The principal objectives of this research project are two-fold: 1) to determine host and viral factors that influence the genetically controlled resistance of inbred mouse strains to street rabies virus (SRV) and 2) to determine the mechanism(s) by which recombinant vaccines expressing different structural proteins of the rabies virus protect mice against lethal infection. In a continuation of our cytotoxic T lymphocyte (CTL) studies, we have determined that CD8+ T cells had no effect on the spontaneous recovery of BALB/c mice from central nervous system (CNS) infection whether or not anti-CD8 Mab was administered throughout the disease course or only during the interval of recovery from CNS infection. In contrast, depletion of CD4+ T cells during these same intervals resulted in 100% mortality which correlated with a complete depression of virus-specific neutralizing antibody production. In contrast to euthymic mice, SRV-infected T-lymphocyte-deficient (nude) mice did not develop hindlimb paralysis prior to death. We documented the role of T lymphocytes in rabies virus-associated paralysis by showing that recipient nude mice which received spleen cells from normal immunocompetent euthymic mice became paralyzed following SRV infection. Depletion of T cells from the donor spleen suspension prior to transfer abrogated the development of paralysis, but did not prevent the deaths of recipient animals. CNS inflammatory responses in nude mice immunologically reconstituted with normal spleen cells were characterized by diffuse cellular infiltrates in the parenchyma and extensive perivascular cuffing. Perivascular infiltrates included CD8+ and CD4+ T lymphocytes and Mac-1+ macrophage- microglial cells. Inflammatory cells in the parenchyma were limited to CD8+ lymphocytes and Mac-1+ cells. These observations indicate that paralysis of rabies-infected mice is dependent on T lymphocytes. Studies with recombinant vaccines expressing either the rabies virus glycoprotein (G) or nucleoprotein (N) are continuing. In vitro and in vivo models to detect protective activity in sera of N vaccinated mice have been established. Although we are certain that protection of N- vaccinated mice is due to anti-N antibody, the mechanism of protection is unclear. Our results suggest that the N protein, because of its remarkable sequence conservation among N proteins of different rabies viruses, should be incorporated with the G protein in recombinant vaccines.