The relationship between binding affinity for HLA class I molecules and immunogenicity of discrete peptide epitopes has been analyzed in two different experimental approaches. In the first approach, the immunogenicity of potential epitopes ranging in MHC binding affinity over a 10,000-fold range was analyzed in HLA-A*0201 transgenic mice. In the second approach, the antigenicity of approximately 100 different hepatitis B virus (HBV)-derived potential epitopes, all carrying A*0201 binding motifs, was assessed by using PBL of acute hepatitis patients. In both cases, it was found that an affinity threshold of approximately 500 nM (preferably 50 nM or less) apparently determines the capacity of a peptide epitope to elicit a CTL response. These data correlate well with class I binding affinity measurements of either naturally processed peptides or previously described T cell epitopes. Taken together, these data have important implications for the selection of epitopes for peptide-based vaccines, and also formally demonstrate the crucial role of determinant selection in the shaping of T cell responses. Because in most (but not all) cases, high affinity peptides seem to be immunogenic, our data also suggest that holes in the functional T cell repertoire, if they exist, may be relatively rare.

No abstract

Inasmuch as the hepatitis B virus (HBV) is not directly cytopathic for the infected hepatocyte, it is generally presumed that viral clearance and liver cell injury during viral hepatitis are due to a CTL response to HBV encoded Ag presented by HLA class I molecules. We have previously examined the peripheral blood CTL response to two HBV nucleocapsid epitopes in patients with acute and chronic viral hepatitis, one of which is restricted by HLA-A2, whereas the other is dually restricted by HLA-A31 and Aw68. In this study, we defined the HLA-A2-restricted CTL response to the hepatitis B surface Ag (HBsAg) by using a panel of HBsAg-derived synthetic peptides containing the ideal HLA-A2.1 binding motif (-L------V). Several novel aspects of HBV immunobiology and pathogenesis are evident from this study. First, the peripheral blood CTL response to HBV-encoded Ag is remarkably polyclonal and multispecific in most patients with acute hepatitis. Indeed, HLA-A2-restricted CTL specific for as many as four envelope epitopes and one nucleocapsid epitope were found to be present simultaneously in individual patients with acute viral hepatitis. Second, HBV-specific CTL are not detectable in the peripheral blood in a minority of patients with acute hepatitis, nor have we detected a CTL response in any of the patients with chronic hepatitis that we have studied thus far. Although the cellular and molecular basis for CTL nonresponse remains to be determined, the data suggest that it may contribute to viral persistence. Third, the diversity and the specificity of the CTL response is determined in part by the coding sequence of the viral genome present in each infected patient. Indeed, the apparent nonresponse of some acutely infected patients to at least one HBsAg-specific CTL epitope actually reflects infection by a viral variant that contains a critical substitution in one of the anchor residues within the epitope. Finally, at a fundamental level, the data suggest that the presence of the HLA-A2.1-binding motif in a peptide may not be sufficient for binding; and the capacity of a peptide to bind the class I molecule does not guarantee that it will be immunogenic.

No abstract

No abstract