Rapid shift from virally infected cells to germinal center-retained virus after HIV-2 infection of macaques.

Lymphoid tissues are the primary target during the initial virus dissemination that occurs in HIV-1-infected individuals. Recent advances in antiretroviral therapy and techniques to monitor virus load in humans have demonstrated that the early stages of viral infection and host response are major determinants of the outcome of individual infections. Relatively little is known about immunopathogenic events occurring during the acute phase of HIV infection. We analyzed viral dissemination within lymphoid tissues by in situ hybridization and by combined immunohistochemistry/in situ hybridization during the acute infection phase (12 hours to 28 days) in pig-tailed macaques (Macaca nemestrina), challenged intravenously with a virulent strain of HIV-2, HIV-2(287). Two stages in viral dissemination were clearly evident within the first 28 days after HIV-2(287) infection. First, a massive increase in individual HIV-2-infected cells, mostly CD3+ T lymphocytes and a smaller percentage of macrophages and interdigitating dendritic cells, was identified within lymph nodes which peaked on the 10th day after HIV-2 infection. A shift of HIV-2 distribution was demonstrable between day 10 and day 14 after HIV-2 infection. Coincident with a marked reduction in individual HIV-2 RNA+ cells by day 14 postinfection, there was a dramatic increase in germinal center-associated HIV-2 RNA. High concentrations of HIV-2 RNA persisted in germinal centers in all animals by days 21 and 28 postinfection. Thus, HIV-2 appears to go through an initial, highly disseminated cellular phase followed by localization in the follicular dendritic cell network with relatively few infected cells. In this nonhuman primate model of HIV-associated immunopathogenesis, using a virus derived from a human pathogen, we identified a significant shift in the pattern of HIV-2 localization within a narrow time frame (day 10 to day 14). This shift in virus localization and behavior indicates that there may be a discrete but remarkably narrow window for therapeutic interventions that interrupt this stage in the natural course of HIV infection. Reproducibility and the accelerated time course of disease development make this model an excellent candidate for such intervention studies.