Group M-based HIV-1 Gag peptides are frequently targeted by T cells in chronically infected US and Zambian patients : AIDS

Secondary Logo

Journal Logo


Group M-based HIV-1 Gag peptides are frequently targeted by T cells in chronically infected US and Zambian patients

Bansal, Anjua,*; Gough, Ethana,*; Ritter, Douga,*; Wilson, Craigb; Mulenga, Josephc; Allen, Susand; Goepfert, Paul Aa,e

Author Information
AIDS 20(3):p 353-360, February 14, 2006. | DOI: 10.1097/01.aids.0000206501.16783.67
  • Free



One of the greatest obstacles to the development of a vaccine to combat HIV is the enormous genetic diversity of this virus [1–3]. Even within the same subtype, viral strains can vary by as much as 25% in the amino acids of the variable proteins [4–6]. The diversity among different subtypes is even higher, so determining which HIV immunogens are to be included in a vaccine has become a major problem. Furthermore, the development of reagents that are needed to evaluate vaccine induced responses has been a challenging task due to this tremendous viral diversity.

The development of an antibody-based preventative HIV vaccine has been hampered by antigen-specific responses that have been type specific and generally unable to neutralize primary isolates [7,8]. In contrast, vaccine or infection-induced HIV-specific CD8 T-cell responses can respond to target cells infected by a variety of primary isolates from different subtypes [9–12]. These responses have also been demonstrated to play an important role in the protection of HIV disease progression and were demonstrated in animal models to be protective against AIDS [13,14]. Therefore, numerous HIV vaccine strategies are attempting to elicit antigen-specific CD8 T-cell responses in uninfected human participants [15–17].

Despite the fact that CD8 T-cell responses are cross-subtype, HIV is so diverse that vaccine design based on the knowledge of the local circulating strains is felt to be necessary [18,19]. This strategy has led to an ever expanding number of different viral strains and subtypes becoming incorporated into HIV vaccine products ( In order to logically decrease the amount of diversity among the many subtypes, several groups have designed immunogens based on computational methods that minimize sequence diversity among viral isolates [20–22]. All of these products are designed based upon information derived from the current HIV sequence database. Consensus M sequences represent the most frequent amino acid residue seen across all viral isolates irrespective of subtype, but subtype consensus sequences (i.e. consensus A, B, C, etc.) have also been designed [23]. Other examples include ancestral sequences (constructed from an evolutionary model) and center of tree (similar to ancestor but uses an unrooted phylogeny) [22,24]. These predicted sequences were originally designed as potential vaccine immunogens, but may also be useful for the purposes of designing reagents to measure T-cell immune responses from HIV vaccine recipients.

Because it is immunogenic and relatively conserved, the Gag protein is a component in virtually all CD8 T-cell-based vaccines [25–30] and We hypothesized that immune responses to consensus or ancestral M Gag peptides will be of equal frequency and magnitude in subtype B-infected patients when compared with consensus B peptides. The goal of the current study was to compare the ability of different peptide sets to elicit Gag-specific responses in T cells obtained from patients with chronic HIV infection using 15-mer peptides overlapping by 11 amino acids based upon the Gag sequences of varying design (consensus A, B, C, M, ancestral B, M, and HIVHXB2).

Materials and methods

Study subjects

Peripheral blood mononuclear cells (PBMCs) were obtained from 43 HIV-1-seropositive patients attending the University of Alabama at Birmingham (UAB) 1917 HIV Clinic. Nineteen subjects were on antiretroviral therapy (ART) and 24 were treatment naive or had discontinued ART. PBMC from 11 healthy seronegative donor subjects in the US were used as negative controls. For HIV-1 subtype C-infected subjects, thirteen seropositive and five seronegative individuals were chosen from the Zambia-Emory HIV Research Project (ZEHRP). These studies were approved by the institutional review board at UAB and University of Zambia Research Ethics Committee in Lusaka, Zambia. Informed consent was obtained from all study participants in accordance with the human experimentation guidelines of the US Department of Health and Human Services.

Blood processing and shipment

PBMC from fresh blood were processed at the ZEHRP laboratory in Lusaka, Zambia. Frozen PBMC from Zambia were then shipped on liquid nitrogen to the US where the enzyme-linked immunosorbent spot (ELISpot) assays were performed on these samples in parallel with the samples obtained from UAB AIDS clinic.


Sets of 15-mer peptides (overlapping by 11) spanning the complete Gag consensus protein sequences for HIV-1 subtypes A (con A), B (con B), C (con C), and the subtype-B clone HXB2 were obtained from the National Institutes of Health AIDS Reagent Program (NIH; Bethesda, Maryland, USA). For designing consensus M (con M), ancestral B (anc B) and M (anc M) peptides, year 2002 sequence alignments were used from the HIV sequence database at The unique peptides in this group that differed from the reagents obtained above from the NIH were synthesized by Polymx (Huntsville, Alabama, USA). All the peptides were reconstituted at 2 mg/ml and the final concentration of each peptide used was 2 μg/ml.

Interferon-γ ELISpot assay

An interferon (IFN)-γ ELISpot assay was performed as described in prior studies [31,32]. In brief, ELISpot plates (MAHAN# 4550) from Millipore Corporation (Bedford, Massachusetts, USA) were coated overnight with anti-IFN-γ mAb (no. 1DIK) from Mabtech Inc. (Mariemont, Ohio, USA). The following day, the cells were plated at 1 × 105 cells per well and incubated with peptide pools/peptide for 20–24 h. For spot detection, biotinylated anti-IFN-γ (Mabtech no. 7B61) followed by streptavidin-AP and BCIP substrate (no. 7100-04 and no. 3032-01, respectively; Southern Biotech, Birmingham, Alabama, USA) were used. Spots were counted using a CTL ‘Immunospot’ plate reader (CTL Analyzers, Cleveland, Ohio, USA) and expressed as SFC (spot forming cells) per 106 PBMC. The cut-off for a positive response was the average spot count from unstimulated ‘media only’ wells plus 3 × standard deviation of all media wells per individual. Positive responses from two overlapping peptides were scored as a single response. The unstimulated control (i.e. media alone) was done in quadruplicate. The experimental and the positive control (PHA) were done in duplicate.

For the US subtype B-infected samples, cryopreserved PBMC were stimulated with 22 pools (10–13 peptide/pool) representing the entire Gag proteome in a matrix format, for each of the above seven different Gag sequences. A second IFN-γ ELISpot assay using individual peptides was performed on positive responses as detected in the matrix format to further elucidate and confirm results.

For the Zambian samples, we were limited by the amount of PBMCs available for some subjects. Hence in such instances, the mapping of T-cell responses with pools of peptides in the first step did not always include all seven Gag sets. In most of these subjects the responses were tested to Consensus C, Consensus B and Consensus A. However, in the second step when the responses were narrowed down to the individual peptides, we included peptides representing all seven different Gag sequences. Breadth in this study is defined by the total number of peptide pools recognized. The magnitude is defined either as the sum total of the reactivity of all the peptides in a pool or just to a single peptide pool.

Amino-acid conservation

The amino-acid conservation between con A, con B, con C, con M, anc M, anc B and HXB2 in a pair wise combination was determined by using the align query program at


A Spearman rank correlation test was used to compare the linearity of the magnitude of Gag and its subunit protein responses directed towards consensus B relative to the other six Gag reagent sets. Differences in peptide recognition between the seven sets of immunogens were analyzed using the Fisher exact test.


T-cell responses in HIV-infected US patients

Based on the percentage amino-acid conservation among the seven different sets of Gag proteins (Table 1), it is predicted that in a US HIV-1 subtype B-infected cohort, the con M and anc M peptides will be recognized in similar frequencies as con B, anc B and HXB2 peptides. Similarly, for consensus based on subtype C, it is predicted that in individuals infected with subtype C, peptides based on con M and anc M will show similar cross reactivity as con C.

Table 1:
Amino-acid conservation (%) among seven different HIV-1 Gag sequencesa.

PBMC from 43 chronically subtype B-infected US patients were used in IFN-γ ELISpot assays to test the cross-reactivity of cellular immune responses to consensus and ancestral sequence based peptides. Of these subjects 72% were males; the median age of the participants and the duration of infection were 38 and 7 years, respectively. The median absolute CD4 T-cell count (cells/μl) was 437 with a plasma viral load of 733 RNA copies/ml. HIV Gag-specific responses were detectable in 34 of the 43 (79%) HIV-infected patients tested using at least one of the peptide sets. The mean and median responses for each of the seven reagent sets tested were similar suggesting that adequate cross reactivity existed. In contrast, only one of 11 seronegative volunteers had a weakly positive low magnitude response (105 SFC/million). This low frequency response did not map to any 15-mer peptide upon routine confirmatory testing.

T-cell responses to 42 of the total 123 tested peptides (34% responses using any of the seven Gag immunogens) were detected in these 34 subjects: eight in p17, 28 in p24, and six in the p2, p7, p1, and p6 proteins of Gag. Due to the low frequency of responses seen in these smaller proteins, they are cumulatively referred to as the p15 throughout. The number of subjects recognizing each peptide varied from one (3%) to 11 (32%) with the latter peptide encompassing the well-described HLA-A2-restricted SL9 peptide in p17. The second most recognized peptide (18%) contains the A3 restricted RK-9 epitope. Interestingly, several individuals targeted these two peptides that did not carry the A2 or A3 alleles; however, all of these individuals carried other HLAs that are known to restrict epitopes within these two peptides (data not shown). The T-cell responses were mostly clustered in the p24 region, which has relatively low sequence variability as shown by prior studies [33]. The patient cohort was not selected to determine clinical correlates of protection since several patients were on therapy and many had been on therapy previously. Nevertheless, when looking at either the entire cohort or patients not currently receiving ART, no significant association between the magnitude of the IFN-γ response and plasma viral load, absolute CD4 T-cell count, or ART was seen.

The magnitude of positive responses varied greatly between peptides (Fig. 1). Fifteen of the 42 peptides were uniquely recognized by lone subjects. Of the eight peptides recognized in p17, nearly 50% responses were greater than 500 SFC/million PBMC. Only 10 of 28 (36%) and one of six (17%) responses in p24 and p15, however, had greater than 500 SFC/million PBMC. Overall of the 42 recognized peptides, 29 (69%) were completely cross-reactive (for all peptide sets tested) for at least one subject. At the subunit protein level this translated into five of eight (63%); 20 of 28 (71%) and four of six (67%) cross-recognition for p17, p24 and p15, respectively. Note that thirteen of the 42 responding peptides displayed varying degrees of cross reactivity for the seven different peptide sets. Remarkably, only one Gag peptide (2%) was only recognized by a single peptide set (HXB2 peptide 120).

Fig. 1:
Mean magnitude of the interferon (IFN)-γ enzyme-linked immunosorbent spot (ELISpot) response to Gag peptides. The magnitude of response to each of the seven Gag peptide sets in an IFN-γ ELISpot assay for all the 42 recognized peptides is shown. For each peptide, the peptide number is shown as well as its first amino-acid number position in HXB2 (shown in parenthesis). The peptides eliciting cross-reactive responses are marked with an asterisk (*). The peptides underlined represent data from lone individuals only. PBMC, peripheral blood mononuclear cells; SFC, spot forming cells.

In general, the frequency of individual peptide responses was similar, as detected by the different peptide sets (Fig. 2a). An exception is seen for the subtype A p24 peptides which were recognized less frequently when compared to the B and M derived p24 peptides (P < 0.05 by Fisher exact, Fig. 2a). The con M and ancestral M peptides were comparable to con B in eliciting T-cell responses for both the variable (p17 and p15) and conserved (p24) regions of Gag.

Fig. 2:
Peptides eliciting a positive interferon (IFN)-γ response in Gag. The percentage of peptides that elicited a positive IFN-γ enzyme-linked immunosorbent spot (ELISpot) to each of the seven peptide sets is shown. The responses are shown for p17, p24, p15 and Gag (a). The data represent the percentage of peptides that resulted in a positive response compared to all 123 peptides tested. Venn diagrams depicting Gag-specific T-cell responses in 43 clade B infected US patients that were detected by one, two or all three peptides sets using an IFN-γ ELISpot assay are shown in panels (b)–(d). The number of responses elicited against con B, con M and HXB2 (b); con B, con M and anc M (c) and con B, con C and con M (d) are shown.

Using the peptides based on con B, con M and HXB2 (Fig. 2b), a total of 101 responses were observed. Nearly 90% of these responses were detected using all three peptide sets. Five responses were common to con B and HXB2 and four and two peptide-specific responses were unique to con M and HXB2, respectively. When the group M sequence-based peptide sets (consensus and ancestral) were compared against con B reagents, a total of 98 responses were detected of which 89% were again detected by all three peptide sets (Fig. 2c). Consensus B and anc M had four and two responses that were recognized by these two peptide sets, respectively. No response was unique to con M although it shared responses with either con B (two) or anc M (three). Importantly, only seven responses (7% of the total) would have been missed if con M was solely used to analyze IFN-γ ELISpot responses when compared with con B or HXB2 in subtype B-infected patients. Comparing the recognition of T-cell responses by con B, con C and con M-based reagents yielded similar findings (Fig. 2d).

Next, we correlated the magnitude of responses to each recognized peptide in Gag among the different groups using the Spearman rank's correlation (Table 2). Significant correlation was observed for all comparisons between con B and the remaining six peptides sets. As expected, the highest correlations for Gag or its subunit protein were noted when the magnitude of con B responses was compared to either anc B or HXB2. This was followed closely by the con M-based sequences with the strongest correlations seen for p24 (Table 2).

Table 2:
Correlation of the magnitude of responses as measured by interferon-γ enzyme-linked immunosorbent spot assay.

T-cell responses in HIV-infected Zambians

PBMCs from thirteen HIV-1 seropositive and five seronegative Zambians were also tested for cross-reactivity of Gag-specific T-cell responses. Although CD4 T-cell counts and plasma viral load were not performed on these subjects, the HIV-1 seropositive patients (69% female) were generally in good health with only one having an AIDS-associated illness (varicella zoster), which was not active at the time of blood sampling. Gag-specific responses were detected in 10 of 13 (77%) seropositive subjects at the first tier of mapping. A low frequency response was noted for two Gag peptides (80 and 90 SCF/million) in one seronegative donor; however, this response did not map to a 15-mer peptide upon routine confirmatory testing. Due to sample availability in only six of the 13 infected subjects the responses were mapped to the individual Gag peptides. All six subjects recognized epitopes in p24, with one subject having a single additional response in p17 (Table 3). Nine individual responses were detected among six Gag epitopes, with six being completely cross-reactive for all sequences tested. All nine responses were detected using consensus C and M peptides; eight of nine using consensus A, B, HXB2, and ancestral M peptides; and seven of nine using peptides from the ancestral B sequence (data not shown). In addition, the sum and mean magnitude response tended to be highest for con C, con M and anc M (Table 3). Comparison of responses to con B, con C and con M immunogens in this more limited number of Zambian subjects showed that eight responses were detected that were common to all three and only one response was unique to con C and con M.

Table 3:
Enzyme-linked immunosorbent spot assay responsesa in Zambian subjects infected with HIV-1 clade C virus.


This is the first study that has evaluated the cross reactivity of T-cell responses to Gag when stimulated with seven different sets of immunogens namely the consensus subtypes A, B, C and M; ancestral M and B; and HXB2 sequences. The M group consensus, derived from the most common amino acid residue as obtained from the aligned HIV-1 sequences in the Los Alamos database, reduces the average distance to the various HIV-1 sequences to 5 to 15%. This distance is similar to the intra-subtype sequence distances between contemporary strains and half that of the inter-subtype distance which is 10 to 30% [22–24].

The findings in this study demonstrated that nearly 69% of the responses in Gag for the US subtype B-infected subjects were cross-reactive. Our data also shows that con M and anc M Gag elicits T-cell responses that are equivalent to those that are directed to the infecting subtype and subtype-based consensus sequences missing only 5% of Gag responses in subtype B-infected patients when compared with the same analysis using con B peptides (Fig. 2b). As expected, gag sequencing in three individuals from the US cohort revealed subtype B infection exclusively. The amino acid conservation from these three representative sequences were most similar to con B (94.6%), intermediately similar to con M and anc M (88.6 and 89.2%, respectively), and least similar to con A (83.8%). The latter observation is consistent with our findings that con A was the least recognized by the US patients in our cohort (Fig. 2a). In the limited number of subtype C-infected subjects analyzed, both con M and anc M sequences elicited similar levels of responses as con C. Hence using immunogens based on con M may be able to overcome the limitations imposed by subtype and subtype-based consensus sequences at least for Gag. However, more detailed studies are needed in larger number of patients and also patients infected with other subtypes to extend and confirm these observations.

These findings are in contrast to those seen by Rutebemberwa and colleagues [34] who measured Env-specific responses in HIV-infected US patients using con B and con M and HIV-1MN (strain specific) sequences. This study demonstrated relatively little cross-reactivity among the tested peptide sets with only two (6%) peptides recognized by all three sequences. In contrast, when we compared Gag-specific responses elicited by con B and con M and HIV-1HXB2, 90% were recognized when using any of these peptide sets. The discrepancy in the results between these two studies may be due to the variability inherent in the Env [34] and Gag proteins (this study) of HIV-1. Gag is much more conserved than Env as a whole with p24 being the second most conserved HIV protein (next to Integrase) [33]. Even the more variable proteins within Gag (p17 and others) have a higher degree of sequence conservation than Env. Although consensus sequences effectively narrow the degree of variability among viral isolates, these may still not be sufficient to detect responses elicited by highly variable proteins.

The variability of Gag and Env protein is influenced by the selection pressure driving evolutionary changes in these proteins. In Gag, it is the escape from CTL [35] and for Env it is the escape from neutralizing antibodies [36,37] that shapes the sequence of these two proteins. Since Env has developed to escape both CTL and neutralizing antibodies, it may be much more flexible having little fitness costs associated with escape mutations. However, this would not explain why consensus Env sequences were shown to be excellent reagents to detect cross-subtype binding antibody responses in both subtype B and C-infected patients [22,24]. Furthermore, it will be important to examine the cross-reactivity of peptide sets based on group M for other regions of HIV-1 that have also been included in vaccines (e.g. Pol and Nef).

Rutebemberwa et al. [34] also demonstrated that none of the observed IFN-γ ELISpot responses in their study were due to CD4 T cells. Similarly, we did not confirm the CD8 T-cell phenotype of the IFN-γ ELISpot responses in this study; however, we have previously demonstrated [31,38] that the vast majority of the latter responses are derived from CD8 T cells when performing the assay in chronically infected patients.

Most of the reagents available are subtype-specific and recently peptides based on consensus sequences have been made. This study should aid in the design of a ‘universal’ reagent set that will have widespread application for screening the current and future vaccine candidates using immunogens from diverse subtypes. This peptide set will probably comprise pools sorted based on both the hits on the peptide and the entropy or variability of the peptides. Since this reagent peptide set will be based on selected peptides from the entire Gag pool, this should also reduce the number of PBMC that are needed for the IFN-γ ELISpot assay. This strategy would save on the cost and labor-intensive process of synthesizing peptide sets for every subtype vaccine that will need to be tested.

It is tempting to speculate that an HIV-1 vaccine based on consensus M Gag will be able to elicit cross reactive T-cell responses in seronegative participants living in diverse geographic regions. However, this idea will need to be tested in primate (human and non-human) vaccine recipients to determine whether responses elicited by a consensus immunogen is more cross-reactive than that seen by a single viral isolate. Such a study is being undertaken in non-human primates (Norman Letvin, personal communication). Further studies should address the question whether a Gag peptide pool (based upon consensus M sequences or similar) would be a useful tool for evaluating the immunogenicity of different subtype vaccine constructs for use in diverse geographic regions throughout the world.


We wish to thank Chris Perkins for excellent technical assistance and Sameera Vohra for the demographics and clinical data on the US subjects. Our thanks are also due to Steffanie Sabbaj for providing the PBMC samples from healthy Zambian subjects.

Sponsorship: This study was supported by grants from NIH including R01 AI 49126 (P.A.G.) and U19 AI 28147 (Jiri Mestecky). Support for the assays performed as part of these studies came from the Adolescent Medicine Trials Network for HIV/AIDS interventions (ATN) which is supported by NICHD (U01 HD40533) with additional support from NIDA, NIMH and NIAAA.


1. Garber DA, Silvestri G, Feinberg MB. Prospects for an AIDS vaccine: three big questions, no easy answers. Lancet Infect Dis 2004; 4:397–413.
2. McMichael AJ, Hanke T. HIV vaccines 1983–2003. Nat Med 2003; 9:874–880.
3. Nabel G, Makgoba W, Esparza J. HIV-1 diversity and vaccine development. Science 2002; 296:2335.
4. Walker BD, Korber BT. Immune control of HIV: the obstacles of HLA and viral diversity. Nat Immunol 2001; 2:473–475.
5. Robertson DL, Anderson JP, Bradac JA, Carr JK, Foley B, Funkhouser RK, et al. HIV-1 nomenclature proposal. Science 2000; 288:55–56.
6. McCutchan FE. Understanding the genetic diversity of HIV-1. Aids 2000; 14(suppl 3):S31–S44.
7. Coeffier E, Excler JL, Kieny MP, Meignier B, Moste C, Tartaglia J, et al. Restricted specificity of anti-V3 antibodies induced in humans by HIV candidate vaccines. AIDS Res Hum Retroviruses 1997; 13:1471–1485.
8. Li M, Gao F, Mascola JR, Stamatatos L, Polonis VR, Koutsoukos M, et al. Human immunodeficiency virus type 1 env clones from acute and early subtype B infections for standardized assessments of vaccine-elicited neutralizing antibodies. J Virol 2005; 79:10108–10125.
9. Durali D, Morvan J, Letourneur F, Schmitt D, Guegan N, Dalod M, et al. Cross-reactions between the cytotoxic T-lymphocyte responses of human immunodeficiency virus-infected African and European patients. J Virol 1998; 72:3547–3553.
10. Ferrari G, Humphrey W, McElrath MJ, Excler JL, Duliege AM, Clements ML, et al. Clade B-based HIV-1 vaccines elicit cross-clade cytotoxic T lymphocyte reactivities in uninfected volunteers. Proc Natl Acad Sci USA 1997; 94:1396–1401.
11. Betts MR, Krowka J, Santamaria C, Balsamo K, Gao F, Mulundu G, et al. Cross-clade human immunodeficiency virus (HIV)-specific cytotoxic T-lymphocyte responses in HIV-infected Zambians. J Virol 1997; 71:8908–8911.
12. Wilson SE, Pedersen SL, Kunich JC, Wilkins VL, Mann DL, Mazzara GP, et al. Cross-clade envelope glycoprotein 160-specific CD8+ cytotoxic T lymphocyte responses in early HIV type 1 clade B infection. AIDS Res Hum Retroviruses 1998; 14:925–937.
13. Shiver JW, Fu TM, Chen L, Casimiro DR, Davies ME, Evans RK, et al. Replication-incompetent adenoviral vaccine vector elicits effective anti-immunodeficiency-virus immunity. Nature 2002; 415:331–335.
14. Caulfield MJ, Wang S, Smith JG, Tobery TW, Liu X, Davies ME, et al. Sustained peptide-specific gamma interferon T-cell response in rhesus macaques immunized with human immunodeficiency virus gag DNA vaccines. J Virol 2002; 76:10038–10043.
15. Hanke T, McMichael AJ, Mwau M, Wee EG, Ceberej I, Patel S, et al. Development of a DNA-MVA/HIVA vaccine for Kenya. Vaccine 2002; 20:1995–1998.
16. Egan MA, Pavlat WA, Tartaglia J, Paoletti E, Weinhold KJ, Clements ML, et al. Induction of human immunodeficiency virus type 1 (HIV-1)-specific cytolytic T lymphocyte responses in seronegative adults by a nonreplicating, host-range-restricted canarypox vector (ALVAC) carrying the HIV-1MN env gene. J Infect Dis 1995; 171:1623–1627.
17. Kent SJ, Zhao A, Best SJ, Chandler JD, Boyle DB, Ramshaw IA. Enhanced T-cell immunogenicity and protective efficacy of a human immunodeficiency virus type 1 vaccine regimen consisting of consecutive priming with DNA and boosting with recombinant fowlpox virus. J Virol 1998; 72:10180–10188.
18. Dorrell L, Dong T, Ogg GS, Lister S, McAdam S, Rostron T, et al. Distinct recognition of non-clade B human immunodeficiency virus type 1 epitopes by cytotoxic T lymphocytes generated from donors infected in Africa. J Virol 1999; 73:1708–1714.
19. Thakar MR, Patke D, Lakhashe SK, Bhonge L, Kulkarni SV, Tripathy SP, et al. Consistent subtype-specific anti-HIV type 1 T lymphocyte responses in Indian subjects recently infected with HIV type 1. AIDS Res Hum Retroviruses 2002; 18:1389–1393.
20. Gao F, Korber BT, Weaver E, Liao HX, Hahn BH, Haynes BF. Centralized immunogens as a vaccine strategy to overcome HIV-1 diversity. Expert Rev Vaccines 2004; 3:S161–S168.
21. Gao F, Weaver EA, Lu Z, Li Y, Liao HX, Ma B, et al. Antigenicity and immunogenicity of a synthetic human immunodeficiency virus type 1 group m consensus envelope glycoprotein. J Virol 2005; 79:1154–1163.
22. Nickle DC, Jensen MA, Gottlieb GS, Shriner D, Learn GH, Rodrigo AG, et al. Consensus and ancestral state HIV vaccines. Science 2003; 299:1515–1518 [author reply 1515–1518].
23. Novitsky V, Smith UR, Gilbert P, McLane MF, Chigwedere P, Williamson C, et al. Human immunodeficiency virus type 1 subtype C molecular phylogeny: consensus sequence for an AIDS vaccine design? J Virol 2002; 76:5435–5451.
24. Gaschen B, Taylor J, Yusim K, Foley B, Gao F, Lang D, et al. Diversity considerations in HIV-1 vaccine selection. Science 2002; 296:2354–2360.
25. AIDS Vaccine Evaluation Group. Cellular and Humoral Immune Responses to a Canarypox Vaccine Containing Human Immunodeficiency Virus Type 1 Env, Gag, and Pro in Combination with RGP120.J Infect Dis 2001; 183:563–570.
26. Belshe RB, Gorse GJ, Mulligan MJ, Evans TG, Keefer MC, Excler JL, et al. Induction of immune responses to HIV-1 by canarypox virus (ALVAC) HIV-1 and gp120 SF-2 recombinant vaccines in uninfected volunteers. NIAID AIDS Vaccine Evaluation Group. AIDS 1998; 12:2407–2415.
27. Belshe RB, Stevens C, Gorse GJ, Buchbinder S, Weinhold K, Sheppard H, et al. Safety and immunogenicity of a canarypox-vectored human immunodeficiency virus Type 1 vaccine with or without gp120: a phase 2 study in higher- and lower-risk volunteers. J Infect Dis 2001; 183:1343–1352.
28. Corey L, McElrath MJ, Weinhold K, Matthews T, Stablein D, Graham B, et al. Cytotoxic T cell and neutralizing antibody responses to human immunodeficiency virus type 1 envelope with a combination vaccine regimen. AIDS Vaccine Evaluation Group. J Infect Dis 1998; 177:301–309.
29. Evans TG, Keefer MC, Weinhold KJ, Wolff M, Montefiori D, Gorse GJ, et al. A canarypox vaccine expressing multiple human immunodeficiency virus type 1 genes given alone or with rgp120 elicits broad and durable CD8+ cytotoxic T lymphocyte responses in seronegative volunteers. J Infect Dis 1999; 180:290–298.
30. Gupta K, Hudgens M, Corey L, McElrath MJ, Weinhold K, Montefiori DC, et al. Safety and immunogenicity of a high-titered canarypox vaccine in combination with rgp120 in a diverse population of HIV-1-uninfected adults: AIDS Vaccine Evaluation Group Protocol 022A. J Acquir Immune Defic Syndr 2002; 29:254–261.
31. Bansal A, Sabbaj S, Edwards BH, Ritter D, Perkins C, Tang J, et al. T cell responses in HIV type 1-infected adolescent minorities share similar epitope specificities with whites despite significant differences in HLA class I alleles. AIDS Res Hum Retroviruses 2003; 19:1017–1026.
32. Sabbaj S, Bansal A, Ritter GD, Perkins C, Edwards BH, Gough E, et al. Cross-reactive CD8+ T cell epitopes identified in US adolescent minorities. J Acquir Immune Defic Syndr 2003; 33:426–438.
33. Yusim K, Kesmir C, Gaschen B, Addo MM, Altfeld M, Brunak S, et al. Clustering patterns of cytotoxic T-lymphocyte epitopes in human immunodeficiency virus type 1 (HIV-1) proteins reveal imprints of immune evasion on HIV-1 global variation. J Virol 2002; 76:8757–8768.
34. Rutebemberwa A, Currier JR, Jagodzinski L, McCutchan F, Birx D, Marovich M, et al. HIV-1 MN Env 15-mer peptides better detect HIV-1 specific CD8 T cell responses compared with consensus subtypes B and M group 15-mer peptides. AIDS 2005; 19:1165–1172.
35. Fernandez CS, Stratov I, De Rose R, Walsh K, Dale CJ, Smith MZ, et al. Rapid viral escape at an immunodominant simian-human immunodeficiency virus cytotoxic T-lymphocyte epitope exacts a dramatic fitness cost. J Virol 2005; 79:5721–5731.
36. Richman DD, Wrin T, Little SJ, Petropoulos CJ. Rapid evolution of the neutralizing antibody response to HIV type 1 infection. Proc Natl Acad Sci USA 2003; 100:4144–4149.
37. Wei X, Decker JM, Wang S, Hui H, Kappes JC, Wu X, et al. Antibody neutralization and escape by HIV-1. Nature 2003; 422:307–312.
38. Bansal A, Gough E, Sabbaj S, Ritter D, Yusim K, Sfakianos G, et al. CD8 T-cell responses in early HIV-1 infection are skewed towards high entropy peptides. AIDS 2005; 19:241–250.

HIV-1 sequence; Gag protein; consensus; ancestral; cross subtype recognition; vaccine

© 2006 Lippincott Williams & Wilkins, Inc.