In this project the specific objectives are to define the unique epidemiologic, clinical, virologic, and immunologic features of HIV infection in developing countries, to determine the viral kinetics associated with perinatal and heterosexual transmission, and to characterize the molecular strains of HIV throughout the world for infectiousness and the immunologic response to cross-clade vaccines. We have established population based cohorts in Uganda, Democratic Republic of the Congo, South Africa, India, China and the U.S. In one cohort in Rakai, Uganda, 788 HIV-discordant couples have been identified, of whom 253 (31.1%) have transmitted infection during a 10-year period. In 397 couples the index partner was male (50.4%) and transmission to the female partner occurred in 118 couples (29.7%); in 376 couples the index partner was female (47.7%) and male infection occurred in 120 instances (31.9%). HIV viral load was the most significant determinant of transmission and the risk of transmission increased 2.5-fold for each log10 increment in HIV viral load. Factors that enhanced transmission were younger age, presence of genital ulcer disease, and early or acute HIV infection. Co-infection with HSV-2 increased HIV viral load in dually infected individuals and enhanced HIV susceptibility in the uninfected individual. HIV viral loads were determined for 256 subjects with early (incident) HIV infection and for 1,293 subjects with later (prevalent) HIV infection. HIV viral load was highest in subjects 25 to 29 years old with incident HIV infection but increased with age in subjects with prevalent HIV infection. Viremia was also higher after seroconversion than in latency and increased with more advanced disease. Genital ulcer disease was consistently associated with higher viral loads in subjects with incident and those with prevalent HIV infections, suggesting that treatment of genital ulcer diseases might reduce HIV viremia. To document viral transmission among partners, molecular sequencing of HIV was performed in samples from 419 HIV-infected individuals, including 84 transmission pairs. Of the 84 transmission pairs, 86% were infected by their sexual partner as indicated by sequence matches. For individuals who reported a monogamous relationship, viral sequences matched in 92% compared to non-monogamous relationships where transmission linkage occurred in 68%. HIV subtype data were determined for 168 index HIV-positive persons and the distribution was subtype A (12.5%), D (72.6%), and recombinant A/D (14.9%). Co-receptor tropism was determined in 108 monogamous couples and we found that subtype A samples were all exclusively CCR5-tropic, subtype D samples (24.6%) were dual CCR5/CXCR4-tropic, and the recombinants were 14.3% dually tropic. These findings are consistent with increased expression of CXCR4 tropism in subtype D infections, resulting in more rapid progression and decreased transmission compared to subtype A infections. To better understand the molecular evolution of HIV-1 viral diversity, we performed a comparative molecular analysis of HIV strains collected from persons in Kinshasa, Democratic Republic of Congo. Analysis of 83 specimens revealed that 44 (53%) had concordant subtypes in three regions (subtype A1, G, D, C, and F1 and CRF O1_AE) and one each had subtypes H and J and sub-subtype A2. The remaining 47% of the samples had mosaic genomes comprising multiple subtype combinations. Similar multi-subtype patterns were also observed in 24 specimens collected 15 years earlier in 1985. Given the complexity of HIV-1 viruses circulating in central Africa, efforts should focus on the development of vaccines that can result in cross-clade immunity. In addition, this high frequency of recombinant viruses, multiple infections, or both would suggest that most of these strains represent mosaic genomes. This would further suggest that the HIV epidemic was well established in central Africa by the early 1980s and that recombinant viruses most likely seeded the early global epidemic. In collaborative studies we examined the hypothesis that HIV viral blips represent random biological and statistical variation around mean steady-state HIV-RNA levels slightly below 50 copies/ml. rather than biologically significant elevations in viremia. Viral blips were detected in nine of ten patients and statistical analysis was consistent with random assay variation and random mean viral load below 50 copies/ml. Blips were of short duration (< 3 days) and of low magnitude (79 median, 79 copies/ml.). Blips did not occur in relationship to illness, vaccination, or directly measured antiretroviral drug concentrations. In approximately 1,000 independent cloned sequences for both protease and reverse transcriptase, no new resistance mutations were seen before, during, or after blips. Most blips in this population appear to represent random biological and statistical variation around the mean of HIV-1 levels below 50 copies/ml rather than clinically significant elevations in viremia. The significance of these studies is that they provide important epidemiologic, clinical, virologic, and immunologic knowledge of HIV infection in developing countries, which can be utilized for monitoring future trends of the epidemic and developing behavioral and biological interventions to prevent further transmission.
Showing the most recent 10 out of 123 publications
