD43null/C2GlcNAcT-|null C57BL/6 CD45.2 Thy1.2 3 PSGL-1nu" C57BL/6 CD45.2 Thy1.2 3,4&5 PSGL-1nul1/ C2GlcNAcT-l™» C57BL/6 CD45.2 Thy1.2 3&4 P-selectinnu" C57BL/6 CD45.2 Thy1.2 5 IL-7R™" C57BL/6 CD45.2 Thy1.2 5 HY* wild-type C57BL/6 CD45.2 Thy1.2 3 HY«9 C2GlcNAcT-l™» C57BL/6 CD45.2 Thy1.2 3 HY^ CD43nu" C57BL/6 CD45.2 Thy1.2 3 HYl9CD43nul1/ C2GlcNAcT-l™" C57BL/6 CD45.2 Thy1.2 3 2.2 Media used and Antibodies used for experiments Anti-CD43 mAb, 1B11-PE (09695A), was obtained from BD Pharmingen. Conditioned media from the anti-CD43 hybridoma line 1B11 (50) and from the anti-pan-CD43 hybridoma line SI 1 (kindly supplied by Dr. J.Kemp, Department of Pathology, University of Iowa, Iowa City, IA) were used for Western blotting experiments. P-selectin-hlgG fusion protein (28111 A; BD Pharmingen) and E-selectin-hlgG fusion protein (575-ES; R&D Systems) were detected with biotinylated anti-human IgG (109-065-098; Jackson ImmunoResearch Laboratories) and CyChrome-conjugated streptavidin (554062; BD Pharmingen) or just with anti-human IgG-PE (109-116-098; Jackson ImmunoResearch Laboratories). Rabbit polyclonal anti-peptide Ab HI 8 was raised against a peptide 108 corresponding to the C-terminus of CD43 and affinity purified on immobilized peptide. CD8a-allophycocyanin (553035; BD Pharmingen) and CD4-FITC (01064D; BD Pharmingen) were used for FACS staining. Neutralizing anti-PSGL-1 mAb, 2PH-1, was kindly provided by Dr. D. Vestweber (University of Miinster, Miinster, Germany). For FACS staining, cells were suspended in DMEM (Invitrogen Life Technologies) supplemented with 8% FCS (Gibco) and incubated with Abs for 15-30 min on ice in 96-well roundbottom plates (163320; Nunc). Cells were washed twice and analyzed on a FACSCalibur flow cytometer (BD Biosciences). For negative controls, cells were either incubated with the neutralizing anti-PSGL-1 Ab 2PH-1 before P-selectin-hlg staining, or EDTA (5 mM) was added during staining with P-selectin-hlgG or E-selectin-hlgG to prevent binding. 2.3 Cell Isolation and T cell Cultures Splenocytes were cultured in culture plates (BD Biosciences) at 106 cells per millilitre in RPMI 1640 medium, 8% FBS (Gibco) and stimulated with 4p.g-per-millilitre Con A (Sigma-Aldrich) and IL-2 for 48 hours at 37°C in 5% C0 2 . Cells were then harvested, washed, counted and re-plated at various secondary cell culture densities, with 2% IL-2 or 2% IL-4 supernatant. Supernatant was obtained as conditioned medium from the myeloma X 653 transfected with the cDNA for murine IL-2 or murine IL-4 (Dr. F. Melcher, Basel Institute of Immunology, Basel, Switzerland). 2.4 Flow cytometric analysis 2.4.1 P-selectin-hIG and E-selectin-hIG labeling of activated cells For FACS labeling, cells were suspended in DMEM (Invitrogen Life Technologies) supplemented with 8% FBS (Gibco) and incubated with P-selectin for 15 to 30 minutes on ice in a 96-well round-bottom plate (Nunc). Cells were then washed 109 twice and analyzed on a FACSCalibur flow cytometer (BD Biosciences). For negative controls, cells were either incubated with the neutralizing oc-PSGL-1 Ab, 2PH-1 (Dr. D Vestweber, Miinster, Germany), or 5mM EDTA was added during labeling with P-selectin-hlG or E-selectin-hIG to prevent specific binding. P-selectin-hlgG fusion protein (28111 A; BD Pharmingen) and E-selectin-hlgG fusion protein (575-ES;R&D Systems) were detected with biotinylated anti-human IgG (109-065-098; Jackson ImmunoResearch Laboratories) and CyChrome-conjugated streptavidin (554062; BD Pharmingen) or just with anti-human IgG-PE (109-116-098; Jackson ImmunoResearch Laboratories). 2.4.2 Procedure used to harvest and stain tissues in thymic homing experiments Anesthetized mice (400-microliters for 20-25g mouse of 25mg-per-millilitre avertin-2,2,2-tribromoethanol stock solution, Sigma) were perfused with PBS containing 5mM EDTA before thymi, spleens and peripheral and mesenteric lymph nodes were harvested. Single-cell suspensions were stained in PBS, with 5% FBS for the following markers: CD4-FITC (L3T4, #553047), CD8p-biotin (CL8938B, Cedarlane), Thyl.l-PE (CD90.1, #554898), and Thyl.2-APC (CD90.2, #553007), CD45.1-FITC (A20, #553775), CD45.1-PE (A20, #553776), CD45.2-biotin (104, #553771). Biotinylated antibodies were detected using streptavidin-PE-Cy5 (#554062). All antibodies were purchased from PharMingen (BD Biosciences), unless otherwise stated. For the analysis of peripheral chimerism, the frequency of single-positive (CD4+ or CD8+) T cells was assessed. For the analysis of thymic chimerism, double-positive lymphocytes, which are generated and reside uniquely in the thymus (Petrie et al., 1990), were analyzed. 110 2.5 Assays 2.5.1 Core 2 and core 4 enzymatic assays Cells were washed in PBS and lysed in 150 mM NaCl and 0.25% Triton X-100 with protease inhibitors (10-u,g-per-millilitre soybean trypsin inhibitor, 40-p.g-per-millilitre phenylmethyl-sufonylfluoride, 10-p.g-per-millilitre aprotinin, 10-p,g-per-millilitre leupeptin and 5-p,g-per-millilitre pepstatin) at 4°C. Transferases assays were performed in triplicate according to established protocols (Bierhuizen and Fukuda, 1992; Saitoh et al., 1991). The reaction mixtures for the C2GlcNAcT assay contained 50 mM MES (pH 7.0), 0.5 \xd of uridine diphosphate-A^-acetyl-D-glucosamine (glucosamine-6-[3H](N)) (NEN), 1 mM uridine 5'-diphospho-iV-acetylglucosamine (Sigma-Aldrich), 0.1 M GlcNAc, 1 mM acceptor and 25-microliters of cell lysate, in a total volume of 50-microliters of cell lysate in a total volume of 50-microliters . The acceptor substrates were Gal pl-3GalNAca-pNp and GlcNAcpGalNAccc-pNp (Toronto Research Chemicals) for the core 2 and core 4 assays, respectively. Negative control samples were run for each assay, with all the reactants in the absence of acceptor substrates to determine the background signal for normalization purposes. The mixtures were incubated for three hours at 37°C, processed by CI8 Sep-Pak (Waters) column chromatography, and eluates were counted in a scintillation counter. 2.5.2 CD43 Western blots Cells were washed with PBS and lysed at 108 cells per millilitre on ice in buffer containing 20mM Tris (pH 7.5), 0.15 M NaCl, 0.5% Triton X-100, 10-u.g-per-millilitre aprotinin, 10-u,g-per-millilitre leupeptin, 174-u\g-per-millilitre PMSF and 5-p,g-per-millilitre-pepstatin. Lysates were rotated for one hour at 4°C before centrifugation at 111 14,000 rpm for five minutes in an Eppendorf centrifuge 5402. Supernatants were then recovered for SDS-PAGE or used for immunoprecipitations. Abs were bound to protein G-coupled Sepharose beads. After washing beads free of unbound Ab with PBS, cell lysates were mixed with beads and incubated at 4°C for one to two hours. Immunoprecipitates were washed in TBS containing 0.1% Tween 20 (Fisher Scientific), and an immunoprecipitated protein was extracted from beads with Laemmli-loading buffer. 2-Me was added to reduce the Ab where necessary, and samples were then loaded for SDS-PAGE with a 4% stacking gel and 8% resolving gel. Resolved protein was transferred to PROTRAN-BA85 nitrocellulose membrane (Schleicher & Schuell Microscience), that was subsequently blocked with 5% BSA. Blots were probed with Abs diluted in TBS containing 0.5% BSA and 0.5% Tween for 60 minutes. Blots were washed in TBS with 0.1% Tween and were detected with goat anti-mouse Ig-HRP or goat anti-rabbit HRP, where appropriate (Invitrogen Life Technologies), washed and developed with ECL reagent (Amersham Biosciences) for autoradiography with Biomax film (Eastman Kodak), according to the manufacturer's instructions. Molecular mass standards in the high range were used (Bio-Rad). 2.5.3 In vitro flow chamber Activated CD8 + T cell interactions with immobilized P-selectin and E-selectin under physiologic flow conditions were assessed using an in vitro flow chamber, according to established protocols (Bakowsky et al , 2002; Bendas et al , 1997; Lawrence and Springer, 1991; Reinhardt and Kubes, 1998; Vogel et al , 1998). Petri dishes (Corning) were coated with P-selectin-hIG or E-selectin-hIG (5 p,g per millilitre) in PBS. After overnight incubation at 4°C, the coated spot was rinsed four times with PBS and blocked with one millilitre of 1% BSA in HBSS for one hour at 37°C. The substrate-112 coated petri dishes were mounted in a parallel-plate flow chamber. For most experiments, H B S S was drawn through the chamber at a shear stress of two dyne per square centimetre (0.55 millilitre per minute) with a syringe pump (Harvard Apparatus). Activated CD8 + T cells (106 cells per millilitre) were suspended in DMEM with or without EDTA (5 mM) and then perfused through the chamber for a six-minute period. Cell rolling was observed using an inverted-phase contrast microscope (Zeiss) and was videotaped using an Exwave HAD colour video camera (model SSC-DC50A; Sony) with a Super VHS ET video recorder (model HR S991 IU; JVC) and an attached time-date ID generator (Crest electronics Inc.) at XI0 objective. Multiple random fields were recorded for at least 20 seconds at the end of the four-minute perfusion period. The total number of rolling cells within each 0.16-square-millimetre field was determined by analyzing the videotapes, with a minimum of five fields analyzed for each experiment. Stable adhesion was defined as: attachment without movement for a minimum of 10 seconds. Assays were run at least five times for each cell type studied. Student's t test was used to determine the significance of differences between sample means. For flow chamber assays, the unpaired Student's t test was used (Microsoft Excel). Values ofp < 0.05 were considered significant. Data are presented as mean ± SEM. 2.6 Adoptive Transfer models and Parabiosis 2.6.1 H Y adoptive transfer model Thymocytes from TCR-transgenic HY female mice (HY t g wild-type, HY t gC2GlcNAcT-I n u l 1, or HY t gCD43 n u l lC2GlcNAcT-I n u 1 1) (3 x 107 cells) were treated with 5 pM CFSE (Molecular Probes) and adoptively transferred into normal male or female C57BL/6-recipient mice. After 45 or 72 hours, spleen cells were harvested and stained for CD8, T3.70 (hybridoma kindly supplied by Dr. H Teh and biotinylated in our 113 laboratory), 1B11, and P-selectin-hlG. The responding TCR-transgenic cells were assessed by flow cytometric analysis, based T3.70 positivity. Those cells that were T3.70 positive (expressed the TCR-transgene) were assessed for their ability to bind P-selectin-hlG and 1B11 and their proliferative response was measured through tracking CFSE dilution. Those cells that are injected into a female recipient do not see male HY antigen and hence do not become activated and do not proliferate (one CFSE bright population). Whereas, those cells that are injected into male recipients see male HY antigen and hence become activated and proliferate (CFSE dilution is evident). These studies are presented as part of CHAPTER 3. 2.6.2 Competitive Repopulation Results from these experiments are presented in CHAPTER 5 of this thesis. Into IL-7Rnu" Non-irradiated eight-to-10-week-old IL-7Rn u" mice received intravenously 2x107 total bone marrow cells from five-week-old donors at a 1:1 ratio of the following donor strains: C2GlcNAcT-I n u l lC57BL/6C D 4 5 2 versus C2GlcNAcT-I n u l l C57BL/6 C D 4 5 1 / C D 4 5 ' 2 ; C2GlcNAcT-I n u l ,C57BL/6C D 4 5- 2 versus wild-type C 5 7 B L / 6 C D 4 5 1 / C D 4 5 2; C2GlcNAcT-I n u l l C 5 7 B L / 6 C D 4 5 1 / C D 4 5 2 versus PSGL-l n u , 1 C57BL/6 C D 4 5 - 2 and; wild-type C 5 7 B L / 6 C D 4 5 1 / C D 4 5 2 versus PSGL-l n u l , C57BL/6 C D 4 5 2 . Thymi were harvested three weeks after bone marrow transfer and single-cell suspensions stained for CD4+, CD8 + and congenic markers (CD45.1, CD45.2 and Thyl.2). The endogenous contribution to the DP T cells population is insignificant consistently below 1%. Into irradiated wild-type-C57BL/6 or P-selectinnu" 114 Lethally irradiated (9.5Gy) eight-to-10-week-old C57BL/6 mice (CD45.1/5.2 or P-selectinnu11) received intravenously lxlO 6 total bone marrow cells from five-week-old donors at a 1:1 ratio of the following donor strains: C57BL/6 T h y l 1 versus Thyl.2 congenic C57BL/6, C2GlcNAcT-In u" or PSGL-l n u 1 1 mice. Thymus, spleen and lymph nodes were harvested three to six weeks after bone marrow transfer and single-cell suspensions stained for CD4, CD8 and congenic markers. The frequency of endogenous DP T-cells that survived the total irradiation was estimated by staining for CD45.1, and was determined as consistently below 5%. The frequency of Thy 1.1 and Thyl.2 cells derived from donor bone marrow was analyzed, as described above. 2.6.3 Parabiosis Parabiotic pairs of mice were surgically generated by Lin Y i from Rossi's lab, as previously described (Abkowitz et al., 2003; Wright et al., 2001). All mice were female, age- and weight-matched and housed together for two weeks prior to surgery. Briefly, a skin incision running from elbow to knee along the flank was generated on opposite sides of the mice to be paired. The mice were first joined with a suture through the shoulder and thigh muscles, and next the inside face of the skin flaps were juxtaposed and sutured. Complete blood sharing (approximately 50% chimerism in each of the paired mice) was observed between four and seven days later. Mice were sacrificed five to six weeks after surgery and processed for flow cytometric analysis. Raw data were analyzed by one-way ANOVA. P values smaller than 0.05 were considered significant. All results are expressed as mean and all error bars represent SEM. These studies are also presented in CHAPTER 5. 2.7 Real-time RT-PCR Results from real-time RT-PCR are shown in Figure 3-9 and Figure 3-11 of CHAPTER 3. 115 AutoMACS purification of C D 4 + and C D 8 + cells Activated splenocytes (outlined in section 2.3) underwent positive selection to enrich for CD8 + or CD4 + T cells before RNA extraction using the AutoMACS cell-separation system (Miltenyi Biotec). Activated C2GlcNAcT-In u" and C57BL/6 T cell cultures were stained with ocCD4-PE or ocCD8-PE, washed, and resuspended in 80-microliters of buffer (0.5% BSA and 5 mM EDTA in PBS) per lx 107 total cells. 20-microliters of MACS anti-PE microbeads were added per l x l0 7 total cells, and the suspension was incubated for 15 minutes at 4°C, washed, and finally resuspended in 500-microliters of buffer per 1x10s total cells before sorting on the AutoMACS. RNA extraction and RT-PCR RNA was extracted from CD8 + T cells using TRIzol Reagent (Invitrogen), according to the manufacturer's instructions. Reverse transcription was performed with approximately 1 [ig of DNase-treated total RNA using Superscript (Invitrogen) in a 20-microliters volume. The reverse-transcription reaction was then diluted 20-fold. 15-microliters of a LightCycler Reaction Mastermix containing 9-microliters of water, 1-microliter of forward primer (0.5 uM), 1-microliter of reverse primer (0.5 pM) and 4-microliters of LightCycler FastStart DNA Master Plus SYBR Green I (Roche Diagnostic Systems) and 5-microliters of cDNA, added as a PCR template, were mixed and transferred into the LightCycler glass capillaries. The LightCycler experimental-run protocol used was as follows: denaturation program (95°C for five minutes), amplification, quantification program repeated 40 times (95°C for 10 seconds, 55°C for 10 seconds and 72°C for 30 seconds, with a single fluorescent measurement), melting curve program (60°C to 95°C with a heating rate of 0.2°C per second and a continuous fluorescent measurement), and finally a cooling step to 40°C. Transcription levels of different mRNAs were determined by preparing a standard curve for each of the 116 genes of interest and using hypoxanthine phosphoribosyltransferase (HPRT) as an internal reference gene to normalize the results. PCR primers for the core 2 isoenzymes were purchased from Qiagen, optimized to an annealing temperature of 55°C and were as follows: C2GlcNAcT-I (EMBL accession no. D87332), forward primer, 5' GAAGGACCTGTACAGAATGAATG-3', and reverse primer, 5'-ACTTGTTGCTTGAGGGGAAAGAA-3'; C2GlcNAcT-II (NCBI accession no. NM028087), forward primer, 5'-CTTGCTTCAGAGCCCCGTGCC-3', and reverse primer, 5'-GTTGCCGGGCTTTTGAGTTACTG- 3'; and C2GlcNAcT-III (Ensembl accession no. ENSMUST00000049586), forward primer, 5'-CCTCCTCAAGTCTTCCGTTCAG-3', and reverse primer, 5' -GGAGACGTTCGTCTTTACTGG-3'. PCR primers for HPRT (EMBL accession No. BC004686) were exon 7 forward primer, 5'-CTC GAAGTGTTGG AT AC AGG-3', and HPRT exon 9 reverse primer, 5'-TGGCCTATAGGCTCATAGTG-3' (Invitrogen), and were specific for separate exons to be able to detect potential contaminating genomic DNA. Genomic contamination was tested by performing HPRT-specific PCR on DNase-treated RNA samples for presence of a 1100-bp product that would have indicated that genomic DNA had resisted the DNase treatment. Duplicate samples that had no cDNA added were included with each batch of PCR to check for external contamination. Data Analysis Data analysis was performed with the relative expression software tool (REST) using HPRT as the internal reference gene (Pfaffl et al., 2002). This software package compares the levels of transcript present in different tissue samples and performs a pairwise-fixed reallocation randomization test to determine the statistical significance of the changes in transcript level. The results for target and reference genes were additionally corrected 117 according to their calculated PCR efficiencies. Finally, the PCR product was verified for size by gel electrophoresis on a 1.5% agarose Tris Borate EDTA gel, and their identity confirmed by sequencing of the PCR products. 118 CHAPTER 3 P-SELECTIN BINDS CD8+ T CELLS IN THE ABSENCE OF C2GlcNAcT-I1 3.1 Introduction PSGL-1 is the best-characterized selectin ligand and, while all three selectins interact with PSGL-1, it is P-selectin's primary ligand. PSGL-1 is expressed as a functional ligand of P-selectin on myeloid cells and T cell subsets. The interaction of PSGL-1 with P-selectin requires core 2-modified branching of a threonine residue in the amino terminus of PSGL-1, as discussed in the introduction and illustrated in Figure 1-7. Selectin-binding studies reveal that C2GlcNAcT-I is essential for P-selectin ligand activity, while only a subset of L-selectin and E-selectin ligands depend on C2GlcNAcT-I for recognition. Despite C2GlcNAcT-I's essential role in P-selectin ligand formation, C2GlcNAcT-Inu11 mice possess a relatively mild phenotype. They show only partial reduction in selectin ligands, while lymphocyte homing in these mice is normal, reflecting the possible redundancy of C2GlcNAcT isoenzymes. There are two other isoenzymes capable of forming core 2 branches, C2GlcNAcT-II and C2GlcNAcT-III. The role of these enzymes in modifying selectin ligands has yet to be addressed. The objective of our studies is to determine whether these isoenzymes have a role in P-selectin ligand formation in the absence of the C2GlcNAcT-I enzyme. In this chapter I show, in vitro and in vivo cell activation conditions that stimulate P-selectin binding to PSGL-1, in the absence of C2GlcNAcT-I, are defined. Activated C2GlcNAcT-Inu" CD8+ T cells form functional PSGL-1 recognized by P-selectin, while activated C2GlcNAcT-Fu" CD4+ T cells do not; this implicates alternate C2GlcNAcTs in the biosynthesis of functional P-selectin ligands in CD8+ T cells. ' A version of this chapter has been published. Merzaban, J.S. and Zuccolo, J., Corbel, ML, Williams, M.J., and Ziltener, H.J. (2005) An Alternate Core 2 p%6-ALAcetylglucosammyltransferase Selectively Contributes to P-Selectin Ligand Formation in Activated CD8 T Cells. Journal of Immunology. 174:4051-4059. 119 3.2 In vitro results 3.2.1 P-selectin binds PSGL-1 on C2GlcNAcT-Inu" Con A-activated Splenocytes Mice lacking C2GlcNAcT-I exhibit a defect in leukocyte recruitment to sites of inflammation, due to a deficiency in P-selectin ligand formation on neutrophils (Ellies et al., 1998) and CD4+ T cells (Snapp et al , 2001). Studies on wild-type (C57BL/6), activated CD8 + T cells show that P-selectin ligand is preferentially expressed when cells are grown with IL-2 due to induction of C2GlcNAcT-I. These studies also display that cell culture densities ranging from 0.05x106 to 0.25x106 cells per millilitre significantly affect the level of P-selectin ligand formation in activated splenocytes. Cells grown at the higher densities express a higher level of P-selectin binding than do cells grown at lower densities (Carlow et al., 2001a). In addition, these studies do not provide evidence for P-selectin binding on C2GlcNAcT-In u" cells grown at either of the densities tested (Carlow et al , 2001a). Striving to search for functional P-selectin ligand expression on T cells in the absence of C2GlcNAcT-I, different in vitro culture stimulation conditions were tested. Splenocytes from C2GlcNAcT-In u 1 1 mice were activated with Con A for 48 hours and then subcultured in IL-2 for an additional 48 hours, under varying, high-density culture conditions, ranging from 0.25xl06 to 2.0xl06 cells per millilitre. Figure 3-1 outlines the in vitro culture stimulation conditions that are the basis for many studies performed in this chapter. 120 Day 0 Day 1 Day 2 Day 3 Day 4 ^0-5x10 6 cells/ml —> —> Spleen Cells > > — • 1X106 cells/ml Con A ^ IL-2 > 2x106 cells/ml Figure 3-1: In vitro culture stimulation conditions. Spleens from C57BL/6, C2GlcNAcT-In u 1 1, PSGL-l n u " and PSGL-1 n u l l/C2GlcNAcT-I n u 1 1 mice were harvested on day 0 and cultured with con A for 48 hours. On day 2, cells were washed and re-plated at the indicated cell culture densities with IL-2. Cells were harvested at various time points for analysis. These experiments indicate that cells cultured under the highest densities exhibit significant P-selectin binding, as shown in Figure 3-2 A. Time course studies demonstrate tight temporal control of P-selectin ligand formation with significant P-selectin binding on days 3 and 4, followed on day 5 by a loss of binding (Figure 3-2B). P-selectin binding is specific for PSGL-1 since binding is inhibited with either the neutralizing a-PSGL-1 antibody, 2PH-1 (Figure 3-2B), or treatment with 5mM EDTA (Figure 3-2A and B). To further confirm PSGL-1 as the P-selectin ligand in question, PSGL-1 n u l 1 splenocytes were tested under these assay conditions and conveyed no detectable P-selectin ligand when activated under high-density conditions (Figure 3-2C). Using a non-glycosylation-dependent antibody to measure PSGL-1 expression, similar levels of PSGL-1 glycoprotein exist on activated splenocytes grown at different cell culture densities from C57BL/6 and C2GlcNAcT-l n u U mice (Figure 3-2D). This indicates that any discrepancies in P-selectin binding result from differences in 121 modifications on the PSGL-1 glycoprotein, not variances in the level of PSGL-1 expression on the surface of each of the cell types. The in vitro culture conditions tested above (Figure 3-1) lead to induction of functional P-selectin ligand on activated T cells in the absence of core 2 activity modulated by the C2GlcNAcT-I enzyme. These data indicate that high-density conditions induce core 2 enzymatic activity independent of C2GlcNAcT-I, and this alternate core 2 enzyme can modify PSGL-1. Figure 3-2: P-selectin ligand formation in Con A-activated C2GlcNAcT-Inul splenocytes is cell-culture-density dependent. A, P-selectin ligand expression (red line) on mitogen-stimulated C2GlcNAcT-In u" spleen cells maintained at culture densities of 0.5x106, l.OxlO6 and 2.0x106 cells per millilitre. B, P-selectin ligand expression on mitogen-stimulated C2GlcNAcT-In u 1 1 splenocytes maintained for 3, 4 and five days (2.0x106 cells per millilitre, red line); P-selectin-hIG binding is blocked by the anti-PSGL-1 mAb 2PH-1 (black line). C, P-selectin-hIG binding on activated PSGL-1 n u" splenocytes cultured at either 0.25xl06 cells per millilitre (black line) or 2.0xl06 cells per millilitre (red line). D, PSGL-1 expression on day 4-activated C57BL/6 and C2GlcNAcT-In u 1 1 splenocytes cultured at 0.5xl06 cells per millilitre (dashed line), l.OxlO6 cells per millilitre (black line) or 2.0xl06 cells per millilitre (red line). Five mM EDTA (dashed line) was added during incubation with P-selectin-hlG as a negative control in all experiments. These experiments were done >30 times with the same reproducible results. 122 o 03 relative cell number relative cell number "0 GO G) relative cell number 2 i w CD. CD O m 1 ''"""it—. D QJ *< CO C/> CD_ CD O 5' 1 Q relative cell number i 1 to 3.2.2 Increased Core 2 O-glycan branching on CD43 and CD45 correlates with increased P-selectin ligand formation To confirm that core 2 enzyme activity correlates with P-selectin ligand formation observed in the experiments outlined in the previous section, several approaches were taken: 1) Core 2 and core 4 in vitro enzyme assays were performed using substrates based on physiological substrates for core 2- and core 4-branching enzymes (outlined substrates and enzyme activities are depicted in Figure 1-3); 2) Core 2 modifications on CD43 were measured by reactivity with the anti-CD43 mAb, 1B11 both by FACS and by Western analyses; and 3) Core 2 modifications on CD45RB were measured by reactivity with the 1B11 mAb on CD43 n u 1 1 cells. Core 2 enzyme activity was measured using a standard enzyme assay of lysates taken from activated C2GlcNAcT-In u 1 1 cells. Activated splenocytes cultured at the highest densities (2.0 x 106 cells per millilitre) reveal significantly more, about twice as much, core 2 activity than those cultured at lower densities (0.25 x 106 cells per millilitre) for both C57BL/6 and C2GlcNAcT-In u 1 1 cells (Figure 3-3). Not surprisingly, C2GlcNAcT-I n u" cells express lower activity than corresponding wild-type cells. 124 Figure 3-3: High-density (HD) cultured C2GlcNAcT-Inu" Con A blasts express significant core 2 activity. Lysates from H D (2.0x10 6 cells per millilitre) and low-density ( L D ; 0 .25xl0 6 cells per millilitre) cultured C57BL/6 or C 2 G l c N A c T - I n u 1 1 splenocytes were tested for core 2 enzyme activity. Values were normalized against background levels obtained from assays done on samples in the absence of the acceptor substrates. C2GlcNAcT- I I has exhibited core 2 and core 4 O-glycan branching activity (Schwientek et al., 1999; Y e h et al., 1999), whereas C2GlcNAcT-II I is specific for core 2 O-glycan branching (Schwientek et al., 2000). In order to help identify which isoenzyme is responsible for the core 2 branching activity in absence of C 2 G l c N A c T - I , lysates were tested in vitro for core 4 enzyme activity using the core 3 substrate, G l c N A c | 3 l -3Ga lNAca-pNp , as an acceptor (refer to Figure 1-3). N o core 4 activity is found in these lysates, although, as expected, lysates prepared from intestinal tissue, which express high levels of C2GlcNAcT- I I enzyme, display significant levels of activity, as shown in Figure 3-4. This suggests that the enzyme involved in functional P-selectin ligand formation is most likely not the C2GlcNAcT- I I isoenzyme. 125 Figure 3-4: HD-cultured C2GlcNAcT-In u u Con A blasts do not contain core 4 activity. Lysates from HD (2.0x106 cells per millilitre) and LD (0.25xl06 cells per millilitre) cultured C57BL/6 or C2GlcNAcT-I n u U splenocytes were tested for core 4 enzyme activity. Intestinal tissue lysate from a C57BL/6 mouse was used as a positive control. Values were normalized against background levels obtained from assays done on samples in the absence of the acceptor substrates. CD43 is the most-highly expressed sialoglycoprotein on the surface of most T cells (Fukuda, 1991). It is a well-established substrate of C2GlcNAcT-I (Barran et al., 1997; Higgins et al., 1991) and, as discussed in section 1.6.1 above, there are two glycoforms of CD43 that result due to lack or presence of C2GlcNAcT activity: a 115 kDa form and a 130 kDa form, respectively (Figure 1-9). Ectopic expression of the two human core 2 isoenzymes, hC2GlcNAcT-II and hC2GlcNAcT-III, increases CD43 O-glycan branching (Schwientek et al., 2000; Yeh et al., 1999), showing that CD43 is not just a predominant substrate for C2GlcNAcT-I. To determine whether the residual core 2 activity observed in C2GlcNAcT-In u" cells targets physiological substrates other than PSGL-1, we evaluated properties of CD43 with regard to changes in its electrophoretic mobility and to reactivity with the core 2 glycosylation-sensitive anti-CD43 mAb 1B11 (Carlow et al., 1999; Ellies et al., 1996; Jones et al., 1994). The anti-CD43 Abs used and their specificities are outlined in Figure 3-5. 126 CD43 immunoprecipitated from cultures of activated, wild-type and C2GlcNAcT-I n u 1 1 splenocytes with the pan-CD43-specific antibody, HI 8, is detected with the pan CD43 mAb SI 1 in lysates of unstimulated wild-type and C2GlcNAcT-In u l 1 lymph node cells (LNC), while mAb 1B11 reactive core 2-branched CD43 is virtually absent (Figure 3-6). mAb 1B11 reactive 130-kDa CD43 is strongly detected in activated wild-type cells. CD43 precipitated from activated C2GlcNAcT-Fu 1 1 cells cultured at 0.5x106 cells per millilitre display a relatively modest 1B11 signal, while a significant increase in the 1B11-CD43 signal is observed with CD43 precipitated from cultures grown at higher densities (l.OxlO6 cells per millilitre and 2.0xl06 cells per millilitre). Figure 3-5: CD43 antibody recognition. HI8 polyclonal antibody recognizes CD43 irrespective of glycosylation status and is effectively used for immunopreciptations of CD43. mAb SI 1 also recognizes CD43 peptide and is not glycosylation dependent; therefore SI 1 recognizes all forms of CD43. Glycosylation-specific mAb 1B11 recognizes the core 2-branched form of CD43 that results from activity of the C2GlcNAcT enzymes; it is, therefore an excellent marker of the presence of C2GlcNAcT activity. 127 200kD—[ Day 4 Con A IL-2 LNC B6 C2GlcNAcT-l n u l 1 B6 C2 2 2 1 0.5 xlO6 culture density S11 IP: pan-CD43 IB: S l l or 1B11 1B11 Figure 3-6: Western blot showing that CD43 is modified by core 2 branches in the absence of C2GlcNAcT-I. CD43 was immunoprecipitated from day 4-activated C2GlcNAcT-I n u 1 1 splenocytes grown at 0.5xl0 6 (0.5), l.OxlO 6 (1) or 2.0xl0 6 (2) cells per millilitre using the pan-CD43 Ab, H18. Samples were subsequently immunoblotted with apan-CD43 mAb, S l l and the core 2-dependent mAb, 1B11. Activated splenocytes and naive lymph node cells (LNC) (negative control; no C2 activity in naive cells) from wild-type C57BL/6 (B6) (positive control; activated, wild-type cells have elevated C2GlcNAcT-I enzyme) cultures were run in parallel as controls. MAb 1B11 has dual specificity for core 2 O-glycan-dependent epitopes on both CD43 and CD45RB. Core 2 O-glycans are required for 1B11 recognition of the CD43 epitope (Figure 3-5), whereas 1B11 recognition of CD45RB (Carlow et al., 1999), exclusively expressed on CD8 + T cells, indicates lack of core 2 O-glycans (discussed in section 1.6.1). On CD43 + C D 8 + T cells, 1B11 reactivity with CD45RB is obscured by reactivity with CD43. FACS analysis of activated CD43 +C57BL/6 and CD43 + C2GlcNAcT-I n u l 1 splenocytes shows that P-selectin-positive cells have an increased 1B11 signal (Figure 3-7A). These results indicate that core 2 activity present in 128 these cultures is able to produce functional P-selectin ligand and modify CD43, which correlates with Figure 3-6. On CD43 n u 1 1 CD8 + T cells, 1B11 reactivity is exclusively CD45RB-dependent and FACS analysis of activated CD43 n u l l C2GlcNAcT-I n u 1 1 splenocytes conveys that P-selectin-positive cells have a decreased 1B11-CD45RB signal (Figure 3-7B) (indicating an increased C2GlcNAcT enzyme activity). These data suggest that the residual core 2 activity detected in these C2GlcNAcT-I n u 1 1 splenocytes targets CD43 and CD45, in addition to PSGL-1. C2GlcNAcT-l n u» C57BL/6 itf-(3 sz c X TJ <D u_ <u CO io1-CL B 10° 101 1# 103 1 Cf* FL2-H 1(P 101 102 103 FL2-H C2GlcNAcT-l n u l CD43 n u " 1B11 (CD43) I CD43 n u " 1B11 (CD45) T ™ , — I ' I—' 1(P 101 102 103 > FL2-H 10» Figure 3-7: CD43 and CD45 on C2GlcNAcT-jnuii C D g + j c e U s a r e modified by an alternate C2GlcNAcT enzyme. Spleen cells prepared from C2GlcNAcT-I n u " (A), C57BL/6 (A), CD43 n u l l /C2GlcNAcT-I n u " (B) and CD43 n u 1 1 (B) were activated as before and harvested on day 4. A three-colour flow cytometric analysis was performed by staining for CD8, P-selectin-hIG and 1B11 to determine both P-selectin ligand expression and the glycosylation status ofCD43 (1B11-CD43) (A) and CD45 (1B11-CD45) (B) on CD8 + T cells. 129 3.2.3 C2GlcNAcT-I RNA and C2GlcNAcT-III RNA are expressed in activated splenocytes Identity between C2GlcNAcT isoenzymes Core 2 O-glycan branching is required for binding of P-selectin to PSGL-1, thus, activity of an alternate C2GlcNAcT isoenzyme is the likely cause for P-selectin ligand formation observed in C2GlcNAcT-Fu 1 1 splenocytes. In the human system, two additional enzymes capable of catalyzing core 2 O-glycan structures are identified, expressed and characterized. A database search identified the murine orthologs of C2GlcNAcT-II and C2GlcNAcT-III. Figure 3-8 shows multiple-amino-acid sequence alignment (ClustalW) of murine C2GlcNAcT-I, C2GlcNAcT-II and C2GlcNAcT-III. There is significant amino acid sequence identity among the three mouse C2GlcNAcT isoenzymes. C2GlcNAcT-II has 47% identity to C2GlcNAcT-I, and C2GlcNAcT-III has 41% identity to C2GlcNacT-I. High sequence similarity occurs in the catalytic domain. The spacing of nine cysteine residues is conserved in all three C2GlcNAcTs, as shown in Figure 3-8. There is one conserved potential N-linked glycosylation site located in the stem region of all three isoenzymes which is shown as essential for catalytic function of C2GlcNAcT-I (Toki et a l , 1997). 130 Figure 3-8: Murine C2GlcNAcT-I, C2GlcNAcT-II and C2GlcNAcT-III show significant sequence identity. Multiple amino-acid-sequence analysis (Clustal W) of the murine C2GlcNAcT is shown. Dark-gray regions depict identity and light-gray regions depict similarity. N-terminus begins with residue 1. Beginning at the N-terminus, there is a short cytoplasmic tail, followed by a stem region and transmembrane domain (predicted by a TMpred-algorithm-MacVector; underlined in red) and ending with a large catalytic domain. The C-terminal catalytic domain displays the most homology, while N-terminal cytoplasmic, stem and transmembrane regions show much less homology. The positions of conserved cysteines are indicated by asterisks. One conserved N-glycosylation site is indicated by a green circle. C2GlcNAcT-l C2GlcNAcT-ll C2GlcNAcT-lll C2GlcNAcT-l C2GlcNAcT-ll C2GlcNAcT-lll 1 I M L riN L F R Q R [ ! M K I FLB C 46 27 . J C P T K Y \ 1 B M E E - - - -LB Q S. K Y T L Q Q K L F I L Q N Q sLll rQ. Y C R D L • LTJ T Q W T R I H offl P E F F S V R H 45 L K ii P A L B - - - - - 26 L K U L N V G R L L F P Q RTD I V L \ l 47 48 I F . Y S I , S T S P F V R N " ! F P E O * L E L Al G D D P Y l S N V N C J T K l L - 1 fs~i1 N C S G V J I SGli A A R g N V N C S f i ^ Y E El E ll c f j V KpO K A V T O A L!_|J LI F. il C L H S - -IlT V Q F L E ll K K I E ll R R 82 54 93 C2GlcNAcT-l C2GlcNAcT-ll C2GlcNAcT-lll Boo P L S R S I I D P FJ D \j I E A D_S L N M l R M l L E D d_rj V V A] M l 3 D C A D C E D O D V Y Y a I V V H M V V H E C2GlcNAcT-l 133 C2GlcNAcT-ll 104 C2GlcNAcT-lll 144 C2GlcNAcT-l 183 C2GlcNAcT-ll 154 C2GlcNAcT-lll 194 T T S L J R L L R i TTN FTE R L L R A ll MVI F. R Tl TI R A I Y A0 } T| A I -MN Q P Q N F Y C I H M d R) !>j VJ Y cQ E V I C K ; V C 1 B MJJ L j j , *_ O sff l LJ A A V) 0 d l A S C I P F K3 A_y R A! I |M S C 1 p D TI F K A 4M N N L L J K I X J S V V Y A s vtrJ R VLKI A D L N C _5 V V Y A S W S R V Q A D L N C T I V T E I Y J I I H ll S ITU O A QW IN P L * D L N C M K D L | Y R J N A N W K Y L l N L C C V D F P I K T 1> L| D L N CJB E D L l l ( f s p[\j P W K Y [TO i Q c C T D F P I K T ? D W N O L SI D L ll K|_a S N d W K V M I N L C d (j D F pTJ M s l 1> t I i L| 132 103 V V H PC D 143 S  FLd N V F|_VJ A sLd~il 182 F P N V F I A S K I 153 r F P l f i l F I A S K l l 193 M R K 232 K A 203 T E 243 C2GlcNAcT-l 233 C2GlcNAcT-ll 204 C2GlcNAcT-lll 244 c s TT3 E[~3 M L E T Lpf l K c q * s"3 G SLJJ oLa RI N \ T L E ii K j v l p rt p P_pj v RLPFTS N H ALM H E|_E R W I TI E rt F L R q vl P Y P Y M ri L KTIIT NTI T L J U H Pv 1 p v iqj G I Vi S K R| N V H TL G 276 246 293 C2GlcNAcT-l 277 C2GlcNAcT-ll 247 C2GlcNAcT-lll 294 L K T P 1 I fTS L T V I K|_a i q VLJ GIS A Y_F V o 3 A \ L M V G S A Y F V I N I Q [ K A R C L H H L V E D F F •3 A d D T Y S P D E L F J L W A i t l 326 E J j J ^ W A 3 S I K P T Y S P D E I W A T l l 343 C2GlcNAcT-l 327 C2GlcNAcT-ll 297 C2GlcNAcT-lll 344 i C2GlcNAcT-l 377 C2GlcNAcT-ll 347 C2GlcNAcT-lll 388 376 346 387 R| K Al L E N L 426 FU£_a I Y G T 396 E Q Q R K L I 437 C2GlcNAcT-l 427 f l H 428 C2GlcNAcT-ll 397 [ E L ] 398 C2GlcNAcT-lll 438 A I L S S E K F M T E G T R O S H T L H I 456 131 C2GlcNAcT isoenzyme RNA expression in activated T cells - low-density versus high-density cultures We carried out real-time RT-PCR to identify and quantify levels of C2GlcNAcT isoenzyme RNA in activated CD8 + T cells from wild-type and C2GlcNAcT-I n u l 1 splenocytes. The analysis reveals significant levels of C2GlcNAcT-I RNA in activated T cells from wild-type control mice, while no signal was detected in C2GlcNAcT-In u" T cells (Figure 3-9A). RNA for C2GlcNAcT-III is detected while that for C2GlcNAcT-II is absent in all splenocyte samples tested, consistent with data for human C2GlcNAcT-II described as present in mucosal tissue and absent in spleen (Yeh et al., 1999). RNA for C2GlcNAcT-III occurs at comparable levels in activated T cells from both wild-type and C2GlcNAcT-In u 1 1 mice (Figure 3-9B), implicating C2GlcNAcT-III as the most likely candidate source of residual core 2 activity found in the activated C2GlcNAcT-In u" T cells. These data imply that the enzyme responsible for modifying PSGL-1 in high-density cultures, in absence of C2GlcNAcT-I, is C2GlcNAcT-III (Figure 3-9A). Although we expected to see a measurable difference in the level of C2GlcNAcT-III expression in the high-density cultures versus the low-density cultures, this was not the case (Figure 3-9B). Neither the C2GlcNAcT-I enzyme nor the C2GlcNAcT-III enzyme was significantly upregulated in high-density cultures from wild-type and C2GlcNAcT-In u 1 1 mice. This statistical analysis was determined using a software program, REST, which uses a pairwise-fixed reallocation randomization test to determine statistical significance of the changes in transcript level with reference to the HPRT house-keeping gene. Although it is generally believed that control of the key glycosyltransferases occurs via gene induction and that there is no significant 132 posttranslational mechanisms that control the activity of these enzymes (Ley and Kansas, 2004), data outlined here suggests that post-transcriptional and/or posttranslational mechanisms do exist; cell preparations that differ approximately ten fold in core 2 enzyme activity do not show significant differences in C2GlcNAcT-III RNA. Indeed evidence for posttranslational control of C2GlcNAcT-I is documented (Ben-Mahmud et a l , 2004; Chibber et al., 2000; Chibber et al., 2003). 1000 850 C2GlcNAcT-l C2GlcNAcT-lll 1 2 311 2 311 2 3 j 1 2 3|1 2 3 | L HD LD HD LD C2GlcNAcT-l n u l 1 C57BL/6 No cDNA B C2- HP W T - H D C2-LD WT - LD W T - H D W T - L D C2-HD C2-LD Figure 3-9: RT-PCR demonstrates expression of C2GIcNAcT-I and C2GlcNAcT-ffl RNA in activated CD8+ T cells. A, RT-PCR was conducted for C2GlcNAcT-I (lane 1), C2GlcNAcT-II {lane 2) and C2GlcNAcT-III (lane 3) from RNA isolated on day 3 from CD8 + T cell-enriched C2GlcNAcT-I n u" or C57BL/6 high-density (HD; 2.0xl0 6 cells per millilitre) and low-density (LD; 0.25xl0 6 cells per millilitre) cultures. RT-PCR products were separated on an agarose gel; D N A ladder is standard (L). No cDNA controls were performed as described in Materials and Methods. B, Comparisons of relative C2GlcNAcT-III RNA expression levels were determined by real-time RT-PCR. C D 8 + T cells from HD and LD splenocyte cultures are compared among and between C2GlcNAcT-I n u " (C2) and wild-type C57BL/6 (WT). Data analysis was carried out with a relative quantification software tool using HPRT as internal reference gene (see section 2.7). Differences were not statistically significant. 133 3.2.4 P-selectin ligand formation in activated C2GlcNAcT-Inu" cells is restricted to CD8+ T cells Activated GD4+ and CD8 + T cells express high levels of P-selectin ligand. In the case of CD4 + T cells, previous data (Snapp et al., 2001; Wagers et al., 1998) indicate that C2GlcNAcT-I expression is essential for generating ligands for P-selectin in primary CD4 + T cells. A comparative analysis (Snapp et al., 2001) of rolling of activated CD4 + T cells from C2GlcNAcT-In u 1 1 and wild-type mice shows that activated C2GlcNAcT-In u" CD4 + T cells display no rolling. Given that Con A-mitogenic stimulation of splenocytes results in cultures that predominantly contain CD8 + T cells, we examine whether P-selectin ligand formation is restricted to CD8 + T cells. A comparison of P-selectin ligand formation on CD4 + against CD8 + cells in cultures of C2GlcNAcT-In u" splenocytes was performed. Both CD4 and CD8 markers were used in the staining along with P-selectin, and the cells were analyzed using FACS. As Figure 3-1 OA demonstrates, CD8 + T cells display significant P-selectin binding, whereas CD4 + T cells display only weak P-selectin staining. To discern whether the CD4 + T cells have less C2GlcNAcT activity than CD8 + T cells, 1B11 reactivity with CD43 was measured on each cell subset. As shown in Figure 3-10B, C2GlcNAcT-In u 1 1 CD8 + cells exhibit significantly more 1B11 reactivity than C2GlcNAcT-In u 1 1 CD4 + cells, indicating that, at least in terms of modifying PSGL-1 and CD43, core 2 activity in CD4 + cells is considerably lower or even absent compared with CD8 + cells. 134 Figure 3-10: P-selectin ligand formation in Con A-activated C2GlcNAcT-ruu cells is restricted to CD8+ T cells. A, Activated C2GlcNAcT-In u" spleen cells were analyzed by FACS to determine P-selectin ligand formation on CD4 + and CD8 + T cell subsets (red line). As a negative control, binding of P-selectin-hIG, was inhibited with either EDTA (dotted line) or by blocking with the anti-PSGL-1 mAb, 2PH-1 (black line). B, 1B11 binding of CD43 (1B11-CD43) was determined on CD8 + and CD4 + subsets from C2GlcNAcT-In u 1 1 (shaded graph)-activated splenocytes and wild-type (red line)-activated splenocytes. Isotype control staining for 1B11 is shown in the dashed line. 135 The data in Figure 3-10 clearly convey that activated CD4 + cells differ from activated CD8 + cells. Residual core 2 activity in C2GlcNAcT-In u 1 1 CD8 + cells but not CD4 + cells modifies both PSGL-1 and CD43. To further understand the differences that exist between the two subsets of T cells, RNA expression profiles for the C2GlcNAcT isoenzymes were determined in parallel with core 2 enzyme activity assays. As shown in Figure 3-11 A, activated CD8 + T cells expressed significantly more C2GlcNAcT-III transcript (normalized to the housekeeping gene, HPRT) than activated CD4 + T cells, which correlates with a similar difference in core 2 enzymatic activity (Figure 3-1 IB). This suggests that these culture conditions selectively induce an alternate core 2 enzyme, likely C2GlcNAcT-III, in CD8 + T cells, while CD4 + T cells do not express this enzyme. Figure 3-11: mRNA expression and activity of activated CD4+ and CD8+ C2GlcNAcT-Inu" T cells. A, RT-PCR demonstrates expression of C2GlcNAcT-III RNA in activated CD8 + and CD4 + cells, grown in the presence of IL-2 at high density (2.0xl06 cells per millilitre). C2GlcNAcT-III transcript levels present in CD8 + are normalized to 100% and the level of C2GlcNAcT-III transcript in CD4 + cells is expressed as a percent of what is expressed in CD8+. B, Lysates from either CD4 + or CD8 + cells were tested for core 2 enzyme activity. Data analysis was carried out with REST using HPRT as an internal reference gene. 136 3.2.5 Induction of core 2 activity by a secreted/soluble factor within high-density culture supernatant Induction of P-selectin ligand on CD8 + T cells is not easily noticed. T cell culture conditions that are normally used for splenocyte activation do not result in P-selectin ligand formation (Carlow et al., 2001a). Only when five-to-10-fold higher cell culture densities are applied is there a reproducible induction of P-selectin ligand (Figure 3-2A). In addition, P-selectin ligand expression appears to be under tight temporal control; cells analyzed at days 3 and 4, after mitogen stimulation, display significant binding, whereas cells outside of this time window do not bind P-selectin (Figure 3-2B). To elucidate the mechanism responsible for inducing the activity of alternate C2GlcNacTs, we completed a number of experiments. To determine whether induction is based on cell-cell contact or due to a soluble factor, cell culture supernatant from day 3 and day 4 cultures was added to C2GlcNAcT-In u 1 1 cells grown at low density (0.25x106 cells per millilitre), along with IL-2, as shown in Figure 3-1. Figure 3-12 illustrates that when supernatant from high-density cultures is added to low-density cultures P-selectin ligand is induced and is evident for both CD8 + as well as CD4 + subsets. The 1B11 CD43 signal also increases on incubation with high-density supernatant, indicating that the core 2 activity induced by this soluble factor is, in addition to PSGL-1, acting on CD43 (Figure 3-12). 137 LD + SN o 0 to CD8 CD4 10u 10 10' 1B11 Figure 3-12: High-density supernatant induces P-selectin ligand formation and core 2 branching on CD43 in C2GlcNAcT-InulI-activated T cells grown at low density Spleen cells prepared from C2GlcNAcT-Inu11 mice were cultured, as in Figure 3-1, at low density (LD; 0.25xl06 cells per millilitre) or high density (HD; 2.0xl06 cells per millilitre), or at low density along with 50% supernatant (SN) from a previous high-density culture. On day 4, cells were stained for CD4 and CD8, 1B11 and P-selectin-hlG. 3.2.6 IL-2 and IL-4 differentially regulate the P-selectin ligand formation Polarization toward activated T H I (Tcl) and TH2 (TC2) subsets is known to affect C2GlcNAcT-I and FucT-VII enzyme expression required to generate P- and E-selectin ligands (Blander et al., 1999; Carlow et al, 2001a; Lim et al., 1999; Lim et al, 2001; Wagers et al., 1998; White et al, 2001). Previous studies in our lab exemplify that IL-2 and IL-4 differentially regulate PSGL-1 formation in activated CD8+ T cells from wild-type mice by regulating the expression of C2GlcNAcT-I (Carlow et al, 2001a). Since the residual core 2 activity found in the absence of the C2GlcNAcT-I enzyme appears to mimic C2GlcNAcT-I in substrate specificity, we were interested to determine if there are differences in its regulation by cytokines. C2GlcNAcT-Inu11 splenocytes were cultured at high density with either IL-2 or IL-4 and analyzed for P-selectin ligand formation. Figure 3-13 A shows that IL-2 preferentially results in P-selectin binding on C2GlcNAcT-InuI1-138 activated cells cultured at 2.0x106 cells per millilitre, whereas IL-4 does not. This differential regulation of PSGL-1 formation on activated T cells is reminiscent of the regulation of C2GlcNAcT-I described by Carlow et al. (Carlow et al., 2001a) (Figure 3-13B), and suggests that alternate C2GlcNAcT enzymes may be receptive to similar cytokine regulation. Figure 3-13: Regulation of C2GlcNAcT by IL-2 and IL-4. Splenocytes from C2GlcNAcT-In u" (A) and C57BL/6 (B) mice were activated with Con A and either IL-2 (red-shaded graph) or IL-4 (black line) at high density (2.0x106 cells per millilitre). P-selectin-hlG binding was assessed by FACS on day 4 of the culture. EDTA controls were negative for P-selectin-hIG binding, but are not shown. 139 3.2.7 Activated C2GIcNAcT-InuU cells roll on immobilized P-selectin in a PSGL-1-dependent manner Physiologically relevant selectin interactions with their ligands can be identified by testing the tethering and rolling of leukocytes on endothelial cells or on recombinant molecules in flow chamber systems in vitro or in microvessels in vivo. In vitro flow chamber assays are used to effectively determine the functional ability of cell populations to roll under defined shear flow conditions. Activated C2GlcNAcT-In u 1 1 T cells, from cultures grown at high density, attach and roll on immobilized P-selectin-hIG at 2 dyne per square centimetre, while attachment and rolling observed for cells activated under lower-density conditions are significantly reduced (Figure 3-14). The number of rolling cells from C2GlcNAcT-In u 1 1 mice is approximately half the number of rolling cells observed for wild-type C57BL/6 mice. Adding EDTA to this system completely abolishes all rolling, reflecting the bivalent cation dependency of the rolling process. As expected, activated T cells from PSGL-1 n u l 1 mice show no rolling on immobilized P-selectin, demonstrating its dependence on PSGL-1. As a further control, cells from mice with genetic deletions in both PSGL-1 and C2GlcNAcT-I were tested to eliminate the possibility of alternate P-selectin ligands being up regulated in response to the deletion of PSGL-1. Cells from these doubly deficient mice also failed to roll on P-selectin. These results further confirm that, independent of the C2GlcNAcT-I enzyme, PSGL-1 is modified by core 2 - a modification that allows for P-selectin-mediated rolling under shear flow. 140 Figure 3-14: Activated C2GlcNAcT-Inu" CD8+ T cells roll on immobilized P-selectin-hIG. Splenocytes from C57BL/6, C2GlcNAcT-l n u 1 1, PSGL-r u 1 1 and C2GlcNAcT-r u l l /PSGL-l n u 1 1 mice activated at low density (0.25xl06 cells per millilitre; white bars) or high density (2.0xl06 cells per millilitre; gray bars) were perfused through an in vitro flow chamber over immobilized P-selectin-hIG at a shear force of 2 dyne per square centimetre. Data are expressed as the mean number of rolling cells/mm2+SEM. Differences between results for C2GlcNAcT-Fu" cells cultured at low density versus high density, as well as wild-type cells cultured at low density versus high density, were found to be significant (*,p < 0.05). 100 • 0.25x106 cells/ml 2.0x106 cells/ml C57BL/6 C2GlcNAcT-1 n u" PSGL-1 ™« PSGL-1 n u " / C2GlcNAcT-1 n u" 141 3.3 In vivo T cell receptor signalling induces Core 2 O-glycan-branched modifications independent of C2GlcNAcT-I 3.3.1 In vivo activation of C2GlcNAcT-Inu" CD8+ T cells induces P-selectin ligand In order to determine whether P-selectin ligand formation in C2GlcNAcT-In u" CD8 + T cells occurs under in vivo stimulation conditions, we compared CD8 + T cell responses from C2GlcNAcT-In u" and wild-type control mice transgenic for the apTCR specific for male Ag (HY) presented by the class I MHC molecule H-2Db (Teh et al., 1988). CFSE-labeled H Y t g thymocytes were transferred from female donors into male recipients, resulting in TCR activation of the transferred CD8 + T cells due to stimulation by endemic male antigen. The responses were analyzed three days after transfer. H Y t g wild-type and HY t g C2GlcNAcT-In u 1 1 CD8 + T cells were compared for P-selectin ligand formation, as well as for their respective proliferation rates, by FACS. Figure 3-15 A illustrates a strong increase in P-selectin ligand formation in wild-type H Y t g CD8 + T cells transferred into male recipients, while the same cells transferred into female control recipients are unresponsive. There was also a notable level of P-selectin binding on HY t g C2GlcNAcT-In u 1 1 CD8 + T cells transferred into male recipients, as compared with the control female recipients. Nevertheless, this increase was less pronounced than that observed for HY t g wild-type CD8 + T cells. No difference was observed in the CFSE dilution 48 hours after injection, indicating that the proliferative response of C2GlcNAcT-In u 1 1 and wild-type HY t g CD8 + T cells was equal (Figure 3-15B). 142 H Y * WT fe H Y * C2GlcNAcT-lnu" P-selectin P-selectin B CD E ID CD o CD > m CD CU HYtg WT CD E z "CD o CD > ro CD H Y * C2GlcNAcT-lnu" 100" 100 101 102 103 10< FL1-H C F S E Figure 3-15: In vivo activated C2GlcNAcT-ruU CD8+ T cells can form functional P-selectin ligand. CFSE-labeled mymocytes from TCR-tg HY (HY t g) female mice (wild-type (WT) or C2GlcNAcT-Inu") were adoptively transferred into male (shaded graph) or female (black line) C57BL/6 recipients. Seventy-two hours later, spleens were harvested and stained for CD8, T3.70 (TCR-tg), and P-selectin-hIG (A). As a control, EDTA was used to abrogate binding of P-selectin (dotted line). B, The proliferative response was monitored by analyzing the CFSE dilution of injected HY t gWT or HY t gC2GlcNAcT-P u 1 1 cells 48 hours after injection into male (shaded graph) and female (dashed line) recipients. Data are representative of at least five separate experiments each with at least three recipient mice each (three male and three female). 143 3.3.2 CD45 is modified upon in vivo activation of CD8 + T cells Analysis of in vivo activated C2GlcNAcT-In u 1 1 CD8 + T cells for core 2 modifications on CD45 reveals some intriguing differences. Data from Figure 3-15 suggest that core 2 activity exists in absence of C2GlcNAcT-I and can modify PSGL-1 O-glycans. Using the same adoptive-transfer model, we were interested in determining if this core 2 activity present in C2GlcNAcT-In u 1 1 mice also targets CD45. Analyzing HY t 8CD43 n u l lC2GlcNAcT-I n u" cells shows that P-selectin binding to the in vivo activated CD8 + T cells is positive while cells injected into control female recipients do not bind P-selectin (Figure 3-16A), confirming data illustrated in Figure 3-15A. Also, upon analyzing core 2 O-glycan branching of CD45 on these in vivo activated cells, we observe a decrease in 1B11 binding (indicative of increased C2GlcNAcT branching on CD45RB), Figure 3-16B portrays. Cells that were activated in the male recipients displayed a lower 1B11 signal - hence increasing C2GlcNAcT activity - than cells injected into female controls. The observed altered 1B11-CD45 reactivity provides direct evidence that antigen stimulation in vivo leads to C2GlcNAcT-I-independent induction of core 2 enzyme activity. In summary, our data demonstrate that both PSGL-1 and CD45 can serve as substrates for C2GlcNAcT-III induced under physiological conditions. 144 Figure 3-16: In vivo activation of CD8+ T cells modifies CD45 in absence of C2GlcNAcT-I. CFSE-labelled thymocytes from HY t g CD43 n u l l/C2GlcNAcT-r u" female mice were adoptively transferred into male (shaded graph) or female (black line) C57BL/6 recipients. Seventy-two hours later, spleens were harvested and stained for CD8, T3.70,1B11, and P-selectin-hIG, and analyzed by FACS. A, P-selectin-hIG staining of CD8 + T3.70+ cells is shown; B, 1B11-CD45 staining of CD8 + T3.70+ cells is shown. Data are representative of at least five separate experiments with three recipient mice per experiment (three male and three female). 145 3.4 Discussion These studies are the first to associate the role of C2GlcNAcT-III in forming functional P-selectin ligands (Merzaban et a l , 2005). C2GlcNAcT branching on PSGL-1 is essential for recognition by P-selectin, as expressed in Figure 1-7. Until recently (Merzaban et al., 2005), this branching was solely attributed to C2GlcNAcT-I, although evidence of other isoenzymes involved in the creation of selectin ligands exists (Ellies et al., 1998; - Kumar et al., 1996; Sperandio et al., 2001a; Sperandio et al., 2001b). Studies of C2GlcNAcT-In u 1 1 mice unveil that this enzyme's absence leads to varying deficiencies in functional selectin ligand formation for each of L-, E- and P-selectins. C2GlcNAcT-I deficiency causes a slight reduction of L-selectin ligands in peripheral-node HEV, yet normal lymphocyte homing to lymph nodes (Ellies et al., 1998) and Peyer's patches (Sperandio et al., 2001a). C2GlcNAcT-I does contribute to L-selectin ligands in HEV; 6-sulfo-sLex is expressed on core 2 O-glycans as well as on core 1 branches (Rosen, 2004), and both co-operate in controlling L-selectin-mediated lymphocyte homing to secondary lymphoid organs (Yeh et al., 2001). Therefore, loss of core 2 branching does not lead to a loss of L-selectin ligand recognition since core 1 extentions remain. Lack of C2GlcNAcT-I yields only a partial loss of E-selectin ligands on neutrophils (Broide et al., 2002; Ellies et al., 1998; Snapp et al., 2001; Sperandio et al., 2001b) and eosinophils (Broide et al., 2002; Symon et al., 1996; Wein et al., 1995). E-selectin binding to its ligands mainly depends on FucT-VII (Snapp, 1997 Blood) activity and much less on C2GlcNAcT activity. In contrast, studies that focus mainly on neutrophils and CD4 + T cells show that P-selectin-dependent cell adhesion almost exclusively depends on C2GlcNAcT-I (Ellies et al., 1998; Snapp et al., 2001; Sperandio et al., 2001b). Our studies now demonstrate that, in activated CD8 + T cells, P-selectin ligand expression can occur independently of 146 C2GlcNAcT-I. Using a neutralizing anti-PSGL-1 mAb or PSGL-1 n u l -mouse-derived cells has identified the C2GlcNAcT-I-independent P-selectin ligand as PSGL-1, removing the possibility that P-selectin recognizes an alternate ligand. Rolling assays in a parallel flow chamber confirm the functionality of the P-selectin ligand expressed on C2GlcNAcT-In u 1 1 T cells. Inducing P-selectin ligand formation in C2GlcNAcT-In u" CD8 + T cells was not achieved under standard T cell activation conditions, but requires very specific conditions that entail growth at a higher density than normal (Carlow et al., 2001a). Attempting to characterize the stimuli responsible for inducing core 2 activity via the alternate enzyme, our laboratory added C2GlcNAcT-III supernatant harvested from high-density cultures to cells grown at a low density that do not typically induce P-selectin ligand formation (Carlow et a l , 2001a). This approach demonstrates that a soluble factor is present in high-density supernatant capable of supporting P-selectin ligand formation. Continuing work in our laboratory shows that this soluble factor may be a very-small, stress-related molecule, but its identity is yet to be discovered. Recent evidence exists for the presence of an unknown soluble factor believed to be released by specific DCs that induce expression of E-selectin ligands, as well as the integrin 0:407, in activated CD4 + and CD8 + T cells (Dudda et al., 2005). Whether this is the samefactor that we are searching for is currently unknown. Although extreme high-density in vitro culture conditions are required to induce functional P-selectin ligand, more physiological in vivo conditions were also found to induce P-selectin ligand formation. Using the H Y t g mouse model, H Y t g CD8 + T cells from both C2GlcNAcT-In u 1 1 and wild-type mice proliferated equally when injected into male recipient mice (in response to HY antigen). There was also a significant induction of P-selectin ligand in HY t gC2GlcNAcT-I n u 1 1 CD8 + T cells, albeit at a lower level than observed for H Y , g wild-147 type CD8 + T cells. The level of induction is consistent but lower than that found in vitro. Manipulating the incubation time after injection as well as the number of cells injected, among other parameters, may result in more-enhanced selectin ligand formation. The histogram of selectin binding observed for in vivo activated wild-type cells shows two peaks, while that for activated C2GlcNAcT-In u 1 1 cells displays only one peak. It is tempting to hypothesize that the high-P-selectin-binding peak found with wild-type cells may be due to modification of P-selectin ligands by C2GlcNAcT-I, whereas the lower P-selectin binding peak may result from modification of P-selectin ligands by C2GlcNAcT-III. One way to test this could be to FACS sort the populations of cells within each peak and measure the RNA transcripts for both isoenzymes using RT-PCR. The murine genome encodes for three C2GlcNAcT isoenzymes that share considerable identity, as is described earlier for the human enzymes (Schwientek et al., 1999; Schwientek et al., 2000; Yeh et al., 1999). Given that P-selectin binding is believed to be wholly dependent on core 2 O-glycan formation (Lowe, 2002; Lowe, 2003), activity of one of the two other core 2 isoenzymes is the most-likely cause for the observed P-selectin ligand formation. Based on the RT-PCR data and the core 4 activity data, contribution of C2GlcNAcT-II can be ruled out as the enzyme responsible, whereas presence of RNA for the C2GlcNAcT-I and C2GlcNAcT-III isoenzymes in activated splenocytes point to C2GlcNAcT-III as the alternate core 2 isoenzyme able to support P-selectin ligand formation. These findings are consistent with the tissue distribution established for the human core 2 isoenzymes: RNA for hC2GlcNAcT-II is associated with digestive tract but not lymphoid tissues (Schwientek et al., 1999; Yeh et al., 1999); and RNA for hC2GlcNAcT-III occurs predominantly in thymus (Schwientek et al., 2000). 148 C2GlcNAcT-III RNA levels determined by real-time RT-PCR in C2GlcNAcT-In u" splenocytes are comparable with RNA levels detected in wild-type cells, indicating that the induction of the C2GlcNAcT-III isoenzyme in C2GlcNAcT-In u ! 1 cells is not a compensatory consequence of the loss of C2GlcNAcT-I. Thus, it is likely that, at least in activated CD8 + T cells, both enzymes may contribute to O-glycan branch formation. Interestingly, real-time RT-PCR analysis indicates that the RNA levels for both C2GlcNAcT-I and C2GlcNAcT-III do not alter significantly between the high-density and low-density cultures analyzed here. However, C2GlcNAcT activity assays of lysates from high-density and low-density cultures do display a noteworthy difference in activity, although, at the RNA level, similar amounts of transcript occur. This may suggest that other glycosyltransferases required for P-selectin ligand formation are the limiting factor or, alternatively, that core 2 enzyme activity may be subject to other control mechanisms under these cell culture conditions. In addition to regulating expression of C2GlcNAcTs at the RNA level, some data imply that regulation occurs at the translational and/or posttranslational level (Chibber et al., 2000; Chibber et al., 2003; Merzaban et al., 2005). Chibber et al. suggest that C2GlcNAcT-I is regulated by phosphorylation. In addition to the control mechanisms of glycosyltransferases outlined in section 1.6.2, there also may be posttranscriptional regulation by naturally occurring siRNA or miRNA (microRNA) (reviewed in (Tang, 2005)); these are small RNA molecules (approximately 22 nucleotides) that regulate genes by binding to their 3'UTR generally repressing gene expression. Therefore, what may be happening is that, under certain conditions, these miRNAs are repressing C2GlcNAcT RNA (low-density) expression whereas in other (high-density) situations, this repression is removed. This is a very intriguing area and is currently under investigation in our laboratory. Determination of C2GlcNAcT isoenzyme protein expression is difficult since there are presently no good antibodies available that are specific for the individual murine 149 C2GlcNAcT enzymes. Antibodies would be useful for determining protein expression by Western blotting of cell lysates and may give us a better understanding of the type of regulation occurring in the cell cultures tested here. Polarization toward activated TH1 (Tcl) and TH2 (TC2) subsets is known to impact the expression of enzymes required to generate functional P-selectin ligands (Blander et al., 1999; Carlow et al , 2001a; Lim et al., 1999; Lim et al., 2001; Wagers et al., 1998; White et al., 2001). Previous studies show that IL-2 and IL-4 differentially regulate functional PSGL-1 formation in activated CD8 + T cells from wild-type mice by regulating the expression of C2GlcNAcT-I (Carlow et al., 2001a). It is interesting to find that C2GlcNAcT-In u 1 1 splenocytes cultured at high density with IL-2 preferentially result in P-selectin ligand formation, whereas IL-4-cultured cells do not (Figure 3-13A). This differential regulation of PSGL-1 formation on activated T cells is reminiscent of the regulation of C2GlcNAcT-I described by Carlow et al. (Figure 3-13B;(Carlow et al., 2001a)) and suggests that the C2GlcNAcT-III enzyme may be receptive to similar cytokine regulation in addition to sharing substrate specificity. CD43 has long been recognized as a major substrate for C2GlcNAcT-I (Saitoh et al., 1991; Sportsman et al., 1985), and ectopic expression of the human core 2 enzymes in Chinese hamster ovary cells reveals that CD43 can serve as a substrate for all three enzymes (Schwientek et al., 2000; Yeh et al., 1999). Our results confirm and extend these findings to the murine cells analyzed here. Earlier studies from our laboratory demonstrate that expression of C2GlcNAcT-I in murine T cells results in increased O-glycan branching on several cell-surface molecules, including CD43 and CD45 (Barran et al , 1997; Carlow et al., 1999). The present data now show that the C2GlcNAcT-III induced in activated CD8 + T cells 150 can, in addition to PSGL-1, also use the leukocyte mucins CD43 and CD45 as acceptor glycoproteins in vitro and in vivo. Analysis of T cell subsets reveals that C2GlcNAcT-I-independent P-selectin ligand formation is observed primarily in Con A-stimulated CD8 + T cells maintained under the high-density culture conditions, whereas CD4 + T cells show only marginal binding of P-selectin, which is consistent with the literature (Snapp et al., 2001). Core 2 O-glycan expression on CD43 is also much lower in the CD4 + subset than in the CD8 + subset, directly demonstrating that core 2 activity is significantly lower in CD4 + cells. RT-PCR analysis confirms this observation, showing that CD4 + cells have extremely low levels of C2GlcNAcT-III compared with CD8 + cells. The current studies have not resolved some key questions, including: whether the weak residual C2GlcNAcT-III activity we observe in CD4 + T cells is functionally relevant; and whether conditions exist under which P-selectin ligand is induced more efficiently in C2GlcNAcT-In u" CD4 + T cells. Why there is co-induction of two core 2 glycosyltransferases with similar substrate specificities in CD8 + T cells is not obvious. There does appear to be some difference in fine specificity for substrates between these two enzymes, at least in vivo. Moreover, it is possible that inducing multiple C2GlcNAcT isoenzymes may be subject to differential regulation in specific subsets of T cells, and that CD8- and/or CD4-specific signalling pathways may result in enzyme induction, which remain to be discovered. We hypothesized that other C2GlcNAcT isoenzymes are responsible for the formation of functional selectin ligands in activated T cells and that due to the distribution of C2GlcNAcT-II and C2GlcNAcT-III in humans, that C2GlcNAcT-III is the likely candidate. These data, in support of our hypothesis, demonstrate that a second C2GlcNAcT enzyme, identified based on RNA expression as C2GlcNAcT-III, can contribute to P-selectin ligand 151 formation and affect the rolling of activated CD8 + T cells but not CD4 + T cells. However, C2GlcNAcT-I remains the key enzyme contributing to P-selectin ligand formation in activated T cells. Nevertheless, C2GlcNAcT-I and C2GlcNAcT-III may co-operate in controlling P-selectin ligand formation in CD8 + T cells. Analysis of mice with genetic inactivation of the C2GlcNAcT-III enzyme must be conducted to determine the true respective contributions of these O-glycan branching enzymes in controlling CD8 + T cell trafficking. 152 CHAPTER 4 E-SELECTIN BINDS ACTIVATED CD8+ T CELLS IN ABSENCE OF C2GlcNAcT-I2 4.1 Introduction E-selectin has a number of different ligands, including L-selectin (Patel et al., 2002; Picker et al., 1991b; Zollner et al , 1997), Mac-1 (Crutchfield et al , 2000), CD44 (Dimitroff et al., 2001a; Katayama et al , 2005; Sackstein, 2004), CD43 (Maemura and Fukuda, 1992) and some glycolipids (Burdick et a l , 2001). The two major glycoprotein ligands for E-selectin are ESL-1 (Levinovitz et al., 1993; Steegmaier et al., 1997; Steegmaier et al., 1995), which binds specifically to E-selectin but not to P-selectin, and PSGL-1 (Asa et al., 1995; Goetz et al., 1997; Hirata et al., 2002; Hirata et al., 2000; Norman et al., 2000; Patel et al., 2002; Somers et al., 2000; Xia et al., 2002a; Zou et al., 2005). These ligands have a common sLex decoration. In addition to the sLex, E-selectin also recognizes CLA on PSGL-1 and CD44 defined by the HECA-452 epitope (Berg et al., 1991; Borges et al , 1997b; Duijvestijn et al., 1988; Fuhlbrigge et al., 1997). E-selectin recognition of PSGL-1 is dependent on the posttranslational modifications made by glycosyltransferases that synthesize the terminal sialofucosylations, recognized by HECA-452, that serve as E-selectin-binding determinants on PSGL-1, including: ST3Gal-IV and FucT-IV, and FucT-VII (Figure 1-7 and Figure 1-8) (Blander et al., 1999; Homeister et al., 2001; Maly et al , 1996; Snapp et al., 1997; Weninger et al., 2000). These terminal sLex structures are expressed on poly-A^-acetyllactosamines found on core 2 O-glycans and asparagines-linked N-glycans on PSGL-1 (Ellies et al., 1998; Kumar et al., 1996; Moore et al , 1994; Nakayama et al., 2000; Snapp et al., 2001; Sperandio et al., 2001a). Although the 2 A version of this chapter has been published. Merzaban, J.S. and Zuccolo, J., Corbel, M., Williams, M.J., and Ziltener, H.J. (2005) An Alternate Core 2 (3l,6-7V-Acetylglucosaminyltransferase Selectively Contributes to P-Selectin Ligand Formation in Activated CD8 T Cells. Journal of Immunology. 174:4051-4059. 153 majority of sLex epitopes appear to reside on PSGL-1 O-glycans (Figure 1-7), there are three potential sites for N-glycosylation (Moore et al., 1994). The precedence of a core 2 branching requirement for E-selectin recognition is somewhat contentious and was one of the objectives of these studies. Once this was established we were also curious as to whether the C2GlcNAcT-III isoenzymes was able to form functional E-selectin ligands in the absence of the C2GlcNAcT-I enzyme. 4.2 E-selectin binds activated T cells in the absence of C2GlcNAcT-I E-selectin can bind to cells in a PSGL-1- and C2GlcNAcT-I-dependent fashion (Asa et al., 1995; Ellies et al., 1998; Li et al., 1996; Snapp et al., 2001; Sperandio et al., 2001b), as well as in a PSGL-1-independent manner, via alternate E-selectin ligands (Levinovitz et al., 1993; Steegmaier et al., 1995; Xia et al., 2002b; Yang et al., 1999). Consistent with this, high levels of E-selectin binding are observed on activated wild-type cells using flow cytometry (Figure 4-1), while binding on activated C2GlcNAcT-In u 1 1 cells is consistently lower. In contrast with P-selectin ligand expression, E-selectin ligand expression on C2GlcNAcT-In u" cells is not influenced by cell culture density (Figure 4-1). The level of E-selectin binding on activated cells from PSGL-l n u 1 1 and C2GlcNAcT-Fu" mice is equivalent, as evidenced in comparisons of the C2GlcNAcT-In u 1 1 graph with the PSGL-l n u 1 1 graph in Figure 4-1. This indicates that E-selectin binding observed in absence of C2GlcNAcT-I is PSGL-1-independent. 154 C57BL/6 C2G lcNAcT - l n u " PSGL-1 n u " > E-selectin-hIG Figure 4-1: E-selectin binds activated CD8+ T cells in the absence of C2GlcNAcT-I and PSGL-1. FACS profiles of E-selectin-hIG binding to day 4-activated splenocytes cultured at either 0.25x106 cells per millilitre (black line) or 2.0x106 cells per millilitre (red line) is shown for C57BL/6, C2GlcNAcT-In u 1 1 and PSGL- l n u " cells. 5 mM EDTA (dashed line) was added during incubation, with E-selectin-hIG as a negative control in all experiments. 4.3 Activated C2GlcNAcT-InuU cells roll on immobilized E-selectin in a PSGL-1-independent C2GlcNAcT-III-independent manner Activated C2GlcNAcT-Pu 1 1 splenocytes were also tested for rolling on E-selectin using an attachment and rolling assay under defined shear flow in vitro. Activated C2GlcNAcT-In u" T cells from high-density and low-density cultures display identical rolling on immobilized E-selectin-hIG at 2.0 dynes per square centimetre (Figure 4-2). This rolling interaction was independent of both PSGL-1 and C2GlcNAcT-I because cells from mice double deficient in PSGL-1 and C2GlcNAcT-I adhered and rolled equally on immobilized E-selectin. These rolling data are consistent with E-selectin binding by FACS (Figure 4-1). In addition, alternate C2GlcNAcT enzyme, C2GlcNAcT-III, does not compensate for E-selectin ligand formation (Figure 3-11). 155 E E "55 <D 5 . o o .c Q. E >1 _l _c O DC 125 100 75 50 25 • 0.25x106 cells/ml T 2.0x106 cells/ml rHfl C57BL/6 C2GlcNAcT-1""« PSGL-1™11 PSGL-1""11/ C2GlcNAcT-1""" Figure 4-2: Activated C2GlcNAcT-ru" CD8+ T cells roU on immobilized E-selectin independent of PSGL-1. Splenocytes from C57BL/6, C2GlcNAcT-l n u 1 1, PSGL-l n u " and C2GlcNAcT-r u l l /PSGL-l n u 1 1 mice activated at low density (white bars) or high density (gray bars) were perfused through an in vitro flow chamber over immobilized E-selectin-hIG at a shear force of 2 dyne per square centimetre. Data are expressed as the mean number of rolling cells per square millimeter ± SEM. 4.4 Activated CD4+ and CD8+ T cells bind E-selectin-hIG differently Since P-selectin ligand formation is restricted to CD8 + T cells in absence of C2GlcNAcT-I, it is relevant to determine if E-selectin ligand formation differs between CD4 + and CD8 + subsets. C2GlcNAcT-I n u l ! splenocytes were activated and maintained at high density (2.0 x 106 cells per millilitre) and, on day 4, stained for CD4 and CD8, in addition to E-selectin-hIG. Interestingly, E-selectin binding to CD4 + T cells is significantly lower than binding to CD8 + T cells in both wild-type and C2GlcNAcT-In u 1 1 cultures (Figure 4-3). Also, on CD4 + T cells, this E-selectin binding is lost in absence of C2GlcNAcT-I, while in CD8 + T cells significant E-selectin binding remains (Figure 4-3). 156 Figure 4-3: Activated CD4+ T cells do not bind E-selectin in the absence of C2GlcNAcT-I whereas CD8+ T cells do. FACS profiles of E-selectin-hIG binding to day 4-activated splenocytes cultured at 2.0x106 cells per millilitre (red line) is shown for C57BL/6 and C2GlcNAcT-In u 1 1 CD4 + and CD8 + T cell subsets. 5 mM EDTA (blue line) was added during incubation with E-selectin-hIG as a negative control in all experiments to inhibit binding. These results show that E-selectin ligands on CD4 + T cells depend on C2GlcNAcT-I activity while, on the other hand, the absence of C2GlcNAcT-I does not result in a complete loss of the ability of CD8 + T cells to bind E-selectin-hIG. 157 4.5 Discussion Data presented in this thesis deduce that both under in vitro and in vivo conditions, C2GlcNAcT-I and -III can catalyze the formation of functional PSGL-1 recognized by P-selectin. To address the question of whether the two enzymes display similar redundancy in E-selectin ligand formation, we also measured binding of E-selectin to activated lymphocytes. In contrast to P-selectin binding and P-selectin-dependent rolling, neither E-selectin binding nor E-selectin-dependent rolling was altered between high-density versus low-density stimulated C2GlcNAcT-In u" splenocytes, indicating that the C2GlcNAcT-III enzyme does not support E-selectin ligand formation under these culture conditions. As this thesis underlines a number of times, P-selectin ligand formation requires core 2 branching on O-glycans, however, this is not the case for E-selectin ligands. A number of studies support that cells deficient in C2GlcNAcT-I are able to bind to E-selectin, although at slightly reduced levels compared with wild-type cells (Ellies et al., 1998; Huang et al., 2000; Merzaban et al., 2005). These studies put forward that 75% to 80% of E-selectin binding is maintained in absence of C2GlcNAcT-I. However, there is a component of binding that likely depends on C2GlcNAcT-I branching on PSGL-1, and is not due to branching on alternate E-selectin ligands, such as ESL-1 (Huang et al., 2000; Merzaban et al., 2005). Our findings also support data indicating that E-selectin ligands other than PSGL-1 can mediate E-selectin-dependent rolling (Xia et a l , 2002a; Yang et al., 1999), and that these ligands do not depend on C2GlcNAcTs. The two major glycoprotein ligands for E-selectin are ESL-1 (Levinovitz et al., 1993; Steegmaier et al., 1997; Steegmaier et al., 1995), which binds specifically to E-selectin but not to P-selectin, and PSGL-1 (Asa et al., 1995; Goetz et al., 1997; Hirata et al., 2002; Hirata et al., 2000; Norman et al., 2000; Patel et al., 2002; Somers et al., 2000; Xia et al., 2002a; Zou et al., 2005). Other ligands proposed include L-selectin 158 (Patel et a l , 2002; Picker et al., 1991b; Zollner et a l , 1997), Mac-1 (Crutchfield et al., 2000), CD44 (Dimitroff et al., 2001a; Katayama et al., 2005; Sackstein, 2004), CD43 (Maemura and Fukuda, 1992) and some glycolipids (Burdick et al., 2001). The specifications for E-selectin binding to ESL-1 differ slightly from those for PSGL-1. ESL-1 binding to E-selectin is sialic acid dependent (Levinovitz et al., 1993) and modifications of one or more of the five putative N-glycans, not O-glycans, are critical for ligand binding to E-selectin (Huang et al., 2000; Steegmaier et al., 1997; Steegmaier et al., 1995). Binding also depends on a(l,3)-fucosylation (Li et al., 1996; Steegmaier et al., 1995; Zollner and Vestweber, 1996); in fact, analysis of FX n u 1 1 mice suggests that E-selectin ligands are much more sensitive to the loss of fucose than P-selectin ligands (Smith et al , 2002). T cell subsets use C2GlcNAcT-I and C2GlcNAcT-III differentially for P-selectin ligand formation, as CHAPTER 3 explains. Distinct T cell subsets differ in their functions and thus may also differ in regulation of their homing behaviour. This differential regulation may be the basis for our observation that CD8 + T cells but not CD4 + T cells can use the C2GlcNAcT-III enzyme for P-selectin ligand formation. This may also be the case for E-selectin ligands. Major differences between CD4 + and CD8 + subsets are uncovered in terms of E-selectin binding in the absence and presence of C2GlcNAcT-I. CD8 + T cells can bind E-selectin in a PSGL-1 - and C2GlcNAcT-I-independent manner due to the presence of alternate E-selectin ligands. However, in CD4 + T cells, E-selectin appears almost entirely dependent on C2GlcNAcT-I under the culture conditions tested here. Based on these data, it is enticing to construct a model where CD8 + T cells expressing PSGL-1 (and ESL-1 or other E-selectin ligands not yet determined) are modified by C2GlcNAcT-I, whereas CD4 + T cells only express PSGL-1 and are modified by C2GlcNAcT-I, but not C2GlcNAcT-III. This infers that CD8 + T cells may bind E-selectin, in 159 addition to P-selectin, better than CD4 + T cells, since they not only express alternate E-selectin ligands but also can use alternate C2GlcNAcT isoenzymes to create core 2-branched ligands for E-selectin (and P-selectin). In summary, these data outline that, in absence of C2GlcNAcT-I, activated CD8 + T cells bind to E-selectin in a PSGL-1-independent manner that likely does not depend on alternate core 2 enzymes. Unlike CD8 + T cells, activated CD4 + cells depend on C2GlcNAcT-I activity in order to bind E-selectin. This is a very interesting point since defining whether or not C2GlcNAcT-I is important for E-selectin ligand formation appears to be cell specific. 160 CHAPTER 5 ROLE OF SELECTINS IN PROGENITOR HOMING TO THE THYMUS3 5.1 Introduction In order to maintain the T cell pool, a constant influx of T cell progenitors is required to enter the thymus from the blood (Donskoy and Goldschneider, 1992; Foss et al., 2001; Scollay et al., 1986). At any given time, the thymus can accommodate a limited number of progenitors, which reside in specific niches; the migration of progenitors, hence, occurs in a highly regulated manner (Donskoy and Goldschneider, 1992; Foss et a l , 2001; Porritt et al., 2003; Prockop and Petrie, 2004). Much emphasis rests on identifying T cell progenitors in the bone marrow (Adolfsson et al., 2001; Adolfsson et al., 2005; Christensen and Weissman, 2001; Perry et al., 2004), blood (Donskoy and Goldschneider, 1992; Schwarz and Bhandoola, 2004) and thymus (Allman et al., 2003). As well, T cell development (Anderson and Jenkinson, 2001; Felli et al., 1999; Harman et al., 2003; Zuniga-Pflucker, 2004) and movement (Lind et al., 2001; Porritt et al., 2003; Prockop et al., 2002) through the thymus have been the subjects of careful analyses. The question remains of how these progenitors enter the thymus. What adhesion molecules are involved? Do these molecules dictate entry of circulating thymic progenitors (CTPs) into the thymus from the blood in an analogous manner to naive T cell homing to the lymph nodes? This chapter strives to address these questions. 5.2 P-selectin ligand is expressed on lymphoid progenitors L-selectin expression on naive T cells and L-selectin ligand expression on HEV are critical to homing of naive T cells to lymph nodes and, thus, in maintaining effective immune responses to pathogens. Our lab is interested in the control of selectins and their ligands over 3 Rossi, F.M.V., Corbel, S.Y.*, Merzaban, J.S.*, Carlow, D.A., Gossens, K., Duenas, J., So, L., Yi, L., Ziltener, H.J. (2005) Recruitment of adult thymic progenitors is regulated by P-selectin and its ligand PSGL-1. Nature Immunology. 6: 262-270. (*these authors contributed equally to this work) 161 cell traffic and homing. Studies designed to examine the role of selectins and their ligands in CTP homing to the thymus initially intended to identify selectin (L-selectin, E-selectin and P-selectin) expression on thymus and CTPs. Through these studies, we discovered that P-selectin expression on thymic endothelium and P-selectin ligand (PSGL-1) expression on progenitor cells have a place in maintaining the thymus's ability to generate T cells. In order to address this, we first determined if these receptors and ligands were expressed on the thymus and relevant progenitor populations. We determined P-selectin ligand expression on bone marrow HSC, circulating KLS (CTP), CLP and ETPs by binding to P-selectin-hIG chimera (Rossi et al., 2005). The parameters used to resolve each population in bone marrow and thymus are outlined in Table 5-1. Bone marrow and circulating KLS cells were defined and included both Flt3" long-term repopulating cells and Flt3+ multipotent progenitors, among which lymphoid progenitors are described (Adolfsson et al., 2001; Christensen and Weissman, 2001; Perry et al., 2004; Schwarz and Bhandoola, 2004; Sitnicka et al., 2002). P-selectin-hIG fusion protein stained the majority of mouse bone marrow and thymic cells, consistent with a possible role in thymic homing. As outlined in the introduction of this thesis, P-selectin recognizes very-specific posttranslational modifications in its ligand, PSGL-1, including core 2 O-glycan branching of crucial threonine residues within the N-terminal region of PSGL-1 (Ellies et al., 1998; Lowe, 2002). This dependency on C2GlcNAcT activity is evident in the inability of C2GlcNAcT-In u" cells from bone marrow and thymus to bind P-selectin-hIG. 162 Table 5-1: Thymic progenitor populations and their phenotypes Tissue Phenotype E T P Thymus Lin , CD3-, CD4", CD8", CD25", CD44+, c-Kit+ C T P Blood stream Lin", c-Kit+, Sca-1+ C L P Bone marrow Lin", c-Kit10, Sca-l'°, IL-7R+ K L S Bone marrow Lin", c-Kit+, Sca-1+ P-selectin is expressed by thymic endothelium (Rossi et al., 2005). Through immunofluorescent staining of frozen thymus sections from wild-type mice using Catalyzed Reporter Deposition (tyramide mediated) to increase detection-sensitivity revealed expression of P-selectin throughout the thymic vasculature, while no staining is observed in control thymi from P-selectinnu11 animals (conducted in Rossi's lab) (Rossi et al., 2005). The constitutive expression of P-selectin on thymic endothelium and C2GlcNAcT-dependent P-selectin ligand on lymphoid progenitors strongly implicate a role for this receptor/co-receptor complex in maintaining the T cell pool. To test the functional importance of this for CTP homing to the thymus, we used a number of mouse models. Since we have established expression of PSGL-1 on all relevant lymphoid progenitor populations, and established that P-selectin is constitutively expressed in the thymus, the functional significance of this expression pattern was our next subject of exploration. 5 .3 Parabiosis Bunster and Meyer first used parabiosis in 1933 (Bunster and Meyer, 1933), and since then its use to study the trafficking of cells, including hematopoietic stem cells (Abkowitz et al., 2003; Wright et al., 2001) and lymphocytes (Donskoy and Goldschneider, 1992; Wright et al , 2001), has become central. This mouse model enables the study of cells in a competitive homeostatic environment, without irradiation or other ablative procedures. The surgically connected parabiotic mice quickly develop an extensive and continuous exchange of blood, allowing the free movement of cells between animals. T cells and other leukocytes are freely exchanged between animals contributing to all peripheral tissues, including lymph nodes and spleen, with chimerism reaching close to 50% as early as seven to 10 days after surgical joining (measured in the blood) (Wright et al., 2001). Experiments in non-irradiated, parabiotic mice typically result in low levels of thymic chimerism, where approximately 5% to 10% of one partner's cells contribute the thymus of the other partner (Donskoy and Goldschneider, 1992; Wright et al., 2001), indicating that thymocytes generally do not move freely between partners. Cells that develop within the thymus of parabiotant-I from parabiotant-II are due to CTP migration from parabiotant-II; also, cells that develop within the thymus of parabiotant-II from parabiotant-I result from the CTP migration from parabiotant-I. In order to track the migration of T cells, CTPs and other precursors, parabionts between C57BL/6 mice expressing either of two allelic variants of the T cell marker Thyl (Thyl .1 or Thyl.2) were created (Rossi et al , 2005). Mice with a null mutation in genes of interest (C2GlcNAcT-I, PSGL-1 and P-selectin) were backcrossed on a C57BL/6 Thyl.2 background. Figure 5-1 illustrates this. 164 Figure 5-1: Diagram of parabiotic mouse model. Parabiotic pairs were surgically generated between congenic Thyl. 1 (orange) and Thyl .2 (green) mice and allowed to freely share circulation for five to six weeks. Peripheral blood and tissues show approximately 50% chimerism within a week following surgery. Both strains of mice were backcrossed on a C57BL-6 background. 5.3.1 Wild-type Thyl.2 parabiosed to wild-type Thyl.l The amount of T cell chimerism in the thymi of each of the parabionts was assessed following sacrifice and perfusion by staining for Thyl.l and Thyl.2, along with CD4 and CD8 cell-surface markers. The amount of chimerism observed in the parabionts between C57BL/6 Thyl.l* and Thyl.2+ wild-type partners five weeks after surgery is 7.6% versus 11.7%, respectively, with no significant difference (p > 0.05) between the two partners in the thymus, spleen and lymph node populations (Figure 5-2A). These data are comparable with that reported by others (Donskoy and Goldschneider, 1992) and confirm that differences in Thyl isoforms do not have an effect on progenitor or mature T cell homing properties, and are a reliable control for further experiments. 165 5.3.2 C2GlcNAcT-Inu" Thyl.2 parabiosed with wild-type Thyl.l Posttranslational modifications on PSGL-1 are crucial for binding to P-selectin, as outlined in section 1.5.2 of this thesis. Among other modifications, C2GlcNAcT branching of the N-terminal Thr 17 (murine form) of PSGL-1 is key for binding to P-selectin (Broide et al., 2002; Ellies et al., 1998; Merzaban et al., 2005; Snapp et al., 2001; Sperandio et al , 2001b; Symon et al., 1996; Wein et al., 1995). To determine if core 2 activity has a role in controlling CTP migration from the bone marrow to the thymus, we analyzed C2GlcNAcT-In u I 1 Thyl .2+ mice that were parabiosed for five weeks with wild-type C57BL/6 Thyl . l + mice. Mice were then sacrificed and the contribution of wild-type and C2GlcNAcT-In u 1 1 T cells to lymphoid organs within each animal was determined. Figure 5-2B shows that in the lymph nodes and spleens of these mice, the ratio of CD4 + and CD8 + T cells derived from the two partners is approximately 1:1, indicating that the mature peripheral T cells circulate freely between the two parabiosed animals independently of C2GlcNacT-I activity. However, analysis of DP thymocytes within the thymi of each partner reveals highly significant differences. First, upon analyzing the thymi of C2GlcNAcT-In u 1 1 Thyl .2+ parabionts (from the C2GlcNAcT-InullThyl.2:wild-typeThyl.l pair) to wild-type Thyl.2+parabionts (from the wild-typeThyl.2:wild-typtThyl.l pair), we find a substantial increase (3.7-fold increase) in the contribution of wild-type Thyl . l + DP thymocytes to the C2GlcNAcT-I n u U Thyl.2+ thymus, compared with the wild-type Thyl .2+ thymus (42.9% versus 11.7% chimerism, respectively; p < 0.001) as illustrated by comparing Figure 5-2A with B. Second, upon analyzing the thymi of wild-type Thyl . l + parabionts from either the wild-type Thyl.2+:wild-type Thyl.T or the C2GlcNAcT-In u" Thyl.2+:wild-type Thyl. 1+pairs, we discover a significant decrease (60% decrease) in C2GlcNAcT-i n u" Thyl.2+'s 166 contribution compared with wild-type Thyl.2 +DP thymocytes to the wild-type Thyl.l thymi (3.1% C2GlcNAcT-In u 1 1 Thyl.2+ versus 7.6% wild-type Thyl.2+ chimerism), as apparent in Figure 5-2B. Figure 5-2: C2GlcNAcT-I, PSGL-1 and P-selectin have a functional role in progenitor homing to the thymus. Parabiotic mice were generated by joining wild-type C57BL/6 Thyl . l + mice with congenic Thyl.2+ wild-type C57BL/6 (A; n=5), C2GlcNAcT-In u 1 1 (B; n=5), PSGL-l n u 1 1 (C; n=5), or P-selectinnu11 (D; n=3) partners. The mice shared circulation for five to six weeks prior to sacrifice and analysis of the thymus, spleen and lymph node by FACS. The frequency of DP thymocytes derived from migrating progenitors in the thymi of parabionts reveals a deficiency in the ability of C2GlcNAcT-In u 1 1 (B) and PSGL-l n u 1 1 (C) progenitors to seed the thymus compared with wild-type (A). There is also a deficiency in the ability of P-selectinnu11 (D) thymi to receive migrating wild-type progenitors. The frequency of CD4 + and CD8 + single positive mature T cells in the spleen (A, B, C, D) and lymph node (A, B, and D; with the exception of C) of parabiotic pairs shows that the thymi from both animals contribute equally to the peripheral T cell pool, and that T cells are able to migrate freely between partners. Dot plots of CD4 + and CD8 + cells in the thymus and lymph nodes with the proportions of Thyl .1+ and Thyl .2+ cells in the DP compartment in the thymus and the CD4 + compartment in the lymph node are shown as one example in each parabiotic group tested. Throughout the figure, green bars represent Thyl.2+ cells present in the Thyl . l + partner (wild-type thymus); orange bars represent Thyl . l + cells present in the Thyl.2+ partner (either wild-type, C2GlcNAcT-fu", PSGL- l n u " or P-selectinnu" thymi). Error bars represent the standard error of the mean. Numbers at end of bars represent the mean frequency of chimerism along with the standard error of the mean. 167 In summary, wild-type T h y l . l + CTPs home significantly better to C2GlcNAcT-I n u 1 1 Thyl .2 + thymi and C2GlcNAcT-I n u 1 1 Thyl .2 + CTPs migrate much less efficiently to corresponding wild-type T h y l . l + thymi. This analysis reveals a key role for C2GlcNAcT-I in thymic progenitor homing since a lack of this enzyme impairs the ability of progenitors to compete for thymic niches. The next question that arises is which substrates of this enzyme activity play a role in the physiological homing of thymic progenitors. 5.3.3 P S G L - l n u " Thyl.2 parabiosed with wild-type T h y l . l Since the P-selectin ligand is expressed on all HSC, including CTPs, and P-selectin ligand expression is lost in C2GlcNAcT-I n u 1 1 HSC, PSGL-1 is the most likely core 2 substrate. We parabiosed P S G L - l n u U Thyl .2 + mice with wild-type T h y l . l + mice in order to establish whether CTPs use PSGL-1 in thymic homing. The results from analyzing these mice are analogous with those obtained from the C2GlcNAcT-I n u 1 1 Thy 1.2+:wild-type T h y l . l + pairs, although at an even greater disparity. Again, on analyzing the thymi of PSGL-1 n u l 1 Thyl .2 + parabionts to wild-type Thyl.2 + parabionts, we uncover a substantial increase (4.3-fold increase) in the contribution of wild-type T h y l . l + DP thymocytes to the PSGL-1 n u l 1 Thyl .2 + thymus, compared with the wild-type Thyl .2 + thymus (50.3% versus 11.7% chimerism, respectively; p < 0.001), as illustrated by comparing Figure 5-2C with A. Also upon analyzing the thymi of wild-type T h y l . l + parabionts from either the wild-type Thyl.2+:wild-type T h y l . l + or the P S G L - l n u 1 1 Thyl.2+:wild-type T h y l . l + pairs, we determine a significant decrease (70% decrease) in the contribution of PSGL-1 n u l 1 Thyl .2 + compared with wild-type Thyl .2 + DP thymocytes to the wild-type Thyl . 1+ thymi (2.3% P S G L - l n u 1 1 Thyl .2 + versus 7.6% wild-type Thyl .2 + chimerism; p < 0.001), as 169 apparent in Figure 5-2C and A. T cells from PSGL-1nu and wild-type mice again contribute equally to the spleens of parabiotic partners, whereas in the lymph nodes of wild-type animals, a small-yet-significant reduction (p<0.001) of PSGL-1 """-derived T cells is observed in Figure 5-2C. These results strongly implicate PSGL-1 as the ligand targeted by C2GlcNAcT-I and involved in CTP progenitor homing to the thymus. Since PSGL-1 is a ligand for all three selectins, the next step is to determine which selectin (or selectins) is responsible. 5.3.4 P-selectinnuU Thyl.2 parabiosed with wild-type Thyl.l To determine if P-selectin is the counter-receptor for PSGL-1 involved in thymic homing, P-selectinnu11 Thyl.2+ mice were parabiosed with wild-type Thyl. l + mice. In addition to endothelial cells, P-selectin is expressed on platelets and megakaryocytes upon activation (McEver and Martin, 1984; Springer, 1994; Vestweber and Blanks, 1999). As a result of this restricted expression pattern, thymic progenitors from both partners should be phenotypically identical and should home at normal rates to the wild-type thymus. In addition, if P-selectin expression in the thymus is required in thymic progenitor homing, migration to the P-selectinnu" thymus would be impaired. Upon analyzing the thymi of P-selectinnu11 Thyl.2+ parabionts to wild-type Thyl.2+ parabionts, we identify a substantial decrease (97% decrease) in the contribution of wild-type Thyl. l + DP thymocytes to the P-selectinnu11 Thyl.2+ thymus compared with the wild-type Thyl.2+ thymus (0.3% versus 11.7% chimerism, respectively; p < 0.001), as illustrated by comparing Figure 5-2D with A. When analyzing the thymi of wild-type Thyl .1+ parabionts from either the wild-type Thyl.2+:wild-type Thyl. l + or the P-selectinnu11 Thyl.2+:wild-type Thyl. l + pairs, the migration of P-selectinnu11 Thyl.2+ cells into wild-type thymi was not significantly reduced (p > 0.05). 170 These data suggest that P-selectin facilitates homing of thymic progenitors to the thymus. 5.4 Competitive Repopulation Studies The data described in the parabiotic mouse models strongly implicate the interaction between C2GlcNAcT-I-decorated PSGL-1 on thymic progenitors with P-selectin present on the thymic endothelium as responsible for the observed competitive advantage of wild-type over PSGL-1 n u" T cell progenitors. To directly confirm this, we carried out competitive repopulation assays in both irradiated recipient mice and in non- irradiated IL-7Rn u 1 1 recipients. 5.4.1 Competitions into irradiated wild-type and P-selectinnu" recipients To determine if the competitive advantage of wild-type progenitors over PSGL-l n u 1 1 and C2GlcNAcT-In u 1 1 progenitors is dependent on the presence of P-selectin, a series of competitive thymic reconstitution experiments were organized. Lethally irradiated wild-type Thyl.2+ or P-selectinnul1 Thyl.2+ recipients were reconstituted with a 1:1 mixture of bone marrow from wild-type Thyl.T mice to C2GlcNAcT-In u" Thyl.2+, PSGL-l n u 1 1 Thyl.2+ or wild-type Thyl.2+ bone marrow. Analysis of thymic chimerism 10 weeks after transplant reveals that in wild-type reconstitution by wild-type cells is fivefold to tenfold more efficient than reconstitution by C2GlcNAcT-In u 1 1 or PSGL-1 n u l 1 (Figure 5-3A). Thus, in agreement with data from the parabiosis experiments (Figure 5-2), lack of C2GlcNAcT-I or PSGL-1 impairs thymic homing of CTPs. On the other hand, when the competitive reconstitution was carried out using lethally irradiated P-selecthT11 recipient mice, C2GlcNAcT-In u 1 1 or PSGL-1 n u l 1 thymic progenitors contributed to thymic reconstitution as efficiently as wild-type progenitors (Figure 5-3B). Thus, the competitive advantage of wild-type cells over C2GlcNAcT-F u U or PSGL-l n u 1 1 cells 171 depends on thymic P-selectin, demonstrating a functional P-selectin-PSGL-1 interaction in the thymic homing process. B 1UU £ i 50 O o 100 C s-56.5 Wild type recipients 15.8 WT C2GlcNAcT-l"u" PSGL-1nu" Competitor Bone Marrow 63.7 P-selectinnu" recipients 53.8 44.3 o o E WT C2GlcNAcT-lnu" PSGL-1™1 Competitor Bone Marrow Figure 5-3: Wild-type cells have a competitive advantage over C2GlcNAcT-Inul1 and PSGL-l n u U cells in repopulating the thymus of wild-type recipient mice, and this competitive advantage is P-selectin dependent. Competitive repopulation studies were set up by injecting a 1:1 ratio of bone marrow from wild-type C57BL/6 Thyl . l + mice with bone marrow from congenic Thyl.2+ C2GlcNAcT-In u 1 1 or PSGL-l n u 1 1 mice, as shown in the schematic (A). These cells were injected into lethally irradiated B , wild-type (CD45.1/CD45.2; n=3) or C, P-selectinnu" (n=3)-recipient mice, and thymi were analyzed 10 weeks later by FACS to determine what frequency (thymic chimerism) of the Thyl.2+ cells comprising the DP compartment. Numbers on top of bars represent the percentage of donor-derived cells within the DP subset and error bars represent the SEM. 172 5.4.2 Competitions into non-irradiated IL-7Rnu recipients Studies show that irradiation alters adhesion molecule expression, including selectins, on vascular endothelium (Mazo et al , 2002). The treatment could, as a result of this expression, alter hematopoietic progenitor cell recruitment processes. To exclude the possibility that irradiation-induced selectin expression interferes with thymic CTP homing in our experiments, we set up additional competitive repopulation experiments using non-irradiated IL-7Rn u 1 1 mice as recipients. IL-7 is a critical cytokine in T cell development required for progression through the DN stages of thymocyte development (DN2 -> DN3 transition) (El Kassar et al., 2004; Peschon et al., 1994; von Freeden-Jeffry et al., 1995). IL-7Rnu" mice lack mature T cells due to a developmental block in thymocyte development, but exhibit otherwise normal thymic microenvironments and overall thymic structure (Prockop and Petrie, 2004). IL-7Rn u 1 1 thymi are highly receptive to wild-type transplanted thymic progenitors in absence of any radiation, displaying normal proportions of all developmental stages (for instance, DN1-3, DP and SP) (Prockop and Petrie, 2004). Thus, the IL-7Rnu11 mouse is an attractive model to study thymic progenitor cell homing, and competitive repopulation experiments were set up using donor mouse bone marrow that is congenic for two different alleles of CD45 (CD45.1 and CD45.2) in order to trace the origin of transplanted cells. C2GlcNAcT-In u" CD45.2+ cells were also competed with C2GlcNAcT-In u" cells from mice heterozygous for CD45.1 and CD45.2 (CD45.1/CD45.2). These were controls in order to verify that the CD45 alleles do not influence the homing or expansion of early thymic progenitors. No significant difference (p>0.05) was observed in these cells' ability to repopulate IL-7Rn u 1 1 thymi (Figure 5-4). 173 Figure 5-4: Wild-type cells have a competitive advantage over C 2 G l c N A c T - I n u " and PSGL-l n u U cells in repopulating the thymus of non-irradiated IL-7RnuU mice. Competitive repopulation studies were set up by injecting a 1:1 ratio of congenic CD45.2 (blue) versus CD45.1/CD45.2 (pink) bone marrow from either wild-type C57BL/6, C2GlcNAcT-In u" or PSGL-1 n u" into IL-7RnulI-recipient mice. Mice were analyzed three weeks after injection to determine the proportion of the competing partner cells in the DP thymocyte subset of the reconstituted IL-7Rn u 1 1 thymus. Thymic chimerism is based on the frequency of DP thymocytes subset derived from the competing partner cells. Numbers on top of bars represent the percentage of donor-derived cells within the DP subset. Error bars represent the standard error of the mean. 174 When C2GlcNAcT-In u" CD45.1/CD45.2 bone marrow cells competed with wild-type CD45.2 bone marrow cells into IL-7Rn u 1 1 recipients, over 97% of the DP thymocytes are generated from the wild-type CD45.2 bone marrow (Figure 5-4). Similarly, when PSGL-1 n u l 1 CD45.2 are competed with wild-type CD45.1/CD45.2 bone marrow cells, over 95% of the DP thymocytes are generated from the wild-type CD45.1/CD45.2 donor (Figure 5-4), in agreement with results obtained from the parabiotic and competitive repopulation studies. . When C2GlcNAcT-In u 1 1 competes with PSGL-l n u 1 1 bone marrow, cells lacking PSGL-1 are generally more severely impaired compared with cells lacking C2GlcNAcT-I alone (Figure 5-4), suggesting that PSGL-1 is not just the target of C2GlcNAcT-I but may also be modified by other C2GlcNAcT isoenzymes (Merzaban et al., 2005), and thus partially rescues the ability of PSGL-1 to bind P-selectin in the absence of C2GlcNAcT-I. 175 5.5 Discussion Intrathymic progenitors have no long-term self-renewing potential; the thymus, therefore, depends on recruiting multipotent progenitors that circulate in the blood (Goldschneider et al., 1986; Scollay et al., 1986). The first step in postnatal T cell production involves two distinct phases of cell migration - out of the bone marrow into the blood and out of the blood into the thymus. Little is known about how this process is regulated in the postnatal thymus. Attempts to identify the signals that support the recruitment of progenitors to the fetal thymic primordium indicate the possible involvement of various chemokines (Bleul and Boehm, 2000; Champion et al., 1986; Wilkinson et al., 1999; Wurbel et al., 2001) and adhesion molecules (Kawakami et al., 1999; Suniara et al., 1999; Wu et al., 1993). The physical characteristics of shear forces in blood vessels associated with recruitment to the postnatal thymus point to the likely involvement of selectin or selectin-like adhesive molecules, especially through analogy to leukocyte extravasation in other tissues. Indeed, recent data from our lab demonstrate that the interaction between PSGL-1 expressed on circulating thymic progenitors (CTP) and P-selectin expressed on thymic endothelium facilitiate the homing of T-cell progenitors to the adult thymus. This is a novel contribution to the field of thymic homing and development, and is illustrated in Figure 1-12 (and Figure A-4). The findings were published recently (Rossi et al., 2005). Establishing the players involved in thymic progenitor homing Characterizing the ETP in the thymus (Allman et al., 2003), the CTP in the blood (Donskoy and Goldschneider, 1992; Schwarz and Bhandoola, 2004) and the ETP-yielding progenitor in the bone marrow (Adolfsson et al., 2001; Adolfsson et al., 2005) has been an elaborate task, yielding several candidate cell populations that represent thymic progenitors. Over all, the ETP, CTP and bone marrow population believed responsible for contributing to 176 thymic T cells is characterized as LSK IL-7Roc" Flt3+ (Adolfsson et al., 2001; Adolfsson et al , 2005; Allman et al., 2003; Donskoy and Goldschneider, 1992; Schwarz and Bhandoola, 2004) . As we are interested in the homing of these progenitors to the thymus and selectins are highly implicated in recruitment of cells where shear forces are involved as would be the case in thymic homing, we surveyed the P-selectin-binding abilities of candidate thymic stem cell populations. As published in a recent article by our group, in collaboration with the laboratory of Dr. Rossi, P-selectin ligand expression occurs on both ETPs (defined as Lin" CD3"CD4"CD8"CD25"CD44+c-Kit+) present in the thymus and their putative precursors in the bone marrow (defined as CLP: Lin"c-Kit loSca-l loIL-7R+ and KLS: Lin"c-Kit+Sca-1+) and bloodstream (Lin"c-Kit+Sca-1+). P-selectin binding depends on many posttranslational modifications of selectin ligands; one key modification is O-glycan branching mediated by C2GlcNAcT-I, as discussed in section 1.5.2. Progenitor cells from C2GlcNacT-Pu11 mice did not bind P-selectin, underlining the importance of C2GlcNAcT activity in P-selectin ligand formation on these cells (Rossi et al., 2005). Our data show that C2GlcNAcT is constitutively expressed in thymic progenitors in contrast with naive T cells, where C2GlcNAcT activity occurs only following T cell activation. These data demonstrate that P-selectin ligands are also present on the putative CTP population responsible for homing to the thymus, implicating PSGL-1 and P-selectin as players in this homing process. Immunofluorescence staining of frozen thymus sections from wild-type mice using a sensitive assay reveals P-selectin expression throughout the thymus vasculature (Rossi et al., 2005) , supporting the premise that P-selectin is involved in the homing of progenitors to the thymus. The presence of these critical players, typically involved mediating the early steps in extravasation, provoked a more thorough investigation into how these molecules are functionally involved in thymic progenitor homing. 177 Mechanism of entry of progenitors into thymus Using mice deficient in PSGL-1, C2GlcNAcT-I and P-selectin, we arranged a number of competitive assays to provide a more comprehensive understanding of the molecular mechanisms involved in thymic progenitor homing. Analysis of PSGL-1 n u l 1 mice parabiosed to wild-type mice uncovered a functional defect of PSGL-l n u 1 1 thymic progenitors to colonize the adult murine thymus; after five weeks of parabiosis, there were only 2% PSGL-1 n u l 1 cells in the wild-type thymus, whereas over 50% of the cells in the thymi of the PSGL-1 n u" partners were derived from the wild-type progenitors. Similar results were obtained for parabiotic pairs between wild-type mice and C2GlcNAcT-In u 1 1 mice, suggesting the importance of C2GlcNAcT modification of PSGL-1 for its binding to P-selectin. When these results were compared with control wild-type to wild-type parabiotic pairs, it became apparent that PSGL-1 n u l 1 and C2GlcNAcT-In u 1 1 cells not only have a significant defect in their ability to home to thymi compared with wild-type cells, but the percentage of wild-type cells colonizing the thymi of either PSGL-1 n u" or C2GlcNAcT-In u 1 1 mice - where the average DP thymocyte chimerism reached up to 50% after five weeks of parabiosis - was considerably higher than that observed for control parabionts analyzed where chimerism was in the range of 7% to 12%, similar to that previously observed for long-term parabiotic pairs (Donskoy and Goldschneider, 1992). The remarkably improved chimerism conveyed in these experiments compared with control parbionts and those reported for long-term parabiotic pairs cannot simply be due to the competitive advantage of wild-type cells over PSGL-1 n u l 1 cells; it is also suggested that PSGL-1 n u l 1 thymi may have a higher number of available progenitor niches compared with wild-type thymi (Rossi et al., 2005). The number of available progenitor niches correlates with the efficiency of CTP recruitment (Prockop and Petrie, 2004), suggesting that PSGL-178 l n u thymi are more proficient than wild-type thymi in recruiting CTPs. This is consistent with the finding that cells lacking PSGL-1 are unable to efficiently fill thymic stromal niches. Since PSGL-1 n u l 1 mice are not reported to have a defect in T cell production, it seems counterintuitive to suggest that PSGL-1 has a critical role in seeding the thymus. A detailed analysis of the thymi of PSGL-1 n u l 1 mice, however, did unveil that the thymus, in addition to containing more unoccupied stromal niches, has significantly reduced numbers of ETPs compared with its wild-type counterparts (Rossi et al., 2005). This suggests that an early progenitor defect compatible with a thymic homing impairment exists. Competitive repopulation studies using wild-type bone marrow cells and PSGL-1 n u l 1 or C2GlcNAcT-In u 1 1 cells into irradiated wild-type recipients yielded results similar to those observed in the parabiotic models. Both PSGL- l n u " and C2GlcNAcT-I n u l ! cells were less efficient than wild-type cells; on average, only 15% and 9% of DP cells in the recipient thymi were of PSGL-1 n u" or C2GlcNAcT-In u" origin, respectively. Since irradiation modifies the expression of selectins and other adhesive receptors on endothelium, potentially altering selectin-mediated recruitment processes (Mazo et al., 2002), non-irradiated IL-7Rn u 1 1 mice were used as recipients in additional competitive repopulation studies. IL-7Rn u 1 1 mice lack mature T cells due to a developmental block in DN cell differentiation, and the thymus is small in size. However, the thymic architecture and microenvironments are normal in these mice and can quickly and effectively become colonized by transplanted wild-type cells (Prockop and Petrie, 2004). Analysis of DP cells in the competitively reconstituted IL-7Rn u 1 1 mice again agrees with the results obtained from parabiosis, and from competition in irradiated wild-type animals where wild-type cells are present at over 95% when competed with cells that lack PSGL-1 or C2GlcNAcT-I. Interestingly in a competition between PSGL-l n u U cells and C2GlcNAcT-In u 1 1 cells, C2GlcNAcT-In u" cells were significantly better at seeding the thymus, suggesting that PSGL-1 is the main substrate of C2GlcNAcT-I involved 179 in thymic homing and that other C2GlcNAcT (for example C2GlcNAcT-III) isoenzymes may compensate for the loss of C2GlcNAcT-I (Merzaban et al., 2005; Schwientek et al., 2000; Yeh et al., 1999). Additionally, PSGL-1 may interact with E-selectin in a C2GlcNAcT-independent fashion (Borges et al., 1997b). To exclude the possibility that the progenitors coming from the PSGL-1 n u l 1 mice are simply expanding at different rates than their wild-type counterparts, we measured the ratio between PSGL-1 n u l 1 and wild-type cells, finding them to be constant across the DN thymic subsets; this indicates that both cells expand at the same rate (Rossi et al., 2005). Competitive intrathymic injection experiments furthermore showed that bone marrow KLS cells from both PSGL-1 n u l 1 and wild-type mice efficiently generate DP thymocytes and do so with similar kinetics (Rossi et al., 2005). Short-term competitive homing assays were also set up using the fluorochrome tracking dyes, CFSE or Cell Tracker Orange (CTO), for the two competing populations. These studies reveal in short-term (two-day) homing assays that P-selectin controls thymic progenitor cell entry. They also illustrate that wild-type cells accumulate up to fivefold more efficiently in the wild-type and IL-7Rnull-recipient thymi than in PSGL-1 n u l 1 or C2GlcNAcT-Inu", and that this competitive advantage is lost in P-selectinnu11-recipient mice (Rossi et al., 2005). Since proliferation can be monitored using such dyes, any preferential expansion can be ruled out. Using P-selectinnu" mice in competitions either in parabiotic models or as recipients in competitive reconstitution studies supports findings that P-selectin is involved in thymic progenitor cell homing. Migration of wild-type cells to the P-selectinnu11 thymus is impaired in the parabiotic model studied, whereas P-selectinnu11 cells migrated normally to the wild-type partner's thymus (relative to control wild-type to wild-type pairs). Thus, P-selectin appears to facilitate the homing of progenitors to the thymus. Competitive reconstitution of 180 P-selectin -recipient mice further reveals that the advantage of wild-type cells over PSGL-l n u 1 1 cells, and C2GlcNAcT-In u 1 1 cells observed in wild-type recipient mice is lost when P-selectin is removed from the equation; this indicates that PSGL-1 interaction with thymic P-selectin is required in the homing process. The ability of CTPs to engraft the adult thymus correlates with the availability of unoccupied stromal niches (Prockop and Petrie, 2004). Prockop and Petrie effectively demonstrate that competition by DN cells for thymic stromal niches regulates thymus size. However, the mechanisms involved in this process are currently unknown. One hypothesis is that CTPs may constitutively home to the thymus, their survival dependent on their ability to engraft in the appropriate stromal niche. Another hypothesis is that a feedback mechanism exists that senses the availability of niches and relates this information to the thymic endothelium, leading to the increased expression of molecules that facilitate CTP recruitment. Since the interaction of P-selectin with PSGL-1 increases the efficiency of thymic progenitor recruitment, we propose that P-selectin is a component of such a feedback loop, where its expression would increase in relation to the number of vacant niches. PSGL-l n u " (Rossi et al., 2005) and IL-7Rn u 1 1 (Prockop and Petrie, 2004) mouse thymi are believed to have more empty progenitor niches. Using P-selectin mRNA levels as a readout, increased levels of P-selectin mRNA are found in the thymi of PSGL-1 n u" as well as IL-7Rn u 1 1 mice, suggesting that P-selectin expression is regulated by the presence of vacant progenitor niches. The demonstration that repopulating such niches in IL-7Rnu" thymi leads to downregulation of P-selectin RNA strongly supports this idea (Rossi et al., 2005). These data, along with all the other data described, indicate that a feedback-control model exists where recruitment of thymic progenitors occurs when progenitors in specific niches are depleted as they differentiate and leave the thymus, supporting data already describing the 181 presence of cyclic waves of progenitor recruitment in adult thymi (Foss et al., 2001; Prockop and Petrie, 2004). In addition to the current work presented here on thymic progenitor homing, prior reports implicate CD44 as a molecule involved in the homing process. A recent study shows that the small GTPase RAP1 mediates the extravasation of mature T cells across vascular endothelium by redistributing CD44 and the chemokines receptor, CXCR4, within the membrane of extravasating cells (Shimonaka et al., 2003). Since both CD44 and CXCR4 are expressed by early thymic progenitors (Kitchen and Zack, 1997; Picker et al., 1990b) and since CD44 may play a role in thymic homing (Khaldoyanidi et a l , 1997; Wu et al., 1993) it is possible that RAP1 may mediate the extravasation events of thymic progenitors. To help confirm or refute the involvement of CD44 in the homing process, we set up a number of competitions using CD44n u 1 1 mouse cells into irradiated, wild-type, CD44n u 1 1, or P-selectinnu" recipient mice and into non- irradiated IL-7Rn u 1 1 mice. These preliminary studies indicate that lack of CD44 does not impact their ability to enter the thymus (Appendix II; Figure A-3) which is consistent with some studies in the literature (Khan et al., 2004; Si-Tahar et al., 2001; Wang et al., 2002). Instead, these studies point to a role for CD44 in negatively regulating progenitor entry into the thymus (see Figure A-4 for model). Although these results challenge the literature, there are a number of differences in our experimental set up and background of mice used for these experiments. Further analysis using parabiotic mice pairs between wild-type and CD44n u 1 1 mice will help resolve some of these issues. 182 PSGL-1 is involved in lymphocyte homing to lymph nodes? Upon analyzing both peripheral and mesenteric lymph node populations from parabionts of wild-type and P S G L - 1 n u l 1 pairs, we found twice as many wild-type C D 4 + and C D 8 + T cells in these lymphoid organs than T cells lacking P S G L - 1 . This observation was made in lymph nodes from either mouse in the pair, indicating that PSGL-1 is partially required for effective homing of nai've cells to lymph nodes. This difference was not observed for C 2 G l c N A c T - I n u " T cells, indicating that lack of PSGL-1 but not loss of C 2 G l c N A c T - I may affect homing to lymph nodes and that P-selectin is not likely the counter-receptor involved in this process. Our data point to a role for PSGL-1 in the physiological homing of naive T cells to lymph nodes. Data in the literature indicate that L -selectin and its ligands are responsible for lymphocyte homing to peripheral organs and no role for PSGL-1 in this process has yet been described. This raises a number of interesting theories: 1) PSGL-1 may be differentially involved T cell homing to lymph node populations. L -selectin-dependent lymphocyte homing is less-prominent in mesenteric lymph nodes and Peyer's patches (Bargatze et al., 1995; Butcher and Picker, 1996). In these assays, all lymph node populations were combined and not analyzed separately. Future experiments wi l l focus on comparing peripheral lymph node populations with mesenteric lymph nodes, and wi l l include Peyer's patches; 2) Secondary lymphocyte tethering may play a significant role in the lymphocyte homing observed in these parabiotic models. Secondary lymphocyte homing implicates PSGL-1 interactions with L-selectin as a recruitment mechanism (Alon et al., 1996; Bargatze et al., 1994; Eriksson et al., 2001; Kunkel et al., 1998; Spertini et al., 1996; Tu et al., 1996; Walcheck et al., 1996) L-selectin expressed on nai've T cells mediates binding of cells to 183 HEV-expressed PNAds. L-selectin on these bound cells also interacts with PSGL-1 on circulating naive T cells, and in doing so propogates the tethering and rolling of T cells over the pre-existing adherent cells. Alternatively, the process could be opposite - L-selectin-expressed circulating T cells could interact with PSGL-1 on adherent cells to yield the same result. Figure 1-5 illustrates this; 3) Platelets that express P-selectin can also mediate PSGL-1-dependent naive T cell homing and this has been investigated in the context of L-selectinnu" mice. P-selectin on circulating activated platelets can mediate simultaneous adhesion to PNAds on HEV and to PSGL-1 on lymphocytes forming a "cellular bridge," which can "rescue" lymphocyte homing in L-selectinnu11 mice (Diacovo et al., 1998; Diacovo et al., 1996; Martinez et al., 2005; Warnock et al., 1998). This process may be partially involved in homing of naive cells within the parabiotic model studied; and 4) PSGL-1 binding to its counter-receptor, either P-selectin or E-selectin (some studies have found that P- and E-selectin are constitutively expressed (Kivisakk et al., 2003)), may, in actuality, not depend on C2GlcNAcT-I, but on other C2GlcNAcT isoenzymes (Merzaban et al., 2005; Schwientek et al., 1999; Schwientek et al., 2000; Yeh et al., 1999), in addition to the other glycosyltransferases involved in creating functional selectin ligands. C2GlcNAcT-I contributes to L-selectin ligand synthesis in high-endothelial venules (Yeh et al., 2001). Whether C2GlcNAcT-III is also involved in L-selectin ligand synthesis is not yet known, but RNA for human C2GlcNAcT-III, most strongly expressed in the thymus, is also detected in lymph nodes (Schwientek et al., 2000). 184 To address whether P-selectin and/or E-selectin ligands and the C2GlcNAcT enzymes are involved in nai've cell homing to the lymph node, future experiments will focus on competitive short-term homing assays looking directly at naive T cell populations and their ability to home to lymph nodes and Peyer's patches. More specifically, harvesting T cells from peripheral and mesenteric lymph nodes from wild-type, PSGL-l n u " , C2GlcNAcT-In u" mice and double null for C2GlcNAcT-I/T-III, and labelling cells with CFDA-SE and a second population with CTO, mixing cells 1:1 and comparing ratios of cells homing to LN in wild-type recipients. If significant differences are observed, competitive homing experiments will be repeated in P- and E-selectin""11 mice. These experiments should help resolve the role PSGL-1 has in homing of T cells to lymph nodes and to what degree C2GlcNAcT-I and C2GlcNAcT-III co-operate in the control of selectin ligand formation. Through analogy with the role of P-selectin and its ligand, PSGL-1, in activated lymphocytes, we propose that engagement of P-selectin with PSGL-1 is likely the first step in the sequence of events leading to the recruitment of thymic progenitors across the thymic endothelium and into the thymus (summarized in Figure 5-5). Circulating thymic progenitors expressing PSGL-1 interact with endothelial P-selectin, thereby allowing increased exposure to chemokines, subsequently leading to arrest due to conformational changes in integrins followed by transmigration through the thymic endothelium. While the events following P-selectin binding to PSGL-1 remain to be identified, these results have provided the first experimental insight into the molecular mechanisms directing homing of T cell progenitors to thymus. 185 Figure 5-5: T cell progenitor entry into the thymus is mediated by C2GlcNAcT-I-modified PSGL-1 and endothelial expressed P-selectin 186 CHAPTER 6 SUMMARY AND FUTURE DIRECTIONS The objective of the work presented here is to examine and determine the relevance of the glycosyltransferases C2GlcNAcTs, an often-overlooked enzyme family, in selectin ligand synthesis. Since C2GlcNAcTs have a pivotal role in controlling leukocyte trafficking, they are ideal candidates in controlling the inflammation that is at the heart of many disorders and diseases. Selectins play a crucial role in the immune response and hematopoiesis, as demonstrated in selectin-deficient and selectin-ligand-deficient patients, as well as mouse models. Disease treatment using selectin ligand inhibition shows most promise in ischemia-reperfusion-type situations in transplantation and in skin diseases. The effects of transient selectin inhibition in well-controlled clinical settings, such as organ transplantation, require further examination. Since selectins are glycan-binding molecules that bind glycoprotein ligands often modulated by C2GlcNAcT, one intriguing suggestion is to produce C2GlcNAcT inhibitors. These inhibitors would mainly target the P-selectin ligand, PSGL-1, hence impacting the activities of all three selectins. CHAPTER 3 and CHAPTER 4 strive to characterize other murine isoenzymes within the C2GlcNAcT family. We determine that C2GlcNAcT-III, in addition to well-characterized C2GlcNAcT-I, is able to create functional P-selectin ligands on activated CD8 + T cells. In addition, the C2GlcNAcT-III isoenzyme is capable of adding core 2 branches to other leukocyte mucins, including CD43 and CD45. This unveils for the first time that C2GlcNAcT-I is not solely responsible for creating functional P-selectin ligands. Both of these enzymes, meanwhile, may function in in vz'vo-to-direct selectin ligand formation and cell trafficking (see cartoon). CHAPTER 5 concentrates on describing the function of C2GlcNAcT-I in mediating progenitor homing to the thymus. By modifying PSGL-1 expressed on the surface of thymic progenitor cells, C2GlcNAcT-I plays a part in directing interactions with P-selectin on the 187 thymic endothelium and, hence, in guiding blood-bome thymic progenitors into the thymus to maintain the adult T cell pool. Analogous with the role of L-selectin expressed on naive T cells, and its place in mediating constitutive lymphocyte recruitment to the HEV of secondary lymphoid organs via interactions with PNAds, these studies define P-selectin and its core 2 glycan-modified ligand, PSGL-1, as essential components in mediating constitutive CTP recruitment to the thymus. These studies are the first to embark on the homing of thymic progenitors to the thymus, and promise to open the field to greater discovery and understanding. The observation that C2GlcNAcT-III can support P-selectin ligand expression in CD8 + T cells points to a place for C2GlcNAcT-III in T cell trafficking. Identifying the relative contributions of C2GlcNAcT-I and C2GlcNAcT-III to P-selectin ligand formation and cell trafficking will demand C2GlcNAcT-III-deficient mice. Generating these currently unavailable mice is a priority in our lab. Based on observations of C2GlcNAcT-In u" mice, C2GlcNAcT-IHn u 1 1 mice will most likely have no overt phenotype. Mice can be crossed with the C2GlcNAcT-In u" mouse to generate double-null mice to determine if loss of both of these enzymes leads to a much-more-pronounced phenotype than that found in the C2GlcNAcT-In u 1 1 alone. Since indications suggest that C2GlcNAcT-III is predominantly expressed in T cells, thymic subsets in C2GlcNAcT-IHn u" and T-I/T-III-double-null mice will be analyzed for abnormalities in cell compartments and distribution. Further delineating the regulation of selectin ligand synthesis, and enhancing our understanding of the actions of key enzymes involved in this process - such as C2GlcNAcT-I and C2GlcNAcT-III - will contribute to developing comprehensive models of lymphocyte migration. 188 R E F E R E N C E S Abkowitz, J. L. , Robinson, A. E. , Kale, S., Long, M . W., and Chen, J. (2003). Mobilization of hematopoietic stem cells during homeostasis and after cytokine exposure. Blood 102,1249-1253. Adema, G. J. , Hartgers, F., Verstraten, R., de Vries, E. , Marland, G., Menon, S., Foster, J., Xu, Y., Nooyen, P., McClanahan, T., et al. (1997). A dendritic-cell-derived C-C chemokine that preferentially attracts naive T cells. Nature 5*7,713-717. Adolfsson, J., Borge, O. J., Bry der, D., Theilgaard-Monch, K., Astrand-Grundstrom, I., Sitnicka, E., Sasaki, Y., and Jacobsen, S. E. (2001). Upregulation of Flt3 expression within the bone marrow Lin(-)Scal(+)c-kit(+) stem cell compartment is accompanied by loss of self-renewal capacity. Immunity 15,659-669. Adolfsson, J., Mansson, R., Buza-Vidas, N., Hultquist, A., Liuba, K., Jensen, C. T., Bryder, D., Yang, L . , Borge, O. J., Thoren, L . A., et al. (2005). Identification of flt3(+) lympho-myeloid stem cells lacking erythro-megakaryocytic potential a revised road map for adult blood lineage commitment. Cell 121,295-306. Aigner, S., Ramos, C. L . , Hafezi-Moghadam, A., Lawrence, M . B., Friederichs, J., Altevogt, P., and Ley, K. (1998). CD24 mediates rolling of breast carcinoma cells on P-selectin. Faseb J 12,1241-1251. Aigner, S., Sthoeger, Z. M. , Fogel, M. , Weber, E. , Zarn, J. , Ruppert, M . , Zeller, Y., Vestweber, D., Stahel, R., Sammar, M. , and Altevogt, P. (1997). CD24, a mucin-type glycoprotein, is a ligand for P-selectin on human tumor cells. Blood 89,3385-3395. 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. Alfalah, M. , Jacob, R., Preuss, U., Zimmer, K. P., Nairn, H., and Nairn, H. Y. (1999). O-linked glycans mediate apical sorting of human intestinal sucrase-isomaltase through association with lipid rafts. Curr Biol 9,593-596. Allenspach, E . J., Cullinan, P., Tong, J., Tang, Q., Tesciuba, A. G., Cannon, J. L . , Takahashi, S. M. , Morgan, R., Burkhardt, J. K., and Sperling, A. I. (2001). ERM-Dependent Movement of CD43 Defines a Novel Protein Complex Distal to the Immunological Synapse. Immunity 15, 739-750. Allman, D., Sambandam, A., Kim, S., Miller, J. P., Pagan, A., Well, D., Meraz, A., and Bhandoola, A. (2003). Thymopoiesis independent of common lymphoid progenitors. Nat Immunol 4,168-174. Alon, R., Fuhlbrigge, R. C , Finger, E. B., and Springer, T. A. (1996). Interactions through L-selectin between leukocytes and adherent leukocytes nucleate rolling adhesions on selectins and VCAM-1 in shear flow. J Cell Biol 135,849-865. Alon, R., Kassner, P. D., Carr, M . W., Finger, E . B., Hemler, M . E. , and Springer, T. A. (1995). The integrin VLA-4 supports tethering and rolling in flow on VCAM-1. J Cell Biol 128,1243-1253. Anderson, G., and Jenkinson, E. J. (2001). Lymphostromal interactions in thymic development and function. Nat Rev Immunol 1, 31-40. Arar, C , Carpentier, V., Le Caer, J. P., Monsigny, M. , Legrand, A., and Roche, A. C. (1995). ERGIC-53, a membrane protein of the endoplasmic reticulum-Golgi intermediate compartment, is identical to MR60, an intracellular mannose-specific lectin of myelomonocytic cells. J Biol Chem 270,3551-3553. Arbones, M . L. , Ord, D. C , Ley, K., Ratech, H., Maynard-Curry, C , Otten, G., Capon, D. J., and Tedder, T. F. (1994). Lymphocyte homing and leukocyte rolling and migration are impaired in L-selectin-deficient mice. Immunity 1, 247-260. 189 Ardavin, C , Wu, L. , L i , C. L . , and Shortman, K. (1993). Thymic dendritic cells and T cells develop simultaneously in the thymus from a common precursor population. Nature 362, 761-763. Ardman, B., Sikorski, M . A., and Staunton, D. E . (1992). CD43 interferes with T-lymphocyte adhesion. Proc Natl Acad Sci U S A 89,5001-5005. Arnold, C. S., Johnson, G. V., Cole, R. N., Dong, D. L. , Lee, M. , and Hart, G. W. (1996). The microtubule-associated protein tau is extensively modified with O-linked N-acetylglucosamine. J Biol Chem 271,28741-28744. Asa, D., Raycroft, L . , Ma, L . , Aeed, P. A., Kaytes, P. S., Elhammer, A. P., and Geng, J. G. (1995). The P-selectin glycoprotein ligand functions as a common human leukocyte ligand for P- and E-selectins. J Biol Chem 270,11662-11670. Asano, M. , Furukawa, K., Kido, M. , Matsumoto, S., Umesaki, Y., Kochibe, N., and Iwakura, Y. (1997). Growth retardation and early death of beta-l,4-galactosyltransferase knockout mice with augmented proliferation and abnormal differentiation of epithelial cells. Embo J16,1850-1857. Asano, M. , Nakae, S., Kotani, N., Shirafuji, N., Nambu, A., Hashimoto, N., Kawashima, H., Hirose, M. , Miyasaka, M. , Takasaki, S., and Iwakura, Y. (2003). Impaired selectin-ligand biosynthesis and reduced inflammatory responses in beta-l,4-galactosyltransferase-I-deficient mice. Blood 102,1678-1685. Axelsson, B., Youseffi-Etemad, R., Hammarstrom, S., and Perlmann, P. (1988). Induction of aggregation and enhancement of proliferation and LL-2 secretion in human T cells by antibodies to CD43. Journal of Immunology 141,2912-2917. Baecher-Allan, C. M. , Kemp, J. D., Dorfman, K. S., Barth, R. K., and Frelinger, J. G. (1993). Differential epitope expression of Ly-48 (mouse leukosialin). Immunogenetics 37,183-192. Baenziger, J. U. (1994). Protein-specific glycosyltransferases: how and why they do it! Faseb J 8,1019-1025. Bakowsky, U., Schumacher, G., Gege, C , Schmidt, R. R., Rothe, U., and Bendas, G. (2002). Cooperation between lateral ligand mobility and accessibility for receptor recognition in selectin-induced cell rolling. Biochemistry 41,4704-4712. Bargatze, R. F., Jutila, M . A., and Butcher, E. C. (1995). Distinct roles of L-selectin and integrins alpha 4 beta 7 and LFA-1 in lymphocyte homing to Peyer's patch-HEV in situ: the multistep model confirmed and refined. Immunity 3,99-108. Bargatze, R. F., Kurk, S., Butcher, E . C , and Jutila, M . A. (1994). Neutrophils roll on adherent neutrophils bound to cytokine-induced endothelial cells via L-selectin on the rolling cells. J Exp Med 180,1785-1792. Barran, P., Fellinger, W., Warren, C. E. , Dennis, J. W., and Ziltener, H. J. (1997). Modification of CD43 and other lymphocyte O-glycoproteins by core 2 N-acetylglucosaminyltransferase. Glycobiology 7,129-136. Bartholdy, C , Marker, O., and Thomsen, A. R. (2000). Migration of activated CD8(+) T lymphocytes to sites of viral infection does not require endothelial selectins. Blood 95,1362-1369. Baum, L. G., Pang, M. , Perillo, N. L . , Wu, T., Delegeane, A., Uittenbogaart, C. H., Fukuda, M. , and Seilhamer, J. J . (1995). Human thymic epithelial cells express an endogenous lectin, galectin-1, which binds to core 2 O-glycans on thymocytes and T lymphoblastoid cells. J Exp Med 181,877-887. Baumheter, S., Singer, M . S., Henzel, W., Hemmerich, S., Renz, M. , Rosen, S. D., and Lasky, L . A. (1993). Binding of L-selectin to the vascular sialomucin CD34. Science 262,436-438. Bazil, V., Brandt, J. , Chen, S., Roeding, M. , Luens, K., Tsukamoto, A., and Hoffman, R. (1996). A monoclonal antibody recognizing CD43 (leukosialin) initiates apoptosis of human hematopoietic progenitor cells but not stem cells. Blood 87,1272-1281. 190 Bazil, V., Brandt, J. , Tsukamoto, A., and Hoffman, R. (1995). Apoptosis of human hematopoietic progenitor cells induced by crosslinking of surface CD43, the major sialoglycoprotein of leukocytes. Blood 86,502-511. Bazil, V., and Strominger, J. L . (1993). CD43, the major sialoglycoprotein of human leukocytes, is proteolytically cleaved from the surface of stimulated lymphocytes and granulocytes. Proceedings of the National Academy of Sciences of the United States of America 90,3792-3796. Bazil, V., and Strominger, J. L . (1994). Metalloprotease and serine protease are involved in cleavage of CD43, CD44, and CD16 from stimulated human granulocytes. Induction of cleavage of L-selectin via CD16. Journal of Immunology 152,1314-1322. Becker, D. J., and Lowe, J. B. (1999). Leukocyte adhesion deficiency type II. Biochim Biophys Acta 1455,193-204. Becker, D. J., and Lowe, J. B. (2003). Fucose: biosynthesis and biological function in mammals. Glycobiology 13,41R-53R. Becker, D. J., Myers, J. T., Ruff, M . M. , Smith, P. L. , Gillespie, B. W., Ginsburg, D. W., and Lowe, J. B. (2003). Strain-specific modification of lethality in fucose-deficient mice. Mamm Genome 14,130-139. Behnia, R., Panic, B., Whyte, J. R., and Munro, S. (2004). Targeting of the Arf-like GTPase Arl3p to the Golgi requires N-terminal acetylation and the membrane protein Syslp. Nat Cell Biol 6,405-413. Beisswanger, R., Corbeil, D., Vannier, C , Thiele, C , Dohrmann, U., Kellner, R., Ashman, K., Niehrs, C , and Huttner, W. B. (1998). Existence of distinct tyrosylprotein sulfotransferase genes: Molecular characterization of tyrosylprotein sulfotransferase-2. PNAS 95,11134-11139. Ben-Mahmud, B. M. , Mann, G. E. , Datti, A., Orlacchio, A., Kohner, E. M. , and Chibber, R. (2004). Tumor Necrosis Factor-{alpha} in Diabetic Plasma Increases the Activity of Core 2 GlcNAc-T and Adherence of Human Leukocytes to Retinal Endothelial Cells: Significance of Core 2 GlcNAc-T in Diabetic Retinopathy. Diabetes 53,2968-2976. Bendas, G., Vogel, J., Bakowski, U., Krause, A., Muller, J. , and Rothe, U. (1997). A Uposome-based model system for the simulation of lectin-induced cell adhesion. Biochim Biophys Acta 1325,297-308. Bennett, E. P., Hassan, H., Mandel, U., Mirgorodskaya, E. , Roepstorff, P., Burchell, J., Taylor-Papadimitriou, J., Hollingsworth, M . A., Merkx, G., van Kessel, A. G., et al. (1998). Cloning of a human UDP-N-acetyl-alpha-D-Galactosamine:polypeptide N-acetylgalactosaminyltransferase that complements other GalNAc-transferases in complete O-glycosylation of the MUC1 tandem repeat. J Biol Chem 273,30472-30481. Berg, E. L . , McEvoy, L . M. , Berlin, C , Bargatze, R. F., and Butcher, E. C. (1993). L-selectin-mediated lymphocyte rolling on MAdCAM-1. Nature 366,695-698. Berg, E. L . , Yoshino, T., Rott, L . S., Robinson, M . K., Warnock, R. A., Kishimoto, T. K., Picker, L. J., and Butcher, E . C. (1991). The cutaneous lymphocyte antigen is a skin lymphocyte homing receptor for the vascular lectin endotheUal cell-leukocyte adhesion molecule 1. J Exp Med 174,1461-1466. Berger, E . G. (1999). Tn-syndrome. Biochim Biophys Acta 1455,255-268. Berlin, C , Bargatze, R. F., Campbell, J. J. , von Andrian, U. H., Szabo, M . C , Hasslen, S. R., Nelson, R. D., Berg, E. L . , Erlandsen, S. L. , and Butcher, E . C. (1995). alpha 4 integrins mediate lymphocyte attachment and rolling under physiologic flow. Cell 80,413-422. Bernimoulin, M . P., Zeng, X. L . , Abbal, C , Giraud, S., Martinez, M. , Michielin, O., Schapira, M . , and Spertini, O. (2003). Molecular basis of leukocyte rolling on PSGL-1. Predominant role of core-2 O-glycans and of tyrosine sulfate residue 51. J Biol Chem 278,37-47. 191 Beum, P. V., and Cheng, P. W. (2001). Biosynthesis and function of beta 1,6 branched mucin-type glycans. Adv Exp Med Biol 491,279-312. Bhandoola, A., Sambandam, A., Allman, D., Meraz, A., and Schwarz, B. (2003). Early T lineage progenitors: new insights, but old questions remain. J Immunol 777,5653-5658. Bieberich, E. , Freischutz, B., Liour, S.-S., and Yu, R. K. (1998). Regulation of Ganglioside Metabolism by Phosphorylation and Dephosphorylation. J Neurochem 71,972-979. Bieberich, E. , and Yu, R. K. (1999). Multi-enzyme kinetic analysis of glycolipid biosynthesis. Biochim Biophys Acta 7452,113-124. Bierhuizen, M . F., and Fukuda, M . (1992). Expression cloning of a cDNA encoding UDP-GlcNAc:Gal beta 1-3-GalNAc-R (GlcNAc to GalNAc) beta l-6GlcNAc transferase by gene transfer into CHO cells expressing polyoma large tumor antigen. Proc Natl Acad Sci U S A 89,9326-9330. Bistrup, A., Bhakta, S., Lee, J. K., Belov, Y. Y., Gunn, M . D., Zuo, F. R., Huang, C. C , Kannagi, R., Rosen, S. D., and Hemmerich, S. (1999). Sulfotransferases of two specificities function in the reconstitution of high endothelial cell ligands for L-selectin. J Cell Biol 145,899-910. Blais, M . E. , Gerard, G., Martinic, M . M. , Roy-Proulx, G., Zinkernagel, R. M. , and Perreault, C. (2004). Do thymically and strictly extrathymic ally developing T cells generate similar immune responses? Blood 103, 3102-3110. Blander, J. M. , Visintin, I., Janeway, C. A., Jr., and Medzhitov, R. (1999). Alpha(l,3)-fucosyltransferase VII and alpha(2,3)-sialyltransferase TV are up-regulated in activated CD4 T cells and maintained after their differentiation into Thl and migration into inflammatory sites. J Immunol 163,3746-3752. Bleul, C. C , and Boehm, T. (2000). Chemokines define distinct microenvironments in the developing thymus. Eur J Immunol 30,3371-3379. Boileau, C., Houde, M . , Dulude, G., Clegg, C. H., and Perreault, C. (2000). Regulation of extrathymic T cell development and turnover by oncostatin M . J Immunol 164,5713-5720. Borges, E. , Eytner, R., Moll, T., Steegmaier, M. , Campbell, M . A., Ley, K., Mossmann, H., and Vestweber, D. (1997a). The P-selectin glycoprotein ligand-1 is important for recruitment of neutrophils into inflamed mouse peritoneum. Blood 90,1934-1942. Borges, E. , Pendl, G., Eytner, R., Steegmaier, M. , Zollner, O., and Vestweber, D. (1997b). The binding of T cell-expressed P-selectin glycoprotein ligand-1 to E - and P-selectin is differentially regulated. J Biol Chem 272,28786-28792. Borges, E. , Tietz, W., Steegmaier, M . , Moll, T., Hallmann, R., Hamann, A., and Vestweber, D. (1997c). P-selectin glycoprotein ligand-1 (PSGL-1) on T helper 1 but not on T helper 2 cells binds to P-selectin and supports migration into inflamed skin. J Exp Med 185,573-578. Bosse, R., and Vestweber, D. (1994). Only simultaneous blocking of the L - and P-selectin completely inhibits neutrophil migration into mouse peritoneum. Eur J Immunol 24,3019-3024. Brockhausen, I., Kuhns, W., Schachter, H., Matta, K. L. , Sutherland, D. R., and Baker, M . A. (1991). Biosynthesis of O-glycans in leukocytes from normal donors and from patients with leukemia: increase in O-glycan core 2 UDP-GlcNAc:Gal beta 3 GalNAc alpha-R (GlcNAc to GalNAc) beta(l-6)-N-acetylglucosaminyltransferase in leukemic cells. Cancer Res 51,1257-1263. Brockhausen, I., Matta, K. L. , Orr, J. , and Schachter, H. (1985). Mucin synthesis. UDP-GlcNAc:GalNAc-R beta 3-N- acetylglucosaminyltransferase and UDP-GlcNAc:GlcNAc beta l-3GalNAc-R (GlcNAc to GalNAc) beta 6-N-acetylglucosaminyltransferase from pig and rat colon mucosa. Biochemistry 24,1866-1874. 192 Brockhausen, I., Toki, D., Brockhausen, J., Peters, S., Bielfeldt, T., Kleen, A., Paulsen, H., Meldal, M . , Hagen, F., and Tabak, L . A. (1996). Specificity of O-glycosylation by bovine colostrum UDP-GalNAc: polypeptide alpha-N-acetylgalactosaminyltransferase using synthetic glycopeptide substrates. Glycoconj J13, 849-856. Broide, D. H., Miller, M. , Castaneda, D., Nayar, J., Cho, J. Y., Roman, M. , Ellies, L . G., and Sriramarao, P. (2002). Core 2 oligosaccharides mediate eosinophil and neutrophil peritoneal but not lung recruitment. Am J Physiol Lung Cell Mol Physiol 282, L259-266. Brown, T. J. , Shuford, W. W., Wang, W. C , Nadler, S. G., Bailey, T. S., Marquardt, H., and Mittler, R. S. (1996). Characterization of a CD43/leukosialin-mediated pathway for inducing apoptosis in human T-lymphoblastoid cells. J Biol Chem 271, 27686-27695. Bruckner, K., Perez, L. , Clausen, H., and Cohen, S. (2000). Glycosyltransferase activity of Fringe modulates Notch-Delta interactions. Nature 406,411-415. Bullard, D. C , Kunkel, E. J. , Kubo, H., Hicks, M . J., Lorenzo, I., Doyle, N. A., Doerschuk, C. M. , Ley, K., and Beaudet, A. L . (1996). Infectious susceptibility and severe deficiency of leukocyte rolling and recruitment in E-selectin and P-selectin double mutant mice. J Exp Med 183,2329-2336. Bunster, E. , and Meyer, R. K. (1933). An improved method of parabiosis. The Anatomical Record 57,339 -343. Burdick, M . M . , Bochner, B. S., Collins, B. E. , Schnaar, R. L. , and Konstantopoulos, K. (2001). Glycolipids support E-selectin-specific strong cell tethering under flow. Biochem Biophys Res Commun 284,42-49. Burns, J. A., Issekutz, T. B., Yagita, H., and Issekutz, A. C. (2001). The alpha 4 beta 1 (very late antigen (VLA)-4, CD49d/CD29) and alpha 5 beta 1 (VLA-5, CD49e/CD29) integrins mediate beta 2 (CD11/CD18) integrin-independent neutrophil recruitment to endotoxin-induced lung inflammation. J Immunol 166,4644-4649. Butcher, E . C , and Picker, L . J. (1996). Lymphocyte homing and homeostasis. Science 272,60-66. Butcher, E . C , Williams, M. , Youngman, K., Rott, L . , and Briskin, M . (1999). Lymphocyte trafficking and regional immunity. Adv Immunol 72, 209-253. Campbell, J. J. , and Butcher, E. C. (2000). Chemokines in tissue-specific and microenvironment-specific lymphocyte homing. Curr Opin Immunol 12,336-341. Campbell, J. J. , Pan, J., and Butcher, E. C. (1999). Cutting edge: developmental switches in chemokine responses during T cell maturation. J Immunol 163, 2353-2357. Capone, M. , Hockett, R. D., Jr., and Zlotnik, A. (1998). Kinetics of T cell receptor beta, gamma, and delta rearrangements during adult thymic development: T cell receptor rearrangements are present in CD44(+)CD25(+) Pro-T thymocytes. Proc Natl Acad Sci U S A 95,12522-12527. Carlow, D. A., Ardman, B., and Ziltener, H. J. (1999). A novel CD8 T cell-restricted CD45RB epitope shared by CD43 is differentially affected by glycosylation. J Immunol 163,1441-1448. Carlow, D. A., Corbel, S. Y., Williams, M . J., and Ziltener, H. J. (2001a). IL-2, -4, and -15 differentially regulate O-glycan branching and P-selectin ligand formation in activated CD8 T cells. J Immunol767,6841-6848. Carlow, D. A., Corbel, S. Y., and Ziltener, H. J. (2001b). Absence of CD43 fails to alter T cell development and responsiveness. J Immunol 166,256-261. Carlow, D. A., Williams, M . J. , and Ziltener, H. J. (2005). P-selectin ligand in activated CD8 T cells proceeds in absence of IL2 and IL12Rp40 in vivo. Journal of Immunology 174, in press. 193 Carlsson, S. R., and Fukuda, M . (1986). Isolation and characterization of leukosialin, a major sialoglycoprotein on human leukocytes. Journal of Biological Chemistry 261,12779-12786. Ceredig, R., and Rolink, T. (2002). A positive look at double-negative thymocytes. Nat Rev Immunol 2,888-897. Cerwenka, A., Morgan, T. M . , Harmsen, A. G., and Dutton, R. W. (1999). Migration kinetics and final destination of type 1 and type 2 CD8 effector cells predict protection against pulmonary virus infection. J Exp Med 189,423-434. Champion, S., Imhof, B. A., Savagner, P., and Thiery, J. P. (1986). The embryonic thymus produces chemotactic peptides involved in the homing of hemopoietic precursors. Cell 44,781-790. Chen, J., Moloney, D. J., and Stanley, P. (2001). Fringe modulation of Jaggedl-induced Notch signaling requires the action of beta 4galactosyltransferase-l. Proc Natl Acad Sci U S A 98,13716-13721. Cheng, J., Baumhueter, S., Cacalano, G., Carver-Moore, K., Thibodeaux, H., Thomas, R., Broxmeyer, H . E. , Cooper, S., Hague, N., Moore, M. , and Lasky, L . A. (1996). Hematopoietic defects in mice lacking the sialomucin CD34. Blood 87,479-490. Chiba, A., Matsumura, K., Yamada, H., Inazu, T., Shimizu, T., Kusunoki, S., Kanazawa, I., Kobata, A., and Endo, T. (1997). Structures of sialylated O-linked oligosaccharides of bovine peripheral nerve alpha-dystroglycan. The role of a novel O-mannosyl-type oligosaccharide in the binding of alpha-dystroglycan with laminin. J Biol Chem 272,2156-2162. Chibber, R., Ben-Mahmud, B. M. , Coppini, D., Christ, E. , and Kohner, E . M . (2000). Activity of the glycosylating enzyme, core 2 GlcNAc (beta 1,6) transferase, is higher in polymorphonuclear leukocytes from diabetic patients compared with age-matched control subjects: relevance to capillary occlusion in diabetic retinopathy. Diabetes 49,1724-1730. Chibber, R., Ben-Mahmud, B. M. , Mann, G. E. , Zhang, J. J. , and Kohner, E . M . (2003). Protein Kinase C {beta}2-Dependent Phosphorylation of Core 2 GlcNAc-T Promotes Leukocyte-Endothelial Cell Adhesion: A Mechanism Underlying Capillary Occlusion in Diabetic Retinopathy. Diabetes 52,1519-1527. Choi, K. H., Osorio, F. A., and Cheng, P. W. (2004). Mucin biosynthesis: bovine C2GnT-M gene, tissue-specific expression, and herpes virus-4 homologue. Am J Respir Cell Mol Biol 30, 710-719. Chou, T. Y., and Hart, G. W. (2001). O-linked N-acetylgtucosamine and cancer: messages from the glycosylation of c-Myc. Adv Exp Med Biol 491,413-418. Chou, T. Y., Hart, G. W., and Dang, C. V. (1995). c-Myc is glycosylated at threonine 58, a known phosphorylation site and a mutational hot spot in lymphomas. J Biol Chem 270,18961-18965. 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. Cinamon, G., Grabovsky, V., Winter, E. , Franitza, S., Feigelson, S., Shamri, R., Dwir, O., and Alon, R. (2001). Novel chemokine functions in lymphocyte migration through vascular endothelium under shear flow. J Leukoc Biol 69,860-866. Clark, R. A., Fuhlbrigge, R. C , and Springer, T. A. (1998). L-Selectin ligands that are O-glycoprotease resistant and distinct from MECA-79 antigen are sufficient for tethering and rolling of lymphocytes on human high endothelial venules. J Cell Biol 140, 721-731. Clegg, C. H., Rulffes, J. T., Wallace, P. M. , and Haugen, H. S. (1996). Regulation of an extrathymic T-cell development pathway by oncostatin M . Nature 384,261-263. 194 Colley, K. J. (1997). Golgi localization of glycosyltransferases: more questions than answers. Glycobiology 7, 1-13. Comer, F. L, and Hart, G. W. (2001). Reciprocity between O-GlcNAc and O-phosphate on the carboxyl terminal domain of RNA polymerase II. Biochemistry 40, 7845-7852. Connolly, E . S., Jr., Winfree, C. J. , Prestigiacomo, C. J., Kim, S. C , Choudhri, T. F., Hoh, B. L. , Naka, Y., Solomon, R. A., and Pinsky, D. J. (1997). Exacerbation of cerebral injury in mice that express the P-selectin gene: identification of P-selectin blockade as a new target for the treatment of stroke. Circ Res 81,304-310. Coutinho, P. M . , Deleury, E. , Davies, G. J. , and Henrissat, B. (2003). An evolving hierarchical family classification for glycosyltransferases. J Mol Biol 328,307-317. Croce, K., Freedman, S. J., Furie, B. C , and Furie, B. (1998). Interaction between soluble P-selectin and soluble P-selectin glycoprotein ligand 1: equilibrium binding analysis. Biochemistry 37,16472-16480. Crutchfield, K. L. , Shinde Patil, V. R., Campbell, C. J., Parkos, C. A., Allport, J. R., and Goetz, D. J. (2000). CDllb/CD18-coated microspheres attach to E-selectin under flow. J Leukoc Biol 67,196-205. Cyster, J . , Somoza, C , Killeen, N., and Williams, A. F. (1990). Protein sequence and gene structure for mouse leukosialin (CD43), a T lymphocyte mucin without introns in the coding sequence. European Journal of Immunology 20,875-881. Cyster, J. G. (1999). Chemokines and the homing of dendritic cells to the T cell areas of lymphoid organs. J Exp Med 189,447-450. Dalziel, M. , Whitehouse, C , McFarlane, I., Brockhausen, I., Gschmeissner, S., Schwientek, T., Clausen, H., Burchell, J. M. , and Taylor-Papadimitriou, J. (2001). The relative activities of the C2GnTl and ST3Gal-I glycosyltransferases determine O-glycan structure and expression of a tumor-associated epitope on MUC1. J Biol Chem 276,11007-11015. Datti, A., and Dennis, J. W. (1993). Regulation of UDP-GlcNAc:Gal beta l-3GalNAc-R beta 1-6-N-acetylglucosaminyltransferase (GlcNAc to GalNAc) in Chinese hamster ovary cells. Journal of Biological Chemistry 268,5409-5416. Delcommenne, M. , Kannagi, R., and Johnson, P. (2002). TNF-alpha increases the carbohydrate sulfation of CD44: induction of 6-sulfo N-acetyl lactosamine on N- and O-linked glycans. Glycobiology 12, 613-622. Delon, J., Kaibuchi, K., and Germain, R. N. (2001). Exclusion of CD43 from the Immunological Synapse Is Mediated by Phosphorylation-Regulated Relocation of the Cytoskeletal Adaptor Moesin. Immunity 15, 691-701. Dennis, J. W., Granovsky, M. , and Warren, C. E . (1999). Glycoprotein glycosylation and cancer progression. Biochim Biophys Acta 1473,21-34. Diacovo, T. G., Catalina, M . D., Siegelman, M . H., and von Andrian, U. H. (1998). Circulating activated platelets reconstitute lymphocyte homing and immunity in L-selectin-deficient mice. J Exp Med 187,197-204. Diacovo, T. G., Puri, K. D., Warnock, R. A., Springer, T. A., and von Andrian, U. H. (1996). Platelet-mediated lymphocyte delivery to high endothelial venules. Science 273,252-255. Dimitroff, C. J., Lee, J. Y., Fuhlbrigge, R. C , and Sackstein, R. (2000). A distinct glycoform of CD44 is an L -seiectin ligand on human hematopoietic cells. Proc Natl Acad Sci U S A 97,13841-13846. Dimitroff, C. J., Lee, J. Y., Rafii, S., Fuhlbrigge, R. C , and Sackstein, R. (2001a). CD44 is a major E-selectin ligand on human hematopoietic progenitor cells. J Cell Biol 153,1277-1286. 195 Dimitroff, C. J., Lee, J. Y., Schor, K. S., Sandmaier, B. M. , and Sackstein, R. (2001b). differential L-selectin binding activities of human hematopoietic cell L-selectin ligands, H C E L L and PSGL-1. J Biol Chem 276, 47623-47631. Dong, Z. M. , Chapman, S. M. , Brown, A. A., Frenette, P. S., Hynes, R. O., and Wagner, D. D. (1998). The combined role of P- and E-selectins in atherosclerosis. J Clin Invest 102,145-152. Donskoy, E. , and Goldschneider, I. (1992). Thymocytopoiesis is maintained by blood-borne precursors throughout postnatal life. A study in parabiotic mice. J Immunol 148,1604-1612. Doyonnas, R., Kershaw, D. B., Duhme, C , Merkens, H., Chelliah, S., Graf, T., and McNagny, K. M . (2001). Anuria, omphalocele, and perinatal lethality in mice lacking the CD34-related protein podocalyxin. J Exp Med 194,13-27. Doyonnas, R., Nielsen, J. S., Chelliah, S., Drew, E. , Hara, T., Miyajima, A., and McNagny, K. M . (2005). Podocalyxin is a CD34-related marker of murine hematopoietic stem cells and embryonic erythroid cells. Blood. Dragone, L . L. , Barth, R. K., Sitar, K. L. , Disbrow, G. L . , and Frelinger, J. G. (1995). Disregulation of leukosialin (CD43, Ly48, sialophorin) expression in the B-cell lineage of transgenic mice increases splenic B-cell number and survival. Proc Natl Acad Sci U S A 92, 626-630. Drew, E. , Merkens, H., Chelliah, S., Doyonnas, R., and McNagny, K. M . (2002). CD34 is a specific marker of mature murine mast ceUs. Exp Hematol 30,1211. Drew, E. , Merzaban, J. S., Seo, W., Ziltener, H. J. , and McNagny, K. M . (2005). CD34 and CD43 inhibit mast cell adhesion and are required for optimal mast cell reconstitution. Immunity 22,43-57. Dudda, J. C , Lembo, A., Bachtanian, E. , Huehn, J., Siewert, C , Hamann, A., Kremmer, E. , Forster, R., and Martin, S. F. (2005). Dendritic cells govern induction and reprogramming of polarized tissue-selective homing receptor patterns of T cells: important roles for soluble factors and tissue microenvironments. Eur J Immunol 35,1056-1065. Duijvestijn, A. M. , Horst, E. , Pals, S. T., Rouse, B. N., Steere, A. C , Picker, L . J., Meijer, C. J., and Butcher, E . C. (1988). High endothelial differentiation in human lymphoid and inflammatory tissues defined by monoclonal antibody HECA-452. Am J Pathol 130,147-155. Ehrhardt, C , Kneuer, C , and Bakowsky, U. (2004). Selectins-an emerging target for drug delivery. Adv Drug Debv Rev 56, 527-549. El Kassar, N., Lucas, P. J. , Klug, D. B., Zamisch, M . , Merchant, M. , Bare, C. V., Choudhury, B., Sharrow, S. O., Richie, E. , Mackall, C. L . , and Gress, R. E . (2004). A dose effect of IL-7 on thymocyte development. Blood 104,1419-1427. Ellies, L . G., Jones, A. T., Williams, M . J., and Ziltener, H. J. (1994). Differential regulation of CD43 glycoforms on CD4+ and CD8+ T lymphocytes in graft-versus-host disease. Glycobiology 4,885-893. Ellies, L . G., Sperandio, M. , Under hi 11, G. H., Yousif, J. , Smith, M. , Priatel, J. J. , Kansas, G. S., Ley, K., and Marth, J. D. (2002). Sialyltransferase specificity in selectin Ugand formation. Blood 100,3618-3625. Ellies, L . G., Tao, W., Feliinger, W., Teh, H. S., and Ziltener, H. J. (1996). The CD43 130-kD peripheral T-cell activation antigen is downregulated in thymic positive selection. Blood 88,1725-1732. Ellies, L . G., Tsuboi, S., Petryniak, B., Lowe, J. B., Fukuda, M. , and Marth, J. D. (1998). Core 2 oligosaccharide biosynthesis distinguishes between selectin ligands essential for leukocyte homing and inflammation. Immunity 9,881-890. 196 Engelhardt, B., Vestweber, D., Hallmann, R., and Schulz, M . (1997). E - and P-selectin are not involved in the recruitment of inflammatory cells across the blood-brain barrier in experimental autoimmune encephalomyelitis. Blood 90,4459-4472. Eriksson, E . E. , Xie, X., Werr, J. , Thoren, P., and Lindbom, L. (2001). Importance of primary capture and L -selectin-dependent secondary capture in leukocyte accumulation in inflammation and atherosclerosis in vivo. J Exp Med 194,205-218. Falkenberg, V. R., Alvarez, K., Roman, C , and Fregien, N. (2003). Multiple transcription initiation and alternative splicing in the 5' untranslated region of the Core2 {beta}l-6 N-acetylglucosaminyltransferase I gene. Glycobiology. Felli, M . P., Maroder, M. , Mitsiadis, T. A., Campese, A. F., Bellavia, D., Vacca, A., Mann, R. S., Frati, L . , Lendahl, U., Gulino, A., and Screpanti, I. (1999). Expression pattern of notchl, 2 and 3 and Jaggedl and 2 in lymphoid and stromal thymus components: distinct ligand-receptor interactions in intrathymic T cell development. Int Immunol 11,1017-1025. Fellinger, W. J., Barran, P., Merkens, H., Corbel, S. Y., and Ziltener, H. J. (1998). In vivo overexpression of Core2 N-acetylglucosaminyltransferase prevents repopulation of the bone marrow with colony forming cells but fails to affect normal T cell development. J Cell Physiol 176,350-358. Fieger, C. B., Sassetti, C. M. , and Rosen, S. D. (2003). Endoglycan, a member of the CD34 family, functions as an L-selectin ligand through modification with tyrosine sulfation and sialyl Lewis x. J Biol Chem 278,27390-27398. Forlow, S. B., and Ley, K. (2001). Selectin-independent leukocyte rolling and adhesion in mice deficient in E- , P-, and L-selectin and ICAM-1. Am J Physiol Heart Circ Physiol 280, H634-641. Foss, D. L . , Donskoy, E. , and Goldschneider, I. (2001). The importation of hematogenous precursors by the thymus is a gated phenomenon in normal adult mice. J Exp Med 193,365-374. Fox, R. I., Hueniken, M. , Fong, S., Behar, S., Royston, I., Singhal, S. K., and Thompson, L . (1983). A novel cell surface antigen (T305) found in increased frequency on acute leukemia cells and in autoimmune disease states. Journal of Immunology 131, 762-767. Fratazzi, C , Manjunath, N., Arbeit, R. D., Carini, C , Gerken, T. A., Ardman, B., Remold-O'Donnell, E. , and Remold, H. G. (2000). A macrophage invasion mechanism for mycobacteria implicating the extracellular domain of CD43. J Exp Med 192,183-192. Frenette, P. S., Mayadas, T. N., Rayburn, H., Hynes, R. O., and Wagner, D. D. (1996). Susceptibility to infection and altered hematopoiesis in mice deficient in both P- and E-selectins. Cell 84,563-574. Frenette, P. S., Subbarao, S., Mazo, I. B., von Andrian, U. H., and Wagner, D. D. (1998). Endothelial selectins and vascular cell adhesion molecule-1 promote hematopoietic progenitor homing to bone marrow. Proc Natl Acad Sci U S A 95,14423-14428. Fuhlbrigge, R. C., Kieffer, J. D., Armerding, D., and Kupper, T. S. (1997). Cutaneous lymphocyte antigen is a specialized form of PSGL-1 expressed on skin-homing T cells. Nature 389,978-981. Fuhlbrigge, R. C., King, S. L. , Dimitroff, C. J., Kupper, T. S., and Sackstein, R. (2002). Direct real-time observation of E - and P-selectin-mediated rolling on cutaneous lymphocyte-associated antigen immobilized on Western blots. J Immunol 168,5645-5651. Fukuda, M . (1991). Leukosialin, a major O-glycan-containing sialoglycoprotein defining leukocyte differentiation and malignancy. Glycobiology 1,347-356. Fukuda, M . (1994). Cell surface carbohydrates: cell-type specific expression (Oxford, Oxford University Press). 197 Fukuda, M. , Bothner, B., Ramsamooj, P., Dell, A., Tiller, P. R., Varki, A., and Klock, J. C. (1985). Structures of sialylated fucosyl polylactosaminoglycans isolated from chronic myelogenous leukemia cells. J Biol Chem 260,12957-12967. Fukuda, M. , Hiraoka, N., Akama, T. O., and Fukuda, M . N. (2001). Carbohydrate-modifying sulfotransferases: structure, function, and pathophysiology. J Biol Chem 276,47747-47750. Fukuda, M. , and Tsuboi, S. (1999). Mucin-type O-glycans and leukosialin. Biochim Biophys Acta 1455,205-217. Fukuda, M. , and Vliegenthart, J. F. (1999). Structure and biosynthesis of glycoproteins. Preface. Biochim Biophys Acta 1473,1-3. Gallatin, W. M. , Weissman, I. L. , and Butcher, E. C. (1983). A cell-surface molecule involved in organ-specific homing of lymphocytes. Nature 304,30-34. Gauguet, J. M. , Rosen, S. D., Marth, J. D., and von Andrian, U. H. (2004). Core 2 branching betal,6-N-acetylglucosaminyltransferase and high endothelial cell N-acetylglucosamine-6-sulfotransferase exert differential control over B- and T-lymphocyte homing to peripheral lymph nodes. Blood 104,4104-4112. Gendler, S. J., and Spicer, A. P. (1995). Epithelial mucin genes. Annual Review of Physiology 57,607-634. Gerken, T. A., Gilmore, M . , and Zhang, J. (2002a). Determination of the site-specific oligosaccharide distribution of the O-glycans attached to the porcine submaxillary mucin tandem repeat. Further evidence for the modulation of O-glycans side chain structures by peptide sequence. J Biol Chem 277,7736-7751. Gerken, T. A., Zhang, J., Levine, J., and Elhammer, A. (2002b). Mucin core O-glycosylation is modulated by neighboring residue glycosylation status. Kinetic modeling of the site-specific glycosylation of the apo-porcine submaxillary mucin tandem repeat by UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferases T l and T2. J Biol Chem 277,49850-49862. Godfrey, D. I., Kennedy, J., Suda, T., and Zlotnik, A. (1993). A developmental pathway involving four phenotypically and functionally distinct subsets of CD3-CD4-CD8- triple-negative adult mouse thymocytes defined by CD44 and CD25 expression. J Immunol 150,4244-4252. Godfrey, D. I., and Zlotnik, A. (1993). Control points in early T-celi development. Immunol Today 14, 547-553. Goetz, D. J., Greif, D. M. , Ding, H., Camphausen, R. T., Howes, S., Comess, K. M . , Snapp, K. R., Kansas, G. S., and Luscinskas, F. W. (1997). Isolated P-selectin glycoprotein ligand-1 dynamic adhesion to P- and E -selectin. J Cell Biol 137, 509-519. Goldschneider, I., Komschlies, K. L . , and Greiner, D. L . (1986). Studies of thymocytopoiesis in rats and mice. I. Kinetics of appearance of thymocytes using a direct intrathymic adoptive transfer assay for thymocyte precursors. J Exp Med 163,1-17. Grabovsky, V., Feigelson, S., Chen, C , Bleijs, D. A., Peled, A., Cinamon, G., Baleux, F., Arenzana-Seisdedos, F., Lapidot, T., van Kooyk, Y., et al. (2000). Subsecond induction of alpha4 integrin clustering by immobilized chemokines stimulates leukocyte tethering and rolling on endothelial vascular cell adhesion molecule 1 under flow conditions. J Exp Med 192,495-506. Gratama, J. W., Sutherland, D. R., and Keeney, M . (2001). Flow cytometric enumeration and immunophenotyping of hematopoietic stem and progenitor cells. Semin Hematol 38,139-147. Green, C. E. , Pearson, D. N., Camphausen, R. T., Staunton, D. E. , and Simon, S. I. (2004). Shear-dependent capping of L-selectin and P-selectin glycoprotein ligand 1 by E-selectin signals activation of high-avidity beta2-integrin on neutrophils. J Immunol 172,7780-7790. 198 Greenberg, A. W., Kerr, W. G., and Hammer, D. A. (2000). Relationship between selectin-mediated rolling of hematopoietic stem and progenitor cells and progression in hematopoietic development. Blood 95,478-486. Guy-Grand, D., Azogui, O., Celli, S., Darche, S., Nussenzweig, M . C., Kourilsky, P., and Vassalli, P. (2003). Extrathymic T cell lymphopoiesis: ontogeny and contribution to gut intraepithelial lymphocytes in athymic and euthymic mice. J Exp Med 197,333-341. Guyer, D. A., Moore, K. L. , Lynam, E . B., Schammel, C. M. , Rogelj, S., McEver, R. P., and Sklar, L . A. (1996). P-selectin glycoprotein ligand-1 (PSGL-1) is a ligand for L-selectin in neutrophil aggregation. Blood 88,2415-2421. Haines, N., and Irvine, K. D. (2003). Glycosylation regulates Notch signalling. Nat Rev Mol Cell Biol 4, 786-797. Hanahan, D. (1998). Peripheral-antigen-expressing cells in thymic medulla: factors in self-tolerance and autoimmunity. Curr Opin Immunol 10,656-662. Hanisch, F. G., and Muller, S. (2000). MUC1: the polymorphic appearance of a human mucin. Glycobiology 10,439-449. Hanisch, F. G., Muller, S., Hassan, H., Clausen, H., Zachara, N., Gooley, A. A., Paulsen, H., Alving, K., and Peter-Katalinic, J. (1999). Dynamic epigenetic regulation of initial O-glycosylation by UDP-N-Acetylgalactosamine:Peptide N-acetylgalactosaminyltransferases. site-specific glycosylation of MUC1 repeat peptide influences the substrate qualities at adjacent or distant Ser/Thr positions. J Biol Chem 274,9946-9954. Hanisch, F. G., Reis, C. A., Clausen, H., and Paulsen, H. (2001). Evidence for glycosylation-dependent activities of polypeptide N-acetylgalactosaminyltransferases rGalNAc-T2 and -T4 on mucin glycopeptides. Glycobiology 11,731-740. Hara-Kuge, S., Seko, A., Shimada, O., Tosaka-Shimada, H., and Yamashita, K. (2004). The binding of VIP36 and alpha-amylase in the secretory vesicles via high-mannose type glycans. Glycobiology 14,739-744. Hare, K. J. , Jenkinson, E . J. , and Anderson, G. (1999). CD69 expression discriminates MHC-dependent and -independent stages of thymocyte positive selection. J Immunol 162,3978-3983. Hare, K. J. , Jenkinson, E . J. , and Anderson, G. (2000). An essential role for the IL-7 receptor during intrathymic expansion of the positively selected neonatal T cell repertoire. J Immunol 165,2410-2414. Harman, B. C , Jenkinson, E. J., and Anderson, G. (2003). Entry into the thymic microenvironment triggers Notch activation in the earliest migrant T cell progenitors. J Immunol / 70,1299-1303. Harrington, L . E. , Galvan, M. , Baum, L . G., Altman, J. D., and Ahmed, R. (2000). Differentiating between memory and effector CD8 T cells by altered expression of cell surface O-glycans. J Exp Med 191,1241-1246. Harris, E. S., Mclntyre, T. M. , Prescott, S. M. , and Zimmerman, G. A. (2000). The leukocyte integrins. J Biol Chem 275, 23409-23412. Harris, R. J., and Spellman, M . W. (1993). O-linked fucose and other post-translational modifications unique to EGF modules. Glycobiology 3,219-224. Harrison, D. E., and Astle, C. M . (1997). Short- and long-term multilineage repopulating hematopoietic stem cells in late fetal and newborn mice: models for human umbilical cord blood. Blood 90,174-181. Hart, G. W. (1997). Dynamic O-linked glycosylation of nuclear and cytoskeletal proteins. Annu Rev Biochem 66,315-335. 199 Hart, G. W., Kreppel, L . K., Comer, F. I., Arnold, C. S., Snow, D. M. , Ye, Z., Cheng, X., DellaManna, D., Caine, D. S., Earles, B. J., et al. (1996). O-GlcNAcylation of key nuclear and cytoskeletal proteins: reciprocity with O-phosphorylation and putative roles in protein multimerization. Glycobiology 6, 711-716. Hattori, R., Hamilton, K. K., Fugate, R. D., McEver, R. P., and Sims, P. J. (1989). Stimulated secretion of endothelial von Willebrand factor is accompanied by rapid redistribution to the cell surface of the intracellular granule membrane protein GMP-140. J Biol Chem 264,7768-7771. Hauri, H., Appenzeller, C , Kuhn, F., and Nufer, O. (2000a). Lectins and traffic in the secretory pathway. FEBS Lett 476,32-37. Hauri, H. P., Kappeler, F., Andersson, H., and Appenzeller, C. (2000b). ERGIC-53 and traffic in the secretory pathway. J Cell Sci 113 (Pt 4), 587-596. Helenius, A., and Aebi, M . (2001). Intracellular functions of N-linked glycans. Science 291,2364-2369. Hemmerich, S., Bertozzi, C. R., Leffler, H., and Rosen, S. D. (1994a). Identification of the sulfated monosaccharides of GlyCAM-1, an endothelial-derived Ugand for L-selectin. Biochemistry 33,4820-4829. Hemmerich, S., Bistrup, A., Singer, M . S., van Zante, A., Lee, J. K., Tsay, D., Peters, M. , Carminati, J. L . , Brennan, T. J. , Carver-Moore, K., et al. (2001). Sulfation of L-selectin ligands by an HEV-restricted sulfotransferase regulates lymphocyte homing to lymph nodes. Immunity 75,237-247. Hemmerich, S., Butcher, E. C , and Rosen, S. D. (1994b). Sulfation-dependent recognition of high endothelial venules (HEV)-ligands by L-selectin and M E C A 79, and adhesion-blocking monoclonal antibody. J Exp Med 180,2219-2226. Hemmerich, S., and Rosen, S. D. (2000). Carbohydrate sulfotransferases in lymphocyte homing. Glycobiology 10,849-856. Hickey, M . J., Granger, D. N., and Kubes, P. (1999). Molecular mechanisms underlying IL-4-induced leukocyte recruitment in vivo: a critical role for the alpha 4 integrin. J Immunol 163,3441-3448. Hicks, C , Johnston, S. H., diSibio, G., Collazo, A., Vogt, T. F., and Weinmaster, G. (2000). Fringe differentially modulates Jaggedl and Deltal signalling through Notchl and Notch2. Nat Cell Biol 2,515-520. Hidalgo, A., Weiss, L . A., and Frenette, P. S. (2002). Functional selectin ligands mediating human CD34(+) cell interactions with bone marrow endothelium are enhanced postnatally. J Clin Invest 110, 559-569. Higgins, E. A., Siminovitch, K. A., Zhuang, D. L. , Brockhausen, I., and Dennis, J. W. (1991). Aberrant O-linked oligosaccharide biosynthesis in lymphocytes and platelets from patients with the Wiskott-Aldrich syndrome. J Biol Chem 266,6280-6290. Hiraoka, N., Kawashima, H., Petryniak, B., Nakayama, J., Mitoma, J., Marth, J. D., Lowe, J. B., and Fukuda, M . (2004). Core 2 branching betal,6-N-acetylglucosaminyItransferase and high endothelial venule-restricted sulfotransferase collaboratively control lymphocyte homing. J Biol Chem 279,3058-3067. Hiraoka, N., Petryniak, B., Nakayama, J., Tsuboi, S., Suzuki, M. , Yeh, J. C , Izawa, D., Tanaka, T., Miyasaka, M. , Lowe, J. B., and Fukuda, M . (1999). A novel, high endothelial venule-specific sulfotransferase expresses 6-sulfo sialyl Lewis(x), an L-selectin ligand displayed by CD34. Immunity 11, 79-89. Hirata, T., Furie, B. C , and Furie, B. (2002). P-, E- , and L-Selectin Mediate Migration of Activated CD8+ T Lymphocytes into Inflamed Skin. J Immunol 169,4307-4313. Hirata, T., Merrill-Skoloff, G., Aab, M. , Yang, J. , Furie, B. C , and Furie, B. (2000). P-Selectin glycoprotein ligand 1 (PSGL-1) is a physiological ligand for E-selectin in mediating T helper 1 lymphocyte migration. J Exp Med 192,1669-1676. 200 Homeister, J. W., Thall, A. D., Petryniak, B., Maly, P., Rogers, C. E. , Smith, P. L . , Kelly, R. J., Gersten, K. M. , Askari, S. W., Cheng, G., et al. (2001). The alpha(l,3)fucosyltransferases FucT-IV and FucT-VH exert collaborative control over selectin-dependent leukocyte recruitment and lymphocyte homing. Immunity 75, 115-126. Hooper, L . V., and Gordon, J. I. (2001). Glycans as legislators of host-microbial interactions: spanning the spectrum from symbiosis to pathogenicity. Glycobiology / / , 1R-10R. Huang, M . C , Zollner, O., Moll, T., Maly, P., Thall, A. D., Lowe, J. B., and Vestweber, D. (2000). P-selectin glycoprotein ligand-1 and E-selectin ligand-1 are differentially modified by fucosyltransferases Fuc-TIV and Fuc-TVII in mouse neutrophils. J Biol Chem 275,31353-31360. Huehn, J., Siegmund, K., Lehmann, J. C , Siewert, C , Haubold, U., Feuerer, M . , Debes, G. F., Lauber, J., Frey, O., Przybylski, G. K., et al. (2004). Developmental stage, phenotype, and migration distinguish naive-and effector/memory-like CD4+ regulatory T cells. J Exp Med 199,303-313. Ikuta, K., and Weissman, I. L . (1992). Evidence that hematopoietic stem cells express mouse c-kit but do not depend on steel factor for their generation. Proc Natl Acad Sci U S A 89,1502-1506. Imai, Y., Singer, M . S., Fennie, C , Lasky, L . A., and Rosen, S. D. (1991). Identification of a carbohydrate-based endothelial ligand for a lymphocyte homing receptor. J Cell Biol 113,1213-1221. Ismaili, J., Antica, M. , and Wu, L . (1996). CD4 and CD8 expression and T cell antigen receptor gene rearrangement in early intrathymic precursor cells. Eur J Immunol 26, 731-737. Issekutz, A. C , and Issekutz, T. B. (2002). The role of E-selectin, P-selectin, and very late activation antigen-4 in T lymphocyte migration to dermal inflammation. J Immunol 168,1934-1939. Issekutz, T. B., Miyasaka, M. , and Issekutz, A. C. (1996). Rat blood neutrophils express very late antigen 4 and it mediates migration to arthritic joint and dermal inflammation. J Exp Med 183,2175-2184. Iwai, T., Inaba, N., Naundorf, A., Zhang, Y., Gotoh, M. , Iwasaki, H., Kudo, T., Togayachi, A., Ishizuka, Y., Nakanishi, H., and Narimatsu, H. (2002). Molecular cloning and characterization of a novel UDP-GlcNAc:GalNAc-peptide betal,3-N-acetylglucosaminyltransferase (beta 3Gn-T6), an enzyme synthesizing the core 3 structure of O-glycans. J Biol Chem 277,12802-12809. Jacobsen, K., Kravitz, J., Kincade, P. W., and Osmond, D. G. (1996). Adhesion receptors on bone marrow stromal cells: in vivo expression of vascular cell adhesion molecule-1 by reticular cells and sinusoidal endothelium in normal and gamma-irradiated mice. Blood 87, 73-82. Johnson, P., Maiti, A., Brown, K. L . , and Li , R. (2000). A role for the cell adhesion molecule CD44 and sulfation in leukocyte-endothelial cell adhesion during an inflammatory response? Biochem Pharmacol 59, 455-465. Johnson, R. C., Chapman, S. M. , Dong, Z. M., Ordovas, J. M. , Mayadas, T. N., Herz, J., Hynes, R. O., Schaefer, E. J., and Wagner, D. D. (1997). Absence of P-selectin delays fatty streak formation in mice. J Clin Invest 99,1037-1043. Johnston, B., Issekutz, T. B., and Kubes, P. (1996). The alpha 4-integrin supports leukocyte rolling and adhesion in chronically inflamed postcapillary venules in vivo. J Exp Med 183,1995-2006. Johnston, B., Walter, U. M. , Issekutz, A. C , Issekutz, T. B., Anderson, D. C., and Kubes, P. (1997). Differential roles of selectins and the alpha4-integrin in acute, subacute, and chronic leukocyte recruitment in vivo. J Immunol 159,4514-4523. Jones, A. T., Federsppiel, B., Ellies, L . G., Williams, M . J. , Burgener, R., Duronio, V., Smith, C. A., Takei, F., and Ziltener, H. J. (1994). Characterization of the activation-associated isoform of CD43 on murine T lymphocytes. J Immunol 153,3426-3439. 201 Jung, U., and Ley, K. (1999). Mice lacking two or all three selectins demonstrate overlapping and distinct functions for each selectin. J Immunol 762, 6755-6762. Jutila, M . A., Watts, G., Walcheck, B., and Kansas, G. S. (1992). Characterization of a functionally important and evolutionarily well-conserved epitope mapped to the short consensus repeats of E-selectin and L-selectin. J Exp Med 775,1565-1573. Kadesch, T. (2004). Notch signaling: the demise of elegant simplicity. Curr Opin Genet Dev 14,506-512. Kamemura, K., Hayes, B. K., Comer, F. I., and Hart, G. W. (2002). Dynamic interplay between O-glycosylation and O-phosphorylation of nucleocytoplasmic proteins: alternative glycosylation/phosphorylation of THR-58, a known mutational hot spot of c-Myc in lymphomas, is regulated by mitogens. J Biol Chem 277,19229-19235. Kansas, G. S. (1996). Selectins and their ligands: current concepts and controversies. Blood 88,3259-3287. Katayama, Y., Hidalgo, A., Chang, J., Peired, A., and Frenette, P. S. (2005). CD44 is a physiological E-selectin ligand on neutrophils. J Exp Med 201,1183-1189. Katayama, Y., Hidalgo, A., Furie, B. C., Vestweber, D., Furie, B., and Frenette, P. S. (2003). PSGL-1 participates in E-selectin-mediated progenitor homing to bone marrow: evidence for cooperation between E -selectin ligands and alpha4 integrin. Blood 102, 2060-2067. Kawakami, N., Nishizawa, F., Sakane, N., Iwao, M . , Tsujikawa, K., Ikawa, M. , Okabe, M . , and Yamamoto, H. (1999). Roles of integrins and CD44 on the adhesion and migration of fetal liver cells to the fetal thymus. J Immunol 163,3211-3216. Kawakubo, M. , Ito, Y., Okimura, Y., Kobayashi, M. , Sakura, K., Kasama, S., Fukuda, M . N., Fukuda, M. , Katsuyama, T., and Nakayama, J. (2004). Natural antibiotic function of a human gastric mucin against Helicobacter pylori infection. Science 305,1003-1006. Kawamoto, H., Ohmura, K., Fujimoto, S., and Katsura, Y. (1999). Emergence of T cell progenitors without B cell or myeloid differentiation potential at the earliest stage of hematopoiesis in the murine fetal liver. J Immunol 162, 2725-2731. Keelan, E. T., Licence, S. T., Peters, A. M. , Binns, R. M. , and Haskard, D. O. (1994). Characterization of E -selectin expression in vivo with use of a radiolabeled monoclonal antibody. Am J Physiol 266, H278-290. Kelly, W. G., Dahmus, M . E. , and Hart, G. W. (1993). RNA polymerase H is a glycoprotein. Modification of the COOH-terminal domain by O-GlcNAc. J Biol Chem 268,10416-10424. Khaldoyanidi, S., Schnabel, D., Fohr, N., and Zoller, M . (1997). Functional activity of CD44 isoforms in haemopoiesis of the rat. Br J Haematol 96,31-45. Khan, A. I., Kerfoot, S. M. , Heit, B., Liu, L. , Andonegui, G., Ruffell, B., Johnson, P., and Kubes, P. (2004). Role of CD44 and hyaluronan in neutrophil recruitment. J Immunol 173, 7594-7601. Kim, C. H., Pelus, L . M. , White, J. R., and Broxmeyer, H. E. (1998). Differential chemotactic behavior of developing T cells in response to thymic chemokines. Blood 91,4434-4443. Kim, M. , Carman, C. V., and Springer, T. A. (2003). Bidirectional transmembrane signaling by cytoplasmic domain separation in integrins. Science 301,1720-1725. Kingsley, P. D., Hagen, K. G., Maltby, K. M. , Zara, J., and Tabak, L . A. (2000). Diverse spatial expression patterns of UDP-GalNAc:polypeptide N-acetylgalactosaminyl-transferase family member mRNAs during mouse development. Glycobiology 10,1317-1323. 202 Kitchen, S. G., and Zack, J. A. (1997). CXCR4 expression during lymphopoiesis: implications for human immunodeficiency virus type 1 infection of the thymus. J Virol 71, 6928-6934. Kivisakk, P., Mahad, D. J. , Callahan, M . K., Trebst, C , Tucky, B., Wei, T., Wu, L. , Baekkevold, E . S., Lassmann, H., Staugaitis, S. M., et al. (2003). Human cerebrospinal fluid central memory CD4+ T cells: evidence for trafficking through choroid plexus and meninges via P-selectin. Proc Natl Acad Sci U S A 100, 8389-8394. Kobzdej, M . M. , Leppanen, A., Ramachandran, V., Cummings, R. D., and McEver, R. P. (2002). Discordant expression of selectin ligands and sialyl Lewis x-related epitopes on murine myeloid cells. Blood 100,4485-4494. Kondo, M. , Weissman, I. L. , and Akashi, K. (1997). Identification of clonogenic common lymphoid progenitors in mouse bone marrow. Cell 91,661-672. Kono, M. , Ohyama, Y., Lee, Y. C , Hamamoto, T., Kojima, N., and Tsuji, S. (1997). Mouse beta-galactoside alpha 2,3-sialyltransferases: comparison of in vitro substrate specificities and tissue specific expression. Glycobiology 7,469-479. Kono, M. , Takashima, S., Liu, H., Inoue, M. , Kojima, N., Lee, Y. C , Hamamoto, T., and Tsuji, S. (1998). Molecular cloning and functional expression of a fifth-type alpha 2,3-sialyltransferase (mST3Gal V: GM3 synthase). Biochem Biophys Res Commun 253,170-175. Korekane, H., Taguchi, T., Sakamoto, Y., Honke, K., Dohmae, N., Salminen, H., Toivonen, S., Helin, J. , Takio, K., Renkonen, O., and Taniguchi, N. (2003). Purification and cDNA cloning of UDP-GlcNAc:GlcNAcbetal-3Galbetal-4Glc(NAc)-R [GlcNAc to Gal]betal,6N-acetylglucosaminyltransferase from rat small intestine: a major carrier of dIGnT activity in rat small intestine. Glycobiology 13,387-400. Krause, D. S., Fackler, M . J., Civin, C. I., and May, W. S. (1996). CD34: structure, biology, and clinical utility. Blood 87,1-13. Kreppel, L . K., Blomberg, M . A., and Hart, G. W. (1997). Dynamic glycosylation of nuclear and cytosolic proteins. Cloning and characterization of a unique O-GlcNAc transferase with multiple tetratricopeptide repeats. J Biol Chem 272,9308-9315. Kubes, P. (2002). Introduction: The complexities of leukocyte recruitment. Seminars in Immunology 14, 65. Kubes, P., Jutila, M. , and Payne, D. (1995). Therapeutic potential of inhibiting leukocyte rolling in ischemia/reperfusion. J Clin Invest 95,2510-2519. Kuhns, W., Rutz, V., Paulsen, H., Matta, K. L. , Baker, M . A., Barner, M. , Granovsky, M . , and Brockhausen, I. (1993). Processing O-glycan core 1, Gal beta l-3GalNAc alpha-R. Specificities of core 2, UDP-GlcNAc: Gal beta 1-3 GalNAc-R(GlcNAc to GalNAc) beta 6-N-acetylglucosaminyltransferase and CMP-sialic acid: Gal beta l-3GalNAc-R alpha 3-sialyltransferase. Glycoconj J10,381-394. Kuijpers, T. W., Hoogerwerf, M. , Kuijpers, K. C , Schwartz, B. R., and Harlan, J. M . (1992). Cross-Unking of sialophorin (CD43) induces neutrophil aggregation in a CD18-dependent and a CD18-independent way. Journal of Immunology 149,998-1003. Kumar, R., Camphausen, R. T., Sullivan, F. X., and Cumming, D. A. (1996). Core2 beta-l,6-N-acetylglucosaminyltransferase enzyme activity is critical for P-selectin glycoprotein ligand-1 binding to P-selectin. Blood 88,3872-3879. Kundra, R., and Kornfeld, S. (1999). Asparagine-iinked oligosaccharides protect Lamp-1 and Lamp-2 from intracellular proteolysis. J Biol Chem 274,31039-31046. Kunkel, E . J. , and Butcher, E. C. (2002). Chemokines and the tissue-specific migration of lymphocytes. Immunity 16,1-4. 203 Kunkel, E. J . , Chomas, J. E. , and Ley, K. (1998). Role of primary and secondary capture for leukocyte accumulation in vivo. Circ Res 82,30-38. Kunkel, E. J. , and Ley, K. (1996). Distinct phenotype of E-selectin-deficient mice. E-selectin is required for slow leukocyte rolling in vivo. Circ Res 79,1196-1204. Labow, M . A., Norton, C. R., Rumberger, J. M. , Lombard-Gillooly, K. M. , Shuster, D. J., Hubbard, J., Bertko, R., Knaack, P. A., Terry, R. W., Harbison, M . L . , and et al. (1994). Characterization of E-selectin-deficient mice: demonstration of overlapping function of the endothelial selectins. Immunity 7,709-720. Laky, K., Lefrancois, L. , von Freeden-Jeffry, U., Murray, R., and Puddington, L . (1998). The role of IL-7 in thymic and extrathymic development of TCR gamma delta cells. J Immunol 161, 707-713. Lasky, L . A. (1995). Selectin-carbohydrate interactions and the initiation of the inflammatory response. Annu Rev Biochem 64,113-139. Lasky, L . A., Singer, M . S., Dowbenko, D., Imai, Y., Henzel, W. J., Grimley, C , Fennie, C , Gillett, N., Watson, S. R., and Rosen, S. D. (1992). An endotheUal ligand for L-selectin is a novel mucin-like molecule. Cell 69,927-938. Laszik, Z., Jansen, P. J., Cummings, R. D., Tedder, T. F., McEver, R. P., and Moore, K. L . (1996). P-selectin glycoprotein ligand-1 is broadly expressed in cells of myeloid, lymphoid, and dendritic lineage and in some nonhematopoietic cells. Blood 88,3010-3021. Lawrence, M . B., and Springer, T. A. (1991). Leukocytes roll on a selectin at physiologic flow rates: distinction from and prerequisite for adhesion through integrins. Cell 65,859-873. Lehmann, J. C , Jablonski-Westrich, D., Haubold, U., Gutierrez-Ramos, J. C , Springer, T., and Hamann, A. (2003). Overlapping and selective roles of endothelial intercellular adhesion molecule-1 (ICAM-1) and ICAM-2 in lymphocyte trafficking. J Immunol 171\ 2588-2593. Lenter, M. , Levinovitz, A., Isenmann, S., and Vestweber, D. (1994). Monospecific and common glycoprotein ligands for E - and P-selectin on myeloid cells. J Cell Biol 725,471-481. Leppanen, A., White, S. P., Helin, J. , McEver, R. P., and Cummings, R. D. (2000). Binding of glycosulfopeptides to P-selectin requires stereospecific contributions of individual tyrosine sulfate and sugar residues. J Biol Chem 275,39569-39578. Leppanen, A., Yago, T., Otto, V. I., McEver, R. P., and Cummings, R. D. (2003). Model glycosulfopeptides from P-selectin glycoprotein ligand-1 require tyrosine sulfation and a core 2-branched O-glycan to bind to L -selectin. J Biol Chem 278,26391-26400. Levesque, J. P., Zannettino, A. C , Pudney, M. , Niutta, S., Haylock, D. N., Snapp, K. R., Kansas, G. S., Berndt, M . C , and Simmons, P. J. (1999). PSGL-l-mediated adhesion of human hematopoietic progenitors to P-selectin results in suppression of hematopoiesis. Immunity 77,369-378. Levinovitz, A., Muhlhoff, J., Isenmann, S., and Vestweber, D. (1993). Identification of a glycoprotein ligand for E-selectin on mouse myeloid cells. J Cell Biol 727,449-459. Ley, K. (2003). The role of selectins in inflammation and disease. Trends in Molecular Medicine 9, 263. Ley, K., Bullard, D. C , Arbones, M . L . , Bosse, R., Vestweber, D., Tedder, T. F., and Beaudet, A. L . (1995). Sequential contribution of L - and P-selectin to leukocyte rolling in vivo. J Exp Med 181,669-675. Ley, K., Gaehtgens, P., Fennie, C , Singer, M . S., Lasky, L . A., and Rosen, S. D. (1991). Lectin-like cell adhesion molecule 1 mediates leukocyte rolling in mesenteric venules in vivo. Blood 77, 2553-2555. 204 Ley, K., and Kansas, G. S. (2004). SELECTINS IN T - C E L L RECRUITMENT TO NON-LYMPHOID TISSUES AND SITES OF INFLAMMATION. Nature Reviews Immunology 4,325. Ley, K., and Tedder, T. F. (1995). Leukocyte interactions with vascular endothelium. New insights into selectin-mediated attachment and rolling. J Immunol 755,525-528. Li , C. L . , and Johnson, G. R. (1995). Murine hematopoietic stem and progenitor cells: I. Enrichment and biologic characterization. Blood 85,1472-1479. Li , C. M. , Adler, K. B., and Cheng, P. W. (1998). Mucin biosynthesis: molecular cloning and expression of bovine lung mucin core 2 N-acetylglucosaminyltransferase cDNA. Am J Respir Cell Mol Biol 18,343-352. Li , F., Wilkins, P. P., Crawley, S., Weinstein, J. , Cummings, R. D., and McEver, R. P. (1996). Post-translational modifications of recombinant P-selectin glycoprotein ligand-1 required for binding to P- and E-selectin. J Biol Chem 271,3255-3264. Liao, F., Ali, J. , Greene, T., and Muller, W. A. (1997). Soluble domain 1 of platelet-endothelial cell adhesion molecule (PECAM) is sufficient to block transendothelial migration in vitro and in vivo. J Exp Med 7*5, 1349-1357. Lim, Y. C , Henault, L. , Wagers, A. J., Kansas, G. S., Luscinskas, F. W., and Lichtman, A. H. (1999). Expression of functional selectin ligands on Th cells is differentially regulated by EL-12 and IL-4. J Immunol 162,3193-3201. Lim, Y. C , Xie, H., Come, C. E. , Alexander, S. I., Grusby, M . J. , Lichtman, A. H., and Luscinskas, F. W. (2001). IL-12, STAT4-dependent up-regulation of CD4(+) T cell core 2 beta-l,6-n-acetylglucosaminyltransferase, an enzyme essential for biosynthesis of P-selectin ligands. J Immunol 767, 4476-4484. Lind, E . F., Prockop, S. E. , Porritt, H. E. , and Petrie, H. T. (2001). Mapping precursor movement through the postnatal thymus reveals specific microenvironments supporting defined stages of early lymphoid development. J Exp Med 194,127-134. Liu, F., Iqbal, K., Grundke-Iqbal, I., Hart, G. W., and Gong, C. X. (2004). O-GlcNAcylation regulates phosphorylation of tau: a mechanism involved in Alzheimer's disease. Proc Natl Acad Sci U S A 101,10804-10809. Liu, W., Ramachandran, V., Kang, J., Kishimoto, T. K., Cummings, R. D., and McEver, R. P. (1998). Identification of N-terminal residues on P-selectin glycoprotein ligand-1 required for binding to P-selectin. J Biol Chem 273,7078-7087. Lowe, J. B. (2002). Glycosylation in the control of selectin counter-receptor structure and function. Immunol Rev 186,19-36. Lowe, J. B. (2003). Glycan-dependent leukocyte adhesion and recruitment in inflammation. Curr Opin Cell Biol 75,531-538. Lowe, J. B. (2005). Cell biology. Does Notch take the sweet road to success? Science 307,1570-1572. Lowe, J. B., and Marth, J. D. (2003). A genetic approach to Mammalian glycan function. Annu Rev Biochem 72,643-691. Luhn, K., Wild, M . K., Eckhardt, M. , Gerardy-Schahn, R., and Vestweber, D. (2001). The gene defective in leukocyte adhesion deficiency n encodes a putative GDP-fucose transporter. Nat Genet 28,69-72. Luo, Y., and Haltiwanger, R. S. (2005). O-fucosylation of notch occurs in the endoplasmic reticulum. J Biol Chem 280,11289-11294. 205 Ma, J., Simonovic, M. , Qian, R., and ColJey, K. J. (1999). Sialyltransferase isoforms are phosphorylated in the cis-medial Golgi on serine and threonine residues in their luminal sequences. J Biol Chem 274,8046-8052. Maemura, K., and Fukuda, M . (1992). Poly-N-acetyllactosaminyl O-glycans attached to leukosialin. The presence of sialyl Le(x) structures in O-glycans. J Biol Chem 267, 24379-24386. Maly, P., Thall, A., Petryniak, B., Rogers, C. E. , Smith, P. L . , Marks, R. M. , Kelly, R. J., Gersten, K. M . , Cheng, G., Saunders, T. L. , et al. (1996). The alpha(l,3)fucosyltransferase Fuc-TVII controls leukocyte trafficking through an essential role in L- , E- , and P-selectin ligand biosynthesis. Cell 86,643-653. Manjunath, N., and Ardman, B. (1995). CD43 regulates tyrosine phosphorylation of a 93-kD protein in T lymphocytes. Blood 86,4194-4198. Manjunath, N., Correa, M. , Ardman, M. , and Ardman, B. (1995). Negative regulation of T-cell adhesion and activation by CD43. Nature 377, 535-538. Manjunath, N., Johnson, R. S., Staunton, D. E. , Pasqualini, R., and Ardman, B. (1993). Targeted disruption of CD43 gene enhances T lymphocyte adhesion. J Immunol 151,1528-1534.. Martin, C. H., Aifantis, I., Scimone, M . L. , von Andrian, U. H., Reizis, B., von Boehmer, H. , and Gounari, F. (2003). Efficient thymic immigration of B220+ lymphoid-restricted bone marrow cells with T precursor potential. Nat Immunol 4, 866-873. Martinez, M. , Joffraud, M. , Giraud, S., Baisse, B., Bernimoulin, M . P., Schapira, M. , and Spertini, O. (2005). Regulation of PSGL-1 Interactions with L-selectin, P-selectin, and E-selectin: Role of human fucosyltransferase-IV and -VII. J Biol Chem 280,5378-5390. Matsuzaki, Y., Gyotoku, J., Ogawa, M . , Nishikawa, S., Katsura, Y., Gachelin, G., and Nakauchi, H. (1993). Characterization of c-kit positive intrathymic stem cells that are restricted to lymphoid differentiation. J Exp Med 178,1283-1292. Mayadas, T. N., Johnson, R. C , Rayburn, H., Hynes, R. O., and Wagner, D. D. (1993). Leukocyte rolling and extravasation are severely compromised in P selectin-deficient mice. Cell 74, 541-554. Mazo, I. B., Gutierrez-Ramos, J. C , Frenette, P. S., Hynes, R. O., Wagner, D. D., and von Andrian, U. H. (1998). Hematopoietic progenitor cell rolling in bone marrow microvessels: parallel contributions by endothelial selectins and vascular cell adhesion molecule 1. J Exp Med 188,465-474. Mazo, I. B., Quackenbush, E. J., Lowe, J. B., and von Andrian, U. H. (2002). Total body irradiation causes profound changes in endothelial traffic molecules for hematopoietic progenitor cell recruitment to bone marrow. Blood 99,4182-4191. McCormick, C , Duncan, G., Goutsos, K. T., and Tufaro, F. (2000). The putative tumor suppressors EXT1 and EXT2 form a stable complex that accumulates in the Golgi apparatus and catalyzes the synthesis of heparan sulfate. Proc Natl Acad Sci U S A 97, 668-673. McEver, R. P., and Cummings, R. D. (1997). Role of PSGL-1 binding to selectins in leukocyte recruitment. J Clin Invest 100, S97-103. McEver, R. P., and Martin, M . N. (1984). A monoclonal antibody to a membrane glycoprotein binds only to activated platelets. J Biol Chem 259,9799-9804. McEvoy, L . M. , Sun, H., Frelinger, J. G., and Butcher, E. C. (1997a). Anti-CD43 inhibition of T cell homing. J Exp Med 185,1493-1498. McEvoy, L. M. , Sun, H., Tsao, P. S., Cooke, J. P., Berliner, J. A., and Butcher, E. C. (1997b). Novel vascular molecule involved in monocyte adhesion to aortic endothelium in models of atherogenesis. J Exp Med 185, 2069-2077. 206 McNagny, K. M. , Pettersson, I., Rossi, F., Flamme, L, Shevchenko, A., Mann, M. , and Graf, T. (1997). Thrombomucin, a novel cell surface protein that defines thrombocytes and multipotent hematopoietic progenitors. J Cell Biol 138,1395-1407. Mentzer, S. J., Remold, O. D. E. , Crimmins, M . A., Bierer, B. E., Rosen, F. S., and Burakoff, S. J . (1987). Sialophorin, a surface sialoglycoprotein defective in the Wiskott-Aldrich syndrome, is involved in human T lymphocyte proliferation. Journal of Experimental Medicine 165,1383-1392. Merzaban, J. S., Zuccolo, J. , Corbel, S. Y., Williams, M . J. , and Ziltener, H. J. (2005). An Alternate Core 2 {beta}l,6-N-Acetylglucosaminyltransferase Selectively Contributes to P-Selectin Ligand Formation in Activated CD8 T Cells. J Immunol 174,4051-4059. Mikecz, K., Brennan, F. R., Kim, J. H., and Giant, T. T. (1995). Anti-CD44 treatment abrogates tissue oedema and leukocyte infiltration in murine arthritis. Nat Med 1, 558-563. Miller, J. F. (2002). The discovery of thymus function and of thymus-derived lymphocytes. Immunol Rev 185, 7-14. Mishiro, E. , Liu, M . Y., Sakakibara, Y., Suiko, M. , and Liu, M . C. (2004). Zebrafish tyrosylprotein sulfotransferase: molecular cloning, expression, and functional characterization. Biochem Cell Biol 82,295-303. Misslitz, A., Pabst, O., Hintzen, G., Ohl, L . , Kremmer, E. , Petrie, H. T., and Forster, R. (2004). Thymic T cell development and progenitor localization depend on CCR7. J Exp Med 200,481-491. Mitoma, J., Petryniak, B., Hiraoka, N., Yeh, J. C , Lowe, J. B., and Fukuda, M . (2003). Extended core 1 and core 2 branched O-glycans differentially modulate sialyl Lewis X-type L-selectin ligand activity. J Biol Chem 278,9953-9961. Mizgerd, J. P. (2002). Molecular mechanisms of neutrophil recruitment elicited by bacteria in the lungs. Semin Immunol 14,123-132. Moloney, D. J., Panin, V. M. , Johnston, S. H., Chen, J., Shao, L. , Wilson, R., Wang, Y., Stanley, P., Irvine, K. D., Haltiwanger, R. S., and Vogt, T. F. (2000a). Fringe is a glycosyltransferase that modifies Notch. Nature 406,369-375. Moloney, D. J., Shair, L. H., Lu, F. M. , Xia, J. , Locke, R., Matta, K. L . , and Haltiwanger, R. S. (2000b). Mammalian Notchl is modified with two unusual forms of O-linked glycosylation found on epidermal growth factor-like modules. J Biol Chem 275,9604-9611. Moore, K. L . (1998). Structure and function of P-selectin glycoprotein ligand-1. Leuk Lymphoma 29,1-15. Moore, K. L. , Eaton, S. F., Lyons, D. E. , Lichenstein, H. S., Cummings, R. D., and McEver, R. P. (1994). The P-selectin glycoprotein ligand from human neutrophils displays sialylated, fucosylated, O-linked poly-N-acetyllactosamine. J Biol Chem 269,23318-23327. Moore, T. A., and Zlotnik, A. (1995). T-cell lineage commitment and cytokine responses of thymic progenitors. Blood 86,1850-1860. Morrison, S. J., Uchida, N., and Weissman, I. L . (1995). The biology of hematopoietic stem cells. Annu Rev Cell Dev Biol 11,35-71. Moser, B., and Loetscher, P. (2001). Lymphocyte traffic control by chemokines. Nat Immunol 2,123-128. Mosmann, T. R., and Sad, S. (1996). The expanding universe of T-cell subsets: Thl , Th2 and more. Immunol Today 17,138-146. 207 Muller, W. A. (2003). Leukocyte-endothelial-cell interactions in leukocyte transmigration and the inflammatory response. Trends Immunol 24,327-334. Munoz, F. M. , Hawkins, E . P., Bullard, D. C , Beaudet, A. L. , and Kaplan, S. L . (1997). Host defense against systemic infection with Streptococcus pneumoniae is impaired in E-, P-, and E-/P-seIectin-deficient mice. J Clin Invest 100,2099-2106. Munro, S. (1998). Localization of proteins to the Golgi apparatus. Trends Cell Biol 8,11-15. Munro, S. (2003). Lipid rafts: elusive or illusive? Cell 115,377-388. Naiyer, A. J. , Jo, D. Y., Ahn, J., Mohle, R., Peichev, M. , Lam, G., Silverstein, R. L. , Moore, M . A., and Rafii, S. (1999). Stromal derived factor-l-induced chemokinesis of cord blood CD34(+) cells (long-term culture-initiating cells) through endothelial cells is mediated by E-selectin. Blood 94,4011-4019. Naka, Y., Toda, K., Kayano, K., Oz, M . C , and Pinsky, D. J. (1997). Failure to express the P-selectin gene or P-selectin blockade confers early pulmonary protection after lung ischemia or transplantation. Proc Natl Acad Sci U S A 94,757-761. Nakayama, F., Teraki, Y., Kudo, T., Togayachi, A., Iwasaki, H., Tamatani, T., Nishihara, S., Mizukawa, Y., Shiohara, T., and Narimatsu, H. (2000). Expression of cutaneous lymphocyte-associated antigen regulated by a set of glycosyltransferases in human T cells: involvement of alphal, 3-fucosyltransferase VII and betal,4-galactosyltransferase I. J Invest Dermatol 115, 299-306. Nishikawa, A., Nanda, A., Gregory, W., Frenz, J. , and Kornfeld, S. (1999). Identification of amino acids that modulate mannose phosphorylation of mouse DNase I, a secretory glycoprotein. J Biol Chem 274,19309-19315. Nong, Y. H., Remold, O. D. E. , LeBien, T. W., and Remold, H. G. (1989). A monoclonal antibody to sialophorin (CD43) induces homotypic adhesion and activation of human monocytes. Journal of Experimental Medicine 170,259-267. Norman, K. E. , Katopodis, A. G., Thoma, G., Kolbinger, F., Hicks, A. E. , Cotter, M . J. , Pockley, A. G., and Hellewell, P. G. (2000). P-selectin glycoprotein ligand-1 supports rolling on E - and P-selectin in vivo. Blood 96,3585-3591. Norman, M . U., Van De Velde, N. C , Timoshanko, J. R., Issekutz, A., and Hickey, M . J. (2003). Overlapping roles of endothelial selectins and vascular cell adhesion molecule-1 in immune complex-induced leukocyte recruitment in the cremasteric microvasculature. Am J Pathol 163,1491-1503. Norment, A. M. , and Bevan, M . J. (2000). Role of chemokines in thymocyte development. Semin Immunol 12, 445-455. Okajima, T., Fukumoto, S., Miyazaki, H., Ishida, H., Kiso, M. , Furukawa, K., Urano, T., and Furukawa, K. (1999). Molecular cloning of a novel alpha2,3-sialyltransferase (ST3Gal VI) that sialylates type H lactosamine structures on glycoproteins and glycolipids. J Biol Chem 274,11479-11486. Okajima, T., Xu, A., and Irvine, K. D. (2003). Modulation of notch-ligand binding by protein O-fucosyltransferase 1 and fringe. J Biol Chem 278,42340-42345. Okajima, T., Xu, A., Lei, L . , and Irvine, K. D. (2005). Chaperone activity of protein O-fucosyltransferase 1 promotes notch receptor folding. Science 307,1599-1603. Omichi, K., Aoki, K., Minamida, S., and Hase, S. (1997). Presence of UDP-D-xylose: beta-D-glucoside alpha-1,3-D-xylosyltransferase involved in the biosynthesis of the Xyl alpha l-3Glc beta-Ser structure of glycoproteins in the human hepatoma cell Une HepG2. Eur J Biochem 245,143-146. 208 Onami, T. M . , Harrington, L . E. , Williams, M . A., Galvan, M. , Larsen, C. P., Pearson, T. C , Manjunath, N., Baum, L . G., Pearce, B. D., and Ahmed, R. (2002). Dynamic regulation of T cell immunity by CD43. J Immunol 168,6022-6031. Opat, A. S., Houghton, F., and Gleeson, P. A. (2000). Medial Golgi but not late Golgi glycosyltransferases exist as high molecular weight complexes. Role of luminal domain in complex formation and localization. J Biol Chem 275,11836-11845. Osawa, M. , Hanada, K., Hamada, H., and Nakauchi, H. (1996). Long-term lymphohematopoietic reconstitution by a single CD34-low/negative hematopoietic stem cell. Science 273,242-245. Ostberg, J. R., Barth, R. K., and Frelinger, J . G. (1998). The Roman god Janus: a paradigm for the function of CD43. Immunol Today 19, 546-550. Ostberg, J. R., Dragone, L . L . , Borrello, M . A., Phipps, R. P., Barth, R. K., and Frelinger, J. G. (1997). Expression of mouse CD43 in the B cell lineage of transgenic mice causes impaired immune responses to T-independent antigens. European Journal of Immunology 27,2152-2159. Ouyang, Y.-B., Crawley, J. T. B., Aston, C. E. , and Moore, K. L . (2002). Reduced Body Weight and Increased Postimplantation Fetal Death in Tyrosylprotein Sulfotransferase-1-deficient Mice. J Biol Chem 277,23781-23787. Ouyang, Y., Lane, W. S., and Moore, K. L . (1998). Tyrosylprotein sulfotransferase: purification and molecular cloning of an enzyme that catalyzes tyrosine O-sulfation, a common posttranslational modification of eukaryotic proteins. Proc Natl Acad Sci U S A 95, 2896-2901. Oxley, S. M . , and Sackstein, R. (1994). Detection of an L-selectin ligand on a hematopoietic progenitor cell line. Blood 84,3299-3306. Panin, V. M. , Shao, L. , Lei, L. , Moloney, D. J., Irvine, K. D., and Haltiwanger, R. S. (2002). Notch ligands are substrates for protein O-fucosyltransferase-1 and Fringe. J Biol Chem 277,29945-29952. Park, J. H., Yu, Q., Erman, B., Appelbaum, J. S., Montoya-Durango, D., Grimes, H. L . , and Singer, A. (2004). Suppression of IL7Ralpha transcription by IL-7 and other prosurvival cytokines: a novel mechanism for maximizing IL-7-dependent T cell survival. Immunity 21,289-302. Patel, K. D. (1998). Eosinophil tethering to interleukin-4-activated endothelial cells requires both P-selectin and vascular cell adhesion molecule-1. Blood 92,3904-3911. Patel, K. D., Cuvelier, S. L. , and Wiehler, S. (2002). Selectins: critical mediators of leukocyte recruitment. Seminars in Immunology 14, 73. Paulson, J. C. (1989). Glycoproteins: what are the sugar chains for? Trends Biochem Sci 14, lll-llb. Penit, C , Lucas, B., and Vasseur, F. (1995). Cell expansion and growth arrest phases during the transition from precursor (CD4-8-) to immature (CD4+8+) thymocytes in normal and genetically modified mice. J Immunol 154,5103-5113. Perez-Vilar, J. , and Hill, R. L . (1999). The structure and assembly of secreted mucins. J Biol Chem 274, 31751-31754. Perry, S. S., Wang, H., Pierce, L . J. , Yang, A. M. , Tsai, S., and Spangrude, G. J. (2004). L-selectin defines a bone marrow analog to the thymic early T-lineage progenitor. Blood 103,2990-2996. Peschon, J. J. , Morrissey, P. J., Grabstein, K. H., Ramsdell, F. J. , Maraskovsky, E. , Gliniak, B. C , Park, L. S., Ziegler, S. F., Williams, D. E. , Ware, C. B., and et al. (1994). Early lymphocyte expansion is severely impaired in interleukin 7 receptor-deficient mice. J Exp Med 180,1955-1960. 209 Petrie, H. T. (2002). Role of thymic organ structure and stromal composition in steady-state postnatal T-cell production. Immunol Rev 189,8-19. Petrie, H. T. (2003). Cell migration and the control of post-natal T-cell lymphopoiesis in the thymus. Nat Rev Immunol 3,859-866. Petrie, H. T., Hugo, P., Scollay, R., and Shortman, K. (1990). Lineage relationships and developmental kinetics of immature thymocytes: CD3, CD4, and CD8 acquisition in vivo and in vitro. J Exp Med 172,1583-1588. Pfaffl, M . W., Horgan, G. W., and Dempfle, L . (2002). Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res 30, e36. Picker, L . J., Kishimoto, T. K., Smith, C. W., Warnock, R. A., and Butcher, E. C. (1991a). ELAM-1 is an adhesion molecule for skin-homing T cells. Nature 349, 796-799. Picker, L . J. , Michie, S. A., Rott, L . S., and Butcher, E. C. (1990a). A unique phenotype of skin-associated lymphocytes in humans. Preferential expression of the HECA-452 epitope by benign and malignant T cells at cutaneous sites. Am J Pathol 136,1053-1068. Picker, L . J., Terstappen, L . W., Rott, L . S., Streeter, P. R., Stein, H., and Butcher, E. C. (1990b). Differential expression of homing-associated adhesion molecules by T cell subsets in man. J Immunol 145,3247-3255. Picker, L . J. , Warnock, R. A., Burns, A. R., Doerschuk, C. M. , Berg, E . L . , and Butcher, E. C. (1991b). The neutrophil selectin LECAM-1 presents carbohydrate ligands to the vascular selectins ELAM-1 and GMP-140. Cell 66,921-933. Piller, F., Piller, V., Fox, R. I., and Fukuda, M . (1988). Human T-lymphocyte activation is associated with changes in O-glycan biosynthesis. J Biol Chem 263,15146-15150. Porritt, H. E. , Gordon, K., and Petrie, H. T. (2003). Kinetics of Steady-state Differentiation and Mapping of Intrathymic-signaling Environments by Stem Cell Transplantation in Nonirradiated Mice. J Exp Med 198, 957-962. Porritt, H. E. , Rumfelt, L . L. , Tabrizifard, S., Schmitt, T. M. , Zuniga-Pflucker, J . C , and Petrie, H. T. (2004). Heterogeneity among DN1 prothymocytes reveals multiple progenitors with different capacities to generate T cell and non-T cell lineages. Immunity 20,735-745. Pouyani, T., and Seed, B. (1995). PSGL-1 recognition of P-selectin is controlled by a tyrosine sulfation consensus at the PSGL-1 amino terminus. Cell 83,333-343. Priatel, J. J. , Chui, D., Hiraoka, N., Simmons, C. J. , Richardson, K. B., Page, D. M. , Fukuda, M. , Varki, N. M. , and Marth, J. D. (2000). The ST3Gal-I sialyltransferase controls CD8+ T lymphocyte homeostasis by modulating O-glycan biosynthesis. Immunity 12, 273-283. Prockop, S., and Petrie, H. T. (2000). Cell migration and the anatomic control of thymocyte precursor differentiation. Semin Immunol 12,435-444. Prockop, S. E. , Palencia, S., Ryan, C. M. , Gordon, K., Gray, D., and Petrie, H. T. (2002). Stromal cells provide the matrix for migration of early lymphoid progenitors through the thymic cortex. J Immunol 169, 4354^ 1361. Prockop, S. E. , and Petrie, H. T. (2004). Regulation of thymus size by competition for stromal niches among early T cell progenitors. J Immunol 173,1604-1611. Puri, K. D., Finger, E. B., Gaudernack, G., and Springer, T. A. (1995). Sialomucin CD34 is the major L -selectin ligand in human tonsil high endothelial venules. J Cell Biol 131, 261-270. 210 Quelle, F. W., Shimoda, K., Thierfelder, W., Fischer, C , Kim, A., Ruben, S. M. , Cleveland, J. L . , Pierce, J. H., Keegan, A. D., Nelms, K., and et al. (1995). Cloning of murine Stat6 and human Stat6, Stat proteins that are tyrosine phosphorylated in responses to IL-4 and LL-3 but are not required for mitogenesis. Mol Cell Biol 75,3336-3343. Randhawa, A. K., Ziltener, H. J., Merzaban, J. S., and Stokes, R. W. (2005). CD43 is required for optimal growth inhibition of Mycobacterium tuberculosis in macrophages and in mice. J Immunol. Reinhardt, P. H., and Kubes, P. (1998). Differential leukocyte recruitment from whole blood via endothelial adhesion molecules under shear conditions. Blood 92,4691-4699. Remold, O. D. E. , and Parent, D. (1994). Two proteolytic pathways for down-regulation of the barrier molecule CD43 of human neutrophils. Journal of Immunology 752,3595-3605. Renkonen, J., Tynninen, O., Hayry, P., Paavonen, T., and Renkonen, R. (2002). Glycosylation might provide endothelial zip codes for organ-specific leukocyte traffic into inflammatory sites. Am J Pathol 767, 543-550. Reya, T., Morrison, S. J. , Clarke, M . F., and Weissman, I. L. (2001). Stem cells, cancer, and cancer stem cells. N a t u r e s , 105-111. Rieu, P., Porteu, F., Bessou, G., Lesavre, P., and Halbwachs-Mecarelli, L . (1992). Human neutrophils release their major membrane sialoprotein, leukosialin (CD43), during cell activation. European Journal of Immunology 22,3021-3026. Robinson, S. D., Frenette, P. S., Rayburn, H., Cummiskey, M. , Ullman-Cullere, M. , Wagner, D. D., and Hynes, R. O. (1999). Multiple, targeted deficiencies in selectins reveal a predominant role for P-selectin in leukocyte recruitment. Proc Natl Acad Sci U S A 96,11452-11457. Rodewald, H. R., Kretzschmar, K., Takeda, S., Hohl, C , and Dessing, M . (1994). Identification of pro-thymocytes in murine fetal blood: T lineage commitment can precede thymus colonization. Embo J13,4229-4240. Rood, P. M. , Gerritsen, W. R., Kramer, D., Ranzijn, C , von dem Borne, A. E. , and van der Schoot, C. E. (1999). Adhesion of hematopoietic progenitor cells to human bone marrow or umbilical vein derived endothelial cell lines: a comparison. Exp Hematol 27,1306-1314. Ropp, P. A., Little, M . R., and Cheng, P. W. (1991). Mucin biosynthesis: purification and characterization of a mucin beta 6N-acerylglucosaminyltransferase. J Biol Chem 266,23863-23871. Rosen, S. D. (2004). Ligands for L-selectin: homing, inflammation, and beyond. Annu Rev Immunol 22,129-156. Rosenstein, Y., Park, J. K., Hahn, W. C , Rosen, F. S., Bierer, B. E. , and Burakoff, S. J. (1991). CD43, a molecule defective in Wiskott-Aldrich syndrome, binds ICAM-1. Nature 354,233-235. Rossi, D., and Zlotnik, A. (2000). The biology of chemokines and their receptors. Annu Rev Immunol 18,217-242. Rossi, F. M . V., Corbel, S. Y., Merzaban, J. S., Carlow, D. A., Gossens, K., Duenas, J., So, L. , Yi, L. , and Ziltener, H. J. (2005). Recruitment of adult thymic progenitors is regulated by P-selectin and its ligand PSGL-1. Nat Immunol 6, 262-270. Roumier, A., Olivo-Marin, J. C , Arpin, M. , Michel, F., Martin, M. , Mangeat, P., Acuto, O., Dautry-Varsat, A., and Alcover, A. (2001). The membrane-microfilament linker ezrin is involved in the formation of the immunological synapse and in T cell activation. Immunity 75,715-728. Rudd, P. M. , Elliott, T., Cresswell, P., Wilson, I. A., and Dwek, R. A. (2001). Glycosylation and the immune system. Science 297, 2370-2376. 211 Sackstein, R. (2004). The bone marrow is akin to skin: H C E L L and the biology of hematopoietic stem cell homing. J Invest Dermatol 122,1061-1069. Sackstein, R., and Dimitroff, C. J. (2000). A hematopoietic cell L-selectin ligand that is distinct from PSGL-1 and displays N-glycan-dependent binding activity. Blood 96,2765-2774. Sackstein, R., Fu, L . , and Allen, K. L . (1997). A hematopoietic cell L-selectin ligand exhibits sulfate-independent binding activity. Blood 89, 2773-2781. Saitoh, O., Piller, F., Fox, R. I., and Fukuda, M . (1991). T-lymphocytic leukemia expresses complex, branched O-linked oligosaccharides on a major sialoglycoprotein, leukosialin. Blood 77,1491-1499. Sako, D., Chang, X. J. , Barone, K. M . , Vachino, G., White, H. M. , Shaw, G., Veldman, G. M. , Bean, K. M. , Ahern, T. J. , Furie, B., and et al. (1993). Expression cloning of a functional glycoprotein ligand for P-selectin. Cell 75,1179-1186. Sako, D., Comess, K. M . , Barone, K. M. , Camphausen, R. T., Cumming, D. A., and Shaw, G. D. (1995). A sulfated peptide segment at the amino terminus of PSGL-1 is critical for P-selectin binding. Cell 83,323-331. Sallusto, F., Mackay, C. R., and Lanzavecchia, A. (2000). The role of chemokine receptors in primary, effector, and memory immune responses. Annu Rev Immunol 18,593-620. Sanchez-Madrid, F., and del Pozo, M . A. (1999). Leukocyte polarization in cell migration and immune interactions. Embo J 7*, 501-511. Sarangapani, K. K., Yago, T., Klopocki, A. G., Lawrence, M . B., Fieger, C. B., Rosen, S. D., McEver, R. P., and Zhu, C. (2004). Low force decelerates L-selectin dissociation from P-selectin glycoprotein ligand-1 and endoglycan. J Biol Chem 279,2291-2298. Sasai, K., Ikeda, Y., Tsuda, T., Ihara, H., Korekane, H., Shiota, K., and Taniguchi, N. (2001). The Critical Role of the Stem Region as a Functional Domain Responsible for the Oligomerization and Golgi Localization of N- Acetylglucosaminyltransferase V. The involvement of a domain homophilic interaction. J Biol Chem 276, 759-765. Sasaki, K., Watanabe, E. , Kawashima, K., Sekine, S., Dohi, T., Oshima, M . , Hanai, N., Nishi, T., and Hasegawa, M . (1993). Expression cloning of a novel Gal beta (1-3/1-4) GlcNAc alpha 2,3-sialyltransferase using lectin resistance selection. J Biol Chem 268,22782-22787. Sassetti, C , Tangemann, K., Singer, M . S., Kershaw, D. B., and Rosen, S. D. (1998). Identification of podocalyxin-like protein as a high endothelial venule ligand for L-selectin: parallels to CD34. J Exp Med 7*7, 1965-1975. Sassetti, C , Van Zante, A., and Rosen, S. D. (2000). Identification of endoglycan, a member of the CD34/podocalyxin family of sialomucins. J Biol Chem 275,9001-9010. Schachter, H., and Brockhausen, I. (1989). The biosynthesis of branched O-glycans. Symp Soc Exp Biol 43,1-26. Schenkel, A. R., Mamdouh, Z., Chen, X., Liebman, R. M. , and Muller, W. A. (2002). CD99 plays a major role in the migration of monocytes through endothelial junctions. Nat Immunol 3,143-150. Schmid, K., Hediger, M . A., Brossmer, R., Collins, J. H., Haupt, H., Marti, T., Offner, G. D., Schaller, J. , Takagaki, K., Walsh, M . T., and et al. (1992). Amino acid sequence of human plasma galactoglycoprotein: identity with the extracellular region of CD43 (sialophorin). Proceedings of the National Academy of Sciences of the United States of America 89,663-667. 212 Schmitt, T. M. , de Pooter, R. F., Gronski, M . A., Cho, S. K., Ohashi, P. S., and Zuniga-Pflucker, J. C. (2004). Induction of T cell development and establishment of T cell competence from embryonic stem cells differentiated in vitro. Nat Immunol 5,410-417. Schwarz, B. A., and Bhandoola, A. (2004). Circulating hematopoietic progenitors with T lineage potential. Nat Immunol 5,953-960. Schweitzer, K. M. , Drager, A. M. , van der Valk, P., Thijsen, S. F., Zevenbergen, A., Theijsmeijer, A. P., van der Schoot, C. E. , and Langenhuijsen, M . M . (1996). Constitutive expression of E-selectin and vascular cell adhesion molecule-1 on endotheUal cells of hematopoietic tissues. Am J Pathol 148,165-175. Schwientek, T., Bennett, E . P., Flores, C , Thacker, J. , Hollmann, M. , Reis, C. A., Behrens, J. , Mandel, U., Keck, B., Schafer, M . A., et al. (2002a). Functional conservation of subfamilies of putative UDP-N-acety!galactosamine:polypeptide N-acetylgalactosaminyltransferases in Drosophila, Caenorhabditis elegans, and mammals. One subfamily composed of l(2)35Aa is essential in Drosophila. J Biol Chem 277, 22623-22638. Schwientek, T., Keck, B., Levery, S. B., Jensen, M . A., Pedersen, J. W., Wandall, H . H., Stroud, M. , Cohen, S. M. , Amado, M. , and Clausen, H. (2002b). The Drosophila gene brainiac encodes a glycosyltransferase putatively involved in glycosphingolipid synthesis. J Biol Chem 277,32421-32429. Schwientek, T., Nomoto, M. , Levery, S. B., Merkx, G., van Kessel, A. G., Bennett, E . P., Hollingsworth, M . A., and Clausen, H. (1999). Control of O-glycan branch formation. Molecular cloning of human cDNA encoding a novel betal,6-N-acetylglucosaminyltransferase forming core 2 and core 4. J Biol Chem 274,4504-4512. Schwientek, T., Yeh, J. C , Levery, S. B., Keck, B., Merkx, G., van Kessel, A. G., Fukuda, M. , and Clausen, H. (2000). Control of O-glycan branch formation. Molecular cloning and characterization of a novel thymus-associated core 2 betal, 6-n-acetylglucosaminyltransferase. J Biol Chem 275,11106-11113. Scollay, R., Smith, J., and Stauffer, V. (1986). Dynamics of early T cells: prothymocyte migration and proliferation in the adult mouse thymus. Immunol Rev 91,129-157. Scudder, P. R., Shailubhai, K., Duffin, K. L . , Streeter, P. R., and Jacob, G. S. (1994). Enzymatic synthesis of a 6'-sulphated sialyl-Lewisx which is an inhibitor of L-selectin binding to peripheral addressin. Glycobiology 4, 929-932. Sekine, M . , Nara, K., and Suzuki, A. (1997). Tissue-specific regulation of mouse core 2 beta-l,6-N-acetylglucosaminyltransferase. J Biol Chem 272,27246-27252. Seko, A., Dohmae, N., Takio, K., and Yamashita, K. (2003). Beta 1,4-galactosy transferase (beta 4GalT)-IV is specific for GlcNAc 6-O-sulfate. Beta 4GalT-IV acts on keratan sulfate-related glycans and a precursor glycan of 6-sulfosialyl-Lewis X. J Biol Chem 278,9150-9158. Serrador, J. M. , Nieto, M. , Alonso-Lebrero, J. L . , del Pozo, M . A., Calvo, J., Furthmayr, H., Schwartz-Albiez, R., Lozano, F., Gonzalez-Amaro, R., Sanchez-Mateos, P., and Sanchez-Madrid, F. (1998). CD43 interacts with moesin and ezrin and regulates its redistribution to the uropods of T lymphocytes at the cell-cell contacts. Blood 91,4632-4644. Setty, S. R., Strochlic, T. I., Tong, A. H., Boone, C , and Burd, C. G. (2004). Golgi targeting of ARF-like GTPase Arl3p requires its Nalpha-acetylation and the integral membrane protein Syslp. Nat Cell Biol 6,414-419. Shamri, R., Grabovsky, V., Gauguet, J. M. , Feigelson, S., Manevich, E. , Kolanus, W., Robinson, M . K., Staunton, D. E. , von Andrian, U. H., and Alon, R. (2005). Lymphocyte arrest requires instantaneous induction of an extended LFA-1 conformation mediated by endothelium-bound chemokines. Nat Immunol. Shao, L. , Luo, Y., Moloney, D. J., and Haltiwanger, R. (2002). O-glycosylation of EGF repeats: identification and initial characterization of a UDP-glucose: protein O-glucosyltransferase. Glycobiology 12,763-770. 213 Shaw, P., Freeman, J., Bovey, R., and Iggo, R. (1996). Regulation of specific DNA binding by p53: evidence for a role for O-glycosylation and charged residues at the carboxy-terminus. Oncogene 12,921-930. Shimaoka, M . , Salas, A., Yang, W., Weitz-Schmidt, G., and Springer, T. A. (2003). Small molecule integrin antagonists that bind to the beta2 subunit I-like domain and activate signals in one direction and block them in the other. Immunity 19,391-402. Shimizu, Y., and Shaw, S. (1993). Cell adhesion. Mucins in the mainstream. Nature 366, 630-631. Shimonaka, M. , Katagiri, K., Nakayama, T., Fujita, N., Tsuruo, T., Yoshie, O., and Kinashi, T. (2003). Rapl translates chemokine signals to integrin activation, cell polarization, and motility across vascular endothelium under flow. J Cell Biol 161,417-427. Shortman, K., Egerton, M. , Spangrude, G. J., and Scollay, R. (1990). The generation and fate of thymocytes. Semin Immunol 2,3-12. Shortman, K., and Wu, L . (1996). Early T lymphocyte progenitors. Annu Rev Immunol 14,29-47. Si-Tahar, M. , Sitaraman, S., Shibahara, T., and Madara, J. L . (2001). Negative regulation of epithelium-neutrophil interactions via activation of CD44. Am J Physiol Cell Physiol 280, C423-432. Simmons, D. L. , Satterthwaite, A. B., Tenen, D. G., and Seed, B. (1992). Molecular cloning of a cDNA encoding CD34, a sialomucin of human hematopoietic stem cells. J Immunol 148,267-271. Singer, A. (2002). New perspectives on a developmental dilemma: the kinetic signaling model and the importance of signal duration for the CD4/CD8 lineage decision. Curr Opin Immunol 14, 207-215. Singh, J., Khan, G. A., Kinarsky, L. , Cheng, H., Wilken, J., Choi, K. H., Bedows, E. , Sherman, S., and Cheng, P. W. (2004). Identification of disulfide bonds among the nine core 2 N-acetylglucosaminyltransferase-M cysteines conserved in the mucin beta6-N-acetylglucosaminyltransferase family. J Biol Chem 279,38969-38977. Sitnicka, E. , Bryder, D., Theilgaard-Monch, K., Buza-Vidas, N., Adolfsson, J., and Jacobsen, S. E. (2002). Key role of flt3 ligand in regulation of the common lymphoid progenitor but not in maintenance of the hematopoietic stem cell pool. Immunity / 7,463-472. Skrincosky, D., Kain, R., El-Battari, A., Exner, M. , Kerjaschki, D., and Fukuda, M . (1997). Altered Golgi localization of core 2 beta-l,6-N-acety!glucosaminyltransferase leads to decreased synthesis of branched O-glycans. J Biol Chem 272,22695-22702. Smith, M . J. , Smith, B. R., Lawrence, M . B., and Snapp, K. R. (2004). Functional analysis of the combined role of the O-linked branching enzyme core 2 betal-6-N-glucosaminyltransferase and dimerization of P-selectin glycoprotein ligand-1 in rolling on P-selectin. J Biol Chem 279,21984-21991. Smith, P. L. , Gersten, K. M. , Petryniak, B., Kelly, R. J., Rogers, C , Natsuka, Y., Alford, J. A., 3rd, Scheidegger, E. P., Natsuka, S., and Lowe, J. B. (1996). Expression of the alpha(l,3)fucosyltransferase Fuc-TVII in lymphoid aggregate high endothelial venules correlates with expression of L-selectin ligands. J Biol Chem 271,8250-8259. Smith, P. L . , Myers, J. T., Rogers, C. E., Zhou, L. , Petryniak, B., Becker, D. J., Homeister, J. W., and Lowe, J. B. (2002). Conditional control of selectin ligand expression and global fucosylation events in mice with a targeted mutation at the FX locus. J Cell Biol 158, 801-815. Smithson, G., Rogers, C. E. , Smith, P. L . , Scheidegger, E. P., Petryniak, B., Myers, J. T., Kim, D. S., Homeister, J. W., and Lowe, J. B. (2001). Fuc-TVII is required for T helper 1 and T cytotoxic 1 lymphocyte selectin ligand expression and recruitment in inflammation, and together with Fuc-TIV regulates naive T cell trafficking to lymph nodes. J Exp Med 194,601-614. 214 Snapp, K. R., Craig, R., Herron, M. , Nelson, R. D., Stoolman, L . M. , and Kansas, G. S. (1998a). Dimerization of P-selectin glycoprotein ligand-1 (PSGL-1) required for optimal recognition of P-selectin. J Cell Biol 142, 263-270. Snapp, K. R., Ding, H., Atkins, K., Warnke, R., Luscinskas, F. W., and Kansas, G. S. (1998b). A novel P-selectin glycoprotein ligand-1 monoclonal antibody recognizes an epitope within the tyrosine sulfate motif of human PSGL-1 and blocks recognition of both P- and L-selectin. Blood 91,154-164. Snapp, K. R., Heitzig, C. E. , Ellies, L . G., Marth, J. D., and Kansas, G. S. (2001). Differential requirements for the O-linked branching enzyme core 2 betal-6-N-glucosaminyltransferase in biosynthesis of ligands for E -selectin and P-selectin. Blood 97,3806-3811. Snapp, K. R., Wagers, A. J., Craig, R., Stoolman, L . M. , and Kansas, G. S. (1997). P-selectin glycoprotein ligand-1 is essential for adhesion to P-selectin but not E-selectin in stably transfected hematopoietic ceU lines. Blood 89,896-901. Somers, W. S., Tang, J., Shaw, G. D., and Camphausen, R. T. (2000). Insights into the molecular basis of leukocyte tethering and rolling revealed by structures of P- and E-selectin bound to SLe(X) and PSGL-1. Cell 103,467-479. Spangrude, G. J., Heimfeld, S., and Weissman, I. L . (1988). Purification and characterization of mouse hematopoietic stem cells. Science 241,58-62. Spangrude, G. J., and Scollay, R. (1990). Differentiation of hematopoietic stem cells in irradiated mouse thymic lobes. Kinetics and phenotype of progeny. J Immunol 145,3661-3668. Sperandio, M. , Forlow, S. B., Thatte, J., Ellies, L . G., Marth, J. D., and Ley, K. (2001a). Differential requirements for core2 glucosaminyltransferase for endothelial L-selectin ligand function in vivo. J Immunol 767,2268-2274. Sperandio, M . , Smith, M . L . , Forlow, S. B., Olson, T. S., Xia, L. , McEver, R. P., and Ley, K. (2003). P-selectin glycoprotein ligand-1 mediates L-selectin-dependent leukocyte rolling in venules. J Exp Med 797,1355-1363. Sperandio, M. , Thatte, A., Foy, D., Ellies, L . G., Marth, J. D., and Ley, K. (2001b). Severe impairment of leukocyte rolling in venules of core 2 glucosaminyltransferase-deficient mice. Blood 97,3812-3819. Sperling, A. I., Green, J. M. , Mosley, R. L. , Smith, P. L. , DiPaolo, R. J., Klein, J. R., Bluestone, J. A., and Thompson, C. B. (1995). CD43 is a murine T cell costimulatory receptor that functions independently of CD28. J Exp Med 182,139-146. Sperling, A. I., Sedy, J. R., Manjunath, N., Kupfer, A., Ardman, B., and Burkhardt, J. K. (1998). TCR signaling induces selective exclusion of CD43 from the T cell-antigen-presenting cell contact site. J Immunol 767,6459-6462. Spertini, O., Cordey, A. S., Monai, N., Giuffre, L . , and Schapira, M . (1996). P-selectin glycoprotein ligand 1 is a ligand for L-selectin on neutrophils, monocytes, and CD34+ hematopoietic progenitor cells. J Cell Biol 135, 523-531. Sportsman, J. R., Park, M . M. , Cheresh, D. A., Fukuda, M. , Elder, J. H., and Fox, R. I. (1985). Characterization of a membrane surface glycoprotein associated with T-cell activation. J Immunol 135,158-164. Springer, T. A. (1994). Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell 76,301-314. Springer, T. A. (1995). Traffic signals on endothelium for lymphocyte recirculation and leukocyte emigration. Annu Rev Physiol 57,827-872. 215 Staite, N. D., Justen, J. M. , Sly, L . M. , Beaudet, A. L . , and Bullard, D. C. (1996). Inhibition of delayed-type contact hypersensitivity in mice deficient in both E-selectin and P-selectin. Blood 88,2973-2979. Stamper, H. B., Jr., and Woodruff, J. J . (1976). Lymphocyte homing into lymph nodes: in vitro demonstration of the selective affinity of recirculating lymphocytes for high-endothelial venules. J Exp Med 144,828-833. Steeber, D. A., Green, N. E. , Sato, S., and Tedder, T. F. (1996). Humoral immune responses in L-selectin-deficient mice. J Immunol 157,4899-4907. Steegmaier, M. , Borges, E. , Berger, J., Schwarz, H., and Vestweber, D. (1997). The E-selectin-ligand ESL-1 is located in the Golgi as well as on microvilli on the cell surface. J Cell Sci 110 (Pt 6), 687-694. Steegmaier, M. , Levinovitz, A., Isenmann, S., Borges, E. , Lenter, M. , Kocher, H. P., Kleuser, B., and Vestweber, D. (1995). The E-selectin-ligand ESL-1 is a variant of a receptor for fibroblast growth factor. Nature 373,615-620. Stockton, B. M . , Cheng, G., Manjunath, N., Ardman, B., and von Andrian, U. H. (1998). Negative regulation of T cell homing by CD43. Immunity 8,373-381. Stoolman, L . M . (1989). Adhesion molecules controlling lymphocyte migration. Cell 56,907-910. Strahl-Bolsinger, S., Gentzsch, M. , and Tanner, W. (1999). Protein O-mannosylation. Biochim Biophys Acta 1426,297-307. Streeter, P. R., Rouse, B. T., and Butcher, E. C. (1988). Immunohistologic and functional characterization of a vascular addressin involved in lymphocyte homing into peripheral lymph nodes. J Cell Biol 107,1853-1862. Strous, G. J. , van Kerkhof, P., Fallon, R. J. , and Schwartz, A. L . (1987). Golgi galactosyltransferase contains serine-linked phosphate. Eur J Biochem 169,307-311. Subramaniam, M. , Koedam, J. A., and Wagner, D. D. (1993). Divergent fates of P- and E-selectins after their expression on the plasma membrane. Mol Biol Cell 4, 791-801. Subramaniam, M. , Saffaripour, S., Van De Water, L . , Frenette, P. S., Mayadas, T. N., Hynes, R. O., and Wagner, D. D. (1997). Role of endothelial selectins in wound repair. Am J Pathol 150,1701-1709. Suniara, R. K., Jenkinson, E. J. , and Owen, J. J. (1999). Studies on the phenotype of migrant thymic stem cells. Eur J Immunol 29, 75-80. Suzuki, A., Andrew, D. P., Gonzalo, J. A., Fukumoto, M. , Spellberg, J., Hashiyama, M. , Takimoto, H., Gerwin, N., Webb, I., Molineux, G., et al. (1996). CD34-deficient mice have reduced eosinophil accumulation after allergen exposure and show a novel crossreactive 90-kD protein. Blood 87,3550-3562. Symon, F. A., Lawrence, M . B., Williamson, M . L. , Walsh, G. M. , Watson, S. R., and Wardlaw, A. J. (1996). Functional and structural characterization of the eosinophil P-selectin ligand. J Immunol 157,1711-1719. Tang, G. (2005). siRNA and miRNA: an insight into RISCs. Trends Biochem Sci 30,106-114. Tang, H. L . , and Cyster, J. G. (1999). Chemokine Up-regulation and activated T cell attraction by maturing dendritic cells. Science 284,819-822. Taylor-Papadimitriou, J. , Burchell, J. , Miles, D. W., and Dalziel, M . (1999). MUC1 and cancer. Biochim Biophys Acta 1455,301-313. Tchernychev, B., Furie, B., and Furie, B. C. (2003). Peritoneal macrophages express both P-selectin and PSGL-1. J Cell Biol 163,1145-1155. 216 Tedder, T. F., Steeber, D. A., Chen, A., and Engel, P. (1995a). The selectins: vascular adhesion molecules. Faseb J 9,866-873. Tedder, T. F., Steeber, D. A., and Pizcueta, P. (1995b). L-selectin-deficient mice have impaired leukocyte recruitment into inflammatory sites. J Exp Med 181,2259-2264. Teh, H. S., Kisielow, P., Scott, B., Kishi, H., Uematsu, Y., Bluthmann, H., and von Boehmer, H. (1988). Thymic major histocompatibility complex antigens and the alpha beta T-cell receptor determine the CD4/CD8 phenotype of T cells. Nature 335,229-233. Ten Hagen, K. G., Bedi, G. S., Tetaert, D., Kingsley, P. D., Hagen, F. K., Balys, M . M. , Beres, T. M. , Degand, P., and Tabak, L . A. (2001). Cloning and characterization of a ninth member of the UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferase family, ppGaNTase-T9. J Biol Chem 276,17395-17404. Ten Hagen, K. G., Fritz, T. A., and Tabak, L . A. (2003). All in the family: the UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferases. Glycobiology 13,1R-16R. Ten Hagen, K. G., and Tran, D. T. (2002). A UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferase is essential for viability in Drosophila melanogaster. J Biol Chem 277,22616-22622. Todeschini, A. R., Nunes, M . P., Pires, R. S., Lopes, M . F., Previato, J. O., Mendonca-Previato, L . , and DosReis, G. A. (2002). Costimulation of host T lymphocytes by a trypanosomal trans-sialidase: involvement of CD43 signaling. J Immunol 168, 5192-5198. Toki, D., Sarkar, M . , Yip, B., Reck, F., Joziasse, D., Fukuda, M. , Schachter, H., and Brockhausen, I. (1997). Expression of stable human O-glycan core 2 beta-l,6-N-acetylglucosaminyltransferase in Sf9 insect cells. Biochem J 325 (Pt 1), 63-69. Tourigny, M . R., Mazel, S., Burtrum, D. B., and Petrie, H. T. (1997). T cell receptor (TCR)-beta gene recombination: dissociation from cell cycle regulation and developmental progression during T cell ontogeny. J Exp Med 185,1549-1556. Tsuboi, S., and Fukuda, M . (1997). Branched O-linked oligosaccharides ectopically expressed in transgenic mice reduce primary T-cell immune responses. Embo J16,6364-6373. Tu, L. , Chen, A., Delahunty, M . D., Moore, K. L . , Watson, S. R., McEver, R. P., and Tedder, T. F. (1996). L -selectin binds to P-selectin glycoprotein ligand-1 on leukocytes: interactions between the lectin, epidermal growth factor, and consensus repeat domains of the selectins determine ligand binding specificity. J Immunol 157,3995-4004. Uchimura, K., Kadomatsu, K., El-Fasakhany, F. M. , Singer, M . S., Izawa, M. , Kannagi, R., Takeda, N., Rosen, S. D., and Muramatsu, T. (2004). N-acetylglucosamine 6-O-sulfotransferase-l regulates expression of L-selectin ligands and lymphocyte homing. J Biol Chem 279,35001-35008. Uchimura, K., Muramatsu, H., Kaname, T., Ogawa, H., Yamakawa, T., Fan, Q. W., Mitsuoka, C , Kannagi, R., Habuchi, O., Yokoyama, I., et al (1998). Human N-acetylglucosamine-6-O-sulfotransferase involved in the biosynthesis of 6-sulfo sialyl Lewis X: molecular cloning, chromosomal mapping, and expression in various organs and tumor cells. J Biochem (Tokyo) 124, 670-678. Ueno, T., Saito, F., Gray, D. H., Kuse, S., Hieshima, K., Nakano, H., Kakiuchi, T., Lipp, M. , Boyd, R. L. , and Takahama, Y. (2004). CCR7 signals are essential for cortex-medulla migration of developing thymocytes. J Exp Med 200,493-505. Vachino, G., Chang, X. J. , Veldman, G. M. , Kumar, R., Sako, D., Fouser, L . A., Berndt, M . C , and Cumming, D. A. (1995). P-selectin glycoprotein ligand-1 is the major counter-receptor for P-selectin on stimulated T cells and is widely distributed in non-functional form on many lymphocytic cells. J Biol Chem 270,21966-21974. 217 van den Berg, T. K., Nath, D., Ziltener, H. J., Vestweber, D., Fukuda, M. , van Die, I., and Crocker, P. R. (2001). Cutting edge: CD43 functions as a T cell counterreceptor for the macrophage adhesion receptor sialoadhesin (Siglec-1). J Immunol 166,3637-3640. Van den Steen, P., Rudd, P. M. , Dwek, R. A., and Opdenakker, G. (1998). Concepts and principles of O-linked glycosylation. Crit Rev Biochem Mol Biol 33,151-208. van Kooyk, Y., and Figdor, C. G. (2000). Avidity regulation of integrins: the driving force in leukocyte adhesion. Curr Opin Cell Biol 12,542-547. van Wely, C. A., Blanchard, A. D., and Britten, C. J. (1998). Differential expression of alpha3 fucosyltransferases in Thl and Th2 cells correlates with their ability to bind P-selectin. Biochem Biophys Res Commun 247,307-311. van Zante, A., Gauguet, J. M. , Bistrup, A., Tsay, D., von Andrian, U. H., and Rosen, S. D. (2003). Lymphocyte-HEV interactions in lymph nodes of a sulfotransferase-deficient mouse. J Exp Med 198,1289-1300. VanderElst, I. E. , and Datti, A. (1998). beta 1,6 N-acetylglucosaminyltransferase (core 2 GlcNAc-T) expression in normal rat tissues and different cell lines: evidence for complex mechanisms of regulation. Glycobiology 8, 731-740. Varki, A. (1993). Biological roles of oligosaccharides: all of the theories are correct. Glycobiology 3,97-130. Varki, A. (1997). Selectin ligands: will the real ones please stand up? J Clin Invest 99,158-162. Varki, A. (1998). Factors controlling the glycosylation potential of the Golgi apparatus. Trends Cell Biol 8, 34-40. Varki, A., Cummings, R. D., Esko, J., Freeze, H., Hart, G. W., and Marth, J. D. (1999). Essentials of Glycobiology (Cold Spring Harbor, New York, Cold Spring Harbor Laboratory Press). Vestweber, D., and Blanks, J . E . (1999). Mechanisms that regulate the function of the selectins and their ligands. Physiol Rev 79,181-213. Vogel, J., Bendas, G., Bakowsky, U., Hummel, G., Schmidt, R. R., Kettmann, U., and Rothe, U. (1998). The role of glycolipids in mediating cell adhesion: a flow chamber study. Biochim Biophys Acta 1372,205-215. von Andrian, U. H. (1996). Intravital microscopy of the peripheral lymph node microcirculation in mice. Microcirculation 3, 287-300. von Andrian, U. H., Chambers, J. D., Berg, E . L . , Michie, S. A., Brown, D. A., Karolak, D., Ramezani, L . , Berger, E. M. , Arfors, K. E. , and Butcher, E. C. (1993). L-selectin mediates neutrophil rolling in inflamed venules through sialyl LewisX-dependent and -independent recognition pathways. Blood 82,182-191. von Andrian, U. H., and Mackay, C. R. (2000). T-cell function and migration. Two sides of the same coin. N Engl J Med 343,1020-1034. von Freeden-Jeffry, U., Vieira, P., Lucian, L . A., McNeil, T., Burdach, S. E. , and Murray, R. (1995). Lymphopenia in interleukin (IL)-7 gene-deleted mice identifies IL-7 as a nonredundant cytokine. J Exp Med 181,1519-1526. Wagers, A. J. , and Kansas, G. S. (2000). Potent induction of alpha(l,3)-fucosyltransferase V H in activated CD4+ T cells by TGF-beta 1 through a p38 mitogen-activated protein kinase-dependent pathway. J Immunol 165,5011-5016. 218 Wagers, A. J. , Waters, C. M. , Stoolman, L . M. , and Kansas, G. S. (1998). Interleukin 12 and interleukin 4 control T cell adhesion to endothelial selectins through opposite effects on alphal, 3-fucosyltransferase VII gene expression. J Exp Med 188, 2225-2231. Walcheck, B., Moore, K. L. , McEver, R. P., and Kishimoto, T. K. (1996). Neutrophil-neutrophil interactions under hydrodynamic shear stress involve L-selectin and PSGL-1. A mechanism that amplifies initial leukocyte accumulation of P-selectin in vitro. J Clin Invest 98,1081-1087. Walter, U. M. , and Issekutz, A. C. (1997). Role of E - and P-selectin in migration of monocytes and polymorphonuclear leucocytes to cytokine and chemoattractant-induced cutaneous inflammation in the rat. Immunology 92,290-299. Wang, B., Fujisawa, H., Zhuang, L. , Freed, I., Howell, B. G., Shahid, S., Shivji, G. M. , Mak, T. W., and Sauder, D. N. (2000). CD4+ Thl and CD8+ type 1 cytotoxic T cells both play a crucial role in the full development of contact hypersensitivity. J Immunol 165,6783-6790. Wang, Q., Teder, P., Judd, N. P., Noble, P. W., and Doerschuk, C. M . (2002). CD44 deficiency leads to enhanced neutrophil migration and lung injury in Escherichia coli pneumonia in mice. Am J Pathol 161, 2219-2228. Wang, Y., Shao, L. , Shi, S., Harris, R. J., Spellman, M . W., Stanley, P., and Haltiwanger, R. S. (2001a). Modification of epidermal growth factor-like repeats with O-fucose. Molecular cloning and expression of a novel GDP-fucose protein O-fucosyltransferase. J Biol Chem 276,40338-40345. Wang, Y., Tan, J., Sutton-Smith, M . , Ditto, D., Panico, M. , Campbell, R. M. , Varki, N. M . , Long, J. M . , Jaeken, J., Levinson, S. R., et al. (2001b). Modeling human congenital disorder of glycosylation type Ila in the mouse: conservation of asparagine-linked glycan-dependent functions in mammalian physiology and insights into disease pathogenesis. Glycobiology 11,1051-1070. Ward, S. G., and Westwick, J. (1998). Chemokines: understanding their role in T-lymphocyte biology. Biochem J 333 (Pt 3), 457-470. Warnock, R. A., Askari, S., Butcher, E. C , and von Andrian, U. H. (1998). Molecular mechanisms of lymphocyte homing to peripheral lymph nodes. J Exp Med 187,205-216. Watson, S. R., Imai, Y., Fennie, C , Geoffroy, J. S., Rosen, S. D., and Lasky, L . A. (1990). A homing receptor-IgG chimera as a probe for adhesive ligands of lymph node high endothelial venules. J Cell Biol 110,2221-2229. Weber, C. (2003). Novel mechanistic concepts for the control of leukocyte transmigration: specialization of integrins, chemokines, and junctional molecules. J Mol Med 81,4-19. Wein, M. , Sterbinsky, S. A., Bickel, C. A., Schleimer, R. P., and Bochner, B. S. (1995). Comparison of human eosinophil and neutrophil ligands for P-selectin: ligands for P-selectin differ from those for E-selectin. Am J Respir Cell Mol Biol 12,315-319. Weissman, I. L. , Anderson, D. J., and Gage, F. (2001). Stem and progenitor cells: origins, phenotypes, lineage commitments, and transdifferentiations. Annu Rev Cell Dev Biol 17,387-403. Wells, L. , Vosseller, K., and Hart, G. W. (2001). Glycosylation of nucleocytoplasmic proteins: signal transduction and O-GIcNAc. Science 291,2376-2378. Wen, D. X., Livingston, B. D., Medzihradszky, K. F., Kelm, S., Burlingame, A. L. , and Paulson, J. C. (1992). Primary structure of Gal beta l,3(4)GlcNAc alpha 2,3-sialyltransferase determined by mass spectrometry sequence analysis and molecular cloning. Evidence for a protein motif in the sialyltransferase gene family. J Biol Chem 267,21011-21019. 219 Weninger, W., Ulfman, L . H., Cheng, G., Souchkova, N., Quackenbush, E . J. , Lowe, J. B., and von Andrian, U. H. (2000). Specialized contributions by alpha(l,3)-fucosyltransferase-IV and FucT-VTI during leukocyte rolling in dermal microvessels. Immunity 12,665-676. White, S. J., Underhill, G. H., Kaplan, M . H., and Kansas, G. S. (2001). Cutting edge: differential requirements for Stat4 in expression of glycosyltransferases responsible for selectin ligand formation in Thl cells. J Immunol 167,628-631. Wilkins, P. P., McEver, R. P., and Cummings, R. D. (1996). Structures of the O-glycans on P-selectin glycoprotein ligand-1 from HL-60 cells. J Biol Chem 277,18732-18742. Wilkins, P. P., Moore, K. L . , McEver, R. P., and Cummings, R. D. (1995). Tyrosine sulfation of P-selectin glycoprotein ligand-1 is required for high affinity binding to P-selectin. J Biol Chem 270,22677-22680. Wilkinson, B., Owen, J. J., and Jenkinson, E . J. (1999). Factors regulating stem cell recruitment to the fetal thymus. J Immunol 762,3873-3881. Winkler, I. G., Snapp, K. R., Simmons, P. J. , and Levesque, J. P. (2004). Adhesion to E-selectin promotes growth inhibition and apoptosis of human and murine hematopoietic progenitor cells independent of PSGL-1. Blood 103,1685-1692. Wolber, F. M. , Curtis, J. L. , Maly, P., Kelly, R. J., Smith, P., Yednock, T. A., Lowe, J. B., and Stoolman, L . M . (1998). Endothelial selectins and alpha4 integrins regulate independent pathways of T lymphocyte recruitment in the pulmonary immune response. J Immunol 767,4396-4403. Wong, J., Johnston, B., Lee, S. S., Bullard, D. C , Smith, C. W., Beaudet, A. L. , and Kubes, P. (1997). A minimal role for selectins in the recruitment of leukocytes into the inflamed liver microvasculature. J Clin Invest 99,2782-2790. Woodman, R. C , Johnston, B., Hickey, M . J. , Teoh, D., Reinhardt, P., Poon, B. Y., and Kubes, P. (1998). The functional paradox of CD43 in leukocyte recruitment: a study using CD43-deficient mice. J Exp Med 7**, 2181-2186. Woodside, D. G., Liu, S., and Ginsberg, M . H. (2001). Integrin activation. Thromb Haemost 86,316-323. Worthylake, R. A., and Burridge, K. (2001). Leukocyte transendothelial migration: orchestrating the underlying molecular machinery. Curr Opin Cell Biol 13, 569-577. Wright, D. E. , Wagers, A. J. , Gulati, A. P., Johnson, F. L. , and Weissman, I. L . (2001). Physiological migration of hematopoietic stem and progenitor cells. Science 294,1933-1936. Wu, L. , Kincade, P. W., and Shortman, K. (1993). The CD44 expressed on the earliest intrathymic precursor population functions as a thymus homing molecule but does not bind to hyaluronate. Immunol Lett 38,69-75. Wu, L. , L i , C. L. , and Shortman, K. (1996). Thymic dendritic cell precursors: relationship to the T lymphocyte lineage and phenotype of the dendritic cell progeny. J Exp Med 184,903-911. Wu, L. , Scollay, R., Egerton, M . , Pearse, M. , Spangrude, G. J., and Shortman, K. (1991). CD4 expressed on earliest T-lineage precursor cells in the adult murine thymus. Nature 349, 71-74. Wurbel, M . A., Malissen, M. , Guy-Grand, D., Meffre, E. , Nussenzweig, M . C , Richelme, M. , Carrier, A., and Malissen, B. (2001). Mice lacking the CCR9 CC-chemokine receptor show a mild impairment of early T- and B-cell development and a reduction in T-cell receptor gammadelta(+) gut intraepithelial lymphocytes. Blood 98,2626-2632. Xia, L . , Ramachandran, V., McDaniel, J. M. , Nguyen, K. N., Cummings, R. D., and McEver, R. P. (2003). N-terminal residues in murine P-selectin glycoprotein ligand-1 required for binding to murine P-selectin. Blood 101,552-559. 220 Xia, L . , Sperandio, M. , Yago, T., McDaniel, J. M. , Cummings, R. D., Pearson-White, S., Ley, K., and McEver, R. P. (2002a). P-selectin glycoprotein ligand-l-deficient mice have impaired leukocyte tethering to E-selectin under flow. J Clin Invest 109,939-950. Xia, L . , Wang, L. , Chung, A. S., Ivanov, S. S., Ling, M . Y., Dragoi, A. M . , Piatt, A., Gilmer, T. M. , Fu, X. Y., and Chin, Y. E. (2002b). Identification of both positive and negative domains within the epidermal growth factor receptor COOH-terminal region for signal transducer and activator of transcription (STAT) activation. J Biol Chem 277,30716-30723. Xie, H., Lim, Y. C , Luscinskas, F. W., and Lichtman, A. H. (1999). Acquisition of selectin binding and peripheral homing properties by CD4(+) and CD8(+) T cells. J Exp Med 189,1765-1776. Yamamoto, Y., Yasumizu, R., Amou, Y., Watanabe, N., Nishio, N., Toki, J. , Fukuhara, S., and Ikehara, S. (1996). Characterization of peripheral blood stem cells in mice. Blood 88,445-454. Yang, J., Galipeau, J., Kozak, C. A., Furie, B. C , and Furie, B. (1996). Mouse P-selectin glycoprotein ligand-1: molecular cloning, chromosomal localization, and expression of a functional P-selectin receptor. Blood 87, 4176-4186. Yang, J., Hirata, T., Croce, K., Merrill-Skoloff, G., Tchernychev, B., Williams, E. , Flaumenhaft, R., Furie, B. C , and Furie, B. (1999). Targeted Gene Disruption Demonstrates That P-Selectin Glycoprotein Ligand 1 (PSGL-1) Is Required for P-Selectin-mediated but Not E-Selectin-mediated Neutrophil Rolling and Migration. J Exp Med 190,1769-1782. Yang, L. , Bryder, D., Adolfsson, J. , Nygren, J., Mansson, R., Sigvardsson, M. , and Jacobsen, S. E. (2005). Identification of Lin(-)Scal(+)kit(+)CD34(+)Flt3- short-term hematopoietic stem cells capable of rapidly reconstituting and rescuing myeloablated transplant recipients. Blood 105,2717-2723. Yang, X., Qin, W., Lehotay, M. , Toki, D., Dennis, P., Schutzbach, J. S., and Brockhausen, I. (2003). Soluble human core 2 beta6-N-acetylglucosaminyltransferase C2GnTl requires its conserved cysteine residues for full activity. Biochim Biophys Acta 1648,62-1A. Yeh, J. C , Hiraoka, N., Petryniak, B., Nakayama, J., Ellies, L . G., Rabuka, D., Hindsgaul, O., Marth, J. D., Lowe, J. B., and Fukuda, M . (2001). Novel sulfated lymphocyte homing receptors and their control by a Corel extension beta 1,3-N-acetylglucosaminyltransferase. Cell 105,957-969. Yeh, J. C , Ong, E. , and Fukuda, M . (1999). Molecular cloning and expression of a novel beta-1, 6-N-acetylglucosaminyltransferase that forms core 2, core 4, and I branches. J Biol Chem 274,3215-3221. Yen, T. Y., Macher, B. A., Bryson, S., Chang, X., Tvaroska, I., Tse, R., Takeshita, S., Lew, A. M. , and Datti, A. (2003). Highly conserved cysteines of mouse core 2 betal,6-N-acetylglucosaminyltransferase I form a network of disulfide bonds and include a thiol that affects enzyme activity. J Biol Chem 278,45864-45881. Young, W. W., Jr., Holcomb, D. R., Ten Hagen, K. G., and Tabak, L . A. (2003). Expression of UDP-GaLNAc:polypeptide N-acetylgalactosaminyltransferase isoforms in murine tissues determined by real-time PCR: a new view of a large family. Glycobiology 13,549-557. Yousefi, S., Higgins, E. , Daoling, Z., Pollex-Kruger, A., Hindsgaul, O., and Dennis, J. W. (1991). Increased UDP-GlcNAc:Gal beta l-3GaLNAc-R (GlcNAc to GaLNAc) beta-1, 6-N-acetylglucosaminyltransferase activity in metastatic murine tumor cell lines. Control of pol> lactosamine synthesis. Journal of Biological Chemistry 266,1772-1782. Yu, R. K., and Bieberich, E . (2001). Regulation of glycosyltransferases in ganglioside biosynthesis by phosphorylation and dephosphorylation. Molecular and Cellular Endocrinology 177,19. Yuan, J. (1997). Transducing signals of life and death. Current Opinion in Cell Biology 9,247-251. 221 Zanardo, R. C , Bonder, C. S., Hwang, J. M. , Andonegui, G., Liu, L. , Vestweber, D., Zbytnuik, L. , and Kubes, P. (2004). A down-regulatable E-selectin ligand is functionally important for PSGL-l-independent leukocyte-endothelial cell interactions. Blood 104,3766-3773. Zannettino, A. C., Berndt, M . C., Butcher, C., Butcher, E . C., Vadas, M . A., and Simmons, P. J. (1995). Primitive human hematopoietic progenitors adhere to P-selectin (CD62P). Blood 85,3466-3477. Zerfaoui, M. , Fukuda, M. , Langlet, C., Mathieu, S., Suzuki, M. , Lombardo, D., and El-Battari, A. (2002). The cytosolic and transmembrane domains of the beta 1,6 N-acetylglucosaminyltransferase (C2GnT) function as a cis to medial/Golgi-targeting determinant. Glycobiology 12,15-24. Zhang, Z. X., Yang, L. , Young, K. J. , DuTemple, B., and Zhang, L . (2000). Identification of a previously unknown antigen-specific regulatory T cell and its mechanism of suppression. Nat Med 6,782-789. Zlotnik, A., and Yoshie, O. (2000). Chemokines: a new classification system and their role in immunity. Immunity 12,121-127. Zollner, O., Lenter, M . C , Blanks, J. E. , Borges, E. , Steegmaier, M. , Zerwes, H. G., and Vestweber, D. (1997). L-selectin from human, but not from mouse neutrophils binds directly to E-selectin. J Cell Biol 136, 707-716. Zollner, O., and Vestweber, D. (1996). The E-selectin ligand-1 is selectively activated in Chinese hamster ovary cells by the alpha(l,3)-fucosyltransferases IV and VII. J Biol Chem 271,33002-33008. Zou, X., Shinde Patil, V. R., Dagia, N. M. , Smith, L. A., Wargo, M . J., Interliggi, K. A., Lloyd, C. M. , Tees, D. F., Walcheck, B., Lawrance, M . B., and Goetz, D. J. (2005). PSGL-1 Derived from Human Neutrophils is a High Efficiency Ligand for Endothelial Expressed E-selectin under Flow. Am J Physiol Cell Physiol. Zuniga-Pflucker, J. C. (2004). T-cell development made simple. Nat Rev Immunol 4,67-72. 222 APPENDICES APPENDIX I: CORE 2 BRANCHING OF PSGL-1 IS OBSERVED ON DOUBLE -NEGATIVE T-REGS? T cell subsets use C2GlcNAcT-I and C2GlcNAcT-III differentially for P-selectin ligand formation. The functions of distinct T cell subsets vary and, thus, these subsets also differentially regulate their homing behaviour. These regulatory discrepencies may form the basis of our observation that CD8 + T cells, but not CD4 + T cells, are able to use the C2GlcNAcT-III enzyme for P-selectin ligand formation, as CHAPTER 3 illustrates. A further lymphocyte subset of interest is the T regulatory cells (Treg) that are recognized to exert a central role in the regulation of the immune response. In mouse, a subset of integrin CCE+ CD25+ Treg has been described that expresses E- and P-selectin ligands (Huehn et al., 2004). In analyzing CD4+CD25+ Treg from C57BL/6 mice, we discovered that a small subset (less than 5%) of ex vivo-analyzed CD4+CD25+ is positive for P-selectin ligand binding. This is significantly less than reported by Huehn et al. (Huehn et al., 2004) and may reflect the SPF condition of our mouse colony. CD4+CD25+ Treg cells from wild-type and C2GlcNAcT-In u l 1 mice were, furthermore, sorted by FACS and activated with anti-TCR. We find that CD4 + Treg gain P-selectin ligand expression in a C2GlcNAcT-I-dependent mechanism (Figure A-1A). Co-incubation of Treg with Con A-stimulated, CFSE-labelled splenocytes was used to confirm their Treg function (Figure A-IB). As observed, these regulatory T cells can significantly slow down the proliferation of these splenocytes, as evidenced in the higher degree of CFSE intensity. 223 Figure A-l: P-selectin binds activated wild-type but not C2GlcNAcT-inu" CD4+CD25+ T reg cells A, CD4+CD25+ T cells were isolated from splenocytes of wild-type and C2GlcNAcT-I n u U mice and then activated for five days using immobilized anti-TCR (2c 11), along with IL-2 prior P-selectin-hIG binding (shaded graph) and analysis by FACS. 5mM EDTA was added during incubation with P-selectin-hIG as a negative control (black line); B, the activated CD4+CD25+ cells were then incubated with CFSE-labelled splenocytes, along with Con A, to test the ability of the Tregs to inhibit the proliferation of the splenocytes as measured by a dilution in the CFSE signal. Representative plots are shown; assays were done in triplicate. CFSE Three- to four-day cultures of Con A blasts cultured in IL-2 and IL-4 give rise to a subset of double-negative (DN) cells that, based on the cell-surface markers (a|3-TCR+, CD3 + , CD4", CD8", CD25 + , CD28", CD44" and NK1.1"), appear to be D N Treg, described by Zhang and colleagues (Zhang et al., 2000). Analysis of these D N from wild-type and C2GlcNAcT-I n u 1 1 cultures reveals that P-selectin ligand expression is maintained on a significant portion of the C2GlcNAcT-I n u 1 1 D N cells (Figure A-2). This population of cells also stains strongly for 1B11, indicating that CD43 on these cells is core 2 glycosylated. These data indicate that CD4 + Tregs use C2GlcNAcT-I and D N Tregs use C2GlcNAcT-I and T-III isoenzymes for P-selectin ligand formation. Future experiments in our lab will examine Treg subsets by real-time PCR for analysis of C2GlcNAcT and other glycosyltransferases to further corroborate this observation. P-C57BL/6 C2GlcNAcT-lnu" ® E 3 E O o I SS P-selectin B Cultured with CD4+CD25+ Cultured with CD4+CD25" 224 selectin binding to these cells also depends on PSGL-1 since cells from PSGL-1 n u mice do not bind P-selectin-hIG, and yet are 1B11+. C2GlcNAcT-lnu" C57BL/6 PSGL-1 null O O <u • QL IL-2 IL-2 & IL-4 1B11 Figure A-2: P-selectin binds double-negative T cells in the absence of C2GlcNAcT-I Splenocytes from C2GlcNAcT-Inu", C57BL/6 and PSGL-l n u " mice were activated with Con A and either IL-2 (top half of diagram) or IL-2 & IL-4 (bottom half of diagram) at high density (2.0xl06 cells per millilitre). P-selectin-hIG binding and CD43-1B11 expression were assessed on the CD4 and CD8 double-negative population of cells by FACS on day 4 of the culture. Our analysis potentially uncovers further differences in control mechanisms potentially allowing Treg subset specific modulation of the homing process. Future experiments in our lab will strive to determine which C2GlcNAcT isoenzymes are used by each Treg subset for P-selectin ligand formation by analyzing these cells in C2GlcNAcT-In u 1 1 mice and also in C2GlcNAcT-IIInuI1 mice (currently being generated). 225 APPENDIX II: LACK OF CD44 DOES NOT APPEAR TO INHIBIT THYMIC PROGENITOR HOMING CD44 has been implicated in controlling thymic progenitor homing to the thymus (Kawakami et al., 1999; Wu et al., 1993). A number of preliminary studies are under way in collaboration with Dr. Pauline Johnson's lab to study the role of CD44 in the homing process. CD44 is expressed on thymic progenitors and endothelial cells. It is therefore important to decipher the role of CD44 not just on the thymic progenitor itself but also on the thymic endothelium. We have not examined CD44 expression on the thymic endothelium directly, but evidence suggests that CD44 is abundantly expressed on many endothelial cells (Johnson et al., 2000; Khan et al., 2004). To begin probing the role of CD44 in thymic progenitor homing, we set up a number of competitive repopulation experiments, as Figure A-3 outlines. These studies compared the ability of CD44n u 1 1 bone marrow with wild-type cells to generate DP thymocytes. Irradiated wild-type CD44n u 1 1 and P-selectinnu11, as well as non-irradiated IL-7R n u l 1 mice, were used as recipients for these experiments. Competitions into wild-type and IL-7Rn u" mice (Figure A-3 A and B) reveal that the lack of CD44 on thymic progenitors does not affect their ability to enter the thymus. This is also true when CD44 is absent from the endothelium (assuming it is present in wild-type mice) since competitions into irradiated CD44 n u" mice do not show a competitive disadvantage for CD44 n u U cells (Figure A-3C). An interesting finding is that when P-selectin is removed from the picture, CD44n u 1 1 cells actually have an easier time getting into the thymus compared with wild-type competing cells (Figure A-3D). This is similar with what is seen when PSGL-1 n u l 1 bone marrow is injected into P-selectinnu" mice - competitive advantage of wild-type cells over PSGL-1 n u l 1 cells is lost, but it is enhanced. It is difficult to speculate what is occurring. Lack of CD44 on thymic progenitors alone does not have an affect on homing, nor does lack of CD44 on the endothelium alone or on both the thymic progenitor and the endothelium. However, lack of CD44 on the thymic progenitor, in addition 226 • WT • CD44nu" A B C D Figure A-3: Lack of CD44 does not inhibit thymic progenitor homing in competitive repopulation experiments Competitive repopulation studies were set up by injecting a 1:1 ratio of (A) bone marrow from wild-type (WT) C57BL/6 CD45.1/5.2+ mice with bone marrow from congenic CD44n u 1 1 CD45.2+ mice into lethally irradiated WT CD45.1+-recipient mice (n=3; p>0.05); (B) bone marrow from WT Thyl . l + mice with bone marrow from congenic CD44n u" Thyl.2+ mice into non-irradiated I L 7 R n u i i r e c i p i e n t m i c e (n=3 ; p<0.05); (C) bone marrow from WT Thyl.l+CD45.1 with bone marrow from CD44n u 1 1 Thyl.2+ CD45.1/5.2+ mice into lethally irradiated CD44n u 1 1 Thyl.2+CD45.2+-recipient mice (n=3; p>0.05); and (D) bone marrow from WT Thyl. 1+ mice with bone marrow from CD44n u 1 1 Thyl.2+ into irradiated-P-selectinnu11-recipient mice (n=3; p<0.001). Mice were analyzed three to four weeks after injection to determine the proportion of the competing partner cells in the DP thymocyte subset of the reconstituted thymi. Thymic chimerism is based on the frequency of DP thymocyte subset derived from the competing partner cells. Values above bars represent the mean frequency of chimerism. Error bars represent the standard error of the mean. Control experiments were also run in parallel (not shown) where in the place ofCD44 n u 1 1 bone marrow donors WT donors of the same background were used; these control competitions showed trends that were similar to those observed for A, B and C. to lack of P-selectin on the endothelium, dramatically enhances entry of thymic progenitors. There is evidence that CD44 deficiency enhances neutrophil migration (Wang et al., 2002), suggesting that the lack of CD44 on thymic progenitors may also enhance their migration. This is indirectly supported by evidence of negative regulation of epithelium-neutrophil interactions via activation of CD44 either through ligation with its ligand or by using the oc-CD44 Ab, IM7 that leads to decreased neutrophil migration (Si-Tahar et al., 2001). In 227 addition, a recent study could not find a role for CD44 in either neutrophil rolling or migration through the interstitium (Khan et al., 2004). Over all, our preliminary data, along with these data, suggest that CD44 may in fact negatively regulate the migration of cells. Si-Tahar et al. suggest that both the antibody and natural ligand for CD44 stimulate cAMP production in neutrophils, a known negative-regulation signal for cell movement (Si-Tahar et al., 2001). One possible explanation is that CD44 expressed on the thymic progenitor and CD44 and/or HA expressed on the endothelium upon interaction lead to signalling events that inhibit migration of the progenitor. As such, removing either CD44 on the progenitor or on the endothelium relieves this negative regulation, allowing cells to transmigrate more freely based on interactions between PSGL-1 and P-selectin (Rossi et al., 2005). Since the thymus has limited space for new progenitors and since only a specific progenitor (CTP) may enter the thymus from the blood, this negative regulatory force may be critical in deterring cells not meant to enter the thymus, allowing for those CTPs expressing PSGL-1 in a properly glycosylated state to enter the thymus via interaction with P-selectin. A model representing these ideas is outlined in Figure A-4. Further studies will first focus on confirming these data and building on these hypotheses. 228 Figure A-4: Molecules involved in CTP homing to the thymus Model describing circulating thymic progenitor (CTP) cell entry into the thymus based on competitions done with mice lacking PSGL-1, C2GlcNAcT-I, P-selectin and/or CD44. A, No migration of blood cell into thymic endothelium since functional PSGL-1/P-selectin interaction is not present and able to overcome the negative regulatory effect of CD44-ligand interactions that inhibit migration of progenitors into the thymus. B, Normal entry of progenitors into the thymus since a functional PSGL-1/P-selectin interaction exists and overcomes the negative regulation of the CD44 interaction with its ligand. 229 LIST O F PUBLICATIONS Merzaban, J. S., J. Zuccolo, S. Y. Corbel, M. J. Williams, and H. J. Ziltener. 2005. An Alternate Core 2 pl,6-Af-Acetylglucosammyltransferase Selectively Contributes to P-Selectin Ligand Formation in Activated CD8 T Cells. Journal of Immunology 174: 4051-4059. Rossi, F. M. V., S. Y. Corbel*, J. S. Merzaban*, D. A. Carlow, K. Gossens, J. Duenas, L. So, L. Y i , and H. J. Ziltener. 2005. Recruitment of adult thymic progenitors is regulated by P-selectin and its ligand PSGL-1. Nature Immunology 6: 626-634. Drew, E., J. S. Merzaban, W. Seo, H. J. Ziltener, and K. M. McNagny. 2005. CD34 and CD43 inhibit mast cell adhesion and are required for optimal mast cell reconstitution. Immunity 22: 43-57. Randhawa, A. K., H. J. Ziltener, J. S. Merzaban, and R. W. Stokes. 2005. CD43 is Required for Optimal Growth Inhibition of Mycobacterium tuberculosis in Macrophages and in Mice. Journal of Immunology. 175: 1805-1812. * These authors contributed equally The contents of the first two publications are closely related to this thesis. The first is the basis of CHAPTER 3 and CHAPTER 4 whereas the second is the basis of CHAPTER 5. 230 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items