UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The Ly-49 family of natural killer cell receptors Brennan, John A. 1996

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

Item Metadata


831-ubc_1996-090485.pdf [ 8.85MB ]
JSON: 831-1.0087916.json
JSON-LD: 831-1.0087916-ld.json
RDF/XML (Pretty): 831-1.0087916-rdf.xml
RDF/JSON: 831-1.0087916-rdf.json
Turtle: 831-1.0087916-turtle.txt
N-Triples: 831-1.0087916-rdf-ntriples.txt
Original Record: 831-1.0087916-source.json
Full Text

Full Text

THE LY-49 FAMILY OF NATURAL KILLER CELL RECEPTORS by J O H N A. BRENNAN B.Sc, The University of Dayton, 1990 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF DOCTOR OF PHILOSOPHY in T H E F A C U L T Y OF GRADUATE STUDIES Genetics Programme We accept this thesis as conforming to the required standard T H E UNIVERSITY OF BRITISH COLUMBIA February 1996 © John A. Brennan, 1996 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 The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract This thesis investigates the nature of receptors involved in target cell recognition by natural killer (NK) cells. N K cells are thought to scan for self class I major histocompatibility complex (MHC) antigens, receive negative signals from interactions with these molecules, and lyse those cells which are unable to deliver protective signals. This system ensures that cells which have extinguished expression of class I M H C and evade T cell immunity are detected by N K cells. Ly-49A was the first N K cell receptor found to receive inhibitory signals from class I M H C on opposing cells. Because Ly-49A is expressed by a subset of N K cells and recognizes only a fraction of class I molecules, additional subsets of N K cells are thought to exist which express receptors for class I molecules not recognized by Ly-49A. Ly-49A is a type II transmembrane protein belonging to a group of related molecules encoded by a murine multigene family. The expression patterns and functional activities of other members of the Ly-49 gene family was the topic of this thesis. Ly-49A and Ly-49C expression was shown to identify distinct single positive, double positive, and double negative subsets of N K cells. Both Ly-49A and Ly-49C were found to mediate adhesion to class I M H C on a variety of cell lines. The specificity of these receptors was shown to be distinct but overlapping, with Ly-49C having a much broader range of class I ligands. A l l members of the Ly-49 family have an extracellular region homologous to the carbohydrate recognition domain of C-type lectins. The functional relevance of this domain was demonstrated by the inhibition of Ly-49C-mediated cell adhesion by both exogenous polysaccharides as well as by the modification of target cell surface carbohydrates. This suggests that the glycosylation of class I M H C may be a critical element in Ly-49 recognition. These combined results support the hypothesis that Ly-49 is a family of N K cell receptors with related, but distinct functions. Ly-49A and Ly-49C are both ii expressed by NK cells, but not by the same NK cells, and they both bind to class I MHC molecules, but not the same class I molecules. These receptors may therefore be responsible for the generation of an NK cell repertoire capable of recognizing cells which lose expression of single class I molecules. iii Table of Contents Page ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES vii LIST OF FIGURES viii LIST OF ABBREVIATIONS x ACKNOWLEDGMENTS xi Chapter 1 INTRODUCTION 1 1.1) Natural Killer Cells 1.1.1) Origin and function A) Discovery 2 B) Surface markers 2 C) Ontogeny 3 D) Cytokine regulation 5 E) Regulation of hematopoiesis and cytokine secretion 5 1.1.2) NK cell mediated cytotoxicity A) Antibody-dependent cellular cytotoxicity 6 B) Anti-viral activity 7 C) Anti-tumor activity 9 D) Cytolytic mechanisms 9 1.1.3) Carbohydrates and NK cell activity A) Exogenous carbohydrates 10 B) Target cell surface carbohydrates 10 1.2) Target Cell Specificity of Natural Killer Cells 1.2.1) The Major Histocompatibility Complex A) Genomic organization 11 B) Structure and function of class I MHC 14 1.2.2) Hybrid resistance A) Immunogenetics of bone marrow graft rejection 16 B) Effector cells 18 1.2.3) Class I MHC and NK cell recognition A) Mouse 19 B) Human 21 1.2.4) The 'missing self hypothesis 23 1.3) Candidate NK Cell Receptors 1.3.1) Mouse A) Ly-49 25 iv B) 5E6 29 C) N K R - P 1 / N K 1 . 1 30 D) NK2.1 32 E) LGL-1 34 F) 2B4 34 1.3.2) Human A) GL183 and E B 6 35 B) N K B 1 39 C) NK3-specific N K cells 40 D) CD94 41 E) N K G 2 41 F) N K - T R 43 1.4) Thesis Objectives 43 Chapter 2 M A T E R I A L S A N D M E T H O D S 46 2.1) Animals 47 2.2) N K Cell Preparations 47 2.3) c D N A Cloning 47 2.4) D N A Sequence Analysis 48 2.5) Cell Lines 48 2.6) Antibodies 49 2.7) Polysaccharides and Glycosidases 49 2.8) Transfections 49 2.9) Flow Cytometry and Cell Sorting 2.9.1) Fresh and IL-2 activated N K cells 50 2.9.2) Cell lines 51 2.9.3) Polysaccharide inhibition 51 2.9.4) Peanut Lectin flow cytometry 51 2.10) Cell Adhesion Assays 52 2.11) Glycosidase Treatments 53 Chapter 3 T H E LY-49 M U L T I G E N E FAMILY: N E W M E M B E R S , A N T I B O D Y SPECIFICITIES A N D N A T U R A L K I L L E R C E L L E X P R E S S I O N 54 3.1) Introduction 55 3.2) Results 3.2.1) Cloning of additional members of the Ly-49 multigene family 56 3.2.2) COS cell expression of Ly-49 cDNAs and reactivity with Ly-49 antibodies 61 3.2.3) Expression of YE1/48 and 5E6 define distinct N K subsets 61 3.2.4) YE1/32 and YE1/48 expression on N K cells of BIO M H C congenic mice 64 3.3) Discussion 64 V Chapter 4 R E C O G N I T I O N O F C L A S S I M H C B Y M E M B E R S O F T H E LY-49 F A M I L Y 69 4.1) Introduction 70 4.2) Results 4.2.1) COS cells transfected with Ly-49A or Ly-49C cDNAs bind a variety of cell lines 71 4.2.2) Ly-49 mediated adhesion is inhibited by anti-Ly-49 and anti-class I M H C antibodies 75 4.2.3) Characterization of Ly-49A-class I M H C interactions in a two way transfection system 78 4.3) Discussion 78 Chapter 5 C A R B O H Y D R A T E R E C O G N I T I O N B Y LY-49C 85 5.1) Introduction 86 5.2) Results 5.2.1) Sulfated polysaccharides inhibit Ly-49C-mediated cell adhesion 87 5.2.2) Polysaccharide inhibition of 5E6 binding to Ly-49C 90 5.2.3) Glycosidase treatment of GM979 90 5.3) Discussion 92 Chapter 6 S U M M A R Y A N D P E R S P E C T I V E S 99 R E F E R E N C E S 108 vi List of Tables Page Table 1 - Immune-genetics of bone marrow graft rejection 17 Table 2 - Homologies between members of the Ly-49 family 60 Table 3 - Adhesive properties of Ly-49A and Ly-49C expressing COS cells 73 Table 4 - Polysaccharide inhibition of Ly-49C-mediated adhesion 89 Table 5 - NK cell receptors for class I MHC 101 Table 6 - Hypothetical interactions between Ly-49 receptors and class I MHC molecules 104 Table 7 - Hypothetical development of the NK cell repertoire 105 vii List of Figures Page Figure 1 - Organization of genes of the major histocompatibility complex 13 Figure 2 - Schematic diagram of a class I major histocompatibility complex molecule 15 Figure 3 - The 'missing self hypothesis 24 Figure 4 - H-2 specificity of Ly-49A+ NK cells 27 Figure 5 - Sequence relatedness of NK cell lectin-like proteins 33 Figure 6 - Alloreactivity of human NK cell clones 37 Figure 7. Nucleotide sequence comparison of Ly-49A-E cDNAs 58 Figure 8 - Amino acid sequence comparisons Ly-49A-E 59 Figure 9 - COS cell expression of the Ly-49 cDNAs 62 Figure 10 - Phenotypic analysis of Ly-49A and Ly-49C populations in C57BL/6 spleen cells 63 Figure 11 - Expression of YE 1/48 on NK cells of MHC congenic mice 65 Figure 12 - Experimental protocol of COS cell adhesion assay 72 Figure 13 - Adhesion of cell lines to COS cells expressing Ly-49A or Ly-49C 74 Figure 14 - Quantitation of cell-cell binding mediated by Ly-49A or Ly-49 C 76 Figure 15 - Distinct class I MHC specificities of Ly-49A and C 77 Figure 16 - Expression of class I MHC by GM979 and its H-2Db, Dd, and D k transfectants 79 Figure 17 - Adhesion of Ly-49A to GM979 expressing transfected class I MHC genes 80 Figure 18 - Quantitation of Ly-49A-H-2Dd/Dk mediated cell adhesion 81 Figure 19 - Polysaccharide dose response inhibition of Ly-49C-mediated adhesion 88 viii Figure 20 - Competition of antibody binding by polysaccharides 91 Figure 21 - Effects of glycosidase treatment and calcium depletion on Ly-49C mediated adhesion 93 Figure 22 - Effects and effectiveness of glycosidase treatments 94 ix List of Abbreviations a.a., amino acid Ab, antibody ADCC, antibody-dependent cellular cytotoxicity A S G M 1 , Asialo-GMl Asn, asparagine B6, C57BL/6 (32m, ^-microglobulin BMC, bone marrow cells bp, base pairs BSA, bovine serum albumin CB6F1, (C57BL/6 x BALB/c )F i cpm, counts per minute CTL, cytotoxic T lymphocyte CRD, carbohydrate recognition domain Cy5, cyanine 5 D M E M , Dulbecco's modified minimum essential media dNTP, deoxyribonucleoside triphosphate DTT, dithiothreitol EDTA, ethylene diamine tetraacetate EGTA, ethylene glycol tetraacetate FACS, fluorescence activated cell sorting FCS, fetal calf serum FITC, fluorescein isothiocynate GM-CSF, granulocyte macrophage-colony stimulating factor HBSS, Hanks' balanced salt solution IFN, Interferon Ig, immunoglobulin IgSF, immunoglobulin superfamily IL-2, Interleukin 2 kD, kilodalton LGL, large granular lymphocyte Lys, lysine mAb, monoclonal antibody MCMV, murine cytomegalovirus M H C , major histocompatibility complex mw, molecular weight NK, natural killer PBL, peripheral blood lymphocyte PCR, polymerase chain reaction PE, phycoerethrin PI, propidium iodide RBC, red blood cells RFLP, restriction fragment length polymorphism RPMI, Roswell Park Memorial Institute SD, standard deviation SDS, sodium dodecyl sulfate SEM, standard error of the'mean Ser, serine TCR, T cell receptor TNF, tumor necrosis factor Acknowledgments I would like to thank the following people for their many forms of support and contribution over the years: Steve Becker Peter Lansdorp Hisham Bismar Dixie Mager Elaine Brennan Gwen Mahon John Brennan Guitta Maki Marian Brennan Patricia Marin Michael Brennan Gabriel Garcia Marquez Carmine Carpenito Stephanie Masterman Eibhlin Conneally Parvine Moayedi Colleen Dailey David Nelson Adriana DiFranco Aaron Nothnagle Graeme Dougherty Shera Paterson Julie Follosco Todd Pfeiffer Pe Follosco Ron Price Doug Freeman Jim Quinn Don Geiger Lou Reed Ahmad Ghaffari Ann Rose Saghi Ghaffari Lily Rosenberg Glenn Graham Cam Smith Maureen Graham Jack Sullivan Ernest Hemingway Jeannie Sullivan Donna Hogge Fumio Takei Wilf Jefferies Gayle Thornbury Rob Kay Ed Vebelun Christine Kelly Ian Whitehead Dan Kilian Simon Wong Kevin Kinser Dave Wilkinson Hans Klingemann Nick Zoumbos Paul Kowalski xi Chapter 1 Introduction i 1.1) Natural Killer Cells 1.1.1) Origin and function A) Discovery In 1975 a novel cell type was described which was functionally defined by its ability to lyse a variety of tumor cell lines without any prior sensitization or self MHC restriction (Herberman et al., 1975; Kiessling et al., 1975). This spontaneous cytotoxicity was termed natural cytotoxicity and the effectors were called natural killer cells. N K cells are a component of the natural immune system, which also includes phagocytic cells, complement, and physical barriers such as mucous membranes. These defense mechanisms are generally present and fully functional prior to infection, and therefore constitute a first line of defense (Abbas et al., 1991). Acquired immunity mediated by T and B lymphocytes, however, is highly specific and involves a series of cognition, proliferation, and differentiation events which proceed the effector phase of immune function. T cell mediated cytotoxicity specifically differs from N K cell killing in that the former requires self M H C antigens on the surface of the target cell as well as an activation period prior to the acquisition of lytic potential (Trinchieri, 1989). B) Surface markers For many years N K cells remained functionally defined as a class of lymphocytes with large granular morphology which mediates natural, non-MHC restricted cytotoxicity. This activity was attributed to "null" cells because N K cells were found to be devoid of typical T and B cell surface antigens (Lanier et al., 1986b). In addition to their CD3- cell surface phenotype, N K cells neither rearrange nor express T cell receptor genes (Biron et al., 1987; Lanier et al., 1986a). In humans, N K cells are most accurately identified by the CD3CD56+ 2 phenotype. The CD56 antigen is the product of the differentially spliced N C A M gene, and is expressed by virtually all peripheral blood cells capable of non-MHC restricted cytotoxicity (Lanier et al., 1989a). The most specific marker of murine N K cells is the NK1.1 alloantigen (Koo and Peppard, 1984). In NK1.1+ strains of mice this antigen is expressed on all CD3- N K cells capable of lysing YAC-1 lymphoma cells in vitro (Hackett et al., 1986b). AsialoGMl is expressed on all N K cells in a strain-independent manner (Young et al., 1980), but this antigen is a less specific marker than NK1.1 because it is also found on both cytotoxic T lymphocytes and macrophages. The majority of human and murine N K cells express CD 16, the low affinity receptor for the Fc region of IgG (Perussia and Trinchieri, 1984). This antigen is not N K cell specific, however, and is also found on macrophages and neutrophils (Trinchieri, 1989). Many monoclonal antibodies have recently been produced which identify discrete subsets of both human and murine N K cells. Most of the antigens identified by these antibodies appear to function in N K cell target specificity and will be discussed further in the section on candidate N K cell receptors (Section 1.3). C) Ontogeny Bone marrow transplantation into lethally irradiated recipients restores N K cell activity. This restoration has been described for both experimental animals and humans receiving allogeneic marrow transplants, and demonstrates that N K cells are derived from bone marrow precursors (Ault et al., 1985; Hackett et al., 1985). The selective ablation of murine bone marrow with 8 9 Sr (bone seeking isotope) or by the administration of 17-0 estradiol results in a specific N K cell defect (Haller and Wigzell, 1977; Seaman et al., 1978). Although T cell, B cell, and macrophage functions remain normal following such treatments, those NK1.1+ cells which remain are nonlytic. These observations suggest that the 3 differentiation of mature NK1.1+ cells is dependent upon the presence of an intact bone marrow microenvironment. It is uncertain, however, if N K cell differentiation occurs entirely in the bone marrow, although development proceeds normally in the absence of a functional thymus (Hackett et al., 1986a). After release from the bone marrow, N K cells generally circulate in the peripheral blood or migrate to the spleen, liver, or lung (Trinchieri, 1989). The developmental relationship of N K cells to other hematopoietic cells is unclear. Although N K cells share surface antigens with both myeloid and T cells, N K progenitors can be transplanted from mutant strains of mice which lack T and B cell progenitors or myeloid progenitors (Hackett et al., 1985, 1986a). Additionally, both SCID mice and RAG -/- mice, which lack T and B cells due to a defect in the recombinase system responsible for TCR and Ig gene rearrangements, have normal numbers and functions of N K cells (Hackett et al., 1986a; Mombaerts et al., 1992; Shinkai et al., 1992). In vivo transplantation studies have shown that mature N K cells can be generated from AsialoGMl- bone marrow precursors, and that this reconstitution can be accelerated by IL-1, IL-2, and IFN-oc/(3 (Hackett et al., 1985). N K cell differentiation has also been studied in the context of in vitro cell culture assay systems. Both the bone marrow and the thymus have been found to contain N K cell progenitors which will produce mature N K cells in vitro in the presence of IL-2. In the human system, CD34+CD16CD56- bone marrow precursors differentiate into mature CD56 + N K cells which are indistinguishable from those found in the peripheral blood (Miller et al., 1994). Several lines of investigation have also detected N K cell progenitors in the thymus. One such study described a CD34+CD1CD4CD8CD56- thymic cell population which contains bipotential NK/T cell progenitors (Sanchez et al., 1994). Single cells of this phenotype were isolated, expanded for a few cell divisions, and then split into either T or N K cell culture conditions. The subsequent development of either 4 mature T or N K cells depending upon the environment, and the lack of myeloid potential of these cells, supports the hypothesis that the thymus contains committed N K / T cell progenitors. The significance of such a thymic N K cell progenitor is unclear, however, in light of the apparent irrelevance of the thymus in N K cell development. D) Cytokine regulation The majority of N K cells constitutively express the intermediate affinity IL-2 receptor chain (p75), and are therefore stimulated by high concentrations of IL-2 (Tsudo et al., 1987). Optimal stimulation of cytolytic activity occurs within 18-24 hours (Trinchieri et al., 1984), and is accompanied by de novo synthesis of the low affinity IL-2 receptor chain (p55) (Siegel et al., 1987). A high affinity IL-2 heterodimer receptor is formed by the pairing of p55 and p75 (Smith, 1988), and this complex appears to be required for the subsequent IL-2 induced proliferation. The recently characterized cytokine IL-12 is similar to IL-2 in its ability to stimulate the lysis of both NK-sensitive and NK-resistant targets (Robertson et al., 1992) . To a lesser extent interferons also augment N K cytolytic activity, although they fail to induce proliferation. This enhancement occurs within 6 hours, with greater stimulation detected by exposure to IFNoc/p than by IFN-y (Trinchieri, 1989). E) Regulation of hematopoiesis and cytokine secretion N K cells have been implicated in the regulation of hematopoiesis both in vitro and in vivo (Trinchieri, 1989). Recent advances in the understanding of NK/target cell interactions suggests that these previous findings may result from direct cytotoxic effects mediated specifically by alloreactive N K cells (Bellone et al., 1993) . Recognition of this sort will be discussed later in the context of N K cell recognition mechanisms and cell surface receptors (Sections 1.2-1.3). 5 In addition to cell mediated cytotoxicity, N K cells have been found to produce a variety of cytokines which may play a role in the regulation of other cell types. Under different experimental settings, N K cells have been shown to produce cytokines which either promote (GM-CSF) or inhibit (TNF) hematopoiesis in vitro (Pistoia et al., 1989). Activated N K cells have also been shown to produce IFN-y(Anegon et al., 1988), which may participate in ensuing adaptive immune responses by contributing to inflammatory processes and upregulating the expression of class II MHC and costimulatory molecules on antigen presenting cells. 1.1.2) NKcell mediated cytotoxicity A) Antibody-dependent cellular cytotoxicity N K cells mediate both antibody-dependent and antibody-independent cellular cytotoxicity, each through distinct cellular recognition pathways. ADCC is the process whereby a target cell which has been precoated with antibody is recognized and lysed by an N K cell. The receptor responsible for this recognition is CD16 (Perussia et al., 1984), which binds to the Fc regions of IgGl and IgG3 antibodies with low affinity (Unkeless, 1989). Because of the weakness of the interaction between CD 16 and monomeric IgG, N K cells are only activated when the antibody is aggregated on the surface of a cell. Although N K cells do not express the CD3y- or 8-subunits, the CD3£ homodimer is present on the cell surface in association with CD16 (Lanier et al., 1989b). Transfection studies have shown that in addition to their molecular association, CD3£ is a requirement for membrane expression of CD16. Activation through CD 16 initiates a series of biochemical events important during lymphocyte activation, including a rise in cytosolic calcium and the production of inositol -1,4,5 triphosphate (Cassatella et al., 1989). Antibody blocking studies have 6 demonstrated that the CD16 pathway appears to be solely responsible for ADCC mediated by N K cells (Perussia et al., 1984). B) Anti-viral activity N K cells function as a first line of defense during viral infections, and have been found to accumulate and proliferate at the site of active viral replication (Natuk and Welsh, 1987). The N K cell response peaks at 3 days post infection and is then followed by a C T L response which peaks around day 7-9 (Welsh, 1978). Many experimental approaches have been pursued which have demonstrated the role of N K cells in resistance to viral infection in mice. In SCID mice, which lack T and B cells, the depletion of N K cells results in increased viral titers and decreased survival of infected animals (Welsh et al., 1991). Mutant beige mice, which have a congenital impairment of N K cell function (Roder and Duwe, 1979), also show increased susceptibility to viral infection (Shellam et al., 1981). A role for N K cells in the control murine cytomegalovirus (MCMV), vaccinia virus, mouse hepatitis virus, herpes simplex virus, influenza virus, and coxsackievirus infections has been suggested by N K cell depletion and reconstitution experiments (Welsh and Vargas-Cortes, 1992). Studies of M C M V infection in inbred strains of mice have demonstrated that resistance is under multigenic control, with different strains having significantly different susceptibilities (Grundy et al., 1981). N K cell depletion by anti-NKl. l or anti-ASGM1 injections abrogates these intra-strain differences and results in a greatly enhanced viral replication. Genetically determined resistance to M C M V infection is controlled by both MHC and non-MHC linked genes. Mice of the H-2 k haplotype are 10 times more resistant to infection than are congenic H-2b or H-2 d mice (Grundy et al., 1981). Genetic studies have identified an autosomal dominant non-H-2 gene on the distal region of chromosome 6, termed Cmv-1, which is associated with low (C57BL/6 7 allele) or high (BALB/c allele) levels of splenic M C M V replication (Scalzo et al., 1990). Because the anti-MCMV effect is N K cell mediated, it has been postulated that the Cmv-1 gene encodes a polymorphic N K cell antigen which regulates target cell killing (Scalzo et al., 1992). As will be discussed in Section 1.3.1, independent studies have found genes linked to the Cmv-1 locus to encode polymorphic N K cell receptors involved in target cell recognition. In vitro studies of human N K cells have established their ability to distinguish between normal and virally infected cells. N K cell clones were found to specifically recognize and lyse autologous T cells infected in culture by human herpesvirus-6 (HHV-6) and to spare uninfected cells. Interestingly, this recognition event was found to be controlled by both the nature of the N K cell clone as well as polymorphic elements on the target cell (Malnati et al., 1993). Human N K cell deficiencies (in the absence of other immune dysfunctions) are extremely rare and have therefore prevented thorough investigations of the in vivo antiviral activity of human N K cells. The most compelling evidence in this field has come from case reports of isolated patients with N K cell defects (Biron et al., 1989; Fleisher et al., 1982). One of these, a teenage girl, suffered a series of independent life threatening viral infections over a 5 year period. These included primary disseminated varicella, primary cytomegalovirus pneumonia, and severe primary cutaneous herpes simplex virus infection. Analysis of her peripheral blood lymphocytes found them to be entirely CD56 and CD 16 negative, and totally devoid of N K activity either before or after IL-2 stimulation. No defects in T cells, B cells, or neutrophils were detected. This patient was treated with antiviral therapy during each infection, and eventually recovered following a normal T cell response and a rise in specific antibody titer (Biron et al., 1989). This individual case supports the hypothesis that N K cells limit the total pathogenic burden early in an infection, and T and B cells ultimately eliminate the virus. 8 C) Anti-tumor activity The importance of N K cells in the elimination of spontaneously arising tumors has been intensively studied, but remains poorly understood. Although N K cells readily lyse syngeneic transformed cell lines in vitro, fresh tumor cells are resistant to killing unless the effectors are first activated with IL-2 (Grimm et al., 1982). Experimental systems have been established in mice which measure the growth of injected malignant tumors following various immune manipulations. Initial studies reported a direct correlation between the level of N K activity and resistance to transplanted syngeneic tumors (Haller et al., 1977a). It was later demonstrated that the adoptive transfer of N K cells (but not T cells) can make immunosuppressed mice resistant to malignant cells (Barlozzari et al., 1985). Impaired N K cell function in either beige mice or in normal mice following anti-NK1.1 antibody injections also results in decreased immunity against injected tumors (Seaman et al., 1987; Talmadge et al., 1980). Although all of these studies were performed under highly experimental conditions and are unable to address the question of reactivity against spontaneous tumors, they suggest that tumor surveillance may be a function of N K cells. D) Cytolytic mechanisms The cytolytic pathways of N K cells (antibody dependent and independent) and CTLs appear to involve common terminal lytic mechanisms (Young and Cohn, 1987). The process can be divided into several discrete stages: adhesion of the effector to the target cell (Mg + + dependent), activation of the killer cell (Ca + + dependent), delivery of an irreversible lethal hit, and detachment of the killer cell from the target cell (Trinchieri, 1989). N K cells have preformed cytotoxic granules that can be discharged immediately after exposure to sensitive target cells. One of the components released during granule exocytosis, perforin, is initially inserted into the target cell membrane as a monomer. In the presence of calcium, perforin 9 polymerizes to form large cylindrical transmembrane pores that result in the osmotic lysis of the cell (Young and Cohn, 1987). In addition to pore-formation, N K cells secrete soluble toxins which induce apoptosis (programmed cell death) of the target cell (Shi et al., 1992). This pathway involves the active participation of the target cell in the production of endonucleases that degrade genomic DNA. 1.1.3) Carbohydrates and NKcell activity A) Exogenous carbohydrates Many of the early studies of the mechanisms of N K cell mediated cytotoxicity suggested that carbohydrates may be involved in target cell recognition. This was frequently evaluated by testing the ability of exogenous carbohydrates to modulate cytotoxicity in an in vitro assay. Stutman et al. (1980) initially found that the addition of several simple sugars (D-mannose, D-galactose, N-acetylglucosamine, and D-glucose) inhibited the killing of YAC-1 by murine N K cells, but did not affect cytotoxicity mediated by allosensitized CTLs. In a similar study of human N K cells, Forbes et al. (1981) found that mannose-6-phosphate, fructose-l-phosphate, and fructose-6-phosphate blocked the lysis of NK-sensitive target cell lines. This inhibitory effect was shown to occur at a postbinding stage, as demonstrated by the inability of these sugars to block effector-target conjugate formation (Ortaldo et al., 1984). Subsequent studies have demonstrated that sulphated forms of various simple sugars also inhibit N K cytotoxicity (more effectively than phosphorylated forms), and that this is also through interference with a postbinding event (Chambers and Oeltmann, 1986). B) Target cell surface carbohydrates The inhibition of N K activity observed by the addition of exogenous sugars suggested that normal interactions between effector and target cells may be 10 mediated through carbohydrates. This hypothesis has been evaluated by studying the changes in N K sensitivity of target cell lines following the modification of their cell surface carbohydrates. Attempts to directly modulate the carbohydrate structures of target cells with glycosylation inhibitors such as tunicamycin have had various and often contradictory effects on N K cell sensitivity, and are therefore difficult to interpret (McCoy and Chambers, 1991). Studies of variants of the murine lymphoma YAC-1 found that increased sialylation was associated with resistance to N K killing, while decreased sialylation was associated with N K sensitivity (Yogeeswaran et al., 1981). An NK-protective role for sialic acid has been further supported by the lysis of normally resistant autologous lymphocytes by N K cells following sialidase treatment (Rooney and Munro, 1984). Defects in oligosaccharide biosynthesis have also been associated with increased N K sensitivity in concanavalin A-resistant Chinese hamster ovary cells (Pohajdak et al., 1984) and oc(l—»2) fucosyltransferase deficient tumor cell lines (Labarriere et al., 1993). Although several additional studies have also found changes in cell surface carbohydrates to affect N K sensitivity (McCoy and Chambers, 1991), the nature and biological relevance of oligosaccharide recognition by N K cells remains unknown. 1.2) Target Cell Specificity of Natural Killer Cells 1.2.1) The Major Histocompatibility Complex A) Genomic Organization The major histocompatibility complex (MHC) is a region of highly polymorphic genes located on chromosome 17 in mice and chromosome 6 in humans. The products encoded by the genes of this complex are expressed on the surface of a variety of cell types, participate in the activation of antigen specific T 11 cells, and are the principal determinants of graft rejection (Kimball and Coligan, 1983; Lawlor et al., 1990). MHC molecules are divided into two groups (class I and II) according to their structure and corresponding function. Class I molecules are composed of a highly polymorphic a chain (-45 kD) which is encoded by the M H C and a noncovalently linked 12 kD (3 chain ( p 2 microglobulin) which is encoded by a nonpolymorphic gene located outside the MHC. The polymorphic a and (3 chains of class II molecules are of similar size (~30 kD) and structure, and are both encoded by genes located within the MHC (Trowsdale, 1993). Alleles of human leukocyte antigens (HLA) have been given numerical designations (i.e. HLA-B7), whereas alleles in mice have been designated by superscript letters (i.e. H-2Kd) (Figure 1). An MHC haplotype refers to the total set of M H C alleles present on each chromosome. Due to their homozygosity at all loci, inbred strains of mice therefore have single haplotypes. For example, an H-2 d mouse possesses the following MHC molecules: K d , D d , L d , I-Ad, and I-E d (Kimball and Coligan, 1983). Outbred populations generally have complex, heterozygous haplotypes which are not as easily categorized. In every vertebrate species analyzed, M H C genes are the most polymorphic of all known genes in the genome, with as many as 100 alleles having been identified at a single locus (Stroynowski, 1990). Class I and class II genes form two distinct multigene families. Aside from the classical MHC genes shown in Figure 1, there are many additional members of each of these gene families. These non-classical M H C molecules are generally poorly characterized and are thought to have different functions than classical M H C because of their limited polymorphism and differing expression patterns (Stroynowski, 1990). The number of genes in a given multigene family has been found to vary significantly among inbred strains of mice. As shown in Figure 1, mice of the H-2 d haplotype have three classical class I genes, as opposed to H-2b and H-2 k mice which have only two (Pullen et al., 1992). The total number of class 12 A) Human (HLA) C l a s s II D P D Q D R 11 1 1 I I 1 1 1 1 cytokines/ complement C l a s s I B C A I I 1 1 1 1 1 I I 1 1 1 1 I I 1 B ) Mouse (H-2) C l a s s I H-2 d: C l a s s II cytokines/ C l a s s I K d I I -A d I - E d 1 I I I I I complement D d L d • • 1 K k I 1 I I I I I I - A k I - E k I I I I I I 1 1 D k • 1 K b 1 1 I I I I I I -A b I - E b I I I i l l 1 D b • 1 1 I I I I I 1 Figure 1. Organization of the genes of the major histocompatibility complex. Classical class I and class II genes of the human (A) and murine (B) M H C are shown. In addition to allelic variation of genes at this locus, inbred strains of mice of different haplotypes have variable numbers of class I genes (Trowsdale, 1993; Stroynowski, 1990). 13 I and class I-like genes in a given strain of mice varies between 20-50 per haploid genome. B) Structure and function of Class I MHC Class I and class II molecules are fundamentally similar in that they both bind short peptides which, in association with the MHC molecule itself, are recognized by antigen specific T cell receptors. Mature T cells are activated only following the recognition of a cell in which a non-self peptide is presented in the context of a self MHC molecule. Class II molecules are expressed on a few restricted cell types (B lymphocytes, macrophages, dendritic cells, and endothelial cells), and function by presenting peptides of endocytosed protein antigens to CD4+ T cells (Trowsdale, 1993). In addition to T cell activation, class I molecules appear to have a central role in N K cell recognition (discussed in Sections 1.2.2-1.2.4). These antigens are expressed on virtually all nucleated cells in the body, and provide the immune system with a means by which a cell which is producing a non-self protein antigen can be identified. The ocl/a2 domain (Figure 2) of class I M H C forms a platform of eight strands of (3-pleated sheets which support two parallel strands of a helix. The a-helices form the side of a cleft and the (3 sheets the floor, which together bind a 10 to 20 amino acid fragment of a processed protein (Bjorkman et al., 1987; Madden et al., 1991). The ccl/a2 domain is the polymorphic region of a class I molecule, and this polymorphism determines which peptides will be presented. The epitope recognized by the T cell receptor is contributed by both the processed peptide and sequences in the ocl/a2 domain (Allen et al., 1984). The nonpolymorphic oc3 domain is recognized by the CD8 co-receptor on T cells. All class I molecules have a conserved N-linked glycosylation site in the al/a2 domain at amino acid position 86 (Figure 2). Murine class I molecules have an additional N-linked glycan at position 176 (Kimball and Coligan, 1983). 14 TCR Figure 2. Schematic diagram of a class I major histocompatibility complex molecule. Oligosaccharide units are identified by T (asparagine 86; all class I molecules) and ? (asparagine 176; all murine class I molecules). The processed peptide is represented by I, and the domains recognized by CD8 (oc3) and TCR (otl/a2) are indicated by brackets (Bjorkman et a l , 1987; Kimball and Coligan, 1983). 15 Class I molecules present peptides of endogenously synthesized proteins to CD8+ T cells. This system allows for the identification of aberrant cells which are internally producing antigens such as viral proteins (Townsend et al., 1986). Both CD8 + and CD4+ T cells are educated to be self tolerant and self M H C restricted. This results a repertoire of T cells which recognizes non-self peptides presented by self MHC molecules. Recognition of this sort by an activated CD8 + T cell results in a delivery of a "lethal hit" and the subsequent death of the conjugated target cell (Trowsdale, 1993). 1.2.2) Hybrid resistance A) Immunogenetics of bone marrow graft rejection According to the classical laws of transplantation, a donor graft will not be rejected if it contains only antigens which the recipient immune system recognizes as self. Although true of solid tissue grafts, the rule is often broken in the case of hematopoietic grafts. This phenomenon has been extensively studied in the mouse by use of an experimental system which follows the survival of a donor's marrow when transplanted into a lethally irradiated recipient. Cudkowicz and Stimpfling (1964) used such a system whereby the donor marrow was derived from a homozygous inbred mouse and the recipient was an F i hybrid offspring. In these studies the strains of mice used were C57BL/6 (H-2b), C3H (H-2k), and (C57BL/6xC3H)Fi (H-2b/k). Although the F i hybrid expresses all histocompatibility antigens of both parents, B6 marrow was rejected by the (C57BL/6xC3H)Fi recipient (Table IA). Backcrossing F i hybrids to parentals and further transplantation demonstrated that heterozygosity at the H-2 locus was required for the rejection. This unexpected phenomenon was therefore called hybrid resistance and the relevant locus was named hybrid histocompatibility-1 (Hh-1). 16 Table 1. Immunogenetics of Bone Marrow Graft Rejection Donor Recipient Fate of Graft A) B) H-2b H-2b /k rejected H-2k H-2b /k rejected H-2b /k H-2b /k accepted H-2b H-2b / k rejected H-2d H-2b /k rejected H-2b /d H-2b/k accepted The experiments summarized in A) (Cudkowicz and Stimpfling, 1964) and B) (Bennett, 1972) describe the fate of murine bone marrow transplanted into lethally irradiated recipients (accepted vs. rejected). The donors and recipients in these studies were inbred strains of mice, and the haplotypes of each are provided. 17 Further studies found that donor H-2 heterozygosity can permit the acceptance of a BMC graft in an F i hybrid even if the two differ at the H-2 locus (Bennett, 1972). This was demonstrated by the acceptance of an H-2 b / d allogeneic graft in an H-2 b / k recipient, although each of the parental homozygous H-2 b and H-2d grafts were rejected (Table IB). The interpretation of these combined observations was that there must be homozygosity at the Hh-1 locus in order for the theoretical Hh-1 antigens to be expressed and thus allow for recognition and elimination of the graft. Although this hypothesis has long been used to explain the peculiarity of hybrid resistance, it relies upon the assumption that graft rejection must result from the recognition of non-self antigens. As will be discussed in Section 1.2.3, elimination of cells lacking the appropriate self antigens may be an equally valid explanation of these phenomena. B) Effector cells The effector cells of hybrid resistance were shown to be marrow dependent prior to the discovery of N K cells. Destruction of the bone marrow microenvironment by 8 9 Sr administration was first found to abrogate hybrid resistance (Bennett, 1973), and only later shown to cause a selective loss of N K cell function (Haller and Wigzell, 1977). A role for N K cells in this process was further supported by the poor allograft rejection demonstrated by N K cell impaired Beige mice (Harrison and Carlson, 1983). Additionally, the depletion of N K cells by the in vivo administration of anti-ASGMl or anti-NKl.l antibodies also resulted in an inability to reject marrow allografts (Lotzova et al., 1983; Saito et al., 1984). Although found mainly on N K cells, both ASGM1 and NK1.1 are also expressed by a small percentage of T cells. In order to examine N K cells in isolation from T cells (and B cells), SCID mice were tested as recipients of allogeneic marrow. The N K cells of these mice are both phenotypically and functionally normal, whereas they lack both T and B cells 18 (Hackett et al., 1986b). SCID mice were found to reject marrow according to the peculiar laws of hybrid resistance. As in normal mice, this could be abrogated by N K cell depletion with anti-ASGMl (Murphy et al., 1987). Additionally, the transfer of SCID N K cells into an N K cell depleted lethally irradiated recipient was shown to transfer the ability to reject donor marrow (Murphy et al., 1990). These experiments demonstrated that N K cells alone can mediate hybrid resistance. Sentman et al. (1989) provided the first indication that distinct subsets of N K cells mediate bone marrow graft rejection. Characterization of a panel of anti-N K cell antibodies found one of these, 5E6, to be expressed by approximately 50% of B6 N K cells. In vivo administration of the 5E6 antibody resulted in the selective depletion of the 5E6 + N K subset and was therefore used to evaluate the role of these cells in allogeneic resistance. Whereas a B10.BR (H-2k) recipient rejected H-2b grafts in the presence and absence of 5E6+ N K cells, H-2 d grafts were accepted only when 5E6+ cells were eliminated. This demonstrates that this N K cell subset is required for the elimination of H-2d, but not H-2b, grafts (Sentman et al., 1989; Yu et al., 1992). The importance of N K cells in these processes, regardless of donor haplotype, is demonstrated by the acceptance of both grafts following the depletion of all NK1.1+ cells. 1.2.3) Class I MHC and NKcell recognition A) Mouse N K cells differ from T cells in that class I M H C antigens need not be present on a target cell for an N K cell to recognize it and lyse it. N K cell killing has therefore long been called MHC unrestricted cytotoxicity, as a description of this lack of requirement for syngeneic class I MHC. The term can be misleading, however, because N K cells actually appear to recognize class I molecules, although in a manner distinct from and possibly complementary to T cells. 19 One of the earliest clues into the nature of N K cell recognition came from an investigation by Karre et al. (1986) into murine lymphoma variants mutagenized in vitro and selected for loss of H-2 expression. Three independent class I negative sublines were established from two H-2b tumor cell lines. Both of the wild-type lines were N K resistant (in vitro) and highly malignant when injected into syngeneic hosts (in vivo). The three H-2 loss variants, however, were all found to be N K sensitive as well as entirely non-tumorigenic when injected in mice. These observations suggested that class I molecules may provide a cell with protection from N K recognition and elimination. This hypothesis was directly tested on a mutant of the lymphoma EL-4 expressing low to undetectable levels of class I on the cell surface due to a defect in [32m (Glas et al., 1992). As with the previous set of mutants, the wild-type line ((32m+ / class I+) was highly tumorigenic, whereas the H-2 loss mutant was not. Transfection of the (32m gene restored both class I expression as well as the tumorigenic phenotype. The in vivo clearance of the H-2 loss mutant was demonstrated to be N K cell mediated, in that depletion of NK1.1+ cells resulted in tumorigenicity levels equal to the wild-type. The role of class I in N K cell education and recognition has been further studied in two genetically altered strains of mice. The first of these, D8, is a C57BL/6 (H-2b) mouse transgenic for H-2D d (Ohlen et al., 1989). These mice express the transgene in all tissues in a manner identical to that of endogenous H-2b antigens. Bone marrow transplantation into irradiated recipients has shown that N K cells of D8 mice acquire the ability to reject C57BL/6 marrow, whereas they are unable to reject self (D8). This experiment demonstrated that N K cells can be educated in such a way that they will see as foreign a cell which is absolutely identical to self in every way except for the absence of a single class I molecule (Dd). D8 mice were also evaluated as graft donors in a series of marrow transplantation experiments. C57BL/6 marrow was rejected by N K cells in both B10.D2 (H-2d) and (B10.D2 x C57BL/6)Fi mice, whereas D8 marrow was not. The 20 presence of a single class I molecule (Dd) can therefore also protect a graft from recognition and elimination by N K cells educated in an H-2 d mouse. The second important strain of mice contains a targeted mutation in the pV microglobulin gene (Bix et al., 1991; Hoglund et al., 1991; Liao et al., 1991). These mice are almost entirely devoid of class I molecules on the cell surface, with the vast majority of a chains accumulating in intracellular compartments, p2m -/-hematopoietic grafts were found to be rejected by N K cells when transplanted into P2m +/+ hosts. Additionally, p2m -/- T cell blasts were lysed by 02m +/+ N K cells in vitro. This serves as conclusive evidence that, even in otherwise normal cells, the absence of class I MHC makes a cell susceptible to N K cell elimination. Interestingly, P2m -/- mice have normal numbers of N K cells, but demonstrate no hematopoietic abnormalities and show no self-reactivity in either in vitro cytotoxicity assays or in vivo transplantation studies. This demonstrates that N K cells are educated to be tolerant not only to a specific repertoire of self class I molecules, but also in the absence of such antigens. Bone marrow chimera studies have shown that the elements which define the N K cell education process are bone marrow derived (Hoglund et al., 1991), as opposed to T cells whose education is dictated by the recipient thymus. B) Human In vitro studies in humans have closely paralleled simultaneous discoveries in mouse. Many laboratories have described an inverse correlation between target cell class I expression and N K sensitivity. This was initially documented in H L A loss sublines which showed increased N K sensitivity compared to the HLA+ wild-type (Harel-Bellan et al., 1986; Storkus et al., 1987). Cytokine induced restoration of class I expression was consistently accompanied by an acquisition of resistance to NK-mediated lysis. As a more formal proof of this phenomenon, H L A genes were transfected into a class I-deficient B cell line (Storkus et al., 1989a). HLA-A 21 and -B alleles were found to provide protection from N K cells, whereas H-2 genes were not. Chimeras between HLA-B7 and H-2DP were constructed to exploit the structural similarities and functional differences between these molecules. Expression of these constructs showed that the ocl/a2 domain (see Figure 2) is responsible for the protection mediated by the H L A molecule (Storkus et al., 1989b). Subsequent analysis demonstrated that some, but not all, HLA-A alleles were protective in this system. Analysis of amino acid differences between protective and non-protective alleles and subsequent site-directed mutagenesis identified position 74 in the a l domain as crucial for HLA-mediated N K resistance (Storkus et al., 1991). The change of a histidine to asparagine was sufficient to render a non-protective allele protective. In a separate line of investigation, Ciccone et al. (1988) made the pivotal discovery that T-cell depleted peripheral blood cell populations can acquire the ability to lyse allogeneic cells following activation in a mixed lymphocyte reaction. The resulting CD3-/TCR7CD7+/CD2+ clones were found to be highly cytotoxic against both the stimulating allogeneic blasts as well as the standard N K target K562, but not against autologous blasts. Analysis of a specific N K clone revealed that, in addition to the stimulating donor, it is capable of lysing cells from some, but not all, allogeneic donors. To understand the target cell elements which are responsible for 'susceptibility to lysis' by this N K clone, inheritance of this character was studied in a large family spanning 3 generations and 38 members. Susceptibility was found to be inherited in an autosomal recessive fashion, whereas resistance was a dominantly inherited character (Ciccone et al., 1990a). Detailed genetic analysis demonstrated that the locus responsible for these characteristics maps to the class I region of the M H C (Ciccone et al., 1990b). 22 1.2.4) The 'missing self hypothesis The experiments in both mouse and human provided a significant amount of data to suggest that the presence of the appropriate class I molecule on a target cell can protect it from lysis by an N K cell. This led Klas Karre to formulate a mechanistic explanation for N K cell recognition known as the 'missing self hypothesis. Two possible descriptions of the inhibitory nature of class I molecules were proposed (Ljunggren and Karre, 1990; Moretta et al., 1992). The first of these, the effector inhibition model, states that an N K cell receptor engages a class I molecule on a target cell and this interaction delivers an inactivating signal that blocks lysis (Figure 3A). This explains studies in which class I negative/NK sensitive cell lines become N K resistant following the transfection of a class I or (32m gene. An N K cell would therefore be educated to recognize and scan for self class I molecules, and would kill any cell unable to deliver the appropriate negative signal. Accordingly, N K cells of D8 mice are proposed to receive negative signals from H-2D d, and reject C57BL/6 bone marrow because they are deficient in this regard. Hybrid resistance may be explained in a similar manner: homozygous H-2 b / b or H-2 k / k grafts are rejected in an F i host (Table IA) because both lack the full complement of self class I molecules (H-2b/k) and are therefore unable to deliver the full range of negative signals required by an F i N K cell. This model requires the presence of accessory molecules which trigger the effector cell and will result in target cell lysis if an overriding negative signal is not received. The second model, target interference, interprets class I mediated protection in terms of an activating N K cell receptor (Figure 3B). The ligand for such a receptor may normally be masked by class I MHC, thereby preventing N K cell activation. When class I is absent (downregulation) or unable to properly mask the target structure (allogeneic cells), then the receptor-ligand interaction will occur and lysis will follow. This sufficiently explains all previous observations, but 23 A) Effector Inhibition NK B ) Target Interference © /Targetf / (MHC D • Recognition Lysis yes no no yes no yes no yes Figure 3. The 'missing self hypothesis. In the effector inhibition model (A), a dominant negative signal is proposed to result from the interaction of an N K cell receptor (NKCR) with class I MHC (with an activating signal mediated through accessory molecules). In the target interference model (B), class I molecules are proposed to mask an NKCR from interacting with its activating ligand (^H). Adapted from Ljunggren and Karre (1990) and Moretta et al. (1992). 24 requires that individual class I molecules have differing abilities to mask different target structures. Only under such circumstances could the lysis of normal allogeneic cells and their protection by self class I be explained. 1.3) Candidate Natural Killer Cell Receptors 1.3.1) Mouse A) Ly-49 Ly-49 is a family of molecules with highly homologous amino acid and nucleotide sequences. The first member of this family to be characterized was Ly-49A, as identified by the antibodies YE1/32, YE1/48, A l , and JR9-318 (Nagasawa et al., 1987; Roland and Cazenave, 1992; Takei, 1983). FACS analysis showed the antigen (a disulfide-linked dimer of 45 kD subunits) to be expressed at high levels on several transformed T cell lines (MBL-2 and EL-4), but undetectable on thymocytes, spleen cells, bone marrow cells, or mitogen-stimulated spleen cells (Chan and Takei, 1986, 1988; Nagasawa et al., 1987). Molecular cloning revealed Ly-49A to be a type II integral membrane protein with an extracellular domain homologous to the carbohydrate recognition domain of C-type animal lectins (Chan and Takei, 1989; Yokoyama et al., 1989). Although expression of Ly-49 A on various normal cell types of hematopoeitic origin was undetectable by flow cytometry, small quantities of the antigen could be successfully immunoprecipitated from spleen cells and thymocytes (Chan and Takei, 1986; Nagasawa et al., 1987). This suggested that the molecule is either expressed at high levels by a small number of cells or that many cells express it at low levels. Splenic subpopulation analysis proved the former to be true, in that approximately 20% of NK1.1+ cells co-express Ly-49A (Yokoyama et al., 1990). This population therefore makes up less than 1% of total spleen cells. 25 Karlhofer et al. (1992) examined the potential functional significance of this N K cell subset expression by sorting N K cells into Ly-49A+ and Ly-49A- fractions, and testing each in cytotoxicity assays against an extensive panel of tumor targets. Although all of the cell lines tested were sensitive to lysis by Ly-49A- N K cells, many were resistant to the Ly-49A+ subset. Target cell resistance was found to correlate with its haplotype, such that H-2 d and H-2 k cell lines were resistant to Ly-49A+ N K cells, whereas H-2 b targets were lysed equally well by both fractions (Figure 4). In light of the 'missing self hypothesis, it was hypothesized that H-2 d and H-2k antigens were providing protection to these tumor targets against Ly-49A+ N K cells, possibly through an interaction with Ly-49 A itself. This possibility was tested by establishing transfectants of an H-2 b cell line expressing cDNAs encoding either H-2D d, K d , or L d (Karlhofer et al., 1992). Although the K d and L d transfectants were lysed in the same manner as the parental line, the D d transfectants became resistant to killing by the Ly-49A+ subset (sensitivity to lysis by Ly-49- N K cells was unaltered). This transfected resistance could be abrogated by antibodies against either Ly-49A (Al) or H-2D d. Only an antibody against the ocl/oc2 domain of D d restored killing (Figure 4), whereas one recognizing the a3 domain had no effect (see Figure 2). This is consistent with previous work implicating the ocl/oc2 domain as the protective portion of a class I molecule. Although Karlhofer's experiments suggested that H-2D d is an inhibitory ligand for Ly-49A, it remained possible that this class I molecule functions by preventing Ly-49A from recognizing an activating ligand on the target cell (masking hypothesis). Kane (1994) directly addressed this question by testing the ability of an Ly-49A+ lymphoma (EL-4) to bind to a panel of purified class I molecules immobilized on a plastic surface. This work demonstrated that EL-4 bound specifically to D d and D k and that this adhesion was inhibitable with the anti-Ly-49A antibody A l (binding of EL-4 to K 3 was also detected, although 26 C57BL/6 NK cells 0 o o o o 0 0 n 0 ° (Cytotoxicity Assays] (Lysis: +/-) Tumor Targets Lv-49A+ Lv-49A~ H-2b + H-2d H-2k H-2b + K d H-2b + L d H-2b + Dd* + + + + Figure 4. H-2 specificity of Ly-49A + N K cells. C57BL/6 N K cells were sorted into Ly-49A+ and Ly-49A- fractions, and tested for cytolytic activity against tumor cell lines of various haplotypes (H-2b, H-2 d, H-2 k, and H-2 b transfected with individual class I MHC cDNAs). The transfected resistance of the H-2 b + D d cell line (*) was abrogated by anti-Ly-49A or anti-Dd (al/a2 domain) antibodies (Karlhofer et al., 1992). 27 apparently not through Ly-49A). The finding that Ly-49A physically interacts with class I molecules lends strong support to the hypothesis that missing self recognition is in fact the result of an inhibitory interaction between an N K cell receptor and a class I molecule. The Ly-49A+ N K cells used in the experiments of Karlhofer et al. (1992) were derived from an H-2 b mouse (C57BL/6), whereas the protective class I molecules were of H-2 d and H-2 k origin (non-self). In consideration of previous studies which have shown that the class I makeup of a mouse can dictate the N K cell education process, it was interesting to examine the nature of Ly-49A+ N K cells in H-2 congenic mice. Whereas approximately 15% of NK1.1+ cells of BIO mice (H-2b) co-express Ly-49A, the antigen was undetectable on N K cells of B10.D2 (H-2d), B10.BR (H-2k), or B10.A (H-2a) mice (Karlhofer et al., 1994; Sykes et al., 1993). Although paradoxical, it nonetheless appeared that in mice expressing its inhibitory ligand, Ly-49A expression was either downregulated or the entire N K cell subset itself was deleted. Southern analysis of genomic DNA hybridized with a radiolabeled probe derived from the Ly-49A cDNA revealed multiple cross-hybridizing bands (Chan and Takei, 1989; Yokoyama et al., 1989), allowing for the prediction that the genome contains multiple Ly-49-related genes. Interestingly, the pattern of these bands differed between certain inbred strains of mice, suggesting polymorphism within the gene family. The RFLP patterns were studied in recombinant inbred strains of mice, and the entire gene family was mapped to the distal portion of chromosome 6, in the vicinity of the proline-rich protein locus (Yokoyama et al., 1990). To further characterize the multigene family, two related Ly-49 cDNAs were isolated from a lung cDNA library (Wong et al., 1991). The predicted proteins encoded by Ly-49B (288 a.a.) and Ly-49C (266 a.a.) were found to be 52 and 65 percent identical to Ly-49A (262 a.a.) respectively. Sequence analyses showed that 28 all three of the Ly-49s contained an extracellular lectin-like domain sharing 18-25 percent amino acid identity with other members of the C-type lectin superfamily. Although Ly-49 molecules lack several of the residues conserved in the majority of C-type CRDs, all of the four invariant cysteines residues of this domain are present. Genomic cloning of Ly-49A (Kubo et al., 1993) has shown the gene to contain 7 exons spanning approximately 19 kilobases, with exons 5, 6 , and 7 encoding exclusively and entirely the lectin-like domain. This is significant from an evolutionary perspective in that all known CRDs are encoded by either one (mannose binding proteins, selectins) or three (hepatic lectins, CD23) exons (Bezouska et al., 1991). Those CRDs containing three exons have conserved intron positioning nearly identical to that of Ly-49A. Additional CRD-encoding genomic clones of Ly-49-related genes have also been isolated (Wong et al., 1991), all containing the same intron-exon boundaries of Ly-49A. These clones also serve as evidence that there are additional, and as yet uncharacterized, members of the Ly-49 multigene family. B) 5E6 The 5E6 antigen was identified in studies designed to define N K cell subsets that mediate hybrid resistance. 129/J (H-2b) mice were immunized with purified NK1.1+ cells from C57BL/6 (H-2b) mice, and hybridoma clones were subsequently isolated that produced antibodies which bound exclusively to N K cells (Sentman et al., 1989). An antibody designated 5E6 met these criteria in that it identified -50% of both fresh and IL-2 propagated NK1.1 + cells, and 5E6 + and 5E6- N K cells were equally lytic against the standard murine N K target YAC-1. The molecule precipitated by the antibody is a dimer composed of 54 kD subunits. As had been shown previously, in vivo depletion of N K cells by means of anti-NKl.l injections resulted in the loss of both N K activity and the ability of an 29 irradiated hybrid mouse to reject parental bone marrow grafts. Injection of anti-5E6 was found to have a more specific effect in that it eliminated the ability of a hybrid or allogeneic mouse to reject H-2d, but not H-2b, bone marrow (Sentman et al., 1989; Yu et al., 1992). Since the recipients tested in this system normally reject both H-2 d and H-2 b grafts, it follows that 5E6 + N K cells are required for the rejection of the former but not the latter. Adoptive transfer of N K cell subsets has in fact shown that 5E6 + cells are neither required for the rejection of H-2 b grafts, nor do they participate in this process (Sentman et al., 1991). Although a fine H-2 specificity has been demonstrated for 5E6+ N K cells in vivo, they show no difference from 5E6- N K cells in the lysis of H-2 d and H-2 b tumor targets in vitro. ONKR-Pl I NK1.1 NKR-P1 is the most useful cell surface marker of rat N K cells. The monoclonal antibody 3.2.3 was produced by a strategy that screened for antibodies which affect the cytolytic activity or proliferation of N K cells (Chambers et al., 1989). This antibody, which recognizes all rat N K cells, was found to enhance the lysis of target cells that express FcR and to induce the exocytosis of granules from N K cells. Studies utilizing an NKR-P1+ N K cell line (RNK-16) demonstrated that F(ab')2 fragments of 3.2.3 stimulate phosphoinositide turnover and a rise in intracellular calcium in the cell line (Ryan et al, 1991). This rise was augmented by cross linking of the F(ab')2 antibody. These findings gave further support to the notion that NKR-P1 may be an activating receptor on N K cells. The initial biochemical studies of the NKR-P1 protein found it to be a 60 kD dimer composed of two 30 kD chains (Chambers et al., 1989). Subsequent molecular cloning revealed it to be a type II transmembrane protein with significant homology to the C-type lectin supergene family (Giorda et al., 1990). Although this is the same superfamily to which Ly-49 belongs, NKR-P1 and Ly-49 30 are clearly not from the same multigene family. This point will be discussed in more detail later. In order to study this interesting molecule in a more well-defined experimental system, two groups succeeded in cloning the murine homologue of NKR-P1. Giorda and Trucco (1991) isolated three related cDNAs (gene-2, gene-34, and gene-40), thereby demonstrating that NKR-P1 is actually a family of related molecules. Yokoyama et al. (1991) isolated a single cDNA identical to that of gene-2. All of the mNKR-Pl cDNAs were found to be highly homologous to each other as well as to rNKR-Pl at both the nucleotide and amino acid levels. Northern analysis showed that these transcripts are highly expressed in IL-2 activated N K cells of C57BL/6 mice, but undetectable in other normal cell types and N K cells derived from several other strains of mice (Giorda et al., 1992). Although negative for mNKR-Pl on a Northern blot, RT-PCR demonstrated that BALB/c mice express at least a minimal level of these transcripts. A genomic clone for gene-2 was isolated and shown to have the same intron-exon organization as Ly-49A, with the lectin-like domain encoded by three exons (Giorda et al., 1992). Genomic southern analysis found that the mNKR-Pl family, also like Ly-49, identifies several RFLP variants in inbred strains of mice (Yokoyama et al., 1991). The distribution of these patterns among strains of mice was noted to be highly similar to that of Ly-49, suggesting a possible genetic linkage. Mapping of NKR-P1 in recombinant inbred mice placed the gene family on the distal portion of chromosome 6, within 0.5 centi-Morgans of the Ly-49 locus (Yokoyama et al., 1991). Several lines of evidence suggested that the NK1.1 antigen may be encoded by a member of the NKR-P1 family. In addition to its expression on all murine N K cells (like NKR-P1 in rat), the anti-NKl.l antibody PK136 induces the lysis of otherwise resistant targets by fresh or IL-2 activated N K cells (Karlhofer and Yokoyama, 1991). NK1.1 has also been shown to have a dimeric structure, strain-31 specific expression, and map to the proline-rich protein locus on chromosome 6. Multiple mNKR-Pl cDNAs were therefore expressed and tested for reactivity with the antibody PK136. When expressed in insect cells, the mNKR-Pl.9 cDNA produced a 56 kD homodimer reactive with the antibody by flow cytometry (Ryan et al., 1992). mNKR-Pl.9 is identical to the gene-40 cDNA reported by Giorda and Trucco (1991), except for a single amino acid difference and a three amino acid deletion in the extracellular domain of mNKR-Pl.9. Transfection of the gene-2 cDNA in insect cells did not produce the NK1.1 epitope, and the gene-34 cDNA has not been tested in this system. These studies have revealed several important similarities between the Ly-49 and NKR-P1 multigene families: 1) membership in the C-type lectin supergene family; 2) expression as type II integral membrane proteins; 3) disulfide-linked dimeric structure; 4) selective expression on N K cells; and 5) chromosomal location. It must be emphasized, however, that Ly-49 and NKR-P1 are distantly related gene families (see dendogram in Figure 5). Amino acid comparisons between members of the two families show that similarity is restricted to the putative CRD (-25% identity), and that there is no significant homology at the nucleotide level. Within each of the gene families, however, high levels of similarity are maintained throughout the entire reading frame, with members sharing 52-87% of amino acids and 70-90% of nucleotides. D)NK2.1 In an attempt to identify allotypic N K cell surface structures, 129/SvJ mice were immunized with purified N K cells derived from C57BL/6 mice (Lemieux et al., 1991). One of the hybridomas produced by the subsequent fusion, 4L03311, secreted an antibody reactive with an N K cell surface antigen designated NK2.1. Although B6 mice express both the NK1.1 and NK2.1 epitopes, all other strains tested were found to be either NK1.1+NK2.1-, NK1.1NK2.1+, or NK1.1-NK2.1-. 32 Ly-49A Ly-49C Ly-49B NKG2-C NKG2-A/B mNKR-Pl (2) mNKR-Pl (40) mNKR-Pl (34) rNKR-P l NKG2-D F i g u r e 5. Sequence re latedness of N K ce l l lect in- l ike p ro te ins . The entire amino acid sequence of each of the members of the Ly-49, N K R - P 1 , and N K G 2 families was used in the generation of the dendogram by the G C G program Pileup. 33 Functionally, F(ab')2 fragments of the anti-NK2.1 antibody were found to enhance the lysis of NK-sensitive targets, and immobilization of the antibody induced granule exocytosis by IL-2 activated N K cells (Morelli and Lemieux, 1993). These observations suggest that NK2.1 may be an activating structure. The antigen is a 130 kD disulfide-linked dimer composed of two extensively glycosylated 65 kD subunits (-32 kD following the removal of N-linked carbohydrates) (Gosselin et al., 1993). E) LGL-1 The LGL-1 molecule, although poorly characterized, has some significant similarities to all of the N K cell antigens thus far described. The antibody, 4D11, was produced by immunizing rats with LGL-enriched spleen cells from C57BL/6 mice (Mason et al., 1988). Hybridomas were selected which secreted antibodies reactive with N K cells but not with thymocytes. LGL-1 was found to be expressed by 50% of splenic NK1.1+ cells as well as by rare CD3 + cells. When expression on total spleen was examined, only 2% of cells were found to express the molecule, suggesting that the antigen is almost exclusively restricted to an N K cell subset. Functionally, 4D11 was able to induce killing of an otherwise resistant FcR + target by IL-2 activated, but not fresh, LGL-1+ N K cells (Mason et al., 1994). Biochemical characterization revealed the protein to be an 87 kD disulfide-linked dimer. LGL-1 has recently been cloned and was found to belong to the Ly-49 gene family (Mason et al., 1995). This will be discussed further in Chapters 4 and 6. F) 2 B 4 2B4 is 66 kD monomeric protein, and molecular cloning has revealed it to be a novel member of the Ig supergene family, most closely related to CD48 and LFA-3. The 2B4 gene maps to chromosome 1, and Southern blot analysis showed 34 hybridization of the cDNA to multiple bands, suggesting that there may be additional 2B4-related genes (Mathew et al., 1993). In freshly isolated spleen cells of C57BL/6 mice, expression of the 2B4 antigen mirrors that of NK1.1. They are both expressed by all N K cells as well as 4-7% of CD3 + cells, and all N K activity is contained within the NK1.1V2B4+ fraction. Following culture of total spleen cells in IL-2 for 7 days, 12-25% of CD3 + cells co-express NK1.1, whereas 20-40% co-express 2B4. At this stage, some CD3+NK1.1- can kill targets without self MHC restriction, whereas all of this type of activity is still contained within the CD3 +2B4 + fraction. 2B4 therefore appears to be a marker of "NK-like" activity (Garni-Wagner et al., 1993). Although the function of 2B4 is unknown, it has been proposed to be an activating structure. The anti-2B4 antibody greatly augments the lytic activity of cultured N K cells against both FcR + and FcR- targets. The concentration of antibody required for activation (5 ng/ml) is much lower than that required to detect activation by anti-NKl.l (25 Ltg/ml). Both the secretion of IFN-y and granule exocytosis are also induced by the anti-2B4 antibody (Garni-Wagner et al., 1993). 1.3.2) Human A) GL183 and EB6 A crucial step in understanding the recognition mechanisms of human N K cells was the discovery that alloreactive N K cell clones can be generated in mixed lymphocyte culture from CD3 - lymphocytes (Ciccone et al., 1988). These N K clones were characterized by their ability to lyse cells derived from the stimulating donor and some but not all other allogeneic donors. To identify N K cell receptors, mice were immunized with clones generated by this process, and hybridoma cell lines were selected which secreted antibodies that modulated N K cell cytolytic function. 35 This strategy yielded two monoclonal antibodies, GL183 and EB6, both of which identify subsets of N K cells (Moretta et al., 1990a, 1990b). Flow cytometric analysis of a panel of donors found the antigens recognized by each of these antibodies to be present on 1-13% of peripheral blood lymphocytes, with significant variation among individuals. N K clones were found to be either single positive, double positive, or double negative when these antibodies were examined in combination. The proportion of each type of population was also seen to vary significantly among donors. Biochemical characterization of the antigens recognized by GL183 and EB6 demonstrated that they are both 58 kD molecules (Moretta et al., 1990a, 1990b). Comparison of the GL183 and EB6 peptide maps demonstrated that although most of the major peptides showed identical migration, there were some peptides unique to each. This strongly supports the hypothesis that these two structures are molecularly related. Further analysis of clonal N K cell populations has suggested that the p58 molecules are functional N K cell receptors. N K cell clones were generated from CD3- CD16+ peripheral blood lymphocytes (PBL) in the absence of mixed lymphocyte culture and tested for reactivity against PHA blasts derived from 7 allogeneic donors (Figure 6). Clones could be placed into five distinct groups based upon their ability to kill the allogeneic targets. Interestingly, all clones sharing the same specificity were homogeneous in their expression of GL183 and EB6 (double positive, double negative, or single positive). All groups were characterized by their reactivity against some non-self blasts and their inability to lyse self and certain non-self targets (Ciccone et al., 1992a; Moretta et al., 1990b). As discussed earlier (Section 1.2.3), resistance to killing by these N K cell clones was shown to be a dominantly inherited character, whereas susceptibility is recessively inherited (Ciccone et al., 1990a). Genetic mapping identified the class I region of the MHC to be responsible for these characters (Ciccone et al., 1990b). 36 CD3CD16+ PBL (Donors: A, 1) limiting dilution (PHA + IL-2) f NK Cell Clones 0 0 0 0 0 0 0 0 0 0 0 0 Test for Lytic Activity (against PHA blasts: A, 1, 2, 3, 4, 5, 6) T Assign to Groups Based Upon Targets Lysed Group Donor Targets Phenotype Inhibited Lysed By 1 (NKI) 1 A, 6 GL183-EB6+ HLA-C (Cw4) 2 (NK2) A 1,2, 3,5 GL183 +EB6 + HLA-C (Cw3) 3 A 1, 2, 4, 6 GL183-EB6-4 A 1, 2,3,4 GL183+EB6+ 5 A 6 GL183-EB6+ Figure 6. Alloreactivity of human NK cell clones. CD3CD16+PBL were cloned and tested for lytic activity against self and non-self P H A blasts. N K cell clones and or PHA blasts were generated from donors designated A, 1,2, 3, 4, 5, and 6. Patterns of alloreactivity fell into five distinct groups, and each specificity was homogeneous in its expression of GL183 and EB6 (Ciccone et al., 1992a; Moretta et al., 1990b). Groups 1 and 2 were inhibited by specific HLA-C alleles (Ciccone et al., 1992b; Colonna et al., 1993b). 37 Clones of specificity 1 and 2 were found to be reciprocally associated with diallelic polymorphisms in HLA-C at amino acid positions 77 and 80 (Colonna et al., 1992). Moreover, clones of specificity 1 could be generated from individuals homozygous for HLA-C (Asn7 7-Lys8 0) following stimulation with target cells homozygous for HLA-C (Ser77-Asn80), and clones of specificity 2 were generated by a reciprocal stimulation (Colonna et al., 1993a). This suggested that either HLA-C or a closely linked gene dominantly prevents or recessively permits recognition by N K clones of specificity 1 and 2. Transfection of HLA-C alleles into target cell lines proved that class I molecules were providing dominant protection from lysis by certain N K clones (Ciccone et al., 1992b; Colonna et al., 1993b). This work made use of class I deleted cell lines that are sensitive to killing by both NK1- and NK2- specific N K cell clones. Transfection of HLA-C alleles encoding Asn 7 7 -Lys 8 0 (HLA-Cw4, -Cw5, and -Cw6) provided targets with protection against NKl-specific clones, but not NK2. Alternatively, transfection with HLA-C alleles encoding Ser 7 7-Asn 8 0 (HLA-Cwl, -Cw3, -Cw7, and -Cwl3) provided specific protection from NK2-specific effectors (Figure 6). Although the transfection studies demonstrated the protective nature of specific class I M H C to a target cell, it remained unclear whether N K cell receptors recognized class I molecules or a structure masked by them. In light of the correlation found between GL183/EB6 expression on N K cell clones and their class I specificity, a receptor function for these molecules was proposed. Antibody blocking studies demonstrated that F(ab')2 fragments of the GL183 or EB6 antibodies were sufficient to restore lysis of cells that had acquired protection from N K cell lysis by transfection of an HLA-C allele (Moretta et al., 1993). It was therefore hypothesized that these molecules function as receptors for HLA-C, and that antibodies block their inhibitory function and restore killing. This was supported by studies using a class I7FcR+ cell line which is efficiently lysed by p58+ 38 clones. The addition of F(ab ')2 fragments to this assay had no effect, presumably because of the initial absence of an inhibitory ligand. When entire anti-p58 antibodies were added, however, the N K clones were unable to kill the target cell (Moretta et al., 1993). This was thought to be the result of crosslinking of p58 molecules by an interaction with FcR that mimics the effect of class I M H C . B) NKB1 The previous studies of GL183VEB6+ N K cell clones demonstrated their ability to recognize specific H L A - C antigens. In an effort to understand the specificity of other N K cell populations, L i twin et al. (1993) generated over 200 N K cell clones and tested their reactivity against a class I deficient B cell line transfected with several H L A - A , -B , or -C genes. The pattern of protection provided by these transfected genes, representing both self and non-self alleles, was complex and prevented N K cell clones from being assigned to a small number of categories according to their recognition profile. Class I specificities were found to be overlapping, but distinct, such that a single N K clone has the potential to recognize multiple alleles of H L A - A , -B , and -C antigens (both self and non-self). Over 70% of the N K clones examined by Litwin et al. (1993) were inhibited by H L A - B alleles. To identify the N K cell receptors responsible for the H L A - B mediated protection, antibodies were generated against one such N K clone and selected on the basis of their ability to permit lysis of the normally resistant H L A -B + target. The antibody DX9 was generated by this screen, and recognizes a 70 kD N K cell surface antigen designated N K B 1 (Litwin et al., 1994). Expression of the antigen is highly heterogeneous among individuals, in a manner similar to GL183 and EB6 . Analysis of 18 adult donors found the antigen to be present on <0.1-61% (mean=14%) of CD3CD56+ N K cells and <0.1-3% of C D 3 + lymphocytes. The H L A specificity of N K B 1 was investigated in a panel of cell lines transfected with several allelic forms of H L A - B and -C genes (Litwin et al., 1994). 39 All NKB1 + N K cell clones were inhibited by a subset of HLA-B alleles, and this inhibition was eliminated by the addition of the DX9 mAb (intact antibody, F(ab)2, and Fab fragments). Other class I molecules (HLA-B and -C alleles) were shown to provide protection from some but not all NKB1+ clones. This inhibition was not affected by addition of the DX9 antibody, suggesting that another N K cell receptor is involved in these processes. NKB1 therefore appears to be an inhibitory receptor for a subset of HLA-B molecules. C) NK3-Specific NK cells NKI- and NK2-specific N K cells can be generated from individuals homozygous for a particular HLA-C allele by stimulation with target cells homozygous for a different HLA-C allele. This allogeneic activation requires amino acid differences at positions 77 and 80 (Colonna et al., 1993a). A similar strategy was used to identify N K cell clones which are specific for HLA-B (Cella et al., 1994). This screen made use of the HLA-Bw4 and Bw6 public specificities that are shared by several HLA-B alleles. The Bw4/Bw6 epitopes are characterized by polymorphisms at amino acid residues 77-83. N K cells were cloned from an HLA-Bw4 homozygous donor in the presence of irradiated targets derived from an HLA-Bw6 donor (Cella et al., 1994). Analysis of the cytotoxic profile of these clones against class I transfectants found a subset of them to be inhibited by HLA-Bw4, but not HLA-Bw6 allotypes. Heterogeneity in the degree of protection provided by Bw4 alleles was strongly correlated with the amino acid at position 80. Isoleucine was associated with total inhibition, threonine with partial protection, and asparagine with complete susceptibility to lysis. This new N K cell specificity for HLA-Bw4 alleles with isoleucine 80 has been designated NK3. Although the characterization of the NK3 specificity did not investigate the nature of the N K cell receptor(s) involved, NKB1 appears to function as a receptor for HLA-Bw4 molecules (Litwin et al., 1994). 40 D) CD94 CD94 (previously named Kp43) was originally identified by immunizing mice with purified human N K cells and selecting hybridomas that secreted antibodies specific for N K cells (Aramburu et al., 1990). The CD94 antigen is a disulfide-linked dimer composed of two 43 kD subunits and is expressed by a small subset of T cells and nearly all N K cells following activation with IL-2. The initial functional characterizations found that an anti-CD94 antibody interferes with the ability of N K cells to proliferate in response to IL-2. Moretta et al. (1994) later identified the same antigen in a screen designed to characterize novel receptors for HLA-B. Having shown the specificity of GL183+ and EB6 + clones for different HLA-C alleles, p58 negative N K cell clones were analyzed for their ability to recognize class I MHC. A subset of these clones were found to be inhibited by target cell expression of HLA-B7, but not any HLA-C alleles. Monoclonal antibodies were generated against these clones and selected on the basis of their ability to reconstitute lysis of HLA-B7+ targets by these clones. Two of the antibodies selected were shown to be directed against the CD94 molecule. Both intact and F(ab')2 fragments of these antibodies were shown to restore killing of the HLA-B7 specific clones. Additionally, antibody crosslinking of the antigen in a redirected killing assay strongly inhibited the cytolytic activity of these clones. These observations suggested that CD94 functions as an inhibitory receptor for HLA-B7. E) NKG2 As an alternative approach to generating monoclonal antibodies against purified N K cells, Houchins et al. (1990) employed a subtractive screening strategy to identify genes which are specifically expressed in N K cells. This method utilized cDNA libraries constructed from both an N K cell clone and an Epstein-Barr virus transformed B cell line, and isolated those cDNAs unique to the former. Analysis 41 of these clones identified one, NKG2, whose transcript was highly abundant in N K cells, but absent in all of the T and B cell lines tested (Houchins et al., 1990). The original NKG2 isolate was a few hundred base pairs shorter than the 1.7 kilobase transcript seen on a Northern blot, and was therefore used as a probe to re-screen the N K cell library and identify a full length clone. Varying intensities of hybridization were noted among the positive plaques during this screen, and examples of each were studied (Houchins et al., 1991). Four related, but distinct, cDNAs were subsequently identified and were designated NKG2-A, -B, -C, and -D. Members of this family were found to encode code type II transmembrane proteins, ranging in size from 215-233 amino acids, with an extracellular region homologous to the CRD of C-type lectins. All of the NKG2 genes map to chromosome 12 (homologous to mouse chromosome 6), suggesting that humans have the equivalent of the murine N K cell gene complex (Yabe et al., 1993). NKG2-A and -B are identical to one another, apart from the absence of an 18 amino acid segment in NKG2-B, and are thus proposed to be products of the same alternatively spliced gene. A more detailed analysis of the cDNAs shows that NKG2-D is distantly related to the other members of the family (21% amino acid identity). In fact, its only region of nucleotide similarity to the others is a 130 base pair stretch in the 5' UTR which is 95% identical with a segment from the coding region of NKG2-A, -B, and -C (Houchins et al., 1991). Although no explanation for the placement of this segment in the cDNA has been provided, Northern blot studies suggest that it is not part of the normal NKG2-D transcript. High stringency genomic Southern blots suggests that there are possibly two or more additional, and as yet unidentified members of the gene family. Northern analysis with transcript specific probes has shown that the genes are expressed in some T cell populations, but that the individual genes are differentially regulated. Of fourteen T cell clones analyzed, 9 expressed NKG2-D, 3 expressed NKG2-A/B, and one expressed NKG2-C (Yabe et al., 1993). 42 Although NKG2 is similar to Ly-49 and NKR-P1 in its structure, C-type lectin homology, N K cell expression, and chromosomal location, low sequence homology suggests that it is not a human homologue of either one of these murine gene families (see dendogram in Figure 5). These multigene families are, however, more similar to one another than to other C-type lectins, and thus form their own sub-branch of this superfamily (Chambers et al., 1993). F) NK-TR NK-TR is a 150 kD cyclophilin-related protein expressed on the surface of human N K cells (Anderson et al., 1993). As opposed to the inhibitory receptors for class I MHC, evidence suggests that this molecule may function in the activation of N K cells. In an attempt to directly assess the importance of NK-TR as a recognition/triggering receptor, the antisense cDNA was expressed in an NK-TR+ T cell clone which displays NK-like activity (Chambers et al., 1994). The transfectants showed a decreased expression of NK-TR which was associated with a decreased ability to lyse NK-sensitive targets. This suggests that the presence of this molecule may be a requirement for N K cell lytic activity. 1.4) Thesis Objectives The above introduction summarizes the state of the field at the time that I initiated the studies reported in this thesis. Although numerous reports had suggested that N K cells recognize class I molecules, the nature of the receptors involved remained largely unknown. At the time, Ly-49A was the only molecularly characterized receptor for class I MHC. Ly-49A had been shown to be expressed by a subset of N K cells and to recognize only a fraction of class I molecules. It was therefore expected that other N K cell subsets exist that express receptors specific for. class I molecules not recognized by Ly-49A. A major question in the field was 43 the nature of the receptors expressed by these subsets and the identity of the class I molecules that they recognize. My work was aimed at studying other members of the Ly-49 gene, family as putative N K cell receptors for class I M H C . The in i t ia l hypothesis was that the N K cell repertoire (for class I M H C ) was made up of N K cell clones expressing different Ly-49 family members and having different class I specificities. The specific objectives of this thesis were therefore to characterize additional members of this family wi th regard to their expression patterns and receptor functions. In Chapter 3, experiments were performed to identify novel members of the Ly-49 family and to characterize their patterns of expression. The essential questions asked were "are these molecules expressed by N K cells, and i f so, how does this expression relate to that of Ly-49A?" The functional characteristics of indiv idual Ly-49s was the subject of Chapter 4. Hav ing observed distinct N K cell subset expression of two indiv idual Ly-49s (Chapter 3), the next question was "do both of these related molecules bind to class I MHC ? " This question was evaluated by measuring class I-dependent cell-cell adhesion mediated by transfected Ly-49 cDNAs. Chapter 5 was an investigation of the functional relevance of the extracellular lectin-like domain of Ly-49. Although Ly-49 had been shown to bind to class I M H C (Chapter 4), the possibility that oligosaccharides might be involved i n this recognition was suggested by the relatedness between Ly-49 and C-type lectins. This was an especially intr iguing hypothesis i n l ight of previous studies which had suggested that N K cells may recognize carbohydrate structures. Whi le this work was being performed, numerous advances have been made i n the field of N K cell receptors that are related to or i n some cases stem from the results described i n this thesis. These recent findings, which include further characterizations of Ly-49 as wel l as the cloning of human receptors for class I M H C , have therefore been integrated into the Introduction and Discussion sections 44 of Chapters 4 and 5, and are collectively discussed in Chapter 6 (Summary and Perspectives). 45 Chapter 2 Materials and Methods 46 2.1) Animals C57BL/6, C57BL/10 (BIO), BIO A , B10.BR, and B10.D2 mice were purchased from the Jackson Laboratory (Bar Harbor, ME). 2.2) NK Cell Preparations Activated splenic N K cells were isolated by a two step adherence system. Single cell suspensions of spleen cells were suspended in ammonium chloride to lyse red blood cells, followed by a 24 hour incubation to remove adherent cells according to their adherence to plastic. The non-adherent layer was then cultured for 3 days in the presence of IL-2 (1,000 units/ml). Activation of N K cells with IL-2 has been shown to specifically induce their adherence to plastic (Gunji et al., 1989). After 3 days all non-adherent cells were removed and the remaining cells were cultured with IL-2 for 4-6 days. This procedure results in a cell population which is 70-80% NKI. 1+. 2.3) cDNA Cloning PCR primers were designed corresponding to sequences in the 5' and 3' untranslated regions of the Ly-49A cDNA, which amplify the entire open reading frame (Chan and Takei, 1989). The sequences of the oligonucleotides were 5'-AGTACCGCGGCATTTGAACTGAGAACA-3' and 5*-TACTCCGCGGCAGACTAAGTCCAATGG-3' . The first strand of cDNA was generated in a 25 ul mixture containing 10 ug total RNA, 0.5 mM dNTPs, 4 ug/ml random hexamers, 10 mM DTT, and 200 units of reverse transcriptase. The reaction was incubated at 37°C for 1 hour followed by 5 minutes at 95°C to inactivate the enzyme. 1 ul of this reaction was then subjected to PCR in a 50 ul volume of 25 mM KC1, 1.5 mM MgCl 2 , 0.2 mM dNTPs, 1 uM each primer, and 1 unit of Taq polymerase. 35 cycles were carried out as follows: 1 minute denaturation at 94°C; 1 minute annealing at 45 °C; and 3 minute extension at 47 72°C. This was followed by a final extension at 72°C for 5 minutes. The PCR products of expected size (950bp) were purified from a 1% agarose gel, cut with Sstll to remove the primer tails, and subcloned into the Sstll site of Bluescript-KS. Ly-49D was isolated from an amplified (C57BL/6 x CBA)Fi lung cDNA library purchased from Stratagene (La Jolla, CA). 6xl0 5 plaques were screened, using a 639bp PstI Ly-49A fragment (nucleotides 442 to 1082) as a probe. Overnight hybridizations were done at 55°C in 6xSSC (lxSSC is 0.15 M NaCl, 0.015 M sodium citrate, pH 7.5), 0.5% SDS, 0.02% polyvinylpyrrolidone, 0.02% ficoll, 0.02% BSA, 10 pg/ml denatured salmon sperm DNA, and 1x10s cpm/ml denatured probe. This library screening was performed by a summer student, John Dunlop. 2.4) DNA Sequence Analysis cDNA clones were sequenced by the dideoxy method modified for use with T7 polymerase (Tabor and Richardson, 1987). Sequence analysis was performed with the G C G software package (Devereux et al., 1984). 2.5) Cell Lines The cell lines 2.4G2, 3-83P, 20-8-4S, 28-11-5S, 28-14-8S, 34-1-2S, 34-2-12S, 34-5-8S, A20, B8-24-3, C1498, CTLL-2, COS- l , GM979, IC-21, K7-65, MBL-2, P388D1, PK136, R l . l , Yac-1, and Y3 were purchased from American Type Culture Collection (Rockville, MD) and maintained in D M E M + 5% FCS, except for CTLL-2 which was grown in RPMI supplemented with 10% FCS and 20 units/ml IL-2. B10/A2/2.2 is an Abelson virus-induced leukemia cell line derived from a B10.A mouse (Takei, 1983) and was grown in D M E M + 5% FCS. 48 2.6) Antibodies The monoclonal antibodies Y E 1/32, YE1/48 (Takei, 1983), and PK136 (Koo and Peppard, 1984) were purified from tissue culture supernatants using Protein G Sepharose 4 Fast Flow (Pharmacia, L K B , Piscataway, NJ). YE1/32 and YE1/48 were labeled with biotin (Sulfosuccinimidobiotin; Pierce Chemical Company, Rockford, IL) and used for cell staining at 2 pg/ml. PK136 was labeled with FITC and cyanine 5 (Cy5), and used at 8 fig/ml and 25 pg/ml respectively. R-PE conjugated 5E6 and FITC conjugated CD3 were purchased from PharMingen (San Diego, CA) and were both used at a concentration of 10 p,g/ml. All anti-class I MHC antibodies were used as hybridoma tissue culture supernatants and are listed in Section 2.5. The monoclonal antibody 3cl2 (anti-CD44H) (Dougherty et al., 1994) was provided by Dr. Graeme Dougherty (Terry Fox Laboratory, Vancouver, Canada). 2.7) Polysaccharides and Glycosidases Dextran sulfate (mw 500,000), fucoidan, ^-carrageenan, chondroitin sulfates A, B, and C, heparan sulfate, hyaluronan, dextran (mw 500,000), neuraminidase (from Vibrio cholerae, type II), a-L-fucosidase (from bovine kidney), and FITC-labeled Peanut Lectin (from Arachis Hypogaea) were purchased from Sigma (St. Louis, MO). 2.8) Transfections Ly-49 cDNAs were subcloned into pAX142, a variant of the pAX114 expression vector (Kay and Humphries, 1991). Plasmid DNA was transfected into COS- l cells by DEAE-dextran transfection as has been described elsewhere (Hammarskjold et al., 1986). Three days later cells were analyzed by FACS® or assayed for acquired adhesive properties. 49 Cell lines permanently expressing transfected class I M H C genes were established by co-transfection of 10 ug of a plasmid containing an entire class I gene and 0.5 ug of a plasmid containing the hygromycin resistance gene (pRC6; provided by Dr. Robert Kay, Terry Fox Lab, Vancouver, BC) into the H-2S cell line, GM979. Genomic constructs of the H-2D d, D k , and D b genes were provided by Dr. Wilfred Jefferies (University of British Columbia, Vancouver, BC). Cells were electroporated in a Gene Pulser Transfection Apparatus (Bio Rad, Hercules, CA), and subsequently grown in media containing 0.5 mg/ml hygromycin for three weeks. 2.9) Flow Cytometry and Cell Sorting 2.9.1) Fresh and IL-2 activated NK cells All staining procedures were carried out at cell concentrations of lxlOVml. Spleen cells were first incubated at 4°C for 30 minutes with the antibody 2.4G2 (Unkeless, 1979) to block Fc receptor binding, then 30 minutes with the desired antibody, followed by two washes with PBS containing 2% FCS and 0.1% sodium azide. In the case of biotinylated antibodies, streptavidin-PE or streptavidin-FITC was then added and cells incubated for an additional 30 minutes at 4°C. Indirect secondary antibody staining techniques were used for single color analysis of IL-2 activated N K cells with the antibodies Y E 1/32 and Y E 1/48 only. After the first set of washes, cells were incubated with a goat anti rat-FITC antibody. In all cases the final wash was done with propidium iodide added at a concentration of 1 ug/ml. Dead cells stained with propidium iodide were gated out. The cells were analyzed on a FACStar Plus® (Beckton-Dickinson & Co., Mountain View, CA) equipped with a 5-W Argon and a 30-mW helium neon laser. 50 2.9.2) Cell lines GM979 was grown for 3 weeks in 0.5 mg/ml hygromycin as bulk populations following transfection with class I genes. When analyzed by flow cytometry, 0.5-10% of the hygromycin resistant cells were found to be expressing the transfected class I gene. The positive cells were sorted on a FACStar Plus® (Beckton-Dickinson & Co., Mountain View, CA) and expanded further in tissue culture. Following this sorting procedure, cells generally maintained a uniform and high level expression of the transfected class I genes. The antibodies used for identification of specific class I molecules were 34-5-8S (Dd), 15-5-5S (Dk), and 28-14-8S (Db). All analyses required a secondary staining step with an FITC-labeled anti-mouse IgG antibody. 2.9.3) Polysaccharide inhibition Transfected COS cells were incubated with carbohydrate for 1 hour at room temperature (control samples were treated with an equal volume of HBSS), followed by a 30 minute incubation with either 5E6-PE or 3cl2. A secondary goat anti-mouse IgG-PE staining step was required with the antibody 3cl2. Cells were washed twice following antibody staining, and dead cells were gated out with propidium iodide. Mean fluorescence values were determined for each condition, and compared to the control which did not receive carbohydrate. Each polysaccharide was tested in at least three independent experiments (transfections and FACS® analysis). 2.9.4) Peanut Lectin flow cytometry Neuraminidase and sham treated GM979 cells (Section 2.11) were washed twice and stained with 100 ng/ml FITC-labeled Peanut Lectin in a 100 ixl volume for 30 minutes at 4°C. This was followed by 2 washes in PBS/2% FCS, with propidium iodide added to the final wash at 1 pg/ml. Cells were analyzed on a 51 FACSort equipped with Lysis II software (Becton Dickinson & Co.), and dead cells were gated out. 2.10) Cell Adhesion Assays COS cells were transfected with Ly-49 cDNAs as described in Section 2.8. After 24 hours, transfectants were trypsinized and 2xl0 5 cells were transferred to 6 cm dishes (3002; Falcon Labware, Oxnard, CA). 48 hours later, the adherent layer of cells was washed once, followed by an overlay of 1.5 ml of test cells at lx lO 6 -lx lO 7 cells/ml. Cells were incubated for 2 hours at 37°C, and photographed following five washes with prewarmed media to remove all unbound cells. To quantitate cell adhesion, 5xl0 5 labeled cells (luCi Na5 1CrO4/105 cells) were incubated with adherent, transfected COS cells for 2 hours. After several washes, cells that remained attached to the adherent layer were lysed with 10% Triton X-100 and radioactivity was determined. Experiments which measure cpm bound were done in triplicate and values presented are the mean ± SD. In antibody blocking studies, COS cells or test cell lines were pre-incubated with anti-Ly-49 or anti-class I MHC antibodies for 30 minutes following 5 1 C r labeling but prior to the start of the adhesion assay. Purified YE1/48, YE1/32, and 5E6 were added at a concentration of 5 ug/ml and all anti-class I MHC antibodies were used as 1:10 dilutions of tissue culture supernatants. Adhesion assays were carried out for 2 hours at 37°C, after which plates were washed three times with pre-warmed media, bound cells were lysed with 10% Triton X-100, and radioactivity was determined. Values are expressed as percent control adhesion relative to plates tested under identical conditions except for a pre-incubation step with media rather than an antibody. These numbers represent the mean ± S E M of at least three independently performed experiments. Polysaccharide inhibition of cell adhesion was tested in an identical manner, except for the absence of a pre-incubation step before adding test cell lines to the transfected COS cells. 52 In experiments involving EGTA, assays were performed in C a + + free HBSS containing 1% BSA/2 mM MgCk 2 mM CaCk was added to control plates, whereas 5 mM E G T A was added to experimental plates. Both COS cells and GM979 were washed twice in the appropriate buffers before the start of the assay. 2.11) Glycosidase Treatments GM979 cells were labeled with 5 1 Cr, washed twice in a 150 mM sodium chloride/4 mM calcium chloride/pH 5.5 buffer, and then resuspended. Neuraminidase treatment was for 1 hour at 37°C in a 50 ul reaction volume with 0.40 or 0.22 units/ml of enzyme. Fucosidase treatment was for 1 hour at room temperature with 0.28 or 0.19 units/ml in a 100 ul reaction volume. Immediately following enzymatic treatment, cells were tested in the cell adhesion assay or analyzed by flow cytometry. 53 Chapter 3 The Ly-49 Multigene Family: New Members, Antibody Characterization and Natural Killer Cell Expression Data presented in this chapter has been incorporated into the following manuscript: Brennan, J . , D. Mager, W. Jefferies, and F. Takei. 1994. Expression of different members of the Ly-49 gene family defines distinct natural killer cell subsets and cell adhesion properties. J . Exp. Med. 180:2287. 54 3.1) Introduction Murine N K cells are capable of recognizing and eliminating diverse cell types, including tumor cell lines, virally infected cells, and MHC-disparate bone marrow grafts (Trinchieri, 1989; Yu et al., 1992). Although N K cell cytotoxicity is not MHC-restricted, the expression of certain class I MHC on target cells may provide protection from N K lysis. It has therefore been hypothesized that N K cells possess receptors which deliver negative signals upon interaction with class I MHC molecules, and that N K cells normally function by recognizing an absence of self (Ljunggren and Karre, 1990; Moretta et al., 1992). Recently, an N K cell surface molecule termed Ly-49, expressed on approximately 20% of NK1.1+ cells from C57BL/6 mice, has been proposed to be a receptor on N K cells that recognizes class I MHC molecules on target cells and transduces negative signals to N K cells. Ly-49+ N K cells were shown to be unable to lyse tumor targets expressing H-2D d, and this inhibitory effect was reversed with either the anti-Ly-49 monoclonal antibody A l or an antibody against the ocl/a2 domain of H-2D d (Karlhofer et al., 1992). The adhesion of Ly-49+ lymphoma cells to purified and immobilized D d and D k , which is inhibitable with the antibody A l , has also been demonstrated (Kane, 1994). These results further indicate that Ly-49 is an N K cell receptor for class I M H C on target cells. Ly-49 is actually a family of closely related molecules which share similar amino acid and nucleotide sequences. The original member of this family has been termed Ly-49A, and two other related molecules have been designated Ly-49B and Ly-49C (Wong et al., 1991). The Ly-49 family has sequence similarity with another family of N K cell-associated molecules termed NKR-P1. They both encode type II transmembrane proteins with lectin-like domains (Giorda et al., 1990). The 5E6 antigen is expressed by the subpopulation of N K cells that mediates the rejection of bone marrow cells from H-2d, but not H-2b, mice transplanted into irradiated hosts (Sentman et al., 1989). The genes regulating the rejection have 55 been located within the M H C and are thought to encode hemopoietic histocompatibility (Hh) antigens or to regulate the expression of Hh antigens (Yu et al., 1992). However, the identity of Hh antigens and their receptors on N K cells are unknown. The 5E6 antigen has recently been reported to be a member of the Ly-49 family (Stoneman et al., 1993), suggesting that it may function as a receptor molecule on N K cells. In this chapter I describe the cloning of two additional members of the Ly-49 multigene family as well as the specificities of three of the existing Ly-49 antibodies, YE1/32, YE1/48, and SW5E6. These results suggest that 5E6 is identical to Ly-49C, and that murine N K cells can be divided into discrete subpopulations on the basis of their expression of Ly-49A and Ly-49C. 3.2) Results 3.2.1) Cloning of additional members of the Ly-49 multigene family In a previous study which reported the isolation of the novel cDNAs Ly-49B and C, analysis of partial genomic clones suggested the existence of other Ly-49 related genes (Wong et al., 1991). In an effort to further characterize this multigene family, additional cDNAs were isolated by PCR and library screening. RT-PCR was performed on RNA isolated from IL-2 activated N K cells of B10.A mice. Several clones were isolated, one of which differed from all other Ly-49s in restriction fragment pattern and was therefore analyzed in more detail. In addition, a lung cDNA library, from which Ly-49B and C were isolated, was re-screened using a 639bp PstI fragment from the 3' end of the Ly-49A cDNA as a probe. This library yielded one clone differing from all other Ly-49 sequences. The cDNA from the lung library has been designated Ly-49D and the PCR generated cDNA from B10.A N K cells is Ly-49E. The Ly-49D cDNA was isolated by John 56 Dunlop and sequenced by Doug Freeman. A comparison of the nucleotides sequences of Ly-49A-E is shown in Figure 7. An alignment of the predicted proteins of the five Ly-49 molecules shows several well conserved features (Figure 8). With the exception of Ly-49B, which contains 17 additional C-terminal amino acids, the molecules are almost identical in size. Additionally, there is one N-linked glycosylation site as well as 11 extracellular cysteine residues common to all Ly-49s. These cysteines are likely important in disulfide bond formation which would function to give these molecules similar three dimensional structures. As has been reported previously, Ly-49A contains an extracellular domain homologous to the carbohydrate recognition domain (CRD) of C-type animal lectins (Wong et al., 1991; Yokoyama et al., 1989). Although members of the Ly-49 family lack several of the amino acids common to the CRD of other C-type lectins (Chambers et al., 1993; Drickamer et al., 1993), the four invariant cysteines of this domain are conserved in all Ly-49s. Table 2 provides pairwise comparisons of the five Ly-49 molecules at the amino acid and nucleotide levels. In all cases homology between members of this family is significantly higher at the nucleotide than at the amino acid level. Ly-49C shares 83% amino acid identity with Ly-49D, a homology higher than that seen between any of the other members of this family. It is interesting to note that the region of striking similarity between Ly-49C and Ly-49D is in the extracellular region (90%), whereas the cytoplasmic and transmembrane regions are equally homologous with all other Ly-49 molecules (65%). Ly-49E shares 76% of its amino acid residues with Ly-49A, with the majority of these falling in the lectin-like domain (83%). All other sequence comparisons between members of this family show similar ranges of identity in the cytoplasmic and transmembrane domains and significantly greater divergence in the extracellular lectin-like domain. 57 9A CATGAGGTTGAGTATCTCTCAGTGGAAATTTAGTTCTACCGTTTATTTTGGAGACACTTAOGGGATATCAACCAGAAAAAG 9C GAATTCCCGGGGAAAACG 9D GAATTCCCGGGGAGAACTAGTCTCGAATTTTTTTTTTTTTTTTTTTT^ 9A CCAACTTTTTCCTCCACCAGAACCACTTCTTGCTAGCGACACAGAAACCACTCGAGGCACCATTTGAACTGAGAACATACTTTATATATCAATCCCAAGA 9* ACTTCATACATCATTCCCAAGA 9C CCAACGTTTCAGACAAATTTTCCCTCCACCAGCATCACTCCGG TAGAGACACAGACTTCTTGTAC . . . TCCCACGA. 9D ACCCCAATTCAGGCATCCATTCTTTCTACCGGCATCACTTCAGGGTGGACACACAGGACATATTTTTTAAAAGAACATACTCTACGTAT. . . TCCCAAGA, 9B . . GGAATTCCCGGGCACCAGAATTACTTCCTGGTAGAGACACAGAAATCACTCGAGGCCACATTTTAATACAAATCGTACTTTACATAC . . . TTCCAAGA. 9A ISAGTGAGCAGGAGGTCACTTATTCAATGGTGAGATTTCATAAATCTGCMGATTGCAGAAACAAGTGAGACCTGAGGAGACTAAAGGGCCCAGAGAAGC 91 IflAGTGAGCAGGAGGTCACTTACTCAACTGTGAGATTTCATGAGTCTTCAAGGTTGCAGAAACTAGTGAGGACTGAGGAGCCTCAAAGGCCCAGAGAAGC 9C ISAGTGAGCCAGAGGTCACTTACTCAACTGTGAGACTTCATAAGTCTTCAGGGTTGCAGAAACAAGTAAGGCATGAGGAGACTCAAGGGCCCAGAGAAGT 9D IflAGTGAGCAGGAGGTCACTTTCCCAACTATGAGATTCCACAAGTCTTCAGGGTTGAACAGCCAGGTGAGACTTCAGGGgAcTcAGAGATCTAGAAAAGC 9B IflAGTGAGCAGGAGGTCACTTACACAACTCTGAGATTTCATAAGTCTTCAGGGTTGCAGAACCCAGTGAGGCCTGAGGAGACTCAAAGGCCCAGAGATGT 9A TGGCTACAGAAGGTGTTCATTCCACTGGAAGTTCATTGTGATAGCTCTTCGCATCTTCT 9E TTGCTACAGAGAGTATTCAGTCCCCTGGcAGCTCATTGTGATAGCTTGTQGAATCCTCTGTTTCCTTCTCCTGGTAACTGTTGCATTGTTGGCAATAAC 9C TGGCAACAGAAAGTGTTCAGCACCCTGGCAACTCATTGTGAAAGCTCTTGGAATC^ 9D TggcCTAAGAGtGtGlTCAGTCCCTTGGCAGCTCATTGTGATAGCTCTTGGAATCCTCTGTtcCCTTCGGCTGGTAATTGTTGCAGTGTTTGTGACAAAG 9B GGGCCACAGAGAGTGTTCAGTCCCCTGGAAGTTCATTGTGATAGTTCTTOGAATCCTCTGTTTCCT^ 9A ATTTTTCAGTATGATCAACAAAAA. . . AAACTGCAGGAATTTCTAAACCAC . . . CACAATAACTGCAGCAACATGCAAAGTGACATCAACTTGAAGGATG 9E ATTTTTCAGCATAGTCAACAAAAACATGAACTACAGGAAACTCTAAATTGC . . . CACGATAACTGCAGCCCCACGCAAAGTGACGTCAACTTGAAGGATG 9C ATTTTTCAGTATAATCAACACAAACAAGAAATCAATGAAACTCTAAACCAT. . . CACCATAACTGCAGCAACATGCAAAGTGATTTCAACTTAAAGGAAG 9D TTTTTTCAGTATAGTCAACACAAACAAGAAATCAATGAAACTCTCAACCAC . . . CGCCATAACTGCAGCAACATGCAAAGGGATTTCAACTTAAAGGAAG 9B ATTTTCCGGGATGGACAAGAGAAACATGAACAGGAGAAAACTCTAAATAACCTCCGTCAAGAGTACCAGGTCATGAAAAATGACAGCTCCTTAATGGAAG 9A AAATGCTGAAAAATAAGTCTATAGAGTGT GATCTTCTQGAATCCCTCAACAGGGATCAGAACAGATTGTATAATAAAACCAAGACTGT 9E AACTGCTGAGAAATAAGTCTATAGAGTGTAGGCCAGGCAATGATCTTCTGGAATCCCTCAGCAGGGATCAGAACAGATGGTACAGTGAAACCAAGACTTT 9C AAATGTTGACAAATAAGTCTATAGATTGTAGGCCAAGCAATGAAACTCTGGAATATATCAAAAGAGAACAGGACAGATGGGACAGTAAAACAAAGACTGT 9D AAATGTTGACAAATAAGTCTATAGATTGTAGGCCAAGCTATGAACTTCTQGAATACATCAAAAGAGAACAGGAGAGATGGGACAGTGAAACCAAGAGTGT 9B AAATGTTAAGAAATAAGTCTTCAGAGTGTAAGGCCCTCAATGATAGCCTGCACTACCTCAACAGAGAACAGAACAGATGCCTCAGGAAAACCAAGATTGT 9A TTTAGATTCCTTACAGCACACAGGCAGAGGTGATAAAGTATACTGGTTCTCCTATGGTATGAAATGTTATTATTTCGTCATGGACAGAAAAACATGGAGT 9E TTCAGATTCCTCACAGCACACAGGTAGAGGTTTTGAAAAATATTGGTTCTGTTATGGTATAAAATGTTATTATTTCAACATGGACAGAAAAACGTGGAGT 9C TTTAGATTCCTCACGGGACACAGGCAGAGGTGTTAAA. . . TACTGGTTCTGCTATAGTACTAAATGTTATTATTTCATCATGAACAAAACTACATGGAGT 9D TTC AGATTCTTCACGAGACACAGGC AGAGGTGTTAAA. . . TACTGGTTCTGCTATGGTACTAAATGTTATTATTTCATCATGAATAAAACTACATGGAGT 9B TTTAGATTGCTCACAGAACAAAGGCAaGCAAGTGGAAGGATACTGGTTCTCCTGTGGCATGAAATGTTATTATTTCATCATGGATGATAAAA 9A GGATGTAAACAGACCTGCCAGAGTTCCAGTTTATCCCTTCTCAAGATAGATGATGAGGATGAACTGAAGTTCCTTCAGCTCGTGGTTCCTTCA 9E GGATGTAAACAGACCTGCCAGATTTCCAGCTTATCCCTTCTGAAGATAGACAATGAGGATGAACTGAAGTTCCTTCAGAACCTGGCTCCTTCAGACATTT 9C GGATGTAAAGCGAACTGCCAGCATTTTAGCGTTCCCATTCTGAAGATAGAAGATGAAGATGAACTGAAATTCCTTCAACGCCATGTTATTCCAGAGAATT 9D GGATGTAAAGCGAACTGCCAGCATTACAgcGTTCCCATTGTGAAGATAGAAGATGaAGATGAACTGAAATTCCTTcAACgCCATGTTATtCTAGAGAGTT 9B GGATGTAAACAGATCTGCCAGGATTACAACTTAACTCTTTTGAAGACAAATGATGAGGATGAATTGAAGTTCCTTAAATCCCAACTTCAAAGAAACACA 9A GCTGGGTTGGATTGTCATATGATAATAAGAAAAAAGATTGGGCATGGATTGACAATCGCCCATCTAAACTTGCCTTGAACACAAGGAAATACAATATAAG 9E CCTGGATTGGATTGTCATATGACAATAAGAAAAAAGATTGGGTATGGATTCACAATGGCCCATCTAAACTTGCCTTGAACACAACGAAATATAATATAAG 9C ACTGGATTGGATTGTCTTATGATAAGAAAAAAAAGGAATGGGCATGGATT3ACAATGGCCCATCTAAACTTGACATGAAAATAAGGAAAATGAACTTTAA 9D ACTGGATTGgATTGTCATATGATAAGAAAAAAAAGGAATGGGCATGGATTCACAATGGCCAATCTAAACTTGACATGAAAATAAAGAAAATGAACTTTAC 9B ACTGGATTTCACTGACACATCATAAAAGCAAAGAGGAATCGCAACAGATTCGTGATAGACCATCTAAACTTGATTCAGCAGCAAGGAATTCAGTACCTAA 9A AGATGGGGGATGTATGTTGTTATCTAAAACAAGACTAGACAATGGTAACT3TGATCAAGTATTCATCTGTATTTGTGGGAAGAGACTGGATAAATTCCCT 9E AGATGGATTATGTATGTCGTTATCTAAAACAAGACTAGACAATGGGGACT3TGATAAATCATACATCTGTATTTGTGGTAAGAGACTGGATAAATTTCCT 9C GTCTAGAGGATGTGTATTTTTATCTAAAGCAAGAATAGAAGATATTGACT3TAATATTCCCTACTACTGTATTTGTGGGAAGAAACTGGATAAATTCCCT 9D GTCTAGAGGATGTGTATTTTTATCTAAAGCAAGAATAGAAGATACTGACTGTAATACTCCCTACTACTGTATTTGTGGGAAGAAACTGGATAAATTCCCT 9B CAGACAAAAGTGTGCATATCTAAGTTCATTTTCTACAGAAGAGGATGACT3TGCTAGAACTCATGGTTGTATT 9A CATigACTCTCCAATGAGTGTTAAAGG. . . AAAAAGTGAAATTTTCTTACTCTCATTTGTTTCCTGTAT 9E CAT1SACTCTCCAATGAGTCTAAAAGGTAAAAAAAATGAAATTTTTTTACTCACATTTGTTTCCTGCAA 9C GATlAiTTTTCCAACCAGAGTTAAAGGT. . . AAAAATGGAATGAGTTGATCCTTATTCGTTTCTTGTAA 9D GATIAACTTTCCAACCAGAGTTAA 9B ATTCCAGGGAGCTGTGCCAAGGGAAGAACTCAATCTGCTCTGCAGAGGGATGAAGATGAAAGTIAAGAAATGTTTTGACAACAGACTTAACAGAGCAGAA 9A TAATTAATGACACCTTGCAAACAAGTGTTTTGACCATTGGACTTAGTCTGCAGTGCAAAGAGAG AGAGA 9E TAATTCATTATTCCTTACAAACAAGTGTTTTGA 9C TAATTCATGACTCC . AACAAACAAGTATTTTGATTACAGAACATAGTCTCCTGTG AAG AGAAA 9D AGAACATAGTCTGCTGTG AAG AGAAA 9B GCCATCTTCCTTCCTGGTAAACAGAGAAGCTTGTGTCTGAAGAATGGACACTCTTTGCTCTAAGCTTTGAAGAATC 9A GAAAATCTGGAAGATTTTGGGAATATTCTCTGAAACATGACATGACAGAGCAGATGACATCTTCCTTCCCTGTTG AGACTGGAC 9C CAAAGCTGCAAGAACATTGGGACTGTACTCTCCTATCTATCTTGACAGAACAGAGGTCA.TTTTCTATCCTGTTGGAGAGAGGACGCATCTACTCGGGTG »9D CAAAGCTGCAAGAACATTGGGACTGTCCTCTCCTATCTATCTTGACAGAACAGAGGTCATTTTTCTATCCTGTTGGAGAGAGGACAAATCTACTCAGGTG 19B GAACAATATACAATACTCAGACAATCACTCACATTCTTAATTATCACTAQGATAATCCATTTCTATAACCTGTTGCTCTGAAATCACTTAAGAACCAGGA 9A AGATCTTCTCTGATACCCCAAAGCTTGGACGAATCTGTTTTATTTGTTTGCATAAACTCTA 9C AATGGGCACGCTTTGCCCTAAAGCCTTCAGAATTGTGTTCTTTCTGATTTCTTAAACTCCCATAAAACTGAATAAAGAGTCCTCCCTAAAATAA 9D AATGGGCATGCTTTGCCCTAAAGCCTTCAGAATTGTGTTCTTTCTGATTTCTTAAACTCCCATAAAACTGAATAAAGAGTCCCCCCTAAAAAAAAAAAAA 9B TGCAGTTGCTGTTTGTTAGTTGGTTGGTAACTTTTTATTTTTGTTTTTTGGAGGTAGGAGAATTAACCTGA 9C AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA. 9D AAAAAAAAAAACTCGAG 9B TTTATAACTTGAAAAAAAAAAAAAAAAAACTCGAG Figure 7. Nucleotide sequence comparison of Ly-49A-E cDNAs. Initiation and termination codons are designated by underlined bold text. The Ly-49D and E sequences are available under accession numbers U12889-90. 58 1 50 Ly-49C MSEPEVTYST VRLHKSSGLQ KQVRHEETQG PREVGNRKCS APWQLIVKAL Ly-49D MSEQEVTFPT MRFHKSSGLN SQVRLEGTQR SRKAGLRVCS VPWQLIVIAL Ly-49A MSEQEVTYSM VRFHKSAGLQ KQVRPEETKG PREAGYRRCS FHWKFIVIAL Ly-49E MSEQEVTYST VRFHESSRLQ KLVRTEEPQR PREACYREYS VPWQLIVIAC Ly-49B MSEQEVTYTT LRFHKSSGLQ NPVRPEETQR PRDVGHRECS VPWKFIVIVL MSE-EVT -R-H-S--L- --VR-E -R R--S —W--IV— 51 100 Ly-49C GILCFLLLVT VAVLAVKIFQ YNQHKQEINE TLNH.HHNCS NMQSDFNLKE Ly-49D GILCSLRLVI VAVFVTKFFQ YSQHKQEINE TLNH.RHNCS NMQRDFNLKE Ly-49 A GIFCFLLLVA VSVLAIKIFQ YDQQK.KLQE FLNH.HNNCS NMQSDINLKD Ly-49E GILCFLLLVT VALLAITIFQ HSQQKHELQE TLNC.HDNCS PTQSDVNLKD Ly-49B GILCFLLLLT VAVLVIHIFR DGQEKHEQEK TLNNLRQEYQ VMKNDSSLME GI-C-L-L-- V F- --Q-K -LN D--L--101 150 Ly-49C EMLTNKSIDC RPSNETLEYI KREQDRWDSK TKTVLDSSRD TGRGVK.YWF Ly-49D EMLTNKSIDC RPSYELLEYI KREQERWDSE TKSVSDSSRD TGRGVK.YWF Ly-49A EMLKNKSIEC . . . . DLLESL NRDQNRLYNK TKTVLDSLQH TGRGDKVYWF Ly-49E ELLRNKSIEC RPGNDLLESL SRDQNRWYSE TKTFSDSSQH TGRGFEKYWF Ly-49B EMLRNKSSEC KALNDSLHYL NREQNRCLRK TKIVLDCSQN KGKQVEGYWF E-L-NKS^-C L -R-Q-R TK D -G YWF 151 200 Ly-49C CYSTKCYYFI MNKTTWSGCK ANCQHFSVPI LKIEDEDELK FLQRHVIPEN Ly-49D CYGTKCYYFI MNKTTWSGCK ANCQHYSVPI VKIEDEDELK FLQRHVILES Ly-49A CYGMKCYYFV MDRKTWSGCK QTCQSSSLSL LKIDDEDELK FLQLWPSDS Ly-49E CYGIKCYYFN MDRKTWSGCK QTCQISSLSL LKIDNEDELK FLQNLAPSDI Ly-49B CCGMKCYYFI MDDKKWNGCK QICQDYNLTL LKTNDEDELK FLKSQLQRNT C KCYYF- M W-GCK —CQ -K EDELK FL 201 250 Ly-49C YWIGLSYDKK KKEWAWIDNG PSKLDMKIRK MNFKSRGCVF LSKARIEDID Ly-49D YWIGLSYDKK KKEWAWIHNG QSKLDMKIKK MNFTSRGCVF LSKARIEDTD Ly-49A CWVGLSYDNK KKDWAWIDNR PSKLALNTRK YNIRDGGCML LSKTRLDNGN Ly-49E SWIGLSYDNK KKDWVWIDNG PSKLALNTTK YNIRDGLCMS LSKTRLDNGD Ly-49B YWISLTHHKS KEESQQIGDR PSKLDSAARN SVPNRQKCAY LSSFSTEEDD -W--L K 1 -SKL C— LS 251 285 Ly-49C CNIPYYCICG KKLDKFPD Ly-49D CNTPYYCICG KKLDKFPD Ly-49 A CDQVFICICG KRLDKFPH Ly-49E CDKSYICICG KRLDKFPH Ly-49B CARTHGCICE KRLNKFPIPG SCAKGRTQSA LQRDE C CIC- K-L-KFP Figure 8. Amino acid sequence comparisons of Ly-49A-E. Only those residues found i n a l l members are indicated i n the consensus sequence. A bold C identifies the cysteine residues found i n a l l C-type carbohydrate recognition domains (CRD). transmembrane region; 52555 potential N-l inked glycosylation site; and = = region homologous to C-type CRD. 59 Table 2. Homologies Between Members of the Ly-49 Family Ly-49A Ly-49B Ly-49C Ly-49D Ly-49E Ly-49A -- 70 81 78 88 Ly-49B 52 -- 71 70 71 Ly-49C 65 52 -- 91 80 Ly-49D 59 49 83 -- 78 Ly-49E 76 50 63 59 Percent nucleotide identity within the coding region is above the dashes, and percent amino acid identity is below. 60 3.2.2) COS cell expression of Ly-49 cDNAs and reactivity with Ly-49 antibodies To determine whether different anti-Ly-49 antibodies can specifically recognize unique members of this multigene family, I expressed the five Ly-49 cDNAs in COS cells and tested for reactivity with YE1/32, Y E 1/48, and 5E6 (Figure 9). Both YE1/48 (Figure 9A) and YE1/32 (not shown) react specifically with Ly-49A and no other members of the family. 5E6, which has recently been shown to bind an Ly-49 molecule (Stoneman et al., 1993), reacts specifically with Ly-49C (Figure 9F). This lack of cross-reactivity between related but distinct proteins indicates that these antibodies should prove useful in dissections of the Ly-49 family. 3.2.3) Expression ofYEl/48 and 5E6 defines distinct NK subsets I next examined the expression of Ly-49A and Ly-49C on N K cells from C57BL/6 spleen by three color FACS® analysis using Y E 1/48, 5E6, and NK1.1. Figure 10B demonstrates that NK1.1+ cells are either Ly-49ALy-49C- (50.8%), Ly-49A+Ly-49C" (18.7%), Ly-49C+Ly-49A- (25.3%), or Ly-49A +Ly-49C + (5.2%). This is the first demonstration that expression of members of the Ly-49 family of molecules defines separate N K cell subsets. Analysis of NKI. 1 and Ly-49A within Ly-49C+ cells showed a complex pattern of expression (Figure 10D). Two Ly-49A+ populations were evident within the Ly-49C+ population, one Ly-49A "high" (most of which are NK1.1+) and the other Ly-49A "low" (mostly NK1.1). Ly-49C+Ly-49A-cells were also divided into NK1.L and NK1.1+ populations. Similarly, Ly-49A "high" cells could be subdivided into NK1.1-Ly-49C", NK1.1+Ly-49C-, NK1.1 +Ly-49C+, and NKl.l"Ly-49C + populations (Figure 10F). Gating on Ly-49A+ or Ly-49C+ spleen cells has shown that significant numbers of these cells are NK1.1 negative (Figures 10D and 10F). I have examined these populations for expression of CD3 (Figures 10H and 10J). Three color analysis shows that substantial portions of both Ly-49A+ and Ly-49C + cells are CD3 + , mainly in the NK1.T window. However, in each case there remains a 61 Y E 1/48 5E6 Ly-49A Ly-49B Ly-49C Ly-49D Ly-49E Figure 9. COS Cell Expression of the Ly-49 cDNAs. Each of the cDNAs were transiently expressed in COS cells, and tested for reactivity with the antibodies Y E 1/48 and 5E6 after three days of growth. Solid histograms indicate COS cells transfected with Ly-49 cDNAs, and empty histograms are cells transfected with the vector pAX142. 62 m ->-O so 1 dd iSo 2dd 250 Forward Scatter co t B 25.3% 5.2'/. HI C L •••ijti&i'. '^-v^P? •  18.7% '1 DO ioi 102 io3 j ANTI-LY49A (FITC) LU =r o -CD 50 i dd 150 2dd Z50 Forward Scatter 100 l b ' 10? 103 ANTI-LY-49A (FITC) O H LL < CO T >• E 5'o 1 OD I SO 2do 250 Forward Scatter LO o 2 < 36.0% 11.0% Q-& ^ 4.6% 1 00 i o i i'02 io3 f ANTI-LY-49C (PE) 0 H 50 idd 150 2dd 250 Forward Scatter LLT sp-o. < s CD t " > 3 _J I i — a. z < * 100 101 102 i'o3 i'o4 ANTI-CD3 (FITC) J 50 Idd 150 2dd 250 Forward Scatter IdO 101 102 1103 iio* ANTI-CD3 (FITC) Figure 10. Phenotypic analysis of Ly-49A and Ly-49C populations in C57BL/6 spleen cells. Staining profiles of total splenocytes are shown in parts A, C, E , G, and I. Boxes indicate gated populations from which 5,000 events were collected for analysis in parts B, D, F, H, and J . 63 significant number of Ly-49A+ and Ly-49C + cells which are negative for both NK1.1 and CD3. 3.2.4) YE 1/32 and YE 1/48 expression on NK cells of BIO MHC congenic mice It has been reported previously that Ly-49+ N K cells defined by the mouse monoclonal antibody A l are deleted in certain MHC congenic strains of mice (B10.A, B10.D2, and B10.BR), indicating an MHC-based selection of the N K cell repertoire (Sykes et al., 1993). FACS® analysis of IL-2 activated N K cells, however, demonstrated that the YE1/48+ population is not grossly altered in these strains of mice (Figures 11A-D). In addition, two color analysis of freshly isolated splenic N K cells indicates that YE1/48+ N K cells are not deleted in B10.A mice (Figure H E ) . This expression pattern is in contrast to that of the antibody A l , which stains 15-20% of BIO NK1.1+ cells, but is entirely absent in B10.A mice (Sykes et al., 1993). Analysis of Y E 1/48 expression within the NK1.1+ gate of fresh spleen cells clearly demonstrates that the same 15-20% population persists regardless of haplotype. 3.3) Discussion The initial molecular characterization of Ly-49 demonstrated that the cDNA hybridizes to multiple bands on a genomic Southern blot. These bands likely represent separate genes rather than a complex intron-exon structure, because even small fragments of the Ly-49A cDNA hybridized to multiple bands (Chan and Takei, 1989; Yokoyama et al., 1989). I have continued the elucidation of this complex family by isolating two novel Ly-49 cDNAs. Ly-49D was cloned from the same (C57BL/6 x CBA)Fi lung cDNA library which yielded Ly-49B and Ly-49C (Wong et al., 1991). Although Southern blot analysis between mouse strains indicates that this family may be polymorphic (Yokoyama et al., 1990), the sequence differences between the cDNAs we have isolated is likely too great to be 64 YE1/48 Figure 11. Expression of YE 1/48 on NK cells of MHC congenic mice. Day 9 IL-2 activated N K cells (A-D) stained with monoclonal antibody Y E 1/48 and a secondary antibody (solid histograms) or stained with secondary antibody alone (empty histograms). NK1.1+ gated fresh spleen cells stained with biotinylated YE1/48 followed by Streptavidin-PE (E). 65 allelic. Ly-49E was cloned from a B10.A mouse, whereas Ly-49A was originally cloned from C57BL/6 T lymphoma lines (Chan and Takei, 1989; Yokoyama et al., 1989). Genomic Southern blot analysis with the Ly-49A cDNA has shown that these two strains share the same RFLP pattern (Yokoyama et al., 1990). Although the expression patterns of these genes have not been fully determined, all five cloned Ly-49 genes are detected in NK-enriched spleen cells by RT-PCR, suggesting that all of them may be expressed in N K cells (Masterman et al., unpublished observation). COS cell expression of these five Ly-49 cDNAs has allowed for a precise definition of three Ly-49 monoclonal antibodies. YE1/32 and YE1/48 react with Ly-49A, and 5E6 is specific for Ly-49C. This lack of cross-reactivity between related molecules has allowed for a dissection of N K cell subsets based upon their expression of Ly-49A and C. I have shown that within NK1.1+ spleen cells, there are both Ly-49A single positive, Ly-49C single positive, as well as double negative and double positive cells. This is the first demonstration that expression of distinct members of the Ly-49 family of molecules defines separate N K cell subsets. Previous work has ascribed different functional properties to Ly-49A+ and Ly-49C + N K cells. IL-2 activated Ly-49A+ and Ly-49A- N K cells are equal in their ability to lyse H-2 b tumor target cell lines, whereas Ly-49A+ N K cells are specifically unable to lyse H-2 d and H-2 k targets. It has been demonstrated that expression of D d on the target cell surface protects the cell from lysis by Ly-49A+ N K cells (Karlhofer et al., 1992). Ly-49C+ (5E6+) N K cells, on the other hand, mediate the rejection of H-2d, but not H-2b, bone marrow grafts in lethally irradiated recipients (Sentman et al., 1989). The demonstration that these N K antigens are expressed on distinct subsets confirms the prediction that N K cells may be subdivided into separate functional compartments. The fact that the cell surface molecules which allow for this precise subdivision belong to the same family of molecules suggests that they themselves may be receptors which confer a 66 functional property on the subset. This has been shown to be the case for Ly-49A. It is interesting to note that there are a small number of NK1.1+ cells double positive for Ly-49A and Ly-49C. It remains to be seen if this population has a unique functional characteristic or whether it represents a stage in N K cell development. Ly-49C + spleen cells can be divided into complex subsets, including Ly-49A negative, Ly-49A "low", and Ly-49A "high" populations, each of which can further be divided into NK1.1+ and NK1.T subsets. Significant numbers of Ly-49A+ and Ly-49C+ cells co-express CD3 and NK1.1, demonstrating a true sharing of markers that have been regarded as N K specific (NK1.1) or T cell specific (CD3). Some of the cells expressing Ly-49A or C are NK1.1- and include CD3+ as well as CD3- cells. FACS® analysis of N K cell populations with the antibodies YE1/32 and Y E 1/48 has shown an interesting difference when compared with the anti-Ly-49A monoclonal antibody A l . A l + cells represent 15-20% of NK1.1 + cells in BIO mice, but are absent in the MHC congenic strain B10.A, demonstrating an MHC based selection of the N K cell repertoire (Sykes et al., 1993). My results have shown, however, that N K l . l + Y E l / 3 2 + and N K l . l + Y E l / 4 8 + cells are not deleted in B10.A or any other M H C congenic strains of mice tested (B10.BR and B10.D2). This discrepancy in the expression patterns of these Ly-49 antibodies suggests that Y E 1/32 and Y E 1/48 recognize epitopes distinct from A l and not negatively selected by the H-2d, H-2 k, or H-2 a haplotypes. It is not known whether this epitope is carried by an additional, and as yet uncharacterized, member of this highly related family of molecules or if it is generated by a post-translational modification of Ly-49A, perhaps by formation of dimers with other Ly-49 molecules. In summary, the data presented in this chapter has addressed one of the fundamental objectives of this thesis: a characterization of the expression patterns of other members of the Ly-49 family. An antibody specific for Ly-49C was identified, and Ly-49A and Ly-49C were shown to be expressed on distinct N K cell 67 subsets. The next chapter will address the receptor functions of each of these related molecules. Note: The cloning of several novel Ly-49 cDNAs has recently been reported (Smith et al., 1994), and these have been designated Ly-49D, E , F, and G. The Ly-49D reported in this thesis is distinct from all of these sequences, and will be referred to as Ly-49H in future studies. The Ly-49E described in this thesis corresponds to the Ly-49G group of cDNA clones, which likely arise from alternate splicing of the same gene. 68 Chapter 4 Recognition of Class I MHC by Members of the Ly-49 Family Data presented in this chapter has been incorporated into the following manuscripts: Brennan, J . , D. Mager, W. Jefferies, and F. Takei. 1994. Expression of different members of the Ly-49 gene family defines distinct natural killer cell subsets and cell adhesion properties. J . Exp. Med. 180:2287. Brennan, J . , G. Mahon, D.L. Mager, W.A. Jefferies, and F. Takei. Recognition of class I major histocompatibility complex molecules by Ly-49: specificities and domain interactions (Submitted) 69 4.1) Introduction The recognition of certain class I M H C molecules by natural killer cells may result in the delivery of an inhibitory signal to the N K cell which prevents target cell lysis (Yokoyama, 1995). This inhibition can be overcome by blocking with antibodies against either class I MHC or N K cell receptors. In humans, N K cell receptors for class I MHC include a novel family of molecules (NKAT/p58) whose members contain two or three external immunoglobulin domains (Colonna et al., 1995; D'Andrea et al., 1995; Wagtmann et al., 1995). Individual members of this family are expressed by subpopulations of N K cells and have distinct specificities for either HLA-C (EB6 and GL183) (Moretta et al., 1993; Vitale et al., 1995) or HLA-B alleles (NKB1) (Gumprez et al., 1995; Litwin et al., 1994). CD94 also functions as an N K cell receptor for HLA-B (Moretta et al., 1994), although it appears to be molecularly unrelated to the NKAT/p58 family (Aramburu et al., 1990). Antibodies against all of these receptors block the protection provided to a target cell by specific class I molecules. In mice, the Ly-49 family has been implicated in class I recognition by N K cells. cDNAs encoding eight members of this family have been identified (Ly-49A-H) (Chapter 3; Chan and Takei, 1989; Smith et al., 1994; Wong et al., 1991; Yokoyama et al., 1989), and polymorphism of individual members has also been reported (Ly-49A, C, and G) (Held et al., 1995; Mason et al., 1995; Stoneman et al., 1995). Studies utilizing monoclonal antibodies which recognize specific members of this family have shown overlapping N K cell subset expression for Ly-49A (20%), Ly-49C (30%), and Ly-49G2 (50%) (Chapter 3; Mason et al., 1995). Ly-49A binds to H-2D d and D k (Daniels et al., 1994a; Kane, 1994) and inhibits the lysis of target cells expressing these antigens (Karlhofer et al., 1992). Ly-49G2+ N K cells appear to be inhibited by H-2 d antigens (possibly in a manner distinct from Ly-49A) (Mason et al., 1995), although a physical interaction between Ly-49G2 and H-2 antigens has not been demonstrated. 70 Bone marrow transplantation studies have implicated Ly-49C + (5E6+) N K cells in the recognition of class I MHC. This subset mediates the rejection of BALB/c (H-2d), but not C57BL/6 (H-2b), bone marrow when injected into lethally irradiated F i recipients (Sentman et al., 1989). Data presented in this chapter demonstrates that Ly-49C (BALB/c / CBA allele) functions as a receptor for a broad range of class I molecules. Characterization of the class I ligands of Ly-49 A and Ly-49C has demonstrated that these N K cell receptors have distinct but overlapping specificities. 4.2) Results 4.2.1) COS cells transfected with Ly-49 A or Ly-49C cDNAs bind a variety of cell lines Recent studies have demonstrated a physical interaction between Ly-49A and certain class I MHC molecules (Daniels et al., 1994a; Kane, 1994). To determine whether Ly-49C can mediate similar functions and to further characterize the binding patterns of Ly-49A, I have examined several cell lines for their ability to adhere to COS cells transfected with each of these Ly-49 cDNAs. This was evaluated experimentally by expressing Ly-49 at high levels in COS cells and testing for acquired adhesive properties in a cell-cell adhesion assay (described in Figure 12). The haplotypes of the cell lines tested were H-2 d (A20 and P388D1), H-2 k (Rl.l), H-2 b (C1498, CTLL-2, IC-21, and MBL-2), H-2*(GM979), and H-2* (B10/A2/2.2 and Yac-1). Ly-49A transfected COS cells bound the H-2 d and H-2 k cell lines, but failed to bind to H-2b, H-2 a, and H-2S (Table 3). Ly-49C transfectants bound all of the cells tested, with the exception of the two H-2 a cell lines (which also did not bind Ly-49A). The photographs in Figure 13 show typical binding of individual H-2 d (A20), H-2 b (IC-21), H-2 k (Rl.l), and H-2* (GM979) cell lines to COS cells transfected with 71 incubate with desired transfect cDNA ^ tyPe o 0 o o o COS Cells Figure 12. Experimental protocol of COS cell adhesion assay. DEAE-dextran transfection of COS cells results in high level cell surface expression of Ly-49A and Ly-49C following 72 hours of cell culture (see Figure 8). These adherent, transfected COS cells are then incubated with various non-adherent cell lines for 2 hours at 37°C, after which time all unbound cells are washed away. All cell lines tested were unable to bind to plastic or untransfected COS cells. 72 Table 3. Adhesive Properties of Ly-49A & Ly-49C Expressing COS Cells Cell Line Cell Type H-2 Ly-49A Ly-49C A20 BCel l d +• + P388D1 Macrophage d + + R l . l TCel l k + + CTLL-2 TCel l b - + C1498 Lymphoma b - + IC-21 Macrophage b - + MBL-2 TCel l b - + GM979 Erythroleukemia s - + Yac-1 T Cell a (RkD d) B10A/A2/2.2 immature a myeloid Transfected COS cells were incubated with the above cell lines for 2 hours at 37°C. Adhesion (+/-) was evaluated relative to the negative control (pAX142 transfectants) after five washes with pre-warmed media. 73 Ly-49C A20 ( H 2 d ) IC-21 (H-2 b ) GM979 (H-2 S) R1.1 (H-2 k) Figure 13. Adhesion of cell lines to COS cells expressing Ly-49A or Ly-49C. COS cells transfected with the vector pAX142 (A, D, G, J), Ly-49A (B, E , H , K), and Ly-49C (C, F, I, L), were overlayed with the cell lines A20 (A, B, C), IC-21 (D, E , F), GM979 (G, H , I), and R l . l (J, K, L). Cells were photographed after five washes following a 2 hour co-incubation at 37°C. 74 Ly-49 A, Ly-49C, or pAX142. This cell-cell binding phenomenon was also quantitated (Figure 14) by labeling the input test cells with 5 1 C r , and measuring the amount of radioactivity which remains attached to the adherent transfected COS cells after washing away unbound cells. Specific adhesion was only observed in COS cells expressing an Ly-49 molecule. These results demonstrate that Ly-49C, like Ly-49A, functions as a receptor molecule, and that these related molecules have non-identical ligand specificities. 4.2.2) Ly-49 mediated adhesion is inhibited by anti-Ly-49 and anti-class I MHC antibodies In consideration of previous studies which demonstrated that Ly-49A binds to class I M H C , it seemed possible that such an interaction was responsible for the binding of Ly-49A- and Ly-49C-expressing COS cells to the cell lines described above. This hypothesis was evaluated by testing a panel of anti-class I M H C antibodies for their ability to block cell adhesion mediated by both Ly-49A and Ly-49C. Antibody blocking experiments with the H-2 d cell line demonstrated the specificity of Ly-49A for D d , but not K d (Figure 15A). As reported previously (Daniels et al., 1994a; Karlhofer et al., 1992), an antibody recognizing the al/oc2 domain (34-5-8S) entirely blocked adhesion, whereas an antibody against the a3 domain (34-2-12S) had no effect. Ly-49C-mediated adhesion was partially blocked by antibodies specific for either K d (20-8-4S) or D d (34-5-8S), and entirely blocked by the simultaneous addition of both antibodies (Figure 15B), suggesting that Ly-49C binds to both D d and K d . This was further supported by the total inhibition seen with an antibody recognizing both K d and D d (34-1-2S). As was seen with Ly-49A, the antibody directed against the al/a2, but not a3, domain of D d inhibited Ly-49C-mediated adhesion. Anti-Ly-49 antibodies inhibited the interaction of both Ly-49A (YE1/32 , YE1/48) and Ly-49C (5E6) with the H-2 d cell line (Figure 15A-B). 75 H-2d H-2b H-2S H-2k Ly-49A Ly-49C p AX 142 0 10 20 0 10 20 0 10 20 30 0 .2 .4 cpm bound (x 1000) Figure 14. Quantitation of cell-cell binding mediated by Ly-49A or Ly-49C. Cell lines representing four haplotypes (H-2d/A20, H-2VIC-21, H-2S/GM979, H-2 k/Rl.l) were labeled with 5 1 C r and overlayed on COS cells transfected with an Ly-49 cDNA or vector alone (pAX142). Plates were washed after two hours and radioactivity was determined. The numbers presented are the mean ± SD of triplicate plates. 76 Ly-49A:H-2d B Ly-49C:H-2d 20-8-4S (Kd) 34-2-12S(Dd) 34-5-8S (Dd) 34-1-2S (Kd/Dd) | YE1/32 (Ly-49A) |H YE1/48 (Ly-49A) B8-24-3 (Kb) K7-65 (Kb) Y3 (Kb) 20-8-4S(Kb) 28-14-8S(Db) 28-11-5S (Db) 5E6 (Ly-49C) 20-8-4S (Kd) 34-2-12S(Dd) 34-5-8S (Dd) 34-1-2S (Kd/Dd) 20-8-4S+34-2-12S 20-8-4S+34-5-8S 5E6 (Ly-49C) 20 40 60 80 100 120 Ly-49C:H-2b D 20 40 60 80 100 120 Ly-49C:H-2S 34-1-2S (H-28) 3-83P (H-2S) Ii 5E6 (Ly-49C) i—i—i—r 20 40 60 80 100 120 i i i i r~ 20 40 60 80 100 120 % Control Adhesion Figure 15. Distinct class I MHC specificities of Ly-49A and C. 5 1 C r labeled cell lines (H-2D, H-2B, H-2S) were incubated for two hours with Ly-49A or C transfected COS cells in the presence of antibodies against either class I MHC or Ly-49 (the antigenic specificities of each antibody is in parentheses next to the clone name). 100% control adhesion represents binding in the absence of antibody, and 0% adhesion is the level of the cell line binding to COS cells transfected with vector alone. Values presented are the mean ± S E M of at least three independent experiments. 77 A panel of well characterized anti-H-2b antibodies has shown that those which recognize K*3 entirely inhibit Ly-49C-mediated adhesion, whereas antibodies against D b are non-inhibitory (Figure 15C). Binding of an H-2S cell line was also inhibited by antibodies against class I MHC (Figure 15D), although the nature of the H-2S structures recognized by these antibodies is not characterized. Additionally, 5E6 completely blocked adhesion to both the H-2 b and H-2S cell lines (Figure 15C-D). 4.2.3) Characterization of Ly-49A-class I MHC interactions in a two way transfection system In addition to the antibody blocking studies described above, the class I specificity of Ly-49A was also characterized in a two way transfection cell adhesion assay. Cell lines permanently expressing transfected class I M H C genes (Dd, D b , and Dk) were established and tested for binding to Ly-49A expressing COS cells (Figure 16). An H-2S cell line (GM979) was chosen for class I transfections because of its inability to bind to Ly-49A. As shown in Figure 17, only the combination of Ly-49A expressed on COS cells in conjunction with either D d or D k on GM979, but not D b , resulted in cell-cell interactions. This specificity of Ly-49A for D d and D k was also quantitated by 5 1 C r labeling of test cells as shown in Figure 18. 4.3) Discussion The results presented in this chapter demonstrate that Ly-49A and Ly-49C bind to distinct but overlapping subsets of class I molecules. Ly-49C was found to recognize a wider range of class I MHC than Ly-49A, including K d , D d , and K*3, but not D b (of these, Ly-49A bound only Dd). Binding to as yet undefined structures on H-2S and H-2 k cell lines was also demonstrated by Ly-49C. These findings support the hypothesis that Ly-49 is a family of N K cell receptors with related, but distinct functions: Ly-49A and Ly-49C are both expressed by N K cells, but not by the same 78 GM979 GM.Db GM.Dd GM.Dk noAb 34-1-2S 3-83P 34-5-8S 28-14-8S 15-5-5S L _ L L L L 1 I LA Figure 16. Expression of class I M H C by GM979 and its H-2D b , D d , and D k transfectants. The H-2 8 cell line GM979 was transfected with genes encoding D b , D d , and D k as described in the Materials and Methods (Sections 2.8 and 2.9.2). The antibodies 34-1-2S and 3-83P recognize endogenous H-2S structures. The remaining antibodies are specific for D d (34-5-8S), D b (28-14-8S), or D k (15-5-5S). All analyses required a secondary staining step with an FITC-labeled anti-mouse IgG antibody. 79 pAX142 Ly-49A B GM979 GM.D d GM.D k a* ^ « ' 4 % ' -"6 •-• •* ° " • k e *.f \ 0 * \ ST* » y M i w H GM.D b Figure 17. Adhesion of Ly-49A to GM979 expressing transfected class I M H C genes. COS cells transfected with the vector pAX142 (A, C, E , G) and Ly-49A (B, D, F, H) were overlayed with the cell lines GM979 (A, B), GM.D d (C, D), G M . D k (E, F), and G M . D b (G, H). Cells were photographed after five washes following a 2 hour co-incubation at 37°C. 80 Figure 18. Quantitation of Ly-49A-H-2Dd/Dk mediated cell adhesion. GM.D d , GM.D b , and G M . D k were each labeled with 5 1 C r and overlayed on COS cells transfected with the Ly-49A cDNA or vector alone (pAX142). Plates were washed after two hours and radioactivity was determined. The numbers presented are the mean ± SD of triplicate plates. 81 NK cells (Chapter 3); they both bind to class I MHC molecules, but not the same class I molecules (this chapter). The binding pattern of Ly-49A correlates well with the data on lysis by Ly-49A + N K cells. These cells kill H-2b targets, but not H-2 d or H-2 k (Karlhofer et al., 1992). My data demonstrates that a ligand for Ly-49A is expressed on H-2 d and H-2k, but not H-2 b cell lines. The lack of binding to H-2 b implies a lack of protective class I molecules, which explains why these cells are killed by Ly-49A + N K cells. Ly-49 A also failed to bind the H-2S and H-2 a cell lines. The H-2 a haplotype is a hybrid one (K*1, Dd), and it was therefore unexpected that Ly-49A would not bind a cell expressing D d . Interestingly, these H-2 a cell lines were the only ones tested to which Ly-49C did not bind, although binding was observed to H-2k, H-2d, H-2b, and H-2S lines. It is unknown if the inability of these H-2 a cell lines to bind Ly-49A and Ly-49C is a peculiar property of these particular cell lines, or the H-2 a haplotype in general. The relationship between class I M H C recognition by Ly-49C and the function of Ly-49C + N K cells has been more difficult to establish than was the case for Ly-49A. This may be a result of the much broader range of class I molecules recognized by Ly-49C. Recent studies by Bennett et al. (1995) have utilized an in vitro system which reproduces the in vivo specificity of Ly-49C + (5E6+) N K cells originally described by Sentman et al. (1989). Ly-49C + N K cells derived from (C57BL/6 x BALB/c)Fi (CB6F1) mice were found to lyse parental BALB/c (H-2d), but not B6 (H-2b), Con A blasts (Bennett et al., 1995). Ly-49C has been proposed to transmit negative signals upon interaction with K?. This is supported by the observation that K?3 bearing targets were lysed following pre-incubation with anti-5E6 (Fab')2, implying that the delivery of a negative signal through Ly-49C was being blocked. Additionally, Ly-49C + N K cells derived from a homozygous BALB/c mouse were unable to lyse BALB/c blasts, and pre-incubation with anti-5E6 resulted in the lysis of these targets as well (Bennett et al., 1995). 82 The suggestion that Ly-49C interacts with class I and delivers negative signals is supported by my binding studies which have demonstrated that Ly-49C binds to H-2 b (K^ but not Db) and H-2 d (K d and Dd) structures. However, the lysis of BALB/c blasts by CB6F1 Ly-49C + N K cells indicates that H-2 d class I molecules cannot, at least under some circumstances, provide the proper protective signal to Ly-49C + N K cells. One explanation for this apparent paradox is that allelic forms of Ly-49C may have different class I specificities and or affinities. The Ly-49C used in these binding studies is the BALB/c allelic form, and differs from the B6 form by 14 extracellular amino acids (Held et al., 1995; Stoneman et al., 1995). Four of these differences fall in a 19 amino acid stem region which has been implicated in the determination of Ly-49 ligand specificity (Mahon, 1995). The allelic exclusion which has recently been demonstrated for Ly-49A and C (Held et al., 1995) predicts that a CB6F1 mouse has two subsets of Ly-49C + N K cells: L y - 4 9 C B A L B and Ly-49CB 6. The Ly -49C B 6 fraction may therefore be responsible for reactivity against BALB/c targets, whereas both the L y - 4 9 C B A L B and Ly-49C B 6 fractions appear to be inhibited by H-2 b antigens (Bennett et al., 1995; Sentman et al., 1989). An alternate explanation for the reactivity of CB6F1 Ly-49C + N K cells against homozygous H-2 d targets may be related to the receptor calibration phenomenon recently described for Ly-49A. Expression of Ly-49A on individual N K cells has been found to be downregulated in mice bearing its class I ligand, D d (Olsson et al., 1995). Although these Ly-49A + N K cells are inhibited by D d expressed on self cells, the minimum level of class I on the target cell surface required for recognition is higher than that required by Ly-49A + effectors educated in an H-2 b mouse. This sort of receptor calibration may also regulate the cell surface expression of Ly-49C and thereby affect class I recognition by these effectors. Therefore, if Ly-49C interacts more strongly with H-2 b than with H-2 d antigens, the cell surface levels of this receptor may be reduced in an H-2 b / d mouse 83 to such an extent that homozygous H-2 d cells would be lysed as a result of their inability to provide an inhibitory signal. Resolution of these questions will require an evaluation of the binding properties of the Ly-49C B 6 receptor as well as a more precise characterization of the means by which class I MHC affects lysis by Ly-49C+ N K cells. 84 Chapter 5 Carbohydrate Recognition by Ly-49C Data presented in this chapter has been incorporated into the following manuscript: Brennan, J . , F. Takei, S. Wong, and D. Mager. 1995. Carbohydrate recognition by a natural killer cell receptor, Ly-49C. J . Biol. Chem. 270:9691. 85 5.1) Introduction Recent studies have implicated a group of molecules containing C-type lectin domains as potential N K cell receptors. These include the genetically linked Ly-49 (mouse) (Chapters 3-4; Chan and Takei, 1989; Smith et al., 1994; Wong et al., 1991; Yokoyama et al., 1989, 1990), NKR-P1 (mouse, rat, and human) (Giorda et al., 1990; Giorda and Trucco, 1991; Lanier et al., 1994; Yokoyama et al., 1991), and NKG2 (human) (Houchins et al., 1991) multigene families. These genes encode type II transmembrane proteins, with a distal extracellular region homologous to the carbohydrate recognition domain (CRD) of C-type lectins. Although members of each multigene family are highly related at both the nucleotide and amino acid level, sequence comparisons between multigene families show low but significant similarity only at the amino acid level, and this is mainly limited to the lectin-like domain (-25% identity) (Houchins et al., 1991; Wong et al., 1991; Yokoyama et al., 1991). The Ly-49 multigene family comprises at least eight distinct genes. These related molecules have been termed Ly-49A-H, and are between 49-91% identical at the amino acid level (Smith et al., 1994; Wong et al., 1991). Ly-49A and Ly-49C expression defines distinct single positive and double positive subsets of N K cells (Chapter 3), and both function as receptors for class I MHC molecules (Chapter 4; Daniels et al., 1994a; Kane, 1994). Ly-49A binds to purified, immobilized H-2D d and D k (Kane, 1994), and the recognition of these antigens on the target cell appears to deliver a negative signal to the N K cell which prevents lysis (Karlhofer et al., 1992). This observation is concordant with the prediction that N K cells possess receptors which deliver negative signals upon interaction with class I MHC on the target cell surface (Ljunggren and Karre, 1990). Ly-49C, recognized by the 5E6 monoclonal antibody, binds to class I M H C on cells of diverse haplotypes (Chapter 4), whereas Ly-49A has been shown only to bind H - 2 d a n d k structures (Chapter 4; Daniels et al., 1994a; Kane, 1994). Ly-49C 86 has been implicated in the H-2 directed (and N K cell-mediated) phenomenon of hybrid resistance. Ly-49C+ (5E6+) N K cells eliminate H-2d, but not H-2b, bone marrow cells when transplanted into irradiated hosts (Sentman et al., 1989). Recent in vitro studies have suggested that recognition of class I M H C by Ly-49C may, like Ly-49A, be inhibitory to the N K cell (Bennett et al., 1995). In this chapter I have shown that cell adhesion mediated by Ly-49C is inhibited by several sulfated glycans, each of which blocks the binding of the monoclonal antibody 5E6, and that treatment of target cells with fucosidase drastically reduces their adhesion to Ly-49C expressing COS cells. 5.2) Results 5.2.1) Sulfated polysaccharides inhibit Ly-49C-mediated adhesion In the previous chapter it was demonstrated that Ly-49C expressed on COS cells binds to cell lines of various haplotypes (H-2b / d / k / s), and that these interactions are inhibited by anti-class I M H C and anti-Ly-49C antibodies (Chapter 4). In this chapter I have used this system to test the potential carbohydrate recognition properties of Ly-49C which are suggested by its amino acid similarity to known lectins (Wong et al., 1991; Yokoyama et al., 1989). Several polysaccharides were tested for their ability to inhibit the adhesion of GM979 to COS cells transfected with the Ly-49C cDNA (Figure 19). Of those tested, dextran sulfate (mw 500,000), fucoidan, and A,-carrageenan were found to inhibit at concentrations ranging from 5xl0-10 to 2xl0-8 M (Table 4). Those polysaccharides which inhibited adhesion are all negatively charged sulfated glycans (DeAngelis et al., 1987; Parish et al., 1988). An important role for sulfate is indicated by the failure of non-sulfated dextran to inhibit at a concentration 400 times higher than the ID50 value determined for dextran sulfate (Table 4). Chondroitin sulfates A, B, and C, and heparan sulfate represent a panel 87 1 0 - i i 10-10 10-9 10-8 10-7 10-6 105 104 Polysaccharide Concentration (M) Figure 19. Polysaccharide dose response inhibition of Ly-49C-mediated cell adhesion. 0% inhibition corresponds to adhesion of the cell line GM979 ( 5 1Cr labeled) to Ly-49C transfected COS cells in the absence of carbohydrate, and 100% inhibition is equivalent to GM979 binding to COS cells transfected with the vector pAX142. COS cells were incubated simultaneously with GM979 and dextran sulfate (mw 500,000), X-carrageenan, fucoidan, dextran (a), heparan sulfate (b), hyaluronan (c), chondroitin sulfate B (d), chondroitin sulfate A (e), or chondroitin sulfate C (f). Each point represents the mean + S E M of at least 3 independent experiments. 88 Table 4. Polysaccharide Inhibition of Ly-49C-Mediated Adhesion Polysaccharide Molecular SO4 groups/ Charge ID50 (M)c Weight Monomer* Densityb Fucoidan 100,000 2 0.3 2 x IO 8 Dextran Sulfate 500,000 4.6 2.8 5 x 10-10 X-C arr ageenan 300,000 3 0.8 3 x IO 9 Hyaluronan 160,000 0 0.5 >3.1xl0-6d Heparan Sulfate 7,500 1-2 >1.3xl0- 6 d Chondroitin Sulfate A 25,000 1 1.0 >2x 10-5d Chondroitin Sulfate B 25,000 1-2 1.0 >4x 10-6d Chondroitin Sulfate C 25,000 1 1.0 >2xl0 -s d Dextran 500,000 0 >2x 10 7 d a Parish et al., 1988 b negative charge per monomer (DeAngelis et al., 1987) c ID50 represents the concentration giving 50% inhibition of adhesion between COS cells expressing Ly-49C and the cell line GM979 d maximum concentration tested gives <50% inhibition 89 of sulfated anionic polysaccharides which do not affect Ly-49C mediated adhesion. Hyaluronan is of intermediate size and charge compared to fucoidan and X-carrageenan (DeAngelis et al., 1987). Its lack of inhibitory activity in this system rules out the possibility of a non-specific charge effect. 5.2.2) Polysaccharide inhibition of5E6 binding to Ly-49C The monoclonal antibody 5E6 appears to recognize a functional domain of Ly-49C, as demonstrated by its ability to completely inhibit Ly-49C-mediated cell adhesion (Chapter 4). The polysaccharide inhibition seen in the cell binding assays suggests that certain carbohydrates may specifically interact with Ly-49C. I therefore tested the series of carbohydrates which strongly inhibit Ly-49C mediated adhesion for their ability to compete with the antibody 5E6 for binding to the molecule. COS cells expressing Ly-49C were pre-incubated with each polysaccharide for 1 hour, followed by the 5E6 antibody for 30 minutes. Similar to what was observed in the cell-cell assay, dextran sulfate (mw 500,000), fucoidan, and X-carrageenan all significantly inhibited the binding of 5E6 to Ly-49C, whereas hyaluronan had no effect (Figure 20). As a control, the common form of CD44 (CD44H) was expressed on COS cells and the same set of experiments was performed using the anti-CD44 antibody 3cl2. All polysaccharides tested had no effect on the interaction of 3cl2 with CD44, indicating that the effect is both Ly-49C and carbohydrate specific. These results suggest that Ly-49C binds to dextran sulfate, fucoidan, and A-carrageenan. 5.2.3) Glycosidase treatment ofGM979 To directly evaluate the role of cell surface carbohydrates in Ly-49C-mediated adhesion, GM979 cells were treated with fucosidase or neuraminidase, and tested in the cell-cell adhesion assay. Treatment of cells with fucosidase 90 Fucoidan ^-Carrageenan Dextran Sulfate Hyaluronan 0 20 40 60 80 100 120 % Control Fluorescence Figure 20. Competition of antibody binding by polysaccharides. COS cells expressing Ly-49C or CD44H were exposed to fucoidan, X,-carrageenan, dextran sulfate (mw 500,000), or hyaluronan for 1 hour at room temperature, followed by the antibodies 5E6 (Ly-49C) or 3cl2 (CD44H) for 30 minutes. Cells were washed twice, analyzed by FACS®, and mean fluorescence was determined for each condition. Antibody binding is shown as the percentage relative to control levels (in the absence of carbohydrate). Values shown are the mean ± S E M of at least 3 independent transfections, stainings, and FACS® analysis. 91 resulted in a dose-dependent decrease in cell adhesion (Figure 21). Cell surface class I MHC levels were unaltered by fucosidase treatment, as evaluated by staining with the Ab 34-1-2S (Figure 22A). Removal of sialic acid resulted in an enhanced adhesion to Ly-49C expressing COS cells (Figure 21). The increased binding of desialylated cells was Ly-49C specific, because neuraminidase treated GM979 showed no altered binding to control COS cells. The effectiveness of the sialidase was verified by the binding of Peanut Lectin only to enzymatically treated cells (Figure 22B). This lectin binds to the terminal galactose of glycoproteins following neuraminidase treatment. One of the defining features of C-type lectins is that ligand binding is calcium dependent (Drickamer, 1988; Drickamer and Taylor, 1993). Upon testing Ly-49C-mediated adhesion for this property, I have found that there is no requirement for exogenous Ca + + , as demonstrated by the persistent binding in 5 mM E G T A (Figure 21). Although this is in sharp contrast to nearly all C-type lectins studied, it appears to be a property common to the N K cell lectin branch of this superfamily (Bezouska et al., 1994a; Daniels et al., 1994b). 5.3) Discussion Ly-49C, like all members of the Ly-49 multigene family, contains a domain homologous to C-type animal lectins (Wong et al., 1991). Many proteins have been shown to bind carbohydrates through this conserved domain, including the hepatic lectins, mannose binding proteins, and the selectins (Drickamer, 1988; Drickamer and Taylor, 1993). Experiments presented in this chapter provide evidence that Ly-49C is also a carbohydrate binding protein. Interestingly, those polysaccharides that are recognized by Ly-49C also inhibit selectin functions (Varki, 1994). Selectins and Ly-49 share amino acid homology only in the putative CRD (-20% identity), and no nucleotide similarity (Wong et al., 1991). Both fucoidan and dextran sulfate inhibit adhesion mediated by L- and P-selectin at 92 Fucosidase (units/ml) Sialidase (units/ml) 5 mM EGTA Control 0 40 80 120 160 % Control Adhesion Figure 21. Effects of glycosidase treatment and calcium depletion on Ly-49C-mediated adhesion. GM979 cells were treated with neuraminidase or fucosidase as described in the materials and methods. In the calcium depletion experiments, adhesion was tested in C a + + free HBSS containing 1% BSA/2 mM MgCl2/5 mM EGTA. 100% control adhesion is equal to binding of GM979 in the absence of treatments, and 0% adhesion is the level of binding detected to COS cells transfected with the vector pAX142 Values shown are the mean ± SEM of at least 3 independent experiments. { { 0.28 0.19 0.40 0.22 93 A Log Fluorescence Figure 22. Effects and effectiveness of glycosidase treatments. (A) GM979 cells were treated with 0.28 units/ml of fucosidase, washed, stained with the anti-class I M H C antibody 34-1-2S, washed, stained with a secondary FITC-labeled anti-IgG antibody, washed, and then analyzed by flow cytometry. NS indicates non-stained cells (B) GM979 cells were treated with neuraminidase, washed, stained with FITC-labeled Peanut Lectin, washed, and then analyzed by flow cytometry. Empty histograms represent unstained cells and black histograms are cells treated with the amount of enzyme shown (units/ml) and then stained with Peanut Lectin. 94 concentrations nearly identical to the ID50 values of Ly-49C (Rochon et al., 1994; Varki, 1994; Yednock et al., 1989). Fucoidan also potently blocks the binding of the antibody Mel-14 to L-selectin (Yednock et al., 1987). In this study, those polysaccharides that inhibit cell adhesion were also found to inhibit the binding of 5E6 to Ly-49C, suggesting that 5E6, like Mel-14 binding to L-selectin, recognizes the CRD ofLy-49C. My studies have shown that A-carrageenan is an equally potent inhibitor of Ly-49C-mediated adhesion. The monosaccharide cores of the inhibitory polysaccharides are fucose, galactose, and glucose (fucoidan, X,-carrageenan, and dextran sulfate, respectively) (DeAngelis et al., 1987; Patankar et al., 1993). These polysaccharides are obviously not the natural ligands for Ly-49C, as they are not found in mammalian cells. It remains to be seen if the natural ligands consist of these monosaccharides, or if they are structurally analogous to the true ligand(s). Previous studies have shown that many sugars inhibit N K activity in a standard cytotoxicity assay. Both soluble and cell surface carbohydrates appear to interfere with cytotoxicity by inhibiting the activation of lysis at a post cell binding stage (McCoy and Chambers, 1991). Without having clarified the mechanisms of N K cell recognition, these studies suggested that N K cells may possess lectin receptors which recognize carbohydrate structures on the target cell surface. The recent molecular characterization of lectin-like receptors expressed by N K cells (Ly-49, NKR-P1, and NKG2) has provided a basis by which carbohydrate binding properties of putative N K cell receptors can be evaluated. Carbohydrate recognition has recently been reported for both rat NKR-P1 (Bezouska et al., 1994a, 1994b) and Ly-49A (Daniels et al., 1994b). Interestingly, these molecules appear to behave differently than all other C-type lectins (Drickamer, 1988; Drickamer and Taylor, 1993) in that normal ligand binding occurs in the absence of exogenous C a + + (Bezouska et al., 1994a). Decalcification of the purified proteins by high pH dialysis has shown that carbohydrate binding is 95 indeed C a + + dependent, but that C a + + is tightly bound at pH 8, and is not released in the presence of E G T A or EDTA (Bezouska et al., 1994a; Daniels et al., 1994b). I have also observed that Ly-49C-mediated adhesion is not inhibited by the removal of exogenous C a + + . N K cell lectin-like receptors define one of at least five distinct subgroups of the C-type lectin superfamily (Chamber et al., 1993). It appears that one property of this class may be a high affinity for C a + + , which expresses itself as exogenous calcium-independent ligand binding. Carbohydrate recognition by NKR-P1 appears to activate N K cell lytic functions (Bezouska et al., 1994b), which follows from previous studies demonstrating that cross-linking of the antigen activates N K cells (Chambers et al., 1989). Ly-49C and NKR-P1 are distantly related molecules, which suggests that they have distinct ligand specificities. Indeed, both chondroitin sulfate and sulfated heparin oligosaccharides inhibit NKR-P1 binding at very low concentrations (Bezouska et al., 1994b), but have no effect on Ly-49C. I have found that fucosidase treatment of GM979 significantly reduced its adhesion to Ly-49C (Figure 21). This observation, along with the inhibition by fucoidan, suggests that fucose may serve as a ligand for Ly-49C. Daniels et al. have shown that Ly-49A mediated adhesion is also inhibited by fucoidan, at concentrations similar to the Ly-49C ID50 value (Daniels et al., 1994b). Additionally, tunicamycin-treated target cells showed reduced adhesion to Ly-49A and increased sensitivity to Ly-49A+ N K cells (Daniels et al., 1994). I have seen that tunicamycin also reduces Ly-49C-mediated adhesion, but consider this observation difficult to interpret due to the associated decrease in class I MHC expression. Daniels et al. also reported that sialidase treatment of D d + target cells had no effect on their adhesion to Ly-49A (Daniels et al., 1994b). Contrary to this, Ly-49C adhesion is increased following sialic acid removal (Figure 21), suggesting that a precursor structure may be masked by sialic acid, the exposure of which allows 96 for a stronger binding. This is reminiscent of another C-type lectin, the asialoglycoprotein receptor, which binds to the terminal galactose of glycoproteins following the loss of sialic acid (Drickamer, 1988; Drickamer and Taylor, 1993). Alternatively, the enhanced adhesion may simply be charge related, resulting from the removal of negatively charged sialic acid moieties. Control COS cells (not expressing Ly-49C), however, showed no increased binding to sialidase treated cells. Nonetheless, Ly-49C recognition is clearly distinct from the sialic acid dependent adhesion mediated by the selectins (Varki, 1994). Ly-49A and C have both been shown to bind to class I M H C on target cells (Chapter 4; Daniels et al., 1994a; Kane, 1994). The highly polymorphic al/a2 domain of this molecule has repeatedly been shown to provide target cells with protection from lysis by N K cells (Cella et al., 1994; Ciccone et al., 1992; Colonna et al., 1993a, 1993b; Sentman et al., 1994; Storkus et al., 1991), and is likely the region of H-2D d recognized by Ly-49A (Daniels et al., 1994a; Karlhofer et al., 1992). A recent report by Correa and Raulet (1995) has carefully examined the possibility that Ly-49A may recognize specific peptides bound by D d . This work has established that Ly-49A recognizes all class I-peptide complexes, rather than any particular subset of them. It is clear that although recognition is not peptide specific, it is peptide dependent, suggesting that a conformation consisting of an a/(3 chain plus peptide is recognized by Ly-49A. All class I MHC molecules contain a conserved N-linked glycosylation site (Asn 86) which lies near the junction of the al/a2 domain. Murine class I molecules have an additional conserved N-linked glycan at Asn 176, and no O-linked sugars (Kimball et al., 1983). Although the oligosaccharide makeup of class I molecules is not well characterized, they are known to be complex sialylated, fucosylated structures (Alting-Mees and Barber, 1986; Kimbal et al., 1983; Misra et al., 1987; Swiedler et al., 1985). If Ly-49 molecules are indeed recognizing cell 97 surface carbohydrates, it is likely that such structures are class I MHC-associated and are located in the ocl/a2 domain. Class I MHC oligosaccharides are known to be heterogeneous, as displayed by the distinct carbohydrate structures of and D k at conserved N-linked sites (Swiedler et al., 1985). It is therefore possible that these oligosaccharide differences (which presumably result from different amino acid sequences) are the basis of Ly-49 recognition. It is also possible that Ly-49 recognizes the combination of a specific carbohydrate and protein scaffolding, as appears to be the case for both L- and P-selectin (Varki, 1994). Ly-49A appears to act as a negative regulator (Karlhofer et al., 1992), and recent studies have suggested that Ly-49C recognition is also inhibitory to the N K cell (Bennett et al., 1995). It has been proposed that N K cell target recognition may be a complex mixture of positive and negative signals, and that these signals may be delivered by molecules such as NKR-P1 and Ly-49 (Yokoyama and Seaman, 1993). The characterization of Ly-49C as a carbohydrate binding protein in conjunction with similar recent findings for NKR-P1 and Ly-49A suggests that lectin interactions may be central to this process. 98 Chapter 6 Summary and Perspectives 99 The field of N K cell biology has undergone remarkable advancements in recent years. At the time that the studies in this thesis were initiated, the mechanisms of target cell recognition by N K cells were largely unknown. Today, N K cell receptors for class I MHC have been identified in both human and mouse. These studies have revealed several layers of complexity in the N K cell recognition system. The most obvious of these is that humans and mice possess molecularly unrelated receptors (Table 5) that perform what appear to be the same function (inhibitory signaling associated with class I recognition). In humans a novel gene family (NKAT/p58) has been identified that encodes cell surface molecules with external immunoglobulin domains (Colonna and Samaridis, 1995; D'Andrea et al., 1995; Wagtmann et al., 1995). These receptors have been shown to recognize distinct subsets of HLA-B and HLA-C alleles. The HLA-C receptors (EB6 and GL183) have two immunoglobulin domains and the HLA-B receptor (NKB1) has three immunoglobulin domains. Recognition by these receptors appears to have at least some specificity for the peptide bound by class I M H C (Malnati et al., 1995), but is not dependent upon oligosaccharides attached to Asn 86 (Gumprez et al., 1995). Although less well characterized, studies have suggested that CD94, a member of the C-type lectin superfamily (Chang et al., 1995), may also be a human H L A receptor (Moretta et al., 1994; Perez-Villar et al., 1995). In mice, all candidate N K cell receptors for class I M H C are members of the Ly-49 family (Table 5). In the introduction of this thesis it was proposed that other members of this family (in addition to Ly-49A) are expressed by N K cell subsets and recognize class I MHC. As described in Chapters 3 and 4, Ly-49A and Ly-49C were found to be expressed on overlapping N K cell subsets and to bind to different class I molecules. Recent studies of the N K cell antigen LGL-1 have shown it to be encoded by Ly-49G.2, and have suggested that this receptor may recognize both L d and D d (Mason et al., 1995). Ly-49G.2 function has been investigated by in vitro cytotoxicity assays, but a direct physical interaction with class I M H C has not yet 100 Table 5. NK Cell Receptors for Class I MHC Receptor Molecular Structure IgSF C-Type Lectin Ligand A) Human EB6 GL183 NKB1 CD94 58 kD monomer S 58 kD monomer S 70 kD monomer S 70 kD dimer HLA-Cw3* HLA-Cw4* HLA-Bw4* HLA-B7* B) Mouse Ly-49A Ly-49C Ly-49G.2 90 kD dimer 110 kD dimer 90 kD dimer / D d , D k • K>, K d , D d , H-2*, H-2 k ^ (L d, Dd)* * apparent ligand based upon inhibition in N K cell cytotoxicity assays (physical interaction not demonstrated) 101 been demonstrated. Similar to what has been seen for Ly-49A and Ly-49C, Ly-49A and Ly-49G.2 have been found to be expressed on overlapping N K cell subsets (Mason et al., 1995). These observations together suggest that a repertoire of N K cells exists whose specificity for class I MHC is dictated by the Ly-49(s) expressed by its individual clones. Those Ly-49s which have been characterized thus far have overlapping but distinct specificities for class I molecules, and N K cells are either single positive or double positive for these receptors. The basic function of these receptors is presumably to recognize cells which have lost expression of class I and would otherwise evade detection by CTLs. N K cells as a whole may possess multiple class I receptors to ensure the existence of N K cells which can recognize cells which lose expression of individual class I molecules. For example, in an H-2 d mouse Ly-49A+ N K cells may recognize the absence of D d (but not K d or L d ), and in an H-2 b mouse Ly-49C + N K cells may recognize the loss of K*5 (but not Db). Because Ly-49 genes do not rearrange to generate a large pool of potential receptors, the range of Ly-49 specificities for class I is expected to be by necessity quite broad. Like class I MHC, the Ly-49 family is also polymorphic (Held et al., 1995) and different alleles are thought to encode receptors with differing specificities for class I MHC (Bennett et al., 1995; Stoneman et al., 1995). A given Ly-49 is likely to recognize multiple allelic forms of a class I molecule, and may even recognize more than one class I molecule encoded by a single haplotype (Chapter 4). Such broad specificities would ensure that an individual will be able to formulate an N K cell repertoire with receptors for self class I M H C . The size of the Ly-49 gene family (approximately 10 members) combined with the broad specificities of these receptors makes it likely that multiple clones will arise that are specific for single class I molecules. To illustrate this concept, Table 6 shows a hypothetical chart describing the nature of Ly-49-class I interactions. This table provides artificial gene 102 designations for both Ly-49 (1-10) and class I MHC (I-III). The fact that an Ly-49 may recognize one, more than one, or no class I molecules of a given haplotype has been represented by a random assignment of specificities. Because both Ly-49 and class I are polymorphic and this polymorphism affects their ability to bind to one another, changes in the alleles of either may change the pattern of binding specificities. Using the material provided in Table 6, an N K cell repertoire may be formed as is described in Table 7. Class I molecules function as the target structures and the pool of Ly-49s are the raw materials used to develop the various N K cell clones (designated A-I). Although the mechanisms governing the regulation of Ly-49 expression are unknown, N K cells may express one or more of these receptors. A combination of these principles predicts the development of a series of N K cell clones expressing one or multiple Ly-49s and recognizing one or multiple class I molecules. Although some N K cells may express multiple Ly-49 genes, expression of a given Ly-49 gene has been found to be allele specific within an individual N K cell (Held et al., 1995). This allelic exclusion presumably functions to prevent the development of multiple receptor specificities within individual N K clones in Ly-49 heterozygous mice. Calibration of the level of expression of Ly-49 also appears to regulate the function of the N K cell (Olsson et al., 1995). In Chapter 3, Ly-49A+ cells were shown to persist in mice expressing D d and D k , as opposed to previous reports which described the absence of these cells in H-2 d and H-2 k mice (Karlhofer et al., 1994; Sykes at el., 1993). Recent studies have shown that the cell surface density of Ly-49A is in fact reduced on an N K cell that is educated in an environment where its class I ligand (Dd) is present (Olsson et al., 1995). This process results in Ly-49A+ N K cells that can detect subtle changes in the levels of D d on the surface of a target cell. Ly-49A+ N K cells educated in an H-2 b mouse are Ly-49A "high" and are inhibited by D d "bright" as well as D d "dull" cells, whereas 103 Table 6. Hypothetical Interactions Between Ly-49 Receptors and Class I M H C Molecules Class I M H C (allele v) I II III Ly-49 (allele x) 1 X 2 X 3 4 X X 5 X 6 7 X 8 X X 9 10 X Arabic numbers for Ly-49 and roman numerals for class I M H C represent a hypothetical set of randomly chosen alleles for each of the members of these families. X denotes a combination in which a given Ly-49 can bind the designated class I molecule. 104 Table 7. Hypothetical Development of the NK Cell Repertoire NK Cell Clones Ly-49 Receptors Class I Specificities A 1 II B 2 I C 4 I, III D 5 III E 7 II F 8 I, II G 10 III H 1,5 II, III I 2,8 I, II N K cell clones are given arbitrary letter designations (A-I). Each clone may express one or more Ly-49 receptors and may have one or more specificities for class I MHC. The Ly-49 and class I molecules used in this model are described in Table 6. 105 the same N K cell population educated in a D d + mouse is Ly-49A "low" and is inhibited only by D d "bright" cells. It is not yet known if other members of the Ly-49 family, such as Ly-49C and Ly-49G.2, also undergo this type of receptor calibration with respect to their class I ligands. An important outstanding question in this area involves the basic processes of N K cell education. Although it is known that N K cells can detect the loss of individual class I molecules and that such recognition appears to be mediated by Ly-49, it is unknown how N K cells are taught to recognize self. Some mechanism must ensure that all N K cells either express receptors for self class I MHC or are rendered non-functional if their receptors have no affinity for self class I molecules. Such mechanisms have been suggested to exist by the nature of N K cells that develop in ^2m -/- mice and in D8 mice (C57BL/6 mice transgenic for Dd). N K cells of ^2m -/- mice are rendered self-tolerant, even though these mutant mice are essentially class I negative (Bix et al., 1991; Hoglund et al., 1991; Liao et al., 1991). N K cells derived from normal mice, however, readily lyse cells from their P2m -/-counterparts. In D8 mice (Ohlen et al., 1989), a subset of N K cells is taught that D d is a self class I molecule and will therefore lyse C57BL/6 targets (H-2b) which lack this inhibitory ligand. A recent study by Correa and Raulet (1995) has investigated the importance of cellular peptides bound by D d as Ly-49A recognition structures. These studies have shown that although peptides are indeed required for the inhibition of Ly-49A+ N K cells, most if not all peptide-class I complexes provide a target cell with protection. Ly-49A is therefore thought to recognize a class I conformation that is peptide dependent but not peptide specific. This is apparently in contrast to the human NKAT/p58 receptors which have at least some specificity for the peptides associated with particular class I molecules (Malnati et al., 1995). 106 As described in Chapter 5, Ly-49C is capable of carbohydrate recognition and may recognize oligosaccharide structures associated with class I molecules. Similar findings have also been reported for Ly-49A (Daniels et al., 1994b). Because Ly-49A and C appear to recognize the polymorphic al/a2 domain of class I MHC, the conserved N-linked glycan at Asn 86 in the a l domain (see Figure 2) is a strong candidate for an Ly-49 recognition structure. Although these studies suggest that Ly-49 may bind to class I-associated oligosaccharides, the specificity of such an interaction may also involve the recognition of the peptide backbone of class I MHC. Nonetheless, recognition of carbohydrates by Ly-49 suggests that the presence of class I on a target cell surface may be insufficient to inhibit lysis by N K cells. The glycosylation status of the peptide may also be of importance and may provide a signal for the recognition of aberrant cells. The results presented in this thesis have contributed to the rapidly progressing field of N K cell biology. Although discovered in 1975, N K cells remained poorly characterized until the late 1980's. Mounting evidence of an inhibitory specificity for class I MHC suggested that N K cells constitute a third class of lymphocytes, devoid of T and B cell antigen receptors. Investigations by many laboratories into the nature of the N K cell receptor led to the identification of the many structures described in Table 5. Studies in the upcoming years promise further insight into the education processes that formulate the N K cell repertoire as well as an understanding of the relationship between the NKAT/p58 and Ly-49 families of N K cell receptors. It will be of particular interest to learn whether the corresponding human Ly-49 and murine NKAT/p58 genes have simply not yet been discovered, or if these species have developed distinct receptor systems through a process of convergent evolution. 107 References Abbas, A.K., A .H. Lichtman, and J.S. Pober. 1991. Cellular and molecular immunology. W.B. Saunders Company, Philadelphia, 417 pp. Adamkiewicz, T.V., C. McSherry, F .H. Bach, and J.P. Houchins. 1994. Natural killer lectin-like receptors have divergent carboxy-termini, distinct from C-type lectins. Immunogenetics. 39:218. Allen, H. , D. Wraith, P. Pala, B. Askonas, and R.A. Flavell. 1984. Domain interactions of H-2 class I antigens alter cytotoxic T-cell recognition sites. Nature. 309:279. Alting-Mees, M. , and B.H. Barber. 1986. A structural analysis of the carbohydrate side chains on class I and class II histocompatibility antigens of the swine facilitated by heteroantisera specific for the denatured polypeptides. Mol. Immunol. 23:847. Anderson, S.K., S. Gallinger, J . Roder, J . Frey, H.A. Young, and J.R. Ortaldo. 1993. A cyclophilin-related protein involved in the function of natural killer cells. Proc. Natl. Acad. Sci. USA. 90:542. Anegon, I., M.C. Cuturi, G. Trinchieri, and B. Perussia. 1988. Interaction of Fc receptor (CD 16) with ligands induces transcription of IL-2 receptor (CD25) and lymphokine genes and expression of their products in human natural killer cells. J .Exp. Med. 167:452. Aramburu, J . , M.A. Balboa, A. Ramirez, A. Silva, A. Acevedo, A. Sanchez-Madrid, M.O. DeLandazuri, and M. Lopez-Botet. 1990. A novel functional cell surface dimer (Kp43) expressed by natural killer cells and T cell receptor-y/8+ T lymphocytes. I. Inhibition of the IL-2-dependent proliferation by anti-Kp43 monoclonal antibody. J . Immunol. 144:3238. Ault, K.A., J .H. Antin, D. Ginsburg, S.H. Orkin, J .M. Rappeport, M.L. Keohan, P. Martin, and B.R. Smith. 1985. Phenotype of recovering lymphoid cell populations after marrow transplantation. J . Exp. Med. 161:1483. Barlozzari, T., J . Leonhardt, R.H. Wiltrout, R.B. Herberman, and C.W. Reynolds. 1985. Direct evidence for the role of L G L in the inhibition of experimental tumor metastases. J . Immunol. 134:2783. Bellone, G., N.M. Valiante, O. Viale, E . Ciccone, L. Moretta, and G. Trinchieri. 1993. Regulation of hematopoiesis in vitro by alloreactive natural killer cell clones. J .Exp. Med. 177:1117. 108 Bennett, M. 1972. Rejection of marrow grafts. Importance of H-2 homozygosity of donor cells. Transplantation. 14:289. Bennett, M . 1973. Prevention of marrow allograft rejection with radioactive strontium: Evidence for marrow dependent effector cells. J . Immunol. 110:510 Bennett, M. , Y.Y.L. Yu, E . Stoneman, R.M. Rembecki, P.A. Mathew, K.F. Lindahl, and V. Kumar. 1995. Hybrid resistance: 'negative' and 'positive' signaling of murine natural killer cells. Semin. Immunol. 7:121. Bezouska, K , G.V. Crichlow, J .M. Rose, M.E. Taylor, and K. Drickamer. 1991. Evolutionary conservation of intron position in a subfamily of genes encoding carbohydrate-recognition domains. J . Biol. Chem. 266:11604. Bezouska, K., G. Vlahas, O. Horvath, G. Jinochova, A. Fiserova, R. Giorda, W.H. Chambers, T. Feizi, and M. Pospisil. 1994. Rat natural killer cell antigen, NKR-P l , related to C-type animal lectins is a carbohydrate-binding protein. J . Biol. Chem. 269:16945. Bezouska, K., C.T. Yuen, J . O'Brien, R.A. Childs, W. Chai, A . M . Lawson, K. Drbal, A. Fiserova, M . Pospisil, and T. Feizi. 1994. Oligosaccharide ligands for NKR-P1 protein activate N K cells and cytotoxicity. Nature. 372:150. Biron, C.A., P. Van den Elsen, M.M. Tutt, P. Medveczky, V. Kumar, and C. Terhorst. 1987. Murine natural killer cells stimulated in vivo do not express the T cell receptor a, p, y, T35, or T3e genes. J . Immunol. 139:1704. Biron, C.A., K.S. Byron, and J.L. Sullivan. 1989. Severe herpesvirus infections in an adolescent without natural killer cells. N. Engl. J . Med. 320:1731. Bix, M. , N.S. Liao, M. Zijlstra, J . Loring, R. Jaenisch, and D. Raulet. 1991. Rejection of class I MHC-deficient haemopoietic cells by irradiated MHC-matched mice. Nature. 349:329. Bjorkman, P.J., M.A. Saper, B. Samraoui, W.S. Bennett, J .L. Strominger, and D.C. Wiley. 1987. Structure of the human class I histocompatibility antigen, HLA-A2. Nature. 329:506. Brennan, J . , D. Mager, W. Jefferies, and F. Takei. 1994. Expression of different members of the Ly-49 gene family defines distinct natural killer cell subsets and cell adhesion properties. J . Exp. Med. 180:2287. Brennan, J . , F. Takei, S. Wong, and D.L. Mager. 1995. Carbohydrate recognition by a natural killer cell receptor, Ly-49C. J . Biol. Chem. 270:9691. 109 Cassatella, M.A., I. Anegon, M.C. Cuturi, P. Griskey, G. Trinchieri, and B. Perussia. 1989. Fey R (CD16) interaction with ligand induces C a 2 + mobilization and phosphoinositide turnover in human natural killer cells: role of C a 2 + in Fey R (CD16)-induced transcription and expression of lymphokine genes. J . Exp. Med. 169:549. Cella, M. , A. Longo, G.B. Ferrara, J .L. Strominger, and M . Colonna. 1994. NK3-specific natural killer cells are selectively inhibited by Bw4-positive H L A alleles with isoleucine 80. J . Exp. Med. 180:1235. Chambers, C.A., S. Gallinger, S.K. Anderson, S. Giardina, J.R. Ortaldo, N. Hozumi, and J . Roder. 1994. Expression of the NK-TR gene is required for NK-like activity in human T cells. J . Immunol. 152:2669. Chambers, W.H., and T.N. Oeltmann. 1986. The effects of hexose 6-0-sulphate esters on human natural killer cell lytic function. J . Immunol. 137:1469. Chambers, W.H., N.L. Vujanovic, A.B. DeLeo, M.W. Olszowy, R.B. Herberman, and J.C. Hiserodt. 1989. Monoclonal antibody to a triggering structure expressed on rat natural killer cells and adherent lymphokine-activated killer cells. J . Exp. Med. 169:1373. Chambers, W.H., T. Adamkiewicz, and J.P. Houchins. 1993. Type II integral membrane proteins with characteristics of C-type animal lectins expressed by natural killer (NK) cells. Glycobiology. 3:9. Chan, P.Y., and F. Takei. 1986. Expression of a T cell receptor-like molecule on normal and malignant murine T cells detected by rat monoclonal antibodies to nonclonotypic determinants. J.Immunol. 136:1346. Chan, P.Y., and F. Takei. 1988. Characterization of a murine T cell surface disulfide-linked dimer of 45-kDa glycopeptides (YE 1/48 antigen). J . Immunol. 140:161. Chan, P.Y., and F. Takei. 1989. Molecular cloning and characterization of a novel murine T cell surface antigen, YE1/48. J . Immunol. 142:1727. Chang, C , A. Rodriguez, M . Carretero, M. Lopez-Botet, J .H. Phillips, and L .L . Lanier. 1995. Molecular characterization of human CD94: a type II membrane glycoprotein related to the C-type lectin superfamily. Eur. J . Immunol. 25:2433. Ciccone, E . , O. Viale, D. Pende, M. Malnati, R. Biassoni, G. Melioli, A. Moretta, E.O. Long, and L. Moretta. 1988. Specific lysis of allogeneic cells after activation of CD3- lymphocytes in mixed lymphocyte culture. J . Exp. Med. 168:2403. 110 Ciccone, E . , D. Pende, O. Viale, G. Tambussi, S. Ferrini, R. Biassoni, A. Longo, J . Guardiola, A. Moretta, and L. Moretta. 1990a. Specific recognition of human CDS-CD 16+ natural killer cells requires the expression of an autosomic recessive gene on target cells. J . Exp. Med. 172:47. Ciccone, E . , M. Colonna, O. Viale, D. Pende, C. Di Donato, D. Reinharz, A. Amoroso, M . Jeannet, J . Guardiola, A. Moretta, T. Spies, J . Strominger, and L. Moretta. 1990b. Susceptibility or resistance to lysis by alloreactive natural killer cells is governed by a gene in the human major histocompatibility complex between BF and HLA-B. Proc. Natl. Acad. Sci. USA. 87:9794. Ciccone, E . , D. Pende, O. Viale, C. Di Donato, G. Tripodi, A .M. Orengo, J . Guardiola, A. Moretta, and L. Moretta. 1992a. Evidence of a natural killer (NK) cell repertoire for (alio) antigen recognition: definition of five distinct NK-determined allospecificities in humans. J . Exp. Med. 175:709. Ciccone, E . , D. Pende, O. Viale, A. Than, C. Di Donato, A . M . Orengo, R. Biassoni, S. Verdiani, A. Amoroso, A. Moretta, and L. Moretta. 1992b. Involvement of H L A class I alleles in natural killer (NK) cell-specific functions: expression of HLA-Cw3 confers selective protection from lysis by alloreactive N K clones displaying a defined specificity (specificity 2). J . Exp. Med. 176:963. Colonna, M. , T. Spies, J.L. Strominger, E . Ciccone, A. Moretta, L. Moretta, D. Pende, and O. Viale. 1992. Alloantigen recognition by two human natural killer cell clones is associated with HLA-C or a closely linked gene. Proc. Natl. Acad. Sci. USA. 89:7983. Colonna, M. , E .G. Brooks, M. Falco, G.B. Ferrara, and J.L. Strominger. 1993a. Generation of allospecific natural killer cells by stimulation across a polymorphism of HLA-C. Science. 260:1121. Colonna, M. , G. Borsellino, M. Falco, G.B. Ferrara, and J.L. Strominger. 1993b. HLA-C is the inhibitory ligand that determines dominant resistance to lysis by NKI- and NK2-specific natural killer cells. Proc. Natl. Acad. Sci. USA. 90:12000. Colonna, M. , and J . Samaridis. 1995. Cloning of immunoglobulin-superfamily members associated with HLA-C and HLA-B recognition by human natural killer cells. Science. 268:405-408. Correa, I. and D.H. Raulet. 1995. Binding of diverse peptides to M H C class I molecules inhibits target cell lysis by activated natural killer cells. Immunity. 2:61. Cudkowicz, G., and J .H. Stimpfling. 1964. Deficient growth of C57BL marrow cells transplanted in FI hybrid mice. Association with the histocompatibility-2 locus. Immunology. 7:291. i l l D'Andrea, A., C. Chang, K. Franz-Bacon, T. McClanahan, J .H. Philips, and L.L . Lanier. 1995. Molecular cloning of NKB1. A natural killer cell receptor for HLA-B allotypes. J . Immunol. 155:2306-2310. Daniels, B.F., F .M. Karlhofer, W.E. Seaman, and W.M. Yokoyama. 1994a. A natural killer cell receptor specific for a major histocompatibility complex class I molecule. J . Exp. Med. 180:687. Daniels, B.F., M.C. Nakamura, S.D. Rosen, W.M. Yokoyama, and W.E. Seaman. 1994b. Ly-49A, a receptor for H-2D d, has a functional carbohydrate recognition domain. Immunity. 1:785. DeAngelis, P.L., and C.G. Glabe. 1987. Polysaccharide structural features that are critical for the binding of sulfated fucans to bindin, the adhesive protein from sea urchin sperm. J . Biol. Chem. 262:13946. Deveraux, J . , P. Haeberli, and O. Smithies. 1984. A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12:387. Dougherty, G.J., D.L. Cooper, J.F. Memory, and R.K. Chiu. 1994. Ligand binding specificity of alternatively spliced CD44 isoforms: recognition and binding of hyaluronan by CD44R1. J . Biol. Chem. 269:9074. Drickamer, K. 1988. Two distinct classes of carbohydrate-recognition domains in animal lectins. J . Biol. Chem. 263:9557. Drickamer, K., and M.E. Taylor. 1993. Biology of animal lectins. Annu. Rev. Cell Biol. 9:237. Fleisher, G., S. Starr, N. Koven, H. Kamiya, S.D. Douglas, and W. Henle. 1982. A non-X-linked syndrome with susceptibility to severe Epstein-Barr virus infections. J . Pediatr. 100:727. Forbes, J.T., R.K. Bretthauer, and T.N. Oeltmann. 1981. Mannose 6-, fructose 1-, and fructose 6-phosphates inhibit human natural cell-mediated cytotoxicity. Proc. Natl. Acad. Sci. USA. 78:5797. Garni-Wagner, B.A., A. Purohit, P.A. Mathew, M. Bennett, and V. Kumar. 1993. A novel function-associated molecule related to non-MHC-restricted cytotoxicity mediated by activated natural killer cells and T cells. J . Immunol. 151:60. Giorda, R., W.A. Rudert, C. Vavassori, W.H. Chambers, J.C. Hiserodt, and M. Trucco. 1990. NKR-P1, a signal transduction molecule on natural killer cells. Science. 249:1298. 112 Giorda, R., and M . Trucco. 1991. Mouse NKR-P1: a family of genes selectively coexpressed in adherent lymphokine-activated killer cells. J . Immunol. 147:1701. Giorda, R., E.P. Weisberg, T.K. Ip, and M. Trucco. 1992. Genomic structure and strain-specific expression of the natural killer cell receptor NKR-P1. J . Immunol. 149:1957. Glas, R., H . Sturmhofel, G.J. Hammerling, K. Karre, and H.G. Ljunggren. 1992. Restoration of a tumorigenic phenotype by ^-microglobulin transfection to EL-4 mutant cells. J . Exp. Med. 175:843. Gosselin, P., Y. Lusignan, and S. Lemieux. 1993. The murine NK2.1 antigen: a 130 kD glycoprotein dimer expressed by a natural killer cell subset of the spleen, thymus and lymph nodes. Mol. Immunol. 30:1185. Grimm, E.A., A. Mazumder, H.Z. Zhang, and S.A. Rosenberg. 1982. Lymphokine-activated killer cells phenomenon. Lysis of natural killer-resistant fresh solid tumor cells by interleukin 2-activated autologous human peripheral blood lymphocytes. J . Exp. Med. 155:1823. Grundy, J .E . , J.S. Mackenzie, and N.F. Stanley. 1981. Influence of H-2 and non-H-2 genes on resistance to murine cytomegalovirus infection. Infect. Immun. 32:277. Gumperz, J .E . , V. Litwin, J .H. Philips, L .L . Lanier, and P. Parham. 1995. The Bw4 public epitope of HLA-B molecules confers reactivity with natural killer cell clones that express NKB1, a putative H L A receptor. J . Exp. Med. 181:1133. Gunji, Y., N.L. Vujanovic, J.C. Hiserodt, R.B. Herberman, and E . Gorelik. 1989. Generation and characterization of purified adherent lymphokine-activated killer cells in mice. J . Immunol. 142:1748. Hackett, J . Jr., M. Bennett, and V. Kumar. 1985. Origin and differentiation of natural killer cells: I. Characteristics of a tranplantable N K cell precursor. J . Immunol. 134:3731. Hackett, J . Jr., G.C. Bosma, M.J. Bosma, M. Bennett, and V. Kumar. 1986a. Transplantable progenitors of natural killer cells are distinct from those of T and B lymphocytes. Proc. Natl. Acad. Sci. USA. 83:3427. Hackett, J . Jr., M. Tutt, M. Lipscomb, M . Bennett, G. Koo, and V. Kumar. 1986b. Origin and differentiation of natural killer cells. II. Functional and morphologic studies of purified NK-1.1+ cells. J . Immunol. 136:3124. 113 Haller, O., and H. Wigzell. 1977. Suppression of natural killer cell activity with radioactive strontium: Effector cells are marrow dependent. J . Immunol. 118:1503. Haller, O., M . Hansson, R. Kiessling, and H. Wigzell. 1977a. Role of non-conventional natural killer cells in resistance against syngeneic tumor cells in vivo. Nature. 270:609. Haller, O., R. Kiessling, A. Orn, and H. Wigzell. 1977b. Generation of natural killer cells: An autonomous function of the bone marrow. J . Exp. Med. 145:1411. Hammarskjold, M.L. , S.C. Wang, and G. Klein. 1986. High-level expression of the Epstein-Barr virus EBNA1 protein in CV1 cells and human lymphoid cells using a SV40 late replacement vector. Gene. 43:41. Harel-Bellan, A., A. Quillet, C. Marchiol, R. DeMars, T. Tursz, and D. Fradelizi. 1986. Natural killer susceptibility of human cells may be regulated by genes in the H L A region on chromosome 6. Proc. Natl. Acad. Sci. USA. 83:5688. Harrison, D.E., and G.A. Carlson. 1983. Effects of the beige mutation and irradiation on natural resistance to marrow grafts. J . Immunol. 130:484. Held, W., J . Roland, and D.H. Raulet. 1995. Allelic exclusion of Ly-49-family genes encoding class I MHC-specific receptors on N K cells. Nature. 376:355. Herberman, R.B., M.E . Nunn, H.T. Holden, and D.H. Lavrin. 1975. Natural cytotoxic reactivity of mouse lymphoid cells against syngeneic and allogeneic tumors. II. Characterization of effector cells. Int. J . Cancer. 16:230. Hoglund, P., C. Ohlen, E . Carbone, L. Franksson, H.G. Ljunggren, A. Latour, B. Koller, and K. Karre. 1991. Recognition of pVmicroglobulin-negative (p2m) T-cell blasts by natural killer cells from normal but not from p2nr mice: Nonresponsiveness controlled by ftem- bone marrow in chimeric mice. Proc. Natl. Acad. Sci. USA. 88:10332. Houchins, J.P., T. Yabe, C. McSherry, N. Miyokawa, and F .H. Bach. 1990. Isolation and characterization of N K cell or NK/T cell-specific cDNA clones. J . Mol. Cell. Immunol. 4:295. Houchins, J.P., T. Yabe, C. McSherry, and F .H. Bach. 1991. DNA sequence analysis of NKG2, a family of related cDNA clones encoding type II integral membrane proteins on human natural killer cells. J . Exp. Med. 173:1017. Kane, K.P. 1994. Ly-49 mediates EL4 lymphoma adhesion to isolated class I major histocompatibility molecules. J . Exp. Med. 179:1011. 114 Karlhofer, F .M. and W.M. Yokoyama. 1991. Stimulation of murine natural killer (NK) cells by a monoclonal antibody specific for the NK1.1 antigen. IL-2-activated N K cells possess additional specific stimulation pathways. J . Immunol. 146:3662. Karlhofer, F .M. , R.K. Ribaudo, and W.M. Yokoyama. 1992. M H C class I alloantigen specificity of Ly-49+ IL-2 activated natural killer cells. Nature. 358:66. Karlhofer, F .M. , R. Hunziker, A. Reichlin, D.H. Margulies, and W.M. Yokoyama. 1994. Host M H C class I molecules modulate in vivo expression of a N K cell receptor. J . Immunol. 153:2407. Karre, K., H.G. Ljunggren, G. Piontek, and R. Kiessling. 1986. Selective rejection of H-2-deficient lymphoma variants suggests alternative immune defense strategy. Nature. 319:675. Kay, R., and R.K. Humphries. 1991. New vectors and procedures for isolating cDNAs encoding cell surface proteins by expression cloning in COS cells. Methods Mol. Cell. Biol. 2:254. Kiessling, R., G. Petranyi, G. Klein, and H. Wigzell. 1975. Genetic variation of in vitro cytolytic activity and in vivo rejection potential of non-immunized semi-sygeneic mice against a mouse lymphoma line. Int. J . Cancer. 15:933. Kimball, E.S., and J .E . Coligan. 1983. Structure of class I major histocompatibility antigens. Contemp. Top. Mol. Immunol. 9:1. Koo, G., and J.R. Peppard. 1984. Establishment of monoclonal Anti-Nk-1.1 antibody. Hybridoma. 3:301. Kubo, S., Y. Itoh, N. Ishikawa, R. Nagasawa, T. Mitarai, and N. Maruyama. 1993. The gene encoding the mouse lymphocyte antigen Ly-49: structural analysis and the 5'-flanking sequence. Gene. 136:329. Labarriere, N , J.P. Piau, R. Zennadi, P. Blanchardie, M . Denis, and P. Lustenberger. 1993. Retinoic acid modulation of oc(l—»2) fucosyltransferase activity and sensitivity of tumor cells to LAK-mediated cytotoxicity. In Vitro Cell. Dev. Biol. 29A:140. Lanier, L .L . , S. Cwirla, N. Federspiel, and J.J . Phillips. 1986a. Human natural killer cells isolated from peripheral blood do not rearrange T cell antigen receptor (3 chain genes. J . Exp. Med. 163:209. Lanier, L .L . , J .H. Philips, J . Hackett, Jr., M. Tutt, and V. Kumar. 1986b. Natural killer cells: definition of a cell type rather than a function. J . Immunol. 137:2735. 115 Lanier, L . L . , R. Testi, J . Bindl, and J .H. Phillips. 1989a. Identity of Leu-19 (CD56) leukocyte differentiation antigen and neural cell adhesion molecule. J . Exp. Med. 169:2233. Lanier, L . L . , G. Yu, and J.H. Phillips. 1989b. Co-association of CD3£ with a receptor (CD16) for IgG Fc on human natural killer cells. Nature. 342:803. Lanier, L .L . , C. Chang, and J .H. Philips. 1994. Human NKR-P1A: a disulfide-linked homodimer of the C-type lectin superfamily expressed by a subset of N K and T lymphocytes. J . Immunol. 153:2417. Lawlor, D.A., J . Zemmour, P.D. Ennis, and P. Parham. 1990. Evolution of class-I M H C genes and proteins: from natural selection to thymic selection. Annu. Rev. Immunol. 8:23. Lemieux, S., F. Ouellet-Talbot, Y. Lusignan, L. Morelli, N. Labreche, P. Gosselin, and J . Lecomte. 1991. Identification of murine natural killer cell subsets with monoclonal antibodies derived from 129 anti-C57BL/6 immune spleen cells. Cell. Immunol. 134:191. Liao, N.S., M . Bix, M . Zijlstra, R. Jaenisch, and D. Raulet. 1991. M H C class I deficiency: susceptibility to natural killer (NK) cells and impaired N K activity. Science. 253:199. Litwin, V., J . Gumperz, P. Parham, J .H. Philips, and L . L . Lanier. 1993. Specificity of H L A class I antigen recognition by human N K clones: evidence for clonal heterogeneity, protection by self and non-self alleles, and influence of the target cell type. J . Exp. Med. 178:1321 Litwin, V., J . Gumperz, P. Parham, J .H. Philips, and L.L . Lanier. 1994. NKB1: a natural killer cell receptor involved in the recognition of polymorphic HLA-B molecules. J . Exp. Med. 180:537. Ljunggren, H.G., and K. Karre. 1990. In search of the 'missing self: M H C molecules and N K cell recognition. Immunol. Today. 11:237. Lotzova, E . , C A . Savary, and S.B. Pollack. 1983. Prevention of rejection of allogeneic bone marrow transplants by NK1.1 antiserum. Transplantation. 35:490. Madden, D.R., J .C. Gorga, J.L. Strominger, and D.C. Wiley. 1991. The structure of HLA-B27 reveals nonamer self-peptides bound in an extended conformation: Nature. 353:321. Mahon, G.M. 1995. Ligand binding studies on Ly-49. Thesis, University of British Columbia, Vancouver, BC. 116 Malnati, M.S., P. Lusso, E . Ciccone, A. Moretta, L. Moretta, and E.O. Long. 1993. Recognition of virus-infected cells by natural killer cell clones is controlled by polymorphic target cell elements. J . Exp. Med. 178:961. Malnati, M.S., M . Peruzzi, K.C. Parker, W.E. Biddison, E . Ciccone, A. Moretta, and E.O. Long. 1995. Peptide specificity in the recognition of MHC class I by natural killer cell clones. Science. 267:1016 Mason, L. , S.L. Giardina, T. Hecht, J . Ortaldo, and B.J. Mathieson. 1988. LGL-1: a non-polymorphic antigen expressed on a major population of mouse natural killer cells. J . Immunol. 140:4403. Mason, L .H . , H . Yagita, and J.R. Ortaldo. 1994. LGL-1: a potential triggering molecule on murine N K cells. J . Leuk. Biol. 55:362 Mason, L .H . , J.R. Ortaldo, H.A. Young, V. Kumar, M. Bennett, and S.K. Anderson. 1995. Cloning and functional characteristics of murine large granular lymphocyte-1: a member of the Ly-49 gene family (Ly-49G2). J . Exp. Med. 182:293-303. Mathew, P.A., B.A. Garni-Wagner, K. Land, A. Takashima, E . Stoneman, M . Bennett, and V. Kumar. 1993. Cloning and characterization of the 2B4 gene encoding a molecule associated with non-MHC-restricted killing mediated by activated natural killer cells and T cells. J . Immunol. 151:5328. McCoy, J.P. Jr., and W.H. Chambers. 1991. Carbohydrates in the functions of natural killer cells. Glycobiology. 1:321. Miller, J.S., K.A. Alley, and P. McGlave. 1994. Differentiation of natural killer cells from human primitive marrow progenitors in a stroma-based long-term culture system: Identification of a CD34 +CD7 + N K progenitor. Blood. 83:2594. Misra, D.N., H.W. Kunz, and T.J. Gill III. 1987. Carbohydrate moieties of rat MHC class I antigens. Immunogenetics. 26:204. Mombaerts, P., J . Iacomini, R.S. Johnson, K. Herrup, S. Tonegawa, and V . E . Papaioannou. 1992. RAG-1 deficient mice have no mature B and T lymphocytes. Cell. 68:869. Morelli, L. and S. Lemieux. 1993. Triggering of the cytotoxic activity of murine natural killer and lymphokine-activated killer cells through the NK2.1 antigen. J . Immunol. 151:6783. Moretta, A. G. Tambussi, C. Bottino, G. Tripodi, A. Merli, E . Ciccone, G. Pantaleo, and L. Moretta. 1990a. A novel surface antigen expressed by a subset of human natural killer cells. Role in cell activation and regulation of cytolytic function. J . Exp. Med. 171:695. 117 Moretta, A., C. Bottino, D. Pende, G. Tripodi, G. Tambussi, O. Viale, A. Orengo, M. Barbaresi, A. Merli, E . Ciccone, and L. Moretta. 1990b. Identification of four subsets of human CD3CD16+ natural killer (NK) cells by the expression of clonally distributed functional surface molecules: correlation between subset assignment of N K clones and ability to mediate specific alloantigen recognition. J . Exp. Med. 172:1589. Moretta, A., M . Vitale, C. Bottino, A .M. Orengo, L. Morelli, R. Augugliaro, M . Barbaresi, E . Ciccone, and L. Moretta. 1993. P58 molecules as putative receptors for major histocompatibility complex (MHC) class I molecules in human natural killer (NK) cells. Anti-p58 antibodies reconstitute lysis of M H C class I-protected cells in N K clones displaying different specificities. J . Exp. Med. 178:597. Moretta, A., M . Vitale, S. Sivori, C. Bottino, L. Morelli, R. Augugliaro, M . Barbaresi, D. Pende, E . Ciccone, M . Lopez-Botet, and L. Moretta. 1994. Human natural killer cell receptors for HLA-class I molecules. Evidence that the Kp43 (CD94) molecule functions as receptor for HLA-B alleles. J . Exp. Med. 180:545. Moretta, L. , E . Ciccone, A. Moretta, P. Hoglund, C. Ohlen, and K. Karre. 1992. Allorecognition by N K cells: nonself or no self? Immunol. Today. 13:300. Murphy, W.J., V. Kumar, and M . Bennett. 1987. Rejection of bone marrow allografts by mice with severe combined immune deficiency (SCID). Evidence that natural killer cells can mediate the specificity of marrow graft rejection. J . Exp. Med. 165:1212. Murphy, W.J., V. Kumar, and M . Bennett. 1990. Natural killer cells activated with interleukin 2 in vitro can be adoptively transferred and mediate hematopoietic histocompatibility-1 antigen specific bone marrow rejection in vivo. Eur. J . Immunol. 20:1729. Nagasawa, R., J . Gross, O. Kanagawa, K. Townsend, L .L . Lanier, J . Chiller, and J.P. Allison. 1987. Identification of a novel T cell surface disulfide-bonded dimer distinct from the ot/p antigen receptor. J . Immunol. 138:815. Natuk, R.J., and R.M. Welsh. 1987. Accumulation and chemotaxis of natural killer/large granular lymphocytes at sites of virus replication. J . Immunol. 138:877. Ohlen, C , G. Kling, P. Hoglund, M . Hansson, G. Scangos, C. Bieberich, G. Jay, and K. Karre. 1989. Prevention of allogeneic bone marrow graft rejection by H-2 transgene in donor mice. Science. 246:666. 118 Olsson, M.Y., K. Karre, and C L . Sentman. 1995. Altered phenotype and function of natural killer cells expressing the major histocompatibility complex receptor Ly-49 in mice transgenic for its ligand. Proc. Natl. Acad. Sci. USA. 92:1649. Ortaldo, J.R., T.T. Timonen, and R.B. Herberman. 1984. Inhibition of activity of human N K and K cells by simple sugars: discrimination between binding and postbinding events. Clin. Immunol. Immunopathol. 31:439. Parish, C.R., V. McPhun, and H.S. Warren. 1988. Is a natural ligand of the lymphocytes CD2 molecule a sulfated carbohydrate? J . Immunol. 141:3498. Patankar, M.S., S. Oehninger, T. Barnett, R.L. Williams, and G.F. Clark. 1993. A revised structure for fucoidan may explain some of its biological activities. J . Biol. Chem. 268:21770. Perez-Villar, J.J. , I. Melero, A. Rodriguez, M. Carretero, J . Aramburu, S. Sivori, A .M. Orengo, A. Moretta, and M. Lopez-Botet. 1995. Functional ambivalence of the Kp43 (CD94) N K cell-associated surface antigen. J . Immunol. 154:5779. Perussia, B. and G. Trinchieri. 1984. Antibody 3G8, specific for the human neutrophil Fc receptor, reacts with natural killer cells. J . Immunol. 132:1410. Perussia, B., G. Trinchieri, A. Jackson, N.L. Warner, J . Faust, H . Rumpold, D. Kraft, and L.L . Lanier. 1984. The Fc receptor for IgG on human natural killer cells: phenotypic, functional, and comparative studies with monoclonal antibodies. J . Immunol. 133:180. Pistoia, V., S. Zupo, A. Corcione, S. Roncella, L. Matera, R. Ghio, and M . Ferrarini. 1989. Production of colony-stimulating activity by human natural killer cells: Analysis of the conditions that influence the release and detection of colony-stimulating activity. Blood. 74:156. Pohajdak, B., J.A. Wright, and A.H. Greenberg. 1984. An oligosaccharide biosynthetic defect in concanavalin A-resistant Chinese hamster ovary cells that enhances N K reactivity in vitro and in vivo. J . Immunol. 133:2423. Pullen, J.K., R.M. Horton, Z. Cai, and L.R. Pease. 1992. Structural diversity of the classical H-2 genes: K, D, and L. J . Immunol. 148:953. Robertson, M.J. , R.J. Soiffer, S.F. Wolf, T.J. Manley, C. Donahue, D. Young, S.H. Herrmann, and J . Ritz. 1992. Response of human natural killer (NK) cells to N K cell stimulatory factor (NKSF): cytolytic activity and proliferation of N K cells are differentially regulated by NKSF. J . Exp. Med. 175:779. Rochon, Y.P., S.I. Simon, E.B. Lynam, and L A . Sklar. 1994. A role for lectin interactions during human neutrophil aggregation. J.Immunol. 152:1385. 119 Roder, J . and A. Duwe. 1979. The beige mutation in the mouse selectively impairs natural killer cell function. Nature. 278:451. Roland, J . and P.A. Cazenave. 1992. Ly-49 antigen defines an ap TCR population in i-IEL with an extrathymic maturation. Int. Immunol. 4:699. Rooney, C M . , and A.J . Munro. 1984. N K cells can recognize asialylated autologous lymphocytes and ABO-mismatched lymphocytes. Immunology. 51:193. Ryan, J . C , E . C Niemi, R.D. Goldfien, J . C Hiserodt, and W.E. Seaman. 1991. NKR-P1, an activating molecule on rat natural killer cells, stimulates phosphoinositide turnover and a rise in intracellular calcium. J . Immunol. 147:3244. Ryan, J . C , J . Turck, E.C. Niemi, W.M. Yokoyama, and W.E. Seaman. 1992. Molecular cloning of the NK1.1 antigen, a member of the NKR-P1 family of natural killer cell activation molecules. J . Immunol. 149:1631. Saito, N. , A. Ozaki, Y. Beppa, K. Takahishi, J . Fujita, Y. Sasaki, H . Nomori, M . Kimata. 1984. Analysis of the spread and growth of tumor cells in mice with depressed natural killer activity by anti-asialo GM1 antibody or anti-cancer drugs. J . Cancer Res. Clin. Oncol. 107:157. Sanchez, M.J. , M.O. Muench, M.G. Roncarolo, L. Lanier, and J .H. Philips. 1994. Identification of a common T/NK progenitor in human fetal thymus. J . Exp. Med. 180:569. Scalzo, A.A., N.A. Fitzgerald, A. Simmons, A.B. LaVista, and G.R. Shellam. 1990. Cmv-1, a genetic locus that controls murine cytomegalovirus replication in the spleen. J . Exp. Med. 171:1469. Scalzo, A.A., N.A. Fitzgerald, C R . Wallace, A . E . Gibbons, Y . C Smart, R . C Burton, and G.R. Shellam. 1992. The effect of the Cmv-1 resistance gene, which is linked to the natural killer cell gene complex, is mediated by natural killer cells. J . Immunol. 149:581. Seaman, W.E., M.A. Blackman, T. Gindhart, J.R. Roubinian, M. Loeb, and N. Talal. 1978. P-Estradiol reduces natural killer cells in mice. J . Immunol. 121:2193. Seaman, W.E., M . Sleisenger, E . Eriksson, and G.C. Koo. 1987. Depletion of natural killer cells in mice by monoclonal antibody to NK-1.1. Reduction in host defense against malignancy without loss of cellular or humoral immunity. J . Immunol. 138:4539. 120 Sentman, C.L., J . Hackett, Jr., V. Kumar, and M. Bennett. 1989. Identification of a subset of murine natural killer cells that mediates rejection of H h - l d but not Hh-l b bone marrow grafts. J . Exp. Med. 170:191. Sentman, C.L. , V. Kumar, and M . Bennett. 1991. Rejection of bone marrow allografts by natural killer cell subsets: 5E6+ cell specificity for Hh-1 determinant 2 shared by H-2 d and H-2 f. Eur. J . Immunol. 21:2821. Sentman, C.L., M.Y. Olsson, M . Salcedo, P. Hoglund, U. Lendahl, and K. Karre. 1994. H-2 allele-specific protection from N K cell lysis in vitro for lymphoblasts but not tumor targets; protection mediated by ocl/cc2 domains. J . Immunol. 153:5482. Shellam, G.R., J .E . Allan, J .M. Papadimitriou, and G.J. Bancroft. 1981. Increased susceptibility to cytomegalovirus infection in beige mutant mice. Proc. Natl. Acad. Sci. USA. 78:5104. Shi, L. , R.P. Kraut, R. Aebersold, and A.H. Greenberg. 1992. A natural killer cell granule protein that induces DNA fragmentation and apoptosis. J . Exp. Med. 175:553. Shinkai, Y., G. Rathbun, K.P. Lam, E .M. Oltz, V. Stewart, M . Mendelsohn, J . Charron, A .M. Stall, and F.W. Alt. 1992. RAG-2-deficient mice lack mature lymphocytes owing to inability to initiate V(D)J rearrangement. Cell. 68:855. Siegel, J.P., M. Sharon, P.L. Smith, and W.J. Leonard. 1987. The IL-2 receptor chain (p70): Role in mediating signals for LAK, NK, and proliferative activities. Science. 238:75. Smith, H.R.C., F .M. Karlhofer, and W.M. Yokoyama. 1994. Ly-49 multigene family expressed by IL-2-activated N K cells. J . Immunol. 153:1068. Smith, K.A. 1988. The interleukin 2 receptor. Adv. Immunol. 42:165. Stoneman, E . , V. Kumar, M. Bennett, and P.A. Mathew. 1993. Molecular cloning of an N K cell surface molecule expressed on a subset of cells that mediate specific rejection of H h - l d positive marrow cells. J.Immunol. 150:1465a. (Abstr.) Stoneman, E.R., M . Bennett, J . An, K.A. Chesnut, E.K. Wakeland, J.B. Scheerer, M.J. Siciliano, V. Kumar, and P.A. Mathew. 1995. Cloning and characterization of 5E6 (Ly-49C), a receptor molecule expressed on a subset of murine natural killer cells. J .Exp. Med. 182:305-313. Storkus, W.J., D.N. Howell, R.D. Salter, J.R. Dawson, and P. Cresswell. 1987. N K susceptibility varies inversely with target cell class I H L A antigen expression. J . Immunol. 138:1657. 121 Storkus, W.J., J . Alexander, J.A. Payne, J.R. Dawson, and P. Cresswell. 1989a. Reversal of natural killing susceptibility in target cells expressing transfected class I H L A genes. Proc. Natl. Acad. Sci. USA. 86:2361. Storkus, W.J., J . Alexander, J.A. Payne, P. Cresswell, and J.R. Dawson. 1989b. The al/oc2 domains of class I H L A molecules confer resistance to natural killing. J . Immunol. 143:3853. Storkus, W.J., R.D. Salter, J . Alexander, F .E . Ward, R.E. Ruiz, P. Cresswell, and J.R. Rawson. 1991. Class I-induced resistance to natural killing: identification of nonpermissive residues in HLA-A2. Proc. Natl. Acad. Sci. USA. 88:5989. Stroynowski, I. 1990. Molecules related to class-I major histocompatibility complex antigens. Annu. Rev. Immunol. 8:501. Stutman, O., P. Dien, R.E. Wisun, and E.C. Lattime. 1980. Natural cytotoxic cells against solid tumors in mice: blocking cytotoxicity by D-mannose. Proc. Natl. Acad. Sci. USA. 77:2895. Swiedler, S.J., J .H. Freed, A.L. Tarentino, T.H. Plummer Jr., and G.W Hart. 1985. Oligosaccharide microheterogeneity of the murine major histocompatibility antigens. Reproducible site-specific patterns of sialylation and branching in asparagine-linked oligosaccharides. J . Biol. Chem. 260:4046. Sykes, M. , M.W. Harty, F .M. Karlhofer, D.A. Pearson, G. Szot, and W. Yokoyama. 1993. Hematopoietic cells and radioresistant host elements influence natural killer cell differentiation. J . Exp. Med. 178:223. Tabor, S., and C C . Richardson. 1987. DNA sequence analysis with a modified bacteriophage T7 DNA polymerase. Proc. Natl. Acad. Sci. USA. 84:4767. Takei, F. 1983. Two surface antigens expressed on proliferating mouse T lymphocytes defined by rat monoclonal antibodies. J . Immunol. 130:2794. Talmadge, J .E . , K M . Meyers, D.J. Prieur, and J.R. Starkey. 1980. Role of N K cells in tumor growth and metastasis in beige mice. Nature. 284:622. Townsend, A.R.M., J . Rothbard, F .M. Gotch, G. Bahadur, D. Wraith, and A.J . McMichael. 1986. The epitopes of influenza nucleoprotein recognized by cytotoxic T lymphocytes can be defined with short synthetic peptides. Cell. 44:959. Trinchieri, G., M . Matsumoto-Kobayashi, S.C. Clark, J . Seehra, L. London, and B. Perussia. 1984. Response of resting human peripheral blood natural killer cells to interleukin 2. J . Exp. Med. 160:1147. Trinchieri, G. 1989. Biology of natural killer cells. Adv. Immunol. 47:187. 122 Trowsdale, J . 1993. Genomic structure and function in the MHC. Trends Genet. 9:117. Tsudo, M. , C.K. Goldman, K.F. Bongiovanni, W.C. Chan, E.F. Winton, M. Yagita, E.A. Grimm, and T.A. Waldmann. 1987. The p75 peptide is the receptor for interleukin 2 expressed on large granular lymphocytes and is responsible for the interleukin 2 activation of these cells. Proc. Natl. Acad. Sci. USA. 84:5394. Unkeless, J.C. 1979. Characterization of a monoclonal antibody directed against mouse macrophage and lymphocyte Fc receptors. J . Exp. Med. 150:580. Unkeless, J.C. 1989. Function and heterogeneity of human Fc receptors for immunoglobulin G. J . Clin. Invest. 83:355. Varki, A. 1994. Selectin ligands. Proc. Natl. Acad. Sci. USA. 91:7390. Vitale, M. , S. Sivori, D. Pende, L. Moretta, and A. Moretta. 1995. Coexpression of two functionally independent p58 inhibitory receptors in human natural killer cell clones results in the inability to kill all normal allogeneic cells. Proc. Natl. Acad. Sci. USA. 92:3536-3540. Wagtmann, N., R. Biassoni, C. Cantoni, S. Verdiani, M.S. Malnati, M . Vitale, C. Bottino, L. Moretta, A. Moretta, and E . Long. 1995. Molecular clones of the p58 N K cell receptor reveal immunoglobulin-related molecules with diversity in both the extra- and intracellular domains. Immunity. 2:439-449. Welsh, R.M. 1978. Cytotoxic cells induced during lymphocytic choriomeningitis virus infection of mice. I. Characterization of natural killer cell induction. J . Exp. Med. 148:163. Welsh, R.M., J.O. Brubaker, M. Vargas-Cortes, and C L . O'Donnel. 1991. Natural killer (NK) cell response to virus infection in mice with severe combined immunodeficiency. The stimulation of N K cells and the N K cell-dependent control of virus infections occur independently of T and B cell functions. J . Exp. Med. 173:1053. Welsh, R.M., and M . Vargas-Cortes. 1992. Natural killer cells in viral infection. In The natural killer cell. The natural immune system. C E . Lewis and J.O. McGee, editors. IRL Press, Oxford. 107-150. Wong, S., J.D. Freeman, C. Kelleher, D. Mager, and F. Takei. 1991. Ly-49 multigene family: New members of a superfamily of type II membrane proteins with lectin-like domains. J.Immunol. 147:1417. 123 Yabe, T., C. McSherry, F .H. Bach, P. Fisch, R.P. Schall, P.M. Sondel, and J.P. Houchins. 1993. A multigene family on human chromosome 12 encodes natural killer-cell lectins. Immunogenetics. 37:455. Yednock, T.A., E.C. Butcher, L . M . Stoolman, and S.D. Rosen. 1987. Receptors involved in lymphocyte homing: relationship between a carbohydrate-binding receptor and the MEL-14 antigen. J . Cell Biol. 104:725. Yednock, T.A., and S.D. Rosen. 1989. Lymphocyte Homing. Adv. Immunol. 44:313. Yogeeswaran, G., A. Gronberg, M . Hansson, T. Dalianis, R. Kiessling, and R.M. Welsh. 1981. Correlation of glycosphingolipids and sialic acid in YAC-1 lymphoma variants with their sensitivity to natural killer-cell-mediated lysis. Int. J . Cancer. 28:517. Yokoyama, W., L.B. Jacobs, O. Kanagawa, E . M . Shevach, and D.I. Cohen. 1989. A murine T lymphocyte antigen belongs to a supergene family of type II integral membrane proteins. J . Immunol. 143:1379. Yokoyama, W.M., P.J. Kehn, D.L Cohen, and E . M . Shevach. 1990. Chromosomal location of the Ly-49 (Al, Y E 1/48) multigene family: Genetic association with the N K 1.1 antigen. J . Immunol. 145:2353. Yokoyama, W.M., J.C. Ryan, J.J. Hunter, H.R.C. Smith, M . Stark, and W.E. Seaman. 1991. cDNA cloning of mouse NKR-P1 and genetic linkage with Ly-49: identification of a natural killer cell gene complex on mouse chromosome 6. J . Immunol. 147:3229. Yokoyama, W.M. and W.E. Seaman. 1993. The Ly-49 and NKR-P1 gene families encoding lectin-like receptors on natural killer cells: the N K gene complex. Annu. Rev. Immunol. 11:613. Yokoyama, W.M. 1995. Natural killer cell receptors specific for major histocompatibility complex class I molecules. Proc. Natl. Acad. Sci. USA. 92:3081-3085. Young, J.D., and Z.A. Cohn. 1987. Cellular and humoral mechanisms of cytotoxicity: structural and functional analogies. Adv. Immunol. 41:269. Young, W.W. Jr., S.I. Hakomori, J .M. Durdik, and C.S. Henney. 1980. Identification of ganglio-N-tetraosylceramide as a new cell surface marker for murine natural killer (NK) cells. J . Immunol. 124:199. Yu, Y.Y.L. , V. Kumar, and M . Bennett. 1992. Murine natural killer cells and marrow graft rejection. Annu. Rev. Immunol. 10:189. 124 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items