UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Molecular modulation of the lymphohaemopoietic system Jones, Amanda Tomlinson 1998

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

Item Metadata

Download

Media
831-ubc_1998-271730.pdf [ 17.62MB ]
Metadata
JSON: 831-1.0088803.json
JSON-LD: 831-1.0088803-ld.json
RDF/XML (Pretty): 831-1.0088803-rdf.xml
RDF/JSON: 831-1.0088803-rdf.json
Turtle: 831-1.0088803-turtle.txt
N-Triples: 831-1.0088803-rdf-ntriples.txt
Original Record: 831-1.0088803-source.json
Full Text
831-1.0088803-fulltext.txt
Citation
831-1.0088803.ris

Full Text

MOLECULAR MODULATION OF THE LYMPHOHAEMOPOIETIC SYSTEM by AMANDA TOMLFNSON JONES A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE DEGREE OF DOCTOR IN PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Experimental Medicine We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April 1998 Amanda Tomlinson Jones, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ^pCnmiYV&t iWMCUAl The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract The immune system has evolved to enable recognition of a huge variety of antigens with high specificity via the production of large quantities of reactive antibody. The ability to direct both the specifities and the quantity of antibody production to derive antibodies with high affinities and pre-determined specificities has been an important advance in our exploration of the functions of the lymphohaemopoietic system. This thesis details two separate projects which have as their common theme the production of antibodies with defined specificities and their use as tools to further our understanding of the immune response. In the first part of the thesis, a monoclonal antibody against a cell surface glycoprotein is characterised and its target antigen identified as a unique isoform of CD43, likely the murine homologue of the T305 antigen. This isoform is developmentally regulated with differential patterns of expression on CD4 and CD8 lymphocytes and on activated cells. In addition the molecule is aberrantly expressed in a number of disease models including B-cell lymphoma, allograft rejection, graft-versus-host disease and the Ipr/fas mouse. The pattern of regulation of expression of the antigen is described in addition to studies detailing its biochemical structure, turnover, patterns of phosphorylation and intracellular associations. In addition some speculations as to the function of this specific isoform of CD43 are offered. In the second part of the thesis we describe the production and characterisation of polyclonal anti-peptide antibodies which react with the interleukin 3 (IL-3 molecule). These antibodies can be used as carrier molecules to prolong the in vivo half-life, alter the biodistribution and to enhance the bioactivity of IL 3 up to 7 fold. The characteristics of enhancing antibodies are defined, detailed pharmacokinetics are presented and some speculations as to the potential therapeutic uses and biological significance of such antibodies arising naturally in vivo are presented. ii T A B L E OF CONTENTS ABSTRACT T A B L E OF CONTENTS LIST OF TABLES LIST OF FIGURES AKNOWLEDGEMENTS Chapter I Introduction 1 Chapter II Materials and Methods I 11 II. 1 Animals 11 11.2 Tissue Culture 11.3 Production of monoclonal antibodies 11 11.3.1 ' Derivation of the IB 11 monoclonal antibody 11.3.2 Derivation of the 1F10 monoclonal antibody 13 11.4 Quantifying antibody production 11.4.1 Screening for antibody production using the Idexx screen 13 machine 11.4.2 Isotyping 14 11.4.3 Determination of antibody titre by ELISA 14 11.5 Production of antibodies from ascites 15 11.5.1 Ascites production 15 11.5.2 Purification of antibody from ascites fluid 15 11.6 FITC-labelling and biotinylation of antibodies 16 11.7 Preparation of primary tissues for Fluorescence activated cell sorter 16 (FACS) analysis, functional assays and immunoblotting 11.8 Preparation of human and murine T-lymphocytes from peripheral blood 17 11.9 FACS analysis 17 11.9.1 Antibodies 17 11.9.2 Methods of antibody labelling 17 11.9.3 FACS analysis using triple-stains 19 11.10 Preparation of depleted populations 20 IT. 11 Preparation of cell lysates 20 11.12 Immunoblotting of whole cellly sates 21 11.13 Radiolabelling of IB 11 antigen in cell lysates 21 II.13.1 35s_]yTemionine labelling of IB 11 antigen in cell lysates 21 II. 13.2 35S-Sulphate labelling of IB 11 antigen in cell lysates 22 II. 13.3 Pulse chase experiments 22 11.14 Immunoprecipitations 22 11.15 Cytokines and T-cell activation 24 11.16 Allograft rejection model 24 11.17. Immunofluorescence studies 25 11.18 Histochemistry 25 11.19 Detection of phosphokinase activity 26 11.20 Detection of phospatase activity 26 11.21 Proliferation assays 27 11.21.1 Cell lines and thymocytes 27 11.21.2 Mixed lymphocyte reaction 27 11.22 Adhesion assays .28 11.22.1 Adhesion to endothelial monolayers 28 11.22.2 Conjugate formation 28 11.23 Cytotoxicity assays 29 Page ii iii viii viii xi iii 11.24 Migration studies using FITC labelled activated lymphocytes 30 11.24.1 Determination of optimum FITC concentration 30 11.24.2 Half-life of FITC labelling 30 11.24.3 Dual staining of FITC labelled cells 30 11.24.4 Lymphocyte migration studies in vivo 30 11.25 Deglycosylation methods 31 11.25.1 Deglycosylation of IB 11 antigen using swainsonine 31 endoglycosidase H 11.25.2 Deglycosylation of IB 11 antigen using N-glycanase .32 11.25.3 Removal of O-linkcd glycosylation 32 11.25.4 Neuraminidase treatment of whole cells 33 Chapter III Production and characterisation of the 1B11 monoclonal 34 antibody III. 1 Introduction and Aims 34 111.2 Results - 35 111.2.1 IB 11 monoclonal antibody can be purified from ascites on 35 a sheep anti-mouse immunoglobulin column 111.2.2 IB 11 monoclonal antibodies do not differ in their ability to 35 detect their target antigen in Western irnmunoblots of D26 whole cell lysates 111.2.3 Determination of the optimum IB 11 dilution for FACS 35 analysis ILf.2.4 Determination of the optimum conditions for 38 immunprecipitating IB 11 antigen lif.2.5 IB 11 monoclonal antibody isotype 42 111.3 Discussion 43 Chapter IV Distribution pf the IB 11 antigen 45 IV. 1 Introduction and aims , 45 IV.2 Results 46 IV.2.1 IB 11 antigen expression is confined to lymphohemopoietic 46 cells IV.2.2 IB 11 antigen is differentially expressed by T lymphocyte. 46 sub sets in lymphoid tissue in vivo . IV.2.3 1B11 antigen is expressed at very low levels in B 50 lymphocytes IV.2.4 The pattern of expression of IB 11 immunoreactivity 58 differs from that of CD43, CD44 and CD45 IV.2.5 IB 11 antigen is expressed by activated macrophages and 60 by natural killer (NK) cells IV.2.6 T-lymphocytes expressing the gamma-delta T cell receptor 60 show high levels of IB 11 immunoreactivity IV.2.7 IB 11 mab does not cross-react with human T-lymphocytes 65 IV.3 < Discussion \ ' 66 Chapter V Regulation of 1B-11 antigen expression 70 V. 1 Introduction and aims 70 i v '• . ' V.2 . Results ';• ' 71 . V.2.1* Regulation of expression of 1BT1 antigen in response to 71 , T-cell activation V.2.2 IB 11 antigen expression within T7lymphocyte subsets 71 V.2.3 IB 11 antigen expression in response to IL-2 71 V.2.4 , 1B11 antigen expression in response to IL-1 a,TL-4 78 and PMA V.2.5 IB 1.1 antigen expression in allograft rejection 78 V.2.6 The MRLlpr/fas autoimmune mouse model 82 V.2,6.1 IB 11 antigen expression is increased in 82. autoimmune mouse model V. 2.6.2. IB 11 antigen expression increases with age 96 in the Ipr/fas mice V. 2.7 IB 11 antigen expression in a B-cell lymphoma 100 V. 3 Discussion 108 Chapter VI Functional studies of the IB 11 antigen 115 VI. 1 Introduction and aims 115 VI.2 Results VI. 2.1 Kinase studies . , . 119 VI.2.2 Phosphatase studies 119 VI.2.3 . Proliferation assays 119 VI. 2.3.1 i. Thymocytes and cell lines ' 119 VI.2.3.2 Mixed lymphocyte reaction . 122 VI.2.4 Differential Detergent Solubility 122 : VI.2.5 Immunofluorescence Studies 127 VI.2.6 Adhesion assays: 127 VI.2.61 Adhesion of cells to an endothelial cell 127. monolayer can be detected by thymidine labelling • VI.2.6.2 Effects of mAbs and cytokines 127 VI.2.7 Cytotoxicity assays' 131 VI.2.7.1 Conjugate formation VI.2.7.2 Chromium release assays. 131 VI. 2.8 Lymphocyte migration studies": 131 VI.2.8.1 FITC labelling of effectors 131 VI.2.8.2 Lymphocyte migration studies: Half-life 135 of FITC in vitro VI.2.8.3 Dual staining of effectors 135 . VI.2.8.4 In vivo fate of labelled cells 135 VI. 3 Discussion 140 Chapter VII Biochemistry of lBllantigen 143 VII. l Introduction 143 VII.2 Results 144 VII. 2.1 IB 11 is a single chain glycoprotein with no intra-chain 144 disulphide bridges . . . VII.2.2 Deglycosylation studies 144 v ' VII.2.2.1 N-linked glycosylation makes little 144 or no contribution to its apparent molecular weight VII.2.2.2 IB 11 antigen is a sialylated glycoprotein 148 which shows aberrant mobility on SDS PAGE after removal of sialic acid VII.2.3 IB 11 mAb recognises a 130 kDa isoform of CD43 152 VII.2:4 The pattern of expression of the T305 antigen resembles 155 that of IB 11 antigen VII.2.5 Role of sialic acid in generating the IB 11 epitope: 160 VII.2.5.1 Neuraminidase treament 160 VII.2.5.2 IB 11 immunoreactivity is enhance by 165 removal of terminal 2-3 linked sialic acid residues VII.2.6 IB 11 antigen does not cross-react with sialyl-Lewis X 165 antigen in immunoblots VII.2.7 IB 11 antigen is not a sulphated glycoprotein. 170 VII.2.8 The pi of the IB 11 antigen in activated spleen cells 170 is close to 6 VII.2.9 Pulse chase experiments suggest the turnover of the IB 11 173 antigen is approximately 45-60 minutes VII.2.10 IB 11 is phosphorylated on serine residues in response to 180 Con A and IL-2 VII. 3 Discussion 183 Chapter VIII Comparison of S7 and 1B11 epitope expression in vivo 187 and in vitro VILLI Introduction and aims 187 VIII. 2 Results 198 VII. 2.1 S7 antigen expression is not increased in response 188 to Con A activation of thymus or spleen VIII. 2.2 S7 immunoreactivity is expressed at high levels by the 188 Ipr/fas mouse but the antigen is distinct from IB 11 VIII.2.3 In cell lines, IB 11 epitope is distinct from that 197 recognised by S7 VIII. 3 Discussion 200 Chapter IX Discussion 203 Chapter X Materials and Methods 2 215 X . l Production of antibodies . 215 X . l . l Rabbit anti-1-29 peptide antibodies 215 X.l .2 2E11 and 1 A3 monoclonal antibodies 215 X.2 Affinity purification of antibodies 216 X.2.1" Anti-murineTL-3 antibodies . 216 X.3 Detection of antibody binding to IL-3 216 X.4 Production of recombinant murine LL-3 216 X.5 Bioassay of mIL-3 217 vi X.5.1 Standardising JX-3 preparations 217 X.5.1 Bioassay of recombinant IL-3 in the presence of anti-IL-3 217 antibodies X.6 Gel-filtration profile of antibodies 218 X.7 In vivo pharmacology . 218 X.7.1 Rabbit anti-1-29 antibodies ,218 X.7.2 2E11 monoclonal antibodies 218 X. 8 In vivo biology 219 X.8.1 Quantification of spleen colony formation in vitro 219 X.8.2 Determination of mast cell precursor frequency in vivo 219 X. 8.3 Histology 220 Chapter X I Characterisation of antibodies 221 XI. 1 Introduction and aims 221 XI.2 Results 223 XI. 2.1 Rabbit anti 1-29 (IL-3), 2E11 mAb and 1 A3 mAb 223 recognise synthetic and native IL-3 in solid phase with high titres XI.2.2.1 Rabbit anti-1 -29 antibodies show weak neutralising 223 properties in biassay XI.2.2.2 2E11 and 1 A3 mAbs do not inhibit the bioactivity of 223 murine interleukin 3 on R6X cell line XI.2.3.1 Rabbit anti-1-29 (IL-3) antibodies augment the apparent 225 molecular weight of murine IL-3 as assessed by gel filtration chromatography XI. 2.3.2 2E11 mAb and 1A3 mAb fail to augment the molecular 229 weight of murine interleukin 3 as assessed by gel filtration chromatography XI. 3 Discussion 232 Chapter X I I Pharmacokinetics 234 XII. 1 Introduction and aims 234 XII.2 Results 235 XII. 2.1 2E11 monoclonal antibody does not significantly extend 235 the biological half life of IL-3 in vivo XII. 2.2 Anti 1-29 IL-3 antibodies significantly extend the 235 biological half-life of IL-3 in vivo XII. 3 Discussion 239 Chapter X I I Biological Effects of Enhancing Antibodies 244 XIII. 1 Introduction and aims 244 XIII.2 Results 246 XIII. 2.1 Anti-IL-3 antibodies increase CFU-cs in the spleen 246 XIII.2.2 Mast cell precursors XIIL2.2.1 Anti IL-3 antibodies increase mast cell 246 precursor frequencies in the spleen. vii XIII.2.3 XIII.2.4 XIII.3 Discussion XIII.2.2.2. Anti-IL-3 antibodies decrease mast cell precursor frequencies in the bone marrow Anti-IL-3 antibodies increase the numbers of mature mast cells in the spleen but not in other tissues. JX-3 bioactivity is not detectable in serum of treated mice 24 hours after injection 250 250 250 253 Chapter XIV Chapter XV Discussion References 255 261 LIST OF TABLES Table Table Table Table Table Table 6 Table 7 Table 8 Table 9 Half-lifes of cytokines Cell lines and their derivations Desciption of monoclonal antibodies IB 11 antigen expression in tissue and cell lines 1B11 immunoreactivity in thymus and spleen relative to control antibody IB IT immunoreactivity on thymus: comparison of CD43, CD44 and CD45 1B11 antigen and IL-2 receptor b chain expression in young and old, +/+ and Ipr/fas mice In vivo fate of injected aactivated splenocytes Mast cell and megakaryocyte frequency in the.spleen LIST OF FIGURES Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Purification of IB 11 mAb on sheep anti-mouse immunoglobulin columns IBM mAb reliably detects its target antigen in lOmg of total cellular protein Titration of IB 11 immunoreactivity on D26 cell line: FACs profiles Determination of optimum conditions for the immunoprecipitation of IB 11 antigen IB 11 antigen is differentially expressed in thymus 1B11 antigen is differentially expressed in lymph node IB 11 antigen is differentially expressed in spleen IB 11 antigen is expressed at very low levels by B-lymphocytes in spleen and lymph nodes 1B11 antigen is expressed by activated macrophages and NK cells in peritoneal lavage IB 11 is expressed at high levels by TCR-receptor positive, large, granular CD4-8- cells in IL-2 cultured thymus IB 11 antigen expression is upregulated in thymus in response to Con A activation IB 1.1 antigen expression is upregulated in spleen.in response to Con A activation IB 11 antigen is preferentially up-regulated on CD4+8-thymocytes in response to Con A IB 11 antigen is preferentially up-regulated on CD4+8-splenocytes in response to Con A 7 12 18 47 55 59 99 139 252 36 37 39 41 48 51 53 56 .61 63 72 73 74 76 viii Figure 15 IB 11 antigen expression is upregulated on CD4-8- thymocytes cultured in IL-2 Figure 16 Interleukin 4, interleukin 1 and PMA do not increase IB 11 . expression on thymocytes Figure 17 Induction of allograft rejection is correlated with increased numbers of CD8+ lymphocytes and with increased IB.11 imrnunbreactivity in CD4+ lymphocytes in peritoneal lavage Figure 18 1B-11 immunoreactivity is associated with larger, more granular lymphocytes in allograft rejection Figure 19 IB 11 immunoreactivity in thymus of MRL+/+and MRLlpr/fas mice Figure 20 1B11 immunoreactivity in lymph node of MRL+/+ and MRLlpr/fas mice Figure 21 IB41 immunoreactivity in BL6gld/gld thymus Figure 22 IB 11 immunoreactivity in BL6gld/gld lymph node Figure 23 1B11 expression is increased in B220hi CD4+8- lymphocytes compared to B220'° CD4+8- lymphocytes in MRLlpr/fas mice Figure 24 The CD4-8- population becomes progressively expanded with age in MRLlpr/fas mice Figure 25 Development of lymphoma in BALB/c mouse is associated with expansion of a B220+IgM-CD4-8- subset in the spleen Figure 26 Development of lymphoma in BALB/c mouse is associated with expansion of a B220+IgM-CD4-8- subset in the lymph nodes Figure 27 IB 11 antigen expression is upregulated on CD4+lymphocytes in lymphoma Figure 28 IB 11 mAb is unable to deliver or augment a proliferative signal in T-lymphocytes Figure 29 IB 11 mAb produces modest inhibition of the mixed lymphocyte reaction. which is not significantly different from an isotype matched control mAb Figure 30 IB 11 associates with the cytoskeleton in thymus and activated lymphocytes Figure 31 Adhesion of cells to an endothelial cell monolayer can be detected by H^-Thymidine labelling Figure 32 IB 11 mAb does not inhibit adhesion of the HT-2Y cell line to an endothelial cell monolayer Figure 33 IB 11 mAb does not inhibit conjugate formation in the cytotoxic response Figure 34 IB 11 mab does not inhibit the cytotoxic response to allograft Figure 35 Activated splenocytes can be labelled with FTTC Figure 36 IB 11 antigen binds to wheat germ agglutinin Figure 37 IB H mAb recognises a single chain molecule with no disulphide bridges Figure 38 IB 11 antigen is resistant to swainsonine/endoglycosidase H digestion Figure 39 N-linked glycosylation of the IB 11 antigen makes little or no contribution to its apparent molecular weight Figure 40 IB 11 antigen is resistant to the action of O-glycanase, but responds to neuraminidase treatment with an aberrant shift in mobility on SDS PAGE gel. Figure 41 IB 11 mAb recognises high molecular weight isoforms of CD43 . Figure 42 T305 mAb recognises an antigen on human T-lymphocytes with a similar distribution to 1B.11 mAb Figure 43 Neuraminidase treatment enhances IB 11 mAb immunoreactivity Figure 44 IB 11 mAb immunoreactivity is enhanced by pre-treatment with Newcastle Disease neuraminidase Figure 45 IB 11 immunoreactivity in splenocytes is enhancedby 79 81 83 84 86 88 90 92 94 97 101 103 106 120 123 125 128 129 132 133 136 145 146 149 151 153 156 158 161 163 166 ix pre-treatment with neuraminidase Figure 46 Cell lines expressing high levels of IB 11 antigen do not express the sialyl Lewis X epitope 168 Figure 47 The high molecular weight isoform of CD43 recognised by 171 Figure IB 11 mAb do not contain sulphate 48 Characterisation of the lymphocyte population used to purifiy IB 11 antigen for 2-D gel analysis 174 Figure 49 The pi of the high molecular weight isoform of CD43 recognised 176 by IB 11 mAb is 6.0 Figure 50 1B11 mAb recognises mature forms of the high molecular weight isoform of CD43 178 Figure 51 IB 11 antige is phosphorylated on serine in response to T-lymphocyte activation 181 Figure 52 S7 antigen expression is not up-regulated in thymus or spleen in response to Con A activation 189 Figure 53 S7 immunoreactivity is not increased within subpopulations of activated thymus or spleen 191 Figure 54 S7 antigen is expressed at high levels in lymph nodes and thymus from MRL Ipr/fas mice 193 Figure 55' S7 antigen is normally expressed in young MRL Ipr/fas mice 195 Figure 56 IB 1.1 mAb and S7 mAb recognise isoforms of CD43 which are expressed on different cell types by FACS and by immunoblot 198 Figure 57 Effect of Rabbit anti 1-29 (IL-3) antibody on the proliferative response of R6-XE4 cells to recombinant IL-3 224 Figure 58 Gel filtration elution profile of IL-3 in the presence and absence of rabbit anti 1-29 (IL-3) antibodies 226 Figure 59 Gel filtration profile for IL-3 and rabbit anti 1-29 (IL-3) antibody on Superose 6 column 228 Figure 60 Gel filtration elution profile of IL-3 in the presence and absence of 1 A3 monoclonal antibodies 230 Figure 61 Gel filtration elution profile of IL-3 in the presence and absence of 2E11 monoclonal antibodies 231 Figure 62 Effect of 2E11 monoclonal antibody on the clearance of DL.-3 236 Figure 63 Effect of rabbit anti 1-29 (IL-3) antibody on the clearance of IL-3 237 Figure 64 Frequency of IL-3 CFcs derived from spleens of mice undergoing various treatments 247 Figure 65 Frequency of mast cell precursors in spleens of mice undergoing various treatments 248 Figure 66 Frequency of mast cell precursors in bone marrows of mice undergoing different treatments 251 X Acknowledgments I would like to thank Dr Hermann Ziltener for his support and excellent supervision and to acknowledge the superb technical assistance of Helen Merkens and Michelle Rebbetoy in the performance of the in vivo pharmacology and biology experiments, the excellent technical assistance of Michael Williams with the 2D-gels, and the critical advice of Dr John Schrader, Dr Vince Duronio, Dr Leslie Ellies and Dr Melanie Wellham. I would like to thank Ian Foltz for his advice on computing, Dr Doug Waterfield for proof-reading, Dr Kara Laing for proof-reading and generally putting up with my craziness and most importantly, my husband Mark, and my dog Maurice, for going without food and company. This thesis is dedicated to the memory of Bartolome Federsppiel. xi I. Introduction The immune system has two functional divisions; the innate immune system and the adaptive immune system (1). Innate immunity acts as a first line of defence against infectious agents and most potential pathogens are checked before they can establish an overt infection. If these first defences are breached, the adaptive immune system is activated and produces a specific response to each infectious agent which normally eradicates that agent. In order to protect the host, . the immune system must be able to recognise an agent which it may never have encountered before, distinguish that this is foreign to the body and activate a mechanism capable of neutralising or eliminating this agent. Therefore the immune system must have the capacity to adapt and to recognise an almost infinite number of antigens with a high degree of specificity. The adaptive immune system historically has been divided into two effector components; the humoral and the cell-mediated immune response (1). However there is substantial overlap between these two systems. Antibodies which comprise a family of structurally related glycoproteins produced by B-lymphocytes, mediate the effects of the humoral limb of the immune system. Antibody molecules are capable of recognising and binding to specific target molecules called antigens. Although antibody binding is entirely non-covalent, the strength and specificity of this interaction is far greater than for any other antigen recognising structure e.g. the T cell antigen receptor (TCR) or MHC molecules. Haying bound its target antigen, the antibody is then able to engage other components of the immune system in bringing about the neutralisation or elimination of its target antigen or of the organism carrying that antigen (1). In order to generate the diversity of antibodies required to recognise and adapt to new antigens, immunoglobulin genes show unique features which they share only with the T-cell receptor gene family. Firstly immunoglobulin molecules are comprised of a tetramer of 4 glycoprotein chains, 2 heavy chains and 2 light chains, linked by disulphide bridges. The heavy and the light chains are transcribed from genes on separate chromosomes (2). The combinations of different heavy and light chains contributes to diversity. The type of heavy chain used determines the class of the antibody. Each heavy and light chain comprises constant and variable parts which 1 are derived from different exons within the gene. The constant components show limited diversity and.provide the structural support for the variable regions and in the case of the heavy chain also transduce the effector functions of the antibody, such as Fc binding and activation of complement. The constant region of the light chain is designated either- K or X. The constant region of the heavy chain determines the class of the antibody arid may be either 8, (i, a, yor e (2). The variable components are derived from the recombination of 2-3 exons which are present in multiple variable copies, grouped in families along the chromosome (3-5) Diversity is generated by the multiple recombinatory potentials of these exons. Addition of nucleotides at the sites of exon joining in the heavy chain gene further contributes to diversity (6). Rearrangement of immunoglobulin genes begins at the pre-B cell stage of B cell development and proceeds to produce a functional immunoglobulin gene (7-10). Further diversity can be generated following activation of the immune response in which helper T-lymphocytes stimulate class switching between different heavy chain constant regions (11,12) and somatic mutation of the variable regions of both heavy and light chains occurs (13-16). Almost every kind of biological molecule including simple intermediary metabolites, sugars,, lipids, autocoids, hormones as well as complex carbohydrates, phospholipids, nucleic acids and proteins can serve as an antigen. However, only macromolecules can initiate the lymphocyte activation necessary for an antibody response (17). In order to generate antibodies specific for small molecules (or haptens), immunologists have attached the small molecules to macromolecules called carriers (17). The resultant hapten-carrier complex then acts as an immunogen. Large immunogens carry multiple determinants, each of which may. be bound by an antibody. Specificity may be enhanced by employing a small peptide sequence coupled to carrier molecules to derive antibodies of very limited specificities. The disadvantage of using peptides to generate an antibody response is that antibodies react with 3 dimensional structures and peptide molecules may assume very different three dimensional structures when removed from the body of the protein (18-19): When large immunogens are employed, serum from immunised animals typically contains a variety of antibodies reactive with different determinants on the immunogen and of varying specificities. Although such anti-sera have proved useful for nearly a quarter of a century for the treatment of disease and the detection of antigen, the relevant antibodies comprise only a small fraction of the total protein in serum and exhibit a wide range of sensitivities and specificities. The need for homogeneous antibodies as reproducible reagents was fulfilled by the development of monoclonal technology in the 1970s. This technique, based on the fact that each B-lymphocyte produces antibody of a single specificity, allows normal B-lymphocytes to be propagated indefinitely by the fusion of normal antibody producing B cells with an immortal myeloma cell line. The resultant somatic hybrid, also immortal, can be selected to produce antibodies of a desired specificity. The first description of this technique was by Kohler and Milstein in 1975 (20). They used the Sendai virus (which expresses an envelope protein which fuses cell membranes) to initiate fusion of a murine plasmacytoma cell line with an antibody producing B cell. The plasmacytoma cell line utilised, had been mutagenised so that it was deficient in hypoxanthine-guanine phosphoribosyl transferase. This enzyme is required for the synthesis of purine nucleotides and thymidylates in the presence of folate antagonists such as aminopterin. Normal B-lymphocytes synthesise this enzyme and are therefore able to grow in the presence of aminopterin if hypoxanthine and thymidine are supplied in the media. In the fusion process, the normal B-lymphocytes supply this enzyme to the mutant plasmacytoma cells. Thus successful fusion hybrids, can be isolated by their ability to grow in media containing hypoxanthine, aminopterin and thymidine (HAT). More recent advances in this technique include the use of myeloma cells which do not express their own immunoglobulin and the use of polyethylene glycol as a fusing agent (21). The development of monoclonal antibodies made accessible to biotechnology, the vast recombinatorial potential of the immune system for generating antibodies with predetermined specificities. Subsequently this technology has been extended to allow for bacterial expression of antibody fragments via the isolation and cloning of specific immunoglobulin genes (22) and ultimately the selection of antibody variable domains displayed oh filamentous phage (23). Nevertheless in isolating antibody production from the in vivo immune response and in producing only small fragments of antibodies, some important characteristics of natural antibodies have been lost. These include recruitment of immune effectors, Fc recognition and binding and extended serum half-life. Some recent efforts have been made to. recover these properties by further modification of the structures of these antibody fragments (24,25). The limited variability'of the Fc portion of natural immunoglobulin has been critical to their use in studying the biochemical structure of proteins and is central to the techniques of immunoprecipitation (26), Enzyme-Linked Immunoassay (ELISA) and Western immunoblotting (27). In addition the divalent nature of antigen recognition and the ability of antibodies to form aggregates allows for more effective cross-linkage of receptors and may more accurately mimic the effects of ligand interaction in vivo. It is likely that further refinements of the production of antibodies of defined specificities using phage libraries will allow more of these functions to be recovered in the future. One area in which monoclonal antibodies have been critical to our understanding of the functions of the immune system is in characterising the patterns of expression, functions and regulation of leukocyte adhesion molecules. In 1996, the Sixth International Workshop and Conference on Human Leukocyte Differentiation Antigens held in Kobe, Japan, reported on over 1200 monoclonal antibodies, defined their specificities and described their effects on cell function (28). Antibodies submitted, were grouped together according to their specificities. Many of these antibodies recognized the same antigen but either had differing affinities for the antigen or bound to different epitopes within the.same molecule. This ability of monoclonal antibodies to bind at distinct sites within a target molecule has been exploited in order to study the function of different domains within a molecule. In addition some antibodies e.g. anti-CD45 antibodies, have been shown to recognise different isoforms of the same molecule (29). These isoforms may be produced by either differential glycosylation or differential exon usage. The patterns of expression, ligands and functions of different isoforms may also vary and may be defined using the appropriate monoclonal antibody. Monoclonal antibodies to leukocyte adhesion, molecules recognise structures which subserve a variety of functions, including adhesion e.g. LFA-1, anti-adhesion e.g. CD43, proliferation e.g., CD40, inhibition of proliferation, differentiation and cell death e.g. CD95 (Reviewed in 28). Specific antibodies targeted against these structures may therefore inhibit or enhance these functions by mimicking ligand binding or by inhibiting ligand binding. The various monoclonal antibodies reported at the leukocyte typing workshop differed in their abilities to interfere with these functions. The mAbs also differed with respect to their abilities to recognise their targets in histological sections, by FACS analysis, in immunoprecipitation and in Western immunoblotting. The majority of antibodies showed utility only in FACS, immunohistochemistry and some biological studies. Fewer antibodies were useful in biochemical studies and antibodies were rarely effective in all areas. , In this thesis, two projects are described. In the first an unknown murine leukocyte antigen is characterised using a rat monoclonal antibody produced in this laboratory. This monoclonal is able to recognise its target antigen by FACS, in Western immunoblot and by immunoprecipitation. This has facilitated studies of the distribution, regulation and biochemical structure of the antigen. Finally the target antigenic specificity of the antibody is defined and demonstrated to be a unique isoform of a known molecule. This isoform has not been previously characterised in mouse. The pattern of expression of this isoform is then compared with the more common isoform of the same molecule and some speculations are offered as to its function. In the second part of this thesis we describe the production and characterisation of anti-peptide antibodies to interleukin 3 (IL-3) which are able to enhance the biological activity of IL-3. The use of these antibodies exploits several aspects of antibody structure and function, namely their high affinity for their target, predetermined specificity and high molecular weight. IL-3 is a hemopoietic growth factor produced by activated T-lymphocytes (30), Natural Killer (NK) cells (31) activated mast cells (32,33), eosinophils and neutrophils (34). IL-3 acts on early progenitors to produce cells of all lineages and modulates the growth and effector functions of mature cells such as eosinophils, macrophages, lymphocytes, mast cells and endothelial cells (35-39). IL-3 cannot be readily detected in the circulation of healthy individuals or in normal bone marrow (39) and transgenic mice which lack the IL-3 mRNA maintain relatively normal hemopoiesis (40). Therefore its role in normal hemopoiesis is obscure. IL-3 is detected in the circulation in graft-versus-host disease (41), a condition characterised by widespread activation of T-lymphocytes. A role for IL-3 in so called stress-hemopoiesis following infection has been suggested (42). Interleukin 3 regulates early and intermediate stages of hematopoiesis. Following myeloablative chemotherapy TL-3 is able to accelerate recovery of erythroid and myeloid lineages (43). Clinical trials of this molecule have focused on its.use following bone marrow transplantation (44), after myeloablative chemotherapy (45), in the treatment of congenital and acquired bone marrow aplasia (46) and for mobilising peripheralstem cells for transplantation purposes (47). In addition IL-3 has recently been shown to be of benefit in the therapy of HIV associated cytopenias (48) and. in the treatment of antiphospholipid antibody syndrome (49). For these indications IL-3 is given by sub-cutaneous injection. Interleukin 3 is superior to granulocyte-macrophage colony stimulating factor (GM-CSF) in shortening the duration of thrombocytopenia following chemotherapy (50) however optimal therapy for accelerating the recovery of platelet and neutrophil numbers may require sequential treatment with IL-3 followed by GM-CSF (51,52). In one trial of IL-3 therapy, 25% of patients had to withdraw from therapy because of. toxicity (44). This appears to be more common at higher doses (> 5 ixg/kg/d). • In common with many other cytokines, IL-3 is a low molecular weight molecule (22-36 kDa)(53) and as such is rapidly cleared from the circulation (54). Other cytokines vary in both their half-lifes and their mechanism of clearance (see Table 1). The reported half-life of IL-3 is between 10.and 23 minutes (54-61). This necessitates that IL-3 must be administered by continuous intravenous infusion or by high dose bolus sub-cutaneous therapy, in the therapeutic setting. Both of these strategies are expensive and wasteful. The subcutaneous route is also unpredictable in its absorption and may lead to local tissue reaction (60). Although IL-3 does show some tissue binding, the major mechanism of elimination is by renal clearance (62). The low molecular weight of the molecule results in rapid filtration at the glomerulus and the molecule is then excreted in the urine in the form of low molecular weight metabolites. We reasoned that if the molecular weight of IL-3 could be augmented by coupling to a carrier molecule, this mechanism of elimination could be diminished. Table 1. Half-lifes of Cytokines Cytokine Half-life References IL-2 13-120 minutes 55-58 y-IFN 19-30 minutes 59,60 a-1 FN 1-2 hours 61-63 GM-CSF 30 -90 minutes 64-66 G-CSF 1-2 hours 67,68 IL-3 10-120 minutes 54, 69-73 Epo 5 hours 74, 75 M-CSF 30 - 240 minutes 76 7 One of the first reported attempts to prolong the half-life of a cytokine was by chemically coupling the interleukin 2 (IL-2) molecule to polymers of polyethylene glycol (PEG) (78). The resulting compound had an extended half-life and enhanced biological activity (79). There are a number of disadvantages to this method. The process is not reversible therefore there is significant potential for toxicity. In addition the large molecular weight of the polymer can result in decreased ; or altered biological activityas a result of steric hindrance (80,81). In a recent study of PEGylated interleukin 15 (il-15), prolongation of the elimination half-life of the conjugated cytokine was observed, however the PEGylated molecule became a functional antagonist of IL-15 as a result of interference with binding to the beta subunit but not the alpha subunit of the IL-15 receptor (82). In addition, it has not as yet proved feasible to use polymers to target the interleukin 2 to specific tissues. Thus the biodistribution of the polymer-protein conjugate is essentially the biodistribution of the polymer. -The advantages of polymer conjugates are that the size of the polymer-protein conjugate and thus the extent of half-life prolongation, can be specified in advance. In addition, the combination of the polymer with the protein appears to decrease the immunogenicity of the protein as has been shown for interleukin 2 (81). In 1991, Curtis et al (83) reported on the development of an EL-3-GM-CSF fusion protein called rh-pIXY- 321 . This molecule has a greater molecular weight than IL-3 alone and has the potential to bind additionally to GM-CSF receptors, thus stimulating a wider range of hemopoietic cells (84). The molecule is reported to be able to produce greater biological effects (84) than equivalent doses of GM-CSF and IL-3 independently. However, since both cytokines share a common receptor subunit (85,86) and have been shown to compete for common binding sites on hemopoietic cells, it is possible that some of the enhanced activity of this compound may be due to its greater molecular weight and hence delayed renal excretion. . Pre-clinical and early clinical studies suggest that rh-pIXY-321 is able to enhance neutrophil and platelet recovery after chemotherapy without novel side effects (87-94). It is not clear however that therapy with rh-pIXY-321 will be superior to sequential treatment with IL-3, 8 followed by later acting growth factors such as G M - C S F , G-CSF, Erythropoietin or thrombopoietin, the exact prescription to be determined by the specific lineage deficiency (51). There has been limited experience to date with IL-3 in clinical trials (reviewed in 43). Interleukin 3 as a monotherapy has been shown to variably enhance neutrophil and platelet, recovery following chemotherapy (43-45,51), to modestly increase circulating progenitor cells (47,52) and to variably increase platelet and neutrophil numbers in Diamond Blackfan anaemia (45), an inherited form of aplastic anaemia. Sequential therapy with IL-3 and GM-CSF to mobilise peripheral blood progenitor cells results in accelerated recovery of platelet counts compared to granulocyte colony stimulating factor (G-CSF) or GM-CSF alone (51). This suggests that sequential therapy may be of benefit. There is also recent evidence to support the use of EL-3 as a therapy for anti-phospholipid antibody syndrome (49), a disease characterised by recurrent thrombotic episodes and pregnancy loss. IL-3 may also be important in the response to parasitic infection (42). We have been interested in exploring the feasibility of using antibodies to prolong the half-life of IL-3 and to enhance biological responses to IL-3 in vivo.. There are a number of advantages to using antibodies for this purpose; a) antibodies can be produced against virtually any target molecule b) the interaction between antibody and antigen is reversible c) specific antibodies could be used to enhance endogenous production of a target cytokine d) Afunctional antibodies could be used to target cytokines to specific tissues e) antibodies could be engineered so as not to engender an allergic response. . There are anecdotal reports in the literature of auto-antibodies to interleukin-la (IL-1 a) (95-99), interleukin-6 (IL-6) (98,100-103), interferons (104-108) interleukin 8 (IL-8) (109). interleukin 10 (IL-10) (98) and tumour necrosis factor (TNF) (110,111) arising in normal healthy individuals. In addition auto-antibodies to these and other cytokines may be produced in a variety of pathological conditions (99,103,107-109,111 -114) and are found following prolonged therapy with the target cytokine. Traditionally these antibodies have been thought to inhibit the effects of their target cytokine. However it has also been speculated that these antibodies might function as carrier molecules in the circulation and may even target their cytokine to high affinity receptors (115-119). Since there are few reports of antibodies functioning in this manner, it is likely that enhancing antibodies must have specific properties which differ from those of the neutralising antibodies used in conventional experiments, , We set out to screen a panel of antibodies, both monoclonal and polyclonal for enhancing properties and to determine which features were critical to the enhancing function. In addition we developed polyclonal anti-peptide antibodies which were able to act as carrier molecules and to enhance the biological activity of IL-3 in vivo. The results of these studies are presented in the second part of this thesis and in addition, some speculations are offered as to how this work might proceed further with currently available technology. 10 II. Materials and Methods 11.1. Animals Animals were housed under specific pathogen free conditions in the animal facility of the University Hospital, Vancouver, B.C. Male Wistar rats were obtained from Jackson laboratories and were used for production of monoclonal antibodies. BALB/c mice, C57/Black 6 mice, C3H mice, DBA/2 mice and BLSglaVgld mice and were obtained from Jackson laboratories. MRLlpr/fas mice and MRL +/+ mice were a kind gift ofDr. Di'Waterfield. 11.2. Tissue Culture The 1B11 and 1F10 hybridoma cell lines were maintained by serial passage in RPMI, 10% v/v FCS, 20 mM P-mercaptoethanol and cultured at 37 °C in incubators gassed with 5% v/v CO2. At 5-:6 weekly intervals limiting dilutions of IB 11 and 1F10 hybridoma clones were performed and supernatants assayed for antibody production by ELISA (as described). High producing clones were selected and expanded for use in generating ascites. Aliquots of 1-2 x 10^  cells were stored frozen in 90% FCS 10% DMSO at -70 °C for future use. Other cell lines used in these experiments and their growth medium requirements are outlined in Table 2. 11.3. Production of monoclonal antibodies H.3.1. Derivation of the 1B11 monoclonal antibody A male Wistar rat was immunised sub-cutanepusly with the 107 cells of the IL-3 dependent myeloid tumour cell line, WEHI 274.3 in CFA (134). The rat was subsequently boosted by intra-peritoneal (i.p.) injection 3 times at monthly intervals with 107 cells in phosphate buffered saline (PBS). The spleen was then harvested for fusion with the non-producing murine myeloma cell line, Clone 653, according to the method of Goding (139). Resultant hybridomas were screened for antibody production and specificity using an IDEXX screen machine (IDEXX Laboratories 11 Table 2. Cells lines and their derivations Cell Line Cell Type Culture Medium Growth factor Reference HT-2Y T-cell RPMI 10% v/v FCS IL-2 120 E9D4 121 HDK-1 122 . CTLL 123 EL4 124 CT-4S JX-4 125 Clone 653 B-cell RPMI 10% Nil • 126 NS-1 . v/v FCS 127 WE 231 .128 Bai 17 129 D26 Mast cell RPMI 10% Nil 130 . P 815 v/v FCS 131 R6X IL-3 132 MC9 133 WEHI 274.3 Myeloid cell RPMI 10% v/v FCS IL-3 134 NSF-60 .135 FDCP-1 136 3T3 Mesenchymal . cell RPMI 10% v/v FCS Nil 137 Endd-Dl 138 12 Inc. Westbrook, Maine) (see below). The 1F10 monoclonal antibody was subsequently cloned by 3 rounds of limiting dilution. 11.3. 2. Derivation of the 1F10 monoclonal antibody 1F10 hybridomas were produced in this laboratory as described above. Briefly, a male Wistar rat was immunised with a rabbit polyclonal antibody directed against the 15-118 peptide sequence of murine,IL-3. Resultant hybridomas were screened for antibody, production and specificity using an IDEXX screen machine (see below). The resultant rat monoclonal antibody reacted exclusively with the idiotype of this rabbit anti-peptide antibody. 11.4. Quantifying antibody production H.4.1. Screening for antibody production using the IDEXX screen machine Supernatants from antibody-producing clones were screened for mAb production by binding to pandex beads cokted with sheep anti-mouse antibody. Sheep anti-mouse IgG pandex beads were produced as follows: One milliliter of 5% Particles was incubated in 8 mis 0.1 M 2-(4-Morpholino) ethane sulfonic acid buffer (MES), pH 4.5, together with 5 mg l-ethyl-3,3-dimethylaminopropyl carbodimide (EDC) and lmg sheep anti-mouse immunoglobulin in a 30 ml siliconised glass tube. After a 2 hour incubation, the tube was centrifuged at 4500 g for 20 minutes and the supernatant carefully removed. The particles were then washed twice with PBS, 0.2% sodium azide, 2% FCS and resuspended in 40 mis of PBS, 0.2% sodium azide, 2% FCS to give a final dilution of 0.125% particles. Hybridoma supernatants were aissayed for antibody production by incubation of 40 \i\ of hybridoma supernatant together with beads in the IDEXX screen machine. Beads were then washed and binding to sheep anti-mouse IgG antibody was determined by addition of FITC-labeled goat anti-rat antibody (Kirkegaard & Perry Laboratories, Inc. Gaithersburg, MA) at a dilution of 1:250 w/v.^ Following further washing, residual fluorescence was read by theTDEXX screen machine at 485nm/535nm. 13 Following identification of immunoglobulin producing hybridomas, clones were screened for production of specific antibody against WEHI-274.3 cell line as described in Merckens et al (140). Briefly WEHI-274.3 cells were washed in serum free media and resuspended at 2 x 10^  cells per milliliter (ml). Twenty -five milliliters of this cell suspension were then added to each well of a 96 well pandex tray, together with a 1% solution of 6.2 (I pandex beads. An additional 100 [ll of test hybridoma supernatant added to each well. Cells were then incubated at 4°C for 45 minutes to allow binding of the mAb. Cells were then washed with media and FITC-labeled goat anti-rat antibody (Kirkegaard & Perry Laboratories, Inc.) was added at a dilution of l:250w/v . After a further 15 minute incubation, cells were washed and the residual fluorescence read by the IDEXX screen machine at 485nm/535nm. Hybridoma supernatants containing.mAbs directed against cell surface molecules on WEHI-274.3 were identified by comparison of the fluorescent signal on WEHI-274.3 with controls containing no cells or no antibody and with the signal derived from adding hybridoma supernatant and secondary antibody to trays containing control P815 cells admixed with pandex beads. Cells from wells with a positive signal were then sub-cloned by 3 rounds of limiting dilution. 11.4.2 Isotyping Both IB 11 mAb and IF 10 mAb were isotyped using the Immunoselect rat monoclonal antibody isotyping kit (Gibco BRL, Ontario, Canada). Both antibodies were determined to be IgG2a II.4.3. Determination of antibody titre by ELISA. Whereas the IDEXX screen machine was used for rapid processing of the large numbers of hybridoma supernatants for antibody production, conventional solid phase ELISA was used to determine antibody titers. This method was found to be more accurate and produced consistent and reproducible serial dilution curves. Briefly, 96 well Micotest III flexible assay plates (Becton Dickinson, Oxnard, CA) were coated overnight with 10jig/ml polyclonal sheep anti-mouse immunoglobulin antibody produced in this laboratory. (This polyclonal antibody cross-reacts strongly with rat monoclonal antibodies). Assay plates were then blocked with PBS.containing .0.1% w/v skimmed milk powder (PBS/milk). Serial dilutions.of antibody preparations were applied to the plates in duplicates and incubated at room temperature for 4 hours. Plates were then washed with PBS/milk and peroxidase coupled goat anti-rat antibody (Kirkegaard & Perry Laboratories, Inc.) added at a dilution of 1:2000 v/v in PBS/milk. The peroxidase coupled antibody was allowed to bind for 1 hour at room temperature and plates were then washed twice with PBS/milk, followed by a further wash in DDW. ABTS 0.1% w/v in citrate buffer pH 4.5 with 0.2% v/v H 2 O 2 was then added to the plates.and allowed to react for 30 minutes at 37°C. Light was excluded during this reaction time. The peroxidase enzyme catalysed a colour change in the ABTS dye resulting ih the formation of a dark green solution the absorbance of which could be read at 405nm wavelength in an Microplate autoreader (Biotek Scientific Co. Ltd). II.5. Production of antibodies from ascites 11.5.1. Ascites production Nude mice obtained from Jackson Laboratories were inoculated i.p. with 2-4 x 106 1B11 or 1F10 hybridoma cells 1-4 weeks after priming with 0.1 ml of pristane i.p. Ascites was harvested under aseptic conditions 2-3 weeks following injection, centrifuged at 3000 rpm and the supernatant, containing the antibody, sterilised by serial filtration through a 0.8 mm and a 0.4 mm Surfil filter (Costar, Cambridge, MA) Ascites was stored at -20 °C. 11.5.2. Purification of antibody from ascites fluid Monoclonal antibodies were purified from ascites fluid by passage over a sheep anti-mouse antibody column equilibrated to pH 7.2 with PBS. The column was then washed with 10 times the bed volume of PBS and antibody binding to the column eluted with 0.1M glycine pH 2.5 and neutralised with saturated Tris base. Antibody preparations were then dialysed three times against PBS and the protein content of these preparations determined by spectrophotometry. Standard 15 dilutions of 0.5mg/ml and lmg/ml were stored at -20°C for future use. Original ascites preparations, column drop-throughs and final antibody preparations were assayed by ELISA to ensure depletion of antibody. 11.6. FITC labeling and biotinylation of antibodies 1B11 and IF10 mAbs were FITC labeled and biotinylated according to the methods of Goding (2). Briefly, for FITC labeling purified antibody preparations were dialysed overnight against carbonate/bicarbonate buffer (8.6g Na2C03 and 17.2g NaHC03 in 1 litre H2O) at pH 9.5. 100 mg of FITC (made up to lmg/ml in DMSO) per mg of antibody was then added in a shielded container and. the reaction allowed to proceed at room temperature for 2 hours. After 2 hours, the conjugated material was applied to a Sephadex G-25 column, equilibrated in PBS containing 0.1% azide. The coloured material, filtered from the column was then collected and the extent of conjugation, as determined by the absorbance ratio at 495nm and 280nm, was determined by spectrophotometry. For biotinylation, the antibody was dialysed overnight against 0.1M NaHC03, pH 8.0-8.3 and then diluted to a concentration of lmg/ml. Biotin succinimide ester was made up at a concentration of lmg/ml in DMSO and then 120 ml of solution added to the antibody for every lml of solution After 2 hours incubation at room temperature, the mixture was dialysed against PBS 0.1% azide to remove any non-conjugated biotin. Conjugated antibodies were stored in PBS, 0.1%. azide at 4°C prior to use. 11.7. Preparation of primary tissues for Fluorescence-activated cell sorter (FACS) analysis, functional assays and immunoblotting Spleen, lymph nodes, thymus, 12 day embryo, placenta, kidney, liver and muscle were harvested from BALB/c mice. Spleen, lymph nodes and thymus were harvested from MRLlpr/fas, BL6glaVgld and MRL+/+ mice, spleens were harvested from DBA/2 and C3H mice and peritoneal lavages were harvested from C57/Black 6 mice. Tissues were disaggregated using a wire mesh and red cells removed by hypotonic lysis buffer containing 140 mmol NH4CI, 20 mmol Tris base 16 pH 7.2. Cells were then either washed once in PBS and then resuspended in FACS buffer for FACS analysis or lysed in standard lysis buffer as described below. Cells.for functional assays were resuspended in RPMI 5% FCS v/v and incubated at 37°C in incubators gassed with 5% v/v C02 for use in functional assays. 11.8. Preparat ion of human and murine T-lymphocytes f rom peripheral blood Two milliliters of blood in 50U/mi heparin was mixed with 2 mis of Hanks buffered solution. The diluted blood was then underlayed with 3 mis of Ficoll hypaque (Pharmacia, Uppsala, Sweden) and centrifuged at 2000 rpm for 30 minutes. The buffy coat was then removed and washed twice with PBS containing 2% FCS. The cells were then incubated with appropriate antibodies diluted in FACS buffer at 4°C and subject to FACS analysis. 11.9. F A C S analysis II.9.1. Ant ibod ies Antibodies used in FACS analysis are described in Table 3. Single cell suspensions prepared from primary tissue as described above and murine cell lines as described in Table 2 were subject to FACS analysis using the FACStar machine (Becton Dickinson). H.9.2. M e t h o d of antibody labeling Directly labeled antibodies were incubated with target cells at a concentration of 2 x 106 cells per ml for cell lines or 4 xlO^ cells per ml for primary cells in a volume of 100 jxl for 40 minutes on ice. After labeling, cells were washed once with a 300 \il volume of FACS buffer containing PBS and 5% v/v FCS and subjected to FACS analysis Values for sensitivity, detector wavelengths and compensation were determined in advance for each target cell population using unstained and single stained controls. FSC and SSC profiles were used to gate out dead cells from the analysis. Optimum antibody dilutions were determined for labeled-IB 11 mAb by use of serial titrations and by comparison with the 1F10 isotype matched control. Commercial antibodies were 17 Table 3. Description of monoclonal antibodies used in this study Murine Antigen Clone Source Ref. CD43 S7 Pharmingen 141 CD44 IM7.8.1 ATCC 142 CD45RA 14.8 Pharmingen 29 CD45RB 16A Pharmingen 143 CD 122 TM-(31 Pharmingen 144 Mac-1 a Ml/70 ATCC 145 NK-1.1 PK136 Pharmingen 146 B220 RA3-6B2 Pharmingen 147 (130-135)IL-3 2E11 HZ lab 148 (130-135)IL-3 1A3 HZ lab 148 apTCR s H57-597 Pharmingen 149 y8 TCR GL3 Pharmingen 150 CD3 2C11 ATCC 151 Antibody Idiotype 1F10 HZ lab 152 IB 11 antigen IB 11 HZ lab 152 CD4 GK1.5 / B&D 153 CD8 • '• 53-6.7 y . B & D 154 LFA-1 FD441.8 ATCC 155 CD4 (human) OKT4 B&D 156 CD8 (human) OKT3 B&D 157 Sialyl-Lewis X (human) SNH3/SNH4 Biomembrane Inst 158 18 used at recommended dilutions. Batches of FITC-labeled IB 11 mAbs were standardised prior to use by comparison with standard FITC-labeled IB 11 antibody preparations. Data are presented as either dot plots (for dual stains) or as histograms (for single stains and gated populations). Data on binding of fluorescently labeled antibodies to a target population is presented relative to the binding of an isotype matched control antibody to the same population. Where biotinylated antibodies were used in FACS analysis, cells were incubated for 40 . minutes with the relevant antibody, washed once with FACS buffer as described above and then incubated for 30 minutes on ice with streptavidin conjugated phycoerythrin or streptavidin conjugated cychrome. FACS analysis was performed on washed cells and immunofluorescence compared with an isotype matched biotinylated control antibody. Where unlabeled primary antibodies were used to identify surface antigens, the primary antibody was incubated with the target cell for 40 minutes as described above and then a secondary species specific antibody was added and the cells incubated for a further 30 minutes. In these experiments the secondary antibodies comprised FITC labeled sheep anti-mouse IgG antibody, FITC labeled goat anti-rat IgG antibody, phycoerythrin conjugated sheep anti-mouse antibody and phycoerythrin conjugated goat anti-rat IgG antibody. Where unlabeled antibodies were used in experiments involving dual or triple stains, incubation with unlabeled antibody preceded labeling with directly conjugated or biotinylated antibodies and was immediately followed by a step in which target cells were incubated for 30 minutes with a 1:10 v/v dilution of either mouse or rat sera in order to saturate Fc binding sites on the secondary antibody. Where unlabeled antibodies were used to label spleen cells, population were depleted of both macrophages and B-lymphocytes as described below in order to prevent irrelevant binding of unlabeled antibody to Fc receptors on macrophages and of FITC-labeled anti-mouse/rat IgG secondary antibody, to surface immunoglobulin. II.9.3. FACS analysis using triple-stains IB 11 mab binding within sub-populations of lymphocytes was determined by gating populations using anti-CD4 and anti-CD8 fluorescence profiles to define CD4+8-, CD4+8+, CD4-19 8+ and CD4+8+ sub-sets. Data on IB 11 expression is presented as a histogram and compared with binding of an isotype matched control antibody to the identical gated sub-population within the same experiment. 11.10. Preparation of depleted populations Unlabeled antibodies were used as follows to deplete mixed cell populations of cells. expressing their target antigen. Cell populations were counted and then incubated with the appropriate antibody at a dilution of 1:100 v/y for 30 minutes at 4°C. After incubation, cells were washed once in FACS buffer and then 100 (ll of washed sheep anti-mouse IgG dynabeads (Dynal Inc, Great Neck, NJ.) were added for each TO? cells. Cells and beads were rotated for 30 minutes at 4°C and then the bound population was removed with three rounds of depletion using a fixed magnet. The purity of the depleted population was assessed using both labeled depleting antibody (to determine whether all target cells had been labeled) and FITC-labeled anti-rat IgG or anti-mouse IgG antibody (to determine whether all bound cells had been removed). Sheep anti-mouse IgG dynabeads were used,alone in some experiments to deplete spleen cells of B-lymphocytes. Sheep anti-mouse dynabeads but not goat anti-mouse dynabeads showed substantial cross^ reactivity with surface IgM. 11.11. Preparation of cell lysates Standard cell lysates were prepared by solubilisation of washed, pelleted cells in lysis buffer containing 1% v/v NP-40, 0.02M Tris pH 7.5, 0.15M NaCI, 0,7mg/ml pepstatin, 10 mg/ml. aprotonin, lOmg/ml leupeptin, 40 rnM PMSF and soy bean trypsin inhibitor and incubated for 10-15 minutes on ice. Nuclei and cell debris were removed by centrifugation. Protein concentrations of cell line lysates were standardised by Bradford assay prior to preparation of samples for SDS PAGE. > ' In other experiments lysates were produced by varying the concentration of detergent in the lysis buffer in order to determine the extent to which IB IT antigen was associated with the cytoskeleton. Single cell suspensions were derived from primary tissues as described previously 20 and 2 x 10° cells pelleted and lysed in MCSK buffer containing 20 mmol Hepes pH 6.8, 3 mmol MgCl2, 50 mmol NaCI, 0.3M sucrose, 0.5% NP-40, lmmol CaCl2- After solubilisation, the lysates were centrifuged and the soluble lysate removed and stored. The insoluble debris was then re-solubilised using RIPA buffer (1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, 2mmol EDTA, 150 mmol NaCI, 20 mmol Tris pH 7.5)and then centrifuged and the lysate removed. Lysates were then resuspended in sample buffer and resolved on 7.5% SDS PAGE mini-gel and compared with respect to IB 11 antigen immunoreactivity. 11.12. Immunoblotting of whole cell lysates Whole cell lysates were prepared from primary tissues and from cell lines as described. Protein concentrations were standardised by the Bradford assay and were resolved by SDS PAGE according to the method of Laemmli (158). 7.5% mini polyacrylamide gels were used throughout. Following electrophoresis, samples were transferred to nitro-cellulose at constant current according to the method of Towbin etal. (159) and blocked for 1 hour in 20 mM Tris, pH 7.5, 150 mM NaCI (TBS) containing 5% (w/v) skimmed milk powder (TBS/milk). Replicate blots were then incubated overnight at 4 °C in TBS/milk containing IBM mAb at a concentration of 2.5 fig/ml. Blots were then washed extensively and incubated with alkaline phosphatase conjugated secondary antibody at 1:3000 (v/v) (BRL) for 1.5 h at room temperature. Blots were washed extensively and then incubated with NBT and BCIP for 30-60 minutes according to the method of Blake et al. (27). 11.13. Radio labeling of 1B11 antigen in cell lysates , II.13.1 3 5S-Methionine Labeling of 1B11 antigen in cell lysates Cells were 35S-labeled by washing three times in methionine free media and then incubating overnight in methionine free media in the presence of 0.1 mCi/ml of 35S-methionine. Cultures were maintained at a density of 0.5-1 x 106 cells /ml: at 37 °C with 5% v/v C O 2 and 21 appropriate growth factor. After incubation, cells were washed three times in PBS, pelleted and lysed as described above. ' 11.13.2. 35|§_suipj|ate labeling of IB 11 antigen in cell lysates CTL cells were sulphate labeled by modification of the method of Wilson et al (160). Briefly GTL cells were washed sulphate-free media containing 1% FCS and resuspended at 10^  cells per ml. .Cells were then incubated for 4 hours in sulphate free media containing 150 mCi/ml 35s-Sulphate and 20 mmol Hepes, pH 7.2, at 37°C. After 4 hours, cells were, washed twice with ice cold PBS and lysed in. standard lysis buffer. IB 11 mAb and control mAb was then used to immunoprecipitate its target antigen from the labeled lysate as described in section 11.15, and compared to control mAb immunoprecipitations. 11.13.3. Pulse chase experiments CTL cells were washed three times in methionine free media and then pre-incubated for 1 hour at 37°C in methionine-free media containing 10 U/ml IL-2. Thereafter lmCi ^^ S-methionine was added to each 10^  cells and the cells incubated for a further 10 minutes at 37°C. After 10 minutes, the cells were pelleted and washed 3 times and resuspended in RPMI containing lOU/ml IL-2, 20 mM p-mercaptoethanol and 5% v/v FCS. An aliquot containing 7 xlO^ cells was removed, pelleted, arid lysed using standard lysis buffer. Thereafter aliquots of cells were removed at standard time intervals, lysed and stored on ice. At the termination of the experiments lysates were subjected to immunoprecipitation using IB 11 mAb as described below. 11.14. Immunoprecipitations To determine the optimum conditions for immunoprecipitation using the IB 11 mAb, immunoprecipitation of the IB 11 antigen from whole cell lysates of 3 5S -labeled D26 cells was compared using Protein G-coupled Sepharose beads (Pharmacia, Uppsala, Sweden), IB 11 mab, coupled Sepharose beads and rat anti-mouse kappa coupled Sepharose beads. The latter were produced as follows: Thirty milligrams of Sepharose 4B (Pharmacia) was washed 3 times with 1 22 ml of ImM HC1 and aliquotted into 2 Eppendorf tubes. The beads were then resuspended in 300 •;(il of either 1B11 rriab at 0.6 mg/ml or 300 (il of mouse anti-rat kappa chain antibody at 0.6 mg/ml. The mixture was then rotated for 2 hours at room temperature. At the end of the incubation period, the protein content of the supernatant was determined by spectrophotometric analysis of absorbance at 280nm. Following completion of the reaction, as determined by a marked reduction in the protein content of the supernatant, 10% v/v saturated glycine solution was added to the mixture to saturate non-specific protein-binding sites and the beads rotated for 1 hour at room temperature. The beads were then washed 3 times in PBS and resuspended in 200 (tl PBS containing 0.5% w/v sodium azide prior to use. 35S-labeled D26 whole cell lysates were prepared as previously described., One milliliter aliquot of lysate was pre-cleared with 20 (tl of a 50% slurry of Protein-G coupled Sepharose beads for 30 minutes at 4°C. 200 (ll aliquots of supernatant were then incubated with either a) 20 (il of a 50 % slurry of IB 11-Sepharose beads in PBS and rotated for 1 hour at 4°C,.or (b and c) 10 mg of IB 11 mab for 1 hour at 4°C in duplicate. After 1 hour, 20 (il of protein G coupled Sepharose beads were added to b) and 20 (il of mouse anti-rat kappa-coupled Sepharose beads were added to c) and the beads were rotated for 1 hour at 4°C. Finally, all three preparations were pelleted and .washed sequentially with i) IM NaCI, 0.1 M Tris pH 8, 0.1% NP-40, ii) 0.1M NaCI, 0.1M Tris pH 8, 1%.NP-40, 0.3% SDS, 0.001M EDTA, and iii) 0.01M Tris pH 8, 0.1% NP-40. After washing, bound material was eluted with 30 (ll diethanolamine pH 11.5, 0.05% NP40. Eluates were neutralised with 1/5 volume IM Tris pH 7 and diluted into 2% SDS sample buffer for analysis by SDS PAGE. Immunoprecipitates were resolved on 7.5% SDS PAGE gel and then stained for 30 minutes in staining buffer containing 41% v/v methanol, 7% glacial acetic acid, 0.325g/l of Coomassie Brilliant Blue dye. After staining, gels were transferred to destain (41% v/v methanol, 7% glacial acetic acid) for 1 hour and then transferred to Amplify (Amersham), for fluorography for 30 minutes. The gel was then placed on a double thickness of Whatman 3MM paper, covered with Saran wrap and dried at 80 °C under vacuum. Gels were autoradiographed with Kodak XAR-5 film at -70 °C. 23 In subsequent experiments IB 11 mAb and Protein G beads were used for immunoprecipitations from whole cell lysates. 11.15. Cytokines and T-cell activation Single cell suspensions of thymocytes and splenocytes, prepared as described previously, were cultured in RPMI, 10% FCS, 20 |iM P-mercaptoethanol, in the presence of lOU/ml IL-2, 5 Ug/ml Concanavalin A (Con A), 2% X63-IL-4-conditioned media containing interleukin 4 (DJ-4) (10U/ml),.lng/ml IL-1 or lng/ml PMA. All cell cultures were.performed at a density of 5 xlO6 cells/ml. At 24 hour intervals, aliquots of cells were removed and either lysed in standard lysis buffer and resolved on SDS PAGE gel or resuspended in FACS buffer and subject to FACS analysis. In some experiments, Con A stimulation was reversed using lmg/ml a-methyl-mannoside. IL-2 conditioned media was used as a control in some experiments (where indicated) to maintain cell viability during the period of the experiment. 11.16. Allograft rejection model C57/Black 6 mice express the H-2Db antigen, whereas the P815 cell line is derived from mice expressing the H-2Dd- Injection of P815 cells into C57/Black 6 mice results in an allograft response characterised by local T-lymphocyte activation- In order to determine the effect of T-lymphocyte activation in vivo on 1B11 antigen expression, P815 cells were washed twice with. PBS and injected intra-peritoneally into C57/Black 6 mice at a concentration of 2-4 x 107 cells in 200 ul PBS (161). Control C57/Black 6 mice were injected with an equal volume of PBS. After 10 days, the mice were sacrificed and the peritoneal lavage harvested by irrigating the exposed peritoneal cavity with PBS. Peritoneal lavage cells were then pelleted and resuspended in RPMI, 10% FCS, 20 mM (3-mercaptoethanol and allowed to adhere to tissue culture treated dishes for 1 hour at 37 °C in incubators gassed with 5%CQ2v/v. This process selectively, depletes macrophages from the preparations. After 1 hour the dishes were gently washed with media and the non-adherent cells were collected. Cells were then pelleted and resuspended in FACS buffer and subjected to FACs 24 analysis. IB 11 antigen expression as defined by FITC-1B11 mAb binding within gated populations, was determined relative to control antibody within the same gated populations. 11.17. Immunofluorescence Studies Single cell suspensions from the thymus of BALB/c mice were prepared as described previously and incubated for 48 hours in the presence of either 5 |ig/ml Con A or lOU/ml IL-2. After 48 hours, cells were pelleted and resuspended in freshly made 1% paraformaldehyde in PBS. Cells were then incubated at room temperature with 1 |ig/ml IB 11 or control antibody for 30 minutes. After 30 minutes cells were washed twice with Hanks balanced salt solution containing 10% FCS and resuspended in Hanks containing 1:200 dilution of FITC labeled anti-rat antibody and incubated for 30 minutes at room temperature. Cell suspensions were then underlayed with 100 ul of FCS and cells pelleted for 10 seconds at 2000 g. The upper layer in each tube was then removed until only 20 |Xl remained. The labeled cells were then resuspended in this volume and examined under a microscope at x 40 magnification. Untreated thymocytes were fixed with paraformaldehyde as described and were also examined with respect to IB 11 antibody labeling. These cells functioned as controls. 11.18. Histochemistry Histochemical staining was performed on cytospin preparations of spleen and lymph nodes obtained from a mouse with a spontaneous B-cell lymphoma and from control mice in the same litter. Single cell suspensions, were prepared as described previously and used to derive cytospin slides using Cytospin 2 (Shandon, U.K.). Slides were fixed with methanol for 3 minutes and then air dried. Cells were stained with May-Grunwald stain for 3 minutes, washed with DDW for 1 minute and then counter stained with Wrights stain for 12 minutes in a 1:12 ratio with DDW. Cells were then washed with DDW and viewed under a light microscope at (x 40) magnification. 25 .11.19. Determination of phosphokinase activity D26 whole cell lysates were prepared by lysis iri ice cold solubilisation buffer (50 mM Tris-HCl, pi I 7.7, 1% Triton X-100, 10% glycerol, 100 mM NaCI, 2.5 mM EDTA, 10 mM NaF, 0.2mM Na3V04, 1 mM Na3M04, 0.25 mM PMSF, \\iM pepstatin, 0.5 ug/ml leupeptin, and 10 : (Ig/ml soya bean trypsin inhibitor. Nucleii were removed by centrifugation at 20000 g. for 1 minute. 1B11 antigen was immunoprecipitated from D26 cells as previously described excepting that beads were washed 3 times with solubilisation buffer and orice.with kinase buffer (20 mM HEPES, pH 7.2, 5 mM MgC12, 1'mM EGTA, 5 uM (3-fnercaptoethanol, 2 mM Na3V04, 0.25 mM PMSF and 0.5 Ug/ml leupeptin). Beads were then resuspended in 25 jxl kinase buffer containing 1 mg/fnl of substrates, cc-casein, histone HI and myelin basic protein. Five ml of ATP solution (100|iM ATP, lmCi 32P-ATP in kinase buffer) Was added followed by incubation for 15 minutes at 30°C. Reactions were stopped by addition of 30 ul 2x SDS sample buffer followed by boiling for 2 rhinutes. Potentially phosphorylated substrates were resolved on 7.5% SDS PAGE gels and subject to auto radiography (162). ' ~- . . 11.20. Determination of phosphatase activity IB 11 antigen was immunoprecipitated from D26 cell as described previously, but with the following modifications: Protein G was swelled iri Tris buffered saline (150 mmol NaCI, 20 mmol Tris, pH 7.5) and not in PBS. Following immunoprecipitation, 20 ul of immunoprecipitate were incubated on ice for 15 minutes with 50 mmol/1 fi-mercaptoethanol. 10 ul of safhple was then mixed with 5 \i\ 3 x sample buffer (0.3M Tris, pH 6.8, fj-mercaptoethanol 50 mmol/1), 5 (Xl of src peptide and 50 |xl of malachite green and the colour reaction determined (163)., In a separate assay for phosphatase activity, 20 (il of immunoprecipitate was incubated with sample buffer as above and with 5 ml of PNPP. * Phosphatase activity was compared with activity of control antibody; immunoprecipitates . from the same cell line. - . . 26 1 11.21. Proliferation assays 11.21.1. Cell lines and thymocytes Five microlitres of a serial titration of 1B11 mAb or control antibody or 2C11 antibody were added in duplicate to each well of a 60 well Terasaki plate. Control wells contained no antibody. Five milliliters of Con A at a concentration of 5 Ug/ml final or PMA at a concentration of 1 ng/ml final was added to some of the wells. 25000 thymocytes prepared as previously described were then added in a volume of 5 ul to each well in media containing 10 U/ml TL-2. The Terasaki plates were incubated for 24 hours at 37°C, 5% CO2 and then pulsed for 24 hours with ^ H-thymidine. Labeled cells were harvested on to filter discs and counted in a scintillation counter in separate vials after addition of scintillation fluid. Bioassays for proliferation were also performed with HT-2Y and CTL cell lines which express high levels of IB 11 antigen. . 11.21.2. Mixed Lymphocyte Reaction The mixed lymphocyte reaction was determined as described in (164). Briefly, 1 year old C3H and DBA/2 mice were sacrificed and their spleens harvested. Single cell suspensions of splenocytes were prepared as described previously. Samples were divided and 4 ml aliquots of cells at 5 x 106/ml Were irradiated with 2000 Rads. Cellswere then washed once with Hanks balanced salt solution and resuspended in the original volume. Irradiated DBA/2 spleen cells were then mixed with control C3H cells at a 1:1 ratio in a total volume of 100 ul in a 96 well tissue culture plate (Costar). 1B11 mAb or control antibody was then added at a concentrations of 100 Jig/ml, 50 ug/ml, 25 ug/ml or 12.5 ug/ml to triplicates wells. Wells containing irradiated DBA/2; cells alone, C3H cells mixed 1:1 with irradiated C3H cells and DBA/2 mixed 1:1 with C3H cells served as controls. After either 48, 72 or 96 hours assays were pulsed with 50 uCi of 3 H -thymidine for 6 hours and then harvested on to filter paper discs and counted in a scintillation counter. 27 11.22. Adhesion assays 11.22.1. Adhesion to endothelial monolayers Adhesion was determined by a modification of the method of (165) using ^H-thymidine labeled cells adherent to an endothelial cell monolayer. Endo-Dl cells were grown in semi-confluent monolayers on tissue culture treated flat bottomed 96 well trays (Costar). HT-2Y, EL4, HDK-1 and WEHI 274.3 cells were suspended at 2 xlO^ cells per ml and labeled overnight with 100 uCi/ml of 3H-thymidine at 37 °C, 5% C0 2 v/v. Cells were then washed twice with complete media and resuspended at 1 xlO^ cells per ml. 100.|ll of cells were added to each of the initial wells and the cells titrated across the tray in 1:2 serial dilutions. The trays were then centrifuged at 1500 rpm for 3 minutes and then incubated for 2 hours at 37°C, 5% CO2 v/v. After 2 hours wells were washed 3 times with warm Hanks balanced salt solution and then adherent cells were lysed with 50 uT of 1% NP-40, 150 mmol NaCl and 20 mmol Tris, pH 7.5. The lysed contents of each well were then harvested on filter paper discs and counted in a scintillation counter after addition of 300 fxl scintillation fluid. In subsequent experiments, 10^  HT-2Y cells were labeled as above and added to semi-confluent monolayers of Endo-Dl cells pre-tfeated for 24 hours with either 100 U/ml recombinant IL-4 or 2 ng/ml IL-1 fi or medium alone. Adhesion assays were performed in the presence or absence of IB 11 or control antibody at concentrations of either 50 u.g/ml or 10 ixg/ml. Cells were subsequently lysed and harvested as described. 11.22.2. Conjugate Formation The cytotoxic T cell line, (CTL) was obtained from the laboratory of Dr Hung-Tsia Teh. These cell are reactive with P815 cell line and were demonstrated by FACS to express high levels of IB 11 antigen and LFA-1 (data not shown). Effector cells were also generated in vivo by immunising Black 6 mice intra-peritoneally with P815 cells as previously described (161). 28 Conjugate formation was determined by the method of (166) modified to incorporate caleeine A M . Briefly, P815 target cells were washed twice with RPMI and resuspended at 2 x 10^ cells per ml in RPMI alone. Ten milligrams of calceirie A M stock (6.9 pi of a 1 mg/ml solution in DMSO added to 23.1 pi of PBS ) was added to each ml of cells and the cells incubated for 50 minutes at 37°C, 5% CQ2 v/v and then washed twice with complete media and resuspended at 4 x 10^ cells /ml. Effector cells were diluted to 4 x 10^ cells /ml and incubated for 30 minutes on ice with either IB 11 mAb, control antibody or anti-CD8 antibody at 100 jig/ml. 100 |il of effectors were then mixed with 100 |il of targets and the mixture pelleted in Eppendorf tubes for 3,0 seconds. After a further 10 minute incubation at room temperature, cells were resuspended using 20 aspirations with a 100 | i l pipette and fixedwith 40 ul of 5% paraformaldehyde. Conjugate formation was then scored using the fluorescent microscope. The percentage of fluorescently labeled cells conjugated with non-fluorescent cells, was determined as a function of the total number of fluorescently labeled cells. Cells were counted in a minimum of 3 low.power fields. 11.23. Cytotoxicity Assays Cytotoxicity was determined by standard chromium release assay (167). P815 and EL4 target cells were labeled for 1 hour with 100 (iCi of 51-Chromium per 10^ cells. Cells were then washed twice with media and added to target cells at T:E ratios of 10:1, 5:1, 2.5:1 and 1.25:1. Control wells contained equal numbers of target cells alone (to assay, spontaneous chromium release) and targets lysed with 1% NP-40 (to assay total chromium release). Ascites containing IB 11 mAb or contrpl mAb was added to certain wells at a concentration of 1:100 v/v. After a 2 hour incubation, cells were pelleted and the supernatant form each well harvested and counted in a gamma counter. Percentage specific lysis was calculated for four replicates as follows: Percentage specific lysis = Chromium released in test well - spontaneous chromium released Total chromium released - spontaneous chromium released 29 11.24. Migration studies using FITC labeled activated lymphocytes 11.24.1. Determination of optimum FITC concentration Single cell suspensions were prepared from the spleen of BALB/c mice as previously described. Cells were incubated for 48 hours at 37°C, 5% C02, in RPMI containing 5 jig/ml Con A. After 48 hours, viable cells were separated over Ficbll-Hypaque, washed artd then labeled with concentrations of FITC ranging from 20 p.g/ml to 200 u.g/ml in RPMI. After 1 hour cells were washed 3 times with RPMI containing 5% FCS and resuspended in FACS buffer and subject to FACS analysis. 11.24.2. Half-life of FITC labeling The half life of FITC labeling in Con A activated splenocytes was determined by incubating activated splenocytes. prepared as above and labeled in the presence of 50 u\g/ml of FITC. At specific time intervals aliquots of cells were removed washed twice and resuspended in FACS buffer. FITC labeling was then determined inFACS analysis. 11.24.3. Dual staining of FITC labeled cells The specificity of FITC labeling was determined by performing FAGS analysis on FITC labeled cells prepared as above and then incubated with anti-CD4-PE and anti-CD8-biotin, followed by streptavidin-cychrome. In this manner, appropriate detector settings and compensation settings could be determined to ensure that the fluorescence readings were specific for the antibody used and that high levels of FITC labeling in cells did not affect detection of other fluorochromes. II.24.3. Lymphocytes migration studies in vivo • Lymphocyte migration studies were performed by modification of the methods of Buhrer (168). Splenocytes were activated with 5 uvg/ml Con A in RPMI, 5% FCS 20 |iM (3- . mercaptoethanol at 37°C in incubators gassed with 5% CO2 v/v for 72 hours. Con A activated 30 splenocytes were purified by centrifugation over Ficoll Hypaque and labeled for 1 hour in RPMI containing 50ug/ml FITC and 1 ug/ml a-methyl-mannoside. After 1 hour cells were washed three times with PBS and 5 x IO6 cells in PBS injected into BALB/c mice via.the tail vein. An aliquot of cells was retained for baseline FACS analysis. After 24 hours mice Were sacrificed and single eeli suspensions prepared from lymph nodes and spleen. Cells were then labeled with appropriate antibodies and subject to FACS analysis. Expression of CD4, CD8 and IB 11 cell surface antigens was determined in FITC and non-FITC labeled populations. 11.25. Deglycosylation methods II.25.1 Deglycosylation of 1B11 antigen using swainsonine/endoglycosidase H Viability of the D26 cell line in the presence of swainsonine was determined by establishing duplicate cultures at a density of 2 x 10^  cells per ml. in the presence or absence of 2 Jig/ml swainsonine. At set time intervals aliquots of cells were removed and counted. . ."• . Degylcosylation experiments were performed by incubating the D26 cell line in culture for 48 hours in the presence of 2 Ug/ml of swainsonine. After 48 hours cells Were pelleted, washed and lysed in standard lysis buffer. IB 11 antigen was then immunoprecipitated from the lysate as described previously using IB 11. mAb and Protein G. IB 11 immunoprecipitates were diluted initially into sample buffer containing 1% SDS and 5 mmol DTT (final concentrations). Samples were further diluted 5 times into 0.2M citrate buffer pFI 5.5. 1U of endoglycosidase H was added per 25 ul sample to one of duplicate, samples and the immunoprecipitates were incubated overnight at 37°C. At the end of the incubation period 10% ice cold TCA was added to each sample and the samples incubated a further 48 hours at 4°C. TCA precipitates were then pelleted and washed with acetone and vacuum dried. Samples were resuspended in sample buffer and resolved on a 7.5% SDS PAGE gel. 1B11 immunoreactivity was then determined by immunoblotting. 31 11.25.2 Deglycosylation of IB 11 antigen using N-glycanase Whole cell lysates were prepared as described previously from C(33, a'CTL cell line transfected with a construct which expresses the IL-3 receptor (3 sub-unit AIC-2A. IB 11 antigen was immunoprecipitated from the lysates using Protein G and IB 11 mAb as previously described. AIC-2A was immunoprecipitated from lysates using 4F1 mAb.and Protein G and was used as positive control. Immunoprecipitates were de-salted over a G25 column and buffer exchanged into 15 mmol Na3P04, pH 7.5. Immunoprecipitates were then boiled in the presence of 10 mmol phenanthrolene hydrate, 50 mmol P-mercaptoethanol and 1% SDS. Samples were then diluted into 1% NP-40 and 50U/ml N-glycanase was added to both samples and the samples incubated overnight at 37°C. Following incubation, digests were diluted into SDS sample buffer and resolved on 7.5% SDS PAGE gels. IB 11. antigen was detected in immunoblots using 1B11 mAb and AIC-2A was detected in immunoblots using an anti-peptide antibody. 11.25.3 Removal of O-linked glycosylation CP3 cell line was labeled overnight with 0.1mCi/ml35s-methionine. Whole cell lysates were prepared as described previously. IB 11 antigen was immunoprecipitated from lysates using Protein G and IB 11 mAb as described previously. As a positive control, CD45 was immunoprecipitated from cell lysates using protein G and rat monoclonal anti-CD45 antibody, 13/2. Immunoprecipitates were incubated with 20 MU neuraminidase in a reaction volume of 50 ul in the presence of 10 mmolar calcium acetate. After 1 hour incubation at 37°C, immunoprecipitates were desalted over a G25 column and buffer exchanged into 15 mM Na3P04, pH 7.5. 2 mU O-glycanase was then added to 30 |ll of the digest and the reaction incubated overnight at 37°C. Following incubation with O-glycanase, samples were made up in sample buffer and resolved in duplicate on 7.5% SDS PAGE gels. One duplicate gel was amplified, following staining of molecular weight markers, and subjected to autoradiography. The second gel was transferred to nitrocellulose and probed with IB 1 ImAb and anti-CD45 mAb. 32 II.25.4 Neuraminidase treatment of whole cells EL-4 cells or splenocytes derived from BALB/c mice were washed and resuspended at 10^  cells per ml. in Hanks balanced salt solution in the presence of 10 mmol/l calcium acetate, 20 mmol/l Tris pH 6.8 arid 55 mU of neuraminidase. Cells were incubated for 1 hour at 37°C and then washed and resuspended in FACS buffer. Treated and mock treated cells were subjected to FACS analysis using FITC-labeled Control mAb, FITC-conjugated WGA, FITC-labeled IB 11 mAb or FITC-labeled S7 mAb. In additional experiments,.cells were incubated as above at pH 6.0 in the presence of 200U/rnl Newcastle Disease neuraminidase and then subject to FACS analysis. 33 III. Production and Characterisation of the IB 11 Monoclonal antibody III.l. Introduction and aims The 1B11 antibody was produced in this laboratory by immunisation of male Wistar rats with the WEHI 274.3 cell line, a myeloid leukaemia. The original purpose of these immunisations had been to produce an antibody which would be reactive with the GM-CSF receptor, however the pattern of expression of the 1B11 antigen was clearly distinct from that of the GM-CSF receptor. The antibody was reactive with a 130 kDa protein, could be used in Western imrmmoblots and was also able to immunoprecipitate its target antigen from cell lysates, The nature of the target antigen and its function however remained obscure. The following chapter describes the purification and characterisation of the mAb IB 11 and determines the optimum conditions for its use in Western immunoblotting, in FACS analysis and in immunoprecipitation; techniques which were subsequently employed to determine the structure and function of the target antigen. 34 IH.2. Results III.2.1 1B11 mAb can be purified from ascites on a sheep anti-mouse immunoglobulin column Typical results from purification of IB 11 mAb on a sheep anti-mouse immunoglobulin column are shown iri Figure 1. Passage of diluted ascites over the column resulted in 80-90% depletion of antibody immunoreactivity, with a recovery of approximately 30-50% of antibody immunoreactivity, (accounting for concentration effects). Typically, 1ml of ascites yielded 3-5 rhg of purified antibody, with a titre of 1:100,000 -1:1,000,000 in ELISA against sheep anti-mouse immunoglobulin antibody bound to the solid phase. 111.2.2. IB 11 mAb preparations do not differ in their ability to detect their target antigen in Western immunoblots of D26 whole cell lysates We attempted to standardise IB 11 antibody preparations for use iri immunoblotting by titrating total protein loaded per lane in Western Immunoblots against a standard concentration of IB 11 mAb. IB 11 antibody preparations were used at a concentration of 2.5 Ug/ml. A faint IB 11 reactive signal was observed when as little as lug of total protein from a whole cell lysate of the D26 cell line was loaded on a SDS PAGE mini gel, however the 1B11 antigen was most reliably detected when 10 ug of total protein was loaded. Surprisingly, there was very little variability between IB 11 mab preparations purified'from ascites. A sample titration of IB 11 mab immunoreactivity is shown in Figure 2. . 111.2.3. Determination of the optimum IB 11 antibody dilution for FACS analysis We determined the optimum dilution of a 1 mg/ml solution of FITC-labeled IB 11 mAb to. use in FACS analysis, by examining binding of serial dilutions of FITC-lBll mAb to the D26 cell line and comparing the immunofluorescence observed with the binding of an equal concentration Of control antibody. Immunofluorescence was determined for a gated population of viable cells, '-i 35 Figure 1. Purification of IB 11 mAb on sheep anti-mouse immunoglobulin columns. o o tn 3 Dilution Monoclpnal antibodies were purified from ascites fluid by passage over sheep anti-mouse immunoglobulin coupled Sepharose beads. Antibody was eluted from the column using 0.1M glycine pH 2.5 and neutralised with saturated Tris base. Antibody titers were determined by binding of serial dilutions of antibody to sheep anti-mouse immunglobulin in the solid phase. Bound antibody was detected using a peroxidase coupled goat anti-rat immunoglobulin antibody to catalyse a colour change in ABTS. Absorbance of the coloured product was determined using an ELISA reader. Typical results are shown. Q Ascites £~J Column drop through. ^ Eluate. Relative yields were calculated based dn antibody titer and volume of dilution of initial sample. 36 Figure 2. IB 11 mAb reliably detects its target antigen in lOjig of total cellular protein. Whole cell lysate prepared from D26 cell line was standardised by Bradford assay with respect to protein content and resolved on 7.5% SDS PAGE gels. IB 11 mAb at 2.5 Ug/ml was used to probe immunoblots for IB 11 irnmunoreactivity. Batches of IB 11 mAb prepared from ascites were compared with respect to their ability to detect IB 11 antigen in immunoblots. Typical results are shown. Lane a 100 |lg total cellular protein. Lane b 10 ug total cellular protein. Lane C 1 |ig total cellular protein. 47 viability being determined by patterns of forward scatter (FSC) and side scatter (SSC). Results are shown in Figure 3. At dilutions of 1:1000 (v/v) only minor binding of FITC-IB 11 mAb to D26 was observed compared to control antibody. At dilutions of 1:10 optimum differentiation of control and 1B11 mAb binding occurred, however very few cells out of the total population had the expected FSC/SSC profiles. Subsequent analysis files of all labeled cells revealed that the majority of the labeled cells had formed aggregates which were sufficiently large as to be excluded from the gates set to capture this population. A small amount of high non-specific binding was seen with the control antibody at this concentration, possibly due to Fc receptor binding. The dilution of 1:100 of FITC-1B11 stock was chosen as optimum as this provided an excellent yield of labeled cells within the defined gate and showed clear separation of FITC-control and FITC-1B11 labeled cells. Levels of control antibody binding were also acceptable. The 1:100 dilution of FITC-IB 11 was verified using primary cells and other cell lines (data not shown) and was used in subsequent experiments. Purified 1B11 antibody preparations showed very little variability with respect to binding iat this dilution (data not shown).. III.2.4. Determination of the optimum conditions for immunoprecipitating IB 11 antigen. We determined the optimum conditions for using IB 11 mAb to immunoprecipitate IB 11 antigen from the D26 cell line. IB 11 antigen was immunoprecipitated from 35S-labeled whole cell lysates of D26 using either Protein G Sepharose, mouse anti-rat kappa coupled Sepharose or IB 11 mab directly conjugated to Sepharose beads.' Immunoprecipitates were resolved on SDS PAGE and autoradiographed. Experiments were repeated on 3 occasions. Typical" results are shown in Figure 4. Immunoprecipitates using IB 11 mAb bound to Protein G were the most efficient in recovering 1B11 antigen, and these immunoprecipitates were associated with lower background contamination. A small amount of the IB 11 antigen was immunoprecipitated using the IB 11 38 Figure 3. Titration of I B 11 immunoreactivity on D 2 6 cell line: FAGS profiles To determine optimum dilutions of 1B11 mAb for use in FACS analysis, D26 cells were labeled for 40 minutes at 4°C in the presence of serial dilutions of FITC-labeled IBM mAb or FITC-labeled isotype matched control antibody diluted in PBS/5% FCS. Celis were then washed and subject to FACS analysis. Batches of IB 11,purified from ascites and conjugated with FITC were compared with respect to binding to D26, Typical data are shown. A 1'ixg/ml, BIO |ig/ml and C 100 p:g/ml Binding is shown relative to control antibody. Figure 3 control Fluorescence control c Fluorescence 4 0 Figure 4 a b c d e 35S-labeled D26 whole cell lysates were prepared as described previously. Lysates were pre-cleared with 20 (ll of a slurry containing 30% v/v Protein G Sepharose beads or Sepharose beads. Lysates were then incubated at 4°C with A 10 ug 1B11 mAb followed by mouse anti-rat IgG coupled Sepharose beads: B IB 11 coupled Sepharose beads: C 1F10 coupled Sepharose beads: D 10 ug control antibody followed by Protein G Sepharose beads: or E 10 (lg 1B11 mAb followed by Protein G Sepharose beads: After 1 hour, beads were pelleted, washed three times with wash buffers as described and bound antigen was eluted with diethanolamine, pH 11.5 containing 0.05% NP-40. Eluates were diluted into SDS sample buffer and resolved on 7.5% SDS PAGE gels. Typical results are shown. 200 kDa 115 kDa 98 kDa 41 conjugated beads and little or ho detectable IB 11 antigen was immunoprecipitated using, the mouse anti-rat kappa conjugated beads. Protein G immuno-precipitations were used subsequently in all experiments. • . ; " , 111.2.5. 1B11 mAb isotyping 1B.11 mAb was isotyped using the Gibco rat isotyping kit and was determined by ELISA to be an IgG2a isotype with a k light chain (data not shown). 42 Ill .3. Discussion IB 11 mAb was readily purified from ascites using sheep anti-mouse immunoglobulin columns by taking advantage of the cross-reactivity between mouse and rat immunoglobulins. Although excellent depletion of ascites preparations was observed, yields were low, possibly due to loss of antibody by precipitation during dialysis! Purified antibodies showed very little variability in their ability to detect IB 11 antigen either in Western immunoblots or by FACS analysis. In D26 whole cell lysates IB 11 antigen could be detected when as little as 1 fig of total protein was loaded on SDS PAGE gel. The IB 11 antigen is very abundant in the D26 cell line and in other cell lines it was not possible to detect IB 11 antigen when such small amounts of protein were resolved. This abundance on the D26 cell line together with the presence of high numbers of Fc receptors, probably lead to the formation of the large aggregates of cells which were observed when FITC-1B11 was used at high concentration in FACS analysis. This phenomena was not apparent when FACS was performed on lymphocytes (data not shown) Immunoprecipitation was most successful when IB 11 mAb was used together with Protein G Sepharose beads. Protein G is preferable to Protein A for use with the IB 11 mAb because IB 1 lmAb is a rat IgG2a isotype and will not bind directly to Protein A, Protein A can be used to immunoprecipitate a rat IgG2a antibody if a secondary antibody is used. For the purposes of these experiments, we felt that the immunoprecipitates would be more efficient if a single antibody step were used and in addition there would be less antibody contamination of the eluates. IB 11 coupled Sepharose beads were largely unsuccessful in immunoprecipitating IB 11 antigen. This problem appears to have arisen because the conformation of the coupled antibody was unsatisfactory, since the coupling procedure itself appears to have been complete as evidenced by a decrease in the 1B11 antibody concentration in the supernatant during the coupling process. Likewise, coupling of mouse anti-rat kappa antibody to Sepharose beads appeared technically .successful but was equally ineffective in subsequent immunoprecipitations. In this case, there may also have been a problem with the conformation of the IB 11 mAb as IB 11 antibody appears to 43 show low reactivity with mouse anti-rat kappa antibody on affinity columns which are successful in purifying other rat.monoclonals (data not shown). Isotyping of IB IT mAb revealed that it was an IgG2a isotype, therefore the IgG2a rat mAb, 1F10, was selected as a control antibody for FACS analyses and for immunoprecipitations This antibody recognises the idiotype of a rabbit anti-LL3 antibody and showed no detectable cross-reactivity with cell surface antigens. 44 IV. Distribution of the IB 11 antigen IV.l . Introduction and aims We were unable to correlate the molecular weight and distribution of the IB 11 antigen with any previously reported protein. Therefore we believed that IB 11 mAb recognised a unique cell surface antigen and accordingly we decided to characterise its pattern of expression. This chapter . describes experiments performed to determine the distribution of the IB 11 antigen both in cell lines and in primary tissues. Particular emphasis has been placed on IB 11 antigen expression in T-lymphocytesand their subsets where 1B11 antigen immunoreactivity is developmentally regulated. Comparison is made with other major cell surface antigens expressed by T-lymphocytes including CD43, CD44 and CD45. These cell surface aritigens were chosen as they are expressed by lymphohemopoietic cells and exist in multiple isoforms, thus making them candidates for the identity of the 1B11 antigen. Experiments were performed Using Western immunoblot and FACS analysis. 45 IV.2. Results IV.2.1. 1B11 E x p r e s s i o n is confined to lymphohemopoietic cells 1B11 antigen expression was determined in cell lines and in primary tissues by Western immunoblot and FACS analysis. The distribution and molecular weight of the I B M antigen in whole tissue and in cell lines is summarised in Table 4. 1B11 antigen expression was confined to lymphohemopoietic tissue and derived cell lines. In lysates prepared from whole organs, low levels of IB 11 antigen expression were detected in thymus, spleen and lymph nodes and higher levels in bone marrow. No signal was detected in kidney, liver, lung or muscle. A high molecular weight form of 1B11 antigen with a molecular weight of 150 kDa was detected in placenta. IB 11 recognised, by immunoblotting, a single protein band at .130 .kDa in most T-helper cell lines. In some preparations a second band at 95 kDa was detected. We believe this to be a proteolytic fragment. In C T L cells a single band was detected at 140 kDa. The 1B11 antigen was expressed at high levels by most T cell lines, with the exception of CT-4S and EL4, by all myeloid cell lines tested and by one of the B-cell myeloma cell lines, NS-1, but not by cell lines representing less mature B-lymphocytes, specifically, Bai 17 and WEHI 231. The 1B11 antigen was not expressed by cultured endothelial cells, fibroblasts or embryonic STO cells. IV.2.2. 1B11 antigen is differentially expressed by T - l y m p h o c y t e sub-sets i n l y m p h o i d tissue in vivo . FACS analysis was performed on resting lymphocytes derived from thymus spleen, lymph nodes and whole blood. Only viable cells were examined. Sub-populations of lymphocytes were gated according to CD4 and CD8 profiles in triple stained preparations. Analysis of sub population expression of IB 11 antigen in the thymus (Figure 5) revealed high levels of expression on the CD4-8- population (double negative thymocytes) and on the CD4-8+ population, intermediate to high levels of expression of IB 11 antigen were observed on CD4+8+ population (double positive thymocytes) and low levels of expression were observed in the CD4+8- population. The mean 46 Table 4 : IB 11 Expression in Tissue and Cell Lines T i s s u e / T y p e " I m m u n o b l o t '•• -FACS Cell Line Lymphohaemopoietic Tissue Spleen 130 kD ++ Lymph node 130 kD + Thymus 130 kD . ++ Peritoneal Lavage 130 kD ++ HT-2Y T-cell . .130 kD . .+++ E9D4 . 130 kD ND HDK-1 130 kD + . CT-4S nil ' ND EL4 nil -/ + CTLL 150 kD ++++ Clone 653 B-cell nil +./-. NS-1 ND + WE 231 ' nil -Bai 17 nil -R6X Mast cells 130 kD +++ D26 . 130 kD ++++ MC9 130 kD ND P815 nil + WEHI 274.3 Myeloid 150 kD ND NSF-60 130 kD ND FDCP-1 130 kD ND Embryonic tissue : 12 day embryo nil ND Placenta . • , 150 kD : ND Embryonic stem cells nil ND M e s e n c h y m a l Kidney nil • ND Liver nil ND Muscle - . ' V nil ND' Swiss-3T3 Fibroblast nil ND Endo-Dl Endothelium nil ND 47 Figure 5 1B11 antigen is differentially expressed in thymus. Single cell suspensions were prepared from BALB/c thymus as described. Cells were triple stained using FITC-labeled IB 11 mAb or FITC-labeled control antibody, PE-conjugated anti-CD4 mAb and biotinylated anti-CD8 mAb followed by streptavidin cychrome. IB 11 immunoreactivity was determined within gated subpopulations and expressed as a histogram relative to binding of control antibody within the same.subpopulation. A CD4/CD8 profiles showing gated subpopulations: IB 11 immunoreactivity in B CD4+CD8-population: C CD4+8+ population: D CD4-8+population: E CD4-8-population: : " 48 Figure 5 4 9 fluorescent intensity (MFI) of IB 11 mab binding was consistently 10-20 fold less on CD4+8-thymocytes compared with CD4-8+ thymocytes. In lymph node (Figure 6), spleen (Figure 7) and whole blood (data not shown), this differential expression of the 1B11 antigen between CD4 and CD8 lymphocytes was maintained and in some experiments, IB 11 immunoreactivity on CD4+ lymphocytes, was comparable with control values. In lymph node and spleen, the GD4-8- population is comprised largely of macrophages and B-lymphocytes and showed very low levels of immunoreactivity with IB 11 mab. The CD4+8+ population is essentially absent from peripheral lymphatic tissue. Levels of IB 11 antigen expression were consistent through multiple experiments and data from 5 such experiments is summarised in Table 5. Similar results were obtained for BALB/c, DBA/2 and C57BL6 mice (data not shown). IV.2.3. 1B11 is expressed at very low levels on B-lymphocytes Levels of IB 11 expression were determined by FACS analysis of labeled splenocytes. B-lymphocytes were identified by expression of the B-lymphocyte associated isoform of CD45, B220, which was detected using phycoerythrin-labeled anti-B220 antibody (Pharmingen). Between 20 and 25% of B220 positive cells in the spleen were positive for IB 11 antigen expression. (Figure 8), however less than 5% of IgM positive cells express IB 11 antigen. We suspect that the majority of cells which express IB 11 antigen are IgG positive plasma cells as is suggested by the results found in cell lines. Attempts td demonstrate this in vivo by FACS have proved difficult because of the. high degree of binding to IB 11 mAb by our labeled anti-IgG antibody (even in the presence of excess of unlabeled antibody) and because of the low frequency of IgG positive cells. 50 Figure 6 1B11 antigen is differentially expressed in lymph node. Single cell suspensions were prepared from BALB/c lymph nodes as described. Cells were triple stained using FITC-labeled IB 11 mAb or FITC-labeled control antibody, PE-conjugated anti-CD4 mAb and biotinylated anti-CD8 mAb followed by streptavidin cychrome. , IB 11 immunoreactivity was determined within gated subpopulations and expressed as a histogram relative to binding of control antibody within the same subpopulation. A CD4/CD8 profiles showing gated subpopulations. B CD4+CD8-population: C CD4-8-population: D CD4-8+ population: ' ' 51 Figure 6 52 Figure 7 IB 11 antigen is differentially expressed in spleen. Single cell suspensions were prepared from BALB/c spleen as described. Cells, were triple stained using FITC-labeled 1B11 mAb or FITC-labeled control antibody, PE-conjugated anti CD4 mAb and biotinylated anti CD8 mAb followed by streptavidin cychrome.' IB 11 immunoreactivity was determined within gated subpopulations and expressed as a histogram relative to binding of control antibody within the same subpopulation. A CD4/CD8 profiles showing gated subpopulations: B CD4+CD8-population: C CD4-8-population: D CD4-8+population. 53 Figure 7 54 Table 5. IB 11 immunoreactivity in thymus and spleen relative to control antibody A Thymus Cell Type % Positive +/-SE CD4+8- 23.3 +/- 3.3 CD4+8+ 74.0.+/- 6.0 CD4-8+ 71.0 +/- 7.8 CD4-8- 77.0 +/- 4.8 B Spleen Cell Type. % Positive +/-SE CD4+ 14.5 +/- 2.6 CD4-8- 31.0 +1-2.9 CD8+ 74.3 +/- 5.0 55 Figure 8. 1B11 antigen is expressed at very low levels by B-lymphocytes in spleen and lymph nodes. Single cell suspensions were prepared from BALB/c spleen and lymph nodes and subject to FACS analysis. Cells were labeled with either A FITC labeled anti-IgM and PE labeled anti-B220 mAb or B PE-labeled anti-B220 and FITC-labeled 1B1 ImAb or C FITC labeled anti IgM and PE-labeled IB 11 mAb. Binding of all antibodies was compared with an isotype matched control antibody. Typical results are shown. .56 Figure 8 <:T. c 1CT 10" 10" o o, o C N o o. •aS«A'S*"V.!:»' > I' ***** ~A • . V r v 10" 10' 1B11 • I ' T ' M T W f " 102 10" 10" CO . o CM O . 57 IV.2.4. The pattern of expression of 1B11 immunoreactivity differs from that of CD43, CD44 and CD45 The distribution of IB 11 antigen.expression within T-lymphocyte subsets is described above. We compared this distribution with known antigens CD43, CD44 and CD45, which are widely expressed'within the lymphohemopoietic system and which are known to exist in multiple isoforms. The results of analysis of CD43, CD44 and GD45RA and CD45RB isoform expression in gated populations of thymocytes and splenocytes are summarised in Table 6. The patterns of expression of these antigens were,distinct from that of IB 11 antigen. In the thymus, anti-CD43 antibody preferentially stained the more mature CD4+8- and CD4-8+, single positive cells and showed low levels of reactivity with CD4+8+ and CD4-8- cells. Differential labeling of CD4+8- and CD4-8+ was not observed. Anti-CD44 mAb labeled mature thymocyte subsets with approximately equivalent levels of intensity. Slightly lower levels of immunoreactivity were observed in immature subsets. Very low levels of labeling were observed using the anti-CD45RA antibody, with only a small percentage of the CD4-8- sub-set expressing the antigen at levels above control. The anti-CD45RB antibody was reactive with the majority of all thymocyte sub-sets with higher levels of immunoreactivity being observed in the CD4-8+ and CD4-8-populations. In the spleen, anti-CD43 antibody labeled both CD4+ and CD8+ with approximately the same intensity, with slightly higher MFI observed in the CD8+ sub-set; Anti-CD43 binding by the CD4-8- subset was low. Anti-CD44 and anti-CD45RB antibodies labeled all three subsets with the same intensity. Anti-CD45RA antibody showed high levels of labeling in the CD8+ population, low levels of labeling in the GD4-8- population and identified two sub-populations within the CD4+ population, one of which showed a high level pf labeling and one which showed no labeling. - , These patterns of distribution are distinct from that observed for 1B11 immunoreactivity. The IB 11 antigen does not therefore appear to be identical to any of these known antigens. 58 Table 6. IB 11 immunoreactivity on thymus & spleen: Comparison with CD43, CD44, CD45 isoforms Subset 1B11 CD43 CD44 CD45RA CD45RB CD4+8- - + + - + CD4+8+ + - - + CD4-8+ + + + - • + CD4-8- +/- + - + Subset 1B11 CD43 CD44 CD45RA CD45RB GD4+8- - + + + + CD4-8- - - + - + CD4-8+ + + + + + 59 IV.2.5. IB 11 antigen is expressed by activated macrophages and by natural killer (NK) cells Activated macrophages and NK cells were derived by immunising C57/Black 6 mice with the P815 mast cell line. The resultant inflammatory peritoneal exudate was then harvested and 1B11 antigen expression determined by FACS analysis. Macrophages were identified by binding of the mAb Ml/70 which recognises Mac-1 antigen. NK cells were identified using the PK136 antibody which recognises the NK-1.1 antigen. Both activated macrophages and NK cells bound 1B11 mAb (Figure 9). . ' IV.2.6. T-lymphocytes expressing the gamma-delta T cell receptor show high levels of 1B11 immunoreactivity Mature thymocytes express one of two forms of the T cell antigen recognition receptor (TCR). The majority express a TCR which is composed of an alpha and a beta chain (ocp-TCR). A minority express a form which is composed of a gamma and a delta chain (y8-TCR). T-lymphocytes expressing the y8-TCR comprise only a small percentage (0.5% or less) of total thymocytes. Determination of IB 11 antigen expression in such a small population proved problematic and therefore we decided to take advantage of the selective growth advantage of y8 T-lymphocytes in culture with IL-2. Single cell suspensions of thymocytes were incubated for 96 hours in standard tissue culture medium containing 10 U/ml IL-2. At the end of the incubation period, FACS analysis was performed on triple stained thymocytes. In addition ccp-TCR and y8-TCR expressing cells were identified by binding of biotinylated anti-a(3 TCR and anti-y8 TCR antibodies respectively. Prolonged culture of single cell suspensions of thymocytes resulted in an increase in the relative proportions of double negative thymocytes to 58% of total. Results are shown in Figure 10A. Examination of the flow profiles of the cultured thymocytes revealed two distinct populations (Figure 10B). The first population comprised smaller,.less,granular cells and was . composed predominantly of the mature single positive thymocytes (Figure 10C). The second Figure 9 iB l l antigen is expressed by activated macrophages and NK cells in peritoneal lavage. C57 Black 6 mice were injected with either 2-5 x IO 7 P815 cells i.p or ah equal volume of PBS, After 10 days lavages were harvested and subjected to FACS analysis. Macrophages were identified by expression of Mac-1 antigen and NK cells were identified by expression of NK-1.1 antigen. IB 11 immunoreactivity was determined in labeled populations and compared to binding of control antibody. A Mac-1 positive cells from P815 lavages: B Mac-1 positive cells from control lavages: C NK-1.1 positive cells. Typical results are shown. 61 Figure 9 A 3" IBl'l 62 Figure 10. 1B11 is expressed at high levels by TCR-receptor positive, large, granular CD4-8- cells in IL-2 cultured thymus. Single cell suspensions were prepared from BALB/c thymUs and incubated,at 5 x 10^  cells/ml for 4 days in the presence of 10 U/ml fL-2. Cells were then washed and labeled at 4°C with PE-conjugated anti CD4 mAb, bidtinylated anti CD8 mAb, followed by streptavidin conjugated Cychrome. FACS profiles for IL-2 treated thymocytes A.: Identification of two subsets based on flow profiles B: CD4 and CD8 expression within small thymocytes C and large thymocytes D:. Double negative thymocytes were then enriched from IL-2 cultures by selective depletion of CD4+ and CD8+ thymocytes using anti-CD4 mAb and anti-CD8 mAb and sheep anti-mouse conjugated Dynabeads. Efficiency of depletion was determined by repeat labeling with PE-conjugated anti-CD4 and biotinylated anti-CD8, followed by streptavidin conjugated cychrome E and by PE-conjugated anti-rat IgG F.. Purified CD4-CD8-thymocytes were then labeled for 40 minutes at 4°C in the presence of FITC-IB 11 mAb and PE conjugated anti-y5 TCR mAb G or FITC-IB 11 and PE-conjugated anti-a(3 TCR mAb H. Typical results are shown. 63 Figure 10 1B11 1B11 6 4 population contained cells which were larger and more granular, and was composed predominantly of CD4-8-cells (Figure 10D). Depletion of single positive and double positive thymocytes using rhiagnetic beads resulted in a population which was 95% pure (Figures 10E). This sorted double negative population contained 25-30% y8 TCR expressing cells ;. 85-87.5% of these cells expressed IB 11 mAb immunoreactivity at levels above control (Figure 10F). Interestingly, the predominant phenotype (53%) of the selected double negative thymocyte population was an ocfj TCR expressing cell. These cells showed a bimodal pattern of IB 11 antigen expression with 50-80% of cells expressing very high levels of IB 11 mAb immunoreactivity (Figure 10G) with an MFI consistently of 10^ -lO .^ This is far higher than previously observed for unmanipulated primary cells. IV.2.7. 1B11 mab does not cross-react with human T-lymphocytes. Peripheral lymphocytes from human blood were stained with FITC-IB 11 mab. No demonstrable binding of the fluorescent antibody was observed (data, not shown). 65 IV.3. Discussion We have demonstrated that the I B 11 antigen is restricted in its expression to the lymphohemopoietic system with high levels of expression in T-lymphocytes and cells-of myeloid lineage. This restricted pattern of expression would suggest that the.IB 11 antigen performs a specific role in the generation of the immune response and is not a generic structural protein. The unusual pattern of distribution of I B 11 antigen expression within T-lymphocytes subsets would suggest that this function is highly specialised. . , • Within the thymus, 1 B 1 1 antigen is expressed at high levels by C D 4 - 8 - , C D 4 + 8 + and C D 4 - 8 + thymocytes, but at low levels by C D 4 + 8 - thymocytes. These sub-populations represent different stages in the.maturation of T-lymphocytes. 1 B 1 Tantigen is expressed predominantly by immature thymocytes but curiously is preferentially downregulated on C D 4 + 8 - thymocytes following maturation from the C D 4 + 8 + stage. We were unable to find a similar precedent in the literature, although expression of the C D 4 5 R A and C D 4 5 R B antigens is reported to be skewed towards CD4-8-and CD4r8+thymocytes (29) . In peripheral lymphatic tissue, I B 11 antigen appears to be down-regulated further in C D 4 + lymphocytes, whereas in C D 8 + lymphocytes, I B 11 antigen expression is increased. This subsequent modulation of I B 11 antigen expression may reflect further maturation of the single positive thymocytes. A similar change in C D 4 5 isoform expression has been reported following egress from the thymus. Similarly there is precedence for differential expression of adhesion molecules by C D 8 + lymphocytes ( 1 6 9 ) . Up-regulation of I B 11 antigen expression, in C D 8 . lymphocytes may reflect a distinct migration pattern. It has been reported that C D 8 + lymphocytes show homing patterns which differ from those pf C D 4 + lymphocytes"(170). Alternatively, the I B 11 antigen may be relevant to the function of C D 8 + lymphocytes i.e. cytotoxicity, and its preferential expression on C D 8 + lymphocytes may reflect this function. Limited cytotoxic potential has been demonstrated for immature thymocytes ( 1 7 1 ) and I B 11 expression in these populations may be consistent with this function! The increase in I B 11 antigen 6 6 expression on CD8+ lymphocytes following migration from the thymus may reflect maturation of this function. . A third possibility is that CD8+ lymphocytes may interact with different antigen presenting cells (APCs) to CD4+ lymphocytes (172) and may therefore require different adhesion molecules. These molecules may also be involved in the interaction of immature thymocytes with the APCs responsible for negative and positive selection. There is evidence that the processing requirements for the development of mature CD4+8- and CD4-8+ thymocytes differ . This is in addition to the requirement for distinct MHC molecules. Petrie et al (173) have demonstrated that even in the presence of appropriate MHC molecules, CD4+8+ thymocytes will preferentially mature into CD4-8+ thymocytes in the absence of an additional signal from thymic stroma. Signals unique to the processing pathways of CD4+8- thymocytes might be responsible for the down-regulation of IB 11 antigen on these cells, perhaps mediated by an interaction with thymic stroma. . The pattern of IB 11 expression was compared with that of CD43, CD44 and 2 isoforms of CD45. These antigens were chosen for comparison as they are single chain molecules, whose expression is largely confined to lymphohemopoietic cells, and because they exist in multiple isoforms. In addition, some of these molecules show variability in expression between different mouse strains (174,175) and we considered it possible that the unique pattern of expression observed for IB 11 antigen might be a function of the mouse strain used for the experiments. CD43 is a 115 kDa surface glycoprotein present on all T-lymphocytes, early B-lymphocytes and cells of myeloid lineage (176,177). Its function is Unknown. It is known to exist in at least 2 distinct isoforms, generated by differential patterns of glycosylation (178). CD44 (Ly-24, Pgp-1) is a 80-200 kDa surface glycoprotein expressed on T and B lymphocytes, fibroblasts and erythrocytes (179-181). Considerable intra-strairi variability in CD44 expression is reported and BALB/c mice are reported to be high expressors. Lynch and Ceredig (174) found that roughly 56% of CD4+8- and 59.9% of CD4-8+ thymocytes are reactive with the Ly-24 antibody < whereas 32.7% of CD4-8- are Ly-24 positive. Expression in the CD4+8+ population was not reported. The pattern of CD44 expression observed is similar to our studies, however we report even higher levels of CD44 immunoreactivity in all thymocyte subsets. These 67 difference may have arisen as a result of the use of different anti-CD44 antibodies. We used directly biotinylated antiTPgp-1 whereas Lynch used unlabeled antibody with a secondary anti-rat antibody. In addition, Lynch derived her populations by negative antibody selection, whereas our studies were performed using triple stains of whole thymocyte preparations. These technical differences may explain the slightly different expression profiles observed in our studies Structural heterogeneity of the CD44 molecule is generated by differential glycosylation and exon usage (182-185). The CD44 molecule appears to function as an adhesion molecule, binding hyaluronic acid, fibronectin and collagen types I and IV (186,187). CD44 has been implicated in lymphocyte homing (188). CD45 is a membrane glycoprotein with intrinsic tyrosine phosphatase activity (189). It is present on T and B lymphocytes and myeloid cells in multiple isoforms, identified by mAbs as CD45RA; CD45RB and CD45RC (and CD45RO*)(190). mAb reactivity is determinedby relative exon usage and carbohydrate modification (189-191). In addition a peptide epitope, B220, is expressed almost exclusively on B-lymphocytes (192). The B-lymphocyte adhesion molecule CD22 (193,194) has been demonstrated to be a ligand for the CD45RO isoform of CD45 and possibly other isoforms(s). CD45 has been implicated in thymic selection events (195), T-lymphocyte signaling (189,193,196) and mast cell degranulation (197). Expression of CD45RA and CD45RB isoforms in the thymus is reported to be skewed toward CD4-.8- and CD4-8+ subsets, in a manner analogous to that seen with the IB 11 antigen (198), however CD45RA is expressed by very few thymocytes overall and CD45RB is reported to be expressed by almost all thymocytes (198). In addition the pattern of expression observed for these antigens in peripheral lymphatic tissue is distinct from that of IB 11 antigen. The patterns of expression of the CD43, CD44 and CD45 antigens determined in these studies conformed largely to those reported in the literature. In no case did the pattern of expression of one of these antigens correspond with that of IB 11 antigen. We also examined the level of 1B11 antigen expression on y8 TCR positive thymocytes. Gamma-delta TCR positive thymocytes comprise 0.5% of the normal thymus population, but appear to respond preferentially to LL-2 (199,200). We were able to enrich thymocyte preparations 68 for yS TCR positive cells by long-term culture in IL-2. The antigen specificities and functions of these T-lymphocytes are poorly understood. A small percentage (30%) express either CD8 or CD4 co-receptors when isolated from peripheral lymphatic tissue (201-203). In other respects, they are reported to express largely the same profile of adhesion molecules as a|3-TCR positive lymphocytes (204). In our studies approximately 85% of y8 TCR positive cells expressed the IB 11 antigen. This level of expression is comparable to that observed for CD8+ lymphocytes and is greater than that observed for CD4+ lymphocytes. Gamma-delta TCR positive lymphocytes have been reported to secrete a limited number of cytokines (204,205), to show non-specific cytotoxic properties (206) and to show distinct patterns of homing to skin and to small bowel associated lymphatic tissue (202). It is possible that. IB 11 antigen expression is integral to one of these functions. . IB 11 expression was also detected on NK-cells, activated macrophages and a small, number of B-lymphocytes. Expression of the IB 11 antigen on NK-cells and on macrophages would be compatible with a role in cytotoxicity, however the function of the antigen on B-lymphocytes remains obscure. In FACS analysis of spleen IBM staining in the double negative population was very low, suggesting that the majority of B-lymphocytes and most resting macrophages are not reactive with the antibody. IB 11 expression on surface IgG positive B-lymphocytes would be compatible with the findings in cell lines where the majority of B-cells were not reactive with IB 11 mAb, however a minority of hybridoma cell lines such as NS -1 show some IBM immunoreactivity. The expression of IB 11 antigen by activated macrophages and perhaps some resting macrophages is compatible with the observation of high IB 11 immunoreactivity in myeloid cell lines. Curiously in bone marrow, IB 11 mAb is reactive with larger proportion of both B-lymphocytes and myeloid cells (data not shown). This area is currently the subject of further investigation. In summary IB 11 antigen show a unique distribution on lymphohemopoietic cells. Differential expression within lymphocyte subsets suggests a possible role in cytotoxicity or in lymphocyte migration. 69 V. Regulation of 1B11 antigen Expression V.l. Introduction and Aims , In chapter IV, we described the distribution of the 1B11 antigen in the lymphohemopoietic system with emphasis on expression within T-lymphocyte sub-sets. In this chapter we outline regulation of IB 11 antigen expression in T-lymphocytes in response to various cytokines and to activation both in vivo and in vitro. The response of T-lymphocyte sub-sets in the thymus to JL-2 has already been described briefly with respect to expression by the yb TCR positive subset in chapter IV.2.6. Further details of this response are presented here. In addition we describe the effects of Con A stimulation, IL-lbc and IL-4 and PMA on thymocytes with respect to their expression of IB 11 antigen. The regulation of IB 11 antigen expression was also determined in various disease states. The model systems chosen to investigate IB 11 expression in vivo were 1) allograft rejection, 2)the MRLlpr/fas mouse and 3) a spontaneously arising B-cell lymphoma. The choice of these models was predicated on the findings of a distinct developmentally regulated pattern of antigen expression in T-lymphocytes as. described in chapter 4, The allograft model was chosen because of the high levels of IB 11 antigen expression on CD8+ lymphocytes compared to CD4+ lymphocytes. The MRL Xpr/lpr model Was chosen because of the high levels of expression of IB 11 antigen in the aP+CD4-CD8- lymphocyte population in the thymus. The lymphoma cell line was chosen because of the.higher levels of IB 11 antigen expression observed in immortalised cell lines in comparison to primary cells. This finding suggested that the IB 11 antigen may play a role in oncogenesis. 70 V.2. Results V.2.1. Regulation of expression of IB 11 antigen in response to T-cell activation IB 11 antigen expression by thymocytes and splenocytes cultured in the presence of either 5 |ig/ml Gon A or 20U/ml IL-2 was determined by immunoblotting of standardised whole cell lysates with 1B11 mAb. Results are shown in Figures 11-12. 1B11 antigen expression was increased in response to Con A activation of both thymus and spleen. 1B11 antigen expression increased with duration of activation and was most marked at 72 hours. Thereafter, levels of IB lLantigen expression remained unaltered in long term culture. There was a small delayed increase iri 1B11 antigen expression in response to IL-2 in thymus only. There was a small shift in the apparent molecular weight of the antigen recognised by IB 11 mAb following activation, suggesting that biochemical modification of the molecule was occurring. V.2.2 . IB 11 antigen expression within T-lymphocyte subsets FACS analysis of IB 11 expression within lymphocyte sub-sets following Con A activation revealed that the increase in IB 11 expression occurred primarily in the CD4+8- subset and that, IB 11 antigen expression by this subset increased with duration pf activation. (Figures 13 and 14). The increase in IB 1Tantigen expression occurred most rapidly in the thymus, but also occurred in the spleen, achieving maximum levels at 72 hours which were maintained for up to 7 days in culture (data not shown). Modest increases in the expression of the 1B11 antigen were observed in the CD4+8+, CD4-8- and CD4-8+ populations in both thymus and spleen. V.2.3 . 1B11 antigen expression in response to IL-2 Immunoblots of thymocytes cultured in IL-2 revealed a modest delayed increase in the level of IB 11 antigen expression. This was not apparent in spleen lysates. FACS analysis revealed that this increase was largely due to an increase in the expression of IB 11 antigen by the 71 Figure 11 IB 11 antigen expression is upregulated in thymus in response to Con A activation. — 200 kDa — 115 kDa — 98 kDa A B C D E F G Single cell suspensions were prepared from BALB/c thymus as described. Cells were incubated in either 10 U/ml IL-2 (Lanes B, D, F) or 5 ug/ml Con A (Lanes C,E,G). At 24 hour intervals aliquots of cells were removed and lysed in standard lysis buffer and resolved on 7.5% SDS PAGE gels. 40 \xg total cellular protein was loaded per lane. 1B11 mAb immunoreactivity was determined by immunoblotting. Control. Lane A: 24 hours. Lanes B & C: 48 hours Lanes D & E : 72 hours Lanes F & G: 72 Figure 12 IB 11 antigen expression is upregulated in spleen in response to Con A activation. — 200 kDa — 115 kDa — 98 kDa A B C D E F G Single cell suspensions were prepared from BALB/c spleens as described. Cells were incubated in either 5 ug/ml Con A (Lanes B, D , F) or 10 U/ml IL-2 (Lanes C , E , G ) . At 24 hour intervals aliquots of cells were removed and lysed in standard lysis buffer and resolved on 7.5% SDS PAGE gels. 40 jig total cellular protein was loaded per lane. 1B11 mAb immunoreactivity was determined by immunoblotting. Control. Lane A: 24 hours Lanes B & C : 48 hours Lanes D & E : 72 hours Lanes F & G . 73 Figure 13. IB 11 antigen is preferentially up-regulated on CD4+8- thymocytes in response to Con A. Single cell suspensions were prepared from BALB/c thymus. Cells were incubated for 48 hours in the presence of 5 lig/ml Con A. After 48 hours incubation, cells were washed and subjected to FACS analysis using FITC-labeled IB 11 mAb or FITC-labeled control antibody, PE-labeled anti-CD4 and biotinylated anti-CD8 followed by streptavidin cychrome. IB 11 immunoreactivity was determined within gated populations of thymocytes and compared with control antibody labeling within the same population. FACS profiles on cells prior to Con A activation and thymocytes incubated for 24 hours in iL-2 were used as controls. FACS profiles are shown for IL-2 treated resting thymocytes (A-D) and Con A activated thymocytes ( E - H ) . IB 11 antigen expression is shown relative to control antibody reactivity in CD4+8- (A & E ) : CD4+8+(B & F): CD4-8+(C & G) and CD4-8-(D & H). .74 Figure 13 o Z 13 U [•Jo f6» 162 IB 11 B o Z 13 U 1B11 o Z U u , a I M W H J KV> :S« '.a* 10 1B11 D o Z U 1B11 E o z 13 U l i 1 lw- lu1 lo' 1B11 o Z 13 U 5™» 42 i o 3 ii" 1B11 H o Z 13 U IB 11 l„,l„imni )H„ o Z 13 U 1B11 75 Figure 14. 1B11 antigen is preferentially up-regulated on CD4+8- splenocytes in response to Con A. Single cell suspensions were prepared from BALB/c spleen as described. Cells were incubated for 48 hours in the presence of 5 |ig/ml Con A. After 48 hours incubation, cells were washed and subjected to FACS analysis using FITC-labeled IB 11 mAb or FITC-labeled control antibody, PE-labeled anti-CD4 and biotinylated anti-CD8 followed by streptavidin cychrome. IB 11 immunoreactivity was determined within gated populations of splenocytes and compared with control antibody labeling within the same population. FACS profiles on cells prior to Con A activation and splenocytes incubated for 24 hours in IL-2 were used as controls. FACS profiles are shown for EL-2 treated resting splenocytes (A-C) and Con A activated splenocytes (D-F). IB 11 antigen expression is shown relative to control antibody reactivity in CD4+8- (A & D), CD4-8-(B & E) and CD4-8+( C & F). 76 Figure 14 CD4-8- sub-set of thymocytes and by an increase in the relative proportions of these thymocytes (Figure 15). This did not become apparent until after 48 hours of incubation. The phenotype of the high IB 11 expressing double negative thymocytes was further examined and they were found to comprise of 40%-70% ccp-TCR expressing cells and 30-47% y8-TCR expressing cells. Of the ap-TCR positive cells, 72% expressed the Vp8 TCR and 87% of cells expressed the CD3 molecule. No cells expressed either CD2 or B220 antigens (data not shown). V.2.4. 1B11 antigen expression in response to IL-lot, IL-4 and PMA No increase in the levels of IB 11 antigen expression could be detected in immunoblots of cells exposed to IL-1 a, IL-4 or PMA (Figure 16). No change in the molecular weight of the protein in response to these factors was observed. Analysis of IB 11 expression by thymocyte sub-populations revealed no changes in the levels of IB IT expression in any groups (data not shown). Splenocytes activated for 48 hour with 5 ug/ml Con A and then cultured in the presence of either lOU/ml IL-2, or lOU/ml IL-2 and lOU/ml IL-4, showed high levels of IB 11 immunoreactivity, however the FACS profiles of the two groups of cells were not significantly different (data not shown). . » V.2.5. 1B11 antigen expression in allograft rejection Previous data showed differential regulation of IB 11 expression between CD4+and CD8+ lymphocytes with CD8+ lymphocytes expressing higher levels of the antigen. This differential pattern of expression might suggest a role for the IB 11 antigen in the cytotoxicresponse, which is predominantly a function of CD8+ lymphocytes. In addition, we have previously demonstrated that IB 11 is an activation antigen. We therefore decided to examine expression of the IB 11 antigen during the cytotoxic response to allo-antigen. Peritoneal lavages were obtained from mice injected with either P815 cells or PBS (161) and were analysed by FACS with respect to 1B11 antigen expression within gated sub-popuiatidns. IB 11 expression was up-regulated on both CD4+ and CD8+ in response to allograft, but not in response to control injections. The increase in 78 Figure 15 1B11 antigen expression is upregulated on CD4-8- thymocytes cultured in IL-2. Single cell suspensions were prepared from thymus of BALB/c mice. Cells were incubated for 96 hours in the presence of lOU/ml of IL-2. At the end of the incubation period, cells, were washed with FACS buffer and subject to FACS analysis, using FITC-labeled IB 11 mAb or FITC labeled control mAb, PE-labeled anti-CD4 and biotinylated anti-CD8 followed by streptavidin conjugated cychrome. 1B11 immunoreactivity was determined within gated sub-populations and compared with binding of control antibody within the same population. FACS analysis was also performed on CD4-8- thymocytes selected by removal of CD4+ and CD8+ thymocytes using anti-CD4 mAb and anti-CD8 mAB and Dynabeads. Purity of the selected population was determined by use of a labeled secondary anti-rat IgG antibody to detect cells which had bound antibody but had not been removed by the Dynabeads and by re-labeling with anti-CD4 mAb and antiCD8 mAb. Purified CD4-8- thymocytes were then subject to FACS analysis using biotinylated anti-afiTCR mAb or anti CD3 mAb or anti-Vp8 F23.1 mAb and FITC labeled 1B11 mAb or control . A CD4/8 profiles: B GD4+8-subset: C CD4+8+subset: D CD4-8+subset: ECD4-8-subset: F-H Depleted CD4-8- thymocytes: F ap-TCR expression and 1B11 mAb: G V(38 TCR expression: H CD3 expression. 79 Figure 15 o u l\ 1 ' J VfJ8-TCR PQ CD3 30 Figure 16 Interleukin 4, interleukin 1 and PMA do not increase IB 11 expression on thymocytes. A 200 kDa » I 115 kDa a b c d e f g B 200 kDa 115kDa a b e d Single cell suspensions were prepared from BALB/c thymus. Cells were incubated at 5 x 106 cells/ml in either lOU/ml IL-2 or IL-2 and lOU/ml IL-4 for 72 hours or 1 ng/ml PMA or 1 ng/ml IL-1 for 24 hours. At 24 hour intervals aliquots of cells were removed and lysed in standard lysis buffer. Samples were resolved on 7.5% SDS gels. Samples were standardised such that 40 pig total cellular protein was loaded per lane. IB 11 immunoreactivity was determined by immunoblotting. Panel A: Lane a Control. Lane b 24 hours IL-2 alone. Lane c 24 hours IL-2 and EL-4. Lane d 48 hours IL-2 alone. Lane e 48 hours IL-2 and IL-4. Lane f 72 hours EL-2 alone. Lane g 72 hours IL-2 and IL-4. Panel B: Lane a Control. Lane b 24 medium alone. Lane c 24 hours IL-1. Lane d 24 hours PMA. 81 the level of IB 11 antigen expression appeared to correlate with an increase in the relative proportion of CD8+ to CD4+ lymphocytes, such that 1B11 expression by CD4+ cells was highest when the ratio of CD4:CD8 cells in the lavage was 1 or less (Figure 17). Within the CD4+ and CD8+ populations, IB 11 antigen was expressed on slightly larger, more granulated cells as determined by forward scatter and side scatter profiles respectively (Figure 18). The data suggests that 1B11 antigen is functioning as an activation marker in this model. P815 cells express little or no IB 11 antigen (data not shown). Increased expression of IB 11 antigen by both CD4+ and to a lesser extent CD8 lymphocytes was also demonstrated in the Graft-versus-Host model (207) and the increase in 1B11 antigen expression was shown to be confined to the donor lymphocytes. V.2.6.1 1B11 antigen expression is increased in autoimmune mouse model MRLlpr/fas and C57/Black6gW/gW mice develop a disease characterised by the development of massive lymphadenopathy in which the predominant cell population expresses the phenotype CD4-8-a(3-TCR+CD2-B220+ (208). The disease arises because of abnormalities in the process of deletion of auto-reactive T-lymphocytes (209). The presence of these abnormal T-lymphocytes accelerates the development of an autoimmune disease characterised by arthritis, glomerulonephritis and skin lesions. In light of the finding of high 1B11 antigen expression by a(3-TCR positive double negative thymocytes, we decided to investigate IB 11 antigen expression by Ipr/fas and BL6glaVgld mice and to compare the pattern of expression with age matched MRL +/+ and C57/Black 6 control mice in FACS analysis. Results of FACS analysis are shown in Figures 19-23. Results are shown for thymus and lymph node only, as spleen data was essentially identical to lymph node data, excepting that a bimodal distribution for 1B11 antigen expression was detected in the CD4-8- population, reflecting the smaller proportion of abnormal CD4-8-ap-TCR+ cells in the spleen. Lymph node data was compared for cervical, popliteal and para-aortic nodes. No significant differences were observed (data not shown). Overall levels of IB 11 antigen expression were increased in diseased MRLlpr/fas and BL6gld/gld mice compared with age matched controls in both thymus and lymph node. Within 82 Figure 17 Induction of allograft rejection is correlated with increased numbers of CD8+ lymphocytes and with increased IB 11 immunoreactivity in CD4+ lymphocytes in peritoneal lavage. 100 - i 80 H > •§. 60 a ,-40-^ -20 H o — i ~ r -T~ 2 3 4 Ratio CD4:CD8 5 C57 Black 6 mice were injected i.p. with either 2 x 10^  P815 cells or an equivalent volume of PBS. After 10 days mice were sacrificed and peritoneal mononuclear cells were harvested by irrigating the peritoneum with PBS and separating the lymphocytes by centrifugation over Ficoll-Hypaque. Cells were then washed and subject to FACS analysis using FITC-labeled 1B11 mAb, or control mAb, PE-labeled anti-:CD4 and biotinylated anti-CD8, followed by streptavidin-conjugated cychrome. Relative percentages of CD4 and CD8 cells were calculated and plotted against 1B11 immunoreactivity (expressed relative to control antibody binding) for each mouse. A P815 injected mice. S Control mice. 83 Figure 18. IB 11 immunoreactivity is associated with larger, more granular lymphocytes in allograft rejection. Ailoreactive peritoneal lymphocytes were prepared by immunising C57Black 6 mice with 2-4 x If/7 P815 cell line. After 10 days peritoneal lavages were harvested and subject to FACS analysis using FITC-labeled 1B11 mAb or control mAb, PE-labeled anti CD4 (Panel A) and PE-labeled anti-CD8 (Panel C). Flow profiles were determined for 1B1 l h i (blue) and 1B1 l l c (pink) lymphocytes within gated CD4+ (Panel B) and CD8+ (Panel D) populations. 84 Figure 19. 1B11 immunoreactivity in thymus of MRL+/+ and 'MRLlpr/fas mice. Single cell suspension were prepared from the thymus of MRL+/+ and MRLlpr/fas mice as described. Thymocytes were triple stained with FITC-labeled IBi ImAb (or Control mAb), PE-labeled anti-CD4 and biotinylated anti-CD8 followed by streptavidin cychrome. IB 11 immunoreactivity was determined within gated sub-populations and compared with control antibody binding (indicated respectively by Ml for MRL+/+ and M2 for MRLlpr/fas mice). Panel A CD4/CD8 profile for MRL+/+ thymus. Panel B CD4/CD8 profile for MRLlpr/fas thymus. Panel C IB 11 immunoreactivity in CD4+8- thymocytes. Panel D IB 11 immunoreactivity in CD4+8+ thymocytes. Panel E IB 11 immunoreactivity in CD4-8+ thymocytes. Panel F IB 11 immunoreactivity in CD4-8-thymocytes. 86 Figure 19 B •.- Hi"':. CD8 a " -l f i 0 101 l 6 2 l § 3 16H CD8 D +/+ lpr/lpr 1B11 o T—H U 1B11 F o U T63 16° lpr/lpr l 6 i • 1^ 2 1^3 t 6 H 1B11 1B11 8 7 Figure 20 IB 11 immunoreactivity in lymph node of MRL+/+ and MRLlpr/fas mice ,. Single cell suspension were prepared from the lymph nodes of MRL+/+ and MRLlpr/fas mice as described. Lymphocytes were triple stained with FITC-labeled lBllmAb (or Control mAb), PE-labeled anti-CD4 and biotinylated anti-CD8 followed by streptavidin cychrome. 1B11 immunoreactivity was determined within gated sub-populations and compared with control antibody binding (indicated respectively by Ml for MRL+/+ and M2 for MRLlpr/fas mice). Panel A CD4/CD8 profile for MRL+/+ lymph node: Panel B CD4/CD8 profile for MRLlpr/fas lymph node. Panel C IB 11 immunoreactivity in CD4+8- lymphocytes. Panel D IB 11 immunoreactivity in CD4-8- lymphocytes. Panel E IB 11 immunoreactivity in CD4-8+ lymphocytes. 88 Figure 20 B ' ' ie 1 ' IV CD8 Q CD8 Figure 21. IB 11 immunoreactivity in C57BL6gld/gld thymus. , \, Single cell suspension were prepared from the thymus of C57BL6g/d/gW mice as described. Thymocytes were triple stained with FITC-labeled 1B1 ImAb (or Control mAb), PE-labeled anti-CD4 and biotinylated anti-CD8 followed by streptavidin cychrome. IB 11 immunoreactivity was determined within gated sub-populations and compared with control antibody binding (indicated by Ml). Panel A CD4/CD8 profile for C51BL6gld/gld thymus: Panel B. IB 1.1 immunoreactivity in CD4+8- thymocytes: Panel C IB 11 immunoreactivity in CD4+8+ thymocytes: Panel D IB 11 immunoreactivity in CD4-8+ thymocytes: Panel E IB 11. immunoreactivity in CD4-8-thymocytes: 90 Figure 21 C D t ' i i H S M I I* I 11 mi—i I I H I I I I I • "ini 1 6 ° 1 0 1 ie2 ie3 1 0 S CD8 '111! I ' I n u l l I I I H I I I ' l I mil l ' B O J O «—«< 13 U . Ml 10° ie 1 162 ie j 1B11 o Z, i (U U ie° Ml ..jnttii.. U i f lu i i ' u i0 i i82 iia3 ieH 1B11 D E CO O U Ml 10* 102 i&3 10H O U Ml :Q3 . ttr 1B11 1B11 91 Figure 22. 1B11 immunoreactivity in Bh6gld/gld lymph node. Single cell suspension were prepared from the lymph nodes of C57BL6gld/gld mice as described. Lymphocytes were triple stained with FITC-labeled IB 1 lmAb (or Control mAb), PE-labeled anti-CD4 and biotinylated anti-CD8, followed by streptavidin cychrome. IB 11 immunoreactivity was determined within gated sub-populations and compared with control antibody binding (indicated by Ml). Panel A CD4/CD8 profile for C51BL6gld/gld lymph node: Panel B IB 11 immunoreactivity in CD4+8- lymphocytes: Panel C IB 11 immunoreactivity in CD4-8-lymphocytes: Panel D IB 11 immunoreactivity in CD4-8+ lymphocytes: 92 Figure 22 B ft CZh. Q r j - I S I '1 < iitii*—f i ti mi 1 0 ' iiitu* •° 10 CD8 *n—i i mini 1 102 I f i 3 1 0 s O i—i U IB 11 D o i—i U Ml A i 1 l O 1 - 102 !05 i y H 1B11 o M l 1B11 93 Figure 23 IB 11 expression is increased in B220ni CD4+8- lymphocytes compared to B220l° CD4+8- lymphocytes in MRLlpr/fas mice. Single cell suspensions were prepared from lymph nodes from MRLlpr/fas mice as described. Cells were triple stained with FITC labeled 1B11 mAb (or Control mAb), PE-labeled anti-B220 and biotinylated anti-CD4 mAb followed by streptavidin conjugated cychrome and subject to FACS analysis. Panel A CD4 and B220: Panel B CD4 and 1B11: Panel C IBM immunoreactivity in CD4+B220m population: Panel D IB 11 immunoreactivity in CD4+B220l° population: 94 Figure 23 lymphocyte sub-populations, IB 11 antigen was expressed at very high levels by the CD4-CD8-(double negative) population, and at similar levels to those observed in the a(3-TCR positive CD2-CD4-CD8- population which becomes the predominant population when thymus is cultured in IL-2 (Figure 15). The single positive CD4+ population also showed increased levels of IB 11 antigen expression compared to controls. Comparison with B220 staining in the CD4+ population showed that whereas the B220 negative population exhibited a range of intensities of IB 11 staining, the B220 positive population (which comprises about 40% of the CD4+ population of the adult MRLlpr/fas mouse), expressed uniformly high levels of IB 11 staining (Figure 23). This population has been shown to have the same functional defects as the abnormal B220 positive double negative population (210,211). In contrast, the CD8+ population showed slightly reduced levels of IB 11 antigen expression compared with controls. Similar FACS profiles were observed for Ipr/jas and C57BL6gld/gld mice (Figure 22). • V.2.6.2. 1B11 antigen expression increases with age in the Ipr/fas mice Dual stains of lymph node and thymus from 6 week old Ipr/fas mice revealed normal proportions of CD4+ and CD8+ lymphocytes compared with +/+ controls and no expansion of the abnormal CD4-8- population in either tissue. Dual profiles of lymph node and thymus from 6 month old Ipr/fas mice revealed the presence of an abnormal proportion of CD4-8- lymphocytes in both tissues (Figure 24). Sorting of these cells and restaining revealed that all the cells were B220 positive (data not shown). IB 11 antigen expression was compared in groups of young and old Ipr/fas mice and in age matched +/+ control mice and correlated with expression of the activation antigen, CD25 (DL-2 receptor |3-chain)(212). Results are shown in Table 7: Abnormalities of IB 11 antigen expression were only observed in older Ipr/fas mice, however abnormalities of IL-2 receptor P-chain expression were observed in both older and inyounger Ipr/fas mice The data reveal an initial increase in IL-2 receptor P-chain expression in young Ipr/fas mice in the CD4+8-population in thymus and lymph node and in the CD4-8+ population in the lymph node, which antedates the appearance of significant numbers of abnormal CD4-8- lymphocytes. 96 Figure 24 The CD4-8- population becomes progressively expanded with age in MRLlpr/fas mice. Single cell suspensions were prepared from thymus and lymph nodes from 6 week (Panels. A & B) and 6 month old (Panels C & D) MRLlpr/fas mice as described. Cells were labeled with PE-labeled anti-CD4 mAb and biotinylated anti-CD8, followed by streptavidin conjugated cychrome and subject to FACS analysis. Thymus (Panels A & C): Lymph node (Panels B & D). 97 Figure 24 Table 7. IB 11 antigen and IL-2 receptor p-chain expression in young and old,+/+and lpr/lpr mice THYMUS Cell +/+ 6 weeks .+/+. 6 months lpr/lpr 6 weeks lpr/lpr 6 months Type 1B11 JJL-2R(3 1B11 IL-2R(3 mn TL-2Rp mn IL-2Rp CD4+8- 24.5+/-3.-5 10.4+/-2.4 24.1+/-5.2 4.1+/-0.8 28.6+/-2.0 21.0+/-.2.1 69+/-4.9 4.4+/-1-4 . CD4+8+ 433+/-5.0 5.4+/- •" 3.2 73.0+/-9.6 4.9+/-"1.4 41.6+/-16:9 8.3+/-1.8 27.5+/-3.3 47.8+/-9.5 CD4-8+ 80.5+/-3.2 39.5+/. 8.7 87.5+/-2.7 26.0+/-2.6 85.5+/-4.8 ' 51.0+/-5.0 53.0+/-5.9 41.8+/-5.4 CD4-8- 90.4+/- . 3.3 42.9+/-10.2 79.8+/-8.4 33.5+/-2.8 93.8+/-3:0 37.8+/-7.6 96.4+/-1.6 7.8+/-19 LYMPH NODE Cell •.'+/+ 6 weeks +/+ 6 months lpr/lpr 6 weeks lpr/lpr 6 months Type 1B11 IL-2Rp mn IL-2Rp mil IL-2Rp mil IL-2Rp CD4+ 18.8+/-9.4 4.0+/-1.5 38.4+/-6.9 12.5+/-3.8 17.4+/-1.5 . 22.5+/-1.4 81.5+/-4.5 27.0+/-5.5 CD4-8- 73.8+/-10.7 88.0+/-.3.8 . 51.5+/-16.2 58.8+/-14.3 68.3+/-8.6 74.5+/-7.2. 97.5+/-0.5 4.7+/-1.2 . CD8+ 98.0+/-1.2 18.7+/-5.3 98.3+/-0.75 47:00+/-11.0 88.5+/-6.7 . 56.8+/-. 5.7 62.5+/-11.8 34.6+/-3.7 99 This would suggest that T-lymphocytes in these mice are activated at a stage which precedes the development of significant lymphadenopathy. However, their expression of IB 11 antigen appears normal relative to controls at this stage. When the disease becomes established, IL-2 receptor [3-chain expression remains high in the thymus and the lymph node in all subsets with the exception of the abnormal CD4-8- population. In contrast, IB 11 antigen expression is increased only in the CD4+8- and CD4-8- populations, but is down-regulated in CD4-8+ and CD4+8+ subsets. This would suggest that IB 11 antigen is not functioning solely as a marker of T-lymphocyte activation in this model V.2.7. 1B11 antigen expression in a B-cell lymphoma BALB/c mice occasionally develop spontaneous tumours especially with advancing age (213). We examined a 1 year old mouse exhibiting symptoms associated with such a tumour. Post-mortem on the mouse revealed, massive splenomegaly with foci of tumour throughout the spleen, 2 small macroscopic metastases at the hilum of the liver, massive para-aortic lymphadenopathy, para-aOrtic lymphadenopathy, minor enlargement of the cervical and femoral nodes and no ascites. The clinical diagnosis was lymphoma. We then characterised the tumour by FACS analysis and histochemistry, and determined the level of IBM antigen expression by both the tumour and associated normal cells. Comparisons were made with an unaffected litter mate. Examination of cytospin preparations prepared from lymph nodes revealed, small uniformly sized lymphocytes with a high nuclear to cytoplasmic ratio (data not shown).. FACS analysis of lymph node and spleen revealed expansion of a double negative B220+ IgM- lymphocyte population in both lymph node and spleen (Figures 25 and 26). These cells comprised approximately 80% of the total cells in the lymph node and 50% of the total cells in the spleen. 98% of these cells also expressed high levels of IB 11 antigen whereas in the control spleen and lymph node, only 10% of B220 positive cells expressed 1B11 antigen, at levels above control (Figure 25 and 26). Analysis of 1B11 antigen expression in the CD4+8- T-lymphocyte sub-set revealed that 1B11 antigen expression had increased from 8% to 89% positive in both 100 Figure 25. Development of lymphoma in BALB/c mouse is associated with expansion of a B220+IgM-CD4-8- subset in the spleen. Single cell suspensions were prepared from the spleen of a BALB/c mouse with a spontaneous lymphoma and from an unaffected control litter mate. Cells were washed and labeled with FITC-labeled IB 11 mAb (or control mAb), PE-conjugated anti-CD4 mAb,PE-conjugated anti-B220 mAb, biotinylated anti-CD8 mAb, followed by streptavidin conjugated cychrome, or PE-conjugated anti-IgM and subject to FACS analysis. A & B CD4/CD8 profiles of lymphocytes from Control and Lymphoma mice respectively. C & D B220/1B11 profiles of lymphocytes Control and Lymphoma mice respectively. E & F IB 11/IgM profiles of lymphocytes from Control and Lymphoma mice respectively All fluorescence values are expressed relative to the binding of control antibodies. 101 Figure 25 B D - k ' 1&> CD8 c s C Q I B M O 1 B 1 1 E C Q 1 i IgM 102 Figure 26. Development of lymphoma in BALB/c mouse is associated with expansion of a B220+IgM-CD4-8- subset in the lymph nodes. Single cell suspensions were prepared from the lymph nodes of a BALB/c mouse with a spontaneous lymphoma and from an unaffected control litter mate. Cells were washed and labeled with FITC-labeled IB 11 mAb, (or control mAb), PE-conjugated anti-CD4 mAb, PE-conjugated anti-B220 mAb, biotinylated anti-CD8 mAb, followed by streptavidin conjugated cychrome, or PE-conjugated anti-IgM and subject to FACS analysis. A & B CD4/CD8 profiles of lymphocytes from Control and Lymphoma mice respectively. C & D B220/1B11 profiles of lymphocytes Control and Lymphoma mice respectively. E & F lBll/IgM profiles of lymphocytes from Control and Lymphoma mice respectively All fluorescence values are expressed relative to the binding of control antibodies. 103 Figure 26 X o-8 2* i e ° o CQ CD8 1B11 \ *r -B " " i V " " ! ^ ' 103 CD8 101* E PQ TO" r . IgM TP" id* F CQ •9111 i IgM 1 0 4 lymph node and spleen (Figure 27). IB 11 antigen expression was slightly higher in the CD8+ subset in both lymph node and spleen of diseased mouse (98% vs 85%)(data not shown). The data suggests that IB 11 antigen can be expressed aberrantly in B-cell lymphoma and that development of lymphoma is associated with induction of IB 11 antigen expression in the CD4+ lymphocyte subset and to a lesser extent in CD8+ lymphocytes. 105 Figure 27. IB 11 antigen expression is upregulated on CD4+ lymphocytes in lymphoma. Single cell suspensions were prepared from the lymph nodes and spleen from a BALB/c mouse with a spontaneous lymphoma and from an unaffected control litter mate. Cells were washed and labeled with FITC-labeled 1B11 mAb (or control mAb), PE-conjugated anti-CD4 mAb and biotinylated anti-CD8 mAb, followed by cychrome, and subject to FACS analysis. IB 11 mAb immunoreactivity was determined in gated populations of CD4+8- lymphocytes, and expressed relative to binding of control antibody in the same population. A Control spleen: B Control lymph nodes: C Lymphoma spleen: D Lymphoma lymph node: .106 Figure 27 B lB i i mn D 1B11 1B11 107 V.3. Discussion Activation of T-lymphocytes results in the co-ordinated expression of adhesion molecules, signaling receptors and receptors, for growth factors necessary for both lymphocyte proliferation and for expression of mature effector functions. Many of these molecules have been identified using monoclonal antibodies (214). Cell surface antigens which are induced or upregulated in response to T-lymphocytes activation are termed 'activation antigens'. The 1B11 antigen appears to be an activation antigen. Expression of the IB 11 antigen is increased in response to T-lymphocyte activation with Con A in both thymus and spleen, with maximum levels of expression after 72 hours in culture. Thereafter 1B11 antigen expression remains stable at the same level for up to 7 days in culture, in the absence of further stimulation. The kinetics of the up-regulation of IB 11 antigen expression are similar to that of the EL-2 receptor. Like IB 11, IL-2 receptor induction is apparent 1 day after activation and peaks at 2-3 days (215). This places IB 11 antigen in the category of 'early activation antigens' which are distinct in their kinetics of induction from the integrin molecules or VLAs (very late activation molecules) (216). Both Con A activation and activation with phytohaemaglutinin (data not shown) lead to increased expression of IB 11 antigen. In contrast to lectin induced activation, cross-linking CD3 in thymus had inconsistent effects on IB 11 antigen expression (data not shown) possibly because of the low numbers of responsive cells and the absence of any co-signal effects. The consequences of lectin binding include both cross-linkage of cell surface molecules and induction of expression of cytokines and their receptors (216). IB 11 antigen expression might therefore be a consequence of either direct receptor cross-linking or release of soluble mediators. Our studies using IL-la, arid IL-4 suggest that IB 11 antigen expression is not increased in response to these soluble mediators. IL-la has been shown to increase IL-2 and IL-2 receptor expression (217). IL-4 has been shown to be. a growth factor for T-lymphocytes, particularly the TH2 phenotype (218-220). The failure to demonstrate increased IB 11 antigen expression ih response to activation by these cytokines is consistent with our observations in the.Graft -versus-' 108 Host model(207). In this model up-regulation of IB 11 antigen expression was only observed on donor lymphocytes suggesting that it was a consequence of TCR engagement and not a response to a soluble factor or factors. The mechanism of IB 11 up-regulation does not appear to involve activation of protein kinase C as PMA did not increase IB 11 expression. PMA has been reported to induce IL-2 receptor expression on mature T-cells and to increase the levels of expression of a variety of cell surface adhesion molecules (221). Interestingly in both thymus and spleen, the up-regulation of TBI 1 expression occurred predominantly on CD4+ lymphocytes. We are unable to find a similar precedent in the literature although the 5C8 antigen appears to be expressed on activated CD4+ cells only (222). It is possible that Con A activation has predominant effects on CD4+ lymphocytes and that CD8+ lymphocytes proliferate largely in response to secreted cytokines from activated CD4+ lymphocytes. However Con A stimulation, although presumably activating all CD4+ cells, did not result in uniform expression of the IB 11 antigen within this population arid even after 72 hours of stimulation no more than 60% of CD4+ lymphocytes expressed the IB 11 antigen. Flow analysis of lBll-reactive Con A activated CD4+ lymphocytes revealed similar size and granularity profiles to IB 11-negative Con A activated CD4+ lymphocytes. Thus both populations appeared to contain activated T-lymphocytes. It therefore appears'likely-that IB 1,1 expression, defines a sub-set of Con A. activated CD4.+ lymphocytes. This subset might comprise lymphocytes with distinct effector functions (e.g; memory cells,.cytotoxic cells, T-helper subsets) but as yet we have been unable to characterise these functions. In contrast to the effects of lectin stimulation in vitro, IB 11 expression was increased on both CD4+ and CD8+ lymphocytes in Graft-versus-Host and in the allograft rejection model. In the allograft rejection model, the greatest increase in 1B11 expression in CD4+ lymphocytes was associated with a decrease in the CD4:CD8 ratio to less than 1. In peritoneal CD4+ lymphocytes, 1B11 antigen expression was associated with larger more granular cells, suggesting that theCD4+ population which was induced to express 1B11, comprised a uniform activated population. Again : 109 this data would support a role for IBM antigen in cytotoxicity as it is likely that the CD4+ lymphocytes in allograft rejection would have cytotoxic properties. The effects of IL-2 on IB 11 antigen expression are more complex, increased expression of 1B11 antigen in response to IL-2 was only observed in the thymus and then only at later time points. Sub-population analysis of IB 11 antigen expression by FAGS revealed that the increase observed in IB 11 antigen expression was attributable to an increase in 1B11 antigen expression by . CD4-8- thymocytes and by expansion of this population. Interestingly, the predominant IB 11 antigen expressing population also expressed the aft TCR. This distinct a(3 DN population expresses 1B11 antigen at far higher levels than have been observed previously on other cell types. apTCR+CD4-8- thymocytes (aP DN) normally comprise 0.5% of total thymocytes (223). The majority of these cells express the Vp8.2 TCR recognised by the F23..1 mAb (224). Long term culture of thymus has been reported to result in the outgrowth of CD4-8- thymocytes expressing this phenotype. Alpha-beta TCR positive thymocytes respond to IL-1, IL-2, IL-3 and IL-4 (225). Methylation studies suggest that they have formally expressed the CD8 co-receptor (226) but it is not clear whether the cells which we derived in culture represent expansion of the original ap DN population or whether they are derived from other single positive, double positive or double negative thymocytes. Data from Gause et al (225) suggests the former. They derived ap DN thymocytes from a pre-selected DN population. Our data, showing high levels of F23.1 positive ap DN thymocytes after a relatively short incubation inTL-2, suggests that they may also be derived from either ap TCR negative DN thymocytes or from CD4+8+ or single positive cells which have down:regulated expression of co-receptors. Indeed this may be a default pathway for unselected thymocytes. Although there is evidence that although these ap TCR positive thymocytes have undergone some repertoire selection (226), this population contains a very high frequency of auto-reactive T-lymphocytes. The role of these cells in the immune response is unknown. Functionally these cells synthesise IL-4, y-interferon, TNFa but not IL-2 (227) and retain cytotoxic properties, showing NK-like as well as antibody directed cytotoxicity. Although they also lack CD2, they are likely distinct from the aP DN population expanded in the Ipr/fas mouse. This population expresses the 110 B220 isoform of CD45, which is absent from thymic a(3 DN cells and only about 30% of Ipr/fas DN cells express the Vp8.2 TCR (225, our observations). Nevertheless the finding of very high levels of IB 11 expression in ocP DN thymocytes prompted us to investigate the role of IB 11 antigen in the. Ipr/fas mouse. The MRLlpr/fas mouse and the C3HBL6gld/gld mouse develop a lymphopr.oliferative disease characterised by the expansion of a T-lymphocyte population with the unusual phenotype of CD4-8-B220+CD2-aPTCR+(208). There is evidence from antibody studies that these ap DN lymphocytes are derived from both CD4+ lymphocytes (228) and from CD8+ lymphocytes (229, 230). In vivo transfer studies suggest that purified CD4+ T-cells are able to generate aP DN lymphocytes on transfer into nude mice (231). Methylation studies have suggested that aP DN lymphocytes have previously expressed the CD8 co-receptor (232). Since all mature CD4+ and CD8+ lymphocytes will have expressed both the CD4 and CD8 co-receptors during the double positive phase of thymic maturation, these studies cannot identify the direct precursor of the aP DN cell. The aP DN population contains a high proportion of auto-reactive T-lymphocytes (233). The mechanism whereby these cells escape negative selection in the thymUs or peripheral deletion has been the subject of considerable investigation. .The mutation responsible for the MRLlpr/fas phenotype. is in Fas-1, a cell surface receptor belonging to'the TNF receptor family which mediates apoptosis when cross-linked (234). The mutation responsible for the C3HBL6gld/gld phenotype is in the corresponding ligand for this receptor (235). There is evidence that negative selection is relatively normal in the thymus of these autoimmune mice either because Fas-1 antigen expression is leaky in Ipr/fas mice, or because other deletion mechanisms can compensate for the absence of Fas-1 (236). However there is evidence for defects in peripheral deletion and tolerisation of reactive lymphocytes (237-239). These defects are thought to arise because of failure to express functional Fas-1 or its ligand. The demonstration of Fas-1 ligand expression on activated lymphocytes suggests that since activation can induce expression of both Fas-1 and its ligand, activated lymphocytes are responsible for their own deletion. It is possible to postulate a mechanism whereby expansion of activated T-lymphocyte populations is controlled by Fas-1 111 mediated deletion. This mechanism may be defective in Ipr/fas mice. A defect in activation induced deletion of T-lymphocytes would explain why infected mice, and mice immunised with complete Freund's adjuvant develop the disease at an earlier age. (D. Waterfield, personal communication). However it does not explain the preponderance of auto reactive lymphocytes in the Ipr/fas and BL6gld/gld mouse nor does it explain the thymic dependence of the disease in the first 3 weeks (240,241). Down-regulation of CD4 and CD8 co-receptor expression and CD2 expression may provide a means for mitigating the effects of activated auto-reactive effector cells. This may explain why ap DN cells express a "partially activated phenotype". The absence of critical adhesion molecules would also impair effector functions leading to an anergic phenotype. Very high" levels of IB 11 antigen have been demonstrated in aP DN cells. It is not clear whether this high level of expression reflects the activation or the anergic component of the phenotype, since the level of expression observed is far higher than that normally seen in activated lymphocytes. Other antigens reported to be aberrantly expressed at high levels by ap DN cells include JlId (242), CD45RA (243) HI.2F3 (244) and 9F3(245). High IB 11 antigen expression was also observed in the CD4+ subset of Ipr/fas and C3HBL6gld/gld mice. This increased level of 1B11 expression may reflect an activated phenotype, however the aetiology of this activation state is unclear. A proportion of CD4+ lymphocytes in the Ipr/fas mouse are reported to express B220 (210, 211). The functional properties of this CD4+B220+ subset are very similar to those of the aP DN cells. Very high levels of IB 11 expression were observed in this.sub-set. However, IB 11 expression was also increased in CD4+B220- lymphocytes, although to a lesser extent. However it is possible that 1B11 antigen expression is an early marker for the development of the CD4+B220+ lymphocyte and ultimately for the ap DN lymphocyte. In direct contrast, CD8+ lymphocytes expressed reduced levels of IB 11 antigen compared with controls. If aP DN lymphocytes are derived from CD8+ lymphocytes, as some have suggested (230), it is surprising that these cells appear to be down-regulating IB 11 antigen expression, when extremely high levels of IB 11 antigen expression are observed in the aP DN 112 lymphocyte. The finding of reduced levels of IB 11 antigen expression in CD8+ lymphocytes suggests either that these cells are not activated or that IB 11 antigen is not exclusively a marker of the activation state pf the lymphocyte. Prior to the development of lymphadenopathy, the pattern of IB 11 antigen expression is normal in both Ipr/fas mice and controls. However levels of IL-2 receptor expression are increased suggesting a low level of activation.. The aetiology of this activation is again unknown. The increase in IL-2 receptor expression appears to precede changes in IBM •antigen expression, suggesting that the two processes can.be dissociated. In the adult Ipr/fas mouse, levels of IL-2 receptor expression are not significantly higher than those observed in young Ipr/fas mice and are only significantly higher than age-matched +/+ mice in the CD4+ subset. In the +/+ control mouse there is an age related increase in IL-2 receptor expression in both subsets, but particularly in the CD8+ sub-set. This may be related to the observation that +/+ control mice may develop an age-related autoimmune disease. Control mice.also show a small age related increase in IBM expression in the CD4+ subset, compatible with delayed development of an autoimmune state. We found diminished levels of IL-2 receptor expression in the abnormal a(3 DN population. This is consistent with other reports in the literature (246) and may represent a further mechanism for abrogating the deleterious effects of these Cells. IB 11 antigen was found to be highly expressed in a spontaneous B-cell lymphoma. It is hot clear whether this represents ectopic expression of IB 11 antigen in a tumour cell i.e. IB 11 is a tumour associated antigen, or whether the tumour arose as a result of proliferation of cells belonging to the small B220+1B11+population observed in normal spleen. IBM antigen expression was found on CD4+ lymphocytes associated with the tumour and levels of expression appeared to be related to the extent of tumour burden i.e. IBM antigen expression was lower in the spleen than in more profoundly affected lymph nodes. Infiltrating lymphocytes have been characterised in a number of tumours (247) however their precise function remain obscure. Although many can be stimulated to proliferate and or display cytotoxic activity towards tumour cells, especially after addition of IL-2 (248), the development of tumour per se may imply a breakdown of immunological surveillance. Interestingly in many animal models the 113 infiltrating lymphocytes belong to the TH2 phenotype and associated with antibody production rather than with the cytotoxic response (249). IB 11 antigen expression appears to be induced in both CD4+ and CD8+ lymphocytes in response to this tumour. Whether high IB 11 expression reflects ah effective response to the tumour (possibly cytotoxic as in the allograft rejection model) or anergy (as in the lpr/lpr and BL6gld/gld model ) is unknown, however it is an area which should warrant further study. In summary 1B11 antigen expression, particularly by CD4+ lymphocytes is correlated with activation by lectins in vitro and by allograft in vivo. In the Ipr/fas and BL6gld/gld mouse models and in the lymphoma tumour model, the situation appears more complex. Very high levels of IB 11 expression in these models are correlated with an anergic state, whereas lower levels of expression may be correlated either with activation or with anergy. -114 VI. Functional studies of the 1B11 antigen VI. 1 Introduction and Aims The following chapter describes our attempts to determine the function of the IB 11 antigen on T-lymphocytes. . Many cell surface antigens defined by mAbs have proven to be adhesion molecules. These molecules are involved in all interactions between T-lymphocytes and opposing cells and mediate the many functions of T-lymphocytes, such as antigen recognition, lymphocyte trafficking and provision of help to other effector cells involved in the immune response. Up-regulation of expression of adhesion molecules is a consistent feature of T-lymphocyte activation (with the notable exception of certain adhesion molecules which may inhibit lymphocyte extravasation such as MEL-14 (168, 250-252). Our findings that 1B11 antigen expression was up-regulated in response T-lymphocyte activation was consistent with the hypothesis that IB 11 antigen was an adhesion molecule. In this chapter we have attempted to determine whether 1B11 antigen may function as an adhesion molecule in a number of a different contexts. These include lymphocyte migration, adhesion to endothelial cells, conjugate formation and cytotoxicity. In addition to their role in facilitating physical.interactions between T-lymphocytes and opposing cells, many adhesion molecules are alsd signal transducing molecules. Binding to their ligand may result in a change in the activation state of the T-lymphocyte by augmenting or inhibiting the TCR-initiated activation cascade. These effects of ligand binding may be mimicked by cross-linking adhesion molecules with mAb. Cross-linking of a number of cell surface antigens with mAb leads to either a proliferative response e.g. CD3 (151) or synergises with the proliferative effects of TCR cross-linkage e.g. CD28 (253, 254). Therefore we sought to determine whether IB 11 antigen cross-linking using the IB 1.1 mAb could transmit or modulate growth.signals in primary T-lymphocytes and in cell lines. A more direct role for IB 11 antigen in the signal transduction pathways regulating cell proliferation was examined by immunoprecipitating the antigen and assessing its ability to directly ' 115 affect the phosphorylation status of target signal transduction molecules. Early proliferation signals require a co-ordinated sequence of phosphorylation and de-phosphorylation of cellular proteins. Many cell surface receptors have intrinsic kinase e.g. EGF receptor (255) or phosphatase activity e.g. CD45 (256) and association with their ligand results in modulation of this activity and transmission of a proliferative or inhibitory signal. We therefore examined whether IB 11 antigen, immunoprecipitated from whole cell lysates,.could function as kinase or a phosphatase. A number of cell surface adhesion molecules show a non-homogeneous distribution over the cell surface and may become concentrated in cleavage furrows or on uropods (257-260). Localisation of the antigen requires a tight association with the cytoskeletal elements required to produce these structures. We examined the distribution of IB 11 antigen on the cell surface of splenocytes and thymocyte by immunofluorescent microscopy and determined its association with the cytoskeleton using differential detergent lysis conditions. Two different solubilisation buffers containing differing concentrations of detergent were used to make extracts from IB 11 antigen expressing cells. IB 11 antigen expression in these extracts was then determined by Western immunoblotting. The extent to which 1B11 antigen was tightly-associated with the cytoskeleton could be determined by comparing the amount of 1B11 antigen released into the lysate under weak detergent conditions and the amount of IB 11 antigen present in extracts produced using harsh detergent conditions in which additional IB 11 antigen bound to the cytoskeleton was dissociated. Adhesion to endothelial cells is one of the earliest functions of activated lymphocytes. For this reason we chose to examine the role of IB 11 antigen in adhesion of T lymphocyte cell lines to endothelial cells in culture. Because of technical difficulties in obtaining sufficient endothelial cells from primary tissues in mouse we chose to examine adhesion to an transformed endothelial cell line, Endo-Dl(138). This cell line, derived from MRL Ipr/fas mice, shows the ultra-structural and histochemical features of endothelial cells in vivo, however expression of adhesion molecules by this cell line has not been characterised and may differ from normal endothelial cells. Nevertheless such cell lines have been used successfully to examine leukocyte adhesion by other investigators (165). In order to be able to detect adherent cells, T-lymphocyte cell lines were labeled overnight with thymidine and then allowed to adhere to a semi-confluent monolayer for 2 hours. Attempts 116 were made to pre-activate the endothelial cells to express differing adhesion molecules by pre-incubating with IL-1 and with IL-4. IL-1 has been shown to induce expression of EL AM-1, ICAM-1 and VGAM-1 (261-263) whereas IL-4 has been shown to up-regulate antigens which preferentially function as cell surface adhesion molecules for T-lymphocytes (264). The pattern of IB 11 antigen expression we had observed, specifically, high levels of expression on CD8+ lymphocytes, yS-TCR lymphocytes, a(3-TCR positive DN lymphocytes and immature thymocytes, suggested a role in cytotoxicity (171, 265-267). Accordingly we chose to investigate the role of 1B11 antigen in the cytotoxic response. We chose two models: A CTL line which reacts with the P815 cell line and the allograft rejection model described in chapter V which also employs P815 as the target. One of the advantages of these two models is that the level of IB 11 antigen expression on the target cell is very low and therefore a.ny inhibition observed is likely a result of IB 11 mAb binding to effector cells. Finally we examined the role of IB 11 antigen in lymphocyte migration and in turnover of activated cells in vivo. Adhesion molecules have been shown to be critical to homing patterns (170, 267-273). For example, CD4+ and CD8+ lymphocytes, and naive and memory T-cells, show distinct patterns of migration and these patterns are determined by differential expression of adhesion molecules (170, 269-273). IB 11 antigen is differentially expressed on activated cells and on CD4+ and CD8+ lymphocytes and might play a role in determining lymphocyte migration patterns. T-lymphocytes are incessantly mobile. Generally a given T lymphocyte moves from one site to another once in a 24 hour period. Few activated lymphocytes can be detected in the circulation, however both naive T-lymphocytes and memory cells derived from activated cells can be detected and show differential patterns of migration. In general, naive cells are unrestricted in their circulation through secondary lymphoid tissue whereas memory cells tend to return to their tissue of origin and in addition may circulate through extra-lymphoid sites. These patterns of migration are determined by expression of specific adhesion molecules. We sought to determine whether expression of IB 11 antigen on CD4+ lymphocytes directed their homing to specific sites or was associated with reduced survival in vivo. Accordingly we determined the fate of activated 117 splenocytes re-infused into syngeneic mice and attempted to correlate the location of these cells at 24 hours with their expression of IB 11 antigen. .118 VI.2. Results VI.2.1. Kinase Studies IB 11 antigen was immunoprecipitated from D26 cells'as previously described and incubated in the presence of and potential kinase substrates; 1B11 antigen, a-casein, histone HI and myelin basic protein. IB 11 antigen immunoprecipitated from D26 failed to stimulate 32p incorporation into any substrates and did not demonstrate autophosphorylation (data riot shown). VI.2.2. Phosphatase studies Reaction of IB 11 immunoprecipitated from CTL. cells with tyrosine -phosphorylated src peptide and malachite green, did not result in the cleavage of phosphate from the peptide as determined by a colour change in the malachite green reagent, nor did IB 11 antigen cleave phosphate groups from PNPP. IL-3 receptor immunoprecipitates from trie same cell line also failed to induce a colour change in the malachite green reagent, however GD45 was effective (data not shown). VI.2.3.1. Proliferation Assays Bioassays for proliferation were performed using the HT-2Y and CTL cell lines which express high levels of IB 11 antigen as well as primary thymocytes. Responses to IB 11.antigen cross-linking were compared with responses to control antibody and to 2C11 a monoclonal antibody, a hamster antibody, which binds.to and activates the T cell receptor. Cross-linking IB 11 antigen alone did not induce proliferation in either thymocytes, HT-2Y or CTL (data not shown): Cross-linking IB 11 antigen did not synergise with either PMA or IL-2-in inducing proliferation in either thymocytes (Figure 28) or cell lines, (data not shown). However PMA, IL-2, 2C11 antibody and Con A did induce proliferation in the thymus and IL-2 induced proliferation in cell lines. Cross-linkirig experiments were not performed with spleen as IB 11 antigen is less abundant in this tissue compared with thymus. 119 Figure 28 1B11 mAb is unable to deliver or augment a proliferative signal in T-lymphocytes. Terisaki trays were, incubated with 10 Ug/ml sheep anti-mouse or riabbit anti-hamster antibody for 2 hours at 37 °C and then washed. Single cell suspensions were prepared from BALB/c thymocytes as described. Cells were incubated in the Terisaki trays with either A lOU/ml IL-2 or B 1 ng/ml PMA or C 5 ug/ml Con A. Serial dilutions of antibodies were added and the cells incubated at 37 °C, 5% C02 for 24 hours. Cells were then pulsed with H^-thymidine for 5 hours and frozen. ^H-Thymidine incorporation was determined by harvesting cells onto filter papers which were placed in vials with scintillation fluid and read in a gamma counter. Experiments were performed in triplicate and mean results are shown, ik. IB 11 mAb, I82C11 mAb, Ocontrol mAb. Backgroung H^-Thymidine incorporation in the absence of stimulation was consistently less than 100 cpm. 120 Figure 28 10000 E cu c o o & o o c E jr. H 1000 H 100 o.oi 0.1 1 10 Antibody concentration (ug/ml) 100 100000• o cu a IOOOOH c E H 1000 0.01 0.1 1 10 Antibody concentration (ug/ml) 100 100000-1 ? o.oi o.i I i o IOO Antibody concentration (ug/ml) 121 VI.2.3.2. The Mixed Lymphocyte Reaction Irradiated DB A2 spleen cells were mixed with control C3H cells in a 1:1 ratio in 100 ul of medium in the presence or absence of IB 11 mAb Or control. At 24 hour time intervals proliferation was determined in aliquots of cells by incorporation of ^ H-thymidine. Both IB 11 mAb and isotype matched control antibody induced a modest dose dependent inhibition of proliferation in the MLR (Figure 29). This inhibition was apparent at all time points sampled. The Fc dependent inhibition of the MLR can be mitigated by using Fab2 fragments. Unfortunately IB 11 mAb is an IgG2a antibody and we were unable to produce Fab2 fragments in sufficient quantities to allow us to pursue these experiments further. VI.2.4. Differential Detergent Solubility Different lysis conditions were used to determine the extent to which IB 11 antigen is associated with cytoskeleton in thymocytes and splenocytes. In mild detergent conditions, only that 1B11 antigen which is not tightly associated with the cytoskeleton would be solubilised. The harsher detergent conditions would be expected to release more material from the cytoskeleton. The amount of IB 11 antigen released under harsh detergent conditions following mild detergent extraction would therefore reflect the amount of 1B11 antigen associated with the cytoskeleton. Approximately twice as much IB 11 immunoreactivity was found in MCSK lysates compared to RIPA lysates prepared from thymus. This suggests that approximately one third of the IB 11 antigen in this tissue was associated with the cytoskeleton. (Fig 30A). On Con A activation, the total amount of IB 11 antigen present increased, but the relative proportion associated with the cytoskeleton did not change dramatically. Lysates of Con A activated spleen cells showed similar proportions of IB 11 antigen with approximately one third of the total IB 11 immunoreactivity extracted under harsh detergent conditions. In resting splenocytes, no additional IB 11 antigen was extracted by RIPA buffer following MCSK solubilisation suggesting that IB 11 antigen is not tightly associated with the cytoskeleton in resting splenocytes (Fig 30 A). 122 Figure 29 IB II mAb prodices a modest inhibition of the mixed lymphocyte reaction which is not significantly different from an isotype matched control mAb. Irradiated DBA2 spleen cells were mixed with control C3H cells in a 1:1 ratio in lOOfil of medium in the presence or absence of serial titrations of IB 11 mAb & or control mAb @ At 24 hour time intervals proliferation was determined in aliquots of cells by pulsing for 5 hours with ^H-thymidine and then freezing. Cells were then harvested onto filter paper and read in vials containing scintillation fluid in a gamma counter. Experiments were performed in triplicate and ;mean results are shown for each time point. A 48 hours.. B 72 hours. C 96 hours. Values for background proliferation, DBA2 splenocytes alone and C3H splenocytes alone, were less than 10^  cpm in all cases. 123 Figure 29 I I I r _ I : h 1 1 0 20 40 60 80 100 1 Antibody concentration (mg/ml) i — : — - i n 1—: 1 1 0 20 40 60 80 100 1 Antibody concentration (mg/ml) Figure 30. 1B11 associates with the cytoskeleton in thymus and activated lymphocytes. A Single cell suspensions were prepared from thymus and spleen from B A L B / c mice. Aliquots of cells were serially lysed using 2 different lysis buffers: initially MCSK (mild detergent) was used to solubilise the cells, the lysate was then removed and the non-soluble precipitate subsequently solubilised in RIPA (harsh detergent). The remaining cells were then incubated in the presence of 5ug/ml Con A for 48 hours, washed and then lysed as described above. Lysates were made up in sample buffer and resolved on 7.5% SDS PAGE gel. IBM immunoreactivity. was determined by immunoblotting. Lanes a-d Control: Lane e-rf Con A: Lanes a, b, e & f Thymus: Lanes c, d, g & h spleen: Lanes a, c, e & g RIPA: Lanes b, d, f & h MCSK: B Single cell suspensions were prepared from thymus as described previously. Cells were incubated at 5 x 10^ per ml in either 10 U/ml IL :2 or 5 ug/ml Con A for 48 hours. At the end of the incubation period, cells were fixed with 1% paraformaldehyde. Following fixation, cells were washed and incubated at room temperature with 1'ug/ml IB 11 or control antibody. After 30 minutes cells were washed and incubated with a 1:200 dilution of FITC-labeled anti-rat antibody for 30 minutes. Cells were then washed through FCS and examined under the microscope at x40 magnification; a IL-2 treated thymus, b Con A treated thymus. C Single cell, Con A treated thymus. • 125 Figure 30 VI.2.5. Immunofluorescence Studies. Immunofluorescent studies of activated thymocytes and splenocytes revealed increased levels of IB 11 expression in response to Con A activation (Figure 30B). In some thymocytes, IBM antigen was clustered in the region of uropod-like extensions from the surface of the cell. These studies were performed in cells which were fixed prior to staining and therefore could not have arisen because of capping of the antigen by the mAb. In spleen no uropod like structures were observed. VI.2.6 Adhesion assays: Vl.2.6.1 Adhesion of cells to an endothelial cell monolayer can be detected by thymidine labeling. HT-2Y, EL4, HDK-1 and WEHI 274.3 cells were labeled overnight with 3H-thymidine. Cells were then washed and allowed to adhere to a semi-confluent monolayer of Endo-Dl cells. Serial titration of cell number over the range 1000 to 5000 produced a linear increase in cpm harvested from cell lysates (Figure 31). Below 1000 cells, counts were not significantly different from background values. Cells varied in their abilities to incorporate 3H-thymidine and in their adherence to the endothelial monolayer, with the highest counts observed for the HT-2Y cell line and the lowest counts observed for EL4. Curiously, this order is identical to that observed for relative expression of IB 11 antigen . The HT-2Y cell line which expresses large amounts of IB 11 antigen and was used subsequently for adhesion assays. VI.2.6.2 Effects of mAbs and cytokines. 1 Adhesion of 10^  cells of the HT-2Y cell line to the endothelial monolayer was not affected by the presence of IBM antibody or control antibody at concentrations of 100 (ig/ml or 20 |ig/ml. Pre-treatment of endothelial monolayers with 2 ng/ml IL-1-(3 or lOOU/ml IL-4 for 24 hours did not increase adhesion of HT-2Y cells, nor was adhesion to 'activated' endothelium decreased in the presence of IB 11 mAb (Figure 32). 127 Figure 31 Adhesion of cells to an endothelial cell monolayer can be detected by ^H-Thymidine labeling. 100000 -| 10 100 1000 10000 Cell no HT-2Y A EL4 O, HDK-1 • and WEHI 274.3 X cell lines were suspended at 2 x 105 cells per ml and labeled overnight with 100 uCi/ml of 3H-thymidine at 37°C, 5% C02- Cells were then washed and serially diluted and then allowed to adhere to a semi-confluent monolayer of Endo-Dl cells in a 96 well tissue-cultured treated plate. After 2 hours wells were washed 3 times with warm Hanks balanced salt solution and adherent cells were lysed with 50 |Lil of 1% NP-40, 150mmol/l NaCI and 20mmol/l tris, pH 7.5. The lysed contents of each well were harvested onto filter paper discs and counted in a scintillation counter after addition of 300 ul scintillation fluid. 128 Figure 32. 1B11 mAb does not inhibit adhesion of the HT-2Y cell line to an endothelial cell monolayer. HT-2Y cell line was labeled overnight with ^H-thymidine as described. Cells were allowed to adhere for 2 hours to a sub-confluent monolayer of Endo-Dl cells in the presence of either 100 u\g/ml or 20 p:g/ml of IB 11 mAb or control mAb in a 96 well plate. After 2 hours the monolayers were washed^ gently with warm Hanks buffered salt solution and the contents of each well was lysed.with 50 ul of 1% NP-40, 150 mmol/l NaCl, 20mMpl Tris, pH7.4. The labeled cell lysates were then harvested on filter discs and counted in a.scintillation counter after addition of 300 pi scintillation fluid. Result are expressed as a mean of 4 replicates (+/- SE). Typical results are shown. Panel A Unstimulated monolayers: Panel B IL-la stimulated monolayers: Panel CIL-4 stimulated monolayers: 129 Figure 32 lOOOOO-i Nil Control Control 1B11 1B11 100/ng/ml20ing/inll00 mg/ml20ing/ml lOOO'00-i c o 8000CH o ft O u a • <u c 6000CH S 40000H •JS 2000CH % Nil Control Control 1B11 1B11 lOQpig/ml 20 /ng/mllOO mg/>nl20 mg/ml 100000-1 80000H 60000H •S 4000CH 2000CH _Lt 'AS PS jgg Nil Control Control 1B11 1B11 . 100/ng/m!20/ng/mll00/ng/ml20/ng/ml 130 VI.2.7. Cytotoxicity: VI.2.7.1. Conjugate formation . Fluorescently labeled P815 targets were ad-mixed with either P815 responsive peritoneal lavage cells or with CTLs specific for P815 in the presence or absence of IB 11 antibody and control antibody. IB 11 mAb at concentrations of 100 P-g/ml did not inhibit conjugate formation between peritoneal lavage effectors or CTL cells and P815 cells. Anti-CD8 antibody did however produce a modest inhibition of conjugate formation in the same system (Figure 33). VI.2.7.2. Chromium release assays. No reduction in percentage specific lysis was observed in the presence of either 1B.11 or control mAbs using either peritoneal lavage cells or P815-specific CTL cells as effector cells over a wide range of E:T ratios. Antibody to LFA-1 nevertheless abolished the cytotoxic response of the CTL cell line to P815 (data not shown). Experiments were performed on 3 separate occasions. A representative experiment is shown in Figure 34 Similar effects were observed for both effectors generated in vivo and for cell lines, although overall percentage specific lysis was lower for in vivo generated effectors, presumably because they are a heterogeneous population. Specificity of the CTL-P815 interaction was confirmed using an irrelevant target cell line, EL-4 (data not shown). VI.2.8. Lymphocyte migration studies : VI.2.8.1. FITC labeling of effectors Single cell suspensions of splenocytes were activated for 48 hours with Con A and then viable cells were labeled for 1 hour with serial dilutions of FITC. After labeling cells were washed and subjected to FACS analysis. A linear relationship between FITC concentration and MFI was noted. The concentration of 10 u\g/ml was chosen as providing a high MFI in FACS (using this 131 Figure 33 1B11 mAb does not inhibit conjugate formation in the cytotoxic response. 30-i 2 5 -p o U No mAb CD8 IB 11 Control mAb Fluorescently labeled P815 target cells were admixed with P815 responsive peritoneal lavage cells in a 1:1 ratio in the presence or absence of 100 p.g/ml of either lBllmAb, control mAb or anti-CD8 mAb. Cells were centrifuged at for 30 seconds and incubated for 1 hour at 37°C, 5% C02- Cells were then resuspended by gently pipetting with a 100 |il pipette 10 x. Conjugates were then scored by counting in a counting chamber under a fluorescent microscope at x 40 magnification. Only those conjugates comprising one fluorescent (target) and one non-fluorescent (effector) cell were scored. Counts were expressed as a percentage of the total number of fluorescently labeled targets. Results are expressed as a mean of 3 high power fields. 132 Figure 34. IB 11 mab does not inhibit the cytotoxic response to allograft. Peritoneal effector cells (prepared by immunising C57/Black6 mice with the P815 cell line) or P815 specific CTL cells were mixed with chromium labeled P815 target cells at effector : target ratios of 10:1, 5:1, 2.5:1 and 1.25:1. Cells were incubated in the presence or absence of a 1:1000 dilution of 1B11 mAb ascites or Control mAb ascites for 2 hours at 37°C. Cells were pelleted and supernatant removed and counted in a gamma counter. Percent specific lysis was calculated for each of 4 replicate experiments. Typical results are shown. A CTL cell line (A) 1B11 mAb. ( • ) Control mAb: B Peritoneal effectors (A ) 1B11 mAb. ( • ) Control mAb: 133 134 concentration profiles for both labeled and unlabeled cells could be viewed on the same scale) and which was associated with high cell viability (Figure 35A). VI.8.2 Lymphocyte migration studies: Half-life of FITC labeling in vitro Activated viable splenocytes labeled with 50 |ig/ml of FITC were washed and examined by FACS at various time intervals to determine the half life of the FITC within the cells. The decay profile of the FITC label was biphasic, with an initial rapid decline followed by a slower decline. The initial rapid decline may be due to loss of dye loosely associated with cells, rather than incorporated into cells. The subsequent slow loss of MFI may be either due to a slow leak and equilibration with the surrounding media, cell death and release of dye or cell division. This slower half life is approximately 12 hours (Figure 35B). -VI.2.8.3. Dual staining of effectors Because the FACScan is often unable to distinguish high FITC signal from high phycoerythrin signal without appropriate compensation, we determined whether it was possible to perform dual stains on FITC labeled effectors. We were able to demonstrate that FACS profiles of FITC labeled splenocytes showed no significant 'spill over ' of signal into PE or Cychrome detectors (Figure 35C), however we were concerned about the effects of adding further fluorescent stains. Accordingly, viable FITC labeled activated splenocytes were incubated with PE-labeled anti-CD4 and biotinylated anti-CD8 antibodies and a streptavidin conjugated secondary agent was added. CD4+ arid CD8+ populations were clearly distinguished in the FITC labeled samples (Figure 35D) and profiles were comparable with unlabeled samples (data not shown). VI.2.8.4 In vivo fate of labeled cells Splenocytes were activated in vitro with. Con A and analysed by FACs for expression of IB 11 antigen by sub-populations. Viable cells were then labeled with 10 |ig/ml FITC and re-injected into identical litter mates. After 24 hours the litter mates were sacrificed and the pattern of 135 Figure 35. Activated splenocytes can be labeled with FITC. A Single cell suspensions were preparedfrom the spleen of BALB/c mice as described. Cells were incubated at 5 x IO0" cells/ml for 48 hours in the presence of 5 (ig/ml Con A. Following incubation, viable cells were purified over Ficoll-Hypaque, washed and then labeled with varying concentrations of FITC for 1 hour. After 1 hour, cells were washed 3 times and subject to FACS analysis. B Viable Con A activated splenocytes were labeled with 50 P-g/ml of FITC for 1 hour at 37°C, 5% C 0 2 - Cells were then washed and incubated for a further 24 hours. At various time points aliquots of cells were removed, washed and subject to FACS analysis. C & D Viable Con A activated splenocytes were labeled with 100 u\g/mlof FITC for 1 hour at 37°C, 5% C O 2 . After 1 hour, cells were washed and incubated in the presence of PE-labeled anti-CD4 mAb and biotinylated anti-CD8 mAb, followed by streptavidin conjugated cychrorrie and subject to FACS analysis. C FITC profile of stained cells. D CD4-PE and C D 8 -cychrome profile of FITC-labeled cells. 136 Figure 35 IB 11 antigen expression by FITC labeled cells determined in spleen, mesenteric nodes and axillary nodes (Table 8). Approximately, 50% of injected cells could be recovered after 24 hours. The majority of these cells had returnedto the spleen, where the frequency of injected cells was 1:100. A slightly lower frequency of injected cells was detected in the mesenteric lymph nodes, with the lowest frequency of injected cells recovered from the axillary nodes (Table 8). This would suggest that activated lymphocytes home to their tissue of origin and secondarily to mesenteric nodes.; No differences were observed in the frequencies of CD4+ and CD8+ lymphocytes between these tissues and the CD4/8 profiles were identical to prior to re-injection. However, almost all of the CD4-8-population re-introduced, returned to the spleen. In axillary nodes, mesenteric nodes and spleen levels of 1B11 antigen expression were similar to background on CD4+ lymphocytes (range 17-28%). This contrasts with the baseline level of expression of IB 11 antigen on re-injected CD4+ lymphocytes which was 69%. No FITC labeled cells were found in non-injected control mice and no 'spill over' into PE or cychrome detectors was observed, suggesting that the FITC labeling is specific for re-injected cells. Flow profiles of the FITC positive cells suggest that they are slightly larger and slightly more granular than host lymphocytes, consistent with their activated state, however their size and granularity is slightly smaller than prior to injection (data not shown). 138 Table 8. In vivo fate of injected activated splenocytes Site Frequency %CD4 %CD8 %CD4-8- CD4/lBll+% Axillary Node • 1:357 79% 19% ... 2% 17% Mesenteric Node 1:238 80% 13.6% 1% 28% Spleen 1:108 78% 9% 11% 17% Activated splenocyes - 71.5% 12.4% 14.8% . 69% 139 V1.3. Discussion Studies examining the potential functions of IB 11 antigen were largely inconclusive. IB 11 antigen showed no kinase or tyrosine phosphatase activity against a variety of common substrates. Although it is possible that IB 11 antigen shows activity against a substrate which was not included in these assays, we did not pursue these investigations further. Cross-linking IB 11 antigen did not result in cell proliferation nor did it synergise with other growth factors. This does not preclude the possibility that the IB 11 antigen transduces or modulates a growth signal as antibody induced proliferation has been shown to be a property of only a limited number of antibodies against activation inducing antigens. It is likely these antibodies are more effective in inducing dimerisation or bind at sites which mimic ligand binding. Because of the possibility that contaminating B-lymphocytes and macrophages will respond to the secondary antibody we did not perform similar studies on splenocytes. In addition, the levels of IB 11 antigen expression on resting spleen are low. IB 11 mAb modestly inhibited the MLR reaction in a dose dependent manner. However a similar inhibition was observed with an isotype matched control antibody. Fc mediated inhibition of MLR is reported in the literature (274), however its effects are usually mitigated by making Fab2 fragments of the antibody. Unfortunately IB 11 mAb is an TgG2a isotype which is difficult . to digest with pepsin without generating large numbers of Fab fragments (275). For this reason and because the level of inhibition observed with control antibody and IB 11 was so similar and low, we did not pursue these experiments further. IB 11 antigen was found associated with the cytoskeleton in thymus but not in spleen. Activation of splenocytes resulted in a tighter association of the IB 11 antigen with the cytoskeleton. This would suggest a role for 1B11 antigen in locomotion, cytokinesis or adhesion. The failure to demonstrate cytoskeleton-associated IB 11 antigen in resting spleen suggests that IB 11 antigen is more consistently associated with cytoskeleton in T-lymphocytes and that in a more heterogeneous population such as the spleen, the presence of B-lymphocytes and myeloid cells expressing IB 11 may obscure this association. Alternatively, resting thymus may show a 140 low level of activation consistent with on-going selection processes. We were also able to demonstrate that IB 11 antigen localises to uropods on some activated thymocytes. This would suggest a role in locomotion or adhesion and is consistent with the finding of an association with the cytoskeleton in thymus. Studies of the role of IB 11 antigen in adhesion were largely uninformative. It is of interest that adherence to the endothelial cell monolayer was proportional to IB 11 expression by various cell lines. However, this may also reflect a greater incorporation of thymidine by high IB 11 antigen expressing cells. Nevertheless 1B11 mAb failed to inhibit adhesion of the T-cell line HT-2Y to the endothelial monolayer. This may be due to the limitations of this particular assay for determining adhesion or it may be due to a lack of neutralising properties for this antibody. Adhesion studies in mouse are restricted by the difficulty in obtaining sufficient endothelial cells to generate a useful monolayer to study adhesion. We took advantage of a transformed endothelial cell line, Endo-Dl which is derived from the Ipr/fas mouse. This transformed endothelial cell line shows the ultra-structural properties of endothelial, cells, however its expression of adhesion molecules may not conform to that observed in vivo. Fdr example, Endo-Dl must be used as a semi-confluent monolayer as it does not exhibit contact inhibition and we observed no increase in adhesion in response to either IL-ip or IL-4. Further complications in demonstrating inhibition of adhesion arise because adhesion is a multi-stage process and depends on the concerted expression of a variety of adhesion molecules over a given time frame and in response to specific cytokines (276). There are also differences in the relative contributions made by different adhesion molecules in static assay systems and under conditions of flow (277, 278). A failure to demonstrate inhibition of adhesion by 1B11 mAb may therefore also arise because the appropriate ligand has not been induced, the contribution of the ligand to overall adhesion may be low or because the affinity of the interaction of IB 11 antigen with its ligand may be too high to allow for effective competition by the mAb at these concentrations. Similarly 1B11 mAb failed to inhibit either conjugate formation or cytotoxicity, whereas anti-CD8 antibody and anti-LFA-1 antibody produced modest inhibition of these processes, consistent with previous reports in the literature (280). This is a disappointing result because the 141 high levels of IB 11 antigen expressed in this system would support a role in cytotoxicity. Again ' the possible explanations include lack of neutralising properties for the antibody or the presence of a high affinity interaction which cannot be overcome by antibody. Alternatively IB 11 antigen may make a relatively minor overall contribution to the engagement process compared with other adhesion molecules. . Finally IB 11 antigen expression was not associated with a distinct homing pattern for . activated lymphocytes. Following injection of labeled activated lymphocytes, themajority of these cells returned to the spleen.which was their original tissue of origin and a slightly smaller number entered mesenteric lymph nodes and fewer still entered axillary lymph nodes. When the injected CD4+ cells were re-examined 24 hours after injection they were found to have lower levels of 1B11 antigen expression compared to baseline values. The decreased level of IB 11 antigen expression observed is unlikely to be due to down-regulation as we; have evidence for long-term maintenance of levels of IB 11 expression in culture for up to 1 week following a-methyl mannoside reversal (our own observations). Rather it suggests that either CD4+1B11+,cells have a,reduced life span in vivo or that they home to a tissue not examined in these experiments. If IB 11 expression is associated with decreased cell viability then this has to be as a result of encountering a ligand Which is not normally present in long-term spleen cultures. In summary studies of IB 11 antigen function have been disappointing. We have been unable to.demonstrate a role for IB 11 antigen in adhesion or cytotoxicity using the systems described. However no firm conclusions can be drawn as to the role of IB 11 antigen in these effector functions as the IB 11 mAb may riot bind at a site which inhibits the function of the target antigen. Studies of lymphocyte homing and turnover are easier to interpret, in that they do not rely on the neutralising properties of the antibody. Here we were unable to determine the fate of CD4+ lymphocytes expressing high levels of IB 11 antigen as these cells could not be detected in the tissues examined. This would suggest that either IB 11 antigen expressing CD4+ lymphocytes are excluded from secondary lymphoid tissues or that they have a reduced survival in vivo. This area warrants further study. 142 VII. The Biochemistry of IB 11 antigen VII.l. Introduction and Aims Previous chapters have outlined the distribution and regulation of IB 11- antigen expression. In this chapter we performed experiments to determine the biochemistry of the antigen in an attempt to gain sufficient information about the antigen to enable its identification. Experiments were performed to determine the biochemical structure of the antigen using simple biochemical methods. Since monoclonal antibodies may react with a variety of structures, we initially determined that the antigen was a protein. We subsequently examined whether the protein comprised single or multiple chains and whether there were any other associated proteins. Since the antigen appeared to recognise a glycoprotein, we attempted to determine what carbohydrate structures were present on the molecule and whether! any of these were additionally modified. We also performed studies of turnover of the antigen and 2D mapping to determine its pi. From the results of these experiments we were able to design further experiments which identified the antigen. The antigen appeared to be a unique isoform of the known glycoprotein, CD43 (Ly 48) which had been generated by. carbohydrate modification of that protein. Further experiments were therefore performed to determine how the epitope recognised by the antibody is formed. 143 VII.2. Results VII.2.1 1B11 is a single chain glycoprotein with no intra-chain disulphide ,;. bridges ' . . ' • ' „ . ' IB 11 antigen, immunoprecipitated from whole cell lysate could be resolved by SDS PAGE gel and detected by silver staining, indicating that the antigen is a protein. In addition, the 1B11 , antigen bound to a WGA column indicating the presence of N-linked glycosylation and/or sialic acid residues,in the molecule and could be eluted with N-ac.etyl glucosamine (Figure 36). The latter feature is characteristic of N-glycosylated proteins. In Western immunoblots, IB 11 antigen often showed a characteristic tracking of the protein to the outer margins of the lane which is also characteristic of heavily glycosylated proteins. When 35S-labeled IB 11 antigen was immunoprecipitated from whole cell lysates prepared under weak detergent conditions, 1B11 antigen precipitated as a single band with no associated proteins, excepting those which were also detected in control preparations (Figure 37A). This would indicate that IB 11 antigen is a.single chain molecule and indicates that it is unlikely that IB 11 belongs to the integrin family of surface adhesion molecules which are heterodimers., When whole cell lysates were resolved in the presence or absence of (3-mercaptoethanol, a reducing agent, effective in cleaving disulphide linkages, there was no shift in the apparent molecular weight.of the 1B.11 antigen, suggesting that there were no intra-chain or inter-chain disulphide linkages present (Figure 37B). VII.2.2. Deglycosylation studies: VII.2.2.1. N-linked glycosylation of the 1B11 antigen makes little or no contribution to its apparent molecular weight Based on the finding that IB 11 antigen bound to WGA columns and could be eluted with N-acetyl glucosamine we attempted to determine the extent of the N-linked glycosylation on the 1B11 glycoprotein. Three methods of deglycosylation were attempted, a) tunicamycin treatment 144 Figure 36 IB 11 antigen binds to wheat germ agglutinin. 200 kDa 115 kDa 98 kDa a b e d Whole cell lysates were prepared from D26 cell line in the presence of protease inhibitors. Lysates were applied to a 1ml WGA column and the flow through and washes collected. Glycoproteins bound to the column were then eluted with N-acetyl glucosamine and eluate was collected in 1 ml fractions. All fractions were then resolved on 7.5% SDS PAGE gel and IB 11 mAb immunoreactivity was detected by immunoblotting. Lane a Original lysate. Lane b Flow through fraction. Lane c Eluate fraction 1. Lane d Eluate fraction 2. 145 Figure 37. IB 11 mAb recognises a single chain molecule with no disulfide bridges. A CTL cells were labeled overnight with 0.5 mCi of 35s-methionine. Cells were then washed and lysed in 0.5% CHAPS, 150fnmol/l NaCl, 20 mmol Tris, pH 7.4 in the presence of protease inhibitors. Lysates were then pre-cleared for 30 minutes with 50% slurry of protein G andthen incubated with either IB 11 mAb 20 fig/ml a or control mAb 20 ixg/ml b at 4°C for 1 hour. After 1 hour, a 30% slurry of protein G beads was added and the mixture was rotated at 4°C for 1.5 hours. Beads were then washed three times with PBS/5%BSA and bound proteins eluted with 0.1 M glycine pH 2.5 and neutralised with saturated Tris base. Eluates were diluted into sample buffer and resolved on 7.5% SDS PAGE gels. 35s-signal was enhanced following staining, by immersing gels in Amplify (TM) and gels were autoradiographed using Kodak XAR film. B Whole celllysates were prepared from D26 using standard lysis buffer. Samples were then resolved on 7.5% SDS PAGE gels in the presence or absence of 10% v/v (3-mercaptoethanol. IB 11 immunoreactivity in lysates was determined by immunoblotting. Typical results are shown, a Non-reduced, b Reduced 146 Figure 37 — 200 kDa 115 kDa — 98 kDa 68 kDa B 200 kDa 115 kDa 98 kDa 1 4 7 of whole cells, b) swainsonine treatment of whole cells followed by endoglycosidase H digestion and c) N-glycanase treatment of IB 11 immunoprecipitates. . We were unable to analyse the effects on the 1B11 antigen of tunicamycin because of extremely low viability of cells treated with tunicamycin. Toxicity was seen at all doses used (100 ng/ml to 10 (ig/ml ), with different tunicamycin preparations and in 2 different IB 11 antigen expressing cell lines, D26 and CTLs. Swainsonine treatment did not result in loss of cell viability, (Figure-38A). Incubation of lBli immunoprecipitates overnight with endoglycosidase H resulted in approximately 5 kDa shift in the apparent molecular weight of the IB 11 antigen when resolved on SDS PAGE gel (Figure 38B). In some gels no apparent shift in molecular weight was Observed. N-glycanase treatment of IB 11 immunoprecipitates also resulted in little or no shift in the apparent molecular weight of IB 11 antigen (Figure 39). Interleukin 3 receptor p-chain immunoprecipitated from the same cell line under the same conditions and digested simultaneously with the same enzyme showed a 20 kDa decrease in apparent molecular weight. (Figure 39). This indicated that the failure to Observe a significant shift in the molecular weight of the IB 11 antigen on removal of N-linked carbohydrate was not a result of loss of enzyme activity. . These results suggest that although IB 11 antigen binds to WGA, N-linked glycosylation accounts for no more than 5 kDa of the molecular weight of the molecule. VII.2.2. 1B11 antigen is a sialylated glycoprotein which shows reduced mobility on SDS PAGE after removal of sialic acid The failure to demonstrate significant N-linked glycosylation on the IB 11 antigen lead to an attempt to demonstrate O-linked glycosylation on the molecule. IB 11 antigen immunoprecipitated from both D26 and CTL cell lines was digested initially with neuraminidase to remove terminal sialic acid residues and then with O-glycanase and resolved on SDS PAGE gel. Treatment of the immunoprecipitates with neuraminidase lead to an apparent increase in the. molecular weight of the IB 11 antigen from 130/140 kDa (D26/CTL) to 180 kDa, whereas subsequent treatment with O 148 Figure 38. IB 11 antigen is resistant to swainsonine /endoglycosidase H digestion. A D26 cell line was established in duplicate cultures at a density of 2.x 10^  cells per nil. 2 Ug/ml swainsonine was added to one culture. At set time intervals, aliquots of cells were removed and counted. Results are expressed as a mean of two counts. B D26 cell line was incubated in culture for 48 hours in the presence Of 2 jxg/ml of swainsonine. After 48 hours cells were pelleted, washed and lysed in standard lysis buffer. 1B11 antigen was then immunoprecipitated from the lysate . IB 11 immunoprecipitates were diluted initially into sample buffer containing 1% SDS and 5 mmol DTT (final concentrations). Samples were further diluted-5•times into 0.2M citrate buffer pH 5.5. 1U of endoglycosidase H was added per 25 ul sample to one of duplicate samples and the immunoprecipitates were incubated overnight at 37°C. At the end of the incubation period 10% ice cold TCA was added to each sample and the samples incubated a further 48 hours at 4°C. TCA precipitates were then pelleted and washed with acetone and vacuum dried. Samples were resuspended in sample buffer and resolved on a 7.5% SDS PAGE gel. IB 11 immunoreactivity was then determined by immunoblotting. Lane a.Control: Lane b Swainsonine & endoglycosidase H treated: 149 Figure 38 115 k D a a b 150 Figure 39 N-linked glycosylation of the IB 11 antigen makes little or no contribution to its apparent molecular weight. 200 kDa 115 kDa 98 kDa 66 kDa a b c d Whole cell lysates were prepared as described previously from C[33, a CTL cell line transfected with a construct which expresses the IL-3 receptor p sub-unit AIC-2A. IB 11 antigen was immunoprecipitated from the lysates using Protein G and 1B11 mAb as previously described. AIC-2A was immunoprecipitated from lysates using 4F1 mAb and Protein G. Immunoprecipitates were de-salted over a G25 column and buffer exchanged into 15 mmol Na3PC»4, pH 7.5. Immunoprecipitates were then boiled in the presence of 10 mmol phenanthrolene hydrate, 50 mmol P-mercaptoethanol and 1% SDS. Samples were then diluted into 1% NP-40 and 50U/ml N-glycanase was added to both samples and the samples incubated overnight at 37°C. Following incubation, digests were diluted into SDS sample buffer and resolved on 7.5% SDS PAGE gels. IB 11 antigen was detected in immunoblots using IB 11 mAb and AIC-2A was detected in immunoblots using an anti-peptide antibody. Lanes a & b 1B11. Lanes C & d AIC-2A. Lanes a & C Controls. Lanes b & d N-glycanase treatment. 151 glycanase lead to no apparent change.in the molecular weight of the molecule (Figure 40). Enzymatic digests were repeated in the presence of 6M Urea to determine whether the de-sialylated glycoprotein was aggregating. However, the same phenomenon was also observed in.these preparations (data not shown). Although O-glycanase digestion failed to produce a further shift in the apparent molecular weight of the IB 11 antigen, this did not appear to be due to decreased activity of the O-glycanase enzyme, but rather appeared to be as a result of some inherent resistance to digestion of the desialylated molecule. CD45 immunoprecipitates from the same cell line were readily digested by O-glycanase under the same conditions and showed a characteristic 100 kDa decrease in apparent molecular weight (Figure 40). This apparent increase in molecular weight oh desialylation is characteristic of only a few heavily O-glycosylated glycoproteins, specifically glycophorin and CD43 (280). VII.2.3. 1B11 m A b recognises a 130 k D a isoform of C D 4 3 We had previously demonstrated that the IB 11 mAb showed a different pattern of immunoreactivity on cell lines and on T-lymphocytes sub-sets when compared with the commercially available anti-CD43 antibodies and recognised a protein with a higher molecular weight than reported for the commercial preparations. However, we speculated that IB 11 mAb might be recognising an isoform of CD43 which was differentially expressed by T-lymphocyte subsets and which was up-regulated on T-lymphocyte in response to activation. There was some precedent for this in the literature in that a number of the antibodies raised against CD43 also recognise a 130 kDa isoform of the molecule which appears only on activated lymphocytes in addition to the 115 kDa resting form (177,281). In the human CD43 system a monoclonal antibody, T305 (282-284), is reactive with a 130 kDa activation isoform of CD43 and appears from references in the published data to be differentially expressed by T-lymphocyte sub-sets. To test this hypothesis, we obtained an L-cell line from the laboratory of Dr. Fumio Takei, which had been transfected with murine CD43 and prepared whole cell lysates for Western immunoblotting. In addition we performed FACS analysis On transfected and non-transfected cells using the S7 (anti-CD43) and IB 11 mAbs. 152 Figure 40 1B11 antigen is resistant to the action of O-glycanase, but responds to neuraminidase treatment with an aberrant shift in mobility on SDS PAGE gel. CTL cell line was labeled overnight with O.lmCi/ml 35s_memionine. Whole cell lysates were prepared as described previously. 1.B11 antigen was immunoprecipitated from lysates using Protein G and IB 11 niAb as described previously. CD45 was immunoprecipitated from cell lysates using protein G and Rabbit ant-CD45 antibody. Immunoprecipitates were incubated with 20 MU neuraminidase in a reaction volume of 50 uT in the presence of 10 mM calcium acetate. After 1 hour incubation at 37°C, immunoprecipitates were desalted over a G25 column and buffer exchanged into 15 mM Na3P04, pH 7.5. 2 mU 6-glycanase was then added to 30 |il of the digest and the reaction incubated overnight at 379C. Following incubation with O-glycanase, samples were made up in sample buffer and resolved in duplicate on 7.5% SDS PAGE gels. One duplicate gel was amplified, following staining of molecular weight markers, and subjected to autoradiography. The second gel was transferred to nitrocellulose and probed with IB 1 ImAb arid anti-CD45 mAb. Panel A: Immunoblots 1CD45. 2 IB 11. Lanes a control b Neuraminidase treated C 6-glycanase treated Panel B: Autoradiographs: Lanes a - C CD45. Lanes d - f IB 11. Lanes a & d. control. Lanes b & e Neuraminidase treated. Lane C & f O-glycanase treated. 153 Figure 40 4 * j f y 200 kDa 200 kDa 115 kDa « " ' 115 kDa — 98 kDa 98 kDa a b c a b c B 200 kDa 115 kDa 98 kDa a b c d e f 1 5 4 In FACS analysis, both IB 11 mAb and S7 bound to transfected but not to non-transfected control cells (Figure 41A). In Western immunoblots both mAbs recognised a protein band in transfected cell lines but not in control cells (Figure 41B). Interestingly, the bands recognised by 1B11 mAb and S7 mAb were quite broad and non-overlapping. 1B11 mAb recognised proteins with apparent molecular weights of 115 to 130 kDa, whereas S7 recognised proteins with apparent molecular weights of 100 to 115 kDa, suggesting that the epitope recognised by the respective mAbs was determined by the glycosylation state of the molecule and that the 1B11 mAb recognises the higher molecular weight forms. Such a broad range of molecular weights for the CD43 molecule is not usually observed in cell lines or in primary cells which express CD43 under physiological conditions. This suggests that the pattern of glycosylation of the CD43 molecule expressed in a fibroblast line is abnormal. VII.2.4. The pattern of expression of the T305 antigen resembles that of IB 11 antigen The T305 mAb recognises a 130 kDa isoform of CD43 which is expressed by activated T-lymphocytes, and at particularly high levels in cytotoxic effector cells (282). The molecule is aberrantly expressed at high levels in auto-immune disease, in HIV infection and in the hereditary immunodeficiency disease, Wiskott-Aldrich syndrome (283). Unfortunately no FACS data comparing T305 immunoreactivity in T-lymphocyte sub-sets have been published therefore we decided to examine T305 immunoreactivity within T-lymphocyte sub-sets and in PHA activated T-lymphocytes from human peripheral blood. Results are shown in Figure 42. T305 antibody bound to human CD8+ lymphocytes at low levels and showed little or no binding to CD4+ lymphocytes. In activated lymphocytes, T305 showed very high levels of immunoreactivity with activated CD8+ lymphocytes and increased levels of immunoreactivity with CD4+ lymphocytes. This differential pattern of expression superficially resembles that observed in the murine system for IB 11. 155 Figure 41. IB 11 mab recognises high molecular weight isoforms of CD43. A Whole cell lysates were prepared from a murine L-cell line transfected with a cDNA expressing CD43 and wild type control cells. Lysates were resolved on 7.5% SDS PAGE gels and immunoblotted with S7 m Ab (ahti-CD43) (lanes 1 & 2) and 1B11 mAb (lanes 3 & 4): Lanes 1 & 3 Control: Lanes 2 & 4 Transfected cells: B Murine L-cell line transfected with a cDNA expressing CD43 and wild type control cells were subject to FACS analysis using FITC-labeled S7 mAb (anti-CD43), FITC-labeled IB 11 mAb and control antibody. Panels a & b transfected cells: Panels C & d Control cells: Panels a & C S.7 mAb: Panels b & d IB 11 mAb: Control mAb IB 11/S7 mAb 156 Figure 41 is1 162 Fluorescence 10 1 li Fluorescence TP" Fluorescence d TP lP~ Fluorescence 1 5 7 Figure 42. T305 mAb recognises an antigen oh human T-lymphocytes with a similar distribution to IB 11 mAb. Human lymphocytes were harvested by centrifugation of peripheral blood cells over Ficoll-Hypaque. Purified lymphocytes were activated in culture for 48 hours in the presence of 5 |lg/ml PHA. Following activation lymphocytes were washed and incubated in the presence of either PE-labeled anti-CD4, FITC-labeled anti-CD8 and unlabelled T305, followed by either FITC anti-mouse IgG or PE-labeled anti-mouse IgG.and then subject to FACS analysis. Freshly harvested lymphocytes were used as a control. Panel A Control CD4 lymphocytes: Panel B Control CD8 lymphocytes: Panel C PHA activated CD4: Panel D PHA activated CD8: 158 Figure 42 159 VII.2.5. Role of sialic acid in generating the 1B11 epitope: VII.2.5.1 1B11 immunoreactivity is enhanced by removal of terminal a 2-3 linked sialic acid residues Immunoreactivity with the T305 antibody has been demonstrated to be critically dependent on the presence of sialic acid, and neurominadase pre-treatment abolishes T305 immunoreactivity (284). Accordingly we attempted to determine whether removal of sialic acid would abolish IB 11 immunoreactivity. FACS analysis was performed on neuraminidase treated viable CTL cells which normally show high levels of IB 11 immunoreactivity. FITC labeled WGA was used to determine the extent of de-sialylation. (WGA has been shown to have a low affinity for sialic acid residues). The optimum dilution of the WGA-FITC was determined prior to staining by comparing the FACS signal of serial dilutions of WGA-FITC on CTL cells (data not shown). Surface treatment of CTL cells with neuraminidase resulted in no shift in the binding of the IB 11 mAb, however there was a decrease in the binding of FITC labeled WGA to the CTL cells, indicating that de-sialylation had occurred (Figure 43A-C). These results suggested that in comparison to T305 mAb, sialic acid is not critical to the binding of IB 11 mAb.. We then compared IBM mAb binding in FACS, before and after treatment with neuraminidase for a cell line which did not express significant levels of 1B11 immunoreactivity, but did express the S7 reactive isoform. Results are shown in Figure 43D-F. Neuraminidase digestion was performed on viable EL4 cells arid binding of IB i 1 mAb, 3E8 mAb (an anti-CD43 mAb which recognises both 115 kDa and 130 kDa isoforms of the CD43 molecule), S7 and WGA-FITC compared for treated and mock-treated controls. Surprisingly, IB 11 immunoreactivity was enhanced by pre-treatment of cells with neuraminidase. S7 and WGA-FITC reactivity were reduced by neuraminidase treatment and 3E8 immunoreactivity was unaltered. This suggests that the conformational epitope recognised by 1B11 can be induced in non-IB! 1 reactive isoforms by removal of sialic acid. It does not preclude the possibility that the CD43 isoform recognised by 1B11 mAb is the murine equivalent of T305, even though the epitopes which are recognised by the two mAbs are dissimilar. 160 Figure 43. <•' Neuraminidase treatment enhances IB 11 mAb immunoreactivity. CTL cells and EL4 cells were washed arid suspended in Hanks balanced salt solution, 20 mmol Tris pH6.8, 10 mmol calcium acetate at 10^  cells per ml. Cells were incubated for 1 hour at 37°C in the presence of 25 mU of neuraminidase. At the end of the incubation, the cells were pelleted, washed and resuspended in FACS buffer. Neuraminidase treated and mock treated cells were then labeled with FITC-labeled IB 11 mAb, FITC-labeled control mAb, FITC-conjugated WGA or FITC-labeled S7 mAb and subjected to FACS analysis. Panels A-C CTL cells: Panels D-F EL4 cells; H I Mock-treated. Q Neuraminidase treated. 161 Figure 43 B 162 Figure 44 IB 11 mAb immunoreactivity is enhanced by p re-treatment with Newcastle Disease virus neuraminidase. EL-4 cells were washed and resuspended at 10^  cells per ml in Hanks balanced salt solution in the presence of 10 mmol/1 calcium acetate, 20 mmol/1 Tris pH 6 and 200U/ml Newcastle Disease virus neuraminidase. Cells were incubated for 1 hour at 37°C and then washed and subjected to FACS analysis using FITC-labeled control mAb, FITC-conjugated WGA, FITC-labeled 1B11 mAb or FITC-labeled S7 mAb. FACS profiles were also obtained on mock treated cells. Panel A Control: Panel B WGA: Panel C IB 11 mAb: Panel D S7 mAb: 163 Figure 44 o U N e u r . C o n t r o l B o 13 U N e u r . C o n t r o l 1 S J Control W G A C o n t r o l N e u r . D N e u r . C o n t r o l -t i i i T i i r - \ • t i <• t j I No. 1 I 1 n i -4 I i i i t t V / I ' / T—1 U I 1 1 I J «H T*-, 10" o 13 U 1B11 S7 164 In light of the finding that neuraminidase treatment enhances IB 11 immunoreactivity in the EL4 cell line, we compared the effect of pre-treatrhent of cells with Newcastle disease virus neuraminidase (285) on 1B11 mAb immunoreactivity. This enzyme removes predominantly a 2-3 linked sialic acid residues. Results are shown in Figure 44. Pre-treatment of EL4 cells with Newcastle Disease virus Neuraminidase resulted in a similar increase in IB 11 immunoreactivity as was observed for standard neuraminidase. However, S7 binding was less markedly affected than when standard neuraminidase was used. This suggests that 2-3 linked sialic acid residues are important in masking the IB 11 epitope, but are not a significant part of the S7 epitope, In addition to studying the EL4 cell line, we also compared IB 11 immunoreactivity before and after neuraminidase treatment of splenocytes, in order to determine whether IB 11 immunoreactivity could be induced on CD4+ lymphocytes by treatment with neuraminidase. Results are shown in Figure 45. We were able to show that following neuraminidase treatment, IB 11 immunoreactivity increased up to 200 fold on both CD4+ and CD8+ cells as measured by MFI in FACS analysis. In contrast S7 immunoreactivity decreased 2-7 fold on both CD4+ and CD8+lymphocytes. VII.2.6. 1B11 antigen does not cross-react with sialyl-Lewis X antigen in immunoblots Expression of the T305 antigen has been demonstrated to be dependent on expression of the enzyme UDP-GlcNAc:Gaipi-3GalNAc-R pi-6-A/-acetylglucosaminyltransferase (core 2 GlcNAc-transferase, C2GnT) and the T305 antigen has been demonstrated to contain the core 2 carbohydrate structure in carbohydrate analysis (286). We have been able to show that 1B11 antigen expression correlates in cell lines with C2GnT enzyme activity (287), suggesting that IB 11 mab binding may .reflect expression of the core 2 carbohydrate structure. It has been suggested in human cell lines, that this structure may be subject to additional modification and may incorporate the sialyl Lewis X epitope, making this form of CD43 a potential ligand for selectins (288). In the 165 Figure 45. 1B11 immunoreactivity in splenocytes is enhanced by pre-treatment with neuraminidase. Single cell suspensions were prepared from BALB/c spleen as described and resuspended at 1()7 cells per ml in Hanks balanced salt solutionin the presence of 10 mmol/l calcium acetate, 20 mmol/l Tris pH 6.8 and 250U/ml neuraminidase. Cells were incubated for 1 hour at 37°C and then washed and subjected to FACS analysis using PE-labeled anti-CD4 or PE-labeled anti-CD8, FITC-labeled Control mAb, FITC-labeled IB 11 mAb or FITC-labeled S7 mAb. Gated CD4+ and CD8+ populations were examined with respect to their binding of Control mAb WGA, IB 11 mAb and S7 mAb.. FACS profiles were also obtained on mock treated cells. Panels (A, C, E ) CD4+ splenocytes. Panels (B,D,F,) CD8+splenocytes. Panels (A-B) Control; Panels (C-D) 1B11: Panels E - F S7: fl| Controls. \~[ Neuraminidase treated. 166 Figure 45 157 Figure 46. Cell lines expressing high levels of IB 11 antigen do not express the Sialyl Lewis X epitope. Single cell suspension were prepared from D26, CTL and HL-60 cell lines. Cells were incubated with either SN3 mAb or SN4 mAb followed by FITC labeled goat anti-mouse IgG. Binding of anti-sialyl Lewis X mAbs was compared with control mAb. Panels A & B D26 cells: Panels C & D CTL cells: Panels E & F HL-60: .Panels (A, C, E ) SN3 mAb: Panels (B, D , F) SN4 mAb: : ' 168 Figure 46 1 6 9 murine system very few selectins have been characterised and the significance of sialyl Lewis X has not been demonstrated. We attempted to determine whether 1B11 antigen immunoprecipitated . from D26 and from CTL cells would immunoblot with SN3 and SN4 antibodies which recognise sialyl-Lewis X (157). We were unable to demonstrate any immunoreactivity (data not shown) and in fact the anti-sialyl Lewis X antibodies did not bind to D26 or CTL cells in FACS analysis, although they did • bind to HL-60 cells (Figure 46). The data suggest that the sialyl Lewis X epitope is not present in the IB 11 antigen. VII.2.7. IB 11 antigen is not a sulphated glycoprotein Studies of a murine lymphoma cell line have suggested that the CD43 molecule may be sulphated and that sulphation may occur only on those molecules expressing the core 2 carbohydrate structure (160,289). To investigate the possibility that the IB 11 antigen was sulphated, we labeled CTL cells with 35S-sulphate and immunoprecipitated IB 11 from whole cell lysates. Results are shown in Figure 47 . No radioactivity could be detected in immunoprecipitates even after exposure of auto radiographs for 6 weeks. Successful immunoprecipitation of the IB IT antigen was confirmed by Western immunoblotting of immunoprecipitates. Whole cell lysates did however show some incorporation of labeled material in a broad band which barely entered the resolving gel and which was apparent after 4 days exposure of auto radiographs. The data suggest that at least in CTL cells, the IB 11 antigen is not sulphated. VII.2.8. The pi of the 1B11 antigen in activated spleen cells is close to 6 CD43 is a heavily sialylated glycoprotein and is reported to be the single largest contributor to net surface negative charge on lymphocytes. The pis reported for CD43 in the literature range from 4-5.7 on various cell lines and show considerable heterogeneity depending on the antibody used to detect the antigen and the cell line from which it is derived (178, 281). We were interested to determine the pi of the IB 11 isoform expressed by activated T-lymphocytes and accordingly 170 Figure 47. The high molecular weight isoforms of CD43 recognised by 1B11 mAb do not contain sulfate. CTL cells were washed with sulfate free media containing 1 % FCS and resuspended at 10^  cells per ml. Cells were then incubated for 4 hours in sulfate-free media containing 150 mCi/ml 35S-Sulfate and 20 mmol HEPES pH7.2 at 37°C. After 4 hours, cells were washed twice with ice cold PBS and lysed in standard lysis buffer. IB 11 mab was used to immunoprecipitate its target antigen from the labeled lysates and control immunoprecipitations were performed simultaneously. Immunoprecipitates and whole cell lysates were resolved in duplicate on 7.5% SDS PAGE gels. Gels were divided and one replicate was transferred to nitrocellulose and subject to immunoblotting with IB 11 mAb to determine the efficiency of the immunoprecipitation. The second replicate was stained and then amplified and subjected to autoradiography. Panel A .Autoradiograph: Lane a Whole cell lysate; Lane b 1B11 immunoprecipitate; Lane c control immunoprecipitate. Panel B Immunoblot: Lane a Whole cell lysate; Lane b IB 11 immunoprecipitate; Lane c control immunoprecipitate; 171 Figure 47 200 kDa 115 kDa 98 kDa 66 kDa 45 kDa a b c 200 kDa 115 kDa 98 kDa 66 kDa 45 kDa a b 1 7 2 resolved 3 5S labeled 1B11 immunoprecipitates derived from activated splenocyte cultures on 2-D mini-gels. The cells expressing IB 11 were characterised prior to lysis and were shown to comprise of 90% T-lymphocytes, 50% of which were reactive with IB 11 mAb at levels above control antibody (Figure 48). IB 11 immunoprecipitates were compared with control antibody immunoprecipitates, in order to identify the IB 11 immunoreactive spot, and with 2-D standards resolved in the same experiment. Whole cell lysates were not used as it was felt that with the limitations on the amount of material that could be loaded on the gel, implicit in this 2-D system, it would be difficult to detect the IB 11 antigen. Immunoblotting of the immunoprecipitates was attempted but was unsuccessful, presumably because of the small quantities of protein present. Results are shown in Figure 49 . 2D analysis of 1BH Immunoprecipitates revealed a single unique spot on 2-D gels corresponding to a molecular weight of 130 kDa in IB 1.1 immunoprecipitates but not in control immunoprecipitations. The pi of this spot was approximately 6. Simultaneous experiments using D26 cell lysates revealed a spot with a similar pi (data not shown). The data suggests that the isoform of CD43 recognised by IB 11 in activated splenocytes does not carry a strong negative charge, making it a possible candidate for an adhesion molecule. VII.2.9. Pulse chase experiments suggest the turnover of the IB 11 antigen is approximately 45-60 minutes Studies of IB 11 antigen turnover were performed using the CTL cell line which expresses high levels of the IB 11 antigen. Following a 10 minute pulse with 35S-methipnine, aliquots of cells were removed at various time intervals, lysed and IBM immunoprecipitations performed. Results are shown in Figure 50 . Although there was unfortunately a strong background signal on these gels, the maximum labeling of 1B11 antigen was detected at 15 minutes and was diminished close to baseline by 60 minutes No low molecular weight forms reactive with 1B11 mAb were observed, suggesting that the conformation recognised by 1B11 is dependent on late modifications in the carbohydrate 173 Figure 48 Characterisation of the lymphocyte population used to purify IB 11 antigen for 2-D gel analysis. Single cell suspensions were prepared from BALB/c spleens as described. Cells were activated in culture with 5 |lg/ml Con A for 48 hours at 37°C, 5% C02. At the end of the 24 hour period, viable cells were isolated over Ficoll-Hypaque and FACS analysis was performed on an aliquot of cells using PE-labeled anti-CD4, biotinylated anti-CD8, followed by cychrome, FITC-labeled IB 11 mAb or control antibody. Overall IB 11 mAb immunoreactivity and. IB 11 immunoreactivity within gated subpopulations was determined. Lymphocytes were subsequently labeled overnight with 3 -^methionine and IB 11 immunoprecipitates analysed on 2D gels. Panel A CD4/CD8 profiles: Panel B 1B11 immunoreactivity in CD4+ lymphocytes: Panel C IB 11 immunoreactivity in CD8+ lymphocytes; Panel D Overall IB 11 immunoreactivity: In Panels (B-D) binding is shown relative to the binding of control antibody. 174 Figure 48 r mil mn 175 Figure 49. The pi of the high molecular weight isoform of CD43 recognised by 1B11 mAb is 6.0. Single cell suspensions were prepared from BALB/c spleens as described. Cells were activated in culture with 5 p:g/ml Con A for 48 hours at 37°C, 5% C02. At the end of the 48 hour period, viable cells were isolated over Ficoll-Hypaque. Viable cells were the washed 3 times in methionine free media and incubated overnight in the presence of lOU/mlTL-2 and 0.05 mCi/ml of 35S -methionine. After 16 hours incubation, cells were pelleted and lysed using standard lysis buffer! IBM mAb was then used to immunoprecipitate its target antigen from the whole cell lysate as previously described. Control immunoprecipitations were performed in parallel on equal volumes of lysate. Immunoprecipitations were then resolved on 2-D mini-gels containing standard ampholytes and subject to autoradiography. Panel A Control immunoprecipitation: Panel B 1B11 immunoprecipitation: 1B11 antigen is indicated by the arrow. Typical results are shown. 176 Figure 49 -200 kDa -115 kDa -98 kDa -66 kDa 45 kDa -200 kDa -115 kDa -98 kDa -66 kDa -45 kDa 177 Figure 50 I B 11 mAb recognises mature forms of the high molecular weight isoform of CD43, CTL cells were washed three times in methionine-free media and pre-incubated in methionine-free media containing lOU/ml IL-2 for 1 hour. Thereafter cells were labeled for 10 minutes in the presence of lmCi 3 -^methionine per 10^  cells at 37°C. At the end of the incubation period, the cells were pelleted , washed three times and then resuspended in RPMI containing 10 U/ml IL-2, 20uM (3-mercaptoethanol and 5% v/v FCS. At various time intervals aliquots of cells were removed, pelleted and lysed in standard lysis buffer. These lysates were then subject to immunoprecipitation using IB 11 mAb. Immunoprecipitates were resolved in duplicate on 7.5% SDS PAGE gels. One half of each gel was then transferred to nitrocellulose and immunoblotted with IB 11 mAb. The other half was stained and amplified and then subject to autoradiography. Panel A Immunoblot. Panel B Autoradiograph. Time intervals are indicated. 178 Figure 50 200 k D — 115 k D — 98 k D 0 5 10 15 30 45 60 179 structure of the glycoprotein. Similar amounts of total IB 11 (labeled and unlabeled) were immunoprecipitated at each time point as demonstrated by Western immunoblots (Figure 50A). VII.2.10. 1B11 is phosphorylated on serine residues in response to Con A and IL-2 The human CD43 molecule is constitutively phosphorylated on serine residues and phosphorylation is increased in response to activation (290-293). Similar studies have not been performed on murine CD43 as no immuno-precipitating monoclonal antibodies were available. We compared 32P-phosphate incorporation into the CD43 antigen in the HDK-1 cell line activated with IL-2 or Con A. This cell line expresses low levels of IB 11 antigen but can be induced to express IB 11 antigen at higher levels with Con A and thus resembles the in vivo derived CD4+ lymphocyte (Figure 51 A). HDK-1 cells showed low levels of constitutive expression of phosphorylated 1B11 antigen but showed a considerable increase in phosphorylation in response to IL-2 and to Con A. (Figure 5 IB) Phospho-amino acid analysis,revealed that the 3 2P was incorporated exclusively into serine residues (Figure 51C). Therefore the pattern of phosphorylation of murine CD43 appears to be identical to that observed for the human homologue. 180 Figure 51. IB 11 antigen is phosphorylated on serine in response to T-lymphocyte activation. A HDK-1 cells were activated in the presence of 5 Ug/ml Con A for 24 hours and whole cell lysates prepared as described previously. Lysates were standardised so that 40 ug total cellular protein was loaded per lane and were resolved on 7.5% SDS PAGE gels. 1B11 immunoreactivity in lysates determined by immunoblotting using IB 11 mAb. Lane A Control HDK-1. Lane B Con A activated HDK-1. B HDK-1 cells were washed in phosphate free media and resuspended at 10^  cell/ml. Samples were labeled for 3 hours at 37°C with 0.5mCi 3^p p e r \Q7 cells. Cells were then stimulated for 10 minutes with either Medium alone (Lane a) 5ug/ml Con A (Lane b),;500u/ml IL-2 (Lane c) or PMA lng/ml. After stimulation, cells were pelleted and lysed using standard lysis buffer containing protease and phosphatase inhibitors. IB IT antigen was immunoprecipitated from labeled lysates and resolved on 7.5% SDS PAGE gels. Gels were stained and amplified and subject to autoradiography. C Gels containing 32p labeled IB 11 antigen were prepared as described above. Following autoradiography gels, and autoradiographs were aligned and the areas of gel containing the labeled bands were excised. The gel strips were then diced into 1 mm cubes and transferred to Eppendorf tubes and vacuum desiccated. 1ml of TPCK trypsin at a concentration of lOmg/ml in 50 mM NH4C03, pH 8, was added to each tube and incubated for 24 hours at 37°C. Digests were then pelleted and the supernatants transferred into separate Eppendorfs and again vacuum desiccated. The lyophilate was then washed and resuspended in constant boiling HC1. The mixtures were then incubated for 1 hour at 105°C, cooled and dried overnight. Samples were then resuspended in DDW and spotted onto nitrocellulose paper. Samples were resolved by flat bed electrophoresis together with standard phosphopeptides. The latter were identified by ninhydrin. Resolved gels were subject to autoradiography and compared with the standards indicated. Lane a Control. Lane b PMA. Lane c IL-2. Lane d Con A. 181 Figure 51 a b 200 kDa 115 kDa 98 kDa 66 kDa 200 kDa 115 kDa 98 kDa Serine Treonine Tyosine a 182 VII.3. Discussion We have identified the IB 11 antigen as a high molecular weight form of CD43. The murine CD43 molecule was cloned and sequenced by 2 groups in 1990 (176,177). The biochemical characteristics 1B11 antigen are almost identical to those reported in the literature for CD43. CD43 is a single chain molecule. It does not contain cysteine residues, but contains 8 methionine residues and therefore can be radio-labeled with 35s-methionine. The molecule has only 1 possible site for N-linked glycosylation and it is not known in mouse whether this site is glycosylated. We were unable to demonstrate significant N-linked glycosylation in the IB 11 antigen, although in some experiments a small shift of possibly 5 kDa was observed following N-glycanase treatment. In addition, IBM antigen bound to WGA columns and could be eluted with N-acetylglucosamine. These results may be consistent with N-linked glycosylation at a single site. CD43 has 75 potential O-linked glycosylation sites, which may also contain sialic acid (294). Removal of sialic acid from CD43 results in a decrease in its apparent mobility on SDS PAGE. We were unable to determine the extent to which O-linked glycosylation contributes to the molecular weight of IB 11 antigen because of the apparent increase in the molecular weight of IB 11 antigen observed after neuraminidase treatment, and because O-glycanase appeared ineffective in further removing the O-linked carbohydrate. This was not due to defective activity of the enzyme as it was able to remove O-linked carbohydrate from CD45. O-glycanase catalyses the removal of Gal-(3(l-3)-GalNAc core disaccharides attached to serine or threonine residues of glycopeptides to give free oligosaccharides and an unsubstituted serine or threonine group. Prior enzymatic hydrolysis of sialic acid residues is essential since substitutions on either the galactosyl or N-acetylgalactoaminyl residues block activity. It is therefore likely that the IB 11 antigen contains more complex O-linked carbohydrate structures which show substitutions at this site, whereas the CD45 molecule contains significant amounts of the unsubstituted disaccharide molecule (295). 183 The carbohydrate structure of human CD43 has been studied extensively. Two major forms of carbohydrate modification occur on the CD43 molecule (296-299). In resting T-lymphocytes, and erythroid cells the predominant carbohydrate modification is the addition of a di-. sialotetrasaccharide. Activated T-lymphocytes, some T-lymphocyte cell lines and myeloid cells carry a more complex di-sialohexasaccharide substitution. We hypothesise that IB 11 mAb recognises the murine homologue of the hexasaccharide isoform: The molecular weight of the tetrasaccharide isoform of CD43 is reported as 115 kDa, whereas the molecular weight of the human hexasaccharide isoform is 130-140 kDa (282,286). The latter is identical to the molecular weight we have reported for the IB 11 antigen. The hexasaccharide isoform of human CD4 is recognised by the T305 monoclonal antibody (282,286), We have compared the distribution of the T305 antigen with that of IB 11 antigen and have found similarities in terms of differential expression on CD4 and CD8+ lymphocytes and on activated cells. In addition, the generation of the hexasaccharide structure has been shown to be critically dependent on expression of the (3 l-6GlcNAc-transferase enzyme (C2GnT). We have been able to demonstrate a significant correlation between 1B11 antigen expression and C2GnT activity (287). We therefore have considerable supportive evidence to suggest that IB 11 mAb recognises the murine homologue of the T305 antigen. The carbohydrate structure of human CD43 may be further modified in some cells to generate the sialyl Lewis X epitope (288). This modification requires the presence of polylactosamine repeats and occurs exclusively on isoform of CD43 which have been modified by the C2GnT enzyme. We have been unable to demonstrate the presence of the sialyl Lewis x epitope on the IB 11 isoform of CD43. This is perhaps not surprising as the sialyl Lewis X epitope has never been convincingly demonstrated in the murine system, despite the identification of putative ligands such as selectin molecules. We were unable to demonstrate that the IB 11 antigen contains sulphated polylactosamine structures as has been reported for a murine lymphoma cell line (160,289). Sulphation has been shown to be critical for the generation of some selectin ligands (290-293). It is possible that the 184 failure to demonstrate sulphation may be a function of the cell lines used for these experiments and we plan to pursue these studies using activated splenocytes. The pi for the IB 11 antigen in activated splenocytes appears higher than previously reported for CD43 (178, 281). Examination of 2 separate gels revealed a spot at a molecular weight of 130 kDa and with a pi of 6.0, which was not present in the control preparation. Two dimensional analysis of affinity purified IB 11 antigen from D26 cell lysates also revealed a spot with a similar pi (our observations). The turnover of the 1B11 antigen appears to be 45 minutes to 1 hour and no low molecular weight isoforms with IB 11 immunoreactivity were detected. These values are comparable with that reported in the literature for human CD43 (303,304), although the majority of antibodies do recognise low molecular weight precursors. (No studies of murine CD43 have been reported). The failure to detect low molecular weight isoforms of 1B11 antigen suggests that the 1B11 mAb immunoreactivity is critically dependent on late modifications in the carbohydrate structure of the molecule or on a conformation which is dependent on such late modifications. We know that these modifications do not require sialic acid as 1B11 immunoreactivity is not reduced but rather is enhanced by removal of sialic acid. Interestingly Cyster etal (303) have demonstrated in rat that the majority of anti-CD43 mAbs recognised the polypeptide backbone of the molecule despite the presence of extensive glycosylation. This included 3/4 mAbs whose immunoreactivity, was found to be neuraminidase sensitive. Since our antibody is one of only two reported to differentially recognise the high molecular weight isoform of CD43 it is tempting to speculate that our antibody may be recognising a novel carbohydrate structure which is accessible after de-sialylation. We believe this to be unlikely as the carbohydrate structures of CD43 in activated and resting T-lymphocytes have been shown to differ and IB 11 immunoreactivity can be conferred on non-IB 11 immunoreactive cells by removal of sialic acid. An alternative explanation is that IB 11 mAb recognises an area of the polypeptide backbone which is accessible in the hexasaccharide isoform but is masked by sialic acid in the tetrasaccharide form. We have demonstrated that IB 11 antigen is phosphorylated in response to Con A activation, PMA and in response to IL-2. The time frame of these experiments was not sufficiently 185 long for the Con A-induced up-regulation of IB 11 antigen expression to contribute to the observed increase in 32p incorporation. Activation induced phosphorylation of CD43 occurred exclusively on serine residues. This data is consistent with that reported for human CD43 (290-293). The function of IB 11 antigen phosphorylation may be modulation of a CD43 mediated signal, alteration in association with cytoskeletal elements (possibly allowing for 1B11 capping in uropods) or stabilisation of IB 11 antigen in the membrane, increasing the half life of the molecule. However the latter effect cannot account for the mechanism of IB 11 antigen up-regulation in response to activation, as PMA and EL-2 induced phosphorylation but did not up-regulate IB 11 antigen expression. In summary, IB 11 mAb recognises a high molecular weight isoform of CD43 with a relatively neutral pi. The turnover and phosphorylation pattern of the molecule resembles that of human CD43 (304). Using comparisons with human CD43 we have considerable evidence to support the hypothesis that IB 11 mAb recognises the murine homologue of the T305 antigen. The IB 11 antigen does not appear to recognise a CD43 isoform which has been modified by incorporation of sulphate or by generation of the sialyl Lewis X epitope. Identification of the precise carbohydrate structure of the IB 11 reactive isoform of CD43 will require detailed carbohydrate analysis. 186 VIII. Comparison of CD43 glycoform expression in vivo and in vitro VIII.l. Introduction and aims The demonstration that IB 11 mAb recognised the 130 kDa glycoform of CD43 lead to a more detailed examination of the relationship between expression of the 115 kDa isoform recognised by the S7 mAb and the isoform recognised by the 1B11 mAb in vivo and in vitro. In particular we wished to determine whether both epitopes were regulated in the same manner and whether both epitopes could be present simultaneously on the same molecule. There is no potential for alternate exon usage in the CD43 gene (176) therefore the protein backbone of the molecules recognised by S7 and IB 11 mAbs must be identical. This is supported by the results of our transfection experiments. The epitope recognised by IB 11 and S7 mAbs must therefore differ structurally as a result of carbohydrate modification of the CD43 protein backbone. Nevertheless, Cyster et al (303) presented evidence in the rat model that most antibodies to the CD43 molecule actually recognise the protein backbone, even 3/4 mAbs whose epitope was shown to be sialic acid dependent. If this holds true for 1B11 and S7 mAbs it is possible that both epitopes may be present simultaneously on the same molecule, depending upon the carbohydrate modifications which are present. Equally it would be possible for other mAbs to recognise molecules expressing either epitope. These possibilities are explored in this chapter. In section IV.2.4. we showed that the resting pattern of S7 antigen expression is distinct from that of IB 11. Here we show that S7 and 1B11 epitope expression are also regulated independently. 187 VIII.2. Results VII. 2.1. S7 antigen expression is not increased in response to ConA activation of thymus or spleen We performed identical studies of S7 antigen expression in Con A activated thymus and spleen as described in section V.4 for IB 11. In immunoblots of activated thymus and spleen no increase in the level of expression of the S7 antigen was observed (Figure 52), nor was there a shift in the apparent molecular weight of the antigen recognised by S7 mAb. This would suggest that because there is no change in the level of S7 expression on activation, the IB 11 isoform cannot be generated at the expense of the S7 isoform. These findings were confirmed by FACS analysis (Figure 53). Analysis of S7 immunoreactivity in sub-populations shows a small increase in S7 immunoreactivity in CD4+ lymphocytes and a small decrease in immunoreactivity in CD8 lymphocytes as reflected by changes in MFI (data not shown). This pattern is quite distinct from the changes observed with IB 11 immunoreactivity under the same conditions. VIII. 2.2. S7 immunoreactivity is expressed at high levels by the Ipr/fas mouse but the antigen is distinct from IB 11 We had previously demonstrated that the abnormal double negative T-lymphocyte population that is expanded in the MRL Ipr/fas mice expresses high levels of the IB 11 antigen. Here we confirm that similar high levels of S7 immunoreactivity are present in this population (Figure 54A-G). Examination of S7 immunoreactivity in CD4+ and CD8+ populations, however, reveals that S7 antigen expression is uniformly high in both CD4+ and CD8+ lymphocytes and does not increase with development of disease as the same level of S7 immunoreactivity is present in young mice prior to the development of any significant increase in the double. negative lymphocyte population (Figure 55). Immunoblots of lysates of lymph nodes from Ipr/fas and +/+ control mice reveal that S7 is recognising the 115 kDa isoform of CD43 in Ipr/fas and mice whereas IB 11 recognises the 130 kDa isoform (Figure 54). Immunoreactivity to both mAbs was 188 Figure 52. S7 antigen expression is not up-regulated in thymus or spleen in response to Con A activation. Single cell suspensions were prepared as described from BALB/c thymus (Panels A & B) and spleen (Panels C & D). Cells were activated in vitro with 5 Ug/ml Con A. At 24 hour time intervals aliquots of cells were removed and lysed using standard lysis buffer. Samples were then diluted into sample buffer and resolved in duplicate on 7.5 % SDS PAGE gels. Duplicate gels were probed separately for IB 11 (Panels A & C) and S7 (Panels B & D) immunoreactivity. Samples were standardised by Bradford assay prior to loading such that 40 ug of total cellular protein was loaded in each lane. Lane 1 Control: Lane 2 24 hours Con A: Lane 3 48 hours Con A: Lane 4 72 hours Con A: 189 Figure 52 B 200 2 3 4 2 3 4 — 200 — 190 Figure 53. S7 immunoreactivity is not increased within subpopulations of activated thymus or spleen. Single cell suspensions were prepared from BALB/c thymus (Panels A-D) and spleen (Panels E-G) as described. Cells were activated in culture in the presence of 5 |lg/ml of Con A. After 48 hours cells were washed and subjected to FACS analysis using FITC-labeled S7 mAb (or control), PE-labeled anti-CD4 mAb and biotinylated anti-CD8 mab, followed by streptavidin conjugated cychrome. S7 immunoreactivity was determined within gated subpopulations and expressed relative to binding of control mAb. Panel A CD4+8-: Panel B CD4+8+: Panel C CD4-8+: Panel D CD4-8-: Panel E CD4+: Panel F CD4-8-: Panel G CD8+: 191 1 9 2 Figure 54. S7 antigen is expressed at high levels in lymph nodes and thymus from MRLlpr/fas mice. , Single cell suspensions were prepared from the thymus and lymph nodes of 6 month old MRLlpr/fas mice as described. Cells were incubated with FITC-labeled S7 or control antibody, PE-labeled anti-CD4 and biotinylated anti-CD8, followed by streptavidin conjugated cychrome and subject to FACS analysis. S7 immunoreactivity was determined within gated subpopulations and expressed.relative to binding of control antibody. Panels (A-D) Thymus: Panels ( E - G ) Lymph nodes: Panel A CD4+8-. Panel B CD4+8+. Panel C CD4-8+. Panel D CD4-8-. Panel E CD4+. Panel F CD4-8-. Panel G CD8+. H Single cell suspensions from MRLlpr/fas (A & D) and MRL+/+ mice (B & C) were solubilised in lysis buffer and resolved on 7.5% SDS PAGE gels as previously described. Immunoblots were probed with either A & B S7 mAb or IB 11 mAb (C & D) 193 Figure 54 B S7 D F — 200 kDa — 115 kDa — 98 kDa 194 Figure 55. S7 antigen is normally expressed in young MRLlpr/fas mice. Single cell suspensions were prepared from the thymus and lymph nodes of 6 week old MRLlpr/fas mice as described. Cells were incubated with FITC-labeled S7 or control antibody, PE-labeled anti-CD4 and biotinylated anti-CD8, followed by streptavidin conjugated cychrome and subject to FACS analysis. S7 immunoreactivity was determined within gated subpopulations and expressed relative to binding of control antibody. Typical results are shown. Panels A-D Thymus. Panels E-G Lymph nodes. Panel A CD4+8-: Panel B CD4+8+: Panel C CD4-8+: Panel D CD4-8-: Panel E CD4+: Panel F CD4-8-: Panel G CD8+: 195 Figure 55 B 160 ' T 6 1 ' 16? " T 6 3 ibH S7 o fc 13 U M l ' A U°" ' T e 1 ' "T62 " i & 3 i ' 6 M o fc I—t 13 U D o fc 13 U 160' "i'61""i'G2' " 'iV "'iidH S7 111 1G1 0 ' 1 102 iV 10 S7 F S7 o fc 13 Ml S7 o fc 13 U Ml 16° 1&4 S7 o fc 13 U Ml S7 196 demonstrated in lysates prepared from Ipr/fas cells but not +/+ cells as the high S7 and IBM expressing double negative population predominates in Ipr/fas nodes but levels of expression of both isoforms of CD43 by normal lymph nodes are low. VIII.2.3. In cell lines, 1B11 epitope is distinct from that recognised by S7 In contrast to S7, IB 11 mAb did not detect a protein in immunoblots of whole cell lysates prepared from the EL4 cell line and in FACS analysis IB 11 expression by EL4 was low or equivalent to control values (Figure 56). This was not due to reduced affinity compared to S7 mAb as under identical conditions IB 11 mAb was more sensitive in detecting expression of its target antigen in the D26 cell line (Figure 56) and in the CTL cell line, no immunoreactivity with S7 could be demonstrated. (Figure 56). This suggested that IB 11 and S7 mAbs were recognising different isoforms of the CD43 molecule and that expression of these isoforms was differentially regulated. In order to confirm that the epitope recognised by the mAb IB 11 was distinct from the epitope recognised by mAb S7 and to determine whether they were present on different molecules, IB 11 was used to immunoprecipitate its target antigen from 3 different cell lines (D26, CTL and EL4) and the immunoprecipitates were resolved by SDS PAGE and then identified by immunoblotting using IB 11 and S7 (Figure 56 D &E). S7 did not recognise IB 11 antigen immunoprecipitated from D26, however S7 did recognise a similar sized molecule in whole cell lysates prepared from D26 but not from CTL. In addition S7 recognised a 115 kDa protein in whole cell lysates of EL4, which was not detected by IB 11. S7 mAb is unable to immunoprecipitate its target antigen and therefore the converse experiment could not be performed. 197 Figure 56. IB 11 mAb and S7 mAb recognise isoforms of GD43 which are expressed on different cell types by FACS and immunoblotting. CTL cells (A), D26 cells (B) and EL4 cells (C) were incubated with FITC-labeled, IB 11 mAb, FITC-labeled S7 mAb or FITC labeled control mAb. Cells were subject to FACS analysis. Whole cell lysates were prepared from CTL, D26 and EL4 as described. 1B11 antigen was immunoprecipitated from each cell lysate as described previously. In Panel D, whole cell lysate from CTL (lane 1), D26 (lane 3) and EL4 (lane 5) and 1B11 immunoprecipitates from CTL (lane 2), D26 (lane 4) were probed with 1B11 mAb. In panel E, whole cell lysates from EL4 (lane 1), D26 (lane 2) and CTL (lane 3) and 1B11 immunoprecipitates from D26 (lane 4) were probed with S7 mAb. 198 Figure 56 199 VIII.3 Discussion We have demonstrated that both IB 11 mAb and S7 mAb recognise CD43. However the two antibodies recognise distinct isoforms: S7 recognises a 115 kDa isoform which is expressed on all resting T-lymphocytes (although it appears to be expressed at very low levels in the CTL-c cell line). This isoform is not upregulated in response to Con A stimulation at the total protein level and only minor changes in levels of expression are observed in individual T-lymphocyte subsets on Con A activation. In contrast, 1B11 mAb recognises a 130 kDa isoform which is present on resting CD8 lymphocytes and at very low levels on resting CD4 lymphocytes and is induced in response to Con A activation. Immunoprecipitation experiments using cell lines have shown that IB 11 immunoprecipitates cannot be recognised by S7 mAb suggesting that under physiological conditions the IB 11 epitope and the S7 epitope do not co-exist on the same molecule. The existence of two distinct isoforms of CD43 has been previously demonstrated in the human (282,284,286) and in mouse (178,281). In mouse, the antibody 3E8 recognises two isoforms of CD43, a 115 kDa isoform present on resting T-lymphocytes and a 130 kDa isoform which is induced on Con A activation of lymphocytes (178,281). We have confirmed that these two isoforms correspond to the S7 and IB 11 immunoreactive isoforms respectively using immunoblots (our own observation). 3E8 mAb is considerably less sensitive than IB 11 mAb in detecting the higher molecular weight isoform and does not detect expression of this isoform in resting T-lymphocytes. In humans, the high molecular weight isoform is recognised by the mab T305 (282,284). Expression of T305 immunoreactivity has been correlated with expression of the core 2 enzyme in. both activated cells (286) and in Chinese hamster ovary cells transfected with human core 2 enzyme (305). Data from our laboratory confirms that transfection of the EL4- cell line with the murine core 2 enzyme results in abolition of S7 immunoreactivity and generation of IB 11 immunoreactivity (287). Thus IB 11 antigen expression appears to correlate with expression of the core 2 enzyme. The exact nature of the epitopes recognised by IB 11 mAb and S7 mAb are unknown. However it appears that the expression of the respective epitopes depends on post-translational 200 modifications of the CD43 protein by glycosylation and that core 2 enzyme is critical to this process. Nevertheless it appears from our studies using neuraminidase that S7 immunoreactivity can also be decreased and IB 11 immunoreactivity increased by removal of sialic acid residues. If 1B11 mAb recognises a carbohydrate epitope, specifically the hexasaccharide carbohydrate moiety of CD43, then neuraminidase treatment should not affect binding. There are a number of potential explanations for this observation. IB 11 mAb may recognise partially sialylated forms of the low molecular weight isoform. This explanation is unlikely as 1B11 immunoreactivity can be induced by transfection of the core 2 enzyme into low 1B11 antigen expressing cell lines (287). Also 1B11 antigen does contain sialic acid as demonstrated by the apparent increase in its molecular weight on deglycosylation. A second explanation is that lBllmAb recognises a carbohydrate epitope on CD43 which is more accessible following desialylation and that low IB 11 expressing cell lines express both isoforms but that sialic acid prevents recognition of the 130 kDa isoform by IB 11 mAb. This is unlikely as anti-peptide antibodies to CD43 do not recognise a 130 kDa isoform in EL4 cell line (287) The most likely explanation is that 1B11 and S7 mAbs both recognise protein epitopes in the backbone of the molecule. This explanation would be consistent with the findings of Gyster et al (303) that is that antibodies to rat CD43 recognise protein motifs, including those antibodies whose epitope is neuraminidase sensitive. Therefore the epitope recognised by IB 11 would appear to be generated through conformational changes of the polypeptide backbone induced by either hexasaccharide modification of the CD43 core protein or by desialylation of the low molecular weight isoform. It is likely that the epitope includes some part of the backbone protein as the hexasaccharide and tetrasaccharide structures expressed on CD43 glycoproteins are common modifications in O-glycans and yet the S7 and IB 11 mAbs appear to recognise CD43 only (Ziltener, unpublished observation, B. Aardman personal communication). Resolution of the exact nature of the epitope recognised by IB 11 mAb will require further analysis of the carbohydrate structure of the IB 11 antigen. The demonstration that S7 mAb immunoreactivity and 1B11 mAb immunoreactivity were mutually exclusive and that the cells lines EL4 and CTL express very different amounts of the two isoforms has enabled us to use these cell lines to produce CD43 fusion proteins. The glycosylation 201 patterns of these fusion proteins resembles either the 115 kDa isoform (EL4) or the 130 kDa isoform (CTL). We plan to use these fusion proteins to attempt to determine the ligands for the respective isoforms. 202 IX. Discussion CD43 is a member of a family of leukocyte sialoglycoproteins which includes CD34 and CD45. Molecular weight heterogeneity of members of this family has been confirmed by initial studies using radio labeling and lectin binding and more recently by employing specific monoclonal antibodies. Antibody studies by Baecher et al in 1988 (178,281) identified 2 major isoforms of murine CD43 with molecular weights of 115 kDa (thymocyte,Tymphocyte/monocyte form) and 135 kDa (neutrophil/platelet). We believe that the 135 kDa form is identical to IB 11 antigen and that the slight difference in molecular weight is likely due to different gel conditions. The antibodies used in these studies recognised either exclusively the 115 kDa isoform (S7) or both isoforms (3E8, SI 1, SI5) but with low affinity for the high molecular weight isoform. The S7 (115 kDa) isoform was ubiquitously expressed by T-lymphocytes. The 135 kDa isoform was less well characterised due to the lack of specific mAbs to this isoform. However, Baechers studies did demonstrate that expression of the 135 kDa isoform could be induced by Con A activation of T-lymphocytes (281). We have been able to demonstrate that the IB 11 antigen corresponds to the 130 kDa molecular weight isoform of murine CD43. The specificity of the IB 11 mAb differs from the previously reported anti-CD43 antibodies in that under normal physiological situations IB 11 mAb recognises exclusively the 130 kDa isoform and there is no cross-reactivity with the low molecular weight (115 kDa) isoform. The pattern of expression of 1B.11 immunoreactivity corresponds to that observed by Baecher for the high molecular weight isoforms of CD43, i.e. the isoform is expressed preferentially on activated T-lymphocytes. Iii addition our findings are consistent with those of Fox al who found that the human homologue T305 was preferentially expressed on activated T-lymphocytes (282), cytotoxic T-lymphocytes and on immature thymocytes (284). This, contrasts with the more uniform pattern of expression of the low molecular weight (115 kDa) CD43 isoform. Lectin binding studies have also identified molecular weight species which show a similar size and pattern of distribution as the 1B.11 antigen. Peanut agglutinin (PNA) which binds to Gaipi-3GalNAc residues in O-glycans, recognises a species of appropriate weight in thymocyte 203 lysates (306) as does Helix pomeratia, a lectin which binds to N-acetyl D-glucosamine on glycoproteins on murine lymphocytes (307). Vicia villosa lectin which binds to N-acetylgalactOse identifies a glycoprotein of molecular weight 130 kDa in murine CD8+ lymphocytes, but the same glycoprotein has a molecular weight 145 kDa in cytotoxic T-cell lines (308). We have observed a similar pattern with respect to 1B11 mAb immunoreactivity in immunoblots of lysates prepared from primary CD8+ lymphocytes compared to cytotoxic T-cell lines. Murine CD43 was originally cloned and sequenced by Baecher etal (177) and Cyster et al (176) in 1990. The human CD43 gene was cloned and sequenced 2 years earlier by Shelley'ef al (309) and Pallant et al (3 \0)\ The protein sequence of murine CD43 shows 72% homology with human CD43 protein and 92% homology with rat CD43. . The murine gene is located on chromosome 7 and comprises a single exon with an open reading frame which corresponds to a predicted protein sequence of 395 amino acids, with a 19 amino acid leader sequence. The transcribed product is a single mRNA species of 4.2 KB. The translated protein has a predicted size of 38 kDa. This is significantly less than reported for the mature CD43 glycoprotein. However, the extracellular domain of murine CD43 is composed of 40% serine (Ser) and threonine (Thr) (compared to 12% in normal proteins) and it is likely that the majority of these 93 Ser and Thr residues carry O-linked carbohydrate structures. Therefore, over 60% of the molecular weight of the CD43 molecule is contributed by carbohydrate. Since there is no evidence for differential exon usage in murine CD43, the molecular weight heterogeneity of the different.CD43 isoforms reported must be generated by post-translational modification. The most likely structure to be modified is the carbohydrate moiety. The apparent larger size of the IB 11 reactive isoform might be generated by addition of more complex polysaccharide structures or by removal of sialic acid residues. The latter would lead to an alteration in the charge of the CD43 molecule and a resultant decrease in its apparent mobility on SDS PAGE analysis. Although we cannot completely exclude the possibility that 1B11 mAb recognises a partially sialylated isoform of CD43, we consider that it is unlikely as . a) neuraminidase treatment causes a characteristic shift iri the SDS PAGE mobility Of the protein suggesting that sialic acid residues are present on the molecule; ; 204 b) there is no: evidence (in the human system) for induction of expression of desialylated isoforms of CD43 on activation. In addition auto-antibodies against asialylated forms of CD43 which are present in the sera of HIV infected individuals, do not recognise structures on activated lymphocytes (311). (There is evidence that asialylated forms of CD43 are expressed on the thymus (311) and on HIV-infected lymphocytes (312)). Of interest to us are the recent findings of Brown et al (313) who have demonstrated that Jurkat T-cells express epitopes for a mAb, J393, which recognises a partially sialylated form of CD43 with a similar molecular weight to IB 11 antigen. However Jurkat cells show defective glycosylation of proteins and the isoform recognised contains monosaccharide glycans. This CD43 isoform was not found on peripheral blood lymphocytes. On activated T-lymphocytes the molecular weight of the isoform recognised by J393 was 160 kDa. c) 1B11 mAb immunoreactivity can be conferred on IB 11 mAb negative cells by transfection of those cells with the murine homologue of the C2GnT enzyme (287). This enzyme is known: to generate higher molecular weight isoforms of human CD43 which are recognised by the T305 antibody. : Rather, we believe that the IB 11 mAb recognises a high molecular weight isoform of murine CD43 which is generated by developmentally regulated expression of the C2GnT enzyme, and which may have a function in immune development that is distinct from that of the low molecular weight isoform of CD43. The biosynthesis of the 2 oligosaccharide isoforms of CD43 requires the concerted sequential action of 6 glycosyltransferases (298). The relative activity of these enzymes is determined by the cell type and its activation state (296). Human myeloid cells, platelets, thymocytes and activated human T-lymphocytes express mainly the hexasaccharide isoform, whereas resting T-lymphocytes express mainly the tetrasaccharide isoform (286,297). The critical determinant of hexasaccharide biosynthesis is the activity of the C2GnT enzyme (286,296,297). We have shown that activity of this, enzyme m v/vo and in vitro correlates with IB 11. immunoreactivity (207). 205 Ser/Thr GalNAcoc-Ser/Thr aGalNAc-transferase fih3Gal-transferase Galpl-3GalNAca-Ser/Thr o2-6SANAc-transferase SAAca2 \ -6 Galb 1 -3GalN Aca-Ser/Thr a2-3SAN Ac-transf erase SAAca2 V 6 SANAca2-3Galpl-3GalNAca-Ser/Thr f31 -6GlcNAc-transfera.se .GlcNAcpM 6 Galp 1 -3GalNAca-Ser/Thr f31-4Gal-transferase Galpl-4GlcNAcpi I 6 Galp 1 -3GalNAca-Ser/Thr C/2-3SANAc-transferase S ANAca2-3Galp 1 -4GlcNAcp 1 \ SANAca2-3Galpl-3GalNAca-Ser/Thr In addition we have shown that activation of T-lymphocytes leads to an up-regulation of IB 11 antigen expression both in vivo and in vitro. This upregulation appears to require specific T-cell receptor engagement. At the, same time there are no. significant changes in the level of expression of the low molecular weight isoform of CD43 as recognised by the S7 mAb. Determination of the actual carbohydrate or protein structure recognised by IB 11 mAb within the high molecular weight isoform of CD43 requires additional study which is on-going in 206 our laboratory and will require carbohydrate.sequencing. A recent workshop adhesion sub-panel report on CD43 (314) suggests that there are at least 4 different epitopes on the human leukosialin molecule. These epitopes differ with respect to their sensitivity to neuraminidase and their ability to induce homotypic aggregation. Interestingly one of the antibodies described, L10 mAb, shares the same increased affinity for its epitope in response to neuraminidase treatment as 1B11 mAb. Unfortunately the panel does not report on the molecular weight of the antigen and has not attempted to group the T305 mAb for comparison. The ability to accurately sequence the oligosaccharide units of glycoconjugates has led to an appreciation of the complexity of carbohydrate structure and function. Roles for carbohydrate have been demonstrated in stabilising protein conformations, extending half-lifes, providing ligands for specific binding events and in modifying protein-protein interactions (reviewed in 315). Since all cells are covered in a dense coat of sugars, it is predicted that oligosaccharides must be critical determinants of cell-cell interactions. Interestingly CD43 low molecular weight isoform is one of the most abundant cell surface glycoproteins on T-lymphocytes. As a result of its high sialic acid content it is also the single biggest contributor to net surface charge. Not surprisingly therefore, knock-out experiments in which CD43 gene expression is disrupted have lead to increased T-cell adhesiveness and enhanced proliferation in mixed lymphocyte reactions (316,317). In addition there is evidence that CD43 negatively regulates cell-cell interaction including that required for cytotoxicity (318). In CD43 knockout mice, T-lymphocytes. show enhanced proliferation in mixed lymphocyte reactions and an enhanced response to Con A, 2CHmAb and the super antigen staphylococcal enterotoxin B. (319). Interestingly, despite their hyperproliferative phenotype, CD43 knockout mice were less efficient in containing and clearing a vaccinnia virus infection, suggesting either that there are defects in tissue localisation of effector lymphocytes or defects in elimination involving other branches of the immune system. If GD43.does indeed function in effective tissue localisation of activated lymphocytes then our data suggesting that IB 11 immunoreactive T-lymphocytes are relatively absent from secondary lymphoid tissue might be consistent with a role in. redistribution of activated lymphocytes to rion-lymphoid tissue. This is an area which warrants further investigation. 207 The extent to which the inhibitory effects of CD43 expression are mediated by the strong negative charge on the low molecular weight form of CD43 are unclear from knockout experiments. Evidence suggests that at least part of the inhibitory effects of CD43 on adhesion can be mitigated by neuraminidase treatment of cells resulting in a decrease in surface sialic acid residues (316). In addition, knock-out experiments disrupt the expression of both the high and the low molecular weight isoforms of CD43 and are unhelpful in resolving Whether there are specific receptors for the high and low molecular weight isoforms. Initial studies by Rosenstein (320) suggesting that CD43 might bind ICAM-1 have not been supported by other groups. Recent studies by Stockl (321) have suggested that CD43 binds directly to MHC class I molecules following transactivation of CD43 by stimulatory mAbs to CD2. In addition, Sanchez-Mateos et al (258) have demonstrated that mAb ligation of CD43 leads to up-regulation of expression of (31 and p*2 integrins and enhanced adhesion of lymphocytes to matrix ligands. In addition the T305 immunoreactive form of CD43 has been shown to bind to galectin 1 (322). Thus CD43 expression may not only negatively regulate adhesion through alteration of cell surface charge but may also lead to increased adhesion both directly and by up regulating expression of other adhesion molecules when specifically engaged by its ligand. The functions of CD43 appear complex and are strongly influenced by the antibody used in the assay, the cell type expressing CD43 and its activation state. Expression of different CD43 . isoforms under specific physiological conditions may underlie some of these complexities since not only do the two isoforms differ with respect to carbohydrate structure but they are also differentially expressed between different cell types and on T-lymphocytes sub-sets and the data we have presented demonstrates that they are differentially regulated in vitro and in vivo. Additional post-translational modifications of the hexasaccharide isoform have been demonstrated in humans. The high molecular weight isoform of CD43 contains polylactosamines which may be fucbsylated, generating the Lewis X and sialyl-Lewis X epitopes (288), and sulfated (160,289). These unique and more unusual types of glycosylation structures are more likely to mediate a specific biological function. For example, the presence of the sialyl-Lewis X epitope 208 has been shown to be required for selectin binding. Sulphation has been shown to enhance binding of matrix proteoglycans to anti-thrombin III, N-CAM and perhaps selectins (300-302,323). The presence of the sialyl-Lewis X epitope has been demonstrated on.polylactosamine moieties derived from human CD43 but is likely only present on a very small percentage of all CD43 molecules (324). We have been unable to demonstrate the presence Of the sialyl-Lewis X epitope on IB 11 immunoreactive'cells or in cross-immunoprecipitation experiments, however the role of the Lewis-X antigenic determinants in adult mice has not been well characterised. There is some evidence from rat studies that rat selectins will bind Lewis-X epitopes in lipid membranes (301), however to date the only description of Lewis-X expression in vivo in rodents is in fetal development (325). Sulphation of CD43 (and CD45) has been reported in a murine T-lymphoma cell line (160) and in PHA-derived T-lymphoblasts (289). We have also been unable to demonstrate sulphation of the high molecular weight isoform recognised by IB 11 although this may be a function of the cell line used for these experiments. Wilkins' studies of HL-60 a human cell line which expresses the hexasaccharide isoform of CD43 found no evidence that CD43 was sulphated in this cell line (324). The relatively neutral pi documented here for the IB 11 molecule would also mitigate against the incorporation of highly acidic sulphate groups into its O-glycan side chains. Interestingly, the presence of sulphate groups in the high molecular weight isoforrri of CD43 in human lymphocytes appears to decrease adhesion, likely by increasing surface negative charge in a manner analogous to that reported for sialic acid residues (289). Although the IB 11 isoform does not appear to contain sulphated oligosaccharides or the sialyl Lewis-X epitope it does contain carbohydrate structures which have been shown to function as ligands for cell surface molecules: Macrophage sialic acid-specific receptor (sialoadhesin) recognises sialic acids on surface glycoconjugates in the sequence SAa2-3Galbl-3GalNAc (326-329). This epitope is present on both isoforms of CD43 but may be relatively more abundant or more accessible on one or other isoform. The pattern of expression of this receptor is intriguing. In bone marrow it is expressed at points of contact between macrophages and developing myelomonocytic cells (327). Similar' 209 expression at points of contact between erythroblasts and macrophages are not seen: If sialoadhesin is interacting with CD43 on these cells, it might mediate the apoptosis observed when the CD43 antigen is cross-linked by antibody on hemopoietic stem cells in vitro (330) and also observed in our laboratory when C2GnT enzyme is overexpressed in hemopoietic progenitor cells (our unpublished observations). The pattern of binding sialoadhesin to lymphocytes is also similar to their relative levels of IB 11 antigen expression with the preferential binding to activated T-lymphocytes, thymocytes and lymphoma cell lines (329). . Other candidate ligands for IB 11 antigen include members of the C-type lectin family. The exact carbohydrate structure bound by these molecules are unknown but may include fucose as one determinant. A feature of these receptors is the promiscuity of their ligands. CD23 has also been reported to bind to carbohydrate sequences which may be present within the high molecular weight isoform of CD43. CD23 was originally identified as the low-affinity IgE receptor, however other structures have been show to bind to CD23 including LFA-1, CD21 and 180 kDa protein identified on RPMI 8226 and human K562 cells (331,332). The galectin family of molecules are S-type cation-independent lectins which recognise galactose as a minimal binding determinant (333). Baum et al (322) have demonstrated that adhesion of the MOLT-4, a human lymphoid cell line, to thymic epithelium could be inhibited by antibodies to Galectin 1 and by antibodies to CD45 and to the high molecular weight isoform of CD43 (T305). No additional inhibition could then be.observed when these antibodies were combined. This data suggested that galectin 1 binds to the T305 immunoreactive isoform of CD43. Further, galectin 1 binding correlated with the pattern of expression of T305 on mature and immature thymocytes. What is not apparent from these experiments is why the levels of inhibition should be similar irrespective of whether anti-CD43, anti-CD45 or ariti-galectin antibodies were used. The hypothesis would suggest that the level of inhibition by anti-galectin 1 should be the sum of the inhibition of anti-CD43 and anti-CD45 and yet the level of inhibition is only equivalent to that observed singly with these antibodies. This would suggest that some of.the inhibition • observed is non-specific. Interestingly galectin binding to CD45 was shown to induce apoptosis in thymocytes but comparable data for CD43 have not been published (334). 210 In similar studies we have demonstrated that binding of Galectin 3 'correlate's with the level of IB 11 immunoreactivity of cells. Galectin 3 (or Mac-2) is a 30,000 molecular weight p-galactoside binding protein which is postulated to be involved in homotypic aggregation of tumour cells in the circulation during metastases through attachment to complementary serum glycoproteins (335). This data:does not preclude the possibility that galectin 3 also interacts with cell surface glycoproteins which express the appropriate carbohydrate recognition sequences. Galectin 3 molecules tend to form multivalent aggregates on the cell surface suggesting that their ligand may also be aggregated. The high molecular weight isoform of CD43 contains the requisite sequences for recognition of galectin 3. In addition we have shown that it clusters in activated cells. A number of recent papers (336,337) have also shown that an isoform of CD43, as yet uncharacterised, can be expressed on non-lymphohemopoietic malignant human cell lines and is frequently secreted. It is tempting to speculate that this aberrant expression of CD43 may reflect an adaptive mechanism to ensure survival of metastasising tumour cells through homotypic aggregation of cells in the circulation, mediated via galectin 3 . Whether the galectins constitute the normal physiological ligands of the high molecular weight isoform of CD43 is unclear. The high molecular weight isoform may have a number of different ligands whose expression is similarly developmentally regulated. The observed localisation of the IB 11 immunoreactive isoform to uropods during activation may enable the high . molecular weight isoform to interact with ligands which are distinct from the low molecular weight isoform and by clustering of the high molecular weight isoform to mitigate the anti-adhesive effect of the negative charge associated with the low molecular weight isoform. Resolution of the controversy surrounding the ligand for CD43, high and low molecular weightforms, requires the development of assay systems which employ the correctly glycosylated isoform in ligand binding studies. Our data presented here shows that CD43 transfected into cell line which dp not express the protein under normal physiological conditions is aberrantly glycosylated and is unlikely to be useful for determining the normal ligand of the molecule. Studies on-going in our laboratory are currently using the CTL and EL-4 cell lines identified here 211 as expressing the 1B11 and the S7 reactive isoforms respectively. We are currently performing experiments in which these two cell types are transfected with a cDNA coding for the extracellular domain of CD43 combined with the Fc receptor. It is hoped that the protein product will be glycosylated by the cellular enzymes in such a way that either the 115 kDa or the 135 kDa isoform will be generated with the correct glycosylation pattern. Activation of T-lymphocytes leads to both phosphorylation of the IBM immunoreactive isoform and to increased association with the cytoskeleton. In addition IB 11 immunoreactivity in activated cells could be seen to cluster on cellular uropods in a manner analogous to that observed by Sanchez-Mateos (258) for CD43 activated lymphoblasts, although they were unable to demonstrate CD43 clustering in response to T-cell receptor engagement alone. It is possible that the aggregation of antigen on lymphocytes observed by Sanchez-Mateos in response to activation arose as a result of co-capping of antigen by antibody, as their cells were not fixed prior to staining. In contrast, we fixed our activated cells before incubation with antibody and thus eliminated any effects that might arise from capping of CD43 by antibody. -We hypothesise that activation of T-lymphocytes leads to phosphorylation of the GD43 molecule on serine residues and directly or indirectly leads to increased association with cytoskeletal proteins. This enables the high molecular weight isoform of CD43 to cluster in specific areas of the membrane e.g. uropods. The high density of CD43 high molecular weight isoform, which carry a lesser negative charge than the corresponding low molecular weight isoform facilitates interaction with cells carrying the corresponding ligand. The ligand then interacts with specific carbohydrate epitopes on the oligosaccharide moiety. Clustering of the CD43 isoform may also facilitate interaction with multivalent ligands such as galectin 3. The nature of the ligand remains obscure and is the subject of on-going study in our laboratory. It is possible that the ligand varies with the physiological situation and the nature of the interacting cell. Association of CD43 with the cytoskeleton may. also enable it to form.complexes-with the T cell antigen receptor (TCR) (338,339). ; 212-The nature of the signal delivered by the GD43 molecule remains obscure. Ardman's group (340) have identified a 95 kDa protein which is phosphorylated in response to antibody binding of CD43. It is not clear as to whether this antibody is interacting with the high or the low molecular weight isoform, nor is it clear what the consequence of this phosphorylation is to the cell. Ligation of CD43 by antibody has been reported to result in calcium mobilisation, and protein kinase C activation (338,341-343). The subsequent cellular response has depended on the cell line and the antibody used. Most anti-CD43 antibodies produce homotypic aggregation of lymphocytes (344-346). This may be partly due to the relative abundance pf the antigen which can be cross-linked and partly due to up-regulation of additional adhesion receptors such as LFA-.l (343-345). Some antibodies to CD43 induce proliferation (MEM-59 mAb and L10 mAb) (338, 341-343), or synergise with suboptimalconcentrations of PMA, IL-2 or Gon A (Bl-B6mAb, S7 mAb and El IB mAb)(344,347) in inducing proliferation in T-lymphocytes. The proliferative response to anti-CD43 antibodies requires the presence of monocytes and the antibody must be present in the bivalent form. However the Fc binding portion does not appear to be essential for activity (338; 341-345). The proliferative response requires the presence of the CD3 molecule on the responding T-lymphocyte (338), but can occur in T lymphocytes deficient in expression of CD28 (347). Cross-linking of CD43 on human lymphocytes with MEM-59 mAb results in significant down-regulation of CD43 expression (348). The mechanism of this effect appears to be via proteolytic cleavage of the CD43 molecule. It is not clear, whether MEM-59 reacts with the, high or the low molecular weight isoform of CD43, however, the large apparent size of the soluble fragment suggest that isoform released, is relatively hyposialylated. The functions of soluble CD43 remain obscure. Interestingly the mAb, MEM-59 also inhibits proliferation and induces apoptosis in hemopoietic progenitor cells (330,349). A similar induction of apoptosis. has been described for the J393 mAb which recognises a hyposialylated form of CD43 in Jurkat cells (313). Unfortunately most of these publications do not include data on the molecular weight of the isoform of CD43 recognised, so it is not possible to determine whether the high or the low molecular weight isoforms are involved. Clearly, CD43-mediated responses can differ 213 significantly depending on the cell type and the isoform expressed. We feel that this area warrants further study and that utilising the IB 11 mAb and the S7 mAb may allow us to clarify some of these issues. The increased expression during allograft rejection and during the graft versus host response also suggests a role in cytotoxicity. We are currently exploring the potential functions of the high molecular weight isoform by over-expressing the enzyme responsible for its generation in cell lines. Preliminary data from our laboratory suggest that expression of the 1B11 antigen but not S7 antigen correlates with resistance to cytotoxic killing by CD8+ lymphocytes and by NK cells.' In conclusion, the study of CD43 its ligands and its function is complicated by the existence of multiple isoforms of the molecule. Differences in glycosylation patterns are critical to the generation of these isoforms and specific patterns of glycosylation may determine not only . ligand specificity but also function. We believe that we have characterised a high molecular weight isoform of CD43 which is dependent for its expression on the presence of the C2GnT enzyme. We have shown that the expression of this isoform is specifically and developmentally regulated in T-lymphocytes. We believe that we have developed both the tools and the model systems to now search for the specific ligand of this isoform and to examine its functions in the lymphohemopoietic system. 214 X. Materials and Methods X . l . Production of antibodies. X . l . l . Rabbit anti-1-29 peptide antibodies Synthetic peptides corresponding to the first 29 amino acids of murine interleukin 3, (1-29 mIL-3) and to the full length molecule, (1-140 mlL-3) were synthesised by solid phase methods and coupled to keyhole limpet hemocyanine (KLH), as described elsewhere (350).. Four Dutch Belted rabbits.were each injected in six subcutaneous sites with 300 ug of peptide (total) emulsified in complete Freund's adjuvant. After 4 weeks the rabbits were bled and boosted at.2 weekly intervals with decreasing amounts of antigen (200-50 ug) emulsified in incomplete Freunds adjuvant. Two weeks after boosting, rabbit sera were screened for production of high titre anti-rL-3 Or anti-peptide antibodies by ELISA as described (351). X.1.2. 2E11 and 1A3 monoclonal antibodies BALB/c mice were injected s.c. with a mixture of synthetic pepides corresponding to murine IL-3 residues 1-29C, CGG30-43, 44-75C, 76-89, 64-82, C91-112 and 123-140 together with 13 overlapping hexapeptides covering residues 123-140 and full length 1-140 IL-3 in complete Freunds adjuvant. At 2 week intervals mice were boosted with. 10|ig synthetic IL-3 (1-140) emulsified in incomplete Freunds adjuvant and after 3 boosts spleens were harvested for fusion with the myeloma cell line, X63/A98:653 as described (148). 1 A3 and 2E11 mAbs are known to recognise the 130-135 peptide sequence of murine EL-3 in solid phase ELISA. The affinity of 2E11 mAb for this sequence is greater than that of 1 A3 as demonstrated by a reduced ability to neutralise IL-3 bioactivity and the higher leakage of IL-3 from affinity columns made from 1 A3 antibodies (148). 215 X.2. Affinity purification of antibodies X.2.1. Anti-murine IL-3 antibodies Rabbit anti-1-29 antibodies were affinity purified from serum by passage over activated thiol-Sepharose (Pharmacia, Uppsala, Sweden) to which 1-29 (IL-3) peptide had been coupled. Antigen-specific antibodies were eluted using 0.1M glycine pH 2.5 and dialysed against 0.9% (w/v) saline solution. Purified antibodies were concentrated to 0.5mg/ml and assayed by ELISA for binding to denatured and recombinant antigen. 2E11 and 1 A3 hybridomas were injected i.p. into BALB/c mice in order to produce ascites fluid. 2E11 mAbs and 1 A3 mAbs were affinity purified from ascites fluid by passage over activated thiol-Sepharbse (Pharmacia) to which chemically synthesised EL-3 had been coupled. Antibodies were eluted, dialysed and concentrated as described above. Binding of antibodies to denatured antigen was detected using'a conventional ELISA employing 10 ug/ml .of full length chemically synthesised IL-3 adsorbed to ELISA plates (Bectoh-Dickinson, Oxnard, CA). -X.3. Detection of antibody binding to IL-3 Binding df antibodies to recombinant mUrine IL-3 was detected using a sandwich ELISA method adapted to assay antibody binding (351). Briefly, ELISA plates were coated with lOug/ml polyclonal sheep antibody raised against murine IL-3. Recombinant murine IL-3 at a concentration of 50U/ml was added to the plates, which were incubated for 2 hours at room temperature. Serial dilutions of anti-IL-3 antibodies were transferred to the plate. Antibody binding to captured IL-3 was detected using a peroxidase coupled anti-rabbit antibody as described (351). . . ' X.4. Production of recombinant murine IL-3. The murine myeloma X63Ag8-653 (X63) transfected with the expression vector pBV-1MTHA modified to express BMGNeo-mIL-3 was obtained from F. Melchers (Basel Institute of 216 Immunology, Switzerland). The resultant stable transformant, X63-omIL-3 constitutiyely secretes murine IL-3 in high quantities (352). Conditioned media from this cell line was concentrated 10-fold using an ultra filtration concentrator (Amicon Co., Danvers, MA) with a membrane molecular weight cut off of 5000 Dal tons. The concentrated media was then applied to a Sepharose column (Pharmacia, Uppsala, Sweden) to which purified rabbit polyclonal antibodies raised against the 1-29 (IL-3) peptide had been coupled.. The columns were washed with 10 times their bed volume of phosphate buffered saline (PBS). IL-3 was then eluted with 0.1M glycine pH 2.5, neutralised with saturated Trizma base solution and dialysed three times against 0.9% (w/v) saline solution. Affinity purified recombinant IL-3 preparations were filter sterilised arid assayed as described below. IL-3 preparations were diluted to 10,000 units per ml with 0.9% (w/v) saline, stored at -20 °C and thawed prior to use. Recombinant IL-3 was used for all in vivo studies and was filter sterilised immediately prior to use. X.5. Bioassay of mIL-3. X.5.1. Bioassay of riiIL-3 preparations: Recombinant IL-3 preparations were assayed for biological activity using the IL-3 responsive cell line, R6-XE4 as described in (353). Experiments were performed in triplicate. Results were compared to a standard curve produced by serial dilution of a known amount of recombinant IL-3. One biological unit of IL-3 was defined as that amount producing 50% of maximal 3H-thymidine incorporation. • X.5.1. Bioassay of recombinant IL-3 in the presence of anti-IL-3 antibodies. Rabbit anti-1-29 (IL-3) antibodies were serially diluted 1:4 fold from a stock of 10 Ug/ml and titrated in triplicate against a constant amount of recombinant mIL-3 (lU/ml) over 10 wells of a Terasaki tray and assayed for inhibition of IL-3 bioactivity as described above. Experiments .were performed in triplicate and compared with the proliferative effect of an equivalent concentration of recombinant JL-3 alone. . ' '• • -2i7 ' • > .' -Similar experiments were performed for 1 A3 and 2E11 mAbs. X.6. Gel-filtration profile of antibodies. A 2.5 x 100 cm column of Sephadex G100 (Pharmacia, Uppsala, Sweden), was equilibrated with PBS. Fifty micrograms of Rabbit anti-1-29 (IL-3) antibodies, 2E11 mAb or 1A3 mAb, were pre-incubated for 30 minutes With 2000 biological units of recombinant TL-3 in 200 |il of phosphate buffered saline and then applied to the column and eluted in 1 ml fractions. Fractions were assayed for IL-3 bioactivity using the bioassay described and for antibody activity using the ELISA described. This method was used as a screening test for high affinity antibodies. In order to determine the exact molecular weight of high affinity antibody-cytokine complexes and the number and types of complexes formed, similar analyses, were performed using an FPLC apparatus equipped with a Superose 6 column (Pharmacia, Uppsala, Sweden). X.7. In vivo pharmacology: BALB/c mice were housed in the animal facility of the University of British Columbia under specific pathogen-free conditions with access to sterilised food and water ad libitum. Mice were used at 8-12 weeks of age. X.7.1. Rabbit anti-1-29 antibodies. Groups of mice were injected intravenously into the tail vein with i) 1000 biological units of recombinant IL-3; or ii) 10 (Xg Rabbit anti-1-29 (IL-3) antibodies ; or iii)TL-3 and Rabbit anti-1-. 29 (IL-3) antibodies; or iv) IL-3 and 10 jLtg Rabbit control IgG. All injections were made in a total volume of 120 ul 0.9% w/v saline. At intervals, groups of 4 mice were sacrificed and bled by cardiac puncture to obtain serum for estimation of IL-3 bioactivity and antibody titre, as described above. Experiments were repeated on 3 separate occasions. Clearance data was also obtained following injection of 10*000 biological units of IL-3 alone, to determine the effect of higher doses on clearance. Clearance data were described statistically using the MULTIFORM program supplied by D.W.A. Bourne (OU College of Pharmacy, . OK)(354). 218 X.7.2. 2E11 monoclonal antibodies. Groups of mice were injected intravenously into the tail vein as described above, with i) 1000 biological units of recombinant IL-3; or ii) 1000 U IL-3 and 100 (ig 2E11 mAb. Experiments were repeated on two occasions.. X.8. In vivo biology. DBA2 mice were, allocated to groups of 5 and received intravenously : i) 1000 biological units of mTL-3; ii) lOug of Rabbit anti-1-29 (IL-3) antibodies; ii) both 1000 biological units of IL-3 and 10 jn.g of Rabbit anti-1-29 (IL-3) antibodies ; or iv) 0.9% w/v saline. Injections were repeated at 24 hour intervals for 5 days. On the sixth day the animals were sacrificed and serum, bone marrow lung, jejunum and spleen removed for further analysis. X.8.1. Quantification of spleen colony formation in vitro. Spleens were bisected and single cell suspensions prepared from one half of each spleen (the remainder being fixed for histology). Red cells were removed by hypotonic lysis using Tris.HCl-NH4Cl. Cells were counted and resuspended at a concentration of either 104 or 105 cells per ml in a 0.3% solution of blue bactor agar (Difco Laboratories, Detroit, MI) in RPMI medium containing 10% FCS, (3-mercaptoethanol. 10-fold concentrated WEHI-3B conditioned media was added at 2% (v/v) as a source of IL-3. Agar cultures were incubated for 7 days at 37°C, 5% CO2 and scored blind for colony formation. In separate experiments mice were injected with i) 1000 units of IL-3; ii) 1000 units of IL-3 and 10 |ig Rabbit anti-1-29 (IL-3) antibody; iii) 0.9% w/v saline; or iv) 1000 units pf IL-3 and 10 fig Rabbit control IgG to act as an additional control. X.8.2. Determination of mast cell precursor frequency in vivo. . The frequency of mast cell precursor in bone marrow and spleen was determined as described previously (355). Briefly, single cell suspensions of spleen cells prepared as above and of bone marrow cells eluted from mouse femurs, were plated at 5 different concentrations into wells of 96 well tissue culture treated trays (Costar, Cambridge, MA.) in 150 |il of RPMI medium 219 containing 10% FCS, (3-mercaptoethanOl and 2% (v/v) 10 x WEHI-3B conditioned media as a source of IL-3. After 15 days, trays were scored blind by two scorers for mast cell growth. The frequency of mast cell precursors was calculated from the limit dilution data using a Poisson probability distribution, lines being fitted to the data points by maximal likelihood estimation techniques. Calculations were performed using the dlimit program (356) to determine the frequency of mast cell precursors and the 95% confidence limits. X.8.3. Histology. One centimeter of jejunum, one lung, one popliteal lymph node and one half of the spleen from each animal were fixed in methanol, formalin and acetic acid (85%, 10%, 5% v/v) for 24 hours at 4°C. Sections were stained with either alcian-blue saffranin for mast cell determination (357) or haematoxylin and eosin (H&E) for assessment of histopathological changes. Mast cell numbers and megakaryocytes were enumerated in 40 randomly chosen high power (x40) fields from each group. Average results for each mouse were then analysed by ANOVA to determine differences between the groups. Statistics. Statistical analysis of results was performed by ANOVA using the STATVIEW program (Microsoft word, Seattle, WA). This program calculates significance at the 95% level by both Fisher PLSD and Scheffe F-test. 220 XI. Characterisation of antibodies XI. 1. Introduction and aims We hypothesised that antibodies with enhancing activity in vivo should share certain characteristic properties, that is, that they should recognise their target molecule with high affinity in a non-denatured form and they should not neutralise the bioactivity that of the target cytokine in bioassays. When animals are immunised with a protein, a proportion of that protein will become denatured, that is its secondary structure and 3 dimensional conformation will, differ from its native conformation. Antibodies raised against this molecule may then show a high affinity for the denatured form of the molecule but little or no reactivity with the target in its native soluble form. Conventional ELIS A methods which determine the reactivity of antibody with target bound to plastic in a solid phase, may overestimate the reactivity of antibody with the native target as a proportion of the target becomes denatured in the process of binding to plastic. This is particularly true for anti-peptide antibodies which are raised against isolated peptide sequences and which recognise linear epitopes that may not assume the native conformation (19). We have.devised a sandwich assay for IL-3 in which native IL-3 is captured by a primary anti-IL-3 antibody, bound to .plastic. The IL-3 is then presented as a target to the antibody of interest, in its native conformation. We believe that this method is better able to predict for antibodies which will bind native IL-3 with high affinity (351). Murine IL-3 is a 22-32 kDa glycoprotein (53,358), the exact molecular weight being determined by the extent of N-linked glycosylation. The human protein has a molecular weight of 15-30 kDa and shows 29% homology at the amino acid level.with murine IL-3 (359). The three-dimensional structure of both molecules is thought to resemble that of GM-CSF and comprises a 4 a helical bundle, stabilised by disulphide bridges (360)! Studies using synthetic peptides and antipeptide antibodies have localised the receptor binding portion of murine IL--3 to two sequences within the molecule (361,362). Antibodies 221-which bind within these sequences would be expected to inhibit receptor binding and hence neutralise the effects of IL-3 in vivo. These studies have also confirmed that the first 1.6 amino acids are not required for receptor binding. The Rabbit anti-1-29 (IL-3) antibody binds to sequences which comprise the first 16 amino acids of the interleukin 3 molecule together with adjacent sequences. Rabbit anti-1-29 (IL-3) antibody would therefore be expected to have minimal effects on receptor binding. Indeed we have demonstrated by FACS analysis that Rabbit anti-1-29 (IL-3) antibodies recognise IL-3 when this molecule is bound to its receptor (362). The 2E11 . monoclonal antibody binds to the 130-135 peptide sequence in murine IL-3 (148). Again binding at this site would not be expected to interfere significantly with receptor binding. The 1A3 antibody binds to the same site but with lower affinity. We selected antibodies based on their known binding sites which we anticipated would not interfere with binding of murine IL-3 to its receptor and assayed these antibodies for neutralising properties using the R6X mast cell line (353), which proliferates in response to IL-3. In addition, as we anticipated, that antibodies would enhance the biological activity of IL-3 by augmentation of its molecular weight (thus inhibiting renal clearance), we screened these antibodies for their ability to form complexes with IL-3 and determined the size of these complexes using a Sepharose gel filtration column. We assayed column fractions for both IL-3 bioactivity and for antibody reactivity and compared the peaks of activity obtained with known molecular weight standards. 222 XI.2. Results XI.2.1. Rabbit anti-1-29 (IL-3), 2E11 mAb and 1A3 mAb recognise synthetic and native IL-3 in solid phase with high titers. To determine the reactivity of our antibodies with recombinant IL-3 in a non-denatured state, we used a sandwich.ELISA, which employs a sheep anti IL-3 antibody to capture recombinant IL-3 in solution. Binding of affinity purified Rabbit anti-1-29 (IL-3) antibodies to both recombinant and denaturedTL-3 could be detected to antibody concentrations of less than 0.5 ng/ml. Similar results were obtained with 2E11 mab arid 1A3 mAbs which could be detected to concentrations of less than 1 ng/ml (data not showri). XI.2.2.1. Rabbit anti-1-29 antibodies show weak neutralising properties in bioassay We hypothesised that Rabbit anti-1-29 (IL-3) antibodies would bind at sites which did not compromise the biological activity of IL-3. To test this hypothesis Rabbit anti-1-29 (IL-3) antibodies were assayed for neutralising or enhancing properties in vitro on the IL-3 responsive cell line, R6-XE4. Rabbit anti-1-29 (IL-3) antibodies produced modest neutralisation of the bioactivity of recombinant mIL-3 over the range of concentrations tested (Fig. 57). In repeat experiments, inhibition was only apparent when concentrations of greater than 50 ng/ml were used in combination with 1 unit of IL-3 (approximately 2000 fold excess of antibody combining sites over IL-3). The ratio of conceritrations employed in subsequent experiments, (lOng antibody: IU IL-3 ) were associated with negligible inhibition of biological activity. Control rabbit antibodies do not inhibit IL-3 bioactivity (data not shown). XI.2.2.2. 2E11 and 1A3 mAbs do not inhibit the bioactivity of murine interleukin 3 on R6-XE4 cell line. 2E11 and 1 A3 mAbs at concentrations up to 2 mg/ml did not inhibit the bioactivity of murine interleukin 3 on R6-XE4 cells (data not showri). 223 Figure 58 Gel filtration elution profile of IL-3 in the presence and absence of Rabbit anti-1-29 (IL-3) antibodies. A. 2000 units of recombinant IL-3 were applied to a Sephadex G100 column. 1ml fractions were eluted in PBS and assayed for IL-3 bioactivity ( Q ) on R6-XE4 cells . Results are expressed as mean counts per minute of 3H-thymidine for duplicate wells. A representative elution profile is shown. B. 2000 units of recombinant IL-3 were pre-incubated for 30 minutes at room temperature with 25ug of Rabbit anti-1-29 (IL-3) antibody and then applied to a Sephadex G100 column. 1 ml fractions were eluted in PBS and assayed for IL-3 bioactivity ( O ) on R6-XE4 cells and for antibody titer ( Q ) by ELISA on full length synthetic IL-3. Results of bioassays are expressed as mean counts per minute of 3H-thymidine uptake for duplicate wells.. Titer is expressed as the reciprocal of that dilution at which the ELISA absorbance is twice background. A representative elution profile is shown. 226 Figure 58 227 Figure 59 Gel filtration profile for IL-3 and Rabbit anti-1-29 (IL-3) antibody on Superose 6 column. 8 0 6 0 4 0 2 0 5 0 0 0 4000 r 3 0 00 2000 1000 1000 units of recombinant IL-3 were pre-incubated for 30 minutes at room temperature with lOug of Rabbit anti-1-29 (IL-3) antibody. 1 ml fractions were eluted in PBS and assayed for IL-3 bioactivity ( H ) on R6-XE4 cells and for antibody titer ( Q ) by ELISA against full length synthetic IL-3. JL-3 bioactivity is expressed as Units/ml , derived by comparison of bioassay titration curves with a known standard. Antibody titer is expressed as the reciprocal of that dilution at which ELISA absorbance is twice background. A representative elution profile is shown. Molecuiar weight markers are shown for comparison. 228 XI.2.3.2. 2E11 mAb and 1A3 mAb fail to augment the molecular weight of murine interleukin 3 as assessed by gel filtration chromatography. In order to determine if 2E11 mAb and 1 A3 mAb would form tight complexes with murine interleukin 3, 50 ug mAbs were mixed recombinant 2000 U murine IL-3 and applied to a gel filtration column as described in section X.6. Fractions eluted from the column were then assayed for IL-3 bioactivity using the R6X cell line and for antibody using an ELISA against recombinant IL-3 bound in the solid phase. The molecular weight ocorresponding to the fraction eluted was determined using known molecular weight standards prior to each column run. When murine IL-3 and 1A3 mAb were applied to the column together,TL-3 bioactivity was detected in fractions 10-16 with the peak activity appearing in fraction 13 (Figure 61). This corresponded to a molecular weight of approximately 32 kDa. The peak is sharp. The 1 A3 mAb activity was detected separately in fractions 7-10 with the peak actiovity in fraction 8,corresponding to a molecular weight of 150 kDa. This suggests that 1A3 is unable to stably augment the molecular weight of IL-3. When murine IL3 and 2E11 mAb were applied to the column together, interleukin-3 bioactivity eluted in fractions 10-15 with the peak activity detected in fractions 11-13 (Figure 62). This corresponds to a molecular weight of 32-75 kDa for interleukin-3. The peak is broad and bifid. The 2E11 antibody activity eluted separately in fractions 7-9 with the peak of activity in fraction 8, corresponding to a molecular weight of 150 kDa . This suggests that 2E11 is unable to form stable complexes which augment the molecular weight of murine interleukin-3. However unlike 1A3, 2E11 mAb does appear to be able to retard the elution of IL-3 bioactivity from the column. 229 Figure 60 Gel filtration elution profile of IL-3 in the presence of 1A3 monoclonal antibodies. 200 150 H > o CD .2 100 J3 50 0 f t ' / \ k - • • • • £ 10000 8000 > r6000 | o Q_ >< •4000 -|-CD 2000 •o 10 15 Fraction (ml) 20 2000 units of recombinant IL-3 were pre-incubated for 30 minutes at room temperature with 25ug 1A3 monoclonal antibody and then applied to a Sephadex G100 column. 1 ml fractions were eluted in PBS and assayed for IL-3 bioactivity ( A- ) on R6-XE4 cells and for antibody titer ( ) by ELISA on full length synthetic IL-3. Results of bioassays are expressed as mean counts per minute of 3H-thymidine uptake for duplicate wells. Titer is expressed as the reciprocal of that dilution at which the ELISA absorbance is twice background. A representative elution profde is shown. 230 Figure 61. Gel filtration elution profile of IL-3 in the presence and absence o f 2E11 monoclonal antibodies. Fraction (ml) 2000 units of recombinant IL-3 were pre-incubated for 30 minutes at room temperature with 25ug 2E11 monoclonal antibody and then applied to a Sephadex G100 column. 1 ml fractions were eluted in PBS and assayed for EL-3 bioactivity ( A ) on R6-XE4 cells and for antibody titer ( K ) by ELISA on full length synthetic IL-3. Results of bioassays are expressed as mean counts per minute of 3H-thymidine uptake for duplicate'wells.. Titer is expressed as the reciprocal of that dilution at which the ELISA absorbance is twice background. A representative elution profile is shown. 231 XI.3. Discussion We evaluated 3 antibodies to murine IL-3 produced in our laboratory with respect to their ability to enhance IL-3 bioactivity. We anticipated that antibodies with enhancing properties would show little or no neutralisation of IL-3 bioactivity iri vitro and would bind native IL-3 with high affinity resulting in the formation of stable high molecular weight complexes. The 2E11 and 1 A3 monoclonal antibodies showed no neutralising properties in vitro but were able to recognise native EL-3 in the sandwich assay. However the affinity of these antibodies for IL-3 appeared to be low and complexes formed between IL-3 and 2E11 mAb or 1A3 mAb readily dissociated on the gel filtration column. There was a suggestion that 2E11 mAb was able to retard the elution of IL-3 activity frOm the column and in repeated experiments IL-3 bioactivity eluted in two peaks with molecular weights slightly in excess of those reported for murine IL-3 when comparisons were made with known standards. This suggests that the 2E11 may be able to form complexes with IL-3, which retard its elution from the column, but that these complexes are unstable and dissociate readily. We therefore hypothesised that the 1A3 antibody would not show enhancing properties in vivo and that the enhancing properties of the 2E11 mAb (if any) would be weak. Rabbit anti-1-29 (IL-3) antibodies were able to recognise IL-3 in its native conformation and showed weak neutralising properties at high concentrations. Nevertheless at low concentrations, which still represented a 2000 fold excess of antibody, neutralising activity was not observed. On the gel filtration column, rabbit anti-1-29 (IL-3).antibodies formed stable complexes with IL-3 with molecular weights in excess of 100 kDa. More precise sizing experiments established the weight of these complexes as 200-250 kDa. We have previously shown by silver staining of SDS PAGE gels that the molecular weight of the Rabbit anti-1-29 molecule is 150 kDa and that the molecular weight of the IL-3 molecule is 32 kDa (data riot showri). Therefore we suspect that the majority of these complexes are composed of one antibody molecule and one or two IL3 molecules, (total molecular weight 180-220 kDa). We suspect that the smaller complexes predominate as the antibody is present in excess. It is also likely the sizing experiments 232 overestimate the weight of the complex as glycosylated proteins tend to elute at higher apparent molecular weights. The demonstration of only weak inhibitory properties together with a high affinity for IL-3" and the ability to form stable complexes predicted that the Rabbit anti-1-29 antibodies would have enhancing properties in vivo. These properties are explored in subsequent chapters. ' .233 XII. Pharmacokinetics XII.l. Introduction and aims In previous studies antibodies against a target molecule have been used to eliminate that target by enhanced renal clearance. This has been most clearly demonstrated for Fab.fragments used in the treatment of digoxin overdose (363). We hypothesised that anti-cytokine antibodies could enhance the activity of IL-3 by augmenting its molecular weight and decreasing clearance of TL-3 by glomerular filtration. The rate at which substances are excreted by the kidney depends on their molecular weight and charge. (364). Low molecular weight substances such as TL-3, which has a molecular weight of 24-32 kDa (53) are readily filtered and excreted. The half-life reported for murine IL-3 in the literature ranges from 10 minutes to 40 minutes (54, 57-59). Detailed pharmacokinetic studies establishing its pattern of distribution, volume of distribution and elimination half-life have not been published. The half-life reported for interleukin-3 may vary with the doses of IL-3 administered if some elimination mechanisms are saturated at high doses (54,365). In humans the pharmacokinetics of IL-3 have been studied in more detail and the elimination half life appears to be 23 minutes (61). We used a pharmacokinetic computer program to model the distribution and elimination of murine IL-3 and the IL-3 antibody complex. We hypothesised that prolongation of the half-life of IL-3 in vivo would be reflected primarily in an increase in the (3 elimination half-life as this component of the pharmacokinetic model most closely reflects renal elimination. However we did not know whether the formation of complexes with antibody would affect the rate of distribution of the cytokine in tissue and how modeling of the pharmacokinetics of IL-3 alone would compare with the pharmacokinetics of the antibody-fL-3 complex. 234 XI.2. Results XI.2.1. 2E11 monoclonal antibody does not significantly extend the biological half life of IL-3 in vivo. Simultaneous injection of 100 ug of 2E11 rhAb together with 1000 biological units of IL-3 did not result in significant prolongation of the half-life of interleukin 3 in vivo compared with mice injected with IL-3 alone (Figure 62). Although we obtained only a limited number of data points, ho IL-3 bioactivity could be detected in either group of mice at the 1.5 hour time point. However, bioactivity of IL-3 was enhanced at the 5 minute and the 45 minute time points in the antibody treated mice compared to controls. These results were consistent in repeat experiments. Unfortunately there were inadequate time points to construct a pharmacokinetic model. XI.2.2. Anti 1-29 IL-3 antibodies significantly extend the biological half-life of IL-3 in vivo. In a series of 3 experiments mice were injected with either IL-3, rabbit anti-1-29 (IL-3) antibodies or a combination of both. At specific time intervals following injection, the mice were sacrificed and serum obtained for estimation of IL-3 bioactivity and antibody titre. A representative . experiment is shown (Fig 63). No IL-3 bioactivity was detected at any time point in mice treated with the antibody alone. The data for the clearance of IL-3 alone and for the IL-3-antibody complex was best described by a 2 compartment model and was fitted to this model by the MULTIFORTE program using the Gauss-Newton method and weighting the. data by 1/Cp2 (the reciprocal of the plasma concentration squared). This gave an excellent fit to the data with a weighted sum of the squares (WSS) value of 0.1109 and a R 2 value of 1.00 for IL-3 alone and a WSS value of 0.03885 and a R 2 value of 1.00 for IL-3 and antibody. Using this method the a half-life for IL-3 alone was calculated to be 2.4 minutes and-the $ half-life was calculated to be 8.5. minutes. The a half-life for IL-3 together with antibody was 5.7 minutes and the (3 half-life was 72.5 minutes. No DL-3 bioactivity could be detected in the group treated with IL-3 alone after 235 Figure 62 Effect of 2E11 monoclonal antibody on the clearance of IL-3. 1.000-'-|. 0 2 0 4 0 6 0 8 0 1 0 0 Time (mins) Groups of mice were injected intravenously into the tail vein with 1000 biological units of recombinant IL-3 (A) or 1000 U IL-3 and 100 mg 2E11 mAb (X )• All injections were made in a total volume of 120ml 0.9% w/v saline. At intervals, groups of 4 mice were sacrificed and bled by cardiac puncture to obtain serum for estimation of IL-3 bioactivity and antibody titer. IL-3 bioactivity is expressed as Units per ml. Mean values ± SE are presented. A representative experiment is shown. 236 Figure 63 Effect of Rabbit anti-1-29 (IL-3) antibody on the clearance of IL-3. o Balb/c mice were injected intra-venously with either 1000 units recombinant IL-3 ( M ) or recombinant IL-3 pre-incubated with 10u\g of Rabbit anti-1-29 (IL-3) antibody for 30 minutes at room temperature ( 0 ) At specific time intervals animals were sacrificed and serum obtained by cardiac puncture for estimation of IL-3 bioactivity and antibody titer. IL-3 bioactivity is expressed as Units per ml. Mean values ± SE are presented for 4 mice for each point. (*) P<0.05 values are considered to be significantly different. A representative clearance is shown. 237 1 hour, whereas in the antibody treated group, IL-3 bioactivity could be detected up to 12 hours following injection. The total body clearance of the EL-3-antibody complex was 5.1 ml per hour compared to 45:4 ml per hour for IL-3 alone. The calculated volume of distribution was slightly smaller for the complex at 8.9 ml compared to 9.3 ml for IL-3 alone but was still in excess of the volume of the intravascular compartment (1-2 ml) suggesting that the complex is not restricted to this compartment. Antibody titre remained unchanged throughout the duration of the experiment (data not shown). This is compatible with the expected half-life for heterologous antibody which is 4-5 days. Clearances were performed in the presence of rabbit control IgG. Rabbit control IgG was ineffective in enhancing the half-life of IL-3 (data not shown). The a half-life of IL-3 in these experiments was 2.1 minutes and the P half-life was 15 minutes. To determine the effect of a higher dosage on IL-3 clearance, mice were injected with 10,000 U IL-3. The clearance of this high dose IL-3 was prolonged compared with low dose EL-3. The initial a half-life was 3.6 minutes and the P half-life, 30.8 minutes. The calculated volume of distribution was 13.6 mis and the total body clearance was 18.09 mis/hour (data not shown). 238 XI.3. Discussion A fundamental hypothesis of clinical pharmacokinetics is that a relationship exists between the pharmacological or toxic response to a drug and the serum concentration of that drug (366). For a given drug the rate of drug administration required to a achieve a given steady state concentration will be determined by the rate of drug elimination: Dosing rate = GL x Css . where CL is clearance and Css is the steady state concentration of drug For most drugs, the rate of elimination is a linear function of the concentration of the drug i.e. a constant fraction of the drug is eliminated per unit time. This type of pharmacokinetic profile is said to obey first order kinetics and the observed decrease in plasma levels of the drug is best described by the monoexponential function shown below; X t =Xoe"kt where X t represents the drug concentration at time t, XQ represents the initial drug concentration and k represents the elimination constant, which is unique to each drug. Although pharmacokinetic calculations that involve drug elimination generally use k values, it is more conventional to describe first order drug elimination in quantitative terms with reference to the plasma half life (tl/2) i.e. the time taken for plasma concentrations to fall by 50%. tl/2 is related to k by the following equation: tl/2 = In 2 k ; 239 The disposition of a drug following injection, is best described by the volume of distribution (Yd). This relates the amount of drug in the body to the concentration of the drug in blood or plasma (C). Vd = Amount of drug in body/ C This volume does not refer to an identified physiological volume, but only to the fluid volume Which would be required to contain all of the drug in the body at the same concentration as in the blood or plasma. The volume of distribution varies with the ability of the drug to penetrate tissue, the partition coefficient of the drug in fat and the degree of binding to plasma proteins and other tissues For many drugs the decrease in plasma levels with time after their intravenous administration cannot be accurately described by a monoexponential function. This occurs typically when the elimination phase is preceded by a distributive component involving uptake of drugs by peripheral tissues. In the simplest case, the pharmacokinetic data conform to a two compartment model, in which a central compartment represents distribution into and elimination from blood and or the extracellular fluid space; and the peripheral compartment comprises distribution into and elimination from a tissue reservoir or reservoirs. 240 In this case, elimination of the drug is said to follow multi-exponential kinetics and both an initial a half life (tl/2a), representing predominantly distribution into the peripheral compartment and a terminal or P half-life (tl/2p) representing predominantly elimination from the central compartment may be derived. The elimination of drugs with complex patterns of distribution and variable tissue binding may best be described by multi compartmental models. The MULTIFORTE computer program (354) is able to find the best fit mathematical model to describe a given set of pharmacokinetic data. For the IL-3 and IL-3-Rabbit anti-1-29 antibody complexes, the pharmacokinetic data was best fitted to a two compartment model, allowing us to derive an a and a p elimination half life. Unfortunately there were inadequate points on the curve to model the data for the 2E11 mAb. In these experiments, the 2E11 mAb was ineffective in significantly prolonging the elimination half-life of IL-3. There was some enhancement of IL-3 bioactivity observed at early time points suggesting that 2E11 mAb has some enhancing properties but that these are weak. These results are consistent with the characteristics reported for 2E11 mAb in the preceding chapter: 2E11 is able to recognise IL-3 in its native conformation at relatively high titre in an ELISA but is unable to form stable complexes with IL-3 in solution and readily dissociates from these complexes when passed over a gel filtration column. Nevertheless 2E11 can form temporary complexes with IL-3 which retard the elution of IL-3 bioactivity from a gel filtration column. We suspect that this instability of 2E11-IL-3 complexes underlies the relatively rapid clearance of 2E11 -IL-3 complexes in vivo. In contrast Rabbit anti-1-29 antibodies were able to significantly Increase the terminal elimination half-life of IL-3 in vivo . The p elimination half-life was most profoundly effected by the addition of antibody with a 7-10 fold enhancement. This supports the hypothesis that the antibody exerts its effects.by reducing renal clearance. Additional effects on the a half life, reflecting distribution to.tissue were also observed but were weaker (2-3 fold). This suggests that the antibody may moderately impair the rate of uptake of the IL-3 into tissue. The in vivo P half-life of 5-10 minutes for IL-3 alone, reported here, is in keeping with some (54,58,59,365) but not all studies of IL-3 half-life . Values of 30-60 minutes have also.been • '241 • ' - • - ' - . . . ' reported (62,365). The short half-life we observe may be related to the small amount of IL-3 injected (approximately 5-10ng). However, in combination with antibody, this amount of IL-3 is still able to exert marked biological effects. Higher concentrations of IL-3 exhibit more prolonged half-lives and in our studies the |3 half-life of 10,000 units of IL-3 was 30.8 minutes. This value is consistent with the studies of Metcalf etal. (54) and of Crapperet al, (365), where longer half-lives were been reported following injection of 100-1000 fold greater quantities of IL-3 than we used in these low dose clearances. In the experiments of Crapper et al (365) and Garland et al (59), injection of low dose IL-3 produced similar clearance profiles as observed in our experiments, with no IL-3 bioactivity detectable after 1 hour. A possible explanation for this discrepancy in the half-lives of IL-3 at different doses, is that the higher serum concentrations of TL-3 may saturate normal clearance mechanisms resulting in a prolongation of the elimination half-life. Alternatively, continuing distribution of IL-3 to tissue during the (3 phase may have resulted in a shorter half-life for low dose IL-3 than can be accounted for by elimination alone. Thus, the enhancing effects of the antibody might be attributable to effects on both distribution and elimination. This is suggested by the slightly greater volume of distribution observed for the high dose of IL-3. The half- life of radk>isotopically labeled IL-3 is reported as up to 5 hours (54,62). However this may not represent biologically active protein. The volume of distribution of the IL-3/rabbit anti-1-29 (IL-3) antibody complex appears to be only slightly smaller than the volume calculated for IL-3 alone, suggesting that either the complex can ultimately penetrate tissue effectively or that dissociation of the complex takes place, in vivo with diffusion of IL-3 into tissue. In addition, the volume of distribution of the IL-3/rabbit anti-1-29 (IL-3) antibody complex is clearly greater than the vascular volume suggesting that the antibody does not "trap" IL-3 within the vasculature. Binding of the Fc portion of the antibody-cytokine complex to Fc receptors might be expected to reduce the biological half-life of the complex by initiating clearance by the reticulo-endothelial system, nevertheless, in these experiments, the biological half-life of the complex was increased. The efficiency of Fc receptor binding is determined by antibody class and by the size of the immune complexes formed (367-369). The binding efficiency of this antibody-cytokine 242 complex is likely to.be low as sizing experiments using a Superose-6 column suggest that the majority of the complex comprises one IL-3 molecule bound to one or less likely, two antibody molecules. This type of complex would activate complement with low efficiency (367) and bind only to high affinity Fc receptors (368). Once bound to receptors it would be very slowly internalised (369). The use of anti-peptide antibodies therefore allows for the production of small immune complexes which are more effective as carriers for their target cytokine. Some binding of the complex to Fc receptors in hemopoietic organs and the reticulo-endothelial system may contribute to the large apparent volume of distribution observed for the complex. Binding of IL-3 within tissue would reduce the concentration of IL-3 within the vasculature and would therefore increase the apparent volume of distribution. The findings that Rabbit anti-1-29 could extend the biological half life of TL-3 significantly in vivo prompted us to investigate the effects of these antibodies on the biological response to chronic IL-3 treatment in mice. 243 XIII. Biological Effects of Enhancing Antibodies XIII.l Introduction and Aims The combination of murine IL-3 with Rabbit anti-1-29 (IL-3) antibodies results in a significant prolongation of the elimination half-life of IL-3 without significant effects on the biodistribution of the IL-3. We therefore hypothesised that the combination would result in enhanced biological activity, compared to IL-3 alone. In this chapter we present the results of biological experiments in which series of mice were injected with either IL-3 or a combination of IL-3 and Rabbit anti-1-29 (IL-3) antibodies. The biological readouts used for IL-3 bioactivity in vivo were increases in the numbers of IL-3 responsive colony forming cells (CFC) in the spleen and bone marrow, increases in mast cell numbers in lung, spleen, jejunum, liver and lymph nodes and increases in mast cell precursor frequencies in spleen and bone marrow. IL-3 has been identified as one of the early acting growth factors for hemopoietic colony forming units (CFUs) and consequently affects production of all hemopoietic lineages (370-375). IL-3 is able to support the growth of stem cells, CFU-blast, CFU-GEMM, CFU-MEG, BFU-E, CFU-GM CFU-Eo/Baso, CFU-G and CFU-M colonies. The numbers of colonies which can be established in soft agar culture will therefore reflect the growth promoting activity of IL-3 on hemopoietic precursors in bone marrow and spleen. The types of colonies formed would be expected to be diverse because of the pleiotropic effects of IL-3. At later stages of maturation, growth promoting activity is confined to the neutrophil lineage up to the promyelocyte stage and the terminal differentiation of eosinophils and basophils (376,377). In these more mature CFUs, IL-3 delays the differentiation process for some further cycles and then switches the lineage choice to the eosinophil/basophil lineage (371,374,377,378)). The relationship between basophils and mast cells is unknown. The cells share many functional characteristics and appear to respond in a similar manner to IL-3, however marker studies suggest that they are derived from different lineages (379,380). Mast cells are long-lived cells and persist in long term cultures of lymphohemopoietic tissue, hence the name P-cell or 244 "persisting" cell (381). Their presence in long term cultures of spleen or bone marrow is evidence of the bioactivity of IL-3 and forms the basis of the P-cell assay for IL-3 (355). . Mature murine mast cells also express IL-3 receptors and IL-3 promotes the growth, differentiation and effector functions of these cells (382). Increases in tissue numbers of mast cells may therefore reflect both an increase in mast cell progenitor numbers and proliferation of mature mucosal mast cells in mouse. Interestingly mature human mast cells do not express IL-3 receptors (383). v ." • In order to determine whether prolongation of the biological half-life of IL-3 would result in augmentation of the biological effects of IL-3 treatment, mice were allocated to one of four treatment regimens. In order to maximise any effect attributable to prolongation of IL-3 half-life, the dosing interval chosen for each group was 24 hours. The biological effects of these regimens were assessed using a number of IL-3 responsive assays. 245 XIII.2 Results XIII.2.1 Anti-IL-3 antibodies increase CFCs in the spleen. The frequency of IL-3 responsive CFCs in the spleen of treated mice was determined by an in vitro colony assay (Fig. 64). The frequency of IL-3 responsive colonies obtained from spleen cells from mice injected with IL-3 alone or from mice injected with Rabbit anti-1-29 (IL-3) antibody alone was not significantly different from that of control animals. The combination of this antibody and IL-3,resulted in a 3-7 fold increase in the number of IL-3-dependent CFCs compared to controls. Colonies contained cells exhibiting both macrophage and neutrophil morphology consistent with the known biological effects of IL-3 (370-375). The data indicate that the combination of rabbit anti 1-29 (IL-3) antibody with IL-3 results in an enhancement of the biological effects of IL-3. Control IgG was ineffective in augmenting the response to IL-3 and the number of colonies scored was not significantly different from controls (data not shown). XIII.2.2.1 Anti IL-3 antibodies increase mast cell precursor frequencies in the spleen. The frequency of mast cell colony precursors detected .when spleen cells are maintained in liquid culture for 15 days has been shown to be a sensitive and specific assay of the in vivo response to treatment with IL-3 (355). The observed frequency of mast cell precursors among spleen cells from mice treated with both antibody and IL-3 was 3-5-fold greater than that seen in control animals or in animals treated with IL-3 or Rabbit anti-1-29 (IL-3) antibody alone (Fig. 65A). The frequencies of mast cell precursors in Control IgG and IL-3, Rabbit anti-1-29 (IL-3) antibody alone or IL-3 alone treated mice were not significantly different from control values using these treatment regimens.(Figs 65) 246 Figure 64 Frequency of IL-3 responsive colonies derived from spleens of mice undergoing different treatments. a o u > a o •  a, co • >> o C" <D 3. CT1 OH •50-, 40H 8 30 20 1 0 0 > */« * « »% «*» • saline antibody IL-3 IL-3 + "antibody Balb/c mice were treated with either saline, Rabbit anti-1-29 (IL-3) antibody alone, IL-3 alone, or EL-3 and Rabbit anti-1-29 (IL-3) antibody by daily iv injection. After 5 days mice were sacrificed and spleen cells disaggregated and diluted to 10.5 in 3% agar. 2% WEHI-3B conditioned media was added as a source of EL-3. After 7 days agars were scored blind for colony formation. Mean values ± SE are presented for 5 mice. (*) P<0.05 values are considered to be significantly different. 247 Figure 65 Frequency of mast cell precursors in spleens of mice undergoing different treatments. Balb/c mice were treated with either saline, Rabbit anti-1-29 (IL-3) antibody alone, IL-3 alone or IL-3 and Rabbit anti-1-29 (IL-3) antibody by daily iv injection. After 5 days mice were sacrificed and spleen cells disaggregated and serially diluted into 96 well tissue culture plates containing 150 ill per well of RPMI 10% FCS and 2% WEHI-3B conditioned media. After i5 days incubation at 37'C, 5% C02, plates were scored blind for mast cell colony formation. The frequency of negative wells in each tray.was used to derive a precursor frequency. Precursor frequencies + 95% confidence limits are shown for pooled spleens from 5 mice for each treatment group. B. Similar analysis is shown for mice treated with saline, IL-3 alone, IL-3 and rabbit control IgG, and IL-3 and Rabbit anti-1-29 (IL-3). 248; Figure 65 1-2-n 1 0 -a 8 6 H s a l i n e a n t i b o d y IL-3 IL-3 + antibody 35 1 30 ?5 20 1 5 1 0 5 0 IL-3 + Rabbit; IgG • control IL-3 + . Rl-29 (IL-3) antibody 249 XIII.2.2.2. Anti-IL-3 antibodies decrease mast cell precursor frequencies in the bone marrow. The frequency of mast cell precursors in the bone marrow was slightly reduced in mice treated with both the Rabbit anti-1-29 (IL-3) antibody and IL-3 compared with IL-3 of saline treated controls (Fig. 66). XIII.2.3. Anti-IL-3 antibodies increase the numbers of mature mast cells in the spleen but not in other tissues. Spleen sections taken from mice treated with both IL-3 and Rabbit anti-1-29 (IL-3) antibody showed significant increases in the numbers of mature mast cells observed compared with other groups (Table 9) (p<0.05). Many of the mast cells observed in the EL-3 and Rabbit anti-1-29 (IL-3) antibody treated sections were small immature mast cells with few alcian-blue positive granules. These observations are compatible with a relatively recent increase in proliferation of precursor cells. The numbers of megakaryocytes in the sections examined was not significantly different between the groups. Mastcell numbers were not increased in either lung, jejunum, liver or lymph nodes (data not shown). No abnormal histopathological changes were seen on any sections stained with H&E. . • XIII.2.4. IL-3 bioactivity is not detectable in serum of treated mice 24 hours after injection. IL-3 bioactivity was not detected in serum samples obtained from each group at the conclusion of the experiment. The data indicates that no treatment regimen resulted in accumulation of IL-3. 250 Figure 66 Frequency of mast cell precursors in bone marrows of mice undergoing different treatments. s 2 5 i o 1 -20-saline antibody IL-3 IL-3 + antibody Balb/c mice were treated with either saline, antibody alone, EL-3 alone or IL-3 and antibody by iv injection daily. After 5 days mice were sacrificed and bone marrows serially diluted into 96 well tissue culture plates containing 150|il of RPMI, 10% FCS and 2% WEHI-3B conditioned media. After 15 days incubation at 37'C, 5% C02, plates Were scored blind for mast cell colony formation. The frequency of negative wells in each tray was used to derive a precursor frequency. Precursor frequencies + 95% confidence limits are shown. .Analyses are shown for mice treated with saline, IL-3 alone, IL-3 and rabbit control IgG, and IL-3 and Rabbit anti-1-29 antibody. 251 Table 9. Mast cell and megakaryocyte frequency in the spleen. Treatment Mast cells Megakaryocytes per high power field (x40)' per high power field (x40) Saline 0.27 ±0.09 IA ±0.33 Rabbit anti-1-29 (IL-3) 0.32+0.16 1.2 +.0.37 Antibody TL-3 0.40 ±0.13 0.9 ±0.21 IL-3 & Rabbit anti-1-29 1.92 ±0.36* 1.5 ±0.07 (TL-3) Antibody Balb/C mice were treated with either saline, antibody alone, IL-3 alone or IL-3 and antibody. After 5 days mice were sacrificed and spleens fixed with methanol:formalin:acetic acid (85:10:5 v/v) overnight at 4'C. Sections were stained with either alcian blue and saffranin-0 or haematoxylin and eosin for determination of mast cell and megakaryocyte numbers and histopathological changes. (*) P<0.05 values are considered to differ significantly from other values. 252 XIII.3. Discussion Our results consistently demonstrate between 3 and 10 fold enhancement of the biological activity of IL-3 by combination with Rabbit anti-1-29 (IL-3) antibodies. This effect was not observed when Rabbit anti-1-^ 29 antibodies were given alone and was not observed for control rabbit IgG. These results are consistent .with the findings of a 7-10 fold prolongation of the elimination half-life of IL-3 by Rabbit anti-1-29 (IL-3) antibodies. At these doses minimal biological effects were observed for IL-3 alone and in some experiments the biological response to 1,000 units of IL-3 alone was similar to controls. This probably reflects the rapid renal elimination, of this small dose of IL-3 when administered as a single daily bolus injection. The numbers of IL-3 responsive CFCs derived from the spleens of mice treated with the Rabbit anti-1-29 (IL-3) antibody-IL-3 complex was significantly higher than from controls and significantly higher than observed in mice treated with IL-3 alone or IL-3 in combination with control IgG. Colonies contained both neutrophils and macrophages by morphology, consistent with the known biological effects of IL-3 (370-375). Interleukin-3 responsive CFCs in the bone marrow were not increased by any of the treatments used (data not shown), and this may require higher concentrations of IL-3 (John Schrader-personal communication). The anti-IL-3 antibody-mediated increase in mast cell precursors in the spleen observed in the P-cell assay, is of a similar magnitude to that observed in mice bearing the IL-3-producing tumour WEHI-3B (five fold)(148,365) or following T-lymphocyte activation in graft-versus-host reactions (fourfold to fivefold)(41). Observations in these pathological conditions associated with modest elevations levels of IL-3 in the plasma indicate that increases in the frequency of mast cell and hemopoietic precursor levels of this magnitude are associated with dramatic increases iri hemopoietic tissue e.g. red pulp in the spleen (41). The marked increase in primitive IL-3 responsive precursor cells in the spleen is associated with similar increase in the numbers of mature cells in spleen as shown by histochemical studies on this organ. It is unlikely that this results from stimulation of early progenitors by IL-3 as this would be followed by a delayed increase in mature cells and would not become apparent during the 253 time frame of. these experiments. IL-3 will stimulate proliferation of mature mast cells and of immature megakaryocytes and the increases in the numbers of mast cells observed in the spleen may reflect this effect although increases in megakaryocytes were not observed. No increases in mature mast cells were observed in lung or.jejunum and higher concentrations of IL-3 may be required to produce effects in these tissues. The frequency of mast cell precursors was increased dramatically in spleen following treatment with the antibody-cytokine, complex, however the frequency of mast cell precursors in the bone marrow was consistently diminished. This observation is in agreement with the findings of Moore er al. (58) but in contrast to those of Metcalf et al. (5.4). It appears from our observations that low dose TL-3 treatment causes a release of mast cell precursors from the bone marrow, which redistribute to the spleen. A possible mechanism for this effect is via down-regulation of the c-kit receptor (385). . In summary, the combination of Rabbit anti-1-29 (IL-3) antibodies and IL-3 was effective in enhancing the biological activity of IL-3 and biological effects attributable to IL-3 could be observed with doses of IL-3 as low as 1,000 U daily given intravenously for 5 days. 254 XIV. Discussion We have demonstrated that the combination of IL-3 with an anti-IL-3 antibody administered intravenously results in a 7-10 fold prolongation of the elimination half life of IL-3 and a 3-10 fold enhancement of its bioactivity in vivo. These effects were observed with very small doses of IL-3 (1,000U daily) which are normally associated with rapid in vivo clearance and little or no in vivo bioactivity. Because of its rapid elimination half life in most clinical studies IL-3 has been administered by sub-cutaneous injection (reviewed in 386). Sub-cutaneous injection is associated with a different pharmacokinetic profile to intravenous injection, with a slow absorption phase followed by a similar elimination (3 half life to intravenous (54). Van Gils et al (56) have studied the optimum administration schedule for IL-3 and compared intravenous infusion with bolus subcutaneous administration. They determined that the bioavailability of IL-3 following subcutaneous injection was only 40 % of that following i.v. administration and that in keeping with the more limited bioavailability of IL-3 following s.c. injection, continuous i.v. infusion was more effective in stimulating hemopoiesis. We believe that the combination of EL-3 with antibody given intravenously can mimic some aspects of an intravenous infusion without significant changes in the biodistribution of EL-3 bioactivity. In addition very small doses of EL-3 can be administered and still exert profound biological effects. Alternative mechanisms for enhancing the bioactivity of cytokines have been discussed in the introduction section: There are advantages and disadvantages associated with each method. PEGylation has the advantage of decreasing the immunogenicity of the target cytokine but can interfere with receptor binding and is not reversible (78-82). In addition the conjugated molecule may be restricted to the vascular compartment because of its large size. Fusion molecules may also show a similar restricted biodistribution and theoretically may be more immunogenic because of the creation of novel antigenic epitopes in the fusion process. The extended half-life of fusion molecules is also irreversible (83,84). Similar disadvantages are observed for liposomal encapsulation ( 387). 255 There are a number of advantages to using antibodies to prolong the elimination half life of cytokines: Firstly the interaction between the antibody and the cytokine is reversible and the biodistribution of the cytokine will be similar to the native molecule. Because the association between the cytokine and the antibody is reversible, the prolongation of the half life of the cytokine produced by this association can be reversed by administration of an excess of peptide containing the binding epitope or by aggregation of the antibody molecules using either a secondary antibody or using the biotin-avidin system (vide infra). The resulting large immune complexes are rapidly removed by the reticuloendothelial system. It may also be possible to boost endogenous levels of a cytokine by administration of antibody alone. The presence of antibody would then delay clearance of the target cytokine resulting in enhanced biological activity. In this manner it might be possible to interfere with the normal immune response, for example by directing the immune response towards either a predominantly humoral (TH2) or cytotoxic (TH1) response by affecting the balance of IL-4, EL-10 or y-interferon (388). . In addition currently available technology allows for manipulation of the properties of the antibody. Monoclonal antibodies directed against a specific epitope on the target cytokine can be engineered so that the CDRs (complementarity determining regions) of the monoclonal are incorporated into an expression system encoding human variable and constant regions (389-39). As a result the 'humanised' antibodies contain only a small fraction of the amino acid sequence of the mouse original. This process minimises the potential for an immune response to the antibody and results in significant prolongation of the half life of autologous antibodies- In addition the constant region of the antibody can be selected to minimise complement activation, and Fc receptor binding, resulting in further prolongation of the half-life of the antibody. The ideal isoform for these purposes would be the human IgG4 isotype. In addition the bivalent nature of antibody binding suggests a potential for targeting cytokines to specific cells (394-398). For example, chimeric Afunctional antibodies could be used to target IL-3 to CD34 positive cells, potentially mitigating adverse effects on more mature cells. In the future it may also be possible to build antibody molecules from components derived from phage libraries, bypassing the requirement for animal immunisation and hybridoma technology (22,23). 256 At the same time that we published our results showing enhancement of the bioactivity of IL-3 in vivo using anti-IL-3 antibodies, Finkelman's group published data on enhancement of the bioactivity of IL-4 using neutralising antibodies and less detailed data on enhancement of TL-3 and IL-7 (399). Our two publications are complimentary. Finkelman's group asserted that neutralising activity was a pre-requisite for antibody enhancement of IL-4 activity. The mechanism that they propose for enhancement is identical-to "the one that we propose, i.e. prolongation Of the elimination half life, excepting that they believe that the prolongation they observe, is due to protection of the target molecule from proteolytic degradation This does not appear to be logical ' unless the major elimination mechanism for all three cytokines is proteolytic degradation affecting . sequences within the receptor binding site. At present we have no evidence for this. We suspect that in their publication, neutralising activity was a surrogate for high affinity. We have found that only a limited number of antibodies are able to increase the effective molecular weight of their target cytokine by sizing experiments. As Finkelman's group present data on only one non-neutralising (and ineffective) antibody and do not provide data on affinity for this antibody, we suspect that the failure to observe enhancement, may be due to the inability of the antibody to form stable complexes. The neutralising activity of the antibody has consequences with respect to interpretation of the half life data. Because, the antibody has neutralising activity, conventional bioassays cannot be used to detect the presence of the cytokine:. Finkelman's group used a sandwich assay, similar to ours to determine levels of IL-4 iri the circulation. They modify this assay to omit the second antibody step. Unfortunately this assay will tend to overestimate the amount Of bioactive IL-4 present as proteolytic fragments may also be detected; Metcalf and Nicola (54) found significant differences in the half-life of IL-3 when radioactive material was compared with bioactive material in the same experiment. The half-life of radioactive IL-3 was 5 hours. The half-life of bioactive IL-3 was 50 minutes. The discrepancy between these two values was thought to be due to. proteolytic degradation of the IL-3 There are a number of other interesting.differences between our results and Finkelman's: 257 In Finkelman's paper, significantly greater amounts of cytokine were used in order to produce these biological effects. Finkelman used 1000 fold greater amounts of EL-4, IL-3 and EL-7 and 10 fold greater amounts of antibody. This may result in prolongation of the elimination half-life of the cytokine as we have shown that the clearances of cytokines are prolonged at.higher concentrations. In addition, the ratio of cytokine to antibody used is high (1:5) compared to our studies (1:1000). We found that at higher ratios there was an increase in the amount of unbound cytokine. Nevertheless in Finkelman's studies, the affinity of the antibodies was such that at a ratio of 1:5 (cytokine:antibody), at least"75% of the IL-4 was bound by antibody. This suggests that the affinity of their antibody was far higher than ours and may explain why the prolongation of half-life and enhancement of bioactivity observed was greater. Interestingly the high ratio of cytokine to antibody used in their experiments has other consequences: Finkelman's group do not present any data on the ability of .their antibodies to neutralise IL-4 bioactivity at this ratio and indeed, increasing the amount of antibody present abrogated the enhancement observed. It is possible that at these ratios, the neutralising activity of the antibody is weak. In addition at these ratios, a higher proportion of the antibody-cytokine complexes will contain two IL-4 molecules. This raises the possibility that some of the enhancement may be due to cross-linking of receptors. In our studies the antibody is present in so great an excess that it is unlikely that any of the antibody-cytokine complexes contain 2 cytokine molecules. Since no data is presented on the size of the complexes in Finkelman's paper, it is unclear whether larger complexes can be formed at these ratios by cross-linking through IL-4 molecules. However it seems reasonable to expect that only a single epitope would be present within the DL-4 molecule so that only a single antibody could bind. The finding that cross linking the antibodies by biotinylating the antibody and then injecting avidin is intriguing, and suggests a mechanism whereby the enhancing effects of antibody could be terminated. There are a number of different mechanisms whereby antibodies may enhance the bioactivity of their target molecule (reviewed in 115). The mechanism that we describe here, prolongation of the elimination half life by reducing renal clearance has been reported elsewhere (400,401). In addition antibodies may extend half-lifes by protecting their target from proteolytic 258 degradation by blocking binding of proteolytic enzymes (this would apply equally to neutralising and non-neutralising antibodies). Reviewed in (115). High affinity antibodies may enhance the activity of their target by directing cytokine-to-high affinity receptors. These receptors can compete effectively for cytokine binding, whereas low affinity receptors cannot. This may consequently restrict cytokine binding to specific cell types. This mechanism has been described for human growth hormone (402) and for IL-la (403,404). The presence of even a limited number of antibodies to IL-la is thought to block absorption to fibroblasts and endothelial cells which possess low affinity binding sites (Kd = 5.x 10" 10 M) and to direct the cytokine to high affinity binding sites in lymphocytes (Kd =10"! ^ M) (116). In addition, antibodies may cross-link cytokine receptors through their target cytokine resulting in rapid dimerisation. The dimerisation of receptors results in an enhanced response to the cytokine. This mechanism has been reported for insulin and for EGF (epidermal growth factor)(405,406) and is only observed with bivalent antibody and not with Fab fragments. Antibodies may also enhance bioactivity by stabilising the interaction between the cytokine and the receptor by conformational modification of the cytokine. The evidence for this mechanism is less substantive. However it is hypothesised that this mechanism underlies the enhancement of bioactivity observed with some antibodies to TSH. These antibodies are equally effective as Fab fragments but do not appear to extent the half life of TSH (407). The ability of some antibodies to enhance the bioactivity of their target cytokine has important consequences. It is interesting to note that a substantial proportion of the cytokine binding activity of naturally occurring auto-antibodies to IL-la and IL-6 resides in the IgG4 fraction (11.6). As previously noted, this isotype has the ideal characteristics for an enhancing : antibody i.e. limited lattice formation, long serum half-life and inability to activate complement (408) It is possible that these auto-ahtibodies function as physiological carriers for their target cytokine in a similar manner as has been.described for a2 macroglobulin (409,410) and for soluble cytokine receptors (411,412). Ih addition it is possible that auto-antibodies produced following therapy with a cytokine, may also enhance the activity of the cytokine. Our results, 259 together with those of Finkelman suggest that caution should.be exercised in interpreting the results of in vivo studies in which antibodies are used to neutralise their target cytokine. In summary, we have demonstrated that antibodies to IL-3 may enhance the activity of IL-3 by extending its elimination half life. Detailed pharmacokinetic studies suggest that the E L - 3 anti-E L 3 antibody complex has a similar biodistribution to the native cytokine. We have demonstrated that iri combination with antibody, profound biological effects can be observed with very small doses of E L - 3 . We have identified characteristics which predict for the presence of enhancing properties in antibodies. These studies should facilitate development of therapeutically useful antibodies and work in our own laboratory has identified si monoclonal antibody to human E L - 3 which has properties suggestive of enhancing activity (data not shown). In addition our work suggests that studies using neutralising antibodies should be interpreted with caution and supports the hypothesis that auto-antibodies may function as physiological carriers for their target cytokine. 260 XV. References 1. Abbas, A.K., A.H. Lichtman and J.S. Pober. 1994. General properties of immune responses. In Cellular and Molecular Immunology. W.B. Saunders Co. Philadelphia. 2. Davies, D.R. and H. Metzger. 1983. Structural basis of antibody function. Ann .Rev. Immunol. 1:87, 3. Hengartner, H., T. Meo and E. Muller 1978. Assignment of gene for Igk and H chains to chromosomes 6 and 12 in mouse. Proc. Natl. Acad. Sci .USA. 75:4494. 4. Brodeur, P and R. Riblett. 1984. The immunoglobulin heavy chain variable region (Ig-H-v) in the mouse I. 100 IgH-V genes comprise 7 families of homologous genes. Eur. J. Immunol. 14:922. 5. Honjo, T. 1983. Immunoglobulin genes. Annu Rev Immunol. 1:499 . 6. Reth, M.G. and F.W. Alt. 1984. Novel immunoglobulin heavy chains are produced from DJH gene segment rearrangement in lymphoid cells. Nature 312:418. 7. Brack, C, M. Hirama, R. Lenhard-Schuller and S. Tonegawa. 1978. A complete immunoglobulin gene is created by somatic recombination. Cell 15:1. 8. Early, P., H. Huang, M. Davis, K. Calame and L. Hood. 1980. An immunoglobulin heavy chain variable region gene is rearranged from three segments of DNA: Vh, D and JH. Cell 19:981. 9. Sakano, H., R. Maki, Y: Kurosawa, W. Reoder and S. Tonegawa. 1980. Two types of somatic recombination are necessary for the generation of complete immunoglobulin heavy chain genes. Nature 286: 676. 10. Tonegawa, S. 1983. Somatic generation of antibody diversity. Nature 308:145. 11. Akira, S. H. Sugiyama, M. Yoshida, H. Kitkutani, Y. Yamamura and T. Kishimoto, 1983. Isotype switching in murine pre-B cell lines. Cell34:545. 12. Radbruch, A., C. Burger, S. Klem and W. Muller.1986. Control of immunoglobulin class switch recombination. Immunol. Rev. 89:69. 13. Nossal, G. 1992. The molecular and cellular basis of affinity maturation in the antibody response. Cell 68:1. 14. Griffith, G.M. C. Barek, M. Kaartinen and C. Milstein. 1984. Somatic mutation and the maturation of the immune response to 2-phenyl oxazolone. Nature 312:271. 15. Kocks, C. and K. Rajewsky. 1988. Stepwise intraclonal maturation of antibody affinity through somatic hypermutation. Proc. Natl. Acad. Sci. USA .90:2385. 16. Rajewsky, K. 1996. Clonal selection and learning in the antibody system. Nature 381:751. . 17. Dadi, H.K., R.J. Morris, E.C. Hulme and N.J. Birdsell. 1984. Antibodies to a covalent agonist used to isolate the muscarinic cholinergic receptor from rat brain. In Investigation of Membrane Located Receptor, (Eds. E.Reid, G. M. W. Cook and D.J. Moore). New York Plenium Press. p425 261 1.8. Crumpton, M.J. 1986. In Synthetic Peptides as Antigens (symposium 119) Ciba Foundation. p93. 19. Regenmortel, M.H.V. 1989. Structural and functional approaches to the study of protein antigenicity. Immunol. Today 10:8266. 20. Kohler,G. and C. Milstein. 1975. Continuous cultures of fused cells secreting antibody of pre-determined specificity. Nature 256:495 21. de St. Groth, S.F. & D. Scheidegger. 1980. Production of monclonal antibodies: strategy and tactics. J. Immunol. Methods 35:1. 22. Winter, G. and C. Milstein 1991. Man-made antibodies. Nature 349:293. 23. Winter, G., A.D. Griffith, R.E. Hawkins and H.R. Hoogenboom. 1994: Making -antibodies byphage display technology. Annu Rev Immunol. 12:433, 24. Ghetie, V., S. Popv, J. Borvak, C. Radu, D. Matesoi ef al 1997. Increasing the serum half life of an IgG fragment by random mutagenesis. Nature Biotech 15:637. 25. Hollinger, P., M. Wing, J.D. Pound, H. Boulton and G. Winter. 1997. Retargeting serum immunoglobulin with bispecific diabodies. Nature Biotech 15:633. 26. Kessler, S.W. 1975. Rapid isolation of antigens from cells with a staphylococcal protein A-antibody absorbent: Parameters of the interaction of antibody-antigen complexes with protein. J: Immmunol. 115:1617. 27. Blake, M.S., K.H. Johnston, G.J. Russel-Jones and E.C. Gotschlich. 1984. A rapid sensitive method for the detection of alkaline phosphatase conjugated anti-antibody on Western blots. Anal. Biochem. 136:175. 28. The Vlth International Workshop and Conference on Human Leukocyte Differeritaition Antigens, Garland Publishing Inc. London (in press). . 29. Hathcock, K.S., G. Laszlo, H.B. Dickler, S.O. Sharrow, P. Johnson, I.S. Trowbridge and RJ. Hodes. 1992. Expression of the variable exon A-, B- and C-specific CD45 determinants on peripheral and thymic T cell populations. J. Immunol. 148:19. 30. Schrader, J. W. 1985. The pan-specific hemopoietin of activated T lymphocytes, interleukin 3. Annu. Rev. Immunol. 4:205. 31. Nimer, S. D. and H. Uchida. 1995 Regulation of granulocyte-macrophage colony-stimulating factor and interleukin 3 expression. Stem Cells 13:324 32. Wodnar-Filipowicz, A., CH. Heusser, and C.Moroni. 1989. Production of the haemopoietic growth factors GM-CSF and interleukin 3 by mast cells in response to IgE receptor mediated activation. Nature 339:150. . 33. Razin, E., K.B. Leslie and J.W. Schrader. 1991. Connective tissue mast cells in contact with fibroblasts express interleukin 3 mRNA. Immunology 146:981. 34. Kita, H., T. Ohnishi, Y. Okubo, D. Weiler, J.S. Abrams, and G.J. Gleich. 1991. Granulocyte/Macrophage colony-stimulating factor and interleukin 3 release from human peripheral blood eosinophils and neutrophils. J. Exp. Med .174:745. .262 35. Warringa, R.A.J., L. Koenderman, P.T.M. Kok, J. Kreukniet, and P.L.B. Bruijnzeel. 1991. Modulation and induction of eosinophil chemotaxis by granulocyte-macrophage colony-stimulating factor and interleukin 3. Blood 77:2694. 36. Frendl, G. and D.I. Beller. 1990. IL-3 functions as a macrophage-activating factor with unique properties, inducing la and lymphocyte function-associated antigen -1 but not cytotoxicity. J. Immunol. 144:3392. 37. Schrader, J.W., I. Clark-Lewis, R.M. Crapper, K.B. Leslie, S. Schrader, . G. Varigos and H.J. Ziltener 1988. The pan-specific hemopoietin interleukin 3; Physiology and pathology. Lymphokines 15:281. 38. Xia, X, L. Li and Y.Y. Choi. 1992. Human recombinant EL-3 is a growth factor for normal B cells. J. Immunol. 148:491. 39. Metcalf, D. 1989. The molecular control of cell division, differentiation, commitment and maturation in haemopoietic cells. Nature 339:27. 40. Cockayne, D.A., D.M. Bodine, A. Cline, A. Nienhuis and CE. Dunbar. 1994. Transgenic mice expressing antisense interleukin-3 RNA develop a B-cell lymphoproliferative syndrome or neurological dysfuction. Blood 84:2699. 41. Crapper, R.M., and J.W. Scrader. 1986. Evidence for the in vivo production and release into the serum of a T-lymphokine, persisting-cell stimulating factor (PSF), during graft-versus-host reactions. Immunology 57:553. 42. Chan, W.L., H.J. Ziltener, and F.'Y. Liew. 1990. Interleukin 3 protects mice from acute herpes simplex virus infection. Immunology 71.-358. 43. Ganser, A. 1993. Clinical results with recombinant human interleukin-3. Cancer Invest. 11:212. 44. Nemunaitis, J., F.R. Appelbaum, J.W. Singer et al 1993. Phase I trial with recombinant human interleukin-3 in patients with lymphoma undergoing autologous bone marrow transplantation for malignant lymphoma. Blood 2:3272. 45. Biemsa, B., P.H.B.. Willemse, N.H. Mulder et al. 1992. Effects of interleukin -3 after chemotherapy for advanced ovarian cancer. Blood 80:1141. 46. Ganser, A., G. Seipelt, A. Lindemann et al .1990. Effects of recombinant human interleukin-3 in patients with myelodysplastic syndromes. Blood 6:455. 47. Vose, J.M., A. Kessinger, P.J. Bierman et al. 1992. The use of rhIL-3 for mobilization of peripheral blood stem cells in previously treated patients with lymphoid malignancies. Int. J. Cell Cloning 10:62. 48. Von Roenn, J.H., I.I. Gordon, W. Grace and S. Lee. 1996. Recombinant IL-3 in HIV sero-positive patients with cytopenias: effective long term therapy. Proc. ASCO 15:305. 49. Shoenfeld, Y and P. Fishman. 1994. Role of EL-3 in the antiphospholipid syndrome. Lupus 3:259. 50. Vose, J.M. and J.O. Armitage. 1995. Clinical applications of haemopoietic growth factors. J. Clin. Oncol. 13:4 1023. 263 51. Fay, J. and S.H. Bernstein. 1996. Recombinant human interleukin-3 and granulocyte macrophage colony stimulating factor after autologous bone marrow transplantation for malignant lymphoma. Seminars in Oncol. 23:22. 52. Neumunaitis, J. 1996. Cytokine mobilized peripheral blood progenitors. Seminars in . Oncol. 23:9. 53. Ziltenef, H.J., B. Fazekas de St. Groth, K.B. Leslie and J.W. Schrader. 1988. Multiple glycosylated forms of T cell derived interleukin 3(IL-3). /. Biol. Chem. 263:14511. 54. Metcalf, D., and.N.A. Nicola. 1988. Tissue localization and fate in mice of injected multipotential colony-stimulating factor. Proc. Natl. Acad. Sci. USA. 85:3160. 55. Hovgaard, D.J., M. Folke, B.T. Mortensen and N.I. Nissen. 1994. Recombinant human interleukin-3: pharmcokinetics after intravenous and subcutaneous bolus injection and effects on granulocyte kinetics. Br. J. Haematol. 87:700. 56. van Gils, F.C., Y. Westerman, C. van den Bos, H. Burger et al. 1993. Pharmacokinetic basis for optimal hemopoietic effectiveness of homologous IL-3 administered to rhesus monkeys. Leukaemia 7:1602. 57. Ziltener, H.J., I. Clark-Lewis, A.T. Jones and M. Dy.1994. Carbohydrate does not modulate the in vivo effects of injected interleukin-3. Exp. Haematol. 22:1070. 58. Moore, M.A.S. 1988. Interleukin 3: An overview. Lymphokines 15:219. 59. Garland, J.M., A. Aldridge, J. Wagstaff, J.M. Dexter. 1983. Studies in the in vivo production of a lymphokine activity, interleukin 3 IL-3 elaborated by lymphocytes and a myeloid leukaemic cell line in vitro and fate of IL-3 dependent cell lines. Br. J. Cancer 48:247. 60. Ganser, A. 1993. Clinical results with recombinant human interleukin-3. Cancer Invest.l 1:212. 61. Lindemann, A., A. Ganser, F. Hermann, J. Frisch, G. Seipert, G. Schulz, D. Huezler and R. Mertelsmann. 1991. Biological effects of recombinant human interleukin-3 in vivo. J. Clin. Oncol. 9:2120. 62. Morstyn, G., G.J. Lieschke, W. Sheridan, J. Layton and J. Cebon. 1989. Pharmacology of the colony stimulating factors. Trends Pharmacol. Sci. 10:154. 63. Gibbons, J.A., Z.P. Luo, E.R. Hannon, R.A. Braeckman and J.D. Young. 1995. Quantitation of the renal clearance on interleukin2 using nephrectomised and ureter ligated rats. J. Pharmacol. Exp. Ther. 272:119. 64. Pais, R.C, N.B. Ingrim, M.L. Garcia, A: Abdel-Mageed, J, McKolanis et al. 1990. Pharmacokinetics of recombinant interleukin-2 in children with malignancies: a Pediatric Oncology Group study. J, Biol Resp. Modifiers 9:517. 65. Konrad, M.W., G. Hemstreet, E:M. hersh, P.W. Mansell et al. 1990. Pharmacokinetics of recombinant interleukin-2 in humans. Cancer Res. 50:2009. 66. Sculier, J.P., J.J. Brody, N. Donnadieu, S. Nejai et al. 1990. Pharmacokinetics of repeated bolus administration of high doses of r-met-Hu interleukin-2 in advanced cancer patients. Cancer Chemo & Pharm. 26:355. 264 67. Rutenfranz, I. and H. Kirchner. 1988. Pharmacokinetics of recombinant murine interferon-gamma in mice. J. Interferon Res. 8:573. 68. Johns, T.G., J.A. Kerry, B.A. Veitch, I.R. Mackay et al. 1990. Pharmacokinetics, tissue distribution and cell localisation of methionine-labelled recombinant human and . murine alpha interferons in mice. Cancer Res. 50:4718. . 69. Wills, R.J. and K.F. Soike. 1988. Pharmacokinetics of human recombinant interferon alpa I after iv infusion and im injection in African green monkeys. J. Interferon Res. 8:427. . 70. On file, Schering Corporation, Inc. Pointe Claire, Quebec. 71. Hovgaard, D., B.T. Mortensen, S. Schifter and N.I. Nissen. 1992. Clinical pharmacokinetics of a human hemopoietic growth factor, GM-CSF. Eur. J. Clin Invest. . 22:45. 12. Cebon, J.S., R.W. Bury, G.J. Lieschke and G. Morstyn. 1990. The effects of dose and route of administration oh the pharmacokinetics of granulocyte-macrophage colony stimulating factor. Eur. J. Cancer 26:1064. 73. Kuwabara, T., S. Kobayashi and Y. Sugiyama. 1996. Pharmacokinetics and pharmacodynamics of a recombinant human granulocyte colony stimulating factor. Drug Metab. Rev. 28:625. 74. Watari, K., K. Ozawa, S. Takahashi, A. Tojo'etal. 1997. Pharmacokinetic studies of intravenous glycosylated recombinant human granulocyte colony stimulating factor in various hematological disorders: inverse correlation between half-life and bone marrow myeloid cell pool. Int. J. Haem. 66:57. 75. Yoon, W.H., S.J. Park, I.C. Kim and M.G. Lee. 1997. Pharmacokinetics of recombinant human erythropoietin in rabbits and 3/4 nephrectormsed rats. Res. Commun. Mol. Pathol. Pharmacol. 96:227. 76. Bleuel, FL, R. Hoffman, B. Kaufmann, P. Neubert etal: 1996. Kinetics of subcutaneous versus intravenous epoetin-beta in dogs, rats and mice. Pharmacology 52:329. 7-7. Bauer, R.J., J.A. Gibbon, D.P. Bell, Z.P. Luo and J.D. Young. 1994. Nonlinear pharmacokinetics of recombinant human macrophage colony stimulating factor (M-CSF) in rats. J. Pharmcol. & Exp. Ther. 268:152. 78. Katre, N.V., M.J. Knauf and W.J. Laird. 1987. Chemical modification of . recombinant interleukin 2 by polyethylene glycol increases its potency in the murine Meth A sarcoma model. Proc. Natl. Acad. Sci. USA. 84:1487. 79. Meyers, F.J., C. Paradise, S.A. Scudder, G. Goodmanand M. Konrad. 1991. A phase I study including pharmacokinetics or polyethylene glycol conjugated interleukin-2. Clin Pharmacol. & Ther. 49:307. 80. Knauf, M.J., D.P. Bell, P. Hirtzer, Z.P. Luo, J.D. Young and N.V. Katre. 1988. Relationship of effective molecular size to systemic clearance in rats of recombinant interleukin-2 chemically modified with water soluble polymers. J. Biol.Chem. 263:15064. 265 81. Duncan, R and F. Speafico. 1994. Polymer conjugates. Pharmacokinetic considerations for design and development. Clin Pharmacokinetics 27:290. 82. Pettit, D.K., T.P. Bonner, J. Eisenmann, S. Srinvasan, R. Paxton, C. Beers, D. Lynch, B. Miller, J. Yost, K.H. Grabstein and W.R. Gombotz. 1997. Structure-function studies of interleukin 15 using site-specific mutagenesis, polyethylene glycol conjugation and homology modelling. J. Biol. Chem. 272:2312. 83. Curtis, B., D. Williams, H. Broxmeyer, J. Dunn, T. Farrah, E. Jeffery, W. Clevenger, P. deRoos, U. Martin, D. Friend et al. 1991. Enhanced hemopoietic activity of a granulocyte/macrophage colony stimulating factor/interleukin 3-fusion protein. Proc. Nat. Acad. Sci. USA .88:5809. 84. Williams, D, and L. Park. 1991. Hemopoietic effects of a granulocyte/macrophage colony stimulating factor-interleukin3 fusion protein. Cancer 67:2705. 85. Park, L.S., D. Friend, V. Price, D. Anderson, J. Singer, K.S. Prickett and D.L. Urdal. 1989. Heterogeneity in human interleukin-3 receptors. /. Biol. Chem. 264:5420. 86. Miyajima, A,, A.L.F. Mui, T. Ogorochi and K. Sakamaki. 1993. Receptors for granulocyte-macophage colony stimulating factor, interleukin-3 and.interleukin-5. Blood 82:1960. 87. Williams, D.E., L.S. Park and H.E. Broxmeyer. 1991. Hybrid cytokines as hemopoietic growth factors. Int. J. Cell Cloning 9:542. 88. Jakubowski, A., G. Raptis, T Gilewki etal. A phase LT1 trial of PIXY321(PIXY) in patients with metastatic breast cancer receiving doxorubicin and thiotepa. Blood 80: (Suppl 1) 88q. 89 Runowicz, CD., J. Mandeli, J. Speyer et al 1992. Phase IM study of PIXY321 after cyclophosphamide and carboplatin in the treatment of patients with advanced stage ovarian cancer. Blood 80: (Suppl. 1) 88a. 90. Williams, DE., A. Farese and TJ. MacVittie 1993. PLXY321, but not GM-CSF plus IL-3, promotes haemopoietic reconstitution following lethal irradiation. Blood 82: (Suppl. 1) 366a, 91. Vadhan-Raj, S., S. Jeha, H.E. Broxmeyer et al. 1993. Stimulation of haemopoiesis by PIXY321 (GM-CSF/IL-3 fusion protein) in patients with bone marrow failure. Blood 82: (Suppl. I) 366a. 92. Collins, C R.B. Livingston, G. Ellis et al. 1993. Effects of PIXY321 on haematological recovery after high dose cyclophosphamide (CTX), etoposide (VP-16) and cisplatin (CDDP) (CEP) in women with breast carcinoma. Blood 82 : (Suppl. 1) 366a. 93. Vadhan-Raj, S., NE. Papadopoulos, M.A. Burgess etal: 1994. Effects of PIXY321, a granulocyte/macrophage colony stimulating factor-interleukin3 fusion protein, on chemotherapy-induced multilineage depression in patients with sarcoma. 7. Clin. . Oncol. 12:715. 94. Jones, S., P.Khandelwal, K. Mclntyre, R. Mennel, D. Orr et al. 1997. A randomized, double blind, placebo-controlled trail of PIXY321 after moderate-dose FAC in stage II and III breast cancer. Proc. ASCO 16:145a. 266 95. Svenson, M., L.K.Poulsen, A. Fomsgaard and K. Bendtzen. 1989. IgG autoantibodies against interleukin la in sera of normal individuals. Scand. J: Immunol. 29:489. 96. Svenson, M. M.B. Hansen, and K. Bendtzen. 1993. Binding of cytokines to pharmaceutically prepared human immunoglobulin. J. Clin. Invest. 92:2533. 97. Satoh, H., R. Chizzonite, C. Ostrowski, G. Ni-Wu, H. Kim, B. Payer, N. Mae, R. Nadeau and D.J. Liberato. 1994. Characterization of anti-IL-1 alpha autoantibodies in the sera from healthy humans. Immunopharmacol. 27:107. 98. Bendtzen, K., M.B. Hansen, M. Diamant, C. Ross and M. Svenson. 1994. Naturally occurring autoantibodies directed against interleukin 1 alpha, interleukin-6, interleukin 10 and interferon-alpha. J. Interferon Res. 14:157. 99. Jouvenne, P., F. Fossiez, P. Garonne, O. Djossou, J. Banchereau and P. Miossec. 1996. Increased incidence of neutralizing autoantibodies against interleukin-1 alpha (IL-1 alpha) in non-destructive chronic polyarthritis. J. Clin. Immunol. 16:283. 100. Hansen, M.B., M. Svenson, M. Diamant, K. Bendtzen. 1989. Anti-interleukin 6 antibodies in normal human serum. Scand. J. Immunol. 33:777. 101. Hansen, M.B., M. Svenson, M. Diamant, K. Abell and K. Bendtzen. 1995. Interleukin 6 autoantibodies: possible biological and clinical significance. Leukaemia 9:1113. 102. Hansen, M.B., M. Svenson, K. Abell, K. Yasukawa, M. Diamant and K. Bendtzen 1991. Influence of interleukin-6 (IL-6) autoantibodies on EL-6 binding to cellular receptors. Eur. J. Immunol. 25:348. 103. Homann, C., M.B. Hansen, N. Graudal, P. Hasseqvist, M. Svenson, K. Bendtzen, A.C. Thomsen and P. Garred. 1996. Anti-interleukin 6 autoantibodies in plasma are associated with an increased frequency of infections and increased mortality of patients with alcoholic cirrhosis. Scand. J. Immunol. 44:623. 104. Ross, C, M.B. Hansen, T. Schyberg and K. Berg, 1990. Autoantibodies to crude human leucocyte interferon (IFN), native EFN, recombinant human IFN-alpha 2b and human IFN-gamma in healthy blood donors. Clin. Exp. Immunol. 82:57. 105. Turano, A and A. Caruso. 1993. The role of human autoantibodies against giamrna-interferon. J. Antimicrob. Chemother. 32 : (Suppl. A) 99 . 106. Ross, C. M. Svenson, M.B. Hansen, G.L. Veijsgaard, K. Bendtzen. 1994. Specific autoantibodies directed against intereferon-alpha in pharmaceutically prepared human immunoglobulin preparations. J. Inteferon Res .14:159. 107. Prummer, 0., D. Zillikens and F. Pozsolt. 1996. High titer interferon-alpha antibodies in a patient with pemphigus foliaceus. Exp. Derm. 5:213. 108. Prummer, O., D. Bunjes, M. Wiesneth, B. Hertenstein, R. Arnold, F. Porzsolt and H. Heimpel. 1996. Antibodies to interferon-alpha: a novel type of autoantibody occurring after allogeneic bone marrow transplantation. Bone Marrow Transplantation 17:617. 109. Fomsgaad, A., M. Svenson and K. Bendtzen. 1989. Autoantibodies to tumor necrosis factor a in healthy humans and patients with inflammatory diseases and gram negative bacterial infections. Scand. J. Immunol. 30:219. 267 110. Sioud, M., A. Dybwad, L. Jesperson, S. Suleyman, J.B. Natvig and O. Forre. 1994. Characterization of naturally occurring autoantibodies against tumor necrosis factor alpha (TNF-alpha): in vitro function and precise epitope mapping by phage epitope library. . Clin. & Exp. Immunol. 98:520. 111. Tsuchiyama, L., T. Wong, J. Kieran, P. Boyle, D. Penza and G.D. Wetzel. 1995. Comparison of anuVTNF alpha autoantibodies in plasma and from EBV transformed lymphocytes of autoimmune and normal individuals. Human Antibodies & Hybridomas 6:73. 112. Reitamo, S., A. Remitz, J. Varga, M. Ceska, F. Effenberger, S. Jimenez and J. Uitto. 1993. Demonstration of interleukin 8 and autoantibodies to interleukin 8 in the serum of patients with systemic schlerosis and related disorders. Arch. Derm. 129:189. , 113. Amiral, J., M. Marfaing-Koka, M. Wolf, M.C. Alessi, B. Tardy, C. Boyer-Neumann, A.M. Vissac, E. Fressinaud,M. Poncz and D. Meyer. 1996. Presence of autoantibodies to interleukin-8 or neutrophil-activating peptide-2 in patients with heparin-associated thrombocytopenia. Blood 88:410. 114. • Kurdowska, A. E./J. Miller, J.M. Noble, R.P! Baughman, M.A. Matthay, W.G. Brelsford and A.B. Cohen. 1996. Anti-IL-8 autoantibodies in alveolar fluid from patients with the adult respiratory distress syndrome. J. Immunol. 157:2699. 115. Aston, R., W.B. Cowden and GL. Ada. 1989. Antibody mediated enhancement of hormone activity. Mol. Immunol. 26:435. 116. Bendtzen K., M. Svenson, V. Jonsson and E. Hippe. 1990. Autoantibodies to cytokines-Friends or Foes? Immunol. Today 11:167. 117. Suzuki, H. H. Takemura, K. Yoshizaki, Y. Koishihara, Y. Ohsugi, Y. Akiyama, T. Tojo, T. Kishimoto and H. Kashiwagi. 1994. IL-6-anti-IL-6 autoantibody complexes with IL-6 activity in sera from some patients with systemic schlerosis. /. Immunol. 152:935. 118. Couinho, A., M.D. Kazatchkine and S. Avrameas. 1995. Natural autoantibodies. Curr. Opinion in Immunol. 7:812. . 119. Rowley, D.A. E.T. Becken and R.M. Stach. 1995. Autoantibodies produced spontaneously by young Ipr mice carry transforming growth factor beta and suppress cytotoxic T lymphocyte responses. J. Exp. Med. 181:1875. 120. Ho, S.N., R.T. Abraham, S. Gillis and D.J. McKean.1987. Differential bioassay of interleukin 2 and interleukin4. J. Immunol Methods 98: 99. 121. Kelso, A. and T. Owens. 1988. Production of hemopoietic growth factors is differentially regulated in single T lymphocytes activated with an anti-T cell receptor antibody! J. Immunol. 140:1159. 122. Fabry, Z., M.M. Waldscmidt, S.A. Moore and M.N. Hart. 1990! Antigen presentation by brain micro vessel smooth muscle and endothelium. J. Neuroimmunol. 28:63. 123. Gillis, S. and K.A. Smith. 1977. Long term culture of tumor-specific cytotoxic T cells. Nature 268:154. 268 124. Heberrhan, R.B. 1972. Serological analysis of cell surface antigen of tumors induced by murine leukaemia virus. J. Natl. Cancer Inst.48:265. 125. Received from W.Paul N.I.H. 126. Andersson, J. and F. Melchers. 1978. The antibody repertoire of hybrid cell lines obtained by fusion of X63-Ag8 myeloma cells with mitogen-activated B-cell blasts! Curr. Topics Microbiol. Immunol. 81:130. 127. . Kohler, G., S.C. Howe and C. Milstein. 1976. Fusion between immunoglobulin-secreting and non-secreting myeloma cell lines. Eur. J. Immunol. 6:292. 128. Gutman, G.A., N.L. Warner and A.W. Harris. 1981. Immunoglobulin production by murine B lymphoma cells. Clin. Immunol. Immunopathol. 18:230. 129. Chiles, T.C. and T.L; Rothstein.1992. Surface lg receptor-induced nuclear AP-1 dependent gene expression in B lymphocytes. J. Immunol. 149:825. 130. Received from J.W. Schrader BRC. 131. Peavy, D.L. and CW. Pierce. 1975. Cell-mediated immune responses in vitro. II. Simultaneous generation of cytotoxic lymphocyte responses in two sets of alloantigens of limited cross-reactivity. J. Immunol. 115:1515. 132. Clark-Lewis,.I., S.B.H. Kent and J.W. Schrader. 1984. Purification to apparent homogeneity of a factor stimulating the growth of multiple lineages of haemopoietic cells. J. Biol. Chem. 259:7488 . 133. Musch, M.W. and M.I. Siegel. 1985. Antigenic stimulated release of arachidonic acid, lipoxygenase activity and histamine release in a cloned murine mast cell MC9. Biochem Biophys. Res. Commun. 126:517. 134. Leslie, K.B., H.J. Ziltener and J.W. Schrader. 1991. The role of interleukin-1 and granulocyte-macrophage colony stimulating factor in the paracrine stimulation of an in vivo derived murine myeloid leukaemia. Blood 78:1301. 135. Scher, W., Y. Eto, D. Ejima, T. Den and LA. Svet-Moldavsky. 1990rPhorbol ester-reated human acute myeloid leukaemic cells secrete G-CSF, GM-CSF and erythroid diferentiation factor into serum free media in primary culture: Biochim. Biophys Acta. 1055:278 136. Weinstein, Y., J.N. Ihle, S. Lavu and E.P. Reddy. 1986. Truncation of the c-myb gene by a retroviral integration in an interleukin-3 dependent myeloid leukaemia cell line. Proc. Natl. Acad. Sci. USA: 83:5010. 137. Henry, P., P.H. Black, M.N. Oxman and S.M. Weissman. 1966. Stimulation of DNA synthesis in mouse cell line 3T3 by Simian virus 40. Proc. Natl. Acad. Sci. USA. 56:1170. 138. Moyer, CFi, E. Huggins, S. Sarantopoulos, J.C. Lewis, D. Sajuthi, CA. Biron and C.L.Reinisch. 1991. Cloned endothelium derived from autoimmune vascular disease retain structural and functional characteristics of normal endothelial cells. Exp. Cell Res. 199:63. 139. Goding, J.W. 1986. Monoclonal Antibodies: Principles and Practice. Academic Press, London. • 269 140. Merkens, H., A. Tomlinson Jones, J. Wasylnka and H.J. Ziltener. 1994. Production of monoclonal antibodies specific for cell surface determinants. Proc. of the 1992 IDEXX Users Symposium 91-96. 141. Gulley, M.L., L.C. Ogata, J.A. Thorson, M.O. Dailey and J.D. Kemp. 1988. Identification of a murine pan-T cell antigen which is also expressed during the terminal phases of B cell differentiation. J. Immunl. 140:3751 142. Lesley, J., R. Schulte, J. Trotter and R. Hyman. 1988. Qualitative and quantitative heterogeneity in Pgp-1 expression among murine thymocytes. Cellular Immunol. 112:40. 143. Bottomly, K.,-M. Luqman, L. Greenbaum, S. Cardig* J. West, T. Pasqualini and D.B. Murphy. 1989. A monoclonal antibody to murine CD45R distinguishes CD4 T cell populations that produce different cytokines. Eur. J. Immunol. 19:617. 144. Tanaka, T., M. Tsudo, H. Karasuyama, F. Kitamura, T. Kono, M. Hatakeyama, T. Taniguchi, and M. Miyasaka. 1991. A novel monoclonal antibody against murine IL-2 receptor p chain. Characterisation of receptor expression in normal lymphoid cells and EL4. J. Immunol. 147:2222. 145. Springer, T., G. Galfre, D.S. Secher and C. Milstein. 1979- Mac-1: a macrophage differentiation antigen identified by monoclonal antibody. Eur. J. Immunol. 9:301. 146. KoO,.G. and J. Peppard. 1984. Establishment of monoclonal anti-NK-1.1 antibody. Hybridoma 3:301. 147. Coffman, B. 1982. Surface antigen expression and immunoglobulin rearrangment during pre-B cell development. Immunological. Rev. 69:5 148. Ziltener, H.J., I. Clark-Lewis, B. Fazekas de St. Groth, P.C. Orban, L.E. Hood, S.B.H. Kent and J.W. Schrader. 1988. Monoclonal antibodies recognise interleukin 3 and neutralise its bioactivity in vivo. J. Immunol. 140:1182. 149. Kubo, R.T., W. Born, J.W. Kappler, P. Marrack and M. pigeon. 1989. Characterisation of a monoclonal antibody which detects all murine aP receptors. J. Immunol. 142:2736. 150. Goodman, T., and L. Le Francois. 1989. Intraepithelial lymphocytes: anatomical site, not. T cell receptor form,.dictates phenotype and function. J. Exp. Med. 170:1569. 151. Leo, O., M. Foo, D. Sachs, L. Samuelson and J. Bluesone. 1987. Identification of a monoclonal antibody specific for murine T3 polypeptide. Proc. Natl. Acad. Sci. USA. 84:1374. 152. Jones, A.T., B. Fedefsppiel, L.G. Ellies, M.J. Williams, R, Burgener, V. Duronio, CA; Smith, F. Takei and H.J. Ziltener. 1994. Characterisation of the activation-associated isoform of CD43 on murine T-lymphocytes. J. Immunol.153:3426. 153. Dialynas, D.P., Z.S. Quan, K.A. Wall, A. Pierres, J. Quintans, M:R.'Loken, M. Pierres and F.W. Fitch. 1983. Characterisation of the murine T cell surface molecule designated L3T4, identified by monoclonal antibody GK1.5: similarity to the human Leu3/T4 molecule. J. Immunol. 131:2445. 154. Ledbetter, J., R. Rouse, S. Micklem and L. Herzenberg. 1980. T cell subsets defined by expression of Lyt-1,2,3 and Thy-1 antigens. J. Exp. Med. 15:280. 270 155. Mechanisms of Cell-mediated Cytotoxicity. 1982. In Proceedings of the First International Cytolysis Workshop. (Eds. W.R. Clark and P. Golstein). p547. 156. Hoffman, R.A., P.C. Kung, W.P. Hansen and G. Goldstein. 1980. Simple and rapid measurement of human T-lymphocytes and their subclasses in peripheral blood. Proc. Natl. Acad. Sci. USA. 77:4917. 157. Muroi, K, T. Suda, H. Nojiri, H. Ema, Y. Amemiya, Y. Miura, H. Nakauchi, A. Singhal and S. Hakamori. 1992. Reactivity profiles of leukaemic myeloblasts with monoclonal antibodies directed to sialosyl-Le(x) and other lacto-series type 2 chain antigens: absence of reactivity with normal hematopoietic progenitor cells. Blood 79:713. 158. Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage TA. Nature 227:680. 159. Towbin, H., T. Staehelin and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA .81:573. 160. Wilson , A. and G. Rider! 1992. Evidence thatleukosialin CD43 is intensely sulfated in the murine T lymphoma line RDM-4. J.Immunol. 148:1777. 161. Denizot, F., A. Wilson, F. Battye, G. Berke and K. Shortman. 1986. Clonal expansion of T cells: A cytotoxic T-cell response in vivo that involves precursor cell proliferation. Proc. Natl. Acad. Sci. USA. 83: 6089. 162. Duronio, V., B.E. Huber and S. Jacobs. 1990! Partial down regulation of protein kinase C reverses the growth inhibitory effect of phorbol esters on hep G2 cells. J. Cell Physiolo. 145:381. 163. Harder, K.W., P. Owen, L.K.H. Wong, R. Aebersold, I. Clark-Lewis and F.Jirik. 1994. Characterisation and kinetic analysis of the intracellular domain of human protein tyrosine phosphatase [3 (HPTPP) using synthetic phosphopeptides. Biochem J. 298: 395. 164. Shaw, S., G.E. Luce, R. Quinones, R.E. Gress, T.A. Springer and M.E. Sanders. 1986. Two antigen-independent adhesion pathways used by human cytotoxic T-cell clones. Nature 323:262, 165. Harder, R., H. Uhlig, A. Kashan, B. Schutt, A. Duijvestijn, E.C. Butcher, H.G. Thiele and A. Hamann. 1991. Dissection of murine lymphocyte-endothelial cell interaction mechanisms by SV40-transformed mouse endothelial cell lines: novel mechanisms mediating basal binding and alpha 4-integrin cytokine induced adhesion. Exp. Cell Res. 197:259. ' 166. Stockl, J., O. Majdic, P.Kohl, W.F. Pickl, J.E. Menzel and W. Knapp. 1996. Leukosialin (CD43)- major histocompatibility class I molecule interaction involved in s pontaneous T cell conjugate formation. J. Exp. Med. 184: 1769. 167. Grabstein, K. 1980. In Selected Methods in Immunology. Freeman Press. San Franscisco. . 168. Buhrer, C., C. Berlin, D. Jablonski-Westrich, B. Holzmann, H.G. Thiele and A. Hamann 1992. Lymphocyte activation and regulation of three adhesion molecules with supposed function in homing: LECAM-1 (MEL-14 antigen), LPAM-1/2 (alpha 4-integrin) and CD44 (Pgp-1).Scand. J. Immunol. 35:107. 271 169. Pardi, R., J.R. Bender, C. Dettoric, E. Giannazza and E.G. Engelman. 1989. Heterogeneous distribution and transmembrane signalling properties of lymphocyte associated antigen (LFA-1 j in human lymphocyte subsets./. Immunol. 143:3157. . 170. Mackay, C.R., W.L. Marston and L. Dudler. 1990. Naive and memory T cells show distinct pathways of lymphocyte recirculation. J. Exp. Med. 171:801. 111.. Budd, R.C., J-C. Cerottini and H. Robson MacDonald. 1986. Cultured Lyt-2- L3T4- T lymphocytes from normal thymus or Ipr mice express a broad spectrum of cytolytic activity. J. Immunol. 137:3734. 172. Croft, M. 1994. Activation of naive, memory and effector T cells. Current Opinions in Immunol. 6:431. i 173. Petrie, H. T., A. Strasser, A.W. Harris, P. Hugo and K. Shortman. 1993. CD4+8r and CD4-8+mature thymocyte require different post-selection processing for final development. J. Immunol. 151:1273. 174. Lynch, F., and R. Ceredig. 1989. Mouse strain variation in Ly-24 (Pgp-1) expression by peripheral T cells and thymocytes: implications for T cell differentiation.£wr J. Immunol. 19:223. 175. Weiss, A. and J.B. Imboden. 1987. Cell surface molecules and early events in T lymphocyte activation. Adv. in Immunol. 41:1. 176. Cyster, J., C. Somoza, N. Killeen and A.F. Williams. 1990. Protein sequence and gene structure for mouse, leukosialin (CD43), a T lymphocyte mucin without introns in the coding sequence. Eur. J. Immunol. 20:875. 177. Baecher, CM., K.S. Dorfman, M.G. Mattei and J.G. Frelinger. 1990. cDNA cloning and localization of the mouse leukosialin gene (Ly-48) to chromosome 1. Immunogenetics 31:307. 178. Baecher, CM., A.J. Infante, K.L. Semcheski and J.G- Frelinger. 1988. Identification and characterisation of a mouse cell surface antigen with alternative molecular forms. Immunogenetics 28:295. . 179. Trowbridge, I.S., J. Lesley, R. Shutte, R. Hyman, and J. Trotter. 1982. Biochemical characterization and cellular distribution of a polymorphic, murine cell-surface glycoprotein expressed on lymphoid tissues. Immunogenetics 15:299. 180. Hughes, E.N., A. Columbatti and J.T. August. 1983. Murine cell surface glycoproteins. Purification of the polymorphic Pgp-1 antigen and analysis of its expression on macrophages and other myeloid cells. /. Biol Chem. 258:1014. 181. Zhou, D.F.H., J.F. Ding, L.J. Picker, R.F. Bargatze, E.C Butcher and D.V. Goeddel. 1989. Molecular cloning and expression of Pgp-1, the mouse homologue of the human H-CAM (Hermes) lymphocyte homing receptor. J. Immunol. 143:3390. 182. Jalkanen, S., M. Jalkanen, R. Bargatze, M. Tmmi and E.C. Butcher. 1988. Biochemical properties of glycoproteins involved in lymphocyte recognition of high endothelial venules in man. J. Immunol, 141:1615. 272 183. Goldstein, L.A., D.F.H. Zhou, L.J. Picker, C.N. Minty, R,F. Bargatze, J.F. Ding and E.G. Butcher. 1989. A human lymphocyte homing receptor, the hermes antigen, is related to cartilage proteoglycan core and link proteins. Cell. 56:11063. 184. Dougherty, G.J., P.M. Landorp, D.L. Cooper and R.K.Humphries. 1991. Molecular cloning of CD44R1 and CD44R2, two novel isoforms of CD44 lymphocyte "homing" receptor expressed on haemopoietic cells.7 Exp Med 174:1. 185. Stamenkovic, I., M. Amiot, J.M. Pesando and B. Seed. 1989. A lymphocyte molecule implicated in lymph node homing is a member of the cartilage link protein family. Cell 56:1057. 186. Arrufo, A, I. Stanenkovic, M. Melnick, CB. Underhill and B. Seed. 1990. CD44 is the . principal cell surface receptor for hyaluronate. Cell 61:1303. 187. Haynes, B.F., M.J. Telen, L.P. Hale and S.M. Denning. 1989. CD44: a molecule involved in leukocyte adherence and T cell activation. Immunol. Today. 10:423. 188. Jalkaren, ST., R.F. Bargatze, L.R. Herron and E.C. Butcher. 1986. A lymphoid cell surface glycoprotein involved in endothelial cell recognition and lymphocyte homing in man. Eur. J. Immunol. 16:1195. 189. Trowbridge, IS., M.L. Thomas. 1994. CD45: an emerging role as a protein tyrosine phosphatase required for lymphocyte activation and development. Annu. Rev. Immunol. 12:85. 190. Steuli, M.,C. Monmoto, M. Schreiber, S.F. Schlossman and H. Saito. 1988. Characterisation of CD45 and CD45R monoclonal antibodies using transfected mouse cell lines that expresss individual human leukocyte common antigens. J. Immunol, 141:3910. 191. Pulido, R. and F. Sanchez-Madrid. 1990. Glycosylation of CD45: carbohydrate composition and its role in the acquisition of CD45RO and CD45RB T cell maturation-relatedantigen specificities. Eur. J. Immunol. 20:2667. 192. Steuli, M., T. Matsuyama, C. Monmoto, S.F. Schlossman and H, Saito. 1987. Identification of the sequence required for expression of the 2H4 epitope on the human leukocyte common antigens./. Exp. Med. 166:1567. 193. Aruffo, A, S.B. Kanner; D. Sgroi, J.A. Leabetter and 1. Stamenkovic. 1992. CD22-mediated stimulation of T cells regulates T-cell receptor/CD3-induced signalling.Proc. Natl. Acad. Sci. USA. 89:10242. 194. Stamenkovic, I., D. Sgroi, A. Aruffo, M.S. Sy and T.Anderson. 1991. The B lymphocyte adhesion molecule CD22 interacts with leucocyte common antigen CD45RO on T cells and alpha 2-6 sialyltransferase, CD75, on B cells. Cell 66:1133. 195. Ong, CJ., D. Chui, H.S. Teh and J.D. Marth. 1994. Thymic CD45 tyrosine phosphatase regulates apoptosis and MHC restricted negative selection. /. Immunol. 152:3793. 196. Koretsky, G.A. 1993. Role of the CD45 tyrosine phosphatase in.signal transduction in the immune system.F.AS£VJ /. 7:420. 197. Berger, S.A., T.W. Mak, and CJ. Paige. 1994. Leukocyte common antigen is required for immunoglobulin E-mediated degranulation of mast cells. /. Exp. Med. 180:471. 273 198. Wallace, V.A., W.P. Fung-Leung, E. Timmuns, D. Gray, K. Kishiwara, D.Y. Loh, J. Penninger and T.W. Mak. 1992. CD45RA and CD45RB high expression induced by thymic selection events. J. Exp. Med. 176:1657. 199. Lynch, F. and E.M. Shevach. 1992. Activation requirements of newborn thymic gamma delta T cells. J. Exp. Med. 169: 1271. 200. Aparicio, P., J.M. Alonso, M.L. Toribio, J.C. Guttierrez, L. Pezzi and A.C. Martinez. 1989. Differential growth requirements and effector functions of alpha/beta and gamma/delta human T cells. Immunol Rev. 111:5. 201. Bucy, R.P., C.L.H. Chen, and M.D. Cooper. 1989.Tissue localisation and CD8 accessory molecule expression of T gamma delta cells in humans. J. Immunol. 142.2045. 202. Spits, H., X. Paliard, Y. Vandekerckhove, P. van Vlasselaer and J.E. de Vries. 199L Functional and phenotypic diferences between CD4+ and GD4- cell receptor-y8 clones from peripheral blood. /. Immunol. 147:1180. 203. Groh, V.S., S. Porcelli, M. Fabbi, L.L. Lanier, L.J. Picker, T. Anderson, R.A. Warnke, A.K. Bhan, J.L. Strominger and M.B. Brenner. 1989. Humari lymphocytes bearing T cell gamma/delta are phenotypically diverse and evenly distributed throughout hte ; lymphoid system. J. Exp Med 169:1277. 204. Crohn, R.Q., T.F. Gajewski, S.O.Sharow, F.W. Fitch, L:A. Matis and J.A. Bluestone. 1989. Phenotypic and functional analysis of murine CD3+,CD4,CD8-TCR-gamma delta-expressing peripheral T'cells. J. Immunol. 142:8754.. 205. Spaner, D.K. Migita, A. Ochi, J. Shannon, R.G. Millre, P. Pereira, S. Tonegawa and R.A. Phillips. 1993. Gamma delta T cells differentiate into a functional but nonproliferative state during a normal immune response.Proc. Nad. Acad. Sci. 90: 8204. 206. Kabelitz, D., A. Bender, S. Schondelmaier, M.L. Da Silva Lobo^  and O. Janssen. 1990. Human cytotoxic lymphocytes. V Frequency and specificity of gamma delta-t- cytotoxic .precursors activated by allogeneic or autologous stimulator cells. J. Immunol .145:2827. 207. Ellies, L, A. Tomlinson Jones, M.Williams and H.J. Ziltener. 1994. Differential . regulation of CD43 glycoforms on CD4+ and CD8+ T lymphocytes in graft-versus-host disease. Glycobiology 4:885. 208. Cohen, P.L & R.A. Eisenberg. 1991. Lpr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Ann Rev Immunol. 9: 243. 209. Scott, D.E., W.J. Kisch and A.D. Steinberg. 1993. Studies of T cell deletion and T cell anergy following in vivo adminstration of SEB to normal and lupus-prone mice. J. Immunol. 150:664. 210. Asano, T., S. Tomooka, B.A..Serushago, K. Himeno and K. Nomoto. 1988. A new T cell subset expressing B220 and CD4 in lpr mice: defects in the response to mitogens and in the production of IL-2. Clin. Exp. Immunol. 74:36. 211. Asano, T., Y.K. Matsumoto, G. Matuzaki andf K. Nomoto. 1990. Subpopulations of . CD4+ cells in lpr/lpr mice: differences in expression of T cell receptor/CD3 complex and proliferative responses. Clin. Exp. Immunol. 81.90. 212. Smith, K.A. 1989. The interleukin 2 receptor. Annu Rev. Cell. Biol. 5:397. . 274 213. Pattengale, P.K. and C.H. Frith. 1983. Immunomorphological classification of spontaneous lymphoid neoplasms occurring in female BALB/c mice. J.N.C.I. 70:169. 214. Stein, H-, G. Scwarting, G. Niedobitek, & F. Dallenbach. 1989. In Leucocyte Typing IV: White Cell Differentiation Antigens. (Eds. W.Knapp, B. Dorken, W.R. Gilks, E.P. Rieber, R.E. Schmidt, H. Stein and A.E. G. Kr. von dem Borne) Oxford University Press. New York. p387. 215. Ortega, R.G., RJ. Robb, E.M. Shevach, and T.R. Malek. 1984. The murine IL-2 receptor. I. Monoclonal antibodies that define distinct functional epitopes on activated T cells and react with activated B cells./. Immunol. 133:1670. 216. Weiss, A. and J.B. Imboden. 1987. Cell surface molecules and early events in human T-lymphocyte activation. Advances in Immunol. 41:1. 217. Hagiwara, H., H.S. Huang, N. Arai, L.A. Herzenberg, K. Arai, and A. Zlotnik 1987. Interleukin 1 modulates messenger RNA levels of lymphokines and of other molecules associated with T cell activation in the T cell lymphoma LBRM33-1A5. J Immunol. 138:2514. 218. Lichtman, A.H., E.A. Kurt-Jones, and A.K. Abbas. 1987. B-cell stimulatory factor 1 and not interleukin 2 is the autocrine growth factor for some helper T lymphocytes. Proc. Natl. Acad. Sci.USA. 84:824. 219. Le Gros, G., S.Z. Ben-Sasson, R. Seder, F.D. Finkelman, and W.E. Paul. 1990. Generation of interleukin 4(IL-4)-producing cells in vivo and vitro: EL-2 and EL-4 are required for in vitro generation of IL-4-producing cells./. Exp. Med. 172:921. 220. Tanaka, T., S.Z. Ben-Sasson and W.E. Paul. 1991. IL-4 increases IL-2 production by T cells in response to accessory cell -independent stimuli./. Immunol. 146:3831. 221. Isakov, N:, R.C Bleakley, J.Shaw, and A. Altman. 1985. Teleocidin and phorbol ester tumour promoters exert similar mitogenic effects on human lymphocytes. Biophys. Biochem. Res. Commun. 130:724. 222. Lederman, S., M.J. Yellin, A. Krichevsky, J. Belko, J.J. Lee, and L.Chess.1992. Identification of a novel surface protein on activated CD4+ T cells that induces contact dependent B cell differentiation. /. Exp. Med. 175:1091. 223. Wilson, A., T. Ewing, T. Owens, R. Scollay and K. Shortman.1988. T cell antigen receptor expression by subsets of Ly-2-L3T4- (CD4-8) thymocytes: J.Immunol 140:1470. 224. Fowlkes, B.J., A.M. Kruisbeek, H. Ton-That, M.A. Weston, J.E. Coligan, R.H. Schwartz and D.M Pardoll. 1987. A novel population of T-cell receptor alpha beta -bearing thymocytes which predominantly express a single V beta gene family. Nature 329:251. 225. Gause, W.C, T. Takashi, J.D. Mountz, F.D. Finkelman and A.D. Steinberg. 1988. Activation of CD4-CD8- thymocytes with IL-4 vs EL-l and IL-2. /. Immunol. 141:2240. 'i 226. Giese, T. and W.F. Davidson. 1992. Evidence for early onset polyclonal activation of T cell subsets in mice homozygous for lpr. J. Immunol. 149:3097. 275 227. Mori, K., S. Kobayashi, M. Inobe, W.Y. Jia, M. Tamakoshi, T. Miyazaki, and T. Cede. 1994. In vivo cytokine gene expression in various T cell subsets of the autoimmune MRL/Mp-/pr/7pr mouse. Autoimmunity 17:49. 228. Santoro, T.J., J.P. Portanova, and B.L: Kotzin. 1988. The contribution of L3T4+ T-cells . to lymphoproliferation and autoantibody production in MRL lpr/lpr mice. J. Exp. Med. 167:1713. ':• 229. Jabs, D.A., C. Lynne Burek, Q. Hu, R.C. Kuppers, B.Lee and R.A. Prendergast. 1992. Anti-CD4 monoclonal antibody therapy suppresses autoimmune disease in MRL lpr/lpr mice. Cellular Immunol. 141:496. 230. Giese, T., & W.F. Davidson. 1994. Chronic treatment of C3H-lpr/lpr and C3H-gld/gld . mice with anti-CD8 monoclonal antibody prevents the accumulation of double negative T cells but not autoantibody production. J. Immunol. 152:2000. 231: Laouar, Y & S. Ezine. 1994. In vivo CD4+ lymph node T cells from Ipr mice generate CD4-8-B220+TCR-beta low cells J. Immunol. 153:3948 232. Tutt Landolfi, M.M., N. Van Houten, J.Q. Russell, R. Scollay, J.R. Parnes and R.C. Budd. 1993. CD2-CD4-CD8- lymph node T lymphocytes in MRL lpr/lpr mice are derived from a CD2+CD4+CD8+ thymic precursor. J. Immunol. 151:1086. 233. Prud'homme, G.J., D.C. Bocarro and E.C.H. Luke. 1991. Clonal deletion and autoreactivity in extrathymic CD4-CD8- (double neagtive) T cell receptor-alpha/beta T cells./. Immunol. 147:3314. 234. Watanabe-Fukunaga, R., C.I. Brannan, N. Itoh, S. Yonehara, N.G. Copeland, N.A. Jenkins and S. Nagata. 1992. Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 356: 314. : 235. Lynch, D.H., M.L. Watson, M.R. Alderson, P.R. Baum, R.E. Miller, T. Tough, M. Gibson, T. Davis-Smith, CA. Smith, K. Hunter. 1994. The mouse Fas-ligand gene is mutated in gld mice and is part of a TNF family gene cluster. Immunity 1:131. 236. Takahasi, T., M. Tanaka, C. Brannan, N: Jenkins, N.G. Copeland, T. Suda and S. Nagata. 1994. Generalized lymphoproliferative disease in mice, caused by a point mutation in the Fas ligand. Cell 76: 969. 237. Zhou, T., H. Bluethmann, J. Eldridge, M. Brockhaus, K. Berry and J.D. Mountz. 1991. Abnormal thymocyte development and production of autoreactive T cells in T cell receptor transgenic autoimmune mice. /. Immunol. 147:466. 238. Zhou, T., Bluethmann, H., Zhang, J., Edwards, C.K., & J.D. Mountz. 1992. Defective maintenance of T cell tolerance to a superantigen in MRL-lpr/lpr mice./. Exp. Med. 176:1063. 239. Russel, J.H., Rush, B., Weaver, C and R. Wang. 1993. Mature T cells of autoimmune lpr/lpr mice have a defect in antigen-mediated suicide. Proc. Natl. Acad. Sci. USA . 90:4409. 240. Steinberg, A.D., J.B. Roths, E.D. Murphy, R.T. Steinberg and E.S. Raveche. 1980. Effects of thymectomy or androgen administration upon the autoimmune disease of MRL-lpr/lpr mice. /. Immunol. 125:871. 276 241. Hang, L., A.N. Theofilopoulos, R.S. Balderas, S.J. Francis and F.J. Dixon. 1984. The effect of thymectomy on lupus-prone mice . J. Immunol. 132:1809. 242. Seth, A., R.M. Pyle, M. Nagarkatti and P. S. Nagarkatti. 1988. Expresion of the Jl Id marker on peripheral T lymphocytes of MRL-lpr/lpr mice/. Immunol. 141:1120. 243. Waterfield, J.D. & M. Fairhurst. 1991. Expression of an EL-4 tumour-associated determinant on subpopulations of murine T cells in normal and lympho-proliferative autoimmune mice. Immunol .72:226. 244. Sobel, E., W.M. Yokoyama, E.M. Shevach, R.A. Eisenberg, R.A. and P.L. Cohen. 1993. Aberrant expression of very early activation antigen on MRL-pr/lpr lymphocytes. /. Immunol. 150:673. 245.. Dumont, F.J., R.C. Habbbersett and E.A. Nichols. 1984. A new lymphocyte surface antigen defined by a monoclonal antibody (9F3) to the T cell population expanding in MRL/Mp-lpr/lpr mice. /. Immunol, 133:809. 246. Tanaka, T., Y. Nagasaka, F. Kitamura, K. Kuida, H. Suwa and M. Miyasaka. 1993. The role of the interleukin 2 (IL-2)/ IL-2 receptor pathway in MRL/lpr lymphadenopathy: the expanded CD4-3- T cell subset completely lacks functional IL-2 receptors. Eur. J. Immunol. 23:1378. 247. Nishimura, T., H. Yagi, Y. Uchiyama and H.Y. Hashimoto. 1986. Generation of lymphokine-activated killer (LAK) cells from tumor-infiltrating lymphocytes. Cellular Immunol. 100:149. 248. Rosenberg, S.A., P.J. Speiss and R. Lafreniere. 1986. A new approach to the adoptive immunotherapy of cancer with tumor-infiltrating lymphocytes. Science 223:1318. 249. Ghosh, P., K.L. Komschlies, M. Cippitelli, D. L. Longo, J. Subleski, J. Ye, A. Sica, H.A. Young, R.H. Wiltrout and A.C. Ochoa. 1995. Gradual loss of T-helper 1 populations in the spleen of mice during progressive tumour growth. J.N.C.I. 87:1478. 250. Jung, T.M., W.M. Gallatin, I.L. Weissman and M.O. Dailey. 1988. Down-regulation of homing receptors after T cell activation. /. Immunol. 141:4110. 251. Kishimoto, T.K., MA. Jutila and E.C. Butcher. 1990. Identification of a human peripheral lymph node homing receptor : a rapidly down-regulated adhesion molecule. Proc. Natl. Acad. Sci. USA. 87:2244 252. Hamann, A., D. Jablonski-Westrich, K.U. Scholz, A. Duijvestijn, E.C. Butcher and H.G. Thiele. 1988. Regulation of lymphocyte homing. I. Alterations in homing receptor expression and organ specific high endothelial venule binding of lymphocytes upon activation. /. Immunol. 140:737. 253. Jenkins, M.K., P.S. Taylor, S.D. Norton and K.B. Urdahl. 1991. CD28 delivers a costimulatory signal involved in antigen specific IL-2 production by human T cells. /. Immunol. 147:2461. 254. Damle, N.K., P.S. Linsley and J.A. Ledbetter.1991, Direct helper T cell induced B cell differentiation involves the interaction between T cell antigen CD28 and B cell activation antigen B7. Eur. J. Immunol. 21:1277. 277. 255. Akiyama, T, C. Sudo, H. Ogawara, K. Toyoshima and T. Yamamoto.. 1986. The product of the human c-erbB-2 gene: a 185-kilodalton glycoprotein with tyrosine kinase activity. Science 232:1644. 256. Tonks, N.K., H. Charbonneau, CD, Diltz, E.H. Fischer and K.A. Walsh. 1988. Demonstration that the leucocyte common antigen CD45 is a protein tyrosine phosphatase; Biochemistry 27:8694. -257. Yonemura, S., A. Nagafuchi, N. Sato and S. Tsukita. 1993. Concentration of an integral membrane protein, CD43 (leukosialin, sialophorin) in the cleavage furrow through the interaction of its cytoplasmic domain with actin abased cytoskeietons. J. Cell Biol. 120:437. 258. Sanchez-Mateos, P., M; Campanero, M. del Pozo and F, Sanchez-Madrid. 1995. Regulatory role of CD43 leukosialin on integrin-mediated T-cell adhesion to.endothelial and extracellular matrix ligands and its polar redistribution to a cellular uropod. Blood 86:2228. 259. Dougherty, G.J., S. Murdoch and N. Hogg: 1988. The function of human intercellular adhesion molecule-1 (ICAM-1) in the generation of an immune response. Eur. J. Immunol. 18:35. 260. Campanero, M.R., P. Sanchez-Mateos, M.A. del Pozo and F. Sanchez-Madrid. 1994. ICAM-3 regulates lymphocyte morphology and integrin-mediated T cell interaction with endothelial cell and extracellular matrix ligands. J. Cell Biol. 127 (3):867. 261. Bevilacqua, M. P., J.S. Pober, M.E. Wheeler, R.S. Cotran and M.A. Gimbrone Jr. 1985. Interleukin-1 acts on cultured human vascular endothelium to increase the adhesiOnof polmorphonuclear leukocytes, monocytes and related leukocytelines. J. Clin Invest. 76:2003. .„ . . 262. Bochner, B.S., F.W. Luscinskas, M.A. Gimbrone Jr, W. Newman, S.A. Sterbinsky, CP. Derse-Anthony, D. Klunk and R.P.Schleimer. 1991. Adhesion of human basophils, eosinophils and neutrophils to interleukin-1 activated human vascular endothelial cells: contributions of endothelial cell adhesion molecules. J. Exp. Med. 173:1553. 263. Osborn, L., C. Hession, R. Tizard, C. Vassallo, S. Luhowskyj, G. Chi-Rosso and R. Lobb. 1989. Direct expression cloning of vascular cell adhesion molecule 1, a cytokine-induced endothelial protein that binds to lymphocytes. Cell 59:1203. 264. Thornhill, M.H., U. Kyan-Aung and DO. Haskard. 1990. IL-4 increases human endothelial cell adhesiveness for T cells but not for neutrophils. J. Immunol. 144:3060. 265. Rodrigues, M, R.S. Nussenzweig, P. Romero and F. Zavala. 1992. The in vivo cytotoxic activity of CD8+ T cell clones correlates with their levels of expression of adhesion molecules. J. Exp. Med. 175:895. 266. Mobley, J.L. and M.O. Dailey. 1992. Regulation of adhesion molecule expression by - CD8 T cells in vivo. 1. Differential regulation of gp90MEL-14 (LECAM-1), Pgp-1, LFA-1 and VLA-4 alpha during the differentiation of cytotoxic lymphocytes induced by allografts. J. Immunol. 148:2348. 267. Oppenheimer-Marks, N.„ L.S. Davis and P.E. Lipsky. 1990. Human T lymphocyte adhesion to endothelial cells and trans-endothelial migration. Alteration of receptor use . 278 relates to the activation status of both the T cell and the endothelial cell. J. Immunol. 145:140. 268. Picker, L.J., T.K. Kishimoto, C.W. Smith, R.A. Warnock and E.C. Bitcher. 1991. ELAM-1 is an adhesion molecule for skin-homing T cells. Nature 349:796. 269. Mackay, C. 1993. Homing of naive, memory and effector lymphocytes. Curr. Opinion. Immunol. 5:423. 270. Picker, L.J. 1994. Control of lymphocyte homing. Curr. Opinion Immunol.6:394. 271. Shaw, S. 1993. Regulation of T-cell adhesion to endothelium. Seminars in Hematol. 30:56. . . , 272. Butcher, E. and L.J. Picker. 1996. Lymphocyte homing and homeostasis. Science 272:60. . 273. Picker, L.J., L.W. Terstappen, L.S. Rott, P.R. Streeeter, H. Stein and E.C. Butcher. 1990. Differential expression of homing associated adhesion molecules by T cell subsets in man. J. Immunol .145:3247. 214. Morito, T., T: Nakamura, H. Nakai, K. Tanimoto, Y. Horiuchi and T Juji. 1978. Close association of human mixed lymphocyte culture antigen, la-like antigen and Fc receptor. Int. Arch. Allergy Applied Immunol. 57:521. 215. Parham, P. 1983. On the fragmentation,of monoclonal IgGl, Ig2a and Ig2b from BALB/c mice. J. Immunol. 1.31:2895. 216. Shimuzu, Y. W. Newman, Y.Tanaka and S. Shaw. 1992. Lymphocyte interactions with endothelial cells. Immunol. Today 13:106. 211. Kuijpers, T.W., B.C. Hakkert, J.A. van Mourik and D. Roos. 1990. Distinct adhesive properties of granulocytes and monocytes to endothelial cells under static and stirred conditions. J. Immunol. 145:2588. 278. Lawrence, M.B. and T.A. Springer. 1991. Leukocytes roll on a selectin at physiological flow rates: distinction from and prerequisite for adhesion through integrins. Cell 65:859. 219. Kranz, D.M., D.H. Shermann, M.V. SitkOvsky, M.S. Pasternack and H.N. Eisen. 1984. Immunoprecipitation of cell surface structures of cloned cytotoxic T lymphocytes by clone specific,antisera. Proc. Natl. Acad. Sci. U.S.A. 81:573. 280. Remold-O'Donnell, E. and F. Rosen. 1990. Proteolytic fragmentation of sialophorin (CD43): Localisation of the activation-inducing site and examination of the role of sialic acid. J. Immunol. 145:3372. . 281. Baecher-Allen, CM., J.D. Kemp, KS. Dorfman, R.K. Barth and J.G. Frelinger. 1993. Differential epitope expression of Ly-48 (mouse leukosialin). Immunogenetics 37:183. 282. Fox, R.L, M. Hueniken, S. Fong, S. Behar, I. Royston, S.K. Singhal and L. Thompson. 1983. A novel cell surface antigen (T305) found in increased frequency on acute leukemia cells and in autoimmune disease states. J. Immunol. 131:762. 283. Piller, F., F. Le Deist, K.I. Weinberg, R. Parkman and M. Fukuda. 1991. Altered O-glycan synthesis in lymphocytes from patients with the Wiskott-Aldrich syndrome. J. Exp..Med. 173:1501, 219 284. Sportsman, J.R., M.M. Park, D.A. Cheresh, M. Fukuda, J.H. Elder and R.I. Fox. 1985. Characterisation of a membrane surface glycoprotein associated with T cell activation. J. Immunol. 135:158. 285. Corfield A.P., R.W. Veh, M. Wember, J.C. Michalski and R. Schauer. 1981, The release of N-acetyl- and N-glycolloyl-neuraminic acid from soluble complex carbohydrates and erythrocytes by bacterial, viral and mammalian sialidases. Biochem J. 197:293. 286. Piller, F., V. Piller, R.I. Fox and M. Fukuda. 1988. Human T lymphocyte activation is associated with changes in O-glycan biosynthesis. 7 Biol. Chem. 263:15146. 287. Barran P., W. Fellinger, C.E. Warren, J.W. Dennis and HJ: Ziltener. 1997. Modification of CD43 and other lymphocyte O-glycoproteins by core 2 N-acetylglucosminyltransferase. G/yco^ /o/ogy 7.729. 288. Maemura, K. and M Fukuda. 1992. Poly-N-acetyllactosaminyl O-glycans attached to leukosialin. 7 Biol. Chem. 267: 24379. V: 289. Giodanengo, V., M. Limouse, J-F. Peyron and J.-C. Lefebre. 1995. Lymphocytic CD43 and CD45 bear sulfate residues potentially implicated in cell to cell interactions. Eur. 7 Immunol. 25:274. 290. Chatila, T.A. and R.S. Geha. 1988. Phosphorylation of T cell membrane proteins by activation of protein kinase C. J .Immunol. 140:4308. 291. Piller, V., F. Piller and M. Fukuda. 1989. Phosphorylation of the major leukocyte surface sialoglycoprotein, leukosialin is increased by phorbol 12-myristatel3-acetate. 7 Biol. Chem. 264:18824. 292. Axelsson, B and P. Perlmann. 1989. Persistant superphosphorylation of leukosialin (CD43) in activated T cells and in tumour cell lines.Scand. J. Immunol: 5:539. 293. Wong, R.C.K., E. Remold-O'Donnell, D. Vercelli, J. Sancho, C. Terhorst, F. Rosen, R. Geha and T. Chatila. 1990. Signal transduction via leukocyte antigen CD43 (sialophorin): Feedback regulation by protein kinase C. 7 Immunol. 144:1455. 294. Carlsson, S.R. and M. Fukuda. 1986. Isolation and characterization of leukosialin, a major sialoglycoprotein on human leukocytes. J. Biol. Chem. 261:1277. 295. Cyster, J.G., D. Fowell and A.N. Barclay J994. Antigenic determinants encoded by alternatively spliced exons of CD45 are detremined by the polypeptide but influenced; by glycosylation. Int. Immunol. 6:1875. 296. Carlsson, S.R., H. Sasaki and M. Fukuda. 1986. Structural variations of O-linked oligosaccharides present in leukosialin isolated from erythroid, myeloid and T-lymphoid cell lines. 7 Biol. Chem. 261:12787. 297. Saitoh, O., F. Piller, R.I. Fox and M. Fukuda. 1991. T-lymphocytic leukemia expresses complex, branched O-linked oligosaccharides on a major sialoglycoprotein, leukosialin. Blood 77:1491. 298. Fukuda, M. 1991. Leukosialin, a major O-glycan containing sialoglycoprotein defining leukocyte differentiation and malignancy. Glycobiology. 1:347. 280 299. Fukuda, M. 1993. Cell Surface carbohydrates in cell differentiation and malignancy. In Cell Surface Carbohydrate and Cell Development. C.R.C. Press, Inc. Florida. 300. Aruffo, A., W. Kolanus, G. Walz, P. Fredman and B. Seed. 1991. CD62/P-selectin recognition of myeloid and tumor cell sulfatides. Cell 67:35. 301. Suzuki, Y., T. Tamatani, T. Watanabe, T. Suzuki, T. Nakao, K. Murase, M. Kiso, A. Hasegawa, K. Tadano-Aritomi, I. Ishizuka and M. Miyasaka. 1993. Sulfated glycolipids are ligands for a lymphocyte homing receptor, L-selectin (LECAM-) binding epitope in sulfated sugar chain. Biochem. Biophys. Res. Commun. 190:426 302. Imai, Y., L.A. Lasky and S.D. Rosen. 1993.Sulfation requirement for GlyCAM-1, an endothelial ligand for L-selectin. Nature 361: 555: 303. Cyster, J.G., D.M. Shotton, and A.F. Williams. 1991. The dimensions of the T lymphocyte glycoprotein leukosialin and identification of linear protein epitopes that can be modified by glycosylation. EMBO J. 10:8931. 304. Remold-O'Donnell, E., D. Kenney and F.S. Rosen. 1987. Biosynthesis of human sialophorins and analysis of the polypeptide core. Biochemistry 26:3908. 305. Bierhuizen, M.F.A., K. Maemura and M. Fukuda. 1994 Expression of a differentiation antigen and poly-N-acetyllactosarninyl O-glycans directed by a cloned core 2 (3-1,6-N-. acetylglucosaminyltransferase. J. Biol. Chem. 269:4473. 306. Wu, W., P.H. Harley, J.A. Phunt, S. O. Sharrow and K. P. Kearse. 1996. Identification of CD8 as a peanut agglutinin (PNA) receptor molecule on immature thymocytes. J. Exp. Med. 184:759. 307. Axelsson, B., A. Kimura, S. Hammarstrom, H. Wizgell, K. Nilsson and H. Mellstedt. 1978. Helix pomatia A haemagglutinin: selectivity of binding to lymphocyte surface glycoproteins on T cells and certain B cells. Eur. J. Immunol. 8:757. 308. Conzelmann, A., R. Pink, O. Acuto, J-P Mach, S.Dolivo and M. Nabholtz. 1980. Presence of T145 on cytolytic T cell lines and their lectin-resistant mutants. Eur. J. Immunol. 10:860. 309. Shelley, C.S., E. Remold-O'Donnell, A.E. Davis, G.A.P. Bruns, F.S. Rosen, M.C. Carroll and A.S. Whitehead. 1989. Molecular characterisation of sialophorin (CD43), the -lymphocyte suface sailoglycoprotein defective in Wiskott-Aldrich syndrome. Proc. Natl. Acad. Sci. USA. 86: 2819. • 310. Pallant, A., A. Eskenazi, M-G. Mattei, R.E.K. Fournier, S;R. Carlsoon, M. Fukuda and JD. Frelinger. 1989. Characterization of cDNAs encoding human leukosialin and localization of the leukosialin gene to chromosome 16. Proc. Natl. Acad. Sci. USA. 86:1328. 311. Ardman, B., M.A. Sikorski, M. Settles and D..E. Staunton. 1990. Human immunodeficiency virus type-1 infected individuals make autoantibodies that bind to CD43 on normal thymic lymphocytes. J. Exp. Med 172:1151. 312. Giordanengo V., M. Limouse, L. Desroys Du RRoure, J. Cottalorda, A. Doglio, A. Passeron, J-G Fuzibet and J-C. Lefebvre. 1995. Autoantibodies directed against CD43 molecules with altered glycosylation status on human immunodeficiency virus type 1 (HTV-l)-infected CEM cells are found in all HTV-1 individuals. Blood 86:2302. 281 313. Brown, T.J., W.W. Shuford, W-C Wang, S. G. Nadler, T.S. Bailey, H. Marqiiardt and R.S. Mittler. 1996. Characterization of a CD43/leukosialin mediated pathway for inducing apoptosis in human T-lymphoblastoid cells. /. Biol. Chem. 271:27686. 314. De Smet, W., H. Walter and P. De Baetselier. 1993. Workshop adhesion structure subpanel 9 (CD43) mAb define at least four different epitopes on human leukosialin. In Leucocyte Typing V. White cell differentiation antigens. ("Eds. Schlossman SF. et al.) . OxfordUniversity Press. 315. Vafki A. 1993. Biological roles of oligosaccharides" all of the theories are corrcct.Glycobiology 3:97. 316. Ardman, B., M.A. Sikorski and D.E. Staunton. 1992. CD43 interferes with T lymphocyte adhesion. Proc. Natl. Acad. Sci. USA. 89:5001. 317. Manjunath, N., R. Johnson, D.E. Staunton, R. Pasqualini and B. Ardman. 1993. Targetted disruption of CD43 gene enhances T lymphocyte adhesion. J. Immunol. 151: 1528. • . 318. McFarland, T.A., B. Ardman, N. Manjunath, J.A. Fabry and J. Lieberman. 1995. CD43 diminshes susceptibility to T lymphocyte-mediated cytolysis. J. Immunol. 154:1097. 319. Manjunath, N., M. Correa, M. Ardman and B. Ardman, 1995. Negative regulation of T-cell adhesion and activation by CD43. Nature 377:535. 320. Rosenstein, Y;, J.K. Park, W.C. Hahn, F.S. Rosen, B.E. Bierer and S.J. Burakoff. 1991. CD43, a molecule defective in Wiskott-Aldrich syndrome binds ICAM-1. Nature 354:233:: 321. Stockl, J., O. Majdic, P. Kohl, W.F. Pikl, J.E. Menzel and W. Knapp. 1996. Leukosialin (CD43)-major histocompatibility class I molecule interactions involved in spontaneous T cell conjugate formation. J. Exp. Med. 184:1769: 322. Baum, L:G., M. Pang, N.L. Prillo, T. Wu, A. Delegeane, CH. Uittenbogaart, M. Fukuda and J.J. Seilhamer. 1995. Human thymic epithelial cells express an endogenous lectin, galectin-1, which binds to core 2 O-glycans on thymocytes and T lymphoblastoid cells. /. Exp. Med. 181:877. 323. Brandley, B.K., S.J. Sweidler and P.W. Robbins. 1990. Carbohydrate ligands of the LEC cell adhesion molecules. Cell 63:861. 324. Wilkins, P.P., R.P. McEer, and R.D. Cummings. 1996. Structures of the O-glycans on P-selectih glycoprotein ligand-1 from HL-60 cells. J.Biol Chem. 271:18732. 325. Bird, J.M. and S.J. Kimber. 1984. Oligosaccharides containing fucose linked alpha (1-3) and alpha (1-4) to N-acetylglucosamine causes.decompaction of murine morulae.7. Exp. Med. 160:1591 326. Crocker, P.R. and S. Gordon. 1989. Mouse macrophage haemagluttinin (sheep erythrocyte receptor) with specificity for sialylated glycoconjugates characterised by a . monoclonal antibody. J. Exp. Med. 169:1333, 327. Crocker, P.R., Z. Werb, S. Gordon and D.F, Bainton. 1990. Ultrastructural localization of a macrophage-restricted sialic acid binding haemagglutinin, SER, in macrophage-haemopoietic cell clusters. Blood 76:1131. 282 328. Crocker, P.R., S. Kelm, C. Dubois, B. Martin, A.S. McWilliam, D.S. Shotton, J.C. Paulson and S. Gordon. 1991. Purification and properties of sialoadhesin, a sialic acid binding receptor of murine tissue macrophages. EMBO J. 10:1661. 329. Van den Berg, T.K.-, J.J.P. Breve, J.G.M.C. Damoiseaux, E.A. Dopp, S. Kelm, PR. Crocker, CD. Dijkstra and G. Kraal. 1992. Sialoadhesin on macrophages: Its . identification as a lymphocyte adhesion molecule. /. Exp. Med. 176:647. 330. Bazil, V., J. Brandt, S. Chen, M. Roeding, K. Luens, A. Tsukamoto and R. Hoffman. 1996. A monoclonal antibody recognising CD43 (leukosialin) initiates apoptosis of human hemopoietic progenitor cells but not stem cells. Blood 87:1272. 331. Gordon, J. 1994. B-cell signalling via the C-type lectins CD23 and CD72. Immunol.Today 15:411. , 332. Kolbe, J.P., A. Abadie, A. Proschnicka-Chalufour, A. de Gramont and J. Poggioli. 1994. CD23-mediated cell signalling. J. Lipid Med and Cell Signal 9:27. 333. Barpndes, S.H., D.N.W. Cooper, M.A. Gitt and H. Lefflert. 1994. Galectins: Structure and function of a large family of animal lectins. /. Biol. Chem. 269:1. 334. Perillo, N.L., K.E. Pace, J.J. Seilhamer and L.G. Baum. 1995. Apoptosis of T cells mediated by galectin-1. Nature 378:736. y 335; Inohara, H. and A. Raz. 1995. Functional evidence that cell surface galectin-3 mediates homptypic cell adhesion. Cancer Res. 55:3267. 336. Baeckstrom, D., K. Zhang, N. Asker, U. Ruetschi, M. Ek and and G.C Hansson. 1995. Expression of the leukocyte-associated sialoglycoprotein CD43 by a colon carcinoma cell line. J. Biol. Chem .270:13688. 337. Santamaria, M., A. Lopez-Beltran, M. Toro, J. Penaand I.J. Molina. 1996.Specific monoclonal antibodies against leukocyte-restricted cell surface molecule CD43 react with non-hemopoietic tumour cells. Cancer Res. 56:3526. 338. Alvarado, M., C. Klassen, J. Cerny, V. Horejsi and R:E. Schmidt. 1995. MEM-59 monoclonal antibody detects a CD43> epitope involved in lymphocyte activation. Eur. J. Immunol. 25:1051. 339. Cerhy, J., H. Stockinger and V. Horejsi. 1996. Non-covalent associations of T lymphocyte surface proteins. Eur. J. Immunol. 26:2335. , 340. Manjunath, N and B. Ardman. 1995. CD43 regulated tyrosine phosphorylation of a 93 kD protein in T-lymphocytes. Blood 86:4194-341. Mentzer, S.J., E. Remold-O'Donnell, M.A. Crimmins, B.E. Bierer and S.J. Burakoff. 1987. Sialophorin, a surface sialoglycoprotein defective in the Wiskott-Aldrich syndrome, is involved in T lymphocyte proliferation. J. Exp. Med. 165:1383. 342. Axelsson, B., R. Youseffi-Etemad, S. Hammarstrom and P. Perlmann. 1988. Induction of aggregation and enhancement of proliferation and IL-2 secretion in human T cells by antibodies to^ D43. J. Immunol. 141:2912. 343. Silverman, L.B., R.C. Wong, E. Remold-O'Donnell, D. Vercelli, J. Sancho, C. Terhorst, F. Rosen, R. Geha and T. Chatila. 1989. Mechanism of mononuclear cell activation by an anti-CD43 (sialophorin) agonistic antibody. J. Immunol. 142:4194. 283 344. Park, J.K., Y.J. Rosenstein, E. Remold-O'Donnel, BE. Bierer, F.S. Rosen and S. Burakoff. 1991. Enhancement of T-cell activation by the. CD43 molecule whose expression is defective in Wiskott-Aldrich syndrome. Nature 350:706. 345. Cyster, J. and A.F. Williams. 1992. The importance of cross-linking in the homotypic aggregation of lymphocytes induced by anti-leukosialin (CD43) antibodies. Ear. J. Immunol. 22:2565. 346. Youseffi-Etemad and B. Axelsson. 1996. Parallel pattern of expression of CD43 and of LFA-1 on the CD45RA (naive) and CD45RO 'memory subsets of human CD4 and CD8 cells. Correlation with the aggregative response of the cells to CD43 monoclonal antibodies. Immunology 87:439. 347. Sperling, A.I., J.M. Green, R.L. Mosley, P.L.Smith, R. DiPaolo et al. 1995. CD43 is a murine T cell costimulatory receptor that functions independently of CD28. J. Exp. Med. 182:139. ' 348. Bazil, V., and J.L. Strominger. 1993. CD43, the major sialoglycoprotein of human leukocytes is prOteolytically cleaved from the surface of stimulated lymphocytes and granulocytes. Proc. Natl. Acad. Sci. USA .90:3792.. 349. Bazil, V., J. Brandt, A. Tsukamoto and R. Hoffman. 1995. Apoptosis of human haemopoietic progenitor cells induced by cross-linking of surface CD43, the major sialoglycoprotein of leukocytes. Blood 86:502. 350. Lee, A. C, J. E. Powell, G. W. Tregear, H. D. Niall, and V. C. Stevehs.1980. A method for preparing |3-hCG COOH peptide-carrier conjugates of predictable composition. Mol. Immunol. 17:749. 351. Ziltener, H. J., I. Clarke-Lewis, and S. L. McDonald. 1989. Sandwich enzyme immunoassay for murine EL-3. Cytokine 1.-56. 352. Karasuyama, H., and F. Melchers. 1988. Establishment of mouse cell lines which constitutively secrete large quantitites of interleukin 2, 3,4 or 5, using modified cDNA expression vectors. Eur. J. Immunol. 18:97. 353. Clark-Lewis, I., S. B. H. Kent, and J. W. Schrader. 1984. Purification to apparent homogeneity of a factor stimulating the growth of multiple lineages of haemopoietic cells. J. Biol. Chem. 259:7488. 354. Bourne, D. W. A. 1986. MULTIFORTE, a microcomputer program for modeling and simulation of pharmacokinetic data. Comput. Programs Biomed. 23:277. 355. Crapper, R. M:, and J. W. Schrader. 1983. Frequency of mast cell precursors in normal tissues by an in vitro assay: Antigen induces parallel increases in the frequency of P cell precursors and mast cells. J. Immunol.131:923. 356. Good, M.F., A.W. Boyd and G.J.V. Nossal. 1983. Analysis of the anti-hapten cytotoxic clones in limit dilution rnicroculfures after correction for anti-self activity: precursor frequencies, Lyt-2 and Thy-1 phenotype, specificity and statistical methods. J. Immunol. 130:2046. • 357. Mayrhofer, G. 1980. Fixation and staining of granules in mucosal mast cells and intraepithelial lymphocytes in the rat jejunum, with special reference to the relationship between the acid glycosaminoglycans in the two cell types: Histochemical Jounal 12:513. 284 358. Ihle, J., N. Keller, P.St.Oroszlan, L.E. Henderson, T.D. Copeland, F. Fitch, M.B. Prystowsky, E. Goldwasser, J.W. Shrader, E. Palazynski, M. Dy and B. Lebel. Biological properties of homogeneous interleukin 3:1. Demonstration of WEHI-3 growth factor activity, mast cell growth factor activity, P cell stimulating activity and histamine-producing cell activity. J. Immunol. 131:282. 359. Yang, Y.C.H., A.B. Ciarletta, P.A. Temple, M.P. Chung, S.H. Kovacic, J.S. Witek-Giannotti, A.C. Leary, R. Kriz, R.E. Donahue, G.G. Wong ansd S. Clark. 1986. Human IL-3 (multi-CSF): Identification by expression cloning of a novel hematopoietic growth . factor elated to murine IL-3. Cell 47:3. 360. Kaushansky, K., CB. Brown arid P.J. O'Hraa. 1990. Molecular modelling of human granulocyte-macrophage colony stimulating factor. Int. Cell Cloning 8:26. 361. Clark-Lewis, I., R. Aebersold, H.Z. Ziltener, J.W. Schrader, L.E. Hood and S.B.H. Kent. 1986. Automated chemical synthesis of a protein growth factor for hemopoietic cells, interleukin 3. Science 231.134. 362. Duronio, V, S.R. Granleese, I. Clark-Lewis, J.W. Schrader and H.J. Ziltener. 1991. Antibodies to interleukin 3 as probes for the interaction of interleukin 3 with its receptor. Cytokine 3:414. 363. Smith, T.W., E. Haber, L. Yeatman and V.P. Butler Jr. 1976. Reversal of advanced digoxin intoxication with fab fragments of digoxin specific antibodies. New Eng. J. Med. 294:797. , 364. In Review of Medical Physiology . (Ed. W.F. Ganong). Appleton and Lange. 1987. Norwalk, USA. p581. • 365. Crapper, R.M., Clark-Lewis, I., Schrader, J.W. 1984.The in vivo functions and properties pf persisting cell stimulating factor. Immunology 53:33. 366. In: Goodman and Gillman's the Pharmacological Basis of Therapeutics. 1990. (Eds. A.G. Goodman Gilman, T.W. Rail, A.S. Nies and P. Taylor). Pergamon Press. p3. 367. Hugh-Jones, N.C, Gorick, B.D., Howard, J.C. 1983. The mechanism of synergistic complement-mediated lysis of rat red blood cells by monoclonal IgG antibodies. Eur. J. Immunol. 13: 635. 368. Klassen, R.J.L., Goldschmeding, R., Tetteroo, P.AT., Kr. Von dem Borne, A. E.G. 1988. The Fc valency of an immune complex is the decisive factor for binding to low-affinity Feyreceptors. Eur. J. Immunol. 18:1373. 369. Strickland, R.W., Wahl, L.M. Finbloom, D.S. 1988. Dimers of human, immunoglobulin Gl provide an insufficient signal for their degradation by human monocytes. Clinical Immunol, and Immunopathol. 48:10. 370. Burgess, A.W. and D. Metcalf. 1980. The nature and action of granulocyte macrophage colony stimulating factors. Blood 56:947. 371. Iscove, N.N., CA. Roitsch, N. Williams, and L.J. GuifberL 1982. Molecules stimulating early red cell, granulocyte, macrophage and megakaryocyte precursors in culture: Similarity in size, hydrophobitity and charge. J. Cell Physiol. 65:111 285. 372. Metcalf, D. 1985. Multi-CSF-dependent colony formation by cells of a murine hematopoietic cell line: specicity and action of multi-CSF. Blood 65:357. .'373. Schrader, J.W., I. Clark-Lewis, R.M. Crapper et al. 1988. In Lymphokines 15; Interleukin 3: The Panspecific /lemopo/erin . fE'd J.W. Schrader). Academic Press^  San Diego, CA. p28l. • 374. Metcalf, D. 1991. Lineage commitment of hemopoietic progenitor cells in developing blast colonies: Influence of colony stimulating factors. Proc. Natl. Acad. Sci. USA. 84:5267. 375. Moore, M.A.S. 1991. Clinical implications of positive and negative hematopoietic stem cell regulators. Blood 78:12. 376. Lopez, A.F., P.G. Dyson, L.B. To, M.J. Elliot, et al. 1988. Recombinant human interleukin 3 stimulation of haemopoiesis in humans: Loss of responsiveness with differentiation in the neutrophilic myeloid series: Blood 73:1791. 377. Ema, H., T. Suda, K. Nagayoshi, Y. Miura et al. 1990. Target cells for granulocyte colony-stimulating factor, interleukin 3,.and interleukin 5 in differentiation pathways of neutrophils and eosinophils. Blood 76:1956. 378. Metcalf, D. 1989. The molecular control of cell division, differentiation commitmenf and maturation in haemopoietic cells. Nature 339:27. 379. Denburg, J.A. 1992. Basophil and mast cell lineages in vitro and in vivo. Blood 79:846 380. Rodewald, H.R, M. Dessing, A.M. Dvorak and S.J. Galli. 1996. Identification of a committed precursor for the mast cell lineage. Science 271:818. 381. Nagao, K.^  K. Yokoro and S.A. Aaronson. 1981. Continuous lines of basophil/mast cells derived from normal mouse bone marrow. Science 212:333. 382. Stevens, R.L. and K.F. Austen. 1989. Recent advances in the cell and molecular biology of mast cells. Immunol. Today 10:38.1. 383. Valent, P., J. Besemer, CH. Sillabar, J.H. Butterfield etal. 1990. Failure to detect IL-3-binding sites on human mast cells. J. Immunol. 145:3432. 384. Metcalf, D. 1988. The multipotential colony-stimulating factor, multi-CSF. Lymphokines 15:183. 385. Welham, M.J., Schrader, J.W. 1991. Modulation of c-kit mRNA and protein by hemopoietic growth factors. Mol. Cell. Biol. 11:2901. 386. Gianella-Borradori, A. 1994. Present and future clinical relevance of interleukin3. Stem Cells. 12: (Suppl 1)241. : ./ 387. Kedar, E., Y. Rutkowski, E. Braun and Y. Barenholz. 1994. Delivery of cytokines by liposomes. I. Preparation and charcterization of interleukin-2 encapsulated in long-circulating sterically stabilized liposomes. J. Immunotherwith Emphasis on Tumor Immunol. 16:47. 388. Lucey, D.R., M. Clerici and G.M. Shearer: 1996. Type 1 and type 2 cytokine dysregulation in human infectious, neoplastic and inflammatory disease. Clin. Microbiol. Rev. 9:532. 286 389. Morrison, S., M.J. Johnson, L.A. Herzenberg and V.T. Oi. 1984.Chimeric human antibody molecules: mouse antigen-binding domains and human constant region domains. Proc. Natl. Acad. Sci USA 81:6851 390: Neuberger, M.S., G.T. Willimas and R.O. Fox. 1984. Recombinant antibodies possessing novel effector functions. Nature 312:604. 391. Jones, P.T., P.H. Dear, J. Foote, M.S. Neuberger and G. Winter. 1986: Nature 321:522. " ' 392. Lobuglio, A.F., R.H. Wheeler, J. Tracy, A. Haynes, K. Rogers, E. Harvey etal. 1989. Mouse/human chimeric monoclonal antibody in man: kinetics and immune response. Proc. Natl. Acad. Sci. USA. 86:4220. 393. Jolliffe, L.K. 1993. Humanized antibodies: enhancing therapeutic utility through antibody engineering. Int. Rev,. Immunol. 10:241. 394. Burnett, K.G., J. Martinis and R.M. Bartholomew. 1982. Production of bifunctional antibodies by hybridoma technology. Biotechnology: Appl. Res. 31:401. 395. Corvalen, J.R.F., W. Smith, V.A. Gore and D.R. Brandon. 1987. Specific in vitro and in vivo drug localisation to tumour cells using a hybrid-hybrid monoclonal antibody recognising both carcinoembryonis antigen (CEA) and vinca alkaloids. Cancer Immunol, mmunother. 24:133. 396. Brennan, M., P.F. Davison and H. Paulus. 1985. Preparation of bi-specific antibodies by chemical recombination of monoclonal immunoglobulin Gl fragments. Science 229:81. 397. Phelps, J.L., DE. Beidler, R.A. Jue, B.W. Unger and M.J. Johnson. 1990. Expression and characterisation of a chimeric bifunctional antibody with therapeutic applications. J. Immunol. 145:1200. ' , 398. Fanger, M.W. and P.M. Guyre. 1991. Bispecific antibodies for targeted cellular cytotoxicity. Trends Biotech. 9:375 . 399. Finkelman, F., K.B. Madden, S.C. Morris, J.M. Holmes et al. 1993. Anti-cytokine antibodies as Carrier proteins: Prolongation of in vivo effects of exogenous cytokines by injection of cytokine-anti cytokine antibody complexes. J. Immunol. 151:1235. 400. Rosenblaum, M.G., B.W. Unger, J.U. Guttman, E.M. Hersh, G.S. David and J.M. Frinke. 1985. Modification of human leukocyte interferon pharmacology with a monoclonal antibody. Cancer Res. 45:2421. 401. Sato, J., N. Hamaguchi, K. Doken, K. Gotoh, K. Ootsu, S. Iwasa et al. 1993. Enhancement of anti-tumor activity of recombinant interleukin-2 (rIL-2) by immunocomplexing with the monoclonal antibody against rIL-2. Biotherapy 6:225. 402. Aston, R., T. Holder, M.A. Preece and J. Ivanyi. 1986. Potentiation of the somatogenic and lactogenic actvity of human growth hormone with monoclonal antibodies. J. Endocr. 110:381. 403. Dower, S.K., S.R. Kronheim, C.J. March et al. 1985. Detection and characterisationof high affinity plasma membrane receptors for human interleukin 1. J. Exp. Med. 162:501. 404. Thieme, T.R. and CR. Wagner. 1989. The molecular weight of the endothelial cell IL-1 receptor is 78,000. Mol. Immunol. 26:249. 287 405. Schechter, Y., K-J. Chang, S. Jacobs and P. Cuatrecasas. 1979. Modulation of binding and bioactivity of insulin by anti-insulin antibody: Relation to possible role of receptor self-aggregation in hormone action. Proc. Natl. Acad. Sci. USA. 76:2720. 406. Schecter, Y., L. Hernaez, J. Schlessinger and P. Cuatrecasas. 1979. Local aggregation of hormone receptor complexes is required for activation by epidermal growth factor. Nature 278:835. 407. Holder, A.T., R. Aston, J.R. Rest, D.J. Hill, N. Patel and J. Ivanyi. 1987. Monoclonal antibodies can enhance the biological activity of thyrotropin. Endocrinol. 120:567. 408. Spiegelberg, H.L. 1989. Biological role of different antibody classes. Int. Arch. Allergy & Appl. Immunol. 90 1:22. 409. James, K. 1990. Interaction between cytokines and oc2 macroglobulin. Immunol. Today 11:163. 410. Philip, A. and M.D. O'Connor-McCourt. 1991. Interaction of transforming growth factor (31 with a2 macroglobulin. J. Biol Chem. 266:22290. 411. Fernandez-Botran. R. 1991. Soluble cytokine receptors: their role in immunoregulation. FASEB J. 5:2567. 412. Aderka, D., H. Englmann, Y. Maor, C. Brakebusch and D. Wallace. 1992. Stabilisation of the bioactivity of tumour necrosis factor by its soluble receptor. /. Exp. Med. 175:323. 288 Figure 36 IB 11 antigen binds to wheat germ agglutinin. Whole cell lysates were prepared from D26 cell line in the presence of protease inhibitors. Lysates were applied to a 1ml WGA column and the flow through and washes collected. Glycoproteins bound to the column were then eluted with N-acetyl glucosamine and eluate was collected in 1 ml fractions. All fractions were then resolved on 7.5% SDS PAGE gel and 1B11 mAb immunoreactivity was detected by immunoblotting. Lane a Original lysate; Lane b Flow through fraction: Lane c Eluate fraction 1: Lane d Eluate fraction 2: -: 145: ml of ImM HG1 and aliquotted, into 2 Eppendorf tubes. The beads were then resuspended in 300 |il of either IB 11 mab at 0.6 mg/ml or 300 |il of mouse anti-rat kappa chain antibody at 0.6 mg/ml. The mixture was then rotated for 2 hours at room temperature. At the end of the incubation period, the protein content of the supernatant was determined by spectrophotometric analysis of absorbance at 280nm. Following completion of the reaction, as determined by a marked reduction in the protein content of the supernatant, 10% v/v saturated glycine solution was added to the mixture to saturate non-specific protein-binding sites and the beads rotated for 1 hour at room temperature. The beads were then washed 3 times in PBS and resuspended in 200 ul PBS containing 0.5% w/v sodium azide prior to use. < 35S-labeled D26 whole cell lysates were prepared as previously described. One milliliter aliquot of lysate was pre-cleared with 20 ul of a 50% slurry of Protein-G coupled Sepharose beads for 30 minutes at 4°C. 200 JLLI aliquots of supernatant were then incubated with either a) 20 |il of a 50 % slurry of IB 11-Sepharose beads in PBS and rotated fori hour at 4°C, or (b and c) 10 mg of IB 11 mab for 1 hour at 4°C in duplicate. After 1 hour, 20 ul of protein G coupled Sepharose beads were added to b) and 20 (ll of mouse anti-rat kappa-coupled Sepharose beads were added to c) and the beads were rotated for 1 hour at 4°C. Finally, all three preparations were pelleted and washed sequentially with i) IM NaCI, 0.1M Tris pH 8, 0.1% NP-40, ii) 0.1'M NaCI, 0.1M Tris pH 8, 1%' NP-40, 0.3% SDS, 0.00IM EDTA, and iii) 0.01M tris pH 8, 0.1% NP-40. After washing, bound material was eluted with 30 jil diethanolamine pH 11.5, 0.05% NP40. Eluates were neutralised with 1/5 volume IM Tris pH 7 and diluted into 2% SDS sample buffer for analysis by SDS PAGE. Immunoprecipitates were resolved on 7.5% SDS PAGE gel and then stained for 30 minutes in staining buffer containing 41% v/v methanol, 7% glacial acetic acid, 0.325g/l of Coomassie Brilliant Blue dye. After staining, gels were transferred to destain (41% v/v methanol, 7% glacial acetic acid) for 1 hoUr and then transferred to Amplify (Amersham), for fluorography for 30 minutes. The gel was then placed on a.double thickness of Whatman 3MM paper, covered with Saran wrap and dried at 80 °C under vacuum. Gels were autoradiographed with Kodak XAR-5 film at-70 °C. - . . 23 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088803/manifest

Comment

Related Items