Selection of T-cell vaccine antigens for chronic persistent viral infections has been largely empirical. To define the relationship, at the population level, between the specificity of the cellular immune response and viral control for a relevant human pathogen, we performed a comprehensive analysis of the 160 dominant CD8(+) T-cell responses in 578 untreated HIV-infected individuals from KwaZulu-Natal, South Africa. Of the HIV proteins targeted, only Gag-specific responses were associated with lowering viremia. Env-specific and Accessory/Regulatory protein-specific responses were associated with higher viremia. Increasing breadth of Gag-specific responses was associated with decreasing viremia and increasing Env breadth with increasing viremia. Association of the specific CD8(+) T-cell response with low viremia was independent of HLA type and unrelated to epitope sequence conservation. These population-based data, suggesting the existence of both effective immune responses and responses lacking demonstrable biological impact in chronic HIV infection, are of relevance to HIV vaccine design and evaluation.

Human immunodeficiency virus (HIV)-infected infants in sub- 00083). Gag-specific CD4؉ T-cell responses were minimal to undetectable in the first 6 months of pediatric infection. These data indicate that failure to control HIV replication in in utero-infected infants is not due to an inability to induce responses but instead suggest secondary failure of adaptive immunity in containing this infection. Moreover, the detection of virusspecific CD8 ؉ T-cell responses in the first days of life in most in utero-infected infants is encouraging for HIV vaccine interventions in infants.

In HIV-infected persons, certain HLA class I alleles are associated with effective control of viremia, while others are associated with rapid disease progression. Among the most divergent clinical outcomes are the relatively good prognosis in HLA-B*5801 expressing persons and poor prognosis with HLA-B*5802. These two alleles differ by only three amino acids in regions involved in HLA-peptide recognition. This study evaluated a cohort of over 1000 persons with chronic HIV clade C virus infection to determine whether clinical outcome differences associated with B*5801 (n = 93) and B*5802 ( n = 259) expression are associated with differences in HIV-1-specific CD8 (+) T cell responses. The overall breadth and magnitude of HIV-1-specific CD8(+) T cell responses were lower in persons expressing B*5802, and epitope presentation by B*5802 contributed significantly less to the overall response as compared to B*5801-restricted CD8 (+) T cells. Moreover, viral load in B*5802-positive persons was higher and CD4 cell counts lower when this allele contributed to the overall CD8 (+) T cell response, which was detected exclusively through a single epitope in Env. In addition, persons heterozygous for B*5802 compared to persons homozygous for other HLA-B alleles had significantly higher viral loads. Viral sequencing revealed strong selection pressure mediated through B*5801-restricted responses but not through B*5802. These data indicate that minor differences in HLA sequence can have a major impact on epitope recognition, and that selective targeting of Env through HLA-B*5802 is at least ineffectual if not actively adverse in the containment of viremia. These results provide experimental evidence that not all epitope-specific responses contribute to immune containment, a better understanding of which is essential to shed light on mechanisms involved in HIV disease progression.

HIV-specific CTL play a central role in immune control of HIV. The basis for understanding the success or failure of this immune response requires identification of the specific epitopes targeted by CTL. However, in populations most severely affected by the global epidemic, this fundamental knowledge is hindered by the lack of characterization of many of the HLA class I alleles highly prevalent in such populations. Overall, the peptide-binding motif has been determined for a small minority (9%) of HLA class I alleles, with a strong bias toward those alleles prevalent in Caucasoid populations. These studies therefore set out to define, in a South African Zulu/Xhosa population at the epicenter of the epidemic, the epitopes presented by alleles highly prevalent, but for which the peptide-binding motif had not been characterized. Using a method of motif inference, epitopes presented by four such alleles prevalent in the Zulu/Xhosa population of Durban, South Africa, namely, B*3910, B*4201, B*8101, and Cw*1801, are described. Importantly, this approach may additionally facilitate optimization of epitopes in certain instances where conflicting reports in the literature exist regarding the peptide-binding motif, such as for HLA-A*2902, also highly prevalent in southern African populations. These data indicate that the previously anomalous position of HLA-A*2902 among HLA-A alleles, outside any recognized HLA-A supertype, is artifactual, and the true position of the A*2902 motif overlaps those of the A1 and A24 supertypes.