We have demonstrated that plasmid constructs containing HIV env/rev and gag/pol are safe and immunogenic when administered to HIV- infected subjects. However, responses have been modest and inconsistent, prompting development of second generation constructs, utilizing """"""""optimized"""""""" plasmid backbones. We hypothesize that combining the two new HIV DNA constructs in a """"""""priming"""""""" series of immunizations followed by """"""""boosting"""""""" with an avipox vector containing env, gag, and pol will induce a broader and more robust immune than found with the earlier HIV DNA vaccines alone. Further, we hypothesize that co-administering the HIV DNA vaccines with a construct containing the gene for the immunostimulatory molecule IL-15 will further enhance responses to the HIV vaccines. Our long-range hypothesis is that a vaccine regimen may be sufficiently immunogenic in HIV-infected subjects receiving highly active antiretroviral therapy (HAART) that it may act therapeutically to provide immunological control of their infection, even in the absence of further antiviral treatment. We propose 2 clinical trials to test these hypothesis. First, studying HIV- infected subjects whose viral production is suppressed by HAART, we will perform a pilot dose-escalation study of the safety of an immunoadjuvant construct containing the gene for IL-15 expression, to determine the dose to be ut8ilized in the larger subsequent trial of HIV vaccines. Trial 2 will compare the safety and immune responses to priming with combined """"""""optimized"""""""" HIV DNA constructs, administered with either the IL-15 construct or its placebo, followed by boosting with an HIV avipox vector. Our long-term research goal is to determine the optimal prime-boost and adjuvant strategy for administering the HIV-1 vaccines, to be utilized in a subsequent trial of efficacy, defined as sustained immune control of HIV infection, even after HAART withdrawal.