Alphaviruses cause encephalitis in man and animals and are transmitted by mosquitos. The alphavirus Sindbis infects and kills most cultured cell types, not by lethal parasitism, but by inducing programmed cell death, or apoptosis. However, in mosquito cells and in neurons, both in vivo and in vitro, Sindbis virus can establish a persistent infection in which virus continues to be produced but the cells survive. These results indicate that a cellular factor can modulate the outcome of infection. Neurons express the human oncogene bcl-2 which has been shown to block apoptosis in a variety of cell types and species. We generated a cell line expressing bcl-2 and found that it supported Sindbis virus persistence while controls lacking bcl-2 died. However, neurovirulent strains of Sindbis virus were able to induce apoptosis despite the expression of bcl-2. This ability to kill bcl2-containing cells has been mapped to a single amino acid change in the E2 glycoprotein. To study Sindbis virus neurovirulence, we will examine the molecular mechanisms by which viral E2 and cellular bcl-2 modulate the outcome of Sindbis virus infection. To determine why avirulent strains fail to kill the cells expressing bcl-2, we will identify the viral infection step (binding, entry, replication, maturation) which is affected. A battery of antibodies will be used study the maturation of E2 in bcl-2 positive and negative cells. In addition we will generate cell lines expressing bcl-2 mutants and assay the function of these mutants in Sindbis virus infections. Some effects of bcl-2 can be mimicked by treatment of cells with antioxidants. Therefore, we will study the effects of these agents on viral replication and maturation, and on the structure/function of the E2 glycoprotein. The mechanisms of Sindbis virus neurovirulence are likely to apply to other viruses, and an increased understanding of the bcl-2 oncogene has ramifications for normal cellular processes as well as neurodegenerative disease.