HLA-B63 Presents HLA-B57/B58-Restricted Cytotoxic
http://www.100md.com
病菌学杂志 2005年第16期
Partners AIDS Research Center, Massachusetts General Hospital, Charlestown, Massachusetts 02129
NIH Clinical Center, HLA Typing Laboratory, Bethesda, Maryland 20892
UHIV Pathogenesis Program, University of Natal, Durban, South Africa
Fenway Community Health Center, Boston, Massachusetts 02115
Lemuel Shattuck Hospital, Jamaica Plain, Massachusetts 02130
Queen Elizabeth Hospital, Bridgetown, Barbados
Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance, California 90502
Los Alamos National Laboratory, Los Alamos, New Mexico 87545, and Santa Fe Institute, Santa Fe, New Mexico 87501
ABSTRACT
Several HLA class I alleles have been associated with slow human immunodeficiency virus (HIV) disease progression, supporting the important role HLA class I-restricted cytotoxic T lymphocytes (CTL) play in controlling HIV infection. HLA-B63, the serological marker for the closely related HLA-B1516 and HLA-B1517 alleles, shares the epitope binding motif of HLA-B57 and HLA-B58, two alleles that have been associated with slow HIV disease progression. We investigated whether HIV-infected individuals who express HLA-B63 generate CTL responses that are comparable in breadth and specificity to those of HLA-B57/58-positive subjects and whether HLA-B63-positive individuals would also present with lower viral set points than the general population. The data show that HLA-B63-positive individuals indeed mounted responses to previously identified HLA-B57-restricted epitopes as well as towards novel, HLA-B63-restricted CTL targets that, in turn, can be presented by HLA-B57 and HLA-B58. HLA-B63-positive subjects generated these responses early in acute HIV infection and were able to control HIV replication in the absence of antiretroviral treatment with a median viral load of 3,280 RNA copies/ml. The data support an important role of the presented epitope in mediating relative control of HIV replication and help to better define immune correlates of controlled HIV infection.
INTRODUCTION
Human immunodeficiency virus (HIV) infection has been shown to induce strong virus-specific cytotoxic-T-lymphocyte (CTL) responses restricted by a large number of different HLA class I alleles (9). Among these alleles, some have been associated with slower or faster rates of HIV disease progression, suggesting that the presented epitopes or the restricting HLA class I molecule may play a central role in determining the ability to control viral replication (25, 33). The precise mechanisms by which these epitopes or the restricting HLA alleles mediate their beneficial or detrimental effects remain elusive.
HLA-B57 and HLA-B58 show strong similarities in their HLA allele binding motifs, and several of their suballeles have been associated with significantly slower HIV disease progression in a number of studies (17, 18, 24, 25, 33). In addition, several CTL epitopes presented by both of these alleles have been identified and are considered a major factor in the control of HIV replication (6, 9, 16, 20). Since HLA-B57-restricted CTL responses can be detected early in infection, it has also been suggested that these CTL specificities may be especially well suited to limit the initial viral replication, thereby conserving immune function (2, 19). The strong immune pressure against some HLA-B57 epitopes has also been found to select for specific viral escape variants, leading to, at times, dramatic increases in viral loads (6, 20). These data indicate that the presented epitope may play an important role in limiting viremia and mediating at least partial in vivo control (7, 14). Consequently, one might expect that different HLA class I alleles that can present an identical epitope(s) could potentially be associated with similar HIV disease outcome. As HLA-B57 and HLA-B58 present a comparable set of epitopes, this could explain why both of these alleles are associated with slower disease progression (9). However, HLA-B57 and HLA-B58 alleles also show extensive overall sequence similarities, and other molecular determinants shared by these two presenting HLA molecules could thus influence relative in vivo viral control (8, 23, 29).
HLA-B63 is the serological marker for several HLA-B15 alleles, including HLA-B1516 and HLA-B1517 and a few minor allelic variants. Through past recombination events, the HLA-B1516 and HLA-B1517 alleles have acquired sequence similarities with HLA-B57 and HLA-B58, especially within the 1 helix of the class I heavy chain. Thus, the 10 residues that are centrally involved in shaping the epitope binding B pocket (12, 30) are 100% conserved between HLA-B1516/1517 and HLA-B5701/5702, with an M45T change in both B5801 and B5802 relative to the other alleles (3). The residues forming the F pocket are similarly conserved, and the B63 and B57/58 alleles all favor epitopes with small aliphatic amino acids like alanine (A), serine (S), and threonine (T) at position 2 and the aromatic amino acids phenylalanine (F), tryptophan (W), or tyrosine (Y) at the C-terminal end (3, 9, 31). Based on these common characteristics, HLA-B63 alleles have been grouped into the "HLA-B58 supertype," together with HLA-B57 and HLA-B58 (21, 32). Although no examples of shared CTL epitopes between the HLA-B63 and the HLA-B57/58 alleles have been described, their essentially identical binding motifs suggest that at least some HLA-B57 and HLA-B58 epitopes could be presented on HLA-B63, and vice versa. However, even if epitopes are shared among the alleles in the HLA-B58 supertype, it is unclear whether these responses could also translate into a more favorable HIV disease outcome in individuals expressing HLA-B63.
The present study defined the immunodominant CTL targets in HLA-B63-expressing individuals and addressed whether HLA-B63 can present previously described HLA-B57- and HLA-B58-restricted HIV epitopes and whether HLA-B63 is associated with reduced HIV viral loads in the absence of antiretroviral treatment. Our data show that HIV-infected individuals who express HLA-B63 do, in fact, target previously described HLA-B57 epitopes as well as longer, overlapping peptides containing known or potential new HLA-B63/B57/58 epitopes. Importantly, individuals who expressed HLA-B63 presented with reduced viral loads compared to the remainder of the study cohort. The present study therefore identifies a group of HLA class I alleles that share CTL epitopes and confer relative protection from HIV disease progression, suggesting that the nature of their shared CTL epitopes may be of significant relevance in providing this effect.
MATERIALS AND METHODS
Study subjects. HIV-infected individuals were recruited from HIV clade B-infected cohorts in the United States and an HIV clade C-infected cohort in Durban, South Africa. The HIV clade B-infected cohort included subjects from several ethnicities enrolled at four hospitals in the Boston area, the Harbor-UCLA Medical Center in Torrance, Calif., and the Queen Elizabeth Hospital in Bridgetown, Barbados (10). The HIV clade C-infected cohort included subjects of the Zulu and Xhosa ethnicities recruited at two clinics in Durban, South Africa (18). Infecting clade information and HLA typing were obtained as described previously (10). Clade C-infected individuals were included in this study for the comparison of viral loads only. Insufficient sample prevented a four-digit subtyping of the HLA-B63 allele in seven subjects. Subject characteristics are summarized in Table 1.
Viral load determinations. Viral loads were determined using the Roche Amplicor HIV-1 Monitor system with a lower detection limit of 400 RNA copies/ml. For samples that fell below this threshold, the ultrasensitive assay with a detection limit of 50 RNA copies/ml was applied where sufficient material was available.
Lymphocyte separation and use in ELISPOT assays. Peripheral blood mononuclear cells (PBMCs) were separated from whole blood and used fresh or frozen to test for CTL responses against previously defined HLA-B57-restricted CTL epitopes or a set of 410 overlapping 18-mer peptides spanning the entire HIV genome (9, 10). Gamma interferon enzyme-linked immunospot (ELISPOT) assays were performed as described in the past using Mabtech (Stockholm, Sweden) antibodies (10). The number of spots was determined using the AID ELISPOT Reader Unit (Autoimmun Diagnostika GmbH, Strassberg, Germany), and results were expressed as spot-forming cells (SFC) per million input cells. Thresholds for positive responses were determined as at least 5 spots (50 SFC/106 cells) per well and responses exceeding "mean of negative wells plus three standard deviations" and "three times mean," whichever was higher.
Epitope fine mapping was performed in peptide titration assays using truncated peptides in serial 10-fold dilutions ranging from 100 μg/ml to 10 pg/ml. All cellular assays were performed using samples from clade B-infected subjects and a clade B consensus sequence-based peptide set (10).
Statistical analysis. Statistical analyses were performed using GraphPad Prism version 3.0 for Macintosh. All tests were two tailed. Bonferroni's correction was applied for multiple tests. Viral load differences between groups were determined using the nonparametric Mann-Whitney U test. All viral load values are represented as medians. Viral load values below the limit of detection of 50 RNA copies/ml were assigned a value of 49 for statistical analyses, and for values with a detection limit of 400 copies/ml, a value of 399 was used. Viral loads >750,000 copies/ml were assigned a value of 750,001.
RESULTS
HLA-B63-expressing individuals frequently target HLA-B57-restricted epitopes. In order to investigate whether HLA-B63-positive individuals show a CTL response pattern similar to that of individuals expressing HLA-B57 or HLA-B58, HLA-B63-expressing subjects were tested for their total HIV-specific CTL response and for their ability to target previously described HLA-B57-restricted CTL epitopes. Due to the low allele frequencies of HLA-B1516 and HLA-B1517 (less than 1% in Caucasians and less than 1.5% in African Americans [5]) compared to those of HLA-B57 (about 4% in both ethnicities) and HLA-B58 (1% in Caucasians and 10% in African Americans [5]), sufficient cells from only 10 HIV clade B-infected, HLA-B63-positive individuals were available to test for total HIV responses using an overlapping peptide (OLP) set spanning the entire expressed HIV genome.
In a first analysis, the frequency of recognition among the HLA-B63-expressing individuals was determined for OLPs containing known HLA-B57 epitopes and compared to their overall recognition frequency in a cohort of 254 subjects tested comprehensively. Table 2 lists seven OLPs containing known HLA-B57 epitopes that were frequent targets of the HIV-specific immune response among these 10 subjects. Remarkably, six of the seven peptides were recognized at higher frequencies in HLA-B63-positive subjects than in HLA-B57/58-positive individuals. Although the HLA-B63-expressing individuals targeted all these OLPs more frequently than the remainder of the cohort, statistical significance was achieved only for two OLPs, mainly due to the presence of numerous epitopes restricted by HLA alleles different from HLA-B63 or HLA-B57/58 in the other, frequently targeted peptide sequences (9, 10). To establish that the HLA-B63-positive individuals did indeed target the same epitopes presented by HLA-B57 and HLA-B58, and not other potential epitopes contained in these OLPs, the described HLA-B57 optimal epitopes and truncations thereof were tested in HLA-B63-expressing individuals. These subjects showed strong responses (median, 379 SFC/106 PBMCs; range, 50 to >3,000 SFC/106 PBMCs) against the previously described optimal HLA-B57-restricted epitopes. Titration analyses of single-residue truncated peptide sequences also confirmed that the identical optimal epitopes were targeted in HLA-B63-positive subjects as in HLA-B57-expressing individuals (Fig. 1).
Frequently targeted epitopes in HLA-B63-positive subjects contain putative HLA-B1516- and HLA-B1517-restricted epitopes. The initial analysis of OLP recognition identified additional OLPs that were preferentially targeted by HLA-B63-positive individuals but did not contain previously known HLA-B57 epitopes. Table 3 includes all OLPs that were targeted by at least 30% of the HLA-B63-positive group. Of these 11 OLPs, 8 showed markedly increased frequencies of recognition compared to the HLA-B58 supertype-negative control group (P < 0.05). The remaining three OLPs were located in HIV Nef and were frequently targeted by the overall cohort, rendering the P value of these comparisons insignificant (10). Importantly, each of these 11 OLPs contained a putative or, for OLP 338, a previously defined HLA-B63-restricted epitope that fulfilled the HLA-B58 supertype-specific binding motif. Notably, this included epitopes with phenylalanine in second position, a residue which is not included in the described B58 supertype motif (21, 32). However, the presence of phenylalanine in the anchor position of some epitopes included in the present report, as well as among HLA-B57-, HLA-B58-, and HLA-B1516-restricted epitopes listed in the SYFPEITHI database (28), suggests that the HLA-B58 supertype motif could be expanded to include this additional residue.
HLA-B57- and HLA-B58-positive individuals recognize new B63 epitopes. To demonstrate not only that HLA-B63-expressing individuals target known HLA-B57-restricted CTL epitopes but that, in turn, HLA-B57/58-positive individuals can also target HLA-B63 epitopes, the frequency of recognition of the 11 OLPs in Table 3 was analyzed in B57/58-positive subjects. Indeed, five of the OLPs preferentially targeted in B63-positive individuals were also favored in B57/58-positive subjects (P < 0.05) (Table 3), strongly suggesting that these OLPs contained HLA-B63 epitopes that can be shared by HLA-B57/58 alleles. To assert that both groups really targeted identical epitopes, standard peptide titration assays were performed to define the optimal length of epitopes shared between HLA-B63 and HLA-B57/58. For six OLPs listed in Table 3, sufficient numbers of OLP responders and samples were available to determine the optimal epitope length. The data for five epitopes targeted by HLA-B63- or HLA-B57/58-positive individuals are shown in Fig. 2. Included in this figure are epitopes LL9, to illustrate presentation in HLA-B1517- and HLA-B57-expressing individuals with similar functional avidity; epitope RY11 (RSLYNTVATLY), embedded in OLP 11 and containing the known, but in this case nontargeted, HLA-A02-restricted SL9 epitope (SLYNTVATL); and epitopes YY9 (OLP 84), FF9 (OLP 23), and KV8 (OLP 162), which were identified based on their frequent recognition in HLA-B63-expressing subjects but which are also targeted by HLA-B57 individuals.
Together, the data identify novel HLA-B63 epitopes that are also frequently targeted in HLA-B57/58-positive subjects and confirm that epitopes previously described as HLA-B57 epitopes can also be recognized in HLA-B63-positive subjects. These included HLA-B57 epitopes of considerable interindividual immunodominance and for which an important role in the control of HIV replication has been proposed in the past (e.g., KF11 in OLP 22 and TW10 in OLP 33) (16, 20). Given the similarities in the functional avidity of these responses and the frequent recognition of novel HLA-B63 epitopes in HLA-B57/58-expressing individuals, the data suggest that CTL responses against these shared epitopes could mediate effective immune control in vivo, potentially leading to decreased viral set points in HLA-B63-positive individuals.
Novel B63 epitopes are targeted early in infection. As CTL responses to HLA-B57-restricted epitopes have been shown to be induced early in acute HIV infection (2), we assessed responses to the optimal epitopes emerging from the above-described analyses in two subjects during primary HIV infection. Subjects B19 and B20, tested before and 2 months after seroconversion, respectively, showed a pronounced immunodominance in the HLA-B63 epitopes compared to other known epitopes restricted by their respective HLA alleles. Subject B19 recognized 3 of 8 tested HLA-B63 epitopes and only 1 of 21 epitopes potentially restricted by other class I alleles (HLA-A01, HLA-A26, HLA-B35, and HLA-Cw04), with the magnitude against HLA-B63 epitopes exceeding 90% of the total response (890 of 980 SFC/million PBMCs) (Fig. 3a). Similarly, subject B20 recognized two epitopes, 1 of 8 HLA-B63 epitopes and 1 of 10 HLA-A01, HLA-A26, HLA-B39, and HLA-Cw07 epitopes, with the HLA-B63 response more than three times stronger than the latter (620 versus 180 SFC/million PBMCs) (Fig. 3b). These data demonstrate that immunodominant responses against HLA-B63-restricted epitopes are detectable in primary HIV infection, comparable to the early targeting of HLA-B57 epitopes (2), and suggest that they might play an important role in containing initial viral replication.
HLA-B63-expressing individuals present with reduced HIV viral loads. Given the shared set of optimal epitopes targeted in individuals expressing HLA-B63 or HLA-B57/58, the early induction of these responses upon HIV infection, and the important role that HLA-B57-restricted responses play in natural infection, the clinical outcome of HIV infection may be comparable between HLA-B63- and HLA-B57/58-expressing subjects. To test this hypothesis, HLA-B63-positive individuals not receiving antiretroviral therapy were identified among HIV clade B- and clade C-infected individuals enrolled in cohorts in Boston and Durban (Table 1), and their viral set points were analyzed. High-resolution HLA typing was available for 20 individuals, showing that all the individuals enrolled in Durban expressed the HLA-B1516 allele, whereas 9/13 individuals in the Boston cohort expressed HLA-B1516 and 4/13 were HLA-B1517 positive. Of the 27 individuals expressing either subtype of HLA-B63, 12 untreated HIV clade B-infected individuals were included in the subsequent analysis. All seven subjects in the clade C-infected cohort were treatment nave.
When untreated HLA-B63-expressing individuals were compared to the remainder of the cohort, a significantly lower viral load was observed in the HLA-B63-positive subjects than in the HLA-B63-negative subjects (P = 0.0002; median viral loads, 3,280 versus 32,500 RNA copies/ml) (Fig. 4). This marked control of viral replication is comparable to the one observed in untreated HLA-B57-positive subjects in our cohort (P < 0.0001; median viral loads, 6,645 versus 34,650 RNA copies/ml) (data not shown). The association remained statistically significant when clade B- and clade C-infected individuals were analyzed separately (P = 0.042 [3,912 versus 13,000 RNA copies/ml for clade B] and P = 0.029 [3,280 versus 35,550 RNA copies/ml for clade C]) (data not shown). Furthermore, excluding HLA-B57- or HLA-B58-coexpressing subjects from the analysis did not affect the outcome (P = 0.0002 for exclusion of HLA-B57-positive subjects and P = 0.0003 for exclusion of HLA-B58-positive subjects). The separate analysis of viral set points in the individuals expressing either the HLA-B1516 or HLA-B1517 subtype also reached statistical significance for HLA-B1516 (P = 0.0008; 4,850 versus 31,700 RNA copies/ml) but not for HLA-B1517, as viral load data were only available for two untreated, HLA-B1517-positive individuals. Collectively, the data demonstrate that HLA-B63 is associated with reduced viral load and, by inference from other published studies, with slower HIV disease progression (22). However, the small number of HLA-B63-expressing individuals for whom "time since infection" could be reliably documented did not allow us to directly assess the rate of HIV disease progression in the present cohort.
DISCUSSION
The data presented in this report identify HLA-B63 as a novel HLA class I allele associated with reduced HIV viral loads in chronic, untreated HIV-1 clade B and clade C infection. Since HLA-B63-expressing individuals mount strong immune responses against previously described HLA-B57-restricted epitopes, and given that HLA-B57/58-expressing subjects also show strong CTL activity against HLA-B63 epitopes, the data indicate that epitopes shared by these alleles may be central in mediating the beneficial effects associated with these alleles.
The ability of HLA-B57 and HLA-B58 to present identical optimal epitopes has been known for many years (13), but no HLA-B57 epitopes have been shown thus far to be presented by other members of the B58 supertype. However, sequence analyses of the residues forming the F pocket of HLA-B1516 strongly suggest that at least some HLA-B57 epitopes could be accommodated in HLA-B1516 (3, 27). Indeed, as reported here, HLA-B57 epitopes containing any of the aromatic amino acids as the C-terminal anchor were targeted in HLA-B63-positive individuals, and titration studies using serially truncated peptides show that the same optimal epitopes are being targeted by these subjects. This is further supported by data derived from the crystallographic structure of HLA-B5703 and bioinformatics analyses of related HLA sequences, which suggest that HLA-B63 alleles could cross-present optimal HLA-B57 epitopes. In addition, the newly described HLA-B63 epitopes are also frequently targeted in HLA-B57- or HLA-B58-positive subjects, further indicating that the same optimal epitopes are frequently shared between the different alleles of the B58 supertype.
A number of HLA class I alleles have been associated with slow HIV disease progression, but the relative contribution of the presenting allele versus the restricted epitopes towards this relative protection remains unclear. As the HLA-B58 supertype binding motif has been acquired by the HLA-B63 allele through a recombination event between HLA-B57/58 and HLA-B15 alleles, the sequence similarities between HLA-B63 and HLA-B57/58 are mainly restricted to the 1-helix region. Thus, whereas high overall sequence similarities between HLA-B57 and HLA-B58 do not allow dissection of the relative contribution of the presenting HLA allele or the presented epitopes on the in vivo efficacy of these CTLs, the case of HLA-B63 represents a situation where nature may have provided an opportunity to elucidate these factors in more detail. Although it cannot be ruled out that sequence similarities among the B58 supertype alleles at T-cell receptor contact sites outside the 1 helix are responsible for their common ability to provide relative protection from HIV disease progression, the present data strongly suggest that the nature of the targeted epitope is playing an important role in providing in vivo control (23). Such superior immune control may be due to better recognition of potential viral epitope variants and their increased cost to viral replication fitness (15). However, the level of variability among the epitopes shown in Tables 2 and 3 did not indicate particularly high conservation of these sequences (which would have indicated a high fitness cost for CTL escape), except for the two potential epitopes embedded in the highly conserved protease protein. Instead, high functional avidity and broad T-cell-receptor repertoires may contribute to more effective in vivo immune control mediated by these responses (16, 26) and, together with possible functional constraints that prevent rapid CTL escape or lead to similarly rapid reversion of escape variants, may override the potential effects of the presenting HLA class I molecule (1, 11, 20). However, it remains to be seen how far this observation holds true for other HLA class I alleles and epitopes, especially for epitopes that could be presented by alleles differentially associated with slow and fast HIV disease progression. In the case of HLA-B58, minute sequence changes between HLA-B5801 and HLA-B5802 are clinically associated with different HIV disease outcomes (18). This is also reflected by different CTL response patterns seen in subjects expressing either subtype, as fewer HLA-B5802-expressing individuals than HLA-B5801-expressing individuals responded to the frequently targeted OLPs shown in Tables 2 and 3, although HLA-B5802 was twice as frequently found in the clade C cohort compared to HLA-B5801. Thus, targeting the epitopes described in this report seems to be providing a selective advantage to the HLA-B5801-expressing individuals, as there was only restricted sharing of these epitopes between subjects expressing closely related alleles. Similar observations have been made for other closely related alleles, including HLA-B35 and its subtypes, which are associated with different rates of HIV disease progression. However, the link to CTL response patterns and the induction of qualitatively different responses depending on the presenting allele has thus far not been made and will require more intensive characterization of responses in individuals expressing closely related alleles. Eventually, the determination of the relative contribution of the presenting HLA class I allele versus the presented epitopes, and the identification of epitopes, rather than HLA alleles, associated with slow HIV disease progression, will further our understanding of antiviral immunity and facilitate the design of effective prophylactic and therapeutic immune interventions for HIV infection.
ACKNOWLEDGMENTS
This work has been funded by NIH contract N01-Al-15422, by the Los Alamos National Laboratory (la-ur-03-5892), and by NIH grants AI-043638 and AI-067077.
We gratefully acknowledge the helpful discussions with and guidance by Patricia D'Souza for this project.
REFERENCES
Allen, T. M., M. Altfeld, X. G. Yu, K. M. O'Sullivan, M. Lichterfeld, S. Le Gall, M. John, B. R. Mothe, P. K. Lee, E. T. Kalife, D. E. Cohen, K. A. Freedberg, D. A. Strick, M. N. Johnston, A. Sette, E. S. Rosenberg, S. A. Mallal, P. J. Goulder, C. Brander, and B. D. Walker. 2004. Selection, transmission, and reversion of an antigen-processing cytotoxic T-lymphocyte escape mutation in human immunodeficiency virus type 1 infection. J. Virol. 78:7069-7078.
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, T. M. Allen, M. E. Feeney, J. O. Kahn, R. P. Sekaly, J. A. Levy, J. K. Rockstroh, P. J. Goulder, and B. D. Walker. 2003. Influence of HLA-B57 on clinical presentation and viral control during acute HIV-1 infection. AIDS 17:2581-2591.
Barber, L. D., L. Percival, K. L. Arnett, J. E. Gumperz, L. Chen, and P. Parham. 1997. Polymorphism in the alpha 1 helix of the HLA-B heavy chain can have an overriding influence on peptide-binding specificity. J. Immunol. 158:1660-1669.
Brander, C., and B. D. Walker. 1996. The HLA class I restricted CTL response in HIV infection: systematic identification of optimal epitopes, p. IV50-IV60. In C. B. B. Korber, B. Walker, R. Koup, J. Moore, B. Haynes, and G. Meyers (ed.), HIV molecular immunology database. Theoretical biology and biophysics. Los Alamos National Laboratory, Los Alamos, N.M.
Cao, K., J. Hollenbach, X. Shi, W. Shi, M. Chopek, and M. A. Fernandez-Vina. 2001. Analysis of the frequencies of HLA-A, B, and C alleles and haplotypes in the five major ethnic groups of the United States reveals high levels of diversity in these loci and contrasting distribution patterns in these populations. Hum. Immunol. 62:1009-1030.
Draenert, R., S. Le Gall, K. J. Pfafferott, A. J. Leslie, P. Chetty, C. Brander, E. C. Holmes, S. C. Chang, M. E. Feeney, M. M. Addo, L. Ruiz, D. Ramduth, P. Jeena, M. Altfeld, S. Thomas, Y. Tang, C. L. Verrill, C. Dixon, J. G. Prado, P. Kiepiela, J. Martinez-Picado, B. D. Walker, and P. J. Goulder. 2004. Immune selection for altered antigen processing leads to cytotoxic T lymphocyte escape in chronic HIV-1 infection. J. Exp. Med. 199:905-915.
Feeney, M. E., Y. Tang, K. A. Roosevelt, A. J. Leslie, K. McIntosh, N. Karthas, B. D. Walker, and P. J. R. Goulder. 2004. Immune escape precedes breakthrough human immunodeficiency virus type 1 viremia and broadening of the cytotoxic T-lymphocyte response in an HLA-B27-positive long-term-nonprogressing child. J. Virol. 78:8927-8930.
Flores-Villanueva, P. O., E. J. Yunis, J. C. Delgado, E. Vittinghoff, S. Buchbinder, J. Y. Leung, A. M. Uglialoro, O. P. Clavijo, E. S. Rosenberg, S. A. Kalams, J. D. Braun, S. L. Boswell, B. D. Walker, and A. E. Goldfeld. 2001. Control of HIV-1 viremia and protection from AIDS are associated with HLA-Bw4 homozygosity. Proc. Natl. Acad. Sci. USA 98:5140-5145.
Frahm, N., P. J. R. Goulder, and C. Brander. 2003. Broad HIV-1 specific CTL responses reveal extensive HLA class I binding promiscuity of HIV-derived, optimally defined CTL epitopes, p. 3-24. In B. T. Korber, C. Brander, B. F. Haynes, R. Koup, J. P. Moore, B. D. Walker, and D. I. Watkins (ed.), HIV immunology and HIV/SIV vaccine databases 2003, vol. 1. Theoretical biology and biophysics. Los Alamos National Laboratory, Los Alamos, N.M.
Frahm, N., B. T. Korber, C. M. Adams, J. J. Szinger, R. Draenert, M. M. Addo, M. E. Feeney, K. Yusim, K. Sango, N. V. Brown, D. SenGupta, A. Piechocka-Trocha, T. Simonis, F. M. Marincola, A. G. Wurcel, D. R. Stone, C. J. Russell, P. Adolf, D. Cohen, T. Roach, A. St. John, A. Khatri, K. Davis, J. Mullins, P. J. Goulder, B. D. Walker, and C. Brander. 2004. Consistent cytotoxic-T-lymphocyte targeting of immunodominant regions in human immunodeficiency virus across multiple ethnicities. J. Virol. 78:2187-2200.
Friedrich, T. C., E. J. Dodds, L. J. Yant, L. Vojnov, R. Rudersdorf, C. Cullen, D. T. Evans, R. C. Desrosiers, B. R. Mothe, J. Sidney, A. Sette, K. Kunstman, S. Wolinsky, M. Piatak, J. Lifson, A. L. Hughes, N. Wilson, D. H. O'Connor, and D. I. Watkins. 2004. Reversion of CTL escape-variant immunodeficiency viruses in vivo. Nat. Med. 10:275-281.
Garrett, T. P., M. A. Saper, P. J. Bjorkman, J. L. Strominger, and D. C. Wiley. 1989. Specificity pockets for the side chains of peptide antigens in HLA-Aw68. Nature 342:692-696.
Goulder, P. J., M. Bunce, P. Krausa, K. McIntyre, S. Crowley, B. Morgan, A. Edwards, P. Giangrande, R. E. Phillips, and 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. Retrovir. 12:1691-1698.
Goulder, P. J., R. E. Phillips, R. A. Colbert, S. McAdam, G. Ogg, M. A. Nowak, P. Giangrande, G. Luzzi, B. Morgan, A. Edwards, A. J. McMichael, and S. Rowland-Jones. 1997. Late escape from an immunodominant cytotoxic T-lymphocyte response associated with progression to AIDS. Nat. Med. 3:212-217.
Goulder, P. J., and D. I. Watkins. 2004. HIV and SIV CTL escape: implications for vaccine design. Nat. Rev. Immunol. 4:630-640.
Jansen, C. A., S. Kostense, K. Vandenberghe, N. M. Nanlohy, I. M. Cuyper, E. Piriou, E. H. Manting, F. Miedema, and D. van Baarle. 2005. High responsiveness of HLA-B57-restricted Gag-specific CD8(+) T cells in vitro may contribute to the protective effect of HLA-B57 in HIV-infection. Eur. J. Immunol. 35:150-158.
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, R. Detels, W. Blattner, J. Phair, H. Erlich, and D. L. Mann. 1996. Influence of combinations of human major histocompatibility complex genes on the course of HIV-1 infection. Nat. Med. 2:405-411.
Kiepiela, P., A. J. Leslie, I. Honeyborne, D. Ramduth, C. Thobakgale, S. Chetty, P. Rathnavalu, C. Moore, K. J. Pfafferott, L. Hilton, P. Zimbwa, S. Moore, T. Allen, C. Brander, M. M. Addo, M. Altfeld, I. James, S. Mallal, M. Bunce, L. D. Barber, J. Szinger, C. Day, P. Klenerman, J. Mullins, B. Korber, H. M. Coovadia, B. D. Walker, and P. J. Goulder. 2004. Dominant influence of HLA-B in mediating the potential co-evolution of HIV and HLA. Nature 432:769-775.
Koup, R. A., J. T. Safrit, Y. Cao, C. A. Andrews, G. McLeod, W. Borkowsky, C. Farthing, and 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.
Leslie, A. J., K. J. Pfafferott, P. Chetty, R. Draenert, M. M. Addo, M. Feeney, Y. Tang, E. C. Holmes, T. Allen, J. G. Prado, M. Altfeld, C. Brander, C. Dixon, D. Ramduth, P. Jeena, S. A. Thomas, A. St. John, T. A. Roach, B. Kupfer, G. Luzzi, A. Edwards, G. Taylor, H. Lyall, G. Tudor-Williams, V. Novelli, J. Martinez-Picado, P. Kiepiela, B. D. Walker, and P. J. Goulder. 2004. HIV evolution: CTL escape mutation and reversion after transmission. Nat. Med. 10:282-289.
Lund, O., M. Nielsen, C. Kesmir, A. G. Petersen, C. Lundegaard, P. Worning, C. Sylvester-Hvid, K. Lamberth, G. Roder, S. Justesen, S. Buus, and S. Brunak. 2004. Definition of supertypes for HLA molecules using clustering of specificity matrices. Immunogenetics 55:797-810.
Mellors, J. W., C. R. Rinaldo, Jr., P. Gupta, R. M. White, J. A. Todd, and L. A. Kingsley. 1996. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 272:1167-1170.
Messaoudi, I., J. A. G. Patino, R. Dyall, J. LeMaoult, and J. Nikolich-Zugich. 2002. Direct link between mhc polymorphism, T cell avidity, and diversity in immune defense. Science 298:1797-1800.
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, and M. Connors. 2000. HLA B5701 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.
O'Brien, S. J., X. Gao, and M. Carrington. 2001. HLA and AIDS: a cautionary tale. Trends Mol. Med. 7:379-381.
Price, D. A., S. M. West, M. R. Betts, L. E. Ruff, J. M. Brenchley, D. R. Ambrozak, Y. Edghill-Smith, M. J. Kuroda, D. Bogdan, K. Kunstman, N. L. Letvin, G. Franchini, S. M. Wolinsky, R. A. Koup, and D. C. Douek. 2004. T cell receptor recognition motifs govern immune escape patterns in acute SIV infection. Immunity 21:793-803.
Prilliman, K. R., K. W. Jackson, M. Lindsey, J. Wang, D. Crawford, and W. H. Hildebrand. 1999. HLA-B15 peptide ligands are preferentially anchored at their C termini. J. Immunol. 162:7277-7284.
Rammensee, H., J. Bachmann, N. P. Emmerich, O. A. Bachor, and S. Stevanovic. 1999. SYFPEITHI: database for MHC ligands and peptide motifs. Immunogenetics 50:213-219.
Robinson, J., M. J. Waller, P. Parham, N. de Groot, R. Bontrop, L. J. Kennedy, P. Stoehr, and S. G. Marsh. 2003. IMGT/HLA and IMGT/MHC: sequence databases for the study of the major histocompatibility complex. Nucleic Acids Res. 31:311-314.
Saper, M. A., P. J. Bjorkman, and D. C. Wiley. 1991. Refined structure of the human histocompatibility antigen HLA-A2 at 2.6 A resolution. J. Mol. Biol. 219:277-319.
Seeger, F. H., D. Arnold, T. Dumrese, H. de la Salle, D. Fricker, H. Schild, H. G. Rammensee, and S. Stevanovic. 1998. The HLA-B 1516 motif demonstrates HLA-B-specific P2 pocket characteristics. Immunogenetics 48:156-160.
Sette, A., and J. Sidney. 1999. Nine major HLA class I supertypes account for the vast preponderance of HLA-A and -B polymorphism. Immunogenetics 50:201-212.
Stewart-Jones, G. B. E., G. Gillespie, I. M. Overton, P. Roche, A. J. McMichael, S. Rowland-Jones, and E. Y. Jones. 2005. Structures of three HIV-1 HLA-B5703-peptide complexes and identification of related HLAs potentially associated with long-term non-progression. J. Immunol., in press.
Trachtenberg, E. A., and H. A. Erlich. 2001. A review of the role of the human leukocyte antigen (HLA) system as a host immunogenic factor influencing HIV transmission and progression to AIDS, p. I-43-60. In B. T. K. Korber, C. Brander, B. F. Haynes, J. P. Moore, R. A. Koup, C. Kuiken, B. D. Walker, and D. I. Watkins (ed.), HIV molecular immunology 2001. Theoretical Biology and Biophysics Group, Los Alamos, N.M.(Nicole Frahm, Sharon Adam)
NIH Clinical Center, HLA Typing Laboratory, Bethesda, Maryland 20892
UHIV Pathogenesis Program, University of Natal, Durban, South Africa
Fenway Community Health Center, Boston, Massachusetts 02115
Lemuel Shattuck Hospital, Jamaica Plain, Massachusetts 02130
Queen Elizabeth Hospital, Bridgetown, Barbados
Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance, California 90502
Los Alamos National Laboratory, Los Alamos, New Mexico 87545, and Santa Fe Institute, Santa Fe, New Mexico 87501
ABSTRACT
Several HLA class I alleles have been associated with slow human immunodeficiency virus (HIV) disease progression, supporting the important role HLA class I-restricted cytotoxic T lymphocytes (CTL) play in controlling HIV infection. HLA-B63, the serological marker for the closely related HLA-B1516 and HLA-B1517 alleles, shares the epitope binding motif of HLA-B57 and HLA-B58, two alleles that have been associated with slow HIV disease progression. We investigated whether HIV-infected individuals who express HLA-B63 generate CTL responses that are comparable in breadth and specificity to those of HLA-B57/58-positive subjects and whether HLA-B63-positive individuals would also present with lower viral set points than the general population. The data show that HLA-B63-positive individuals indeed mounted responses to previously identified HLA-B57-restricted epitopes as well as towards novel, HLA-B63-restricted CTL targets that, in turn, can be presented by HLA-B57 and HLA-B58. HLA-B63-positive subjects generated these responses early in acute HIV infection and were able to control HIV replication in the absence of antiretroviral treatment with a median viral load of 3,280 RNA copies/ml. The data support an important role of the presented epitope in mediating relative control of HIV replication and help to better define immune correlates of controlled HIV infection.
INTRODUCTION
Human immunodeficiency virus (HIV) infection has been shown to induce strong virus-specific cytotoxic-T-lymphocyte (CTL) responses restricted by a large number of different HLA class I alleles (9). Among these alleles, some have been associated with slower or faster rates of HIV disease progression, suggesting that the presented epitopes or the restricting HLA class I molecule may play a central role in determining the ability to control viral replication (25, 33). The precise mechanisms by which these epitopes or the restricting HLA alleles mediate their beneficial or detrimental effects remain elusive.
HLA-B57 and HLA-B58 show strong similarities in their HLA allele binding motifs, and several of their suballeles have been associated with significantly slower HIV disease progression in a number of studies (17, 18, 24, 25, 33). In addition, several CTL epitopes presented by both of these alleles have been identified and are considered a major factor in the control of HIV replication (6, 9, 16, 20). Since HLA-B57-restricted CTL responses can be detected early in infection, it has also been suggested that these CTL specificities may be especially well suited to limit the initial viral replication, thereby conserving immune function (2, 19). The strong immune pressure against some HLA-B57 epitopes has also been found to select for specific viral escape variants, leading to, at times, dramatic increases in viral loads (6, 20). These data indicate that the presented epitope may play an important role in limiting viremia and mediating at least partial in vivo control (7, 14). Consequently, one might expect that different HLA class I alleles that can present an identical epitope(s) could potentially be associated with similar HIV disease outcome. As HLA-B57 and HLA-B58 present a comparable set of epitopes, this could explain why both of these alleles are associated with slower disease progression (9). However, HLA-B57 and HLA-B58 alleles also show extensive overall sequence similarities, and other molecular determinants shared by these two presenting HLA molecules could thus influence relative in vivo viral control (8, 23, 29).
HLA-B63 is the serological marker for several HLA-B15 alleles, including HLA-B1516 and HLA-B1517 and a few minor allelic variants. Through past recombination events, the HLA-B1516 and HLA-B1517 alleles have acquired sequence similarities with HLA-B57 and HLA-B58, especially within the 1 helix of the class I heavy chain. Thus, the 10 residues that are centrally involved in shaping the epitope binding B pocket (12, 30) are 100% conserved between HLA-B1516/1517 and HLA-B5701/5702, with an M45T change in both B5801 and B5802 relative to the other alleles (3). The residues forming the F pocket are similarly conserved, and the B63 and B57/58 alleles all favor epitopes with small aliphatic amino acids like alanine (A), serine (S), and threonine (T) at position 2 and the aromatic amino acids phenylalanine (F), tryptophan (W), or tyrosine (Y) at the C-terminal end (3, 9, 31). Based on these common characteristics, HLA-B63 alleles have been grouped into the "HLA-B58 supertype," together with HLA-B57 and HLA-B58 (21, 32). Although no examples of shared CTL epitopes between the HLA-B63 and the HLA-B57/58 alleles have been described, their essentially identical binding motifs suggest that at least some HLA-B57 and HLA-B58 epitopes could be presented on HLA-B63, and vice versa. However, even if epitopes are shared among the alleles in the HLA-B58 supertype, it is unclear whether these responses could also translate into a more favorable HIV disease outcome in individuals expressing HLA-B63.
The present study defined the immunodominant CTL targets in HLA-B63-expressing individuals and addressed whether HLA-B63 can present previously described HLA-B57- and HLA-B58-restricted HIV epitopes and whether HLA-B63 is associated with reduced HIV viral loads in the absence of antiretroviral treatment. Our data show that HIV-infected individuals who express HLA-B63 do, in fact, target previously described HLA-B57 epitopes as well as longer, overlapping peptides containing known or potential new HLA-B63/B57/58 epitopes. Importantly, individuals who expressed HLA-B63 presented with reduced viral loads compared to the remainder of the study cohort. The present study therefore identifies a group of HLA class I alleles that share CTL epitopes and confer relative protection from HIV disease progression, suggesting that the nature of their shared CTL epitopes may be of significant relevance in providing this effect.
MATERIALS AND METHODS
Study subjects. HIV-infected individuals were recruited from HIV clade B-infected cohorts in the United States and an HIV clade C-infected cohort in Durban, South Africa. The HIV clade B-infected cohort included subjects from several ethnicities enrolled at four hospitals in the Boston area, the Harbor-UCLA Medical Center in Torrance, Calif., and the Queen Elizabeth Hospital in Bridgetown, Barbados (10). The HIV clade C-infected cohort included subjects of the Zulu and Xhosa ethnicities recruited at two clinics in Durban, South Africa (18). Infecting clade information and HLA typing were obtained as described previously (10). Clade C-infected individuals were included in this study for the comparison of viral loads only. Insufficient sample prevented a four-digit subtyping of the HLA-B63 allele in seven subjects. Subject characteristics are summarized in Table 1.
Viral load determinations. Viral loads were determined using the Roche Amplicor HIV-1 Monitor system with a lower detection limit of 400 RNA copies/ml. For samples that fell below this threshold, the ultrasensitive assay with a detection limit of 50 RNA copies/ml was applied where sufficient material was available.
Lymphocyte separation and use in ELISPOT assays. Peripheral blood mononuclear cells (PBMCs) were separated from whole blood and used fresh or frozen to test for CTL responses against previously defined HLA-B57-restricted CTL epitopes or a set of 410 overlapping 18-mer peptides spanning the entire HIV genome (9, 10). Gamma interferon enzyme-linked immunospot (ELISPOT) assays were performed as described in the past using Mabtech (Stockholm, Sweden) antibodies (10). The number of spots was determined using the AID ELISPOT Reader Unit (Autoimmun Diagnostika GmbH, Strassberg, Germany), and results were expressed as spot-forming cells (SFC) per million input cells. Thresholds for positive responses were determined as at least 5 spots (50 SFC/106 cells) per well and responses exceeding "mean of negative wells plus three standard deviations" and "three times mean," whichever was higher.
Epitope fine mapping was performed in peptide titration assays using truncated peptides in serial 10-fold dilutions ranging from 100 μg/ml to 10 pg/ml. All cellular assays were performed using samples from clade B-infected subjects and a clade B consensus sequence-based peptide set (10).
Statistical analysis. Statistical analyses were performed using GraphPad Prism version 3.0 for Macintosh. All tests were two tailed. Bonferroni's correction was applied for multiple tests. Viral load differences between groups were determined using the nonparametric Mann-Whitney U test. All viral load values are represented as medians. Viral load values below the limit of detection of 50 RNA copies/ml were assigned a value of 49 for statistical analyses, and for values with a detection limit of 400 copies/ml, a value of 399 was used. Viral loads >750,000 copies/ml were assigned a value of 750,001.
RESULTS
HLA-B63-expressing individuals frequently target HLA-B57-restricted epitopes. In order to investigate whether HLA-B63-positive individuals show a CTL response pattern similar to that of individuals expressing HLA-B57 or HLA-B58, HLA-B63-expressing subjects were tested for their total HIV-specific CTL response and for their ability to target previously described HLA-B57-restricted CTL epitopes. Due to the low allele frequencies of HLA-B1516 and HLA-B1517 (less than 1% in Caucasians and less than 1.5% in African Americans [5]) compared to those of HLA-B57 (about 4% in both ethnicities) and HLA-B58 (1% in Caucasians and 10% in African Americans [5]), sufficient cells from only 10 HIV clade B-infected, HLA-B63-positive individuals were available to test for total HIV responses using an overlapping peptide (OLP) set spanning the entire expressed HIV genome.
In a first analysis, the frequency of recognition among the HLA-B63-expressing individuals was determined for OLPs containing known HLA-B57 epitopes and compared to their overall recognition frequency in a cohort of 254 subjects tested comprehensively. Table 2 lists seven OLPs containing known HLA-B57 epitopes that were frequent targets of the HIV-specific immune response among these 10 subjects. Remarkably, six of the seven peptides were recognized at higher frequencies in HLA-B63-positive subjects than in HLA-B57/58-positive individuals. Although the HLA-B63-expressing individuals targeted all these OLPs more frequently than the remainder of the cohort, statistical significance was achieved only for two OLPs, mainly due to the presence of numerous epitopes restricted by HLA alleles different from HLA-B63 or HLA-B57/58 in the other, frequently targeted peptide sequences (9, 10). To establish that the HLA-B63-positive individuals did indeed target the same epitopes presented by HLA-B57 and HLA-B58, and not other potential epitopes contained in these OLPs, the described HLA-B57 optimal epitopes and truncations thereof were tested in HLA-B63-expressing individuals. These subjects showed strong responses (median, 379 SFC/106 PBMCs; range, 50 to >3,000 SFC/106 PBMCs) against the previously described optimal HLA-B57-restricted epitopes. Titration analyses of single-residue truncated peptide sequences also confirmed that the identical optimal epitopes were targeted in HLA-B63-positive subjects as in HLA-B57-expressing individuals (Fig. 1).
Frequently targeted epitopes in HLA-B63-positive subjects contain putative HLA-B1516- and HLA-B1517-restricted epitopes. The initial analysis of OLP recognition identified additional OLPs that were preferentially targeted by HLA-B63-positive individuals but did not contain previously known HLA-B57 epitopes. Table 3 includes all OLPs that were targeted by at least 30% of the HLA-B63-positive group. Of these 11 OLPs, 8 showed markedly increased frequencies of recognition compared to the HLA-B58 supertype-negative control group (P < 0.05). The remaining three OLPs were located in HIV Nef and were frequently targeted by the overall cohort, rendering the P value of these comparisons insignificant (10). Importantly, each of these 11 OLPs contained a putative or, for OLP 338, a previously defined HLA-B63-restricted epitope that fulfilled the HLA-B58 supertype-specific binding motif. Notably, this included epitopes with phenylalanine in second position, a residue which is not included in the described B58 supertype motif (21, 32). However, the presence of phenylalanine in the anchor position of some epitopes included in the present report, as well as among HLA-B57-, HLA-B58-, and HLA-B1516-restricted epitopes listed in the SYFPEITHI database (28), suggests that the HLA-B58 supertype motif could be expanded to include this additional residue.
HLA-B57- and HLA-B58-positive individuals recognize new B63 epitopes. To demonstrate not only that HLA-B63-expressing individuals target known HLA-B57-restricted CTL epitopes but that, in turn, HLA-B57/58-positive individuals can also target HLA-B63 epitopes, the frequency of recognition of the 11 OLPs in Table 3 was analyzed in B57/58-positive subjects. Indeed, five of the OLPs preferentially targeted in B63-positive individuals were also favored in B57/58-positive subjects (P < 0.05) (Table 3), strongly suggesting that these OLPs contained HLA-B63 epitopes that can be shared by HLA-B57/58 alleles. To assert that both groups really targeted identical epitopes, standard peptide titration assays were performed to define the optimal length of epitopes shared between HLA-B63 and HLA-B57/58. For six OLPs listed in Table 3, sufficient numbers of OLP responders and samples were available to determine the optimal epitope length. The data for five epitopes targeted by HLA-B63- or HLA-B57/58-positive individuals are shown in Fig. 2. Included in this figure are epitopes LL9, to illustrate presentation in HLA-B1517- and HLA-B57-expressing individuals with similar functional avidity; epitope RY11 (RSLYNTVATLY), embedded in OLP 11 and containing the known, but in this case nontargeted, HLA-A02-restricted SL9 epitope (SLYNTVATL); and epitopes YY9 (OLP 84), FF9 (OLP 23), and KV8 (OLP 162), which were identified based on their frequent recognition in HLA-B63-expressing subjects but which are also targeted by HLA-B57 individuals.
Together, the data identify novel HLA-B63 epitopes that are also frequently targeted in HLA-B57/58-positive subjects and confirm that epitopes previously described as HLA-B57 epitopes can also be recognized in HLA-B63-positive subjects. These included HLA-B57 epitopes of considerable interindividual immunodominance and for which an important role in the control of HIV replication has been proposed in the past (e.g., KF11 in OLP 22 and TW10 in OLP 33) (16, 20). Given the similarities in the functional avidity of these responses and the frequent recognition of novel HLA-B63 epitopes in HLA-B57/58-expressing individuals, the data suggest that CTL responses against these shared epitopes could mediate effective immune control in vivo, potentially leading to decreased viral set points in HLA-B63-positive individuals.
Novel B63 epitopes are targeted early in infection. As CTL responses to HLA-B57-restricted epitopes have been shown to be induced early in acute HIV infection (2), we assessed responses to the optimal epitopes emerging from the above-described analyses in two subjects during primary HIV infection. Subjects B19 and B20, tested before and 2 months after seroconversion, respectively, showed a pronounced immunodominance in the HLA-B63 epitopes compared to other known epitopes restricted by their respective HLA alleles. Subject B19 recognized 3 of 8 tested HLA-B63 epitopes and only 1 of 21 epitopes potentially restricted by other class I alleles (HLA-A01, HLA-A26, HLA-B35, and HLA-Cw04), with the magnitude against HLA-B63 epitopes exceeding 90% of the total response (890 of 980 SFC/million PBMCs) (Fig. 3a). Similarly, subject B20 recognized two epitopes, 1 of 8 HLA-B63 epitopes and 1 of 10 HLA-A01, HLA-A26, HLA-B39, and HLA-Cw07 epitopes, with the HLA-B63 response more than three times stronger than the latter (620 versus 180 SFC/million PBMCs) (Fig. 3b). These data demonstrate that immunodominant responses against HLA-B63-restricted epitopes are detectable in primary HIV infection, comparable to the early targeting of HLA-B57 epitopes (2), and suggest that they might play an important role in containing initial viral replication.
HLA-B63-expressing individuals present with reduced HIV viral loads. Given the shared set of optimal epitopes targeted in individuals expressing HLA-B63 or HLA-B57/58, the early induction of these responses upon HIV infection, and the important role that HLA-B57-restricted responses play in natural infection, the clinical outcome of HIV infection may be comparable between HLA-B63- and HLA-B57/58-expressing subjects. To test this hypothesis, HLA-B63-positive individuals not receiving antiretroviral therapy were identified among HIV clade B- and clade C-infected individuals enrolled in cohorts in Boston and Durban (Table 1), and their viral set points were analyzed. High-resolution HLA typing was available for 20 individuals, showing that all the individuals enrolled in Durban expressed the HLA-B1516 allele, whereas 9/13 individuals in the Boston cohort expressed HLA-B1516 and 4/13 were HLA-B1517 positive. Of the 27 individuals expressing either subtype of HLA-B63, 12 untreated HIV clade B-infected individuals were included in the subsequent analysis. All seven subjects in the clade C-infected cohort were treatment nave.
When untreated HLA-B63-expressing individuals were compared to the remainder of the cohort, a significantly lower viral load was observed in the HLA-B63-positive subjects than in the HLA-B63-negative subjects (P = 0.0002; median viral loads, 3,280 versus 32,500 RNA copies/ml) (Fig. 4). This marked control of viral replication is comparable to the one observed in untreated HLA-B57-positive subjects in our cohort (P < 0.0001; median viral loads, 6,645 versus 34,650 RNA copies/ml) (data not shown). The association remained statistically significant when clade B- and clade C-infected individuals were analyzed separately (P = 0.042 [3,912 versus 13,000 RNA copies/ml for clade B] and P = 0.029 [3,280 versus 35,550 RNA copies/ml for clade C]) (data not shown). Furthermore, excluding HLA-B57- or HLA-B58-coexpressing subjects from the analysis did not affect the outcome (P = 0.0002 for exclusion of HLA-B57-positive subjects and P = 0.0003 for exclusion of HLA-B58-positive subjects). The separate analysis of viral set points in the individuals expressing either the HLA-B1516 or HLA-B1517 subtype also reached statistical significance for HLA-B1516 (P = 0.0008; 4,850 versus 31,700 RNA copies/ml) but not for HLA-B1517, as viral load data were only available for two untreated, HLA-B1517-positive individuals. Collectively, the data demonstrate that HLA-B63 is associated with reduced viral load and, by inference from other published studies, with slower HIV disease progression (22). However, the small number of HLA-B63-expressing individuals for whom "time since infection" could be reliably documented did not allow us to directly assess the rate of HIV disease progression in the present cohort.
DISCUSSION
The data presented in this report identify HLA-B63 as a novel HLA class I allele associated with reduced HIV viral loads in chronic, untreated HIV-1 clade B and clade C infection. Since HLA-B63-expressing individuals mount strong immune responses against previously described HLA-B57-restricted epitopes, and given that HLA-B57/58-expressing subjects also show strong CTL activity against HLA-B63 epitopes, the data indicate that epitopes shared by these alleles may be central in mediating the beneficial effects associated with these alleles.
The ability of HLA-B57 and HLA-B58 to present identical optimal epitopes has been known for many years (13), but no HLA-B57 epitopes have been shown thus far to be presented by other members of the B58 supertype. However, sequence analyses of the residues forming the F pocket of HLA-B1516 strongly suggest that at least some HLA-B57 epitopes could be accommodated in HLA-B1516 (3, 27). Indeed, as reported here, HLA-B57 epitopes containing any of the aromatic amino acids as the C-terminal anchor were targeted in HLA-B63-positive individuals, and titration studies using serially truncated peptides show that the same optimal epitopes are being targeted by these subjects. This is further supported by data derived from the crystallographic structure of HLA-B5703 and bioinformatics analyses of related HLA sequences, which suggest that HLA-B63 alleles could cross-present optimal HLA-B57 epitopes. In addition, the newly described HLA-B63 epitopes are also frequently targeted in HLA-B57- or HLA-B58-positive subjects, further indicating that the same optimal epitopes are frequently shared between the different alleles of the B58 supertype.
A number of HLA class I alleles have been associated with slow HIV disease progression, but the relative contribution of the presenting allele versus the restricted epitopes towards this relative protection remains unclear. As the HLA-B58 supertype binding motif has been acquired by the HLA-B63 allele through a recombination event between HLA-B57/58 and HLA-B15 alleles, the sequence similarities between HLA-B63 and HLA-B57/58 are mainly restricted to the 1-helix region. Thus, whereas high overall sequence similarities between HLA-B57 and HLA-B58 do not allow dissection of the relative contribution of the presenting HLA allele or the presented epitopes on the in vivo efficacy of these CTLs, the case of HLA-B63 represents a situation where nature may have provided an opportunity to elucidate these factors in more detail. Although it cannot be ruled out that sequence similarities among the B58 supertype alleles at T-cell receptor contact sites outside the 1 helix are responsible for their common ability to provide relative protection from HIV disease progression, the present data strongly suggest that the nature of the targeted epitope is playing an important role in providing in vivo control (23). Such superior immune control may be due to better recognition of potential viral epitope variants and their increased cost to viral replication fitness (15). However, the level of variability among the epitopes shown in Tables 2 and 3 did not indicate particularly high conservation of these sequences (which would have indicated a high fitness cost for CTL escape), except for the two potential epitopes embedded in the highly conserved protease protein. Instead, high functional avidity and broad T-cell-receptor repertoires may contribute to more effective in vivo immune control mediated by these responses (16, 26) and, together with possible functional constraints that prevent rapid CTL escape or lead to similarly rapid reversion of escape variants, may override the potential effects of the presenting HLA class I molecule (1, 11, 20). However, it remains to be seen how far this observation holds true for other HLA class I alleles and epitopes, especially for epitopes that could be presented by alleles differentially associated with slow and fast HIV disease progression. In the case of HLA-B58, minute sequence changes between HLA-B5801 and HLA-B5802 are clinically associated with different HIV disease outcomes (18). This is also reflected by different CTL response patterns seen in subjects expressing either subtype, as fewer HLA-B5802-expressing individuals than HLA-B5801-expressing individuals responded to the frequently targeted OLPs shown in Tables 2 and 3, although HLA-B5802 was twice as frequently found in the clade C cohort compared to HLA-B5801. Thus, targeting the epitopes described in this report seems to be providing a selective advantage to the HLA-B5801-expressing individuals, as there was only restricted sharing of these epitopes between subjects expressing closely related alleles. Similar observations have been made for other closely related alleles, including HLA-B35 and its subtypes, which are associated with different rates of HIV disease progression. However, the link to CTL response patterns and the induction of qualitatively different responses depending on the presenting allele has thus far not been made and will require more intensive characterization of responses in individuals expressing closely related alleles. Eventually, the determination of the relative contribution of the presenting HLA class I allele versus the presented epitopes, and the identification of epitopes, rather than HLA alleles, associated with slow HIV disease progression, will further our understanding of antiviral immunity and facilitate the design of effective prophylactic and therapeutic immune interventions for HIV infection.
ACKNOWLEDGMENTS
This work has been funded by NIH contract N01-Al-15422, by the Los Alamos National Laboratory (la-ur-03-5892), and by NIH grants AI-043638 and AI-067077.
We gratefully acknowledge the helpful discussions with and guidance by Patricia D'Souza for this project.
REFERENCES
Allen, T. M., M. Altfeld, X. G. Yu, K. M. O'Sullivan, M. Lichterfeld, S. Le Gall, M. John, B. R. Mothe, P. K. Lee, E. T. Kalife, D. E. Cohen, K. A. Freedberg, D. A. Strick, M. N. Johnston, A. Sette, E. S. Rosenberg, S. A. Mallal, P. J. Goulder, C. Brander, and B. D. Walker. 2004. Selection, transmission, and reversion of an antigen-processing cytotoxic T-lymphocyte escape mutation in human immunodeficiency virus type 1 infection. J. Virol. 78:7069-7078.
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, T. M. Allen, M. E. Feeney, J. O. Kahn, R. P. Sekaly, J. A. Levy, J. K. Rockstroh, P. J. Goulder, and B. D. Walker. 2003. Influence of HLA-B57 on clinical presentation and viral control during acute HIV-1 infection. AIDS 17:2581-2591.
Barber, L. D., L. Percival, K. L. Arnett, J. E. Gumperz, L. Chen, and P. Parham. 1997. Polymorphism in the alpha 1 helix of the HLA-B heavy chain can have an overriding influence on peptide-binding specificity. J. Immunol. 158:1660-1669.
Brander, C., and B. D. Walker. 1996. The HLA class I restricted CTL response in HIV infection: systematic identification of optimal epitopes, p. IV50-IV60. In C. B. B. Korber, B. Walker, R. Koup, J. Moore, B. Haynes, and G. Meyers (ed.), HIV molecular immunology database. Theoretical biology and biophysics. Los Alamos National Laboratory, Los Alamos, N.M.
Cao, K., J. Hollenbach, X. Shi, W. Shi, M. Chopek, and M. A. Fernandez-Vina. 2001. Analysis of the frequencies of HLA-A, B, and C alleles and haplotypes in the five major ethnic groups of the United States reveals high levels of diversity in these loci and contrasting distribution patterns in these populations. Hum. Immunol. 62:1009-1030.
Draenert, R., S. Le Gall, K. J. Pfafferott, A. J. Leslie, P. Chetty, C. Brander, E. C. Holmes, S. C. Chang, M. E. Feeney, M. M. Addo, L. Ruiz, D. Ramduth, P. Jeena, M. Altfeld, S. Thomas, Y. Tang, C. L. Verrill, C. Dixon, J. G. Prado, P. Kiepiela, J. Martinez-Picado, B. D. Walker, and P. J. Goulder. 2004. Immune selection for altered antigen processing leads to cytotoxic T lymphocyte escape in chronic HIV-1 infection. J. Exp. Med. 199:905-915.
Feeney, M. E., Y. Tang, K. A. Roosevelt, A. J. Leslie, K. McIntosh, N. Karthas, B. D. Walker, and P. J. R. Goulder. 2004. Immune escape precedes breakthrough human immunodeficiency virus type 1 viremia and broadening of the cytotoxic T-lymphocyte response in an HLA-B27-positive long-term-nonprogressing child. J. Virol. 78:8927-8930.
Flores-Villanueva, P. O., E. J. Yunis, J. C. Delgado, E. Vittinghoff, S. Buchbinder, J. Y. Leung, A. M. Uglialoro, O. P. Clavijo, E. S. Rosenberg, S. A. Kalams, J. D. Braun, S. L. Boswell, B. D. Walker, and A. E. Goldfeld. 2001. Control of HIV-1 viremia and protection from AIDS are associated with HLA-Bw4 homozygosity. Proc. Natl. Acad. Sci. USA 98:5140-5145.
Frahm, N., P. J. R. Goulder, and C. Brander. 2003. Broad HIV-1 specific CTL responses reveal extensive HLA class I binding promiscuity of HIV-derived, optimally defined CTL epitopes, p. 3-24. In B. T. Korber, C. Brander, B. F. Haynes, R. Koup, J. P. Moore, B. D. Walker, and D. I. Watkins (ed.), HIV immunology and HIV/SIV vaccine databases 2003, vol. 1. Theoretical biology and biophysics. Los Alamos National Laboratory, Los Alamos, N.M.
Frahm, N., B. T. Korber, C. M. Adams, J. J. Szinger, R. Draenert, M. M. Addo, M. E. Feeney, K. Yusim, K. Sango, N. V. Brown, D. SenGupta, A. Piechocka-Trocha, T. Simonis, F. M. Marincola, A. G. Wurcel, D. R. Stone, C. J. Russell, P. Adolf, D. Cohen, T. Roach, A. St. John, A. Khatri, K. Davis, J. Mullins, P. J. Goulder, B. D. Walker, and C. Brander. 2004. Consistent cytotoxic-T-lymphocyte targeting of immunodominant regions in human immunodeficiency virus across multiple ethnicities. J. Virol. 78:2187-2200.
Friedrich, T. C., E. J. Dodds, L. J. Yant, L. Vojnov, R. Rudersdorf, C. Cullen, D. T. Evans, R. C. Desrosiers, B. R. Mothe, J. Sidney, A. Sette, K. Kunstman, S. Wolinsky, M. Piatak, J. Lifson, A. L. Hughes, N. Wilson, D. H. O'Connor, and D. I. Watkins. 2004. Reversion of CTL escape-variant immunodeficiency viruses in vivo. Nat. Med. 10:275-281.
Garrett, T. P., M. A. Saper, P. J. Bjorkman, J. L. Strominger, and D. C. Wiley. 1989. Specificity pockets for the side chains of peptide antigens in HLA-Aw68. Nature 342:692-696.
Goulder, P. J., M. Bunce, P. Krausa, K. McIntyre, S. Crowley, B. Morgan, A. Edwards, P. Giangrande, R. E. Phillips, and 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. Retrovir. 12:1691-1698.
Goulder, P. J., R. E. Phillips, R. A. Colbert, S. McAdam, G. Ogg, M. A. Nowak, P. Giangrande, G. Luzzi, B. Morgan, A. Edwards, A. J. McMichael, and S. Rowland-Jones. 1997. Late escape from an immunodominant cytotoxic T-lymphocyte response associated with progression to AIDS. Nat. Med. 3:212-217.
Goulder, P. J., and D. I. Watkins. 2004. HIV and SIV CTL escape: implications for vaccine design. Nat. Rev. Immunol. 4:630-640.
Jansen, C. A., S. Kostense, K. Vandenberghe, N. M. Nanlohy, I. M. Cuyper, E. Piriou, E. H. Manting, F. Miedema, and D. van Baarle. 2005. High responsiveness of HLA-B57-restricted Gag-specific CD8(+) T cells in vitro may contribute to the protective effect of HLA-B57 in HIV-infection. Eur. J. Immunol. 35:150-158.
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, R. Detels, W. Blattner, J. Phair, H. Erlich, and D. L. Mann. 1996. Influence of combinations of human major histocompatibility complex genes on the course of HIV-1 infection. Nat. Med. 2:405-411.
Kiepiela, P., A. J. Leslie, I. Honeyborne, D. Ramduth, C. Thobakgale, S. Chetty, P. Rathnavalu, C. Moore, K. J. Pfafferott, L. Hilton, P. Zimbwa, S. Moore, T. Allen, C. Brander, M. M. Addo, M. Altfeld, I. James, S. Mallal, M. Bunce, L. D. Barber, J. Szinger, C. Day, P. Klenerman, J. Mullins, B. Korber, H. M. Coovadia, B. D. Walker, and P. J. Goulder. 2004. Dominant influence of HLA-B in mediating the potential co-evolution of HIV and HLA. Nature 432:769-775.
Koup, R. A., J. T. Safrit, Y. Cao, C. A. Andrews, G. McLeod, W. Borkowsky, C. Farthing, and 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.
Leslie, A. J., K. J. Pfafferott, P. Chetty, R. Draenert, M. M. Addo, M. Feeney, Y. Tang, E. C. Holmes, T. Allen, J. G. Prado, M. Altfeld, C. Brander, C. Dixon, D. Ramduth, P. Jeena, S. A. Thomas, A. St. John, T. A. Roach, B. Kupfer, G. Luzzi, A. Edwards, G. Taylor, H. Lyall, G. Tudor-Williams, V. Novelli, J. Martinez-Picado, P. Kiepiela, B. D. Walker, and P. J. Goulder. 2004. HIV evolution: CTL escape mutation and reversion after transmission. Nat. Med. 10:282-289.
Lund, O., M. Nielsen, C. Kesmir, A. G. Petersen, C. Lundegaard, P. Worning, C. Sylvester-Hvid, K. Lamberth, G. Roder, S. Justesen, S. Buus, and S. Brunak. 2004. Definition of supertypes for HLA molecules using clustering of specificity matrices. Immunogenetics 55:797-810.
Mellors, J. W., C. R. Rinaldo, Jr., P. Gupta, R. M. White, J. A. Todd, and L. A. Kingsley. 1996. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 272:1167-1170.
Messaoudi, I., J. A. G. Patino, R. Dyall, J. LeMaoult, and J. Nikolich-Zugich. 2002. Direct link between mhc polymorphism, T cell avidity, and diversity in immune defense. Science 298:1797-1800.
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, and M. Connors. 2000. HLA B5701 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.
O'Brien, S. J., X. Gao, and M. Carrington. 2001. HLA and AIDS: a cautionary tale. Trends Mol. Med. 7:379-381.
Price, D. A., S. M. West, M. R. Betts, L. E. Ruff, J. M. Brenchley, D. R. Ambrozak, Y. Edghill-Smith, M. J. Kuroda, D. Bogdan, K. Kunstman, N. L. Letvin, G. Franchini, S. M. Wolinsky, R. A. Koup, and D. C. Douek. 2004. T cell receptor recognition motifs govern immune escape patterns in acute SIV infection. Immunity 21:793-803.
Prilliman, K. R., K. W. Jackson, M. Lindsey, J. Wang, D. Crawford, and W. H. Hildebrand. 1999. HLA-B15 peptide ligands are preferentially anchored at their C termini. J. Immunol. 162:7277-7284.
Rammensee, H., J. Bachmann, N. P. Emmerich, O. A. Bachor, and S. Stevanovic. 1999. SYFPEITHI: database for MHC ligands and peptide motifs. Immunogenetics 50:213-219.
Robinson, J., M. J. Waller, P. Parham, N. de Groot, R. Bontrop, L. J. Kennedy, P. Stoehr, and S. G. Marsh. 2003. IMGT/HLA and IMGT/MHC: sequence databases for the study of the major histocompatibility complex. Nucleic Acids Res. 31:311-314.
Saper, M. A., P. J. Bjorkman, and D. C. Wiley. 1991. Refined structure of the human histocompatibility antigen HLA-A2 at 2.6 A resolution. J. Mol. Biol. 219:277-319.
Seeger, F. H., D. Arnold, T. Dumrese, H. de la Salle, D. Fricker, H. Schild, H. G. Rammensee, and S. Stevanovic. 1998. The HLA-B 1516 motif demonstrates HLA-B-specific P2 pocket characteristics. Immunogenetics 48:156-160.
Sette, A., and J. Sidney. 1999. Nine major HLA class I supertypes account for the vast preponderance of HLA-A and -B polymorphism. Immunogenetics 50:201-212.
Stewart-Jones, G. B. E., G. Gillespie, I. M. Overton, P. Roche, A. J. McMichael, S. Rowland-Jones, and E. Y. Jones. 2005. Structures of three HIV-1 HLA-B5703-peptide complexes and identification of related HLAs potentially associated with long-term non-progression. J. Immunol., in press.
Trachtenberg, E. A., and H. A. Erlich. 2001. A review of the role of the human leukocyte antigen (HLA) system as a host immunogenic factor influencing HIV transmission and progression to AIDS, p. I-43-60. In B. T. K. Korber, C. Brander, B. F. Haynes, J. P. Moore, R. A. Koup, C. Kuiken, B. D. Walker, and D. I. Watkins (ed.), HIV molecular immunology 2001. Theoretical Biology and Biophysics Group, Los Alamos, N.M.(Nicole Frahm, Sharon Adam)