CD1-specific T cells in microbial immunity.

Recent results strengthen evidence that CD1-restricted T cells play important roles in host defense against microbial infections. Human subjects recently infected with Mycobacterium tuberculosis showed elevated responses to CD1c-mediated presentation of a microbial lipid antigen, compared with control donors. Activation of CD1d-restricted NKT cells with a synthetic glycolipid antigen results in improved immune responses to several infectious pathogens.

Murine CD1d-restricted T cell recognition of cellular lipids.

NKT cells are associated with immunological control of autoimmune disease and cancer and can recognize cell surface mCD1d without addition of exogenous antigens. Cellular antigens presented by mCD1d have not been identified, although NKT cells can recognize a synthetic glycolipid, alpha-GalCer. Here we show that after addition of a lipid extract from a tumor cell line, plate-bound mCD1d molecules stimulated an NKT cell hybridoma. This hybridoma also responded strongly to three purified phospholipids, but failed to recognize alpha-GalCer. Seven of sixteen other mCD1d restricted hybridomas also showed a response to certain purified phospholipids. These findings suggest NKT cells can recognize cellular antigens distinct from alpha-GalCer and identify phospholipids as potential self-antigens presented by mCD1d.

Direct binding and functional transfer of NK cell inhibitory receptors reveal novel patterns of HLA-C allotype recognition.

Cytotoxicity of human NK cells is under negative control of killer cell Ig-like receptors (KIR) specific for HLA class I. To determine the specificity of five KIR containing two Ig domains (KIR2D), direct binding of soluble recombinant KIR2D to a panel of HLA class I transfectants was assayed. One soluble KIR2D, derived from an inhibitory receptor with a long cytoplasmic tail (KIR2DL1), bound to HLA-C allotypes containing asparagine 77 and lysine 80 in the heavy chain, as expected, since these allotypes inhibit lysis by NK cells expressing KIR2DL1. Surprisingly, another KIR2D (KIR2DL2), which inhibits NK lysis of cells expressing HLA-C molecules with serine 77 and asparagine 80, bound to HLA-C allotypes carrying either amino acid motif. Expression of the KIR2DL receptors in NK cells using recombinant vaccinia viruses confirmed these patterns of recognition, and identified KIR2DL3 as another KIR reacting with both groups of HLA-C allotypes. Mutagenesis of amino acid 44 in KIR2DL1 and KIR2DL2 suggested this residue controls the affinity of KIR for the 77/80 motif of HLA-C molecules. Two other soluble KIR2D, derived from noninhibitory receptors with short cytoplasmic tails (KIR2DS), did not bind to any of the HLA class I allotypes tested. One of these receptors (KIR2DS2) is closely related in sequence to KIR2DL2. Substitution of tyrosine 45 with the phenylalanine conserved in other KIR was sufficient to permit specific binding of KIR2DS2 to HLA-C. These results show that KIR2DL receptors are specific for HLA-C, but that recognition of HLA-C allotypes appears more permissive than indicated by previous functional experiments.

Conserved and variable residues within the Bw4 motif of HLA-B make separable contributions to recognition by the NKB1 killer cell-inhibitory receptor.

Allotypes from four divergent HLA-B families (B8, B15, B16, and B27) were compared for their inhibition of cytolysis by NK cells expressing the NKB1 receptor. Allotypes differing solely at the Bw4/Bw6 region were examined as were a more divergent subset of B15 allotypes. The capacity to interact with NKB1 correlated precisely with possession of a Bw4 sequence motif at residues 77-83, whereas no correlation was made with the peptide-binding specificities of two Bw4 and four Bw6 allotypes of the B15 family. HLA-B allotypes having four different Bw4 motifs were examined and all interact with NKB1. In contrast, HLA-A allotypes, which have a Bw4 motif identical with one of those present in HLA-B, do not. Mutation at leucine 82 and arginine 83, the residues common to Bw4 motifs, shows they contribute to NKB1 interaction but are not essential. Three types of polymorphism are implicated in formation of the ligand recognized by NKB1: ones shared by Bw4 motifs; ones distinguishing Bw4 motifs; and ones outside the Bw4/Bw6 region that distinguish HLA-B from HLA-A.

Natural inactivation of a common HLA allele (A*2402) has occurred on at least three separate occasions.

HLA-A*2402 is common and widely distributed in human populations. Several individuals were identified who type genotypically for A*2402, but are serologically null for the HLA-A24 Ag. Sequencing and transfection of genomic DNA fragments containing null and wild-type A*2402 alleles, and the related A*2301 allele, revealed three different null alleles (A*2409N, A*2411N, and A*2402(low)), each of which differs from A*2402 by a single nucleotide change within the 6.7-kb sequence. The A*2301 and A*2402 sequences differ by no substitutions additional to those previously determined for the 1.1-kb cDNA. In exon 4, A*2409N has an in-frame stop codon, while A*2411N has a nucleotide insertion that alters the reading frame, causing premature termination. A*2402(low) has a nucleotide substitution near the splice acceptor site for intron 2 that impairs the production of correctly spliced mRNA. For A*2409N and A*2411N, mRNA is undetectable by Northern analysis, whereas A*2402(low) produces a low level of mRNA and a concomitant amount of normal A*2402 protein at the cell surface. The protein expressed from the A*2402(low) allele is sufficient to stimulate an alloreactive T cell response. On a background of unexpected sequence homogeneity, the single nucleotide changes in the A*2409N, A*2411, and A*2402(low) alleles have dramatic effects upon gene expression and are of likely importance for HLA matching in clinical transplantation. Segregation of at least three independently inactivated A*2402 alleles in human populations raises the possibility that loss of A*2402 may be the result of natural selection.

Polymorphism in the alpha 1 helix of the HLA-B heavy chain can have an overriding influence on peptide-binding specificity.

Previously, we reported overlap in the repertoires of peptides endogenously bound by a group of HLA-B allotypes related to HLA-B7. Extending such analysis to four members of the B17 family and seven members of the B15 family shows that allotypes that share sequence identity in the alpha 1 helix of the class I heavy chain possess markedly similar peptide-binding specificities. Members of the B17 family share a preference for peptides with serine, threonine, or alanine at position 2 and aromatic residues at the carboxyl terminus. Strikingly, the presence of a segment of the B17 alpha 1 helix in B*1516 and B*1517 confers the B17-like peptide-binding motif. The strong influence of natural variation in the alpha 1 helix is exemplified by the differences in peptide-binding specificity of B15 allotypes related by conversion events that replaced segments of the alpha 1 helix. In contrast, evolutionary changes that are confined to the alpha 2 domain confer less dramatic change. They do not perturb the primary anchors of the peptide-binding motif but can modulate the specificity through development and diversification of secondary anchors. Our results, in combination with those obtained previously for other HLA-B allotypes, suggest a general trend whereby polymorphism in the alpha 1 helix is the overriding influence on peptide-binding specificity of HLA-B allotypes, while amino acid substitutions in the alpha 2 domain play a more modulatory role.

Specificity of two anti-class I HLA monoclonal antibodies that block class I recognition by the NKB1 killer cell inhibitory receptor.

Cytolysis by NK cells that possess the NKB1 killer cell inhibitory receptor is inhibited by target cell expression of Bw4+ HLA-B molecules. The inhibitory effect can be prevented by addition of mAbs which block recognition of class I molecules by NKB1. The epitopes recognized by two anti-class I mAbs, DX15 and DX16, which inhibit the interaction of NKB1 with class I have been characterized. Binding of DX15 and DX16 to class I allotypes was investigated by flow cytometric analysis of transfected cell lines which express just one HLA-A, B, or C allele, and by immunoprecipitation of class I molecules from HLA typed B-lymphoblastoid cell lines, followed by isoelectric focusing. The DX16 mAb recognizes class I allotypes which possess alanine at position 71 of the alpha 1 helix, and therefore has a specificity resembling that of the ME1 mAb but with broader specificity. Class I recognition by DX15 is affected by polymorphisms of the C-terminal part of the alpha 1 helix, and the N-terminal part of the alpha 2 helix. DX15 thus appears to recognize a complex epitope near the end of the peptide binding groove which may be conformationally determined. Both antibodies are as effective as the anti-NKB1 mAb (DX9) in preventing class I recognition by the NKB1 receptor. DX16 also blocked recognition by a B*0702 allospecific CTL clone, whereas DX15 did not.

Peptides bound endogenously by HLA-Cw*0304 expressed in LCL 721.221 cells include a peptide derived from HLA-E.

The peptide-binding specificity of HLA-Cw*0304 was determined. Sequence analysis of endogenously-bound peptides isolated from Cw*0304 expressed by LCL 721.221 (221 for short) cells transfected with Cw*0304 cDNA revealed this class I allotype preferentially binds peptides possessing alanine at position 2 and leucine or methionine at the C-terminus. One peptide isolated from Cw*0304 expressed by 221 cells has sequence identity to residues 116-126 of HLA-E. Expression of HLA-E by 221 cells was confirmed by isolation of mRNA transcripts for HLA-E*0101 and detection of beta 2-microglobulin (beta 2-m)-associated HLA-E protein.

The inter-locus recombinant HLA-B*4601 has high selectivity in peptide binding and functions characteristic of HLA-C.

The vast majority of new human HLA class I alleles are formed by conversions between existing alleles of the same locus. A notable exception to this rule is HLA-B*4601 formed by replacement of residues 66-76 of the alpha 1 helix of B*1501 by the homologous segment of Cw*0102. This inter-locus recombination, which brings together characteristic elements of HLA-B and HLA-C structure, is shown here to influence function dramatically. Naturally processed peptides bound by B*4601 are distinct from those of its parental allotypes B*1501 and Cw*0102 and dominated by three high abundance peptides. Such increased peptide selectivity by B*4601 is unique among HLA-A,B,C allotypes. For other aspects of function, presence of the small segment of HLA-C-derived sequence in an otherwise HLA-B framework converts B*4601 to an HLA-C-like molecule. Alloreactive cytotoxic T lymphocytes (CTL), natural killer (NK) cells, and cellular glycosidases all recognize B*4601 as though it were an HLA-C allotype. These unusual properties are those of an allotype which has frequencies as high as 20% in south east Asian populations and is associated with predisposition to autoimmune diseases and nasopharyngeal carcinoma.

CD94 and a novel associated protein (94AP) form a NK cell receptor involved in the recognition of HLA-A, HLA-B, and HLA-C allotypes.

Whereas the human killer cell inhibitory receptors (KIRs) for HLA class I are immunoglobulin-like monomeric type I glycoproteins, the murine Ly49 receptors for H-2 are type II homodimers of the C-type lectin superfamily. Here, we demonstrate that human NK cells also express C-type lectin receptors that influence recognition of polymorphic HLA-A, HLA-B, and HLA-C molecules. These receptors are heterodimers composed of CD94 chains covalently associated with novel tyrosine-phosphorylated glycoproteins (94AP). Some NK clones recognize a common HLA-C ligand using both KIRs and CD94-94AP receptors. These findings suggest the existence of human inhibitory MHC class I receptors of the immunoglobulin and C-type lectin superfamilies and indicate overlap in ligand specificity.

Unusual uniformity of the N-linked oligosaccharides of HLA-A, -B, and -C glycoproteins.

MHC class I glycoproteins possess an invariant site for N-linked oligosaccharide addition at position 86 of the heavy chain. For human HLA-A, -B, and -C class I glycoproteins, position 86 is the only site of N-linked glycosylation. Comparison of the size and relative abundance of oligosaccharides associated with nine HLA-A, -B, or -C allotypes isolated from EBV-transformed B cell lines and mixtures of HLA-A, -B, and -C allotypes isolated from pooled PBLs revealed a very restricted set of structures. Allotypes encoded by the HLA-A and -B loci have two predominant glycan structures that were almost exclusively di-sialylated. In contrast, HLA-C allotypes have four glycan structures, comprising those associated with HLA-A and -B and two additional glycans. Identical oligosaccharides were present on different allotypes of a class I HLA locus, and in particular, HLA-C allotypes defining two inhibitory specificities for NK cells were shown to possess the same set of oligosaccharides. The uniformity of oligosaccharide structure associated with different HLA-A, -B, and -C products and the relative lack of heterogeneity for any given allotype are unusual features for a mammalian glycoprotein. Particularly striking is that such conserved oligosaccharide structures juxtapose the major regions of amino acid sequence variation within the Ag recognition site, including the polymorphisms of the alpha 1 helix that determine the inhibitory ligands for human NK cells.

Heterogeneous phenotypes of expression of the NKB1 natural killer cell class I receptor among individuals of different human histocompatibility leukocyte antigens types appear genetically regulated, but not linked to major histocompatibililty complex haplotype.

Natural killer (NK) cells that express the NKB1 receptor are inhibited from killing target cells that possess human histocompatibility leukocyte antigen (HLA) B molecules bearing the Bw4 serological epitope. To investigate whether NKB1 expression is affected by HLA type, peripheral blood lymphocytes of 203 HLA-typed donors were examined. Most donors had a single population of NKB1+ cells, but some had two populations expressing different cell surface levels of NKB1, and others had no detectable NKB1+ cells. Among the donors expressing NKB1, both the relative abundance of NKB1+ NK cells and their level of cell surface expression varied substantially. The percentage of NKB1+ NK cells ranged from 0 to >75% (mean 14.7%), and the mean fluorescence of the positive population varied over three orders of magnitude. For each donor, the small percentage of T cells expressing NKB1 (usually <2%), had a pattern of expression mirroring that of the NK cells. NKB1 expression by NK and T cells remained stable over the 2-yr period that five donors were tested. Patterns of NKB1 expression were not associated with Bw4 or Bw6 serotype of the donor or with the presence of any individual HLA-A or -B antigens. Cells expressing NKB1 are often found in donors who do not possess an appropriate class I ligand, and can be absent in those who express Bw4+ HLA-B antigens. Family studies further suggested that the phenotype of NKB1 expression is inherited but not HLA linked. Whereas identical twins show matching patterns of NKB1 expression, HLA-identical siblings can differ in NKB1 expression, and conversely, HLA-disparate siblings can be similar. Thus NKB1 expression phenotypes are tightly regulated and extremely heterogeneous, but not correlated with HLA type.

The enigma of the natural killer cell.

Natural killer (NK) cells are controlled by receptors specific for polymorphic determinants of class I molecules of the major histocompatibility complex (MHC). The contrasting properties of NK and cytotoxic T cell (CTL) class I receptors provide complementarity in the cytolytic lymphocyte response to viruses, tumours and transplants. Whereas human NK cell class I receptors consist of immunoglobulin domains, their mouse counterparts resemble C-type lectins. This difference may reflect the receptors' diverse and rapidly evolving class I ligands.

Expression of an unusual Bw4 epitope by a subtype of HLA-B8 [B*0802].

The primary structure of a variant HLA-B8 antigen has been determined by cDNA cloning and sequencing. The variant, B*0802 differs, from the common B*0801 subtype at positions 77-83 of the alpha 1 helix that determine the Bw4 and Bw6 public epitopes. Whereas B*0801 has the common Bw6 motif, B*0802 has the Bw4 motif found in B*13 and B*44 allotypes. Serological analysis of B cell lines expressing B*0802 and of a B*0802 transfectant made with the HLA-A,B negative cell line 721.221 shows that B*0802 reacts with Bw4-specific antibodies, but at a level much lower than expected for Bw4 positive HLA-B allotypes.

Low HLA-C expression at cell surfaces correlates with increased turnover of heavy chain mRNA.

In comparison with HLA-A and -B, the protein products of the HLA-C locus are poorly characterized, in part because of their low level of expression at the cell surface. Here, we examine how protein-protein interactions during assembly and regulation of the mRNA level affect cell surface expression of HLA-C. We find that intrinsic properties of the HLA-C heavy chain proteins do not correlate with low cell surface expression: HLA-C heavy chains associate and dissociate with beta 2-microglobulin (beta 2m) at rates comparable to those found for HLA-A and -B, and increased competition for beta 2m does not alter the surface expression of HLA-C. From studies of chimeric genes spliced from the HLA-B7 and -Cw3 genes, we find that chimeric proteins containing the B7 peptide-binding groove can have low cell surface expression, suggesting that inefficiency in binding peptides is not the cause of low cell surface expression for HLA-C. The surface levels of HLA-A, -B, or -C in cells transfected with cDNA can be similar, implicating noncoding regions of HLA-C heavy chain genes in the regulation of surface expression. We find that HLA-C mRNA is expressed at lower levels than HLA-B mRNA and that this difference results from faster degradation of the HLA-C message. Experiments examining chimeric B7/Cw3 and B7/Cw6 genes suggest that a region determining low expression of HLA-C is to be found between the 3' end of exon 3 and a site in the 3' untranslated region, approximately 600 bases downstream of the translation stop codon.

Superantigen-dependent, cell-mediated cytotoxicity inhibited by MHC class I receptors on T lymphocytes.

Bacterial superantigens bind with high affinity to major histocompatibility complex (MHC) class II antigens on antigen-presenting cells and with T cell antigen receptor (TCR) beta chains on T lymphocytes, which results in the T cell activation responsible for toxic shock syndrome and food poisoning. Many cytotoxic T lymphocyte (CTL) clones were shown to have receptors for human leukocyte antigen (HLA) class I molecules that inhibited superantigen-induced cytotoxicity against appropriate class I-bearing target cells. One type of inhibitory receptor, NKB1, was present on CD4+ and CD8+TCR alpha beta+ CTL clones and blocked the killing of staphylococcal enterotoxin B (SEB)-coated targets bearing certain polymorphic HLA-B molecules. Expression of HLA-A, -B, and -C molecules on the SEB-coated targets also protected against cytolysis mediated by many NKB1-negative T cell clones, suggesting the presence of additional inhibitory MHC class I receptors. These HLA class I receptors may limit tissue destruction and possibly autoimmunity caused by activated T lymphocytes.

The NKB1 and HP-3E4 NK cells receptors are structurally distinct glycoproteins and independently recognize polymorphic HLA-B and HLA-C molecules.

NK cells lyse hematopoietic cells that lack expression of MHC class I molecules on the cell surface. Transfection of certain MHC class I negative cell lines with MHC class I genes renders these cells resistant to NK cell-mediated cytotoxicity. Recently, we described an NK cell receptor, NKB1, that inhibits NK cells from killing target cells expressing Bw4-reactive HLA-B molecules (-B*2705, -B*5101, -B*5801). In this study, we have demonstrated that another structurally distinct NK cell membrane glycoprotein, HP-3E4, is involved in the recognition of certain polymorphic HLA-C molecules (-Cw*0401 and -Cw*1503). NK cell clones co-expressing both the NKB1 and HP-3E4 receptors fail to lyse targets expressing HLA-Cw*0401 and -B*5801, but are able to kill the transfectants in the presence of mAbs against both receptors. These studies demonstrate that a single NK cell clone may possess multiple structurally distinct receptors for different polymorphic HLA class I molecules that function independently.

The Bw4 public epitope of HLA-B molecules confers reactivity with natural killer cell clones that express NKB1, a putative HLA receptor.

Although inhibition of natural killer (NK) cell-mediated lysis by the class I HLA molecules of target cells is an established phenomenon, knowledge of the features of class I molecules which induce this effect remains rudimentary. Using class I alleles HLA-B*1502 and B*1513 which differ only at residues 77-83 which define the Bw4 and Bw6 serological epitopes, we tested the hypothesis that the presence of the Bw4 epitope on class I molecules determines recognition by NKB1+ NK cells. HLA-B*1513 possesses the Bw4 epitope, whereas B*1502 has the Bw6 epitope. Lysis by NKB1+ NK cell clones of transfected target cells expressing B*1513 as the only HLA-A, -B, or -C molecule was inhibited, whereas killing of transfectants expressing B*1502 was not. Addition of an an anti-NKB1 monoclonal antibody reconstituted lysis of the targets expressing B*1513, but did not affect killing of targets bearing B*1502. The inhibitory effect of B*1513 could be similarly prevented by the addition of an anti-class I monoclonal antibody. These results show that the presence of the Bw4 epitope influences recognition of HLA-B molecules by NK cells that express NKB1, and suggest that the NKB1 molecule may act as a receptor for Bw4+ HLA-B alleles. Sequences outside of the Bw4 region must also affect recognition by NKB1+ NK cells, because lysis of transfectants expressing HLA-A*2403 or A*2501, which possess the Bw4 epitope but are in other ways substantially different from HLA-B molecules, was not increased by addition of the anti-NKB1 antibody. Asparagine 86, the single site of N-linked glycosylation on class I molecules, is in close proximity to the Bw4/Bw6 region. The glycosylation site of the Bw4-positive molecule B*5801 was mutated, and the mutant molecules tested for inhibition of NKB1+ NK cells. Inhibition that could be reversed by addition of the anti-NKB1 monoclonal antibody was observed, showing the presence of the carbohydrate moiety is not essential for class I recognition by NKB1+ NK cell clones.

NKB1: a natural killer cell receptor involved in the recognition of polymorphic HLA-B molecules.

Natural killer (NK) cells kill normal and transformed hematopoietic cells that lack expression of major histocompatibility complex (MHC) class I antigens. Lysis of HLA-negative Epstein Barr virus-transformed B lymphoblastoid cell lines (B-LCL) by human NK cell clones can be inhibited by transfection of the target cells with certain HLA-A, -B, or -C alleles. NK cell clones established from an individual demonstrate clonal heterogeneity in HLA recognition and a single NK clone can recognize multiple alleles. We describe a potential human NK cell receptor (NKB1) for certain HLA-B alleles (e.g., HLA-B*5101 and-B*5801) identified by the mAb DX9. NKB1 is a 70-kD glycoprotein that is expressed on a subset of NK cells and NK cell clones. DX9 monoclonal antibody (mAb) specifically inhibits the interaction between NK cell clones and B-LCL targets transfected with certain HLA-B alleles, but does not affect recognition of HLA-A or HLA-C antigens. An individual NK cell clone can independently recognize B-LCL targets transfected with HLA-B or HLA-C antigens; however, DX9 mAb only affects interaction with transfectants expressing certain HLA-B alleles. These findings demonstrate the existence of NK cell receptors involved in the recognition of HLA-B and imply the presence of multiple receptors for MHC on an individual NK clone.