To develop therapeutic and vaccine strategies to combat HIV-1, the cause of AIDS, it is essential to understand the cell-mediated immune response to this agent. To pursue this goal, murine model systems for immunization, isolation of antigen-specific T cells, and in vitro analysis of T cells specific for HIV-1 proteins have been devised. Restimulation of spleen cells from H-2d mice immunized with recombinant vaccinia virus containing the HIV-1 (IIIB) gp160 gene using syngeneic transfected fibroblasts expressing the same gene led to the production of CD8+ cytolytic T cells specific for amino acids 315-329. Such CTL could also be elicited by restimulation of primed spleen cells with peptide alone. Primed CD4+ cells were required, unless the cultures were supplemented with IL-2. This same procedure also was successful in generating CTL specific for the HIV-1 (MN) gp160. Both responses were found in H-2d, but not H-2b or H-2k mice, and the CTL elicited were both H-2Dd restricted. The peptide involved in the MN response corresponded to the same region as that involved in the IIIB response, yet the CTL produced in were highly specific and did not cross-react on target cells incubated with the reciprocal peptide. The difference in specificity could be ascribed to a single residue (position 325, V in IIIB, Y in MN) that varies widely among HIV-1 isolates. These results indicate that this is an immunodominant region of the protein in H-2d mice, possibly because of its dual capacity for stimulation of class I and class II-restricted T cells, and that naturally occurring variation in viral protein structure leads to non-overlapping T cell specificities. This has important consequences for vaccine considerations, and for understanding the selection of virus variants during infection that may escape ongoing immune responses. Application of the same methodology has led to the identification of a CTL epitope in the pol protein, restricted by H-2k. This same determinant (residues 203-219) was found to sensitize targets for lysis by PBL from HIV-infected humans. A similar overlap between murine and human responses was observed for the gp160 epitope, suggesting that studying the murine response to HIV is of potential direct value in the design of human therapeutics.
