Structures of Three HIV-1 HLA-B*5703-Peptide Complexes and Identification of Related HLAs Potentially Associated with Long-Term Nonprogression

Guillaume B. E. Stewart-Jones, Geraldine Gillespie, Ian M. Overton, Rupert Kaul, Philippe Roche, Andrew J. McMichael, Sarah Rowland-Jones and E. Yvonne Jones
J Immunol August 15, 2005, 175 (4) 2459-2468; DOI: https://doi.org/10.4049/jimmunol.175.4.2459

Abstract

Long-term nonprogression during acute HIV infection has been strongly associated with HLA-B*5701 or HLA-B*5703. In this study, we present the high resolution crystal structures of HLA-B*5703 complexes with three HIV-1 epitopes: ISPRTLNAW (ISP), KAFSPEVIPMF (KAF-11), and KAFSPEVI (KAF-8). These reveal peptide anchoring at position 2 and their C termini. The different peptide lengths and primary sequences are accommodated by variation in the specific contacts made to the HLA-B*5703, flexibility in water structure, and conformational adjustment of side chains within the peptide-binding groove. The peptides adopt markedly different conformations, and trap variable numbers of water molecules, near a cluster of tyrosine side chains located in the central region of the peptide-binding groove. The KAF-11 epitope completely encompasses the shorter KAF-8 epitope but the peptides are presented in different conformations; the KAF-11 peptide arches out of the peptide-binding groove, exposing a significant main chain surface area. Bioinformatic analysis of the MHC side chains observed to contribute to the peptide anchor specificity, and other specific peptide contacts, reveals HLA alleles associated with long-term nonprogression and a number of related HLA alleles that may share overlapping peptide repertoires with HLA-B*5703 and thus may display a similar capacity for efficient immune control of HIV-1 infection.

Particularly during acute infection (1, 2, 3, 4), CD8+ T lymphocytes are thought to play a pivotal role in the control of HIV-1 replication, yet, the clinical course of chronic disease is heterogeneous, often with no apparent correlation between the strength and focus of the CD8+ T cell-mediated immune response and duration of chronic disease. Both viral and host determinants influence HIV-1 disease pathogenesis, and thereby contribute to long-term clinical outcome (5). The unique ability of HIV-1 to accommodate a large number of mutations throughout its genome (6) frequently results in viral escape from immune recognition through the accumulation of mutations in viral epitopes (7, 8). Host factors that influence disease progression have also been defined. Of note is the strong influence of HLA, in particular the MHC class I loci, on HIV-1 disease progression. Maximum heterozygosity at the MHC class I loci is considered advantageous, and individual MHC class I alleles also associate with favorable clinical outcome (9, 10, 11).

In the context of CD8+ T cell-mediated immunity, the heterozygous advantage at the MHC class I locus probably reflects the availability of diverse MHC class I molecules to present a range of antigenic epitopes to T lymphocytes, inducing multispecific immune responses, delaying the incidence of viral immune evasion. Recently, the HLA-B locus has been found to be the principal focus of HIV-specific CD8+ CTL responses compared with HLA-A or HLA-C alleles and disease progression is strongly associated with particular HLA-B alleles but not HLA-A or HLA-C alleles (12). Of the MHC class I alleles also associated with prolonged AIDS-free survival, HLA-B*27 and HLA-B*57 have been consistently described (10, 12, 13, 14, 15). The GAG p24-derived B*2705-restricted epitope (KRWIILGLNK), is highly immunodominant and induces a potent CD8+ T cell-mediated immune response consistent with slower onset of AIDS in HIV-1-infected donors. In this epitope, a compensatory mutation at peptide position 6 (P6; valine to methionine) (16) is required for P2 anchor mutations to be compatible with virus viability. Once achieved, P2 mutations diminish binding to MHC, allowing immune evasion of infected cells from immunodominant CD8+ T cell responses and progression to AIDS (17, 18).

The molecular factors that form the basis of the association of HLA-B*57 with slower onset of AIDS are likely to be multifactoral. In chronic infection it has been observed that the HLA-B*57+CD8+ T cell responses in long-term nonprogressors (LTNP)5 are highly focused onto HLA-B*57-restricted epitopes. This contrasts with the broader responses to both HLA-B*57 and other MHC class I peptide complexes seen in HLA-B*57-positive progressors (19). Recently, a study of acutely infected HLA-B*57 patients demonstrated a predominance of HIV-1-specific CD8+ T cell responses to HLA-B*57 epitopes that were notably stronger than responses restricted to the combined set of coexpressed MHC class I molecules (20). A wide variety of HLA-B*57-restricted HIV-1 epitopes have been mapped within almost all of the HIV-1 structural and regulatory proteins (21, 22). The CD8+ T cell immune responses directed to gag epitopes have been the focus of most studies involving HLA-B*57+ HIV-1-infected patients, since several of these epitopes are defined immunodominant targets. Five HLA-B*57-restricted epitopes have been mapped to the p24 capsid protein: KAFSPEVIPMF (KAF-11), ISPRTLNAW (ISP), KAFSPEVI (KAF-8), TSTLQEQIAW (TST), and QASQEVKNW (QW-9). CD8+ T cell responses to the gag epitopes have been identified in HLA-B*57+ patients through chronic disease both in HLA-B*57+ normal progressor and LTNP patient cohorts. Many HLA-B*57+ LTNPs can control HIV-1 replication, reducing viral load to undetectable levels (<50 copies/ml) and have normal CD4+ T cell counts and immune functions. The HLA-B*57 molecule, however, is a fundamental component of this protective effect, and enhanced immune control is thus likely to be related to the particular population of antigenic epitopes presented by the HLA-B*57 molecule.

In the HLA-B*57-binding motif, position 2 anchor residues are typically alanine, serine, or threonine, and C-terminal amino acids are usually phenylalanine or tryptophan, but the smaller hydrophobic amino acids such as isoleucine or valine have also been identified (23). HLA-B*57 epitopes have shown variability in length, ranging from between 8 to 11 aa, although most are 9- or 10-aa long. Goulder et al. (24) identified that HLA-B*57 can present an 11-mer gag peptide (KAFSPEVIPMF, hereafter KAF-11) for CTL recognition that entirely overlaps in sequence another epitope: the HLA-B*57 8-mer (KAFSPEVI, hereafter KAF-8). CTL clones derived from a patient were capable of optimal recognition of either one or the other peptide. KAF-8 is the only HLA-B*57 8-mer peptide described to date. Although the KAF-11 epitope is immunodominant compared with the KAF-8 epitope, independent CTL responses to the KAF-8 or KAF-11 peptides have been detected within the single HLA-B*57 patient, in line with the suggestion that epitopes competing in the endoplasmic reticulum for HLA molecules can play a role in determining the immunodominance of viral peptide fragments (25).

To define the specific interactions of HLA-B*57 with epitopes of different lengths, sequence, and anchor usage, we solved the high resolution crystal structures of two immunodominant p24 GAG epitopes (ISP and KAF-11) and the GAG epitope (KAF-8) in complex with the HLA-B*5703 allele. Using the structures we are able to assess the roles of individual MHC amino acid side chains in stabilizing the peptides in the HLA-B*5703-binding groove and identify a unique combination of amino acid side chains that form the basis for the specific binding of HLA-B*5703 epitopes. We combine a bioinformatic analysis of HLA molecules, highlighting HLA-B alleles which may share HLA-B*5703 peptide-binding specificity and consider the capacity of these alleles to present overlapping peptide repertoires which may implicate them in LTNP.

Materials and Methods

Protein preparation

Residues 1–274 from the HLA-B*5703 H chain, cloned into the pGMT7 plasmid, were expressed as inclusion bodies in Escherichia coli. Solubilized HLA B*5703 H chain was refolded with β2 microglobulin (β2m) in the presence of KAFSPEVIPMF (HIV p24 gag 30 to 40), ISPRTLNAW (HIV gag p24 147 to 158), or KAFSPEVI (HIV p24 gag 30 to 37) peptides as previously described (46). Refolded MHC class I molecules were isolated by gel filtration, further purified with anion exchange chromatography and a second round of gel filtration, and finally concentrated to 12–15 mg/ml in 10 mM HEPES, pH 7.0 for crystallization.

Protein crystallization and data collection

All crystallizations were done by the sitting drop vapor diffusion technique. Due to the very low yields of refolded protein complexes, a focused crystallization strategy was used based on conditions known to allow peptide-HLA complexes to crystallize. After several days of equilibration, and cross-seeding from HLA-B*8-nef crystals (G. B. E. Stewart-Jones, unpublished data), single crystals of HLA-B*5703-peptide complexes were obtained at room temperature (21°C) from 16% Peg 8000, 50 mM MES, pH 6.5. These grew as rods or plates of maximal dimensions 250 μm by 170 μm by 50 μm. The largest crystals were soaked briefly and sequentially in mother liquor solutions supplemented with 10% and 20% glycerol, and then flash-cooled and maintained at 100 K in a cryostream (Oxford Cryosystems). The HLA-B*5703-KAFSPEVIPMF, HLA-B*5703-ISPRTLNAW, and HLA-B*5703-KAFSPEVI data sets were collected at station 14.2 of the Synchrotron Radiation Source (SRS, Daresbury, U.K.) with an ADSC-Q4 (Area Detector Systems) CCD detector. All three pMHC crystals belonged to the orthorhombic space group P212121. Data were autoindexed with the program DENZO (26) and scaled with the program SCALEPACK (26), the statistics are summarized in Table I⇓.

View this table:
Table I.

Crystallographic statistics

Structure determination, refinement, and analysis

Crystal structures were determined by molecular replacement using the program EPMR (27). Initially, for the HLA B*57-KAFSPEVIPMF data set, the H chain and β2m domains from the crystal structure of a HLA B8-nef peptide complex (G. B. E. Stewart-Jones and E. Y. Jones, unpublished data) was used as the search probe (with peptide and water coordinates omitted) and yielded an unambiguous solution with a correlation coefficient of ∼0.7 and R factors of ∼35% for data between 30 and 4 Å. The HLA-B*5703 sequence was imposed onto the molecular replacement solution using the SWISSMODEL server (〈www.expasy.org/swissmod/SWISS-MODEL.html〉). Using program CNS (28), the model were subjected to several rounds of rigid body refinement of individual domains (α1/α2, α3, β2m) and a Fo-Fc difference map revealed clearly the presence of the 11-mer peptide when displayed in the graphics program O (29).

The peptides were then modeled into Fo-Fc difference density maps and further rounds of refinement were conducted using standard CNS protocols for bulk solvent correction and overall anisotropic B-factor scaling, positional refinement, simulated annealing, and individual B-factor refinement. Manual refitting of the models was conducted using O and then crystallographic conjugate gradient minimization refinement was performed; final stages of refinement after water picking with ARPw/ARP (30) were done using a restrained refinement algorithm in REFMAC5 (31).

The other datasets were solved in similar ways using the refined HLA- B*57-KAFSPEVIPMF H chain and β2m as the molecular replacement probe with peptide and water molecules omitted. Briefly, EPMR allowed unambiguous solutions to be obtained for the other two HLA-B*5703 data sets and using O, the peptide models were built into Fo-Fc difference density maps. Subsequent refinement was performed as described above and structural superpositions were conducted using program SHP (32). Solvent accessible surface areas were calculated with Naccess (33) and the figures generated in Bobscript (34).

Bioinformatic analysis of related HLA

The B*570301 suballele was searched against all EMBL HLA-B sequences using BLASTP (35, 36). From the resultant alignments, residues in the B*570301 suballele were assigned positional equivalence to residues in the aligned HLA sequences. Each HLA sequence was scored according to the identity of residues at positions inferred to be important by structural analysis; identical residues incremented the score by one unit, nonidentical residues neither incremented nor reduced the score.

ELISPOT analysis of HLA-B*5703 PBMC

Molecular HLA class I genotyping was performed using PCR with sequence-specific primers (37). Peptides were synthesized by F-moc chemistry, using a Zinnser Analytical synthesizer (Advanced Chemtech), and purity was established by high pressure liquid chromatography. Four predefined B*5701-restricted CD8+ epitopes were screened for reactivity in B*5703 individuals: KAFSPEVIPMF (HIV-1 p24), TSTLQEQIGW (HIV-1 p24), ISPRTLNAW (HIV-1 p24), IVLPEKDSW (HIV-1 Pol) (Los Alamos). Overlapping peptide assays were run using 122 15-mer peptides overlapping by 11 aa and spanning HIV-1 Gag (AIDS Research and Reference Reagent Program, National Institute of Allergy and Infectious Diseases, National Institutes of Health). IFN-γ ELISPOT assays were set up using PBMC at 1 × 105 per well. For individual epitopes, the final peptide concentration was 10 mg/ml; overlapping Gag peptides were set up in a matrix of 23 pools, with each peptide at 2 mg/ml in 2 separate pools. All responses were confirmed using individual 15-mer peptides and optimized peptide epitopes.

Results

Residues 1–245 of HLA-B*5703 were expressed as inclusion bodies in E. coli and refolded with three HIV-1 index peptides (KAFSPEVIPMF, KAFSPEVI, and ISPRTLNAW) and β2m. The three pMHC complexes were purified and separately crystallized under identical conditions (see Materials and Methods). The crystals were used for x-ray crystallographic structure determination. All structures were of the same orthorhombic space group (P212121) and almost identical unit cell dimensions, with one HLA-B*5703 peptide complex per crystallographic asymmetric unit (Table I⇑ and Materials and Methods). Crystal structure resolutions were as follows: KAFSPEVIPMF, 1.6Å; KAFSPEVI, 2.0Å; and ISPRTLNAW, 1.35Å. All regions of the HLA-B*5703 and β2m were well ordered and electron density for each of the viral peptides was unambiguous and showed no indication of multiple peptide conformations (Fig. 1⇓). The resolution of these structures allowed precise determination of ordered networks of water molecules integral to the HLA-peptide complexes and accurate definition of the role in peptide binding of HLA amino acid side chains in the binding groove.

FIGURE 1.
FIGURE 1.

The high resolution crystal structures of the three HIV-1 peptides (shown without the HLA-B*5703 peptide-binding groove). a, ISPRTLNAW; b, KAFSPEVIPMF; and c, KAFSPEVI. Representative 2fo-fc electron density maps are shown as green chicken wire, contoured at 2.5 ς.

Overall HLA-B*5703 structure

The overall HLA-B*5703 structure is characteristic of the classical HLA-class I structure, with the α1/α2 domain peptide-binding groove composed of two α-helices supported by an eight-stranded β-sheet, into which the antigenic peptide fits. The structures confirm two dominant anchor residues from the peptide interact with the HLA-B*5703 molecule: the PC-anchor residue (F/W/I) occupying the F-pocket and the smaller B-pocket accommodating the serine or alanine side chains commonly seen at position 2 of the binding motif.

The solvent-accessible surface areas for these peptides (ISP = 255Å2, KAF-8 = 290Å2, and KAF-11 = 608Å2) show a significant difference for the KAF-11, exposing more than double the peptide surface area to the solvent than either ISP or KAF-8 peptides. These solvent accessible surface areas represent the peptide regions most accessible to recognition by specific TCRs. The buried surface areas of the peptides, however, are relatively similar (ISP = 1077Å2, KAF-8 = 937Å2, and KAF-11 = 968Å2).

The central region of the peptide-binding groove is composed of a large number of tyrosine residues (termed “tyrosine bed”; Y7, Y9, Y74, Y99 and Y116 see Figs. 2⇓ and 3⇓a). These residues form a hydrophobic platform in the floor of the groove but through their hydroxyl groups also mediate direct or indirect (via water) polar interactions with the associated peptides. The three index peptides show most variation in their main chain conformations, side chain orientations, and water networks over the portions that interact with this tyrosine bed. This is consistent with peptide elution studies from HLA-B*5701 and HLA-B*5702 which have shown greatest variability in amino acid sequence in the central regions of the peptides (P3 to PC-1). The KAF-8 peptide structure traps two water molecules between the peptide and the tyrosine bed whereas the ISP peptide traps four water molecules and the KAF-11 peptide has six clearly ordered water molecules clustering beneath the peptide and interacting with the tyrosine bed. Only one water molecule occupies the same position in all three HLA-B*5703-peptide complexes; this water makes hydrogen bonds between the P3 carbonyl group, the α1 helix Q65 side chain, and the Q69 hydroxyl group.

FIGURE 2.
FIGURE 2.

Structures of the HIV-1 peptide-HLA-B*5703 complexes viewed in profile with the α1 helix and α3 and β2m domains omitted. The ISP peptide (a) is colored cyan, the KAF-11 peptide (b) is colored orange, and the KAF-8 peptide (c) colored green. MHC residues mediating hydrogen bonds are labeled and shown in yellow. Hydrogen bonds are shown as red dashes. Water molecules are represented as magenta spheres.

FIGURE 3.
FIGURE 3.

a, α1/α2 superimposition of the three viral peptides illustrating the KAF-11 bulge and the cluster of tyrosine side chains in yellow located in the central region of the peptide-binding groove (tyrosine bed); b, the key residues of the HLA-B*5703 B pocket (yellow) that confer anchor specificity at peptide position P2; and c, the structure of the F pocket with the MHC side chains making greatest contacts with the peptide PC-anchors colored in yellow. Side chains that are near the F pocket, but make few interactions with the anchors are colored blue. Peptides are colored according to Fig. 2⇑.

Comparison of KAF-11 and KAF-8 peptide structures

The two related epitopes KAF-11 and KAF-8 show dramatically different conformations when bound to HLA-B*5703. Some CD8+ T cell clones specific for the KAF-8 peptide have been identified that have partial cross-reactivity to the KAF-11 peptide, whereas KAF-11-specific clones do not exhibit any cross-reactivity to the KAF-8 peptide (22). Although the KAF-8 peptide is bound in a fully extended conformation, the KAF-11 peptide contains three extra amino acids at the C terminus (VMF), that induce a switch in the usage of PC anchor residue from Ile8 with the KAF-8 epitope to Phe11 with the KAF-11 epitope. To accommodate the extra residues within the same N- to C-terminal length the KAF-11 peptide forms a large hairpin structure protruding from the peptide-binding groove. This hairpin buckles the P5 to P9 region out of the groove and into the solvent (Figs. 1⇑b and 2⇑a) with the main chain Glu6 forming a sharp bend at the crest of this bulge. The main chain of KAF-11 is thus displaced by up to 7 Å from the position of equivalent regions of KAF-8 in the peptide-binding groove. The bulge in the KAF-11 peptide structure is stabilized by prolines at peptide residues P5 and P9. The ring of Pro5 stacks with the aromatic ring of phenylalanine at P3, buttressing the bulge structure. Both side chains also interact with the side chains of Ile8 and Pro9, forming hydrophobic contacts that further stabilize the bulged peptide conformation. A tightly bound water molecule mediates a hydrogen bond between the main chain nitrogen atom of Glu7 at the top of the bulge and the side chain of Q155.

The first three amino acids of both the KAF-8 and KAF-11 peptides superimpose almost identically between structures except for the ring of the Phe3 residue which is shifted ∼0.75 Å toward the N terminus in the KAF-8 peptide complex compared with the KAF-11 structure. Four conserved hydrogen bonds and one water-mediated hydrogen bond are made between the HLA-B*5703 and the P1-P3 residues in KAF-8 and KAF-11, locking the N-terminal residues of the peptides into very similar conformations. The resultant orientation of the Ala2 side chain into the B pocket is almost identical in the two complexes. The HLA-B*5703 B pocket is relatively shallow with the M45 and M68 side chains filling the space available for bulkier B-pocket anchors in other HLA class I peptide complex structures. After the P3 residue the conformations of the bound KAF-8 and KAF-11 peptides diverge.

The KAF-8 peptide structure exposes 43% less main chain surface area than the KAF-11 (KAF-8 = 49Å2; KAF-11 = 86Å2) and Ser4 and Val7 are fully exposed, whereas the side chains of Pro5 and Glu6 are orientated toward the α1 and α2 helices, respectively. This pattern of side chain exposure is very different to that of the same residues in KAF-11. For example in the KAF-8 structure the ring of the P5 proline lies planar with the putative TCR recognition surface, whereas in the KAF-11 structure Pro5 is elevated out of the groove and orientated perpendicular to the plane of the recognition surface. The side chains of Glu6 and Val7 are most prominently exposed at the crest of the bulge in the KAF-11 structure, and are thus likely to make intimate contacts with specific TCRs. The large surface area of KAF-11 main chain exposed also indicates that there may be significant peptide main chain contacts generated between the TCR and the KAF-11 bulge.

The large volume of the HLA-B*5703 F-anchor pocket makes it suited to binding large aromatic C-terminal side chains. The structures reveal 46 interatomic van der Waal’s contacts exist between the Phe11 of KAF-11 and the HLA-B*5703 F-anchor pocket whereas only 30 occur with the KAF-8 Ile8 anchor, an observation consistent with the reduced stability of the KAF-8 peptide in the HLA-B*57 groove compared with the KAF-11 peptide (24). Indeed peptide elution studies have shown that F or W are exclusively found as the C-terminal amino acids from pooled self HLA-B*5701 and HLA-B*5702 peptide complexes (23). The KAF-11 Phe11 side chain forms extensive ring-stacking interactions with the HLA-B*5703 Y123 side chain, and in addition has significant interactions with the I95, N77, and T143 side chains (see Table II⇓). A more detailed discussion of the role of individual F-pocket side chains in specifying the anchor preference is given below.

View this table:
Table II.

Hydrogen bonds and van der Waal’s interactions for the HLA-B*5703-peptide complexes

ISPRTLNAW has an atypical peptide structure at P4-P5

The immunodominant ISP peptide complexed with HLA-B*5703 has a remarkable number of direct hydrogen bonds (13) between the peptide and MHC compared with other pMHC structures (see Table II⇑). This is the first pMHC structure showing a tryptophan PC side chain bound to the F pocket. The plane of the tryptophan side chain lies parallel with the peptide axis. Extensive van der Waal’s contacts are made by hydrophobic ring-stacking interactions with Y123 and many contacts are made to the I95 and N77 side chains. Of the three PC side chains studied here, tryptophan has the best shape complementarity for the F pocket (ScW(ISP) = 0.73, ScF(KAF-11) = 0.70, ScI(KAF-8) = 0.69) and the greatest number of van der Waal’s contacts (86 contacts). A water molecule (water 35) stabilizes the Trp9 side chain orientation by hydrogen bonding between the Trp NE1 atom and the Y116 hydroxyl group. The B-pocket anchor, serine, forms a pair of bifurcated hydrogen bonds between the Ser2 hydroxyl group and the OE1 and ND1 atoms of the α1 helix residues E63 and N66, respectively. Threonine, another preferred HLA-B*57 B-pocket anchor, is likely to adopt the same hydrogen-bonding network. Because serine and alanine are small side chains, little distortion of the B pocket is apparent in any of the three HLA-B*5703 peptide complexes, the only difference is that the side chain of N66 rotates by some 20° in the KAF-11 structure to avoid steric clashes with the peptide Ser4 side chain.

Usually, in pMHC structures the peptide position 3 side chain is orientated toward the α2 helix and fits into the D pocket. The Pro3 aliphatic ring atoms of the ISP peptide, however, are orientated toward the floor of the peptide-binding groove, causing a conformational rearrangement of the Y99 side chain when compared with either of the KAF complexes (Fig. 2⇑). The Pro3 ring makes close contacts with the ring of Y159 and the peptide Thr5 side chain hydrogen bonds toward the Y99 hydroxyl group, occupying the D pocket. This contrasts with the Pro5 side chain in the KAF-8 peptide structure, which is orientated toward the α1 helix, whereas in the ISP structure P5 is orientated toward the α2 helix. This induces a lateral zig-zag conformation in the ISP nonamer peptide main chain at P4-P7; hence accommodating an extra amino acid when compared with the KAF-8 octamer. In contrast with conformations observed in many nonamer peptide-MHC structures, for which the P4 side chain projects into solvent, the ISP Arg4 curves around to allow the guanadinium group to form a network of hydrogen bonds and salt bridges focused on the acidic E63 side chain at the entrance of the B pocket. A water molecule, conserved between KAF-8 and KAF-11 structures, that hydrogen bonds the E63 carboxylate and the P2 carbonyl is replaced by the ISP Arg4 guanadinium head group. Two salt bridges are formed between the basic Arg4 NH1/NH2 groups and the OE2 oxygen of the acidic E63. The guanadinium primary amine NH1 atom also hydrogen bonds to the P2 carbonyl oxygen and the NH2 atom forms a water-mediated hydrogen bond to the A60 carbonyl group from the α1 helix.

Analysis of the HLA-B*5703 residues conferring preferred anchor specificity

Of all the human MHC class I genes, the most polymorphic and rapidly evolving is HLA-B (23). Polymorphisms can alter peptide-binding specificity, however, analysis of peptide-binding specificity of related HLA-B allotypes shows that some dominant anchor residue preferences are frequently conserved among related allotypes (23). Detailed knowledge of the roles of HLA side chains in specifying anchor preference can provide information on HLA alleles that may share specific peptide-binding motifs and thus have an overlap in presented peptide repertoires. Because S/T/A are the preferred HLA-B*57 B-pocket anchors and F/W are preferred F-pocket anchors, the structures herein provide a sample of the interactions made with four anchor residues (Ser and Ala at P2, Trp and Phe at PC) that correspond to the HLA-B*57-binding motif. The allelic differences between HLA-B*5703 and HLA-B*5701 are at positions 116 (Y to S) and 114 (N to D) near the F pocket. Because very few contacts to peptide are contributed from either of these side chains, the F-anchor residues are likely to adopt very similar conformations in HLA-B*5703 and HLA-B*5701. In the following, HLA-B*5703 residues are selected to define a distinctive peptide-binding motif, based on their interactions with the anchor residues, their role in shaping the anchor pockets, or their contribution in generating other interactions with the bound peptides (Table III⇓).

View this table:
Table III.

The MHC class I residues selected as key in conferring the HLA-B*5703 anchor residue motif and contributing to peptide binding

Analysis of B-pocket residues

Because the HLA-B*5703 B pocket is relatively shallow, it accommodates small side chains. The M45 and M67 side chains occupy the space where larger B-pocket anchors are seen to reside in other pMHC structures, thus sterically limiting the size of potential P2 side chains. The entrance of the B pocket is hydrophilic and acidic, and E63 plays an important role in hydrogen bonding the three peptide P2 main chain nitrogens and the ISP P2 serine hydroxyl group. The N66 side chain ND1 atom also hydrogen bonds the ISP P2 serine hydroxyl, and is involved in hydrogen bonding the peptide main chain P3 carbonyl groups of the three peptides. Thus, residues conferring B-pocket specificity are M45, M67, E63, and N66.

Analysis of F-pocket residues

The HLA-B*5703 F pocket is generally composed of hydrophobic side chains consistent with the binding of hydrophobic side chains at the peptide C termini. Almost all HLA-B*5701 peptides characterized by peptide elution and sequencing or epitope mapping share F or W at their C termini. The KAF-8 structure illustrates the binding of Ile as the F anchor but this peptide has significantly lower binding to HLA-B*57 as the contacts made by the Ile anchor in the main form a subset of the contacts made by the optimal Trp and Phe anchors exemplified in the KAF-11 and ISP structures. Therefore, this analysis focuses on these latter two structures. For both W and F, Y123 engages with close aromatic ring-stacking interactions that orientate and stabilize the F or W aromatic groups parallel with the axis of the peptide-binding groove (Fig. 3⇑c). The Phe side chain is tightly wedged between Y123, I95 and N77, while the Trp accesses additional space at the base of the F pocket to make contacts to the I95 and A81 side chains. Little difference is observed in the side chain conformations of the F-pocket amino acids of the three peptide structures. The side chain of A81, although it does not generate many contacts with the anchor residues, is of particular importance because the small size of alanine allows the F pocket to be of sufficient volume to accommodate these large aromatic anchors. Therefore, any increase in size of HLA residue 81 may have a dramatic effect on the specificity of the F-anchor pocket due to the specific orientation of the aromatic rings. It is noteworthy that Y116 (a serine in HLA-B*5701) makes few contacts with the anchor residues since it is orientated toward the central region of the peptide-binding groove, thus allowing the same dominant F-anchor residues to be shared between these HLA-B*57 subtypes. Therefore, residues considered to be important in conferring W and F specificity in HLA-B*5703 are Y123, A81, T143, I95, and N77.

Other B*5703 residues making specific interactions

The specific peptide population selected by HLA-B*5703 is determined by key amino acids in the B and F pockets but is also modulated by the properties of the central region of the binding groove. Some tyrosine bed residues provide an important contribution to the specific associations with these peptides. The conformational adjustment of the Y9 side chain between KAF and ISP structures suggests that this side chain may provide important flexibility to the peptide-binding groove in accommodating different peptides. Y159 makes numerous contacts to all three P3 side chains, and the Y99 side chain interacts closely with the Pro3 ring from the ISP peptide. All three Tyr residues are therefore considered to participate in the flexibility of the peptide-binding groove to bind the HLA-B*57 range of epitopes. Additionally, V97 is positioned in the central region of the peptide-binding groove and is likely to have an important role in determining the specificity of the peptide-binding groove. In other pMHC structures, side chains at position 97 are observed to frequently interact with the bound peptides (38, 39). V156 is involved in shaping the D pockets, participating in van der Waal’s contacts with the KAF11 and KAF8 peptide P3 phenylalanines.

The amino acids discussed above are therefore the side chains conferring the principal peptide-binding specificity of the HLA-B*5703 peptide-binding groove and are illustrated in Table III⇑.

Bioinformatic analysis of related HLA sequences

Using the 14 selected residues as a combined set (in Table III⇑) every HLA-B sequence from the EMBL database (40) was assigned positional equivalence with the HLA-B*570301 suballele using a BLAST (35, 36) based protocol and ranked according to identity with the selected amino acids. This generated a list of HLA alleles and suballeles that showed decreasing similarity to the combined set of selected amino acids from the peptide-binding groove. A greater number of identical amino acids implies greater similarity in the range of peptides bound and presented by a given allele.

Many HLA-B*57 suballeles share identity with all the amino acids selected from the HLA-B*5703 structure and are thus likely to share similar anchor preferences and present significantly overlapping peptide repertoires in vivo. This is supported by data for peptide repertoires eluted from HLA-B*5701 and HLA-B*5702 (23). These data show identical dominant anchor residue usage and, among the limited number of sequenced peptides (six for HLA-B*5701 and eight for HLA-B*5702) from each HLA subtype, an identical peptide was detected from both suballeles. HLA-B*5702 also shares the two same polymorphic differences between HLA-B*5701 and HLA-B*5703 (S/Y116 and N/D114, respectively) supporting the notion that HLA-B*5703 shares similarity in F-anchor preference.

The HLA-B*58 suballeles show the next best identity based on the selected residues. All HLA-B*58 suballeles have the M45T and V97R/W differences. Two suballeles HLA-B*5806 and HLA-B*5802 have a third alteration: I95L or I95W, respectively. Peptide elution studies from HLA-B*5801 and HLA-B*5802 molecules reveal that HLA-B*5801 shows no difference to the HLA-B*5701 W or F primary anchor preference, whereas HLA-B*5802 shows a modest change in preferring F over W as the F-anchor residue (23). The more distinctive I95W alteration found in HLA-B*5808 and HLA-B*1516 is likely to modify the anchor preference, and this is evidenced by dominant selection of tyrosine in the HLA-B*1516 F pocket (23). The orientation and proximity of I95 to the F-anchor side chain in the ISP structure suggests that W95 may form extensive ring-stacking interactions with a tyrosine F anchor. Little alteration is observed for the dominant B-anchor residue specificity, suggesting that the A/S/T selection is not crucially reliant on either the M45 or T45 side chains in the B pocket.

Members of the subgroup B63 from the B15 family, HLA-B*1516, HLA-B*151701, and HLA-B*151702 match as closely as most B*58 suballeles and carry a V97R alteration and either a I95L (HLA-B*1517 suballeles) or a I95W alteration (HLA-B*1516). Thus, there may be differences in the F-anchor preferences between HLA-B*1516 and HLA-B*1517 alleles, however, these alleles share very similar anchor pocket residues to the HLA-B*58 suballeles. HLA-B*1567 is found to contain identical selected residues as seen in HLA-B*1516, and thus these alleles are likely to share an overlap in presented peptide repertoires. Because the HLA-B*1516 and HLA-B*1517 alleles have been found to present HLA-B*57 peptides and are associated with LTNP (56), these amino acid changes are likely to constitute small functional differences for the presentation of respective peptide repertoires relevant to LTNP in HIV infection.

Because the ISP P2 serine side chain is anchored by two hydrogen bonds to the E63 and N66 side chains, changes to this hydrogen bonding network are likely to affect the specificity of serine or threonine as preferred B-pocket anchor residues in HLA-B*57 alleles. Elution studies of peptides from the HLA-B*15 allele have shown that the N66I change (frequently accompanied by the M67S alteration) is associated with a switch from T/S/A specificity to a range of larger side chains. Thus, alleles bearing these B-pocket differences are unlikely to present similar peptide repertoires as the HLA-B*57 suballeles (23). However, HLAs B*151701, B*151702, B*1567, and B*1516 are found to contain residues conserved with the HLA-B*57 alleles at the N66 and M67 positions. In summary, the analysis suggests HLA alleles B*5707, B*5708, B*570101, B*570302, and B*570102 share complete identity in the selected residues (Table IV⇓) and therefore are likely to display significant overlap in bound peptide repertoires presented to CD8+ T cells. Additionally, the HLA alleles B*5702, B*5704, B*5705, B*5706, B*5709, B*1516, B*151701, B*151701, B*1567, B*5801, B*5804, and B*5805 alleles share considerable identity with the selected residues and therefore are likely to display overlapping peptide repertoires with the HLA- B*5703 allele.

View this table:
Table IV.

Match of residues conferring a HLA-B*5703-like peptide-binding specificitya

Immunodominant recognition of KAF11 and ISP epitopes by HLA-B*5703 patients

Responses to the four predefined HLA-B*5701 epitopes were screened in 12 HLA-B*5703+, HIV-infected, therapy-naive Kenyan women. Two epitopes were strongly immunodominant: KAF was recognized by 12 of 12 participants (100%), with a mean epitope-specific response frequency of 2120 spot-forming units (SFU)/million PBMC (range, 115-5140 SFU/million PBMC); ISP was recognized by 11 of 12 (92%), with a mean epitope-specific response frequency of 1670 SFU/million PBMC (range, 235-4720 SFU/million PBMC). Intracellular cytokine staining and flow cytometric analysis in 11 women demonstrated that responses were CD8+ T cell mediated, and overlapping peptide assays confirmed that these responses were immunodominant within HIV Gag for 7 of 7 subjects. A minority of subjects recognized TST (3 of 10; 30%) or IVL (2 of 9; 22%).

Discussion

The structures of pMHC complexes for three HIV-1 gag epitopes of variable primary sequence and length provide insight into the features of peptide presentation by the HLA-B*5703 allele. We define the structures of two related epitopes, the sequence of one of which is completely contained within the other, and illustrate the mechanisms by which the peptide-binding groove achieves flexibility in binding different antigenic peptides. The B and F pockets are clearly the dominant peptide anchor sites, and, as observed for other pMHC complexes (38), the central regions of the peptides show the greatest differences in bound peptide conformations. In HLA-B*5703 one distinctive mechanism for accommodating variability in the central regions of the epitopes is through a platform of tyrosines (“tyrosine bed”) that 1) contribute some flexibility in side chain conformation and 2) recruit water molecules into the peptide-binding groove. Variable networks of water molecules formed between the peptide and binding groove have been noted in both human and murine pMHC structures (41, 42, 43, 44, 45) and, as in these human HLA-A, HLA-B, and mouse alleles, water clearly plays an important role in modulating the binding of different peptides to HLA-B*5703. The F- and B-anchor pockets, however, specify the peptide anchor residues through direct side chain contacts. The F-anchor pocket of HLA-B*5703 is particularly distinctive in that it confers a preference for tryptophan as the C-terminal residue of the epitope. Tryptophan is encoded by one codon only, suggesting that if the genetically variable HIV virus uses this amino acid there may be functional restraints on mutation. HLA molecules that preferentially present peptides using tryptophan as a dominant anchor may thus select epitopes in which mutation to abolish HLA binding and hence allow immune evasion incur a reduction in viral fitness.

Structural studies on MHC class I molecules have defined the characteristics of peptide binding which are common to all MHC class I molecules, but have also served to highlight features which are distinctive to particular alleles. Generally such features relate to the properties of the binding groove and determine the anchor residue preferences. As the database of structural information has built up it has been increasingly used to help define anchor motifs and hence aid in the identification and refinement of CTL epitopes (53, 54). The interplay between the structural and immunological characterization of epitopes has particular clinical relevance for the design of peptide-based vaccines to boost the immune response to viruses such as HIV-1 or tumors (47). In this arena structural studies continue to provide fresh insights into peptide binding for even relatively well-characterized alleles; for example, two crystal structures of noncanonical peptides complexed with H-Kb indicated alternative strategies to generate binding that are of general relevance to the design of peptide-based vaccines (41, 42). Conversely, studies of previously structurally uncharacterized alleles are providing detailed information on the nature of the binding groove which is of very specific relevance, for example, to the recent crystal structures of HLA-A*1101 complexed with immunodominant nonamer and decamer HIV-1 epitopes (48).

Structures of TCR-pMHC complexes have revealed that TCRs frequently bind pMHCs by centering onto a prominent peptide side chain (e.g., tyrosine (49, 50, 51) methionine or glutamate (G. B. E. Stewart-Jones and E. Y. Jones, unpublished data) and have a relatively conserved docking footprint on the pMHC surface (52, 55). Because the bulky bulge motif of the KAF-11 peptide as presented by HLA-B*5703 is found within the central region of the average TCR-docking footprint, TCRs recognizing the KAF-11 bulge may require formation of a large cavity between the CDR loops to dock onto this motif. Conversely, because the bulge is likely to play a dominant role in TCR recognition of the KAF-11 epitope any mutations which affect its main chain conformation might be predicted to trigger immune escape. The two peptide proline residues at P5 and P9 are particularly important in stabilizing the bulge conformation. The KAF-11 region of the HIV-1 GAG sequence, however, is highly conserved (HIV web site, 〈www.hiv-lanl.gov〉).

The conformations of the KAF-8 and KAF-11 peptides as presented by HLA-B*5703 are radically different. Some cross-recognition of the KAF-11 peptide has been reported to have been detected using KAF-8-specific clones, with a ∼2-log molar peptide concentration difference to achieve equivalent levels of specific lysis (24). This cross-recognition is surprising in the light of the structures of the two HIV-1 index peptides illustrating such significant differences in the main chain and side chain conformations of the peptides and may warrant further assessment.

The HLA-B*5703 peptide structures allow the functional impact of polymorphisms in related HLA sequences to be assessed. HLA-B*57 is associated with LTNP in HIV+ individuals and one suggestion is that this concords with the particular repertoire of epitopes presented to CD8+ T cells. Therefore, HLA molecules sharing similar peptide-binding motifs may present similar peptide repertoires and be associated with LTNP. Our analysis of the structural determinants of the HLA-B*5703 peptide-binding repertoire suggest that many of HLA-B*57 suballeles share identity between residues conferring anchor-binding specificity. The next closest match is the group containing HLA-B*5801, HLA-B*1567, HLA-B*1516, and HLA-B*1517. To date the association of HLA-B*5701 and HLA-B*5703 with LTNP (15, 19) has been formally documented, however recent findings have indicated that the related alleles HLA-B*5801, HLA-B*1516, and HLA-B*1517 are also associated with LTNP (Refs. 12 and 56). These observations support the premise that peptide repertoires are functionally associated with LTNP. However, HLA-B*5701 progressors have shown CTL responses to defined HLA-B*5701 epitopes similar to those recognized by LTNPs, suggesting that other factors act in conjunction with the HLA-B*5701 allele in conferring the LTNP status.

HLA-B*5802, in contrast with HLA-B*5801, has been associated with fast disease progression (12). One polymorphism unique to this allele, not found in alleles associated with LTNP is W97 (Table IV⇑). This side chain is located in the center of the peptide-binding groove and peptide elution studies have found a preference for phenylalanine over tryptophan as a F anchor (23). The effect of the bulky aromatic W97 side chain may alter the structure of the tyrosine bed and the space available in the F-anchor pocket, thus altering the selection of the bound peptide repertoire. Because the 97W polymorphism is also found in HLA-B*5806 (Table IV⇑), this latter allele may exhibit similar functional properties of fast disease progression as HLA-B*5802.

Variations in the anchor-specifying residues in HLA-B*57 suballeles may have effects on the scope and stability of peptides that prompt immune responses conferring LTNP status. Future studies defining precise suballele association with LTNP will provide insight into the effects of polymorphic variations in selecting specific peptide repertoires and their relative abilities to control HIV-1 infection. The HLA-alleles of B*57, B*58, B*1516, B*1517, B*1567 appear to have very similar amino acid compositions that define the primary anchor and peptide-binding selectivity. It is possible that certain related HLA alleles exhibit even better control of HIV-1 but are at such low frequency in human populations that statistically valid associations with LTNP are lacking. A fuller analysis of HLA-associated LTNP by more detailed characterization of the scope of HLAs, degrees of association with LTNP, and characterization of presented epitopes may allow the identification of a super-set of residues that provide a clearer molecular correlate with peptide selection and association with LTNP.

Acknowledgments

We thank the staff of the Daresbury synchrotron facility for assistance with data collection and Simon Hubbard from Manchester University, U.K. for the provision of computing resources.

Disclosures

The authors have no financial conflict of interest.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

  • ↵1 The Medical Research Council (U.K.) funded researchers G.G., S.R.-J., A.J.M., and G.B.E.S.-J.; E.Y.J. is a Cancer Research U.K. Principal Research Fellow. R.K. is supported by a Tier 2 Chair from the Canada Research Chair Programme and I.M.O. is supported by the Biotechnology and Biological Sciences Research Council.

  • ↵2 Accession numbers. Atomic coordinates and structure factor amplitudes for the HLA-B*5703-KAF11, HLA-B*5703-ISP, and HLA-B*5703-KAF8 complexes have been deposited in the Protein Data Bank under accession codes 2bvo, 2bvp and 2bvq, respectively.

  • ↵3 G.B.E.S.-J. and G.G. contributed equally to this work.

  • ↵4 Address correspondence and reprint requests to Dr. E. Yvonne Jones, Division of Structural Biology, Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford OX3 7BN, U.K. E-mail address: yvonne{at}strubi.ox.ac.uk

  • ↵5 Abbreviations used in this paper: LTNP, long-term nonprogression; β2m, β2 microglobulin; SFU, spot-forming unit.

  • Received February 8, 2005.
  • Accepted June 6, 2005.

References

  1. ↵
    Borrow, P., H. Lewicki, B. H. Hahn, G. M. Shaw, M. B. Oldstone. 1994. Virus-specific CD8+ cytotoxic T-lymphocyte activity associated with control of viremia in primary human immunodeficiency virus type 1 infection. J. Virol. 68: 6103-6110.
    OpenUrlAbstract/FREE Full Text
  2. ↵
    Koup, R. A., J. T. Safrit, Y. Cao, C. A. Andrews, G. McLeod, W. Borkowsky, C. Farthing, D. D. Ho. 1994. Temporal association of cellular immune responses with the initial control of viremia in primary human immunodeficiency virus type 1 syndrome. J. Virol. 68: 4650-4655.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    Klein, M. R., C. A. van Baalen, A. M. Holwerda, S. R. Kerkhof Garde, R. J. Bende, I. P. Keet, J. K. Eeftinck-Schattenkerk, A. D. Osterhaus, H. Schuitemaker, F. Miedema. 1995. Kinetics of Gag-specific cytotoxic T lymphocyte responses during the clinical course of HIV-1 infection: a longitudinal analysis of rapid progressors and long-term asymptomatics. J. Exp. Med. 181: 1365-1372.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    Ogg, G. S., X. Jin, S. Bonhoeffer, P. R. Dunbar, M. A. Nowak, S. Monard, J. P. Segal, Y. Cao, S. L. Rowland-Jones, V. Cerundolo, et al 1998. Quantitation of HIV-1-specific cytotoxic T lymphocytes and plasma load of viral RNA. Science 279: 2103-2106.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    O’Brien, S. J., G. W. Nelson, C. A. Winkler, M. W. Smith. 2000. Polygenic and multifactorial disease gene association in man: lessons from AIDS. Annu. Rev. Genet 34: 563-591.
    OpenUrlCrossRefPubMed
  6. ↵
    McGrath, K. M., N. G. Hoffman, W. Resch, J. A. Nelson, R. Swanstrom. 2000. Using HIV-1 sequence variability to explore virus biology. Virus Res. 76: 137-160.
    OpenUrl
  7. ↵
    Klenerman, P., S. Rowland-Jones, S. McAdam, J. Edwards, S. Daenke, D. Lalloo, B. Koppe, W. Rosenberg, D. Boyd, A. A. Edwards, et al 1994. Cytotoxic T-cell activity antagonized by naturally occurring HIV-1 Gag variants. Nature 369: 403-407.
    OpenUrlCrossRefPubMed
  8. ↵
    Koenig, S., A. J. Conley, Y. A. Brewah, G. M. Jones, S. Leath, L. J. Boots, V. Davey, G. Pantaleo, J. F. Demarest, C. Carter, et al 1995. Transfer of HIV-1-specific cytotoxic T lymphocytes to an AIDS patient leads to selection for mutant HIV variants and subsequent disease progression. Nat. Med. 1: 330-336.
    OpenUrlCrossRefPubMed
  9. ↵
    Carrington, M., G. W. Nelson, M. P. Martin, T. Kissner, D. Vlahov, J. J. Goedert, R. Kaslow, S. Buchbinder, K. Hoots, S. J. O’Brien. 1999. HLA and HIV-1: heterozygote advantage and B*35-Cw*04 disadvantage. Science 283: 1748-1752.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Klein, M. R., I. P. Keet, J. D’Amaro, R. J. Bende, A. Hekman, B. Mesman, M. Koot, L. P. de Waal, R. A. Coutinho, F. Miedema. 1994. Associations between HLA frequencies and pathogenic features of human immunodeficiency virus type 1 infection in seroconverters from the Amsterdam cohort of homosexual men. J. Infect. Dis. 169: 1244-1249.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    Gao, X., G. W. Nelson, P. Karacki, M. P. Martin, J. Phair, R. Kaslow, J. J. Goedert, S. Buchbinder, K. Hoots, D. Vlahov, et al 2001. Effect of a single amino acid change in MHC class I molecules on the rate of progression to AIDS. N. Engl. J. Med. 344: 1668-1675.
    OpenUrlCrossRefPubMed
  12. ↵
    Kiepiela, P., A. J. Leslie, I. Honeyborne, D. Ramduth, C. Thobakgale, S. Chetty, P. Rathnavalu, C. Moore, K. J. Pfafferott, L. Hilton, et al 2004. Dominant influence of HLA-B in mediating the potential co-evolution of HIV and HLA. Nature 432: 769-775.
    OpenUrlCrossRefPubMed
  13. ↵
    Haynes, B. F., G. Pantaleo, A. S. Fauci. 1996. Toward an understanding of the correlates of protective immunity to HIV infection. Science 271: 324-328.
    OpenUrlAbstract
  14. ↵
    Kaslow, R. A., M. Carrington, R. Apple, L. Park, A. Munoz, A. J. Saah, J. J. Goedert, C. Winkler, S. J. O’Brien, C. Rinaldo, et al 1996. Influence of combinations of human major histocompatibility complex genes on the course of HIV-1 infection. Nat. Med. 2: 405-411.
    OpenUrlCrossRefPubMed
  15. ↵
    Costello, C., J. Tang, C. Rivers, E. Karita, J. Meizen-Derr, S. Allen, R. A. Kaslow. 1999. HLA-B*5703 independently associated with slower HIV-1 disease progression in Rwandan women. AIDS 13: 1990-1991.
    OpenUrlCrossRefPubMed
  16. ↵
    Kelleher, A. D., C. Long, E. C. Holmes, R. L. Allen, J. Wilson, C. Conlon, C. Workman, S. Shaunak, K. Olson, P. Goulder, et al 2001. Clustered mutations in HIV-1 gag are consistently required for escape from HLA-B27-restricted cytotoxic T lymphocyte responses. J. Exp. Med. 193: 375-386.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    Goulder, P. J., R. E. Phillips, R. A. Colbert, S. McAdam, G. Ogg, M. A. Nowak, P. Giangrande, G. Luzzi, B. Morgan, A. Edwards, et al 1997. Late escape from an immunodominant cytotoxic T-lymphocyte response associated with progression to AIDS. Nat. Med. 3: 212-217.
    OpenUrlCrossRefPubMed
  18. ↵
    Goulder, P. J., C. Brander, Y. Tang, C. Tremblay, R. A. Colbert, M. M. Addo, E. S. Rosenberg, T. Nguyen, R. Allen, A. Trocha, et al 2001. Evolution and transmission of stable CTL escape mutations in HIV infection. Nature 412: 334-338.
    OpenUrlCrossRefPubMed
  19. ↵
    Migueles, S. A., M. S. Sabbaghian, W. L. Shupert, M. P. Bettinotti, F. M. Marincola, L. Martino, C. W. Hallahan, S. M. Selig, D. Schwartz, J. Sullivan, M. Connors. 2000. HLA B*5701 is highly associated with restriction of virus replication in a subgroup of HIV-infected long term nonprogressors. Proc. Natl. Acad. Sci. USA 97: 2709-2714.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    Altfeld, M., M. M. Addo, E. S. Rosenberg, F. M. Hecht, P. K. Lee, M. Vogel, X. G. Yu, R. Draenert, M. N. Johnston, D. Strick, et al 2003. Influence of HLA-B*57 on clinical presentation and viral control during acute HIV-1 infection. AIDS 17: 2581-2591.
    OpenUrlCrossRefPubMed
  21. ↵
    Goulder, P. J., M. Bunce, P. Krausa, K. McIntyre, S. Crowley, B. Morgan, A. Edwards, P. Giangrande, R. E. Phillips, A. J. McMichael. 1996. Novel, cross-restricted, conserved, and immunodominant cytotoxic T lymphocyte epitopes in slow progressors in HIV type 1 infection. AIDS Res. Hum. Retroviruses 12: 1691-1698.
    OpenUrlCrossRefPubMed
  22. ↵
    Klein, M. R., S. H. van der Burg, E. Hovenkamp, A. M. Holwerda, J. W. Drijfhout, C. J. Melief, F. Miedema. 1998. Characterization of HLA-B*57-restricted human immunodeficiency virus type 1 Gag- and RT-specific cytotoxic T lymphocyte responses. J. Gen. Virol. 79: 2191-2201.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    Barber, L. D., L. Percival, K. L. Arnett, J. E. Gumperz, L. Chen, P. Parham. 1997. Polymorphism in the α1 helix of the HLA-B heavy chain can have an overriding influence on peptide-binding specificity. J. Immunol. 158: 1660-1669.
    OpenUrlAbstract
  24. ↵
    Goulder, P. J., Y. Tang, S. I. Pelton, B. D. Walker. 2000. HLA-B*57-restricted cytotoxic T-lymphocyte activity in a single infected subject toward two optimal epitopes, one of which is entirely contained within the other. J. Virol. 74: 5291-5299.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    Tussey, L. G., S. Rowland-Jones, T. S. Zheng, M. J. Androlewicz, P. Cresswell, J. A. Frelinger, A. J. McMichael. 1995. Different MHC class I alleles compete for presentation of overlapping viral epitopes. Immunity 3: 65-77.
    OpenUrlCrossRefPubMed
  26. ↵
    Otwinowski, Z., W. Minor. 1997. Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276: 307-326.
    OpenUrlCrossRef
  27. ↵
    Kissinger, C. R., D. K. Gehlhaar, D. B. Fogel. 1999. Rapid automated molecular replacement by evolutionary search. Acta Crystallogr. D. 55: 484-491.
    OpenUrlCrossRefPubMed
  28. ↵
    Brunger, A. T., P. D. Adams, G. M. Clore, W. L. DeLano, P. Gros, R. W. Grosse-Kunstleve, J. S. Jiang, J. Kuszewski, M. Nilges, N. S. Pannu, et al 1998. Crystallography and NMR system: a new software suite for macromolecular structure determination. Acta. Crystallogr. D Biol. Crystallogr 54: 905-921.
    OpenUrlCrossRefPubMed
  29. ↵
    Jones, T. A., J. Y. Zou, S. W. Cowan, S. W. Kjeldgaard. 1991. Improved methods for building protein models in electron density maps and the location of errors in these models. Acta. Crystallogr A. 47: 110-119.
    OpenUrlCrossRefPubMed
  30. ↵
    Morris, R. J., A. Perrakis, V. S. Lamzin. 2002. ARP/wARP’s model-building algorithms. I. The main chain. Acta. Crystallogr. D Biol. Crystallogr 58: 968-975.
    OpenUrlCrossRefPubMed
  31. ↵
    Murshudov, G. N.. 1997. Refinement of macromolecular structures by the maximum-likelihood method. Acta. Crystallogr. D Biol. Crystallogr 53: 240-255.
    OpenUrlCrossRefPubMed
  32. ↵
    Stuart, D. I., M. Levine, H. Muirhead, D. K. Stammers. 1979. Crystal structure of cat muscle pyruvate kinase at a resolution of 2.6 A. J. Mol. Biol. 134: 109-142.
    OpenUrlCrossRefPubMed
  33. ↵
    Hubbard, S. J., and J. M. Thornton. 1996. ‘NACCESS’ computer program version 2.1.1. University College London, London.
  34. ↵
    Esnouf, R. M.. 1999. Further additions to MolScript version 1.4, including reading and contouring of electron-density maps. Acta Crystallogr. D Biol. Crystallogr. 55: 938-940.
    OpenUrlCrossRefPubMed
  35. ↵
    Altschul, S. F., W. Gish, W. Miller, E. W. Myers, D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215: 403-410.
    OpenUrlCrossRefPubMed
  36. ↵
    Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller, D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25: 3389-3402.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    Bunce, M., C. M. O’Neill, M. C. Barnardo, P. Krausa, M. J. Browning, P. J. Morris, K. I. Welsh. 1995. Phototyping: comprehensive DNA typing for HLA-A, B, C, DRB1, DRB3, DRB4, DRB5 and DQB1 by PCR with 144 primer mixes utilizing sequence-specific primers (PCR-SSP). Tissue Antigens 46: 355-367.
    OpenUrlCrossRefPubMed
  38. ↵
    Madden, D. R.. 1995. The three-dimensional structure of peptide-MHC complexes. Annu. Rev. Immunol. 13: 587-622.
    OpenUrlCrossRefPubMed
  39. ↵
    Lee, J. K., G. Stewart-Jones, T. Dong, K. Harlos, K. Di Gleria, L. Dorrell, D. C. Douek, P. A. van der Merwe, E. Y. Jones, A. J. McMichael. 2004. T cell cross-reactivity and conformational changes during TCR engagement. J. Exp. Med. 200: 1455-1466.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    Kulikova, T., P. Aldebert, N. Althorpe, W. Baker, K. Bates, P. Browne, A. van den Broek, G. Cochrane, K. Duggan, R. R. Eberhardt, et al 2004. The EMBL nucleotide sequence database. Nucleic Acids Res. 32: D27-D30.
    OpenUrlAbstract/FREE Full Text
  41. ↵
    Apostolopoulos, V., M. Yu, A. L. Corper, W. Li, I. F. McKenzie, L. Teyton, I. A. Wilson, M. Plebanski. 2002. Crystal structure of a non-canonical high affinity peptide complexed with MHC class I: a novel use of alternative anchors. J. Mol. Biol. 318: 1307-1316.
    OpenUrlCrossRefPubMed
  42. ↵
    Apostolopoulos, V., M. Yu, A. L. Corper, L. Teyton, G. A. Pietersz, I. F. McKenzie, I. A. Wilson, M. Plebanski. 2002. Crystal structure of a non-canonical low-affinity peptide complexed with MHC class I: a new approach for vaccine design. J. Mol. Biol. 318: 1293-1305.
    OpenUrlCrossRefPubMed
  43. ↵
    Sliz, P., O. Michielin, J. C. Cerottini, I. Luescher, P. Romero, M. Karplus, D. C. Wiley. 2001. Crystal structures of two closely related but antigenically distinct HLA-A2/melanocyte-melanoma tumor-antigen peptide complexes. J. Immunol. 167: 3276-3284.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    Hillig, R. C., P. G. Coulie, V. Stroobant, W. Saenger, A. Ziegler, M. Hulsmeyer. 2001. High-resolution structure of HLA-A*0201 in complex with a tumour-specific antigenic peptide encoded by the MAGE-A4 gene. J. Mol. Biol. 310: 1167-1176.
    OpenUrlCrossRefPubMed
  45. ↵
    Smith, K. J., S. W. Reid, K. Harlos, A. J. McMichael, D. I. Stuart, J. I. Bell, E. Y. Jones. 1996. Bound water structure and polymorphic amino acids act together to allow the binding of different peptides to MHC class I HLA-B*53. Immunity 4: 215-228.
    OpenUrlCrossRefPubMed
  46. ↵
    Garboczi, D. N., D. R. Madden, D. C. Wiley. 1994. Five viral peptide-HLA-A2 co-crystals: simultaneous space group determination and X-ray data collection. J. Mol. Biol. 239: 581-587.
    OpenUrlCrossRefPubMed
  47. ↵
    Apostolopoulos, V., M. Yu, I. F. McKenzie, I. A. Wilson. 2000. Structural implications for the design of molecular vaccines. Curr. Opin. Mol. Ther. 2: 29-36.
    OpenUrlPubMed
  48. ↵
    Li, L., M. Bouvier. 2004. Structures of HLA-A*1101 complexed with immunodominant nonamer and decamer HIV-1 epitopes clearly reveal the presence of a middle, secondary anchor residue. J. Immunol. 172: 6175-6184.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    Garboczi, D. N., P. Ghosh, U. Utz, Q. R. Fan, W. E. Biddison, D. C. Wiley. 1996. Structure of the complex between human T-cell receptor, viral peptide and HLA-A2. Nature 384: 134-141.
    OpenUrlCrossRefPubMed
  50. ↵
    Ding, Y. H., K. J. Smith, D. N. Garboczi, U. Utz, W. E. Biddison, D. C. Wiley. 1998. Two human T cell receptors bind in a similar diagonal mode to the HLA-A2/Tax peptide complex using different TCR amino acids. Immunity 8: 403-411.
    OpenUrlCrossRefPubMed
  51. ↵
    Garcia, K. C., M. Degano, R. L. Stanfield, A. Brunmark, M. R. Jackson, P. A. Peterson, L. Teyton, I. A. Wilson. 1996. An αβ T cell receptor structure at 2.5 A and its orientation in the TCR-MHC complex. Science 274: 209-219.
    OpenUrlAbstract/FREE Full Text
  52. ↵
    Rudolph, M. G, I. A. Wilson. 2002. The specificity of TCR/pMHC interaction. Curr. Opin. Immunol. 14: 52-65.
    OpenUrlCrossRefPubMed
  53. ↵
    Goulder, P. J., S. W. Reid, D. A. Price, C. A. O’Callaghan, A. J. McMichael, R. E. Phillips, E. Y. Jones. 1997. Combined structural and immunological refinement of HIV-1 HLA-B8-restricted cytotoxic T lymphocyte epitopes. Eur. J. Immunol. 27: 1515-1521.
    OpenUrlCrossRefPubMed
  54. ↵
    Flower, D. R.. 2003. Towards in silico prediction of immunogenic epitopes. Trends Immunol. 24: 66-674.
    OpenUrl
  55. ↵
    Stewart-Jones, G. B., A. J. McMichael, J. I. Bell, D. I. Stuart, E. Y. Jones. 2003. A structural basis for immunodominant human T cell receptor recognition. Nat. Immunol. 4: 657
    OpenUrlCrossRefPubMed
  56. ↵
    Frahm, N., S. Adams, P. Kiepiela, C. H. Linde, H. S. Hewitt, M. Lichterfeld, K. Sango, N. V. Brown, E. Pae, A. G. Wurcel, et al. HLA-B63 presents HLA-B57/B58-restricted cytotoxic-T-lymphocyte epitopes and is associated with low human immunodeficiency virus load. J. Virol. In press.
View Abstract
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section