Search

Type 1 diabetes (T1D) is an autoimmune disease characterized by T cell-mediated destruction of insulin-producing pancreatic beta cells. In both humans and the non-obese diabetic (NOD) mouse model of T1D, class II MHC alleles are the primary determinant of disease susceptibility. However, class I MHC genes also influence risk. These findings are consistent with the requirement for both CD4(+) and CD8(+) T cells in the pathogenesis of T1D. Although a large body of work has permitted the identification of multiple mechanisms to explain the diabetes-protective effect of particular class II MHC alleles, studies examining the protective influence of class I alleles are lacking. Here, we explored this question by performing biochemical and structural analyses of the murine class I MHC molecule H-2K(wm7), which exerts a diabetes-protective effect in NOD mice. We have found that H-2K(wm7) molecules are predominantly occupied by the single self-peptide VNDIFERI, derived from the ubiquitous protein histone H2B. This unexpected finding suggests that the inability of H-2K(wm7) to support T1D development could be due, at least in part, to the failure of peptides from critical beta-cell antigens to adequately compete for binding and be presented to T cells. Predominant presentation of a single peptide would also be expected to influence T-cell selection, potentially leading to a reduced ability to select a diabetogenic CD8(+) T-cell repertoire. The report that one of the predominant peptides bound by T1D-protective HLA-A*31 is histone derived suggests the potential translation of our findings to human diabetes-protective class I MHC molecules.

The PD-1 costimulatory receptor inhibits T cell receptor signaling upon interacting with its ligands PD-L1 and PD-L2. The PD-1/PD-L pathway is critical in maintaining self-tolerance. In this study, we examined the role of PD-1 in a mouse model of acute infection withHistoplasma capsulatum, a major human pathogenic fungus. In a lethal model of histoplasmosis, all PD-1-deficient mice survived infection, whereas the wild-type mice died with disseminated disease. PD-L expression on macrophages and splenocytes was up-regulated during infection, and macrophages from infected mice inhibitedin vitroT cell activation. Of interest, antibody blocking of PD-1 significantly increased survival of lethally infected wild-type mice. Thus, our studies extend the role of the PD-1/PD-L pathway in regulating antimicrobial immunity to fungal pathogens. The results show that the PD-1/PD-L pathway has a key role in the regulation of antifungal immunity, and suggest that manipulation of this pathway represents a strategy of immunotherapy for histoplasmosis.

Beta-2 microglobulin (beta2m) is a member of the immunoglobulin-like domain superfamily that is an essential structural subunit of the MHC class I (MHC-I) molecule. beta2m was previously identified as a susceptibility factor for the development of type 1 diabetes (T1D) in NOD mice, whereby transgenic expression of the beta2ma variant, but not the beta2mb variant, restored diabetes susceptibility to normally resistant NOD.beta2mnull mice. Here we report the crystal structures and thermodynamic stabilities of the NOD MHC-I molecule H2-Db containing these two variants. Our results reveal subtle differences in the structures of the beta2m variants, namely in minor loop shifts and in variations in the hydrogen bonding networks at the interfaces between the components of the ternary complex. We also demonstrate that the thermodynamic stabilities of the beta2m variants in isolation differ. However, the conformation of the peptide in the MHC cleft is unchanged in beta2m allelic Db complexes, as are the TCR recognition surfaces. Thus, despite modest structural differences between allelic complexes, the evidence indicates that Db peptide presentation of the representative peptide is unchanged in the context of either beta2m allelic variant. These data suggest that other mechanisms, such as differential association of MHC-I in multiprotein complexes, are likely responsible for the effect of beta2m on T1D development.

CD8 + T cells recognize peptides of eight to nine amino acid residues long in the context of MHC class I molecules on the surface of antigen-presenting cells (APCs). This recognition event is highly sensitive, as evidenced by the fact that T cells can be activated by cognate peptide/MHC complex (pMHC) at extremely low densities (1-50 molecules). High sensitivity is particularly valuable for detection of antigens at low density, such as those derived from tumor cells and intracellular pathogens, which can down-modulate cognate pMHCs from the surface of APCs to evade recognition by the adaptive immune system. T cell activation is only triggered in response to interactions between the T cell receptor (TCR) and the pMHC ligand that reach a specific half-life threshold. However, interactions with excessively long half-lives result in impaired T cell activation. Thus, efficient T cell activation by pMHC on the surface of APCs requires an optimal dwell time of TCR-pMHC interaction. Here, we show that, although this is a requirement at low cognate pMHC density on the APC surface, at high epitope density there is no impairment of T cell activation by extended TCR-pMHC dwell times. This observation was predicted by mathematical simulations for T cell activation by pMHC at different densities and supported by experiments performed on APCs selected for varied expression of cognate pMHC. According to these results, effective T cell activation depends on a complex interplay between inherent TCR-pMHC binding kinetics and the epitope density on the APC.

To assess the roles of serial engagement and kinetic proofreading in T cell receptor (TCR) internalization, we have developed a mathematical model of this process. Our determination of TCR down-regulation for an array of TCR mutants, interpreted in the context of the model, has provided new information about peptide-induced TCR internalization. The amount of TCR down-regulation increases to a maximum value and then declines as a function of the half-life of the bond between the TCR and peptide-major histocompatibility complex (pMHC). The model shows that this behavior, which reflects competition between serial engagement and kinetic proofreading, arises only if it is postulated that activated TCRs remain marked for internalization after dissociation from pMHC. The model also predicts that because of kinetic proofreading, the range of TCR-pMHC-binding half-lives required for T cell activation depends on the concentrations and localization of intracellular signaling molecules. We show here that kinetic proofreading provides an explanation for the different requirements for activation observed in naïve and memory T cells.

The mouse H13 minor histocompatibility (H) Ag, originally detected as a barrier to allograft transplants, is remarkable in that rejection is a consequence of an extremely subtle interchange, P4Val/Ile, in a nonamer H2-Db-bound peptide. Moreover, H13 peptides lack the canonical P5Asn central anchor residue normally considered important for forming a peptide/MHC complex. To understand how these noncanonical peptide pMHC complexes form physiologically active TCR ligands, crystal structures of allelic H13 pDb complexes and a P5Asn anchored pDb analog were solved to high resolution. The structures show that the basis of TCRs to distinguish self from nonself H13 peptides is their ability to distinguish a single solvent-exposed methyl group. In addition, the structures demonstrate that there is no need for H13 peptides to derive any stabilization from interactions within the central C pocket to generate fully functional pMHC complexes. These results provide a structural explanation for a classical non-MHC-encoded H Ag, and they call into question the requirement for contact between anchor residues and the major MHC binding pockets in vaccine design.

The interaction between TCRs and peptides presented by MHC molecules determines the specificity of the T cell-mediated immune response. To elucidate the biologically important structural features of this interaction, we generated TCR β-chain transgenic mice using a TCR derived from a T cell clone specific for the immunodominant peptide of vesicular stomatitis virus (RGYVYQGL, VSV8) presented by H-2Kb. We immunized these mice with VSV8 or analogs substituted at TCR contact residues (positions 1, 4, and 6) and analyzed the CDR3α sequences of the elicited T cells. In VSV8-specific CTLs, we observed a highly conserved residue at position 93 of CDR3α and preferred Jα usage, indicating that multiple residues of CDR3α are critical for recognition of the peptide. Certain substitutions at peptide position 4 induced changes at position 93 and in Jα usage, suggesting a potential interaction between CDR3α and position 4. Cross-reactivity data revealed the foremost importance of the Jα region in determining Ag specificity. Surprisingly, substitution at position 6 of VSV8 to a negatively charged residue induced a change at position 93 of CDR3α to a positively charged residue, suggesting that CDR3α may interact with position 6 in certain circumstances. Analogous interactions between the TCR α-chain and residues in the C-terminal half of the peptide have not yet been revealed by the limited number of TCR/peptide-MHC crystal structures reported to date. The transgenic mouse approach allows hundreds of TCR/peptide-MHC interactions to be examined comparatively easily, thus permitting a wide-ranging analysis of the possibilities for Ag recognition in vivo.

T cell responses against hapten-modified peptides play an important role in the pathogenesis of certain diseases, including contact dermatitis and allergy. However, the structural features of TCRs recognizing bulky, potentially mobile hapten groups remain poorly defined. To analyze the structural basis of TCR recognition of defined hapten-modified peptides, the immunodominant octapeptide derived from vesicular stomatitis virus nucleoprotein (VSV8) was modified with a trinitrophenyl (TNP) group at the primary TCR contact residues (position 4 or 6) and used for immunization of mice carrying either the TCR α- or β-chain of a VSV8 (unmodified)/H-2Kb-specific CTL clone as a transgene. Such mice allow independent analysis of one TCR chain by maintaining the other fixed. The TCR V gene usage of the responding T cell population was specifically altered depending upon the presence of the TNP group and its position on the peptide. The CDR3 sequences of the TNP-modified peptide-specific TCRs showed a preferential J region usage in both the CDR3α and β loops, indicating that the J regions of both CDR3s are critical for recognition of TNP-modified peptides. In contrast to our previous observations showing the prime importance of CDR3β residues encoded by D-segment or N-addition nucleotides for recognition of position 6 of unmodified VSV8, our studies of TNP-modified peptides demonstrate the importance of the Jβ region, while the Jα region was crucial for recognizing both TNP-modified and unmodified peptides. These data suggest that different structural strategies are utilized by the CDR3α and β loops to allow interaction with a haptenated peptide.

The Ag receptor of cytotoxic CD8+ T lymphocytes recognizes peptides of 8–10 aa bound to MHC class I molecules. This Ag recognition event leads to the activation of the CD8+ lymphocyte and subsequent lysis of the target cell. Altered peptide ligands are analogues derived from the original antigenic peptide that commonly carry amino acid substitutions at TCR contact residues. TCR engagement by these altered peptide ligands usually impairs normal T cell function. Some of these altered peptide ligands (antagonists) are able to specifically antagonize and inhibit T cell activation induced by the wild-type antigenic peptide. Despite significant advances made in understanding TCR antagonism, the molecular interactions between the TCR and the MHC/peptide complex responsible for the inhibitory activity of antagonist peptides remain elusive. To approach this question, we have identified altered peptide ligands derived from the vesicular stomatitis virus peptide (RGYVYQGL) that specifically antagonize an H-2Kb/vesicular stomatitis virus-specific TCR. Furthermore, by site-directed mutagenesis, we altered single amino acid residues of the complementarity-determining region 3 of the β-chain of this TCR and tested the effect of these point mutations on Ag recognition and TCR antagonism. Here we show that a single amino acid change on the TCR CDR3β loop can modulate the TCR-antagonistic properties of an altered peptide ligand. Our results highlight the role of the TCR complementarity-determining region 3 loops for controlling the nature of the T cell response to TCR/altered peptide ligand interactions, including those leading to TCR antagonism.

Comparisons of the structures of different mouse MHC class I molecules define how polymorphic residues determine the unique structural motif and atomic anchoring of their bound peptides. Here, Ted Hansen and colleagues speculate that quantitative differences in how class I molecules interact with peptide, β 2 -microglobulin and molecular chaperones that facilitate peptide loading might determine their relative participation in different pathways of antigen presentation.

To study how the T cell receptor interacts with its cognate ligand, the MHC/peptide complex, we used site directed mutagenesis to generate single point mutants that alter amino acids in the CDR3beta loop of a H-2Kb restricted TCR (N30.7) specific for an immunodominant peptide N52-N59 (VSV8) derived from the vesicular stomatitis virus nucleocapsid. The effect of each mutation on antigen recognition was analyzed using wild type H-2Kb and VSV8 peptide, as well as H-2Kb and VSV8 variants carrying single replacements at residues known to be exposed to the TCR. These analyses revealed that point mutations at some positions in the CDR3beta loop abrogated recognition entirely, while mutations at other CDR3beta positions caused an altered pattern of antigen recognition over a broad area on the MHC/peptide surface. This area included the N-terminus of the peptide, as well as residues of the MHC alpha1 and alpha2 helices flanking this region. Assuming that the N30 TCR docks on the MHC/peptide with an orientation similar to that recently observed in two different TCR-MHC/peptide crystal structures, our findings would suggest that single amino acid alterations within CDR3beta can affect the interaction of the TCR with an MHC surface region distal from the predicted CDR3beta-Kb/VSV8 interface. Such unique recognition capabilities are generated with minimal alterations in the CDR3 loops of the TCR. These observations suggest the hypothesis that extensive changes in the recognition pattern due to small perturbations in the CDR3 structure appears to be a structural strategy for generating a highly diversified TCR repertoire with specificity for a wide variety of antigens.

Whether T-cell receptors (TCRs) recognize antigenic peptides bound to major histocompatability complex (MHC) molecules through common or distinct docking modes is currently uncertain. We report the crystal structure of a complex between the murine N15 TCR [1–4] and its peptide–MHC ligand, an octapeptide fragment representing amino acids 52–59 of the vesicular stomatitis virus nuclear capsid protein (VSV8) bound to the murine H-2K b class I MHC molecule. Comparison of the structure of the N15 TCR–VSV8–H-2K b complex with the murine 2C TCR–dEV8–H-2K b [5] and the human A6 TCR–Tax–HLA-A2 [6] complexes revealed a common docking mode, regardless of TCR specificity or species origin, in which the TCR variable V α domain overlies the MHC α 2 helix and the V β domain overlies the MHC α 1 helix. As a consequence, the complementary determining regions CDR1 and CDR3 of the TCR V α and V β domains make the major contacts with the peptide, while the CDR2 loops interact primarily with the MHC. Nonetheless, in terms of the details of the relative orientation and disposition of binding, there is substantial variation in TCR parameters, which we term twist, tilt and shift, and which define the variation of the V module of the TCR relative to the MHC antigen-binding groove.

The major histocompatibility complex class I complex consists of a heavy chain and a light chain (β2-microglobulin, β2m), which assemble with a short endogenously derived peptide in the endoplasmic reticulum. The class I peptide can be directly exchanged, either at the cell surface or, as recently described, in vesicles of the endocytic compartments, thus allowing exogenous peptides to enter the class I presentation pathway. To probe the interactions between the components of the class I molecule, we analyzed the exchange of peptide and β2m by using purified, recombinant H2-Kb/peptide complexes in a cell-freein vitrosystem. The exchange of competitor peptide was primarily dependent on the off-rate of the original peptide in the class I binding groove. Peptide exchange was not enhanced by the presence of exogenous β2m, as exchange occurred to the same extent in its absence. Thus, the exchange of peptide and β2m are independent events. The exchange rate of β2m also was not affected by the dissociation rates of the original peptides. Furthermore, peptides could substantially exchange into class I molecules over a pH range of 5.5 to 7.5, conditions prevalent in certain endocytic compartments. We conclude that the dynamic properties of the components of class I molecules explain its function as a highly peptide-receptive molecule. The major histocompatibility complex class I can readily receive peptides independent of the presence of exogenous β2m, even at a low pH. Such properties are relevant to class I peptide acquisition, which can occur at the cell surface, as well as in specialized endosomes.

The surface residues of the VSV8/Kb complex important for recognition by N15 and N26 alphabeta T cell receptors (TCR) were mapped by mutational analysis and compared to each other and with epitopes of well-characterized Kb specific monoclonal antibodies (mAb). Three features of immune receptor recognition emerge. First, the footprints of the two TCR on VSV8/Kb are similar with more than 80 % overlap between sites. Given that only 8 of 14 surface exposed VSV8/Kb residues identified as critical for TCR interaction are in common, the chemical basis of the N15 and N26 interactions is nevertheless distinct. Second, the cognate peptide is a major focus of TCR recognition: mutation at any of the three exposed side chains (at p1, p4 or p6) abrogates interaction of both TCR as measured by functional T cell activation. Third, in contrast to TCR, mAb bind to discrete segments on the periphery of the alpha1 and/or alpha2 helices without orientational restriction. These findings suggest that unlike soluble antibodies, surface membrane receptor-ligand interactions on opposing cells (i.e. TCR-peptide/ MHC, CD8-MHC) limit the orientational freedom of the TCR in the immune recognition process.

The use of peptides as a vaccine is a potentially powerful immunization strategy. We explored the possibility of inducing an efficient cytotoxic T lymphocyte (CTL) mediated immune response in mice, using in vitro reconstituted major histocompatibility complex (MHC) class I/peptide complexes as the immunogen. Recombinant derived H-2Kb and β2-microglobulin (β2m) were properly folded into an MHC class I complex using the vesicular stomatitis virus (VSV)-8mer from the natural nucleocapsid proteinN52–59 (RGYVYQGL), an immunodominant Kb epitope in C57BL 6 (B6) mice. After immunizing mice with the H-2Kb class I/VSV peptide complex and a subsequent in vitro stimulation with the VSV peptide alone, a specific CTL response was demonstrated. The method was also applicable to other peptides, for example, the Sendai virus (SV) peptideN324–332 (FAPGNYPAL). The CTL response was mediated by CD3 + CD8 + T cells and was shown to be allele specific, as only peptide loaded target cells expressing the H-2Kb allele could be recognized. It is of interest that extremely small amounts of injected MHC class I/peptide complex (i.e. 500 pg) could generate a measurable CTL response. The MHC class I/peptide complex had to be intact and properly folded to elicit an immune response, suggesting that the complex protected the peptide for internalization by antigen presenting cells (APCs) or for delivering to the proper site for peptide exchange on the cell surface of APCs. The described immunizing method can be routinely used to prime a CTL response by employing in vitro folded MHC class I/peptide complexes, without the use of adjuvants. It appears to be efficient, sensitive and specific. By using the recombinant protein system, unlimited amounts of MHC class I/peptide complex can be produced for immunization. Moreover, this protocol permits different in vitro combinations of allelic MHC class I molecules and peptide variants.

As an approach to determine the structural basis of interactions between T cell receptors (TCRs) and MHC class I/peptide complexes, the fine specificities of a panel of vesicular stomatitis virus (VSV)-specific CTL clones recognizing the antigenic peptide (nucleoprotein 52-59) and the class I (Kb) molecule were correlated with the TCR primary structure. Each TCR showed a distinct interaction pattern with N52-59 and the Kb molecule. The large majority of the TCRs expressed by the panel of CTL clones used V beta 13 gene segments that had randomly recombined with D beta and J beta gene segments. The alpha chains were from randomly assorted V alpha and J alpha gene segments. Thus, the panel was found to be a highly heterogeneous set of TCRs, each member of which appeared to have an unique surface interface area, the recognition site, that interacted with a complementary surface formed by the single peptide bound in the class I antigenic groove.

Generation of soluble T-cell receptor (TCR) molecules by a variety of genetic engineering methods has been hampered by inefficient pairing of alpha and beta subunits in the absence of their respective transmembrane regions and associated CD3 components. To overcome this obstacle, we have added 30-amino acid-long segments to the carboxyl termini of alpha and beta extracellular domains via a cleavable flexible linker. These peptide segments (BASE-p1 for alpha and ACID-p1 for beta) have been previously shown to selectively associate to form a stable heterodimeric coiled coil termed a leucine zipper. Homodimeric structures are not permitted due to electrostatic repulsion among amino acid side chains. Expression of a representative TCR-leucine zipper fusion protein in a baculovirus expression system results in production of alpha beta TCR heterodimer at 0.6-1.4 mg/liter. This yield is 5- to 10-fold greater than that of the TCR expressed in the absence of the synthetic leucine zipper sequence. The structure of the TCR component of the fusion heterodimer was judged to be native when probed with a panel of 17 mAbs specific for alpha and beta constant and variable domains. A mAb specific for the isolated BASE-p1/ACID-p1 coiled coil was also generated and shown to react with the TCR fusion protein. The above technology should be broadly useful in the efficient production and purification of TCRs as well as other heterodimeric proteins.

We have identified two forms of a major histocompatibility complex (MHC) class I molecule, H-2Kb, distinguishable by specific antibodies through a study of a genetically engineered mouse cell line that overexpresses these molecules. One form, a complex associated with beta 2-microglobulin (native, beta 2m+ class I), is detectable by conformation-dependent antibodies. The other form, which remains after preclearing cell lysates of native class I, is only poorly, if at all, associated with beta 2-microglobulin (beta 2m- class I) and is detectable by an antiserum against the cytoplasmic tail region of H-2K molecules. Both forms are also present in normal cell lines. The affinity-purified native class I molecules bind short peptides (8 or 9 residues) and assemble tightly with beta 2-microglobulin. In striking contrast, the beta 2m- class I molecules bind peptides that are longer (> 15 residues) than those bound to native class I molecules. This finding is consistent with the recent evidence that peptides longer than 8-10 amino acid residues are transported into the endoplasmic reticulum and suggests the possibility of a control step for peptide presentation by MHC in which the incompletely processed peptides bind to the heavy chain and a selected fraction undergoes final processing and presentation on the cell surface.

Solution at 2.4 A resolution of the structure of H-2Db with the influenza virus peptide NP366-374 (ASNEN-METM) and comparison with the H-2Kb-VSV (RGY-VYQGL) structure allow description of the molecular details of MHC class I peptide binding interactions for mice of the H-2b haplotype, revealing a strategy that maximizes the repertoire of peptides than can be presented. The H-2Db cleft has a mouse-specific hydrophobic ridge that causes a compensatory arch in the backbone of the peptide, exposing the arch residues to TCR contact and requiring the peptide to be at least 9 residues. This ridge occurs in about 40% of the known murine D and L allelic molecules, classifying them as a structural subgroup.

Unlike the highly polymorphic major histocompatibility complex (MHC) class Ia molecules, which present a wide variety of peptides to T cells, it is generally assumed that the nonpolymorphic MHC class Ib molecules may have evolved to function as highly specialized receptors for the presentation of structurally unique peptides. However, a thorough biochemical analysis of one class Ib molecule, the soluble isoform of Qa-2 antigen (H-2SQ7b), has revealed that it binds a diverse array of structurally similar peptides derived from intracellular proteins in much the same manner as the classical antigen-presenting molecules. Specifically, we find that SQ7b molecules are heterodimers of heavy and light chains complexed with nonameric peptides in a 1:1:1 ratio. These peptides contain a conserved hydrophobic residue at the COOH terminus and a combination of one or more conserved residue(s) at P7 (histidine), P2 (glutamine/leucine), and/or P3 (leucine/asparagine) as anchors for binding SQ7b. 2 of 18 sequenced peptides matched cytosolic proteins (cofilin and L19 ribosomal protein), suggesting an intracellular source of the SQ7b ligands. Minimal estimates of the peptide repertoire revealed that at least 200 different naturally processed self-peptides can bind SQ7b molecules. Since Qa-2 molecules associate with a diverse array of peptides, we suggest that they function as effective presenting molecules of endogenously synthesized proteins like the class Ia molecules.

Using Mitomycin C mutagenesis and negative and positive selection with monoclonal antibodies specific for H-2Kb and H-2Kbm10, respectively, a mutant cell line clone, Mitc-182, was isolated. Direct sequencing of uncloned cDNA as well as PCR based cloning and sequencing of the H-2Kb182 transcript from this mutant revealed a single G-->T transversion resulting in the substitution of Trp167 by cysteine. Serologically, the mutant Kb182 and Kbm10 are almost identical as each has lost at least five Kb specific mAb epitopes and gained several new epitopes. Interestingly, the mutant cell line, Mitc-182, is efficiently recognized by alloreactive CTLs raised in reciprocal combinations, e.g. CB6 anti Cbm10 and Cbm10 anti CB6, indicating that Kb182 contains both Kb and Kbm10 specific epitopes. The mutation has not affected the ability of Kb182 to present Kb restricted antigenic peptides of Sendai and vesicular stomatitis viruses. In addition to underscoring the importance of amino acid residue 167 in alloreactivity, these results indicate a positive correlation between the gain of both an mAb epitope and a defined alloreactive CTL epitope.

This study describes serological and biochemical properties of a novel MHC class I molecule. The mutant H-2Ksm1 molecule was discovered in a mouse because of loss of reactivity of its peripheral blood lymphocytes to monoclonal antibodies. This mutation in the H-2Ks molecule is the first in vivo mutation described that has altered an amino acid residue (amino acid 107) distant from the regions generally considered to be peptide or TCR contacts. Cell surface expression of the mutant molecules remains high but the Arg107 to Trp substitution appears to alter the native protein conformation, markedly decreasing cell surface association with beta 2-microglobulin light chains and conferring a loss of recognition by Ks specific antibodies.

Summary The major histocompatibility complex class I molecules are receptors for intracellular peptides, both of self and non-self origin. When non-self peptides (eg., pathogen derived) are bound to the class I molecules, they form ligands for T cell receptors resulting in antigen specific lysis of the infected cells by cytotoxic T lymphocytes. Therefore, an understanding of the process of antigen recognition requires the precise definition of the structural features of the bimolecular complex formed by a single well defined antigenic peptide bound to the class I molecule. A strategy using antibodies was developed to probe the structural features of the H-2K b containing a defined peptide in the antigen cleft. We report that the binding surface area of a K b specific monoclonal antibody (28-13-3s) includes residues in the α1 (Gly56 and Glu58) and α2 (Trp167) helices of K b thus, binding across the antigen binding groove. When cells treated with the antigenic peptide of vesicular stomatitis virus, N52-59, and its alanine substituted analogs were tested for 28-13-3s binding, it was found that position 1 of the peptide also forms a part of the antibody binding site. This finding strongly supports the positioning of the N-terminus of N52-59 proximal to pocket A, thus, assuming an orientation parallel to the a 1 helix.

This study describes a novel MHC class I mouse mutant that was discovered because of loss of reactivity of its cells to monoclonal antibodies. The mutation occurred in the H-2Ks molecule and is the first in vivo mutation described that has a single altered amino acid residue (amino acid 107) distant from the regions considered to be peptide or TCR contacts. Nevertheless, skin grafts from the mutant to the parent are rejected by CD8+ T-cells. In the reciprocal direction, the mutant shows partial tolerance to parental skin grafts, suggesting that the mutant is inefficient in selecting alloreactive T-cells specific for the wild-type Ks molecule.

Major histocompatibility complex (MHC) class I molecules combine with short peptides of defined length and sequence. Here we describe an approach that may be used in the analysis of peptide preference of different allelic MHC class I molecules, and in the determination of T cell epitopes. We produced synthetic "peptide libraries" of limited complexity by standard peptide chemistry. Using these peptide mixtures we show that H-2 Kb molecules can accommodate both 8- and 9-residue peptides, whereas Db molecules are unable to combine with peptides shorter than 9 amino acids present in these libraries. When these peptide mixtures are used to provide "fingerprints" of Db molecules and mutants thereof, both loss and gain of the ability to combine with certain peptides is observed. For the Kbm1 mutant a strong influence of amino acid substitutions in class I molecules on the peptides selected is observed. In these synthetic peptide mixtures, the presence of a specific T cell epitope, known to be represented once, can be detected. This approach may be extended to the identification of new T cell epitopes from larger peptide libraries.

We have examined the effect of diversity in the antigen-binding groove of the Kb, Db, Kbm1, and Kbm8 major histocompatibility complex (MHC) class I molecules on the set of self-peptides they present on the cell surface, by using a procedure we recently developed in our laboratory to isolate endogenously processed peptides bound to MHC class I molecules. We found that such naturally processed peptides are 7-10 amino acids long. A major motif of tyrosine and phenylalanine residues at positions three and five was found for peptides binding to Kb. The availability of Kb mutant molecules Kbm1 and Kbm8, each with localized clustered changes in the antigen-binding cleft, allowed us to probe the effect of such small alterations on peptide selection. We found that such changes in different regions in the antigen-binding groove exert an absolute effect by changing subsets of self-peptides bound to these MHC molecules. In the Kbm1 mutant, the binding of the characteristic major set of Kb-associated peptides with tyrosine at position three or both positions three and five is abrogated, although this MHC molecule still binds peptides with tyrosine at position seven; the latter peptides also bind to Kb. Kbm8 shares the major Tyr-3, Tyr-5 peptide set that binds to Kb but does not bind the peptides with tyrosine at position seven. Thus differences in binding selectivity in Kbm1 and Kbm8 appear to be the major determinant for the observed alterations in in vivo immune responses.

Numerous tumor cell lines of leukemic origin are known to modulate cell surface expression of major histocompatibility complex (MHC) class I antigens resulting in alterations in their immune detection and tumorigenicity. We have been studying the mechanisms responsible for attenuation of MHC class I gene expression in an H-2 heterozygous (H-2b x H-2d) Abelson-Murine leukemia virus (A-MuLV)-transformed leukemic cell line (designated R8). Here we report that treatment of the R8 cell line with the protein synthesis inhibitor cycloheximide (CHX) increased H-2Kb steady-state messenger RNA (mRNA) levels several fold. The induced H-2Kb mRNA transcripts were functional, as demonstrated by their ability to be translated into immunoprecipitable H-2Kb alloantigen. H-2Kb null variants derived from the R8 cell line were shown to be the product of both cis- and trans-acting mechanisms, insomuch as the treatment of R8-derived H-2Kb non-expressor lines with CHX re-established expression of H-2Kb mRNA to the same extent as transfection of the variant cell line with the wild-type H-2Kb gene. Such findings indicate that downregulation of MHC class I gene expression is constitutive for the R8 leukemic cell line, a phenomenon that may be related to the immature pre-B-cell phenotype of this A-MuLV transformant.

Using an approach for isolating and characterizing peptide fractions that are intracellularly associated with major histocompatibility complex class I molecules, the major peptide recognized by cytotoxic T cells specific for the vesicular stomatitis virus has been isolated from the H-2Kb molecule of infected cells. This endogenously processed octapeptide is allele-specific as it does not bind to H-2Db molecules, and contains the core sequence of the epitope of the nucleocapsid protein of the vesicular stomatitis virus identified by testing with exogenous synthetic peptides.

The “bg” series of MHC mutations is the most prevalent type of mutations of Kb in C57BL/6 mice screened by reciprocal tail skin grafting. The basis for identification of this series of mutations is the incompatibility of grafts between the parental B6 and the mutant. This series takes the longest to reciprocally reject the skin grafts. The series can be subdivided into “bg 1” and “bg 2” groups based on Kb-restricted recognition of virus-infected mutant target cells. The biochemical basis for these mutations are amino acid substitutions at residues 116 and 121 of the Kb transplantation antigen. These substitutions do not alter monoclonal antibody binding sites. The structural basis of MAb binding and the genetic basis of the mutation are discussed.

T-cell activation results from engagement of the T-cell receptor (TCR) by cognate peptide-major histocompatibility complex (pMHC) complexes on the surface of antigen-presenting cells (APC). Previous studies have provided evidence supporting the notion that the half-life of the TCR/pMHC interaction and the density of pMHC on the APC are two parameters that can influence T-cell activation. However, whether the half-life of the TCR/pMHC interaction can modulate T-cell activation in response to a pathogen challenge remains unknown. To approach this question, we generated strains of bacteria expressing variants of the ovalbumin (OVA) antigen, carrying point mutations in the SIINFEKL sequence. When bound to H-2K(b), this peptide is the cognate ligand for the OT-I TCR. Variants of the H-2K(b)/SIINFEKL bind to the OT-I TCR with distinct half-lives. Here we show that dendritic cells (DCs) infected with bacteria expressing OVA variants were incapable of activating OT-I T cells when the half-life of the TCR/H-2K(b)/OVA interaction was excessively short. Consistent with these data, T-cell activation was only observed in mice infected with bacteria expressing OVA variants that bound to OT-I with a half-life above a certain threshold. Considered together, our data suggest that the half-life of TCR/pMHC interaction can significantly modulate T-cell activation in vivo, as well as influence recognition of antigens expressed by bacteria. These observations underscore the importance of the TCR/pMHC half-life on the clearance of pathogens.

Summary The T cell immunoglobulin mucin (Tim) family of receptors regulates effector CD4 + T cell functions and is implicated in autoimmune and allergic diseases. Tim-3 induces immunological tolerance, and engagement of the Tim-3 immunoglobulin variable (IgV) domain by galectin-9 is important for appropriate termination of T helper 1-immune responses. The 2 A crystal structure of the Tim-3 IgV domain demonstrated that four cysteines, which are invariant within the Tim family, form two noncanonical disulfide bonds, resulting in a surface not present in other immunoglobulin superfamily members. Biochemical and biophysical studies demonstrated that this unique structural feature mediates a previously unidentified galectin-9-independent binding process and suggested that this structural feature is conserved within the entire Tim family. The current work provided a graphic example of the relationship between sequence, structure, and function and suggested that the interplay between multiple Tim-3-binding activities contributes to the regulated assembly of signaling complexes required for effective Th1-mediated immunity.

The vesicular stomatitis virus (VSV) octapeptide RGYVYQGL binds to H-2Kb and triggers a cytotoxic T cell response in mice. A variant peptide, RGYVYEGL (E6) with a glutamic acid for glutamine replacement at position 6 of the VSV peptide, elicits a T cell response with features that are quite different from those elicited by the wild-type VSV peptide. The differences found in the nature of the T cells responding to the E6 peptide include changes in both the Vβ elements and the sequences of the complementarity-determining region 3 loops of their TCRs. Further experiments found that the E6 peptide can act as an antagonist for VSV-specific T cell hybridomas. To determine whether these differences in Vβ usage, complementarity-determining region 3 sequences, and the switch from agonism to antagonism are caused by a conformational change on the MHC, the peptide, or both, we determined the crystal structure of the variant E6 peptide bound to H-2Kb. This structure shows that the only significant structural difference between H-2Kb/E6 and the previously determined H-2Kb/VSV is limited to the side chain of position 6 of the peptide, with no differences in the MHC molecule. Thus, a minor conformational change in the peptide can profoundly alter the biological outcome of the TCR-peptide/MHC interaction.

Previous work has indicated that an important component for the initiation of autoimmune insulin-dependent diabetes mellitus (IDDM) in the NOD mouse model entails MHC class I-restricted CD8 T cell responses against pancreatic β cell Ags. However, unless previously activated in vitro, such CD8 T cells have previously been thought to require helper functions provided by MHC class II-restricted CD4 T cells to exert their full diabetogenic effects. In this study, we show that IDDM development is greatly accelerated in a stock of NOD mice expressing TCR transgenes derived from a MHC class I-restricted CD8 T cell clone (designated AI4) previously found to contribute to the earliest preclinical stages of pancreatic β cell destruction. Importantly, these TCR transgenic NOD mice (designated NOD.AI4αβ Tg) continued to develop IDDM at a greatly accelerated rate when residual CD4 helper T cells were eliminated by introduction of the scid mutation or a functionally inactivated CD4 allele. In a previously described stock of NOD mice expressing TCR transgenes derived from another MHC class I-restricted β cell autoreactive T cell clone, IDDM development was retarded by elimination of residual CD4 T cells. Hence, there is variability in the helper dependence of CD8 T cells contributing to the development of autoimmune IDDM. The AI4 clonotype represents the first CD8 T cell with a demonstrated ability to progress from a naive to functionally activated state and rapidly mediate autoimmune IDDM development in the complete absence of CD4 T cell helper functions.

Recently, a powerful approach for the detection of MHC/peptide-specific T cells has been made possible by the engineering of soluble-tetrameric MHC/peptide complexes, consisting of singly biotinylated MHC/peptide molecules bound to fluorescent-labeled streptavidin. These tetrameric molecules are thought to compensate for the low affinity and relative fast dissociation rate of the TCR/MHC-peptide interaction by increasing the avidity of this interaction, thus allowing the stable binding of MHC/peptide tetramers to TCR expressing cells. Here we describe a new more simplified procedure for obtaining MHC/peptide tetramers using the well-characterized H-2K(b)/VSV system. This procedure consists of the incorporation of an unpaired cysteine residue at the C-terminus of the H-2K(b) molecule, allowing site-specific biotinylation by a -SH-specific biotinylating reagent. The H-2K(b)/VSV tetramers bound only to hybridomas expressing H-2K(b)/VSV-specific TCRs. When coated on a plate, these tetramers were able to induce IL-2 release by those hybridomas. Furthermore, H-2K(b)/VSV tetramers bound to CTL populations obtained from mice immunized with VSV-peptide. The specificity of the binding was further refined by studying cross-recognition of VSV by CTL populations obtained from mice immunized with single amino acid substituted VSV peptide variants. H-2K(b)/VSV tetramers bound only to those CTL populations that cross-reacted with the wild-type VSV peptide. Our method provides a simple, efficient and inexpensive procedure for making MHC/peptide tetramers, a highly specific and very useful reagent with a number of important applications in basic and clinical T cell research.

The specificity of T cell-mediated immune responses is primarily determined by the interaction between the T cell receptor (TCR) and the antigenic peptide presented by the major histocompatibility complex (MHC) molecules. To refine our understanding of interactions between the TCR and the antigenic peptide of vesicular stomatitis virus (VSV) presented by the class I MHC molecule H-2Kb, we constructed a TCR α chain transgenic mouse in a TCR α-deficient background to define specific structural features in the TCR β chain that are important for the recognition of the VSV/H-2Kbcomplex. We found that for a given peptide, a peptide-specific, highly conserved amino acid could always be identified at position 98 of the complementarity-determining region 3 (CDR3) loop of TCR β chains. Further, we demonstrated that substitutions at position 6, but not position 1, of the VSV peptide induced compensatory changes in the TCR in both the amino acid residue at position 98 and the length of the CDR3β loop. We conclude that the amino acid residue at position 98 of the CDR3β loop is a key residue that plays a critical role in determining the specificity of TCR–VSV/H-2Kbinteractions and that a specific length of the CDR3β loop is required to facilitate such interactions. Further, these findings suggest that the α and β chains of TCRs interact with amino acid residue(s) toward the N and C termini of the VSV peptide, respectively, providing functional evidence for the orientation of a TCR with its peptide/MHC ligand as observed in the crystal structures of TCR/peptide/MHC complexes.

Each T cell receptor (TCR) recognizes a peptide antigen bound to a major histocompatibility complex (MHC) molecule via a clonotypic alphabeta heterodimeric structure (Ti) non-covalently associated with the monomorphic CD3 signaling components. A crystal structure of an alphabeta TCR-anti-TCR Fab complex shows an Fab fragment derived from the H57 monoclonal antibody (mAb), interacting with the elongated FG loop of the Cbeta domain, situated beneath the Vbeta domain. This loop, along with the partially exposed ABED beta sheet of Cbeta, and glycans attached to both Cbeta and Calpha domains, forms a cavity of sufficient size to accommodate a single non-glycosylated Ig domain such as the CD3epsilon ectodomain. That this asymmetrically localized site is embedded within the rigid constant domain module has implications for the mechanism of signal transduction in both TCR and pre-TCR complexes. Furthermore, quaternary structures of TCRs vary significantly even when they bind the same MHC molecule, as manifested by a unique twisting of the V module relative to the C module.

The molecular interactions between the CD8 co-receptor dependent N15 and N26 T cell receptors (TCRs) and their common ligand, the vesicular stomatitis virus octapeptide (VSV8) bound to H-2Kb, were studied to define the docking orientation(s) of MHC class I restricted TCRs during immune recognition. Guided by the molecular surfaces of the crystallographically defined peptide/MHC and modeled TCRs, a series of mutations in exposed residues likely contacting the TCR ligand were analyzed for their ability to alter peptide-triggered IL-2 production in T cell transfectants. Critical residues which diminished antigen recognition by 1000 to 10,000-fold in molar terms were identified in both N15 Valpha (alphaE94A or alphaE94R, Y98A and K99) and Vbeta (betaR96A, betaW97A and betaD99A) CDR3 loops. Mutational analysis indicated that the Rp1 residue of VSV8 is critical for antigen recognition of N15 TCR, but R62 of H-2Kb is less critical. More importantly, the alphaE94R mutant could be fully complemented by a reciprocal charge reversal at Kb R62 (R62E). This result suggests a direct interaction between N15 TCR Valpha E94R and Kb R62E residues. As Rp1 of VSV8 is adjacent to R62 in the VSV8/Kb complex and essential for T cell activation, this orientation implies that the N15 Valpha CDR3 loop interacts with the N-terminal residues of VSV8 with the Valpha domain docking to the Kb alpha2 helix while the N15 Vbeta CDR3 loop interacts with the more C-terminal peptide residues and the Vbeta domain overlies the Kb alpha1 helix. An equivalent orientation is suggested for N26, a second VSV8/Kb specific TCR. Given that genetic analysis of two different class II MHC-restricted TCRs and two crystallographic studies of class I restricted TCRs offers a similar overall orientation of V domains relative to alpha-helices, these data raise the possibility of a common docking mode between TCRs and their ligands regardless of MHC restriction.

Solution at 2.5-Å resolution of the three-dimensional structure of H-2Ldwith a single nine-residue peptide provides a structural basis for understanding its unique interaction with beta-2 microglobulin (β2m) and peptide. Consistent with the biological data that show an unusually weak association of Ldwith β2m, a novel orientation of the α1/α2 domains of Ldrelative to β2m results in a dearth of productive contacts compared with other class I proteins. Characteristics of the Ldantigen-binding cleft determine the unique motif of peptides that it binds. Ldhas no central anchor residue due to the presence of several bulky side chains in its mid-cleft region. Also, its cleft is significantly more hydrophobic than that of the other class I molecules for which structures are known, resulting in many fewer H-bonds between peptide and cleft residues. The choice of Pro as a consensus anchor at peptide position 2 appears to be related to the hydrophobicity of the B pocket, and to the unique occurrence of Ile (which mirrors Pro in its inability to form H-bonds) at position 63 on the edge of this pocket. Thus, the paucity of stabilizing H-bonds combined with poor complementarity between peptide postion 2 Pro and the B pocket contribute to the weak association between Ldand its peptide antigen. The unique structural interactions of Ldwith β2m and peptide could make Ldmore suited than other classical class I molecules to play a role in alternative pathways of antigen presentation.

A strategy to overexpress T cell receptors (TCRs) in Lec3.2.8.1 cells has been developed using the "Velcro" leucine zipper sequence to facilitate alpha-beta pairing. Upon secretion in culture media, the VSV-8-specific/H2-Kb-restricted N15 TCR could be readily immunopurified using the anti-leucine zipper monoclonal antibody 2H11, with a yield of 5-10 mg/liter. Mass spectrometry analysis revealed that all attached glycans were GlcNAc2-Man5. Following Superdex 200 gel filtration to remove aggregates, wild-type N15 or N15(s), a C183S variant lacking the unpaired cysteine at amino acid residue 183 in the Cbeta domain, was thrombin-cleaved and endoglycosidase H-digested, and the two derivatives were termed iN15DeltaH and N15(s)DeltaH, respectively, and sized by Superdex 75 chromatography to high purity. N-terminal and C-terminal microsequencing analysis showed the expected unique termini of N15 alpha and beta subunits. Nevertheless, neither protein crystallized under a wide range of conditions. Subsequently, we produced a Fab fragment of the murine TCR Cbeta-specific hamster monoclonal antibody H57 and complexed the Fab fragment with iN15DeltaH and N15(s)DeltaH. Both N15(s)DeltaH-Fab[H57] and iN15DeltaH-Fab[H57] complexes crystallize, with the former diffracting to 2.8-A resolution. These findings show that neither intact glycans nor the conserved and partially exposed Cys-183 is required for protein stability. Furthermore, our results suggest that the H57 Fab fragment aids in the crystallization of TCRs by altering their molecular surface and/or stabilizing inherent conformational mobility.

Publisher Summary The class I major histocompatibility complex (MHC) molecule consists of a highly variable heavy chain in complex with a light chain (β 2 -microglobulin, β 2 m) and peptides of approximately 8–10 amino acids in length. This chapter describes a novel method for the rapid assembly of properly folded MHC ternary complexes in relatively high yields, utilizing exogenous synthetic peptides and over-expressed polypeptides from inclusion bodies. The method is reproducible, and capable of rapidly and efficiently forming stable ternary complexes. The presence of glutathione in the incubation mixture is essential, emphasizing that during the refolding process proper formation of disulfide bonds is critical, and is consistent with the observation that, there are no free sulfhydryl groups in initially solubilized heavy chain, suggesting that mispaired disulfide bonds prevent the correct folding to ternary complex. The adaptability of the method to analysis of multiple samples facilitates screening of peptide variants, and consequently, the study of the rules for epitope binding. Because of the efficient manner by which MHC ternary complexes can be formed, this method has potential for rapid screening of peptide analogues, possibly by radio assay, or for use in equilibrium and kinetic studies.

The three-dimensional structures of three human and two murine class I molecules, in complex with single peptides and with mixtures of endogenous peptides, have now been determined to high resolution. These structures have afforded important insights into the way in which antigenic peptides are bound by an MHC class I molecule, and how a given MHC molecule can bind a large number and variety of peptides, for presentation to a T cell receptor. Peptides are bound in a cleft located in the alpha 1/alpha 2 domain of a class I molecule. They are tethered by an array of hydrogen bonding interactions many of which are conserved among the different structures. Binding is also accomplished through van der Waals interactions between two or three peptide residues and complementary pockets in the cleft. The location and the characteristics of its pockets are unique to a given MHC class I molecule, and so determine the identity of the anchor residues of the set of peptides that bind. The antigenic epitope recognized by the TCR consists of residues on the MHC as well as the side chains of those peptide residues that point out from the cleft. The various strategies used to expand the repertoire of peptides bound and presented are discussed.

It is now an accepted fact that peptides of self or non-self origin form an essential component of the MHC class I structure. The peptide component of the heterotrimer contains the essential determinants recognized by the T-cell receptors of cytotoxic T lymphocytes, be it an antigen-specific, alloimmune or autoimmune response. Because of the importance of the recognition process, several methods have been developed to characterize naturally processed peptides presented by the class I molecules.

To study the structure of a homogenous major histocompatibility complex (MHC) class I molecule containing a single bound peptide, a complex of recombinant mouse H-2Kb, beta 2-microglobulin (beta 2m), and a fragment of the vesicular stomatitis virus (VSV) nuclear capsid protein, VSV-(N52-59) octapeptide (Arg-Gly-Tyr-Val-Tyr-Gln-Gly-Leu), was prepared by exploiting a high-yield bacterial expression system and in vitro cocomplex formation. The structure of mouse H-2Kb revealed its similarity to three human class I HLA molecules, consistent with the high primary sequence homology and common function of these peptide-presenting molecules. Electron density was located in the peptide-binding groove, to which a single peptide in a unique conformation was unambiguously fit. The peptide extends the length of the groove, parallel to the alpha-helices, and assumes an extended, mostly beta-strand conformation. The peptide is constrained within the groove by hydrogen bonding of its main-chain atoms and by contacts of its side chains with the H-2Kb molecule. The amino-terminal nitrogen atom of the peptide forms a hydrogen bond with the hydroxyl group of Tyr-171 of H-2Kb at one end of the groove, while the carboxyl-terminal oxygen forms a hydrogen bond with the hydroxyl group of Tyr-84 at the other end. Since the amino acids at both ends are conserved among human and mouse MHC molecules, this anchoring of each end of the peptide appears to be a general feature of peptide-MHC class I molecule binding and imposes restrictions on its length. The side chains of residues Tyr-3, Tyr-5, and Leu-8 of the VSV octapeptide fit into the interior of the H-2Kb molecule with no appreciable surface exposure, a finding in support of previous biological studies that showed the importance of these residues for binding. Thus, the basis for binding of specific peptide sequences to the MHC class I molecule is the steric restriction imposed on the peptide side chains by the architecture of the floor and sides of the groove. The side chains of Arg-1, Val-4, and Gln-6 and the main-chain of Gly-7 of the octapeptide are exposed on the surface of the complex, thus confirming their availability for T-cell receptor contact, as previously demonstrated by T-cell recognition experiments.

The basis for the immune response against intracellular pathogens is the recognition by cytotoxic T lymphocytes of antigenic peptides derived from cytosolic proteins, which are presented on the cell surface by major histocompatibility complex (MHC) class I molecules. The understanding of MHC class I-restricted peptide presentation has recently improved dramatically with the elucidation of the structural basis for the specificity of peptide binding to MHC class I molecules and the identification of proteins encoded in the class II region of the MHC that are putatively involved in the production of peptides and their transport into the endoplasmic reticulum, where they assemble with class I molecules.

To study the structural basis for specificity and affinity of cytotoxic T lymphocytes for major histocompatibility complex/peptide complexes, we have employed a cytotoxic T lymphocyte (CTL)-mediated immunoselection approach to obtain H-2Kb structural mutants which are resistant to lysis by a Kb-specific alloreactive CTL clone. In this study we describe the Kb structural mutant, designated R8.60.14, recovered following immunoselection using the CD8-dependent CTL clone 60 as a selective agent. Although serologically unaltered with respect to Kb expression, R8.60.14 is not recognized by CD8-dependent, Kb-specific CTL. DNA sequence analysis revealed a single Glu----Lys amino acid substitution at position 222 in the Kb alpha 3-domain of this variant. To determine if a direct correlation exists between CD8 dependence of a Kb specific CTL and its failure to respond to R8.60.14, we examined the lytic response against R8.60.14 by CD8-independent, Kb-specific CTL obtained from long-term culture in the presence of anti-CD8 monoclonal antibody, 3.155. CD8-independent CTL exhibit no difference in their response against the R8 parent and R8.60.14 variant. This study demonstrates unequivocally that Kb-specific recognition of R8.60.14 by CD8-independent CTL is unaltered, while the response by CD8-dependent CTL is completely abrogated. Thus, the sole basis for emergence of this variant in the CTL-mediated immunoselection approach used in this study resides in the alteration of a single CD8-binding site residue at position 222 in the Kb alpha 3 domain. The functional importance of this Glu222 residue for the interaction between the CD8 molecule on CD8-dependent CTL and the Kb alpha 3 domain is further reinforced by virtue of the recovery of the R8.60.14 variant on the basis of its resistance to lysis by a CD8-dependent CTL clone in this CTL-mediated immunoselection approach.

The immune response against intracellular pathogens is mediated by cytotoxic T lymphocytes (CTL) which recognize components of the invader in combination with self molecules of an infected cell. These self products are encoded in the class I region of the Major Histocompatibility Complex (MHC) and the process is termed MHC restriction (1). Originally it was thought that proteins of the infectious agent were present in close proximity to the surface MHC molecule where they could be simultaneously contacted by the T cell receptor (TCR). Therefore it was hypothesized that spike proteins of viruses which are present on cell surfaces after fusion with the cell membrane at cell entry, or before virus budding, would be likely candidates for such a dual recognition. However, from studies with influenza viruses with variations in the nonsurface exposed nucleoprotein (Np) it was learned that, unlike antibodies, the major population of influenza specific CTL recognize the Np rather than the surface exposed hemagglutinin (2). Studies in which the Np gene or parts thereof were transfected into L-cells further showed that the influenza CTL were specifically recognizing a linear peptide stretch in Np (3). Moreover the specific CTL response could be triggered when short synthetic peptides were mixed in vitro with antigen presenting cells carrying the appropriate MHC class I molecules (4).

The ability to down-regulate major histocompatibility complex class I antigen expression on allografts prior to transplantation would be expected to improve their survival in immunocompetent recipients. In order to identify genetic mechanisms that mediate attenuation of MHC class I antigen expression, we have begun characterizing H-2Kb surface null somatic cell variants derived from an H-2 heterozygous tumor cell line (H-2b X H-2d). These variants have sustained a modification in cell surface MHC phenotype, as evidenced by their failure to be recognized by both anti-H-2Kb antibodies and cytotoxic T lymphocytes. The mutant phenotype for one such variant (designated 69.9.15) was marked by the expression of abundant H-2Kb mRNA and immuno-precipitable H-2Kb protein in cell lysates. The failure in cell surface expression of the H-2Kb antigen was caused by a single base change (G to A transition) in exon 3, encoding the second external domain (alpha 2) of the H-2Kb molecule. The mutation resulted in the substitution of Tyr for Cys at amino acid position 164, thereby disrupting an intrachain disulfide linkage formed between Cys 101 and 164. In contrast to the wild-type H-2Kb gene, DNA-mediated transfer of the mutant H-2Kb gene into mouse L cell fibroblasts failed to result in cell surface expression of the H-2Kb antigen, although both the wild-type and mutant genes were transcribed to equivalent levels. These data indicate that a genetic event as limited as somatic point mutation can abrogate expression of a MHC class I antigen and provide support for the hypothesis that protein folding plays an important role in the cell surface expression of MHC class I molecules.