Recombinant human immunodeficiency virus type 1 (HIV-1) integrase was shown to bind ATP and other nucleoside triphosphates and nucleotide analogs in vitro. Cross-linking of ATP and the photoaffinity analog 8-azido-ATP to integrase occurred in a UV dose-dependent manner. Covalent binding of ATP to integrase was also achieved without UV irradiation when the nucleotide was oxidized to the 2',3'-dialdehyde derivative (oxidized ATP) prior to incubation with the protein, indicating the presence of a reactive lysine residue in the nucleotide binding region of the protein. A number of experimental observations indicate that nucleotides and DNA substrates bind at the same or overlapping site(s) on the integrase protein. For example, the binding of nucleotides or nucleotide analogs to integrase was blocked by prior incubation with DNA substrates, and the covalent cross-linking of 8-azido-ATP to integrase inhibited the DNA binding and oligonucleotide cleavage activities of the protein. Oxidized ATP inhibited the oligonucleotide cleavage activity of integrase at concentrations that had no effect on DNA binding, suggesting that oxidized nucleotides may specifically target the catalytic center of the enzyme. These studies indicate that nucleotide analogs may serve as probes for the DNA binding and catalytic sites of the enzyme and may serve as models for the design of active site inhibitors of retroviral integrase.
Good markers for monitoring the efficacy of antiretroviral therapy in children do not currently exist. This study examined the effect of antiretroviral therapy on human immunodeficiency virus (HIV-1) unintegrated DNA (uDNA), integrated DNA (iDNA), percent uDNA, immune complex dissociated (ICD) p24 antigenemia, and plasma viral titer. Seven children were followed at therapy initiation and at approximately 3- and 10-month intervals. HIV-1 uDNA was detected in all children prior to start of therapy (average percent uDNA, 43%). At 3 months, the percent HIV uDNA decreased in all patients to an average of 18% (p = 0.01) and at 10 months decreased to an average of 1%. In contrast, the amount of HIV iDNA was relatively constant after initiation of therapy. ICD HIV p24 antigen was detected in all patients prior to therapy (average, 538 pg/ml). Over the study period, the ICD p24 antigen level decreased in three patients and remained relatively unchanged in four patients. Plasma cultures of HIV-1 were positive in only one of the seven patients prior to therapy. Among the methods evaluated, measurement of uDNA was the only parameter which reliable decreased after initiation of nucleoside therapy.
We examine changes and stability in risk behaviors and HIV-1 seroprevalence among heterosexual injection drug users (IDUs) over 13 cross-sectional surveys, 1986-1992. Interviews (n = 5,956) were conducted with IDUs in street settings and drug detoxification clinics over 6.5 years, and respondents were tested for HIV-1 antibody. Trends in use of condoms and bleach and HIV seroprevalence were assessed using multiple logistic and linear regression analyses. The percentage of time condoms were reportedly used during intercourse among men increased from 4.5% to 31.0%. Among the declining population of IDUs who reported needle sharing, reported use of bleach increased from 3% to 89%. Significant changes in use of bleach 100% of the time were reported: 29.8% in 1988, 52.8% in 1990, and 40.0% in 1992. HIV seroprevalence doubled from 7% in 1986 to 14% in 1987. Post-1987 fluctuations in HIV seroprevalence were not significant. Significant changes in risk behaviors among IDUs were reported over the study period. These changes coincided with the implementation of HIV prevention in San Francisco, including outreach programs, HIV testing and counseling, bleach distribution, and syringe exchange. The moderate and stable rate of HIV seroprevalence beginning in 1987 parallels self-reported reductions in risk behavior.