Roles of the six peptide-binding pockets of the HLA-A2 molecule in allorecognition by human cytotoxic T-cell clones

Molecular docking to investigate HLA-associated idiosyncratic drug reactions

Idiosyncratic drug reactions (IDRs) pose severe threats to patient health. Unlike conventionally dose-dependent side effects, they are unpredictable and more frequently manifest as life-threatening conditions, such as severe cutaneous adverse reactions (SCARs) and drug-induced liver injury (DILI). Some HLA alleles, such as HLA-B*57:01, HLA-B*15:02, and HLA-B*58:01, are known risk factors for adverse reactions induced by multiple drugs. However, the structural basis underlying most HLA-associated adverse events remains poorly understood. This review summarizes the application of molecular docking to reveal the mechanisms of IDR-related HLA associations, covering studies using this technique to examine drug-HLA binding pockets and identify key binding residues. We provide a comprehensive overview of risk HLA alleles associated with IDRs, followed by a discussion of the utility and limitations of commonly used molecular docking tools in simulating complex molecular interactions within the HLA binding pocket. Through examples, including the binding of abacavir and flucloxacillin to HLA-B*57:01, carbamazepine to HLA-B*15:02, and allopurinol to HLA-B*58:01, we demonstrate how docking analyses can provide insights into the drug and HLA allele-specificity of adverse events. Furthermore, the use of molecular docking to screen drugs with unknown IDR liability is examined, targeting either multiple HLA variants or a single specific variant. Despite multiple challenges, molecular docking presents a promising toolkit for investigating drug-HLA interactions and understanding IDR mechanisms, with significant implications for preemptive HLA typing and safer drug development.

Structural Prediction of Peptide-MHC Binding Modes

The immune system is constantly protecting its host from the invasion of pathogens and the development of cancer cells. The specific CD8⁺ T-cell immune response against virus-infected cells and tumor cells is based on the T-cell receptor recognition of antigenic peptides bound to class I major histocompatibility complexes (MHC) at the surface of antigen presenting cells. Consequently, the peptide binding specificities of the highly polymorphic MHC have important implications for the design of vaccines, for the treatment of autoimmune diseases, and for personalized cancer immunotherapy. Evidence-based machine-learning approaches have been successfully used for the prediction of peptide binders and are currently being developed for the prediction of peptide immunogenicity. However, understanding and modeling the structural details of peptide/MHC binding is crucial for a better understanding of the molecular mechanisms triggering the immunological processes, estimating peptide/MHC affinity using universal physics-based approaches, and driving the design of novel peptide ligands. Unfortunately, due to the large diversity of MHC allotypes and possible peptides, the growing number of 3D structures of peptide/MHC (pMHC) complexes in the Protein Data Bank only covers a small fraction of the possibilities. Consequently, there is a growing need for rapid and efficient approaches to predict 3D structures of pMHC complexes. Here, we review the key characteristics of the 3D structure of pMHC complexes before listing databases and other sources of information on pMHC structures and MHC specificities. Finally, we discuss some of the most prominent pMHC docking software.

Peptides for T cell selection in the thymus

Major histocompatibility complex (MHC)-associated peptides generated and displayed by antigen-presenting cells in the thymus are essential for the generation of functional and self-tolerant T cells that protect our body from various pathogens. The peptides displayed by cortical thymic epithelial cells (cTECs) are generated by unique enzymatic machineries including the thymoproteasomes, and are involved in the positive selection of self-protective T cells. On the other hand, the peptides displayed by medullary thymic epithelial cells (mTECs) and thymic dendritic cells (DCs) are involved in further selection to establish self-tolerance in T cells. Although the biochemical nature of the peptide repertoire displayed in the thymus remains unclear, many studies have suggested a thymus-specific mechanism for the generation of MHC-associated peptides in the thymus. In this review, we summarize basic knowledge and recent advances in MHC-associated thymic peptides, focusing on the generation and function of thymoproteasome-dependent peptides specifically displayed by cTECs.

Crystal structure of suboptimal viral fragments of Epstein Barr Virus Rta peptide-HLA complex that stimulate CD8 T cell response

Peptides presented by Human leukocyte antigen (HLA) class-I molecules are generally 8-10 amino acids in length. However, the predominant pool of peptide fragments generated by proteasomes is less than 8 amino acids in length. Using the Epstein - Barr virus (EBV) Rta-epitope (ATIGTAMYK, residues 134-142) restricted by HLA-A*11:01 which generates a strong immunodominant response, we investigated the minimum length of a viral peptide that can constitute a viral epitope recognition by CD8 T cells. The results showed that Peripheral blood mononuclear cells (PBMCs) from healthy donors can be stimulated by a viral peptide fragment as short as 4-mer (AMYK), together with a 5-mer (ATIGT) to recapitulate the full length EBV Rta epitope. This was confirmed by generating crystals of the tetra-complex (2 peptides, HLA and β2-microglobulin). The solved crystal structure of HLA-A*11:01 in complex with these two short peptides revealed that they can bind in the same orientation similar to parental peptide (9-mer) and the free ends of two short peptides acquires a bulged conformation that is directed towards the T cell receptor. Our data shows that suboptimal length of 4-mer and 5-mer peptides can complement each other to form a stable peptide-MHC (pMHC) complex.

Pro-inflammatory Cytokines Alter the Immunopeptidome Landscape by Modulation of HLA-B Expression

Antigen presentation on HLA molecules is a major mechanism by which the immune system monitors self and non-self-recognition. Importantly, HLA-I presentation has gained much attention through its role in eliciting anti-tumor immunity. Several determinants controlling the peptides presented on HLA have been uncovered, mainly through the study of model substrates and large-scale immunopeptidome analyses. These determinants include the relative abundances of proteins in the cell, the stability or turnover rate of these proteins and the binding affinities of a given peptide to the HLA haplotypes found in a cell. However, the regulatory principles involved in selection and regulation of specific antigens in response to tumor pro-inflammatory signals remain largely unknown. Here, we chose to examine the effect that TNFα and IFNγ stimulation may exert on the immunopeptidome landscape of lung cancer cells. We show that the expression of many of the proteins involved in the class I antigen presentation pathway are changed by pro-inflammatory cytokines. Further, we could show that increased expression of the HLA-B allomorph drives a significant change in HLA-bound antigens, independently of the significant changes observed in the cellular proteome. Finally, we observed increased HLA-B levels in correlation with tumor infiltration across the TCGA lung cancer cohorts. Taken together, our results suggest that the immunopeptidome landscape should be examined in the context of anti-tumor immunity whereby signals in the microenvironment may be critical in shaping and modulating this important aspect of host-tumor interactions.

Dual non-contiguous peptide occupancy of HLA class I evoke antiviral human CD8 T cell response and form neo-epitopes with self-antigens

Host CD8 T cell response to viral infections involves recognition of 8-10-mer peptides presented by MHC-I molecules. However, proteasomes generate predominantly 2-7-mer peptides, but the role of these peptides is largely unknown. Here, we show that single short peptides of <8-mer from Latent Membrane Protein 2 (LMP2) of Epstein Barr Virus (EBV) can bind HLA-A*11:01 and stimulate CD8+ cells. Surprisingly, two peptide fragments between 4-7-mer derived from LMP2(340-349) were able to complement each other, forming combination epitopes that can stimulate specific CD8+ T cell responses. Moreover, peptides from self-antigens can complement non-self peptides within the HLA binding cleft, forming neoepitopes. Solved structures of a tetra-complex comprising two peptides, HLA and β2-microglobulin revealed the free terminals of the two peptides to adopt an upward conformation directed towards the T cell receptor. Our results demonstrate a previously unknown mix-and-match combination of dual peptide occupancy in HLA that can generate vast combinatorial complexity.

Association of HLA alleles (A, B, DRB1) and HIV-1 infection in the Han population of Hubei, China

The HIV susceptibility and resistance alleles in the HLA genes were determined by investigating the distribution characteristics of the HLA alleles (A, B, and DRB1) in HIV-infected individuals of the Han population in Hubei, and by comparing these alleles with HIV-negative individuals from the same area. A cohort of 424 HIV-1 infected individuals were chosen as study subjects, and 836 HIV-negative healthy subjects from the same area served as the control population. HLA-A, B, and DRB1 allele typing was performed using polymerase chain reaction-sequence-specific oligonucleotide probes (PCR-SSOP) and polymerase chain reaction-sequencing based typing (PCR-SBT) techniques. Arlequin ver3.0 was used to analyze the allele and haplotype frequencies of HLA-A, B, and DRB 1, whereas Epi Info 7 and SPSS18.0 was used to analyze the differences in the HLA alleles between the HIV-1 positive and HIV-1 negative groups. A*02:03, DRB1*01:01, and DRB1*15:01 alleles and their haplotypes as well as the HLA_Bw4-Bw6 hybrid showed a protective effect on HIV-1 infection. After adjusting for confounding factors such as age and sex, multivariate logistic regression analysis revealed that B*15:02G, DRB1*01:01, and DRB1*15:01 subtypes were the resistance genes of HIV-1 infection, while B*13:01 might increase susceptibility to HIV-1 infection. The correlation between A*02:06 and B*15:01G subtypes and HIV-1 susceptibility was independent of the age and sex of the host. This study demonstrated the influence of genetic factors in humans such as HLA polymorphism on individuals to resist HIV-1 infection. Association studies of HLA polymorphism, susceptibility/resistance to HIV-1 infection, and hosts' genetic background are of significant importance for research on HIV-1 pathogenesis and vaccine design.

Characterization of the Canine MHC Class I DLA-88*50101 Peptide Binding Motif as a Prerequisite for Canine T Cell Immunotherapy

There are limitations in pre-clinical settings using mice as a basis for clinical development in humans. In cancer, similarities exist between humans and dogs; thus, the dog patient can be a link in the transition from laboratory research on mouse models to clinical trials in humans. Knowledge of the peptides presented on MHC molecules is fundamental for the development of highly specific T cell-based immunotherapies. This information is available for human MHC molecules but is absent for the canine MHC. In the present study, we characterized the binding motif of dog leukocyte antigen (DLA) class I allele DLA-88*50101, using human C1R and K562 transfected cells expressing the DLA-88*50101 heavy chain. MHC class I immunoaffinity-purification revealed 3720 DLA-88*50101 derived peptides, which enabled the determination of major anchor positions. The characterized binding motif of DLA-88*50101 was similar to HLA-A*02:01. Peptide binding analyses on HLA-A*02:01 and DLA-88*50101 via flow cytometry showed weak binding of DLA-88*50101 derived peptides to HLA-A*02:01, and vice versa. Our results present for the first time a detailed peptide binding motif of the canine MHC class I allelic product DLA-88*50101. These data support the goal of establishing dogs as a suitable animal model for the evaluation and development of T cell-based cancer immunotherapies, benefiting both dog and human patients.

Selector function of MHC I molecules is determined by protein plasticity

The selection of peptides for presentation at the surface of most nucleated cells by major histocompatibility complex class I molecules (MHC I) is crucial to the immune response in vertebrates. However, the mechanisms of the rapid selection of high affinity peptides by MHC I from amongst thousands of mostly low affinity peptides are not well understood. We developed computational systems models encoding distinct mechanistic hypotheses for two molecules, HLA-B*44:02 (B*4402) and HLA-B*44:05 (B*4405), which differ by a single residue yet lie at opposite ends of the spectrum in their intrinsic ability to select high affinity peptides. We used in vivo biochemical data to infer that a conformational intermediate of MHC I is significant for peptide selection. We used molecular dynamics simulations to show that peptide selector function correlates with protein plasticity, and confirmed this experimentally by altering the plasticity of MHC I with a single point mutation, which altered in vivo selector function in a predictable way. Finally, we investigated the mechanisms by which the co-factor tapasin influences MHC I plasticity. We propose that tapasin modulates MHC I plasticity by dynamically coupling the peptide binding region and α3 domain of MHC I allosterically, resulting in enhanced peptide selector function.

Introduction of a point mutation into an HLA class I single-chain trimer induces enhancement of CTL priming and antitumor immunity

We previously discovered one particular HLA-A*02:01 mutant that enhanced peptide-specific cytotoxic T lymphocyte (CTL) recognition in vitro compared to wild-type HLA-A*02:01. This mutant contains a single amino acid substitution from histidine to leucine at position 74 (H74L) that is located in the peptide-binding groove. To investigate the effect of the H74L mutation on the in vivo CTL priming, we took advantage of the technology of the HLA class I single-chain trimer (SCT) in which three components involving a peptide, β2 microglobulin and the HLA class I heavy chain are joined together via flexible linkers. We generated recombinant adenovirus expressing SCT comprised influenza A matrix protein (FMP)-derived peptide, β2 microglobulin and the H74L heavy chain. HLA-A*02:01 transgenic mice were immunized with the adenovirus, and the induction of peptide-specific CTLs and antitumor immunity was investigated. It was clearly shown that the H74L mutation enabled the HLA-A*02:01 SCT molecule to dramatically enhance both in vivo priming of FMP-specific CTLs and protection against a lethal challenge of tumor cells expressing FMP. These data present the first evidence that a simple point mutation in the HLA class I heavy chain of SCT is beneficial for improving CTL-based immunotherapy and prophylaxis to control tumors.

HLA-A SNPs and amino acid variants are associated with nasopharyngeal carcinoma in Malaysian Chinese

Nasopharyngeal carcinoma (NPC) arises from the mucosal epithelium of the nasopharynx and is constantly associated with Epstein-Barr virus type 1 (EBV-1) infection. We carried out a genome-wide association study (GWAS) of 575,247 autosomal SNPs in 184 NPC patients and 236 healthy controls of Malaysian Chinese ethnicity. Potential association signals were replicated in a separate cohort of 260 NPC patients and 245 healthy controls. We confirmed the association of HLA-A to NPC with the strongest signal detected in rs3869062 (p = 1.73 × 10(-9)). HLA-A fine mapping revealed associations in the amino acid variants as well as its corresponding SNPs in the antigen peptide binding groove (p(HLA-A-aa-site-99) = 3.79 × 10(-8), p(rs1136697) = 3.79 × 10(-8)) and T-cell receptor binding site (p(HLA-A-aa-site-145) = 1.41 × 10(-4), p(rs1059520) = 1.41 × 10(-4)) of the HLA-A. We also detected strong association signals in the 5'-UTR region with predicted active promoter states (p(rs41545520) = 7.91 × 10(-8)). SNP rs41545520 is a potential binding site for repressor ATF3, with increased binding affinity for rs41545520-G correlated with reduced HLA-A expression. Multivariate logistic regression diminished the effects of HLA-A amino acid variants and SNPs, indicating a correlation with the effects of HLA-A*11:01, and to a lesser extent HLA-A*02:07. We report the strong genetic influence of HLA-A on NPC susceptibility in the Malaysian Chinese.

TCR-independent killing of B cell malignancies by anti-third-party CTLs: the critical role of MHC-CD8 engagement

We previously demonstrated that anti-third-party CTLs (stimulated under IL-2 deprivation against cells with an MHC class I [MHC-I] background different from that of the host and the donor) are depleted of graft-versus-host reactivity and can eradicate B cell chronic lymphocytic leukemia cells in vitro or in an HU/SCID mouse model. We demonstrated in the current study that human allogeneic or autologous anti-third-party CTLs can also efficiently eradicate primary non-Hodgkin B cell lymphoma by inducing slow apoptosis of the pathological cells. Using MHC-I mutant cell line as target cells, which are unrecognizable by the CTL TCR, we demonstrated directly that this killing is TCR independent. Strikingly, this unique TCR-independent killing is induced through lymphoma MHC-I engagement. We further showed that this killing mechanism begins with durable conjugate formation between the CTLs and the tumor cells, through rapid binding of tumor ICAM-1 to the CTL LFA-1 molecule. This conjugation is followed by a slower second step of MHC-I-dependent apoptosis, requiring the binding of the MHC-I α2/3 C region on tumor cells to the CTL CD8 molecule for killing to ensue. By comparing CTL-mediated killing of Daudi lymphoma cells (lacking surface MHC-I expression) to Daudi cells with reconstituted surface MHC-I, we demonstrated directly for the first time to our knowledge, in vitro and in vivo, a novel role for MHC-I in the induction of lymphoma cell apoptosis by CTLs. Additionally, by using different knockout and transgenic strains, we further showed that mouse anti-third-party CTLs also kill lymphoma cells using similar unique TCR-independence mechanism as human CTLs, while sparing normal naive B cells.

Diverse peptide presentation of rhesus macaque major histocompatibility complex class I Mamu-A 02 revealed by two peptide complex structures and insights into immune escape of simian immunodeficiency virus

Major histocompatibility complex class I (MHC I)-restricted CD8(+) T-cell responses play a pivotal role in anti-human immunodeficiency virus (HIV) immunity and the control of viremia. The rhesus macaque is an important animal model for HIV-related research. Among the MHC I alleles of the rhesus macaque, Mamu-A 02 is prevalent, presenting in ≥20% of macaques. In this study, we determined the crystal structure of Mamu-A 02, the second structure-determined MHC I from the rhesus macaque after Mamu-A 01. The peptide presentation characteristics of Mamu-A 02 are exhibited in complex structures with two typical Mamu-A 02-restricted CD8(+) T-cell epitopes, YY9 (Nef159 to -167; YTSGPGIRY) and GY9 (Gag71 to -79; GSENLKSLY), derived from simian immunodeficiency virus (SIV). These two peptides utilize similar primary anchor residues (Ser or Thr) at position 2 and Tyr at position 9. However, the central region of YY9 is different from that of GY9, a difference that may correlate with the immunogenic variance of these peptides. Further analysis indicated that the distinct conformations of these two peptides are modulated by four flexible residues in the Mamu-A 02 peptide-binding groove. The rare combination of these four residues in Mamu-A 02 leads to a variant presentation for peptides with different residues in their central regions. Additionally, in the two structures of the Mamu-A 02 complex, we compared the binding of rhesus and human β(2) microglobulin (β(2)m) to Mamu-A 02. We found that the peptide presentation of Mamu-A 02 is not affected by the interspecies interaction with human β(2)m. Our work broadens the understanding of CD8(+) T-cell-specific immunity against SIV in the rhesus macaque.

Identification and functional perspective of a novel HLA-A allele: A*0279

A novel HLA-A allele, HLA-A*0279, was identified using PCR-SSP and PCR-SBT methods. It is inheritable. HLA-A*0279 differs from HLA-A*020601 by a single nucleotide at position 497 in exon 3, leading to an amino acid change from Threonine to Isoleucine at the alpha2 helix of HLA molecule. To investigate whether the altered amino acid residue could affect its peptide-binding repertoire, we compared the predicted crystal structure of HLA-A*020601 and HLA-A*0279 by Swiss-PdbViewer software analysis. We found that the crystal structure of the two molecules is very similar except for a difference in the number of hydrogen bonds they can possibly form, which in turn could affect their structural stability. To test whether HLA-A*0279 has the ability to cross-present A*0201 - restricted peptides to T cells, the full lenght cDNA of HLA-A*0201, -A* 020601 and -A*0279 were respectively transfected into COS-7 cells, which were then used as targets in IFN-gamma release Elispot assay. A*2079 was found to be able to present A*0201- restricted peptides to and induce the response of CTL, thus it can be classified as member of the HLA-A2 functional supertype family. This finding would benefit the design of peptide vaccines to be applied in broader populations.

Vaccination of macaques with SIV immunogens delivered by Venezuelan equine encephalitis virus replicon particle vectors followed by a mucosal challenge with SIVsmE660

VEE replicon particles (VRP), non-propagating vaccine vectors derived from Venezuelan equine encephalitis virus (VEE), were engineered to express immunogens from the cloned isolate SIVsmH-4, combined in a vaccine cocktail and inoculated subcutaneously to immunize rhesus macaques. The virulent, uncloned challenge stock, SIVsmE660, represented a type of heterologous challenge and the intrarectal challenge modeled infection across a mucosal surface. Prechallenge neutralizing antibodies against SIVsmH-4 were induced in all vaccinates, and a prechallenge cellular immune response could be detected in one of six. Post-challenge, virus loads were reduced at the peak, at set point and at termination (41 weeks post-challenge), although these differences did not reach statistical significance. Significantly elevated levels of CD4+ T cells were observed post-challenge. A strong correlation was noted between a net increase in CD4+ T cell count and lowered virus load at set point.

Unraveling a hotspot for TCR recognition on HLA-A2: evidence against the existence of peptide-independent TCR binding determinants

T cell receptor (TCR) recognition of peptide takes place in the context of the major histocompatibility complex (MHC) molecule, which accounts for approximately two-thirds of the peptide/MHC buried surface. Using the class I MHC HLA-A2 and a large panel of mutants, we have previously shown that surface mutations that disrupt TCR recognition vary with the identity of the peptide. The single exception is Lys66 on the HLA-A2 alpha1 helix, which when mutated to alanine disrupts recognition for 93% of over 250 different T cell clones or lines, independent of which peptide is bound. Thus, Lys66 could serve as a peptide-independent TCR binding determinant. Here, we have examined the role of Lys66 in TCR recognition of HLA-A2 in detail. The structure of a peptide/HLA-A2 molecule with the K66A mutation indicates that although the mutation induces no major structural changes, it results in the exposure of a negatively charged glutamate (Glu63) underneath Lys66. Concurrent replacement of Glu63 with glutamine restores TCR binding and function for T cells specific for five different peptides presented by HLA-A2. Thus, the positive charge on Lys66 does not serve to guide all TCRs onto the HLA-A2 molecule in a manner required for productive signaling. Furthermore, electrostatic calculations indicate that Lys66 does not contribute to the stability of two TCR-peptide/HLA-A2 complexes. Our findings are consistent with the notion that each TCR arrives at a unique solution of how to bind a peptide/MHC, most strongly influenced by the chemical and structural features of the bound peptide. This would not rule out an intrinsic affinity of TCRs for MHC molecules achieved through multiple weak interactions, but for HLA-A2 the collective mutational data place limits on the role of any single MHC amino acid side-chain in driving TCR binding in a peptide-independent fashion.

Allelic variation of MHC structure alters peptide ligands to induce atypical partial agonistic CD8+ T cell function

T cell receptors recognize small changes in peptide ligands leading to different T cell responses. Here, we analyzed a panel of HLA-A2-Tax11-19 reactive T cell clones to examine how small allelic variations of MHC molecules could alter the functional outcome of antigen recognition. Similar to the effects induced by antigenic altered peptide ligands, weak or partial agonistic T cell functions were identified in individual T cell clones with the recognition of MHC-altered peptide ligands (MAPLs). Interestingly, one subtype of HLA-A2 molecules induced an unusual type of partial agonistic function; proliferation without cytotoxicity. Modeling of crystallographic data indicated that polymorphic amino acids in the HLA-A2 peptide binding groove, especially the D-pocket, were responsible for this partial agonism. Reciprocal mutations of the Tax peptide side chain engaging the D-pocket indeed restored the agonist functions of the MHC-peptide complex. Whereas early intracellular signaling events were not efficiently induced by these MAPLs, phosphorylated c-Jun slowly accumulated with sustained long-term expression. These data indicate that MAPLs can induce atypical partial agonistic T cell function through structural and biochemical mechanisms similar to altered peptide ligands.

Downregulation of the proteasome subunits, transporter, and antigen presentation in hepatocellular carcinoma, and their restoration by interferon-gamma

BACKGROUND

In our previous study, expressions of human histocompatibility leukocyte antigens class I molecules (HLA-I) and the transporter associated with antigen processing (TAP) 1/2 genes were investigated in seven hepatocellular carcinoma (HCC) cell lines. Two cell lines, Hep-3B and HuH-7, showed a reduced level of TAP, which might cause the low surface expression of HLA-I. In order to understand the downregulation mechanism of antigen presentation in tumors, the two cell lines were further investigated.

METHODS

Expressions of HLA-I and antigen presentation-related genes were analyzed by flow cytometry and polymerase chain reaction, respectively. Antigen presentation was tested in 51Cr-release assays.

RESULTS

Flow cytometric analyses revealed low surface expression of HLA-I on Hep-3B and HuH-7 cells. Introduction of HLA-A2 gene did not result in a high surface expression of HLA-A2. This suggested the downregulation of HLA-I expression might be related to defects in the antigen presentation machinery. We then examined expression levels of various antigen presentation-related genes. Hep-3B and HuH-7 demonstrated low expression of the low-molecular-weight protein (LMP) 2, LMP7, TAP1, and HLA-I heavy-chain transcripts. The downregulation of these genes was dissolved by treatment with gamma-interferon. Furthermore, allo-specific cytotoxic T lymphocyte (CTL) lines failed to recognize Hep-3B and HuH-7 cells, while they killed IFN-gamma-treated Hep-3B and HuH-7 cells.

CONCLUSIONS

Our results suggest that defects in the antigen presentation-related molecules might cause downregulation of HLA-I expression, antigen presentation, and subsequently, escape from specific CTL killing. The downregulation could be restored by IFN-gamma treatment, suggesting the potential use of IFN-gamma for therapeutic purposes.

Development of a rapid in vitro protein refolding assay which discriminates between peptide-bound and peptide-free forms of recombinant porcine major histocompatibility class I complex (SLA-I)

The extracellular domains of swine leukocyte antigen class I (SLA-I, major histocompatibility complex protein class I) were cloned and sequenced for two haplotypes (H4 and H7) which do not share any alleles based on serological typing, and which are the most important in Danish farmed pigs. The extracellular domain of SLA-I was connected to porcine beta2 microglobulin by glycine-rich linkers. The engineered single-chain proteins, consisting of fused SLA-I and beta2 microglobulin, were overexpressed as inclusion bodies in Escherichia coli. Also, variants were made of the single-chain proteins, by linking them through glycine-rich linkers to peptides representing T-cell epitopes from classical swine fever virus (CSFV) and foot-and-mouth disease virus (FMDV). An in vitro refold assay was developed, using a monoclonal anti-SLA antibody (PT85A) to gauge refolding. The single best-defined, SLA-I restricted porcine CD8(+) T-cell epitope currently known is a 9-residue peptide from the polyprotein of CSFV (J. Gen. Virol. 76 (1995) 3039). Based on results with the CSFV epitope and two porcine haplotypes (H4 and H7), the in vitro refold assay appeared able to discriminate between peptide-free and peptide-occupied forms of SLA-I. It remains to be seen whether the rapid and technically very simple in vitro refold assay described here will prove generally applicable for the screening of virus-derived peptides for SLA-I binding.

High expression of HLA-A2 on an oral squamous cell carcinoma with down-regulated transporter for antigen presentation

Using the reverse transcription PCR, we evaluated expression levels of various antigen presentation-related genes, including LMP2, LMP7, MECL-1, PA28alpha, PA28beta, TAP1, TAP2, and tapasin, in two oral squamous cell carcinoma cell lines, HSC5 and HSC7. Expression levels of LMP2, MECL-1, TAP1, and TAP2 transcripts are reduced in both cell lines in comparison with a normal epithelial cell line. Further, HSC5 and HSC7 show diminished expression of LMP7/tapasin, and PA28alpha/beta, respectively. Surface expression of HLA-B alleles is down-regulated in both lines presumably due to low expression of TAP1/2. However, HLA-A2 surface expression is not significantly down-regulated in HSC5 cells, suggesting an involvement of signal-sequence derived peptides in the TAP-independent pathway. The current study would contribute to our understanding of significance of abnormalities in the antigen presentation machinery of oral squamous cell carcinoma, and provide meaningful information in the design of CTL-based tumor vaccines by intra-cellular delivery system.

Vaccination of macaques against pathogenic simian immunodeficiency virus with Venezuelan equine encephalitis virus replicon particles

Vaccine vectors derived from Venezuelan equine encephalitis virus (VEE) that expressed simian immunodeficiency virus (SIV) immunogens were tested in rhesus macaques as part of the effort to design a safe and effective vaccine for human immunodeficiency virus. Immunization with VEE replicon particles induced both humoral and cellular immune responses. Four of four vaccinated animals were protected against disease for at least 16 months following intravenous challenge with a pathogenic SIV swarm, while two of four controls required euthanasia at 10 and 11 weeks. Vaccination reduced the mean peak viral load 100-fold. The plasma viral load was reduced to below the limit of detection (1,500 genome copies/ml) in one vaccinated animal between 6 and 16 weeks postchallenge and in another from week 6 through the last sampling time (40 weeks postchallenge). The extent of reduction in challenge virus replication was directly correlated with the strength of the immune response induced by the vectors, which suggests that vaccination was effective.

Venezuelan equine encephalitis virus vectors expressing HIV-1 proteins: vector design strategies for improved vaccine efficacy

A live virus vaccine vector has been constructed from a molecularly cloned attenuated strain of Venezuelan equine encephalitis virus (VEE). High levels of foreign protein expression are regulated by an additional copy of the 26 S viral subgenomic RNA promoter. The position of this additional promoter and foreign gene in the VEE genome was predicted to have a major influence on expression level of the heterologous protein. Two sites in the genome were tested to determine the optimal site for expression of the matrix/capsid (MA/CA) coding region of human immunodeficiency virus (HIV-1). One vector contained the additional promoter and the MA/CA genes immediately downstream of the VEE E1 gene at the 3' end of the genome. In the second vector, the additional promoter was introduced immediately upstream from the authentic 26 S subgenomic promoter. Significantly higher levels of MA/CA were expressed from the downstream vector compared to the upstream vector. However, the stability of expression for both vectors was similar following passage in baby hamster kidney cells (BHK) cells. In BALB/c mice, the two vectors elicited similar levels of cellular immune responses to MA/CA as determined by bulk cytotoxic T-lymphocyte assays and precursor frequency analysis, but the humoral response induced by the downstream vector was significantly stronger. At 11 months post boosting with the downstream vector, serum antibody levels against HIV MA/CA were undiminished, and MA/CA specific CTLp were detectable in all mice tested. These findings suggest that VEE vectors can be optimized to elicit strong, balanced and long-lived immune responses to foreign viral proteins.

The metalloproteinase-mediated pathway is essential for generation of soluble HLA class I proteins by activated cells in vitro: proposed mechanism for soluble HLA release in transplant rejection

We and others have found donor-derived soluble beta2m-associated HLA class I proteins (sHLA/beta2m) in the serum of allograft recipients with acute and chronic rejection. Whether appearance of sHLA/beta2m and upregulated expression of donor cell-bound HLA/beta2m during allograft rejection are related events is unknown. Activation-induced upregulation of in vitro HLA/beta2m expression correlates with the surface expression of another form of HLA class I, namely beta2m-free HLA heavy chains (beta2m-free HC). We have shown that beta2m-free HC, but not beta2m-associated HC, are then cleaved by a specific membrane-bound metalloproteinase and released into supernatants as soluble 36 kDa proteins. We show now that activated peripheral blood lymphocytes produce predominantly the 36 kDa form of sHLA proteins which is present in supernatants as both beta2m-free HC and sHLA/beta2m. Importantly, the metalloprotease inhibitor BB-94 blocked not only the release of soluble beta2m-free HC, but also the appearance of sHLA/beta2m in cell supernatants. Low levels of 36 kDa beta2m-free HC were also present in human plasma of healthy donors. These data suggest an important role for the HLA class I-specific metalloproteinase in vivo in healthy individuals and during allograft rejection in the generation of soluble beta2m-free and beta2m-associated HLA proteins.

Automated nucleotide sequencing reveals substantial disparity between the HLA-A2 genes of bone marrow transplant recipients and donors

HLA disparity is associated with immunological complications after bone marrow transplant and it has been demonstrated that a single amino acid substitution can dramatically alter the function or allorecognition of an HLA molecule. Current serological methods for typing Class I HLA do not distinguish between most HLA-A2 variants which can differ by 1-8 amino acid residues. HLA-A2 disparity between bone marrow transplant patients and donors was investigated using automated nucleotide sequencing of the entire coding region of HLA-A2 genes. A total of 122 HLA-A2 alleles were sequenced from 47 patient-donor pairs (94 individuals). HLA-A2 disparity was observed in 10 of 47 pairs (21.3%) and consisted of HLA-A*0201 mismatched with 0202 (n = 2), 0205 (n = 3), 0206 (n = 3), 0217 (n = 1) or 0221 (n = 1). Four of 6 (66.7%) non-Caucasian or mixed race pairs were HLA-A2 disparate, while 6 of 36 (16.7%). Caucasian pairs were HLA-A2 disparate (p = 0.008). Among all individuals HLA-A*0201 was the most frequently observed allele (90.0%) while 0202 (1.6%), 0205 (2.5%), 0206 (4.1%), 0217 (0.8%) and 0221 (0.8%) were also observed. This study illustrates the diversity of HLA-A2 in non-Caucasian individuals and suggests that HLA-A2 subtyping for applications such as bone marrow transplantation, especially in non-Caucasian or mixed-race donor-recipient pairs, may be important.

Differential generation of class I H-2D- versus H-2K-restricted cytotoxicity against a demyelinating virus following central nervous system infection

Despite the fact that both H-2K and D molecules are up-regulated in the central nervous system (CNS) following Theiler's murine encephalomyelitis virus (TMEV) infection, resistance in this virus model of multiple sclerosis maps exclusively to D. To address this paradox, we examined the ability of the K and D molecules to present viral antigens to cytotoxic T lymphocytes (CTL). Whereas no virus-specific CTL were detected in the CNS of susceptible B10.Q and B10.S mice 7 days post-infection, D-restricted CTL were identified readily in the CNS of resistant B10 animals. There was no evidence of K-restricted CTL in the CNS of B10 mice at day 7 post-infection. The presence of both K- and D-restricted virus-specific CTL in the spleen of immunized B10 mice demonstrates that the exclusive use of D molecules by CTL in the CNS of mice 7 days post-infection is not due to the inability of the K molecules to present viral peptides to lymphocytes. We conclude that the prominent role of the D locus in determining resistance or susceptibility to TMEV-induced demyelination is determined by factors governing the regulation of the immune response, and not by the presence or absence of CTL precursors capable of recognizing viral peptides presented by the K and D antigen-presenting molecules, or by differences in the ability of the K and D molecules to present viral peptides.

Humoral, mucosal, and cellular immunity in response to a human immunodeficiency virus type 1 immunogen expressed by a Venezuelan equine encephalitis virus vaccine vector

A molecularly cloned attenuated strain of Venezuelan equine encephalitis virus (VEE) has been genetically configured as a replication-competent vaccine vector for the expression of heterologous viral proteins (N. L. Davis, K. W. Brown, and R. E. Johnston, J. Virol. 70:3781-3787, 1996). The matrix/capsid (MA/CA) coding domain of human immunodeficiency virus type 1 (HIV-1) was cloned into the VEE vector to determine the ability of a VEE vector to stimulate an anti-HIV immune response in mice. The VEE-MA/CA vector replicated rapidly in the cytoplasm of baby hamster kidney (BHK) cells and expressed large quantities of antigenically identifiable MA/CA protein. When injected subcutaneously into BALB/c mice, the vector invaded and replicated in the draining lymphoid tissues, expressing HIV-1 MA/CA at a site of potent immune activity. Anti-MA/CA immunoglobulin G (IgG) and IgA antibodies were present in serum of all immunized mice, and titers increased after a second booster inoculation. IgA antibodies specific for MA/CA were detected in vaginal washes of mice that received two subcutaneous immunizations. Cytotoxic T-lymphocyte responses specific for MA/CA were detected following immunization with the MA/CA-expressing VEE vector. These findings demonstrate the ability of a VEE-based vaccine vector system to stimulate a comprehensive humoral and cellular immune response. The multifaceted nature of this response makes VEE an attractive vaccine for immunization against virus infections such as HIV-1, for which the correlates of protective immunity remain unclear, but may include multiple components of the immune system.

Class I HLA-restricted cytotoxic T lymphocyte responses against malaria--elucidation on the basis of HLA peptide binding motifs

In animal models, CD8+ T cells are a critical effector mechanism in the protective immunity against malaria. Conventional approaches to the development of many vaccines, including those against malaria, have however proved inadequate. In particular, an alternative approach is needed for the development of vaccines designed to induce a cellular immune response mediated by CD8+ T cells. Advances in the field of molecular immunology during the past decade have provided an insight into the presentation of peptides by MHC class I molecules and their recognition by CD8+ T cells. These studies have provided a conceptual basis for the development of efficacious parasitic and viral vaccines. By a combination of immunochemical and cellular immunologic analyses based on specific peptide binding motifs, a subunit malaria vaccine that includes CD8+ T cell epitopes restricted by the most common class I HLA alleles, including HLA-A2, can now be constructed.

Binding of viral antigens to major histocompatibility complex class I H-2Db molecules is controlled by dominant negative elements at peptide non-anchor residues. Implications for peptide selection and presentation

Binding of viral antigens to major histocompatibility complex (MHC) class I molecules is a critical step in the activation process of CD8(+) cytotoxic T lymphocytes. In this study, we investigated the impact of structural factors at non-anchor residues in peptide-MHC interaction using the model of lymphocytic choriomeningitis virus (LCMV) infection of its natural host, the mouse. Altering viral genes by making reassortants, recombinants, and using synthetic peptides, CD8(+) cytotoxic T lymphocytes were shown to recognize only three H-2Db-restricted epitopes, GP amino acids 33-41/43, GP 276-286, and NP 396-404. However, LCMV NP and GP proteins contain 31 other peptides bearing the H-2Db motif. These 34 LCMV peptides and 11 other known H2-Db-restricted peptides were synthesized and examined for MHC binding properties. Despite the presence of the H-2Db binding motif, the majority of LCMV peptides showed weak or no affinity for H-2Db. We observed that dominant negative structural elements located at non-anchor positions played a crucial role in peptide-MHC interaction. By comparative sequence analysis of strong versus non-binders and using molecular modeling, we delineated these negative elements and evaluated their impact on peptide-MHC interaction. Our findings were validated by showing that a single mutation of a favorable non-anchor residue in the sequence of known viral epitopes for a negative element resulted in dramatic reduction of antigen presentation properties, while conversely, substitution of one negative for a positive element in the sequence of a non-binder conferred to the peptide an ability to now bind to MHC molecules.

Effects of HIV-1 Tat on expression of HLA class I molecules

Tat protein of HIV-1 is a potent transactivator of transcription and essential for HIV-1 replication. In addition, Tat has been proposed to possess immunosuppressive functions, suggesting that Tat may play a direct role in the immune dysfunction associated with AIDS. Recently, it has been reported that Tat represses activity of a major histocompatibility complex (MHC) class I gene promoter. Because HIV infection downmodulates expression of class I molecules, this data strongly suggests that Tat downregulates class I expression and leads to loss of CTL activity. Here, we report effects of Tat on class I expression using a human cell line, T0, expressing Tat (TO-Tat). Northern blot analysis shows that levels of MHC class I transcripts are normal in T0-Tat. Flow cytometry analyses indicate that expression of HLA class I molecules is not substantially downregulated to any great extent by Tat in T0-Tat. Further, pulse-chase experiments followed by Endoglycosidase-H treatment show that the rate of maturation and processing of class I molecules in T0-Tat is indistinguishable from that in the original cell line, T0. Taken together, these data suggest that Tat expression does not necessarily result in downregulation of class I expression.

HLA-A2 subtypes are functionally distinct in peptide binding and presentation

Nearly half of HLA-A2-positive individuals in African populations have a subtype of HLA-A2 other than the A*0201 allele. We have isolated the common African HLA-A2 subtype genes from Epstein-Barr virus-transformed B cell lines and have established stable class I reduced transfectants expressing these alleles. We have studied the peptide binding and presentation properties of A*0201, A*0202, A*0205, A*0214, and A*6901 by a combination of approaches: assaying direct binding of labeled synthetic peptides, studying the ability of antigen-specific cytotoxic T lymphocytes to recognize peptide-pulsed cells, and sequencing peptide pools and individual ligands eluted from cells. We find that A*0201-restricted peptides can also bind to A*0202 but do not bind strongly to the other alleles in this study. We show that some cytotoxic T lymphocytes can recognize all subtypes capable of binding an antigenic peptide, whereas others are subtype specific. Sequencing of eluted peptides reveals that A*0202 has a similar peptide motif to A*0201, but that A*0205, A*0214, and A*6901 have different motifs. These data strongly support a model in which residue 9 (Phe or Tyr) of the A2/A68/A69 molecules is a critical factor in determining the specificity of the B pocket of the major histocompatibility complex and the position 2 anchor residue of associated peptides. We conclude that a single-amino acid difference in the major histocompatibility complex can be sufficient to cause a dramatic change in the nature of bound peptides, implying that individuals with closely related HLA subtypes may present very different repertoires of antigenic peptides to T cells in an immune response. It is likely to be a general phenomenon that very similar class I subtypes will behave as functionally distinct HLA allotypes.

Amino-terminal alteration of the HLA-A*0201-restricted human immunodeficiency virus pol peptide increases complex stability and in vitro immunogenicity

Initial studies suggested that major histocompatibility complex class I-restricted viral epitopes could be predicted by the presence of particular residues termed anchors. However, recent studies showed that nonanchor positions of the epitopes are also significant for class I binding and recognition by cytotoxic T lymphocytes (CTLs). We investigated if changing nonanchor amino acids could increase class I affinity, complex stability, and T-cell recognition of a natural viral epitope. This concept was tested by using the HLA-A 0201-restricted human immunodeficiency virus type 1 epitope from reverse transcriptase (pol). Position 1 (P1) amino acid substitutions were emphasized because P1 alterations may not alter the T-cell receptor interaction. The peptide with the P1 substitution of tyrosine for isoleucine (I1Y) showed a binding affinity for HLA-A 0201 similar to that of the wild-type pol peptide in a cell lysate assembly assay. Surprisingly, I1Y significantly increased the HLA-A 0201-peptide complex stability at the cell surface. I1Y sensitized HLA-A 0201-expressing target cells for wild-type pol-specific CTL lysis as well as wild-type pol. Peripheral blood lymphocytes from three HLA-A2 HIV-seropositive individuals were stimulated in vitro with I1Y and wild-type pol. I1Y stimulated a higher wild-type pol-specific CTL response than wild-type pol in all three donors. Thus, I1Y may be an "improved" epitope for use as a CTL-based human immunodeficiency virus vaccine component. The design of improved epitopes has important ramifications for prophylaxis and therapeutic vaccine development.

The relative distribution of B35 alleles and their IEF isotypes in a HLA-B35-positive population

The HLA-B35 serotype represents a group of antigens detectable by IEF, cytotoxic T cells, and by sequencing analysis. Four isotypes and eight alleles have been thus far reported. We have determined the relative frequencies of these B35 subtypes in a group of 203 unrelated people. Dot blot hybridization of PCR amplified products was performed using 23 sequence-specific oligo probes designed based on the EMBL HLA class I sequence database. The amplification was achieved by a pair of group-specific primers, producing approximately 600 bp fragments. By hybridization pattern analysis, we found that four alleles represent over 95% of the B35+ population, with relative frequency of 48.2% for B*3501, 23.7% for B*3502, 15.2% for B*3503, and 8.0% for B*3508. We also identified 3 individuals with B*3504 and one with B*3505, and seven samples with new patterns. B*3501 and B*3503 exactly correlated with the most common isotype B35.3, B*3502 and B*3504 with B35.2, B*3508 may be the B35.1 IEF isotype. The B*3505 was identified from an individual with B35 IEF variant form. Our study shows that the B35 antigen has a wide distribution of alleles, and that many more B35-related alleles may yet to be uncovered.

Functional analysis of avian class I (BFIV) glycoproteins by epitope tagging and mutagenesis in vitro

Similarities between the physical structures of avian and mammalian major histocompatibility complex (MHC) class I glycoproteins have been proposed based on comparative alignment of their amino acid sequences. To investigate the physical structure of the chicken class I glycoprotein, we cloned the cDNA representing the BFIV locus of the B21 haplotype. A unique, chimeric class I glycoprotein was constructed by incorporating an epitope tag (FLAG) at the N terminus. Monoclonal antibodies to the FLAG epitope served to monitor cell-surface expression for functional analysis of the BFIV21 class I glycoprotein. The chimeric class I glycoprotein was expressed in target cells using an avian leukosis virus (ALV)-derived retrovirus vector (RCASBP). The presence of the FLAG epitope did not interfere with either alloantibody recognition or cytotoxic T lymphocyte interaction. Functional analysis employing site-directed mutagenesis identified BF amino acid residues forming serologic epitopes as well as residues important in antigen presentation to ALV-induced cytotoxic T lymphocytes. BF residues 78 and 81, corresponding to HLA 79 and 82, form an antibody epitope with a slight effect on ALV antigen presentation, consistent with their predicted orientation based on the HLA-A2 crystal structure. Alignment of the BFIV21 sequence with previously published BFIV sequences revealed polymorphisms at position 34 (HLA 34), a monomorphic residues in HLA and H-2. Residue 34 is located in pocket B and is predicted to contact the main-chain carbon of peptides bound in HLA-A2. A site-directed substitution in BFIV residue 34 dramatically alters ALV antigen presentation by the BFIV21 class I glycoprotein. These data indicate that the physical molecular structure of the chicken MHC class I glycoprotein is similar to HLA.

H-2M3a violates the paradigm for major histocompatibility complex class I peptide binding

The major histocompatibility (MHC) class I-b molecule H-2M3a binds and presents N-formylated peptides to cytotoxic T lymphocytes. This requirement potentially places severe constraints on the number of peptides that M3a can present to the immune system. Consistent with this idea, the M3a-Ld MHC class I chimera is expressed at very low levels on the cell surface, but can be induced significantly by the addition of specific peptides at 27 degrees C. Using this assay, we show that M3a binds many very short N-formyl peptides, including N-formyl chemotactic peptides and canonical octapeptides. This observation is in sharp contrast to the paradigmatic size range of peptides of 8-10 amino acids binding to most class I-a molecules and the class I-b molecule Qa-2. Stabilization by fMLF-benzyl amide could be detected at peptide concentrations as low as 100 nM. While N-formyl peptides as short as two amino acids in length stabilized expression of M3a-Ld, increasing the length of these peptides added to the stability of peptide-MHC complexes as determined by 27-37 degrees C temperature shift experiments. We propose that relaxation of the length rule may represent a compensatory adaptation to maximize the number of peptides that can be presented by H-2M3a.

Mutations inside but not outside the peptide binding cleft of the H-2 Ld molecule affect CTL recognition and binding of the nucleoprotein peptide from the lymphocytic choriomeningitis virus

In order to investigate the role of residues inside and outside the peptide binding cleft of the Ld molecule in peptide presentation to cytotoxic T lymphocytes (CTL), we constructed a series of point mutations in the Ld gene. We determined the effects of the mutations in the Ld molecule on the binding and recognition of an Ld-restricted CTL epitope derived from the nucleoprotein (NP) of the lymphocytic choriomeningitis virus (LCMV). Each of the mutations within the Ld peptide binding cleft resulted in a complete loss of CTL recognition. Addition of the LCMV NP peptide to cells expressing these mutants did not increase surface Ld expression, suggesting that the mutations altered peptide binding. Mutations involving pockets D and E within the cleft affected LCMV peptide binding and recognition as drastically as those in pocket B, which was predicted to interact with a main anchor residue of the peptide. In striking contrast, the mutations located outside the cleft did not change either recognition or binding. These results demonstrate that the Ld residues in the peptide binding cleft are the main determinants dictating LCMV NP peptide binding, and that the residues in each of the pockets within the cleft play a role in this interaction. Surprisingly, one mutation outside the peptide binding cleft, T92S, abrogated CTL lysis of target cells treated with the LCMV NP peptide, but not virus-infected cells. These data show that this mutation selectively altered the presentation of the LCMV NP peptide introduced to the cell exogenously, but not endogenously. This implies that the pathway by which peptides associate with class I molecules within the cell differs from that of exogenous peptide binding.

Pocket mutations of HLA-B27 show that anchor residues act cumulatively to stabilize peptide binding

Major histocompatibility complex (MHC) class I molecules bind endogenously synthesized peptides for presentation to cytotoxic T-cells. The human class I molecule HLA-B27 consists of a trimolecular complex containing the HLA-B27 heavy chain, a peptide that is usually nine amino acid residues (aa) long, and beta 2-microglobulin (beta 2m). The key interactions for peptide selectivity are between Glu-45, which forms a salt bridge with the Arg at P2 of the peptide, and Asp-116 which favors the binding of peptides containing a Lys or Arg at P9. The t1/2 of dissociation of [125I]beta 2m was measured for peptide-specific HLA-B27 wild-type (wt) and mutant complexes. HLA-B27 wt and HLA-B27 D116F formed relatively stable complexes, with a t1/2 of dissociation on the scale of hours, with appropriate peptides that contained Arg at P2, whereas HLA-B27 E45T required a Gln at P2. Similarly, kinetically stable D116F complexes were formed only with peptides that contained a Leu or Val at P9 instead of Arg or Lys. The [125I]beta 2m dissociation rate data were fit to a set of equations in order to calculate relative binding coefficients for each anchor residue at P2 and P9. The P2 coefficients were sensitive to the E45T mutation but not the D116F mutation, whereas the P9 coefficients were sensitive only to the D116F mutation. Thus, drastic structural changes in one subsite do not affect the other subsite, indicating that the dominant anchor residues at P2 and P9 independently contribute to stabilizing the class I/peptide complex.

Allele-specific B pocket transplant in class I major histocompatibility complex protein changes requirement for anchor residue at P2 of peptide

To investigate the role of an anchoring pocket in allele-specific peptide presentation by a major histocompatibility complex class I molecule, we "transplanted" a B pocket from HLA-A*0201 into HLA-B*2705 by site-directed mutagenesis. The resulting protein, designated B27.A2B, binds a different set of endogenous peptides than B*2705 as evidenced by complete loss of allorecognition as well as restored expression in the antigen processing-defective mutant cell line T2. B27.A2B also fails to present an HLA-B27-restricted influenza virus peptide [nucleoprotein (383-391)] to cytotoxic T lymphocytes (CTLs). However, substitution of leucine, the predominant P2 anchor residue in A*0201-restricted peptides, for arginine, the P2 anchor in nucleoprotein-(383-391) and other B*2705-restricted peptides, restores recognition of B27.A2B by the same B*2705-restricted peptide-specific CTLs. These results demonstrate that a dominant polymorphic pocket in a class I molecule, through interaction with the anchor residue of an antigenic peptide, can distinguish among peptides differing by only a single amino acid and thus determine the allelic specificity of peptide presentation.