We studied the mechanisms by which T cells recognize antigens presented on the surface of cells bound to major histocompatibility (MHC) molecules, the molecular basis of peptide binding to MHC molecules, and the application of these principles to the design of synthetic vaccines for AIDS and cancer. By studying synthetic peptides corresponding to antigenic epitopes, but with amino acid substitutions at different positions, we identified mutations that produced higher affinity for the MHC molecule or for the T-cell receptor. One such peptide from the HIV envelope binds more tightly to a class II MHC molecule and is effective as a vaccine in mice at 10-100-fold lower doses than the natural sequence at eliciting helper T-cells specific for the natural HIV epitope. We also showed that only a few residues on the peptide are critical for positive interaction, but other residues can interfere by adverse interactions, which play a major role in specificity. We found that cytotoxic T lymphocytes (CTL) specific for another HIV peptide from three strains of mice showed degenerate MHC restriction, and also used a limited number of T-cell receptor V-region genes. This same HIV peptide is also presented by both class I and class II MHC molecules to CTL and helper T cells, respectively. We compared the amino acid residues involved in presentation by each, and found a remarkable similarity, that may suggest a similar basis for binding. We used these epitopes to make a synthetic peptide candidate vaccine for HIV. The toxicology and clinical protocols are being written for a human phase I clinical trial. We have also developed a new method of immunizing with peptides without adjuvant, using dendritic cells. We applied this to cancer vaccine development, to show that an endogenously expressed mutant p53 oncoprotein in a tumor cell can serve as a target antigen for CTL, and that such CTL can be elicited by immunization with a synthetic peptide from the mutant p53 sequence. We have also begun determining helper and CTL responses to peptides from human papillomavirus (HPV) oncoproteins E6 and E7, for application to HPV-related cervical cancer. We also found that CTL activity to viruses can be inhibited by Schistosome infection related to a shift in Th1/Th2 helper balance and mediated by a suppressor cell that we are characterizing. This finding may account for the more rapid progression of HIV disease where these parasites are endemic. We have also found that the immune defect in HIV-infected patients in in vitro T-cell response can be overcome by use of anti-IL-10 antibodies, suggesting a possible approach to therapy.
