| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
,¶
* National Institute for Communicable Diseases, and
Perinatal HIV Research Unit, University of the Witwatersrand, Johannesburg, South Africa;
Institute of Infectious Diseases and Molecular Medicine, University of Cape Town, Cape Town, South Africa;
HIV Vaccine and Prevention Trials Unit, Medical Research Council, and
¶
University of KwaZulu Natal, Durban, South Africa;
||
Department of Immunology, University of Zimbabwe, Harare, Zimbabwe;
#
Center for AIDS Research, Stanford University Medical Center, Palo Alto, CA 94305;
**
Zambia-University of Alabama at Birmingham HIV Research Project, Lusaka, Zambia;
Johns Hopkins Research Project, Blantyre, Malawi; and
*
California Department of Health Sciences, Richmond, CA 94804
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Clinical data, from natural history studies, have shown that CTL targeting of Gag may or may not be important for viral control. Some studies have shown that no correlations exist between the overall CD8+ T cell response and plasma viral load (20, 21, 22), and others show that HIV-1 Gag-specific responses associate with a lower viral set point within the first 12 mo of HIV-1 infection (23). Data also exist that show a correlation between CTL responses targeting Gag with proviral DNA, but not plasma viral RNA (24). Other studies have observed an inverse correlation between viremia and the breadth and magnitude of p24 Gag-specific CTL responses (20, 25, 26, 27). These observations alone provide strong evidence that MHC class I presentation of Gag epitopes will be essential for the effectiveness of HIV-1 vaccines and concur with the data found from macaque studies. In agreement with other studies (28, 29), we have recently shown (30) that there is a positive correlation between total HIV-1-specific CD8+ T cell responses and viral load, but a significant correlation between preferential targeting of Gag by CD8+ T cells and control of plasma viremia.
In this study, we extend these observations (30), and the primary objective of this study was to characterize epitope-specific regions in Gag from subtype C HIV-1-infected southern Africans. We have mapped epitopes in Gag and reveal the presence of a number of novel epitopes overlapping with previously described epitopes from subtype B HIV-1 infection. We hypothesize that unique HLA-restricting alleles in the southern African population account for the identity of novel epitopes, and that these occur in highly conserved domains of Gag. These data: 1) highlight the degeneracy of HIV-1 Ag recognition in Gag; 2) highlight the notion that global HLA diversity between populations may not be a limiting factor for common epitope recognition in Gag; and 3) provide baseline information to compare with vaccine-induced Gag-specific CD8+ responses.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Thirty-eight HIV-1-infected individuals were analyzed for responses against Gag, and were derived from those enrolled into HIVNET 028, a natural history study of recent HIV-1 infection from four southern African countries: Malawi, Zimbabwe, Zambia, and South Africa. Table I shows details of the individuals analyzed, including time from seroconversion, CD4 counts, and viremia. The median time from seroconversion was 7 mo, in which time from seroconversion was estimated as the midpoint between the last Ab-negative and the first Ab-positive ELISA, using the ELISPOT assay date as the reference time point. All individuals analyzed were antiretroviral drug naive. The median RNA copies/ml were 12,936 (4.11 log10), and there were no significant associations between country of origin and viremia. Thus, for the purposes of this study, individuals were grouped as one cohort.
|
A total of 66 peptides based on consensus C Gag was used to screen and confirm Gag-specific responses and was synthesized using Fmoc-based solid-phase chemistry (Natural and Medical Sciences Institute, University of Tuebingen, Reutlingen, Germany). All peptides were checked for the correct m.w. using Elektrospray-quadruple time-of-flight-mass spectrometry, and peptide purity ranged from 70 to 80%. Peptides were dissolved in 100% DMSO at an initial concentration of 10 mg/ml and pooled at 40 µg/ml/peptide stock in PBS, in which the final DMSO concentration was always <0.5%. The peptide set consisted of 15–18 mers overlapping by 10 aa and was aliquoted to be arranged in a pool and matrix design, such that each peptide appeared twice in two different pools. In total, there were five major pools containing 14 peptides/pool plus a further 14 matrix pools containing 5 peptides/pool. Pools were arranged so that peptide responses identified in any one of the major pools could be cross-referenced to a response in one of the matrix pools. In so doing, multiple peptide responses could be narrowed down to likely single peptide response. Single peptide responses were then confirmed in subsequent assays. Truncated 9- to 12-mer peptides were used for fine epitope mapping and for HLA restrictions.
Preparation of cells
PBMC were prepared using Ficoll gradient centrifugation and cryopreserved in 90% FCS plus 10% DMSO. At the time of analysis, cryopreserved PBMC were thawed and rested overnight in R10 (RPMI 1640 (Invitrogen Life Technologies, Paisley, U.K.) supplemented with 10% FCS (Invitrogen Life Technologies) and 50 U of gentamicin (Invitrogen Life Technologies) at 2–4 x 106 PBMC/ml at 37°C in 5% CO2 before use in the ELISPOT assay (30). After incubation, cells were counted using trypan blue dye exclusion and used in assays. EBV-transformed B cell lines, for use as class I HLA-matched targets, were established for each individual and maintained in RPMI 1640 supplemented with 20% (R20) FCS and 50 U of gentamicin at 37°C/5% CO2.
IFN-
ELISPOT assays
ELISPOT assays were performed, as previously described (30). Briefly, cryopreserved PBMC were plated on 96-well polyvinylidene difluoride plates (MAIP S45; Millipore, Johannesburg, South Africa) that were coated with 50 µl of anti-IFN- mAb 1-D1k (2 µg/ml; Mabtech, Stockholm, Sweden) overnight at 4°C. Peptides were added directly to the wells at various final concentrations, ranging between 0.002 to 2 µg/ml along with 1–2 x 105 cells in a volume of 50 µl of R10 and incubated at 37°C in 5% CO2. After 16–18 h, plates were extensively washed with PBS and wash buffer (PBS, 1% FCS, 0.001% Tween 20), followed by incubation with biotinylated anti-IFN- mAb (2 µg/ml; clone 7-B6-1; Mabtech) at room temperature for 3 h. After a further six washes with wash buffer, 2 µg/ml streptavidin HRP (BD Pharmingen, Cupertino, CA) was added to the wells, and the plates were incubated for an additional 1 h at room temperature. Spots were visualized using Novared substrate (Vector Laboratories, Burlingame, CA), according to the manufacturer’s instructions. Wells containing PBMC and R10 medium, as well as R10 medium alone, were used as negative controls, in which PBMC and medium controls were run in duplicate. Wells containing PBMC and PHA served as positive controls. The number of spots per well was counted using the Immunospot (Cellular Technology, Cleveland, OH) automated cell counter. Peptide responses were confirmed using individual peptides in the ELISPOT assay. The criteria used to define a positive response are as described previously (30).
Intracellular cytokine staining (ICS)4
PBMC (0.5–1 x 106) were incubated with 2 µg of peptide and anti-CD28 and anti-CD49d mAbs (1 µg/ml; BD Pharmingen) at 37°C and 5% CO2 for 1 h. Brefeldin A (10 µg/ml; Sigma-Aldrich, St. Louis, MO) was added, and cells were incubated for a further 5 h at 37°C and 5% CO2. Cells were transferred to 4°C overnight. PBMC were washed, permeabilized, and stained with mAbs against CD8 (PerCP), CD3 (allophycocyanin), CD69 (PE), and IFN- (FITC) for 30 min at room temperature in the dark. Cells were washed, fixed, and acquired on a FACSCalibur using the CellQuest software. Analysis for IFN--positive cells was done using Paint-a-gate software. For HLA restriction, B cells expressing a single matched HLA allele with autologous B cells were pulsed with 2 µg of peptide for 1 h at 37°C and 5% CO2. B cells were washed four times with RPMI 1640 supplemented with 1% FCS (R1), and the B cells were mixed with Ag-specific cells (effectors) at a ratio of 5:1 T cells to B cells in 1 ml of R10 and incubated with anti-CD28/CD49d mAbs, and the assay continued as described for ICS.
51Cr release assay
PBMC at 1 x 106 were stimulated with 10 µg of peptide and 330 U/ml IL-7 (Roche, Basel, Switzerland), and cells were transferred to a 24-well plate and incubated at 37°C and 5% CO2. Following 1 wk in culture, 25 U/ml IL-2 was added, and after 14–21 days in culture, PBMC (effectors) were harvested for use in a standard 51chromium release assay. HLA-matched B cells were pulsed with 10 µg of peptide and 100–150 µCi of 51Cr for 1 h at 37°C and 5% CO2. Following extensive washing with R1, B lymphoblastoid cell line (BLCL) targets were resuspended in R10 and mixed with effectors in triplicate at a ratio of 50:1 and 25:1 E:T ratio in V-bottom 96-well plate, and the killing assay was incubated for 4–5 h at 37°C and 5% CO2. Twenty-five microliters of the supernatant were harvested onto LumaPlates-96 (Packard Bioscience, Billerica, MA) and allowed to dry overnight, and the radioactivity was detected using a gamma counter (TopCount.NXT; Packard Bioscience). Spontaneous release of 51Cr was determined by incubating targets with R10, and the maximum release was determined from the total radioactivity released by targets in the presence of 0.01% Triton X-100. Specific lysis (%) was calculated as: (experimentalaverage – spontaneousaverage) ÷ (maximumaverage – spontaneousaverage) x 100.
HLA typing
DNA was extracted from EBV-transformed B cell lines using the Pel-Freez DNA Isolation Kit (Pel-Freez Clinical Systems, LLC, Brown Deer, WI). Low to medium resolution typing of class I HLA A, B, and C loci was performed using the Pel-Freez sequence specific primers UniTray PCR-based method (Pel-Freez Clinical Systems). High resolution typing of HLA class I A and B loci was performed using sequencing-based typing kits (Applied Biosystems, Foster City, CA). Sequencing products were purified and analyzed using the ABI PRISM 3100 Genetic Analyzer (Applied Biosystems). Analysis of the resulting sequences and subsequent allele assignment were performed using the MatchMaker Allele Identification Software (Applied Biosystems).
Statistical and data analysis
All data were expressed as medians with interquartile ranges and analyzed using nonparametric statistics. Krukal-Wallis nonparametric ANOVA was used to test for significant differences across pools of Gag and Dunn’s pairwise analysis to identify differences between pools. Fisher’s exact test was used to identify differences in HLA allele frequencies between populations. Statistical analysis was performed using SigmaStat 2.0 (SPSS, Chicago, IL) and Arlequin (Zero G Software, San Francisco, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Table I shows the magnitude and range of IFN- ELISPOT responses (spot-forming units (sfu)/106 PBMC) of each of the HIV-1-infected individuals to the five pools of overlapping Gag peptides. There were no significant differences between responses found in pool 1 (P1) to pool 4 (P4), although there was a significantly (p < 0.05) lower magnitude of recognition to pool 5 (P5, p2p7p1p6 region). Seventy-four percent (28 of 38) of individuals recognized at least one of the peptide pools, where only one individual recognized all five pools. The median cumulative response across all five pools was 380 sfu/106 PBMC for the whole cohort and 908 sfu/106 PBMC for only the 28 responders, and the median number of pools recognized was 1 for the cohort and 3 for the 28 responders. There was no significant difference (p = 0.763) between viral loads in those that did respond to any of the Gag pools (median of 9,402 RNA copies/ml) and those that did not respond to any pool (median of 14,280 RNA copies/ml). There was no correlation between the number of Gag pools recognized and viral load (r = 0.05; p = 0.74), although there was a negative nonsignificant trend between sfu/106 PBMC and log10 viral load (r = –0.353, p = 0.09). Using a matrix approach (outlined in Materials and Methods), a range of possible peptides was identified and single peptides were then used to confirm epitopic region reactivity.
Fig. 1 shows a peptide density plot of the frequency of recognized epitopic regions after confirming the ELISPOT screens with single Gag peptides in 26 of the 28 responders. When the overlapping amino acid sequences were conjoined, four peptide stretches in p17 were identified: 9RGGKLDKWEKIRLRPGGKKHYMLKHL34 was recognized at a frequency of 15.4% (4 of 26); 16KHLVWASRELERFAL30 was recognized at a frequency of 7.7% (2 of 26); 70TGTEELRSLYNTVATLY86 was recognized at a frequency of 34.6% (9 of 26); and the frequency of recognition for 119KAAADKGKVSQNYPIV134 was 11.5% (3 of 26).
|
Optimal epitope recognition and HLA restrictions in p17
To determine the optimal epitopes for the peptide stretches identified in Fig. 1, we used truncated peptides ranging from 9 to 12 aa in length. The optimal epitope was defined as the peptide that gave the highest number of sfu/106 cells at the lowest peptide concentration in the IFN- ELISPOT assay. HLA restrictions were performed using either 51Cr release CTL assays or ICS, and the restricting allele was defined as the single HLA allele that gave comparable or greater responses to autologous target BLCL. Fig. 2 shows that within the parent peptide stretch 16KHLVWASRELERFAL46 (Fig. 1, epitopic region 2), two novel epitopes were identified. The first was 17HLVWASREL25 (HL-9), shown to give the highest sfu/106 PBMC at the lowest peptide concentration upon peptide titration (Fig. 2A). The HLA-restricting allele of the HL-9 epitope was shown to be HLA-Cw*0804 (Fig. 2B), in which comparable target cell lysis with autologous BLCL was observed with partially matched target cells at the –Cw*0804 locus (Fig. 2B). The second epitope within the parent stretch was 18LVWASRELERF28 (LF-11), which was clearly recognized as the only truncated peptide in the titration experiment (Fig. 2C). Restriction experiments using the optimal LF-11 epitope revealed two possible HLA restrictions. Fig. 2D shows titratable CTL killing when BLCL partially matched at HLA-A*3001/3002 or HLA-A*3001/3002/B*5703 were used. These experiments show that LF-11 is restricted by either HLA-A*3001/3002 or HLA-A*57031. The higher percentage of lysis using BLCL expressing both HLA-A*3002 and HLA-B*570301 (Fig. 2D) infers that the epitope may be presented by both HLA restriction alleles in culture. The lower CTL lysis using almost identical HLA-matched targets, except for HLA-A*3001 difference, suggests that the epitope may be restricted optimally through HLA-A*3002 and not HLA-A*3001. The ambiguous nature of these restrictions highlights the promiscuous nature of LF-11.
|
|
Next, we fine mapped two epitopic regions in p24 of subtype C Gag.The first region, 172PMFTALSEGATPQDLNTMLNTVGGH196, is known to contain a previously described TL-9 epitope recognized by both subtype B and C HIV-1-infected individuals (39, 40, 41), and has been shown to be restricted by HLA-B*07 and HLA-B*42 or HLA-B*81. We mapped fine responses within the longer peptide to the known TPQDLNTML (TL-9) epitope (Fig. 4A) and confirmed with the 51Cr release assay that it was restricted by both HLA-B*42 and HLA-B*81 in an individual expressing both alleles (Fig. 4B). Confirmation of these HLA restrictions using ICS, a more ex vivo assay, also revealed that the TL-9 epitope was restricted by both HLA-B*42 and HLA-B*81 in the same individual (Fig. 4C) and confirmed these were CD8+ T cell mediated.
|
|
When each of the epitopic regions in p17, p24, and p2p7p1p6 (Fig. 1) were aligned with the consensus Gag sequence from the HIVNET 028 cohort, consensus C from southern Africa and consensus M (Fig. 6), it was evident that each of the targeted regions was highly conserved. Our rationale for these alignments was to highlight the level of conservancy across these regions. Although there were a few amino acid substitutions identified, mostly in the consensus M sequence, the mapped epitope identities (Fig. 6) revealed the highly conserved nature of the epitopes. Both the RY-11 and LY-9 epitopes in p17 were identified using the RSLYNTVATLY peptide sequence, although the corresponding consensus cohort and autologous sequence was identified to be RSLFNTVATLY. The phenylalanine (F) substitution for tyrosine (Y) is known to abrogate T cell recognition in the context of HLA-A*0201 restriction (37), but appears not to influence recognition for HLA-A*2902/B*44 and HLA-A*30 (Fig. 6). The 289GPKEPFRDYVDRFKTLRAEQATQDV314 epitopic region, at the -C terminus of the major homology region in p24, is completely conserved and possesses four highly targeted overlapping epitopes.
|
Table II shows a summary of the nine optimal epitopes identified in four of the peptide stretches described in Fig. 1. Three epitopes (LVWASRELERF, LYNTVATLY, and TPQDLNTML) were restricted by more than one HLA allele, showing the promiscuity of epitope binding. Five of the epitopes have not been previously reported, and two epitopes (LVWASRELERF and LYNTVATLY) were restricted by frequently expressed HLA-A*30 (HLA-A*3002) (allele frequency = 0.226) and HLA-A*2902 (allele frequency = 0.119) alleles in the cohort. Similarly, three epitopes (RSLYNTVATLY, TPQDLNTML, and YVDRFFKTL) identified were not novel, although the restricting HLA allele was significantly more common in the cohort examined; namely HLA-B*4201, HLA-B*8101, and HLA-B*15 (Table II). Conversely, two further epitopes (LYNTVATLY and TLRAEQATQD) were restricted by HLA-B*4403 and HLA-Cw*0304 that are significantly more common in Caucasian populations (Table II). When we noted the HLA backgrounds of individuals recognizing each peptide, we could make accurate associations for single HLA alleles with the identified restricting allele for eight of the epitopes. For example, all individuals (n = 3) recognizing the LF-11 epitope expressed HLA-A*30, and all individuals (n = 13) recognizing either the RY-11 or LY-9 epitope expressed HLA-A*30, HLA-A*29, or HLA-B*44 (Table II). All individuals (n = 5) recognizing the TL-9 epitope expressed either or both HLA-B*4201 or HLA-B*8101. For the cluster of epitopes within the peptide stretch, GPKEPFRDYVDRFFKTLRAEQATQDV, all individuals (n = 10) expressed either a variant of HLA-B*15 or HLA-B*1401 or Cw*03. These data show that epitopes recognized frequently in the cohort were associated with more than one HLA allele, and partly substantiate the promiscuous nature of epitope restriction within this cohort.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Two separate studies have shown that while p17 is highly targeted in subtype B HIV-1-infected individuals, p24 was mostly targeted by subtype C-infected individuals in Africa (39, 40). In contrast, we identified epitope-rich regions in both p17 and p24 that are targeted by CD8+ T cells, and the peptide stretch, TGTEELRSLYNTVATLYCVHAGIEV, in p17 was recognized at the highest frequency (35%) and contained multiple epitopes. Characterization of targeted regions in p17 revealed the presence of both previously described epitopes, identified from subtype B and C HIV-1 infection, and three novel epitopes. The HLA-A*02-restricted SL-9 epitope has been widely characterized in subtype B-infected individuals, in which earlier studies showed an inverse correlation between the magnitude of SL-9-A02 tetramer-positive cells with plasma viral load (41, 42). Our data show that individuals in this cohort recognized the RY-11 epitope restricted by HLA-A*30 (43) and LYNTVATLY (LY-9), which was found to be restricted by both HLA-A*2902 and HLA-B*4403. Peptide-binding specificities for HLA-A*2902 and B*4403 are the same at positions 2 (E) and 9 (Y) (44), and these observations provide evidence that epitopes in p17 recognized by subtype C HIV-1-infected individuals overlap with epitopes defined from subtype B HIV-1 infection and are promiscuous in nature. The fine nature of LY-9 restriction was further highlighted by the inability of HLA-A*290201 to present the epitope. A previous study demonstrated that a phenylalanine (F) substitution for tyrosine (Y) at position 3 within SL-9 resulted in loss of recognition of an HLA-A*0201 peptide/MHC tetramer folded around SLYNTVATL, suggesting that TCR recognition is sensitive to a single amino acid change at this position (45). It was interesting to note that 52% of the major viral population infecting the individuals investigated had an F to Y substitution at position 3, even though the consensus peptides used in this study contained the SLYNTVATL sequence. It is likely that the LY-9 and RY-11 epitope restrictions and TCR recognition do not have the same constraints as for SL-9 and HLA-A*0201. We also identified two further novel epitopes, HLVWASREL (HL-9) and LVWASRELERF (LF-11), in p17 between aa 9–46, restricted by Cw*0804 for HL-9 and HLA A*30 for LF-11. A similar epitope in this region, WASRELERF, restricted by HLA-B*35, has been described for subtype B-infected individuals (41).
Regions recognized in p24 contain previously described epitopes that also matched previously described restricting HLA alleles. For example, the TL-9 epitope has been shown to be restricted by B*07 in Caucasians and by B*42 or B*81 in Africans (39, 40, 41). Our study has confirmed TL-9 epitope recognition and has further shown that it can be HLA restricted by HLA-B*81 (39, 40, 41). Additionally, we demonstrate that the TL-9 epitope can be restricted by both HLA- B*81 and HLA-B*42 alleles in the same individual with a B*42/B*81 haplotype. The highly targeted region in p24 (aa 289–314) located on the C-terminal side of the major homology region (MHR) has been shown to contain multiple epitopes from subtype B HIV-1 infection. The MHR is a 20-aa highly conserved stretch among all replication-competent retroviruses, with the exclusion of spumaretroviruses, and is virtually identical between different HIV-1 subtypes, HIV-2, and SIV (41, 46). The functional importance of this region has been highlighted when site-specific mutations abolished viral replication and significantly reduced the particle-forming ability of the mutant gag gene products. The DA-9 epitope, an immunodominant epitope described in a long-term nonprogressing individual (47), is located at the C terminus of the MHR, and we report that it was recognized by subtype C-infected individuals and is restricted by HLA-B*14. Mutations in this epitope that abrogated CTL recognition strongly impaired viral replication, whereas replication-competent viral variants were recognized by CTL (47). Previous studies have shown that the YVDRFFKTL epitope variant is restricted by A*2601 and B*70 in Caucasians (41, 44), and the YVDRFFKRL variant is restricted by B*1510 in Africans (32), and both variants are equally recognized by A*0207 in Thais (45). We show that the YVDRFFKTL variant is restricted by B*1503 in this cohort (40, 45, 48, 49). It was interesting to note that some of the individuals with an HLA-B*1510 background and who recognized the peptide stretch aa 289–314 recognized the TD-10 epitope, which we show to be restricted by -Cw*0304, suggesting preferential epitope processing. The known peptide-binding specificity for HLA-Cw*0304 indicates a requirement for alanine at position 2 and either L or M at position 9, and judging from the sequence of this novel epitope, it is evident that the epitope does not follow this prediction (44). A second novel epitope, PF-10, also located on the C-terminal end of the MHR region, was identified, but due to shortage of cells, we were unable to determine the restricting allele.
These observations suggest that there might be preferences in epitope processing and presentation, and would argue against the use of a polyepitope vaccine to prime the immune system, despite taking into account a comprehensive HLA coverage of the target population group. We have provided evidence that it is possible to measure CTL responses to Gag using consensus peptides in subtype C-infected individuals, and that the majority of these responses target conserved regions in Gag. However, fine epitope mapping of these regions has revealed the presence of a number of novel and promiscuous epitopes that overlap or are embedded within previously described epitopes from subtype B-infected Caucasians. We propose that HLA allele divergence between populations will dictate which epitopes are recognized in infected individuals, and the presence of a specific allele does not always translate into presentation of a predicted epitope. For example, in this study, an individual expressing HLA-B*1510 recognized the peptide stretch that contained the B*1503-restricted YL-9 epitope. However, this individual restricted the TD-10 epitope through HLA-Cw*0304, and not the B allele. Thus, although both B*1510 and B*1503 belong to the same supertype family, we did not observe the expected restriction. Our observations suggest that a vaccine based on known poly-epitopes may not elicit expected epitope-specific responses, or possibly no responses, from different populations with unique HLA backgrounds. It is therefore important to allow expression and processing of epitopes using full-length Gag. In summary, despite the highly conserved nature of this region, we have shown the presence of novel epitopes and novel restricting HLA alleles, confirming the immunogenicity of this region of p24. These data highlight the degenerate nature of epitope binding in which different class I HLA molecules across -A, -B, and -Cw alleles obviously share some level of common sequence motifs. Structural similarities in MHC-binding clefts have given rise to HLA supertype groupings (50, 51), although our data show that degenerate epitope binding does not adhere to known supertypes. These data extend and confirm recent data, inferring that similar epitope specificities are observed between different populations expressing divergent HLA class I alleles (52).
To conclude, we have characterized responses in Gag and provide evidence that subtype C HIV-1-infected southern Africans target conserved regions in Gag. Epitope mapping has identified the presence of novel epitopes that are embedded and overlap with previously described epitopes. We have identified novel restricting alleles, due to the unique HLA class I distribution in southern Africa, and provided evidence for promiscuous CTL epitope recognition. Collectively, these data support our hypothesis that HLA diversity between populations can shape fine epitope specificity, but may not represent a constraint for eliciting universal responses, regardless of HIV-1 subtype or host genetic diversity.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Funding was through National Institutes of Health Grant N01-AI-45202. A.M.M. is a South African AIDS Vaccine Initiative fellow; A.M.M. and T.M. are both Fogarty International Fellows. 
2 Address correspondence and reprint requests to Dr. Clive M. Gray, Private Bag X4, Sandringham 2131, Johannesburg, South Africa. E-mail address: cgray{at}nicd.ac.za
3 The HIVNET 028 Study Team is comprised of: Clive Gray (co-chair); Haynes Sheppard (co-chair); Rosemary Masondu, Susan Allen, and Michelle Klautzman (Zambia University Teaching Hospital, Lusaka, Zambia; University of Alabama, Birmingham, AL); Newton Kumwenda (Malawi College of Medicine, Blantyre, Malawi); Taha Taha (Johns Hopkins University, Baltimore, MD); Lynn Zijenah, Victoria Aquino, Michael Chirenje, Mike Mbizo, and Ocean Tobaiwa (University of Zimbabwe, Harare, Zimbabwe); David Katzenstein (Stanford University, Palo Alto, CA); Glenda Gray, James McIntyre, and Armstrong Mafhandu (Perinatal HIV Research Unit, Chris Hani Baragwanath Hospital, Johannesburg, South Africa); Efthyia Vardas, Mark Colvin, Dudu Msweli, Wendy Dlamini, Gita Ramjee, Salim Abdool Karim, and Quarraisha Abdool Karim (Medical Research Council, Durban, South Africa); Lynn Morris and Natasha Taylor (National Institute for Communicable Diseases, Sandringham, Johannesburg, South Africa); Carolyn Williamson, Helba Bredell, and Celia Rademayer (University of Cape Town, Cape Town, South Africa); Jorge Flores (Division of AIDS, National Institutes of Health, Besthesda, MD); and Ward Cates, Linda McNeil, and Missie Allen (Family Health International, Triangle Park, NC).
4 Abbreviations used in this paper: ICS, intracellular cytokine staining; MHR, major homology region; sfu, spot-forming unit; BLCL, B lymphoblastoid cell line.
Received for publication April 8, 2004. Accepted for publication August 2, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
interferon-secreting CD8+ T cells in primary HIV-1 infection. J. Virol. 77:6867.-interferon-producing CD8+ T cells in infected children correlate positively with plasma viral load. J. Virol. 76:12414.This article has been cited by other articles:
|
|
 |
|
  A. Abraha, I. L. Nankya, R. Gibson, K. Demers, D. M. Tebit, E. Johnston, D. Katzenstein, A. Siddiqui, C. Herrera, L. Fischetti, et al. CCR5- and CXCR4-Tropic Subtype C Human Immunodeficiency Virus Type 1 Isolates Have a Lower Level of Pathogenic Fitness than Other Dominant Group M Subtypes: Implications for the Epidemic J. Virol., June 1, 2009; 83(11): 5592 - 5605. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Gray, M. Mlotshwa, C. Riou, T. Mathebula, D. de Assis Rosa, T. Mashishi, C. Seoighe, N. Ngandu, F. van Loggerenberg, L. Morris, et al. Human Immunodeficiency Virus-Specific Gamma Interferon Enzyme-Linked Immunospot Assay Responses Targeting Specific Regions of the Proteome during Primary Subtype C Infection Are Poor Predictors of the Course of Viremia and Set Point J. Virol., January 1, 2009; 83(1): 470 - 478. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. K. Chege, E. G. Shephard, A. Meyers, J. van Harmelen, C. Williamson, A. Lynch, C. M. Gray, E. P. Rybicki, and A.-L. Williamson HIV-1 subtype C Pr55gag virus-like particle vaccine efficiently boosts baboons primed with a matched DNA vaccine J. Gen. Virol., September 1, 2008; 89(9): 2214 - 2227. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L. Perez, M. V. Larsen, R. Gustafsson, M. M. Norstrom, A. Atlas, D. F. Nixon, M. Nielsen, O. Lund, and A. C. Karlsson Broadly Immunogenic HLA Class I Supertype-Restricted Elite CTL Epitopes Recognized in a Diverse Population Infected with Different HIV-1 Subtypes J. Immunol., April 1, 2008; 180(7): 5092 - 5100. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Turk, M. M. Gherardi, N. Laufer, M. Saracco, R. Luzzi, J. H. Cox, P. Cahn, and H. Salomon Magnitude, Breadth, and Functional Profile of T-Cell Responses during Human Immunodeficiency Virus Primary Infection with B and BF Viral Variants J. Virol., March 15, 2008; 82(6): 2853 - 2866. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. O. Peters, M. G. Mendoza, R. E. Capina, M. Luo, X. Mao, M. Gubbins, N. J. D. Nagelkerke, I. MacArthur, B. B. Sheardown, J. Kimani, et al. An Integrative Bioinformatic Approach for Studying Escape Mutations in Human Immunodeficiency Virus Type 1 gag in the Pumwani Sex Worker Cohort J. Virol., February 15, 2008; 82(4): 1980 - 1992. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Nakagawa, K. H. Kim, T. M. Gillam, and A.-B. Moscicki HLA Class I Binding Promiscuity of the CD8 T-Cell Epitopes of Human Papillomavirus Type 16 E6 Protein J. Virol., February 1, 2007; 81(3): 1412 - 1423. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Li, A. D. Gladden, M. Altfeld, J. M. Kaldor, D. A. Cooper, A. D. Kelleher, and T. M. Allen Rapid Reversion of Sequence Polymorphisms Dominates Early Human Immunodeficiency Virus Type 1 Evolution J. Virol., January 1, 2007; 81(1): 193 - 201. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Honeyborne, A. Rathod, R. Buchli, D. Ramduth, E. Moodley, P. Rathnavalu, S. Chetty, C. Day, C. Brander, W. Hildebrand, et al. Motif Inference Reveals Optimal CTL Epitopes Presented by HLA Class I Alleles Highly Prevalent in Southern Africa. J. Immunol., April 15, 2006; 176(8): 4699 - 4705. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Zuniga, A. Lucchetti, P. Galvan, S. Sanchez, C. Sanchez, A. Hernandez, H. Sanchez, N. Frahm, C. H. Linde, H. S. Hewitt, et al. Relative Dominance of Gag p24-Specific Cytotoxic T Lymphocytes Is Associated with Human Immunodeficiency Virus Control J. Virol., March 15, 2006; 80(6): 3122 - 3125. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Geels, S. A. Dubey, K. Anderson, E. Baan, M. Bakker, G. Pollakis, W. A. Paxton, J. W. Shiver, and J. Goudsmit Broad Cross-Clade T-Cell Responses to Gag in Individuals Infected with Human Immunodeficiency Virus Type 1 Non-B Clades (A to G): Importance of HLA Anchor Residue Conservation J. Virol., September 1, 2005; 79(17): 11247 - 11258. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), A*0205/2902, B*1401/4403, Cw*07/08; (*), A*23/66, B*39/42, Cw*12/17; (
), A*2301/6802, B*0702/4403, Cw*03/07; (
), A*2301/7401, B*1503/4504, Cw*02/07; (
), A*02/30, B*4501/5802, Cw*06/16; (
), A*2301/3004, B*4403/5301, Cw*02/06; D, (
), A*2902/4301, B*15/44, Cw*04/07; (
), A*2301/2902, B*08/13, Cw*06/07; (
), A*290201/6802, B*1302/440301, Cw*04/06; (
