It is apparent that a safe and effective HIV vaccine is an important component in the development of rational approaches for the control and prevention of HIV transmission. Given the fact that the virus most often encounters a mucosal surface during sexual transmission, a vaccine designed to stimulate both the systemic and mucosal immune systems is essential. Poliovirus is attractive as a delivery system because of several biological features inherent to the virus. First, the pathogenesis of the virus has been well studied, and important features have been identified. The virus is naturally transmitted by a fecal-oral route and is stable in the harsh conditions of the gastrointestinal tract. Second, previous studies using attenuated vaccine strains of poliovirus showed that a long-lasting systemic and mucosal immunity is generated after administration of the vaccines. Studies have demonstrated the presence of circulating T cells that proliferate to whole inactive poliovirus or peptides corresponding to amino acids of the VP1 proteins in previously immunized individuals. These results established that immunization with poliovirus stimulates both the humoral and cell-mediated components of the immune system. Third, the attenuated strains of poliovirus are safe for humans and are given to infants as early as 6 months of age. The incorporation of foreign genes into the attenuated strains would be an attractive feature that should pose no more of a health risk than that associated with administration of the attenuated vaccines. Finally, studies from this laboratory, as well as others, have established the feasibility of incorporating foreign genes into the poliovirus cDNA. We have described the expression of fragments of HIV-1 gag, pol, and env genes in poliovirus replicons and the procedures required for encapsidation and serial passage of these replicons. The administration of these replicons to mucosal surfaces, as well as intramuscular injection, results in the stimulation of antibodies to the foreign gene expressed from the replicons. To date, we have expressed fragments of HIV-1 gag, pol, or env gene. Although these replicons are useful, a replicon that expresses native proteins probably will be essential for an effective vaccine. With respect to this, we have expressed the full-length HIV-1 gp160, using a poliovirus replicon (D.C. Porter and C.D. Morrow, unpublished observations, 1994). Experiments are underway to encapsidate this replicon and to test its immunogenicity.