Skip to main content

Immunological basis of rational vaccine design for hepatitis C virus


Since HCV does not efficiently infect human cells in-vitro and primary infected hepatocytes cannot be used as stimulator/target cells for CTL analysis, the study of CTL activity has so far been severely hampered by the lack of systems to activate and expand CTL in-vitro, reproducing antigen presentation to CTL occurring during natural infection. To overcome this limitation, recombinant expression vectors represented by five different defective adenoviruses carrying structural and non-structural HCV genes (core, core-E1-E2, E2, NS3-NS4A, NS3-NS5A) were used to induce the endogenous synthesis of HCV proteins in human cells in vitro and to allow their entry into the HLA class I cytosolic pathway of antigen processing. Our study defined a combined and sequential strategy of CTL induction with recombinant adenoviruses followed by target cell sensitisation with synthetic peptides or recombinant vaccinia viruses that represents a powerful tool to study the cytotoxic T cell response against endogenously processed immunodominant epitopes likely involved in protection against HCV. This represents a reproducible system to stimulate and expand HCV-specific CTL in-vitro, that mimics the conditions of antigen encounter in vivo and should therefore permit to develop novel pathogenic information essential for the design of strategies to prevent HCV infection and to manipulate the immune response in patients with chronic HCV infection. The major exploitation of these results is for the design of a T cell vaccine against the hepatitis C virus (HCV). Since the analysis of the HLA class I HCV-specific T cell response induced by viral vectors allows the identification of endogenously processed immunodominant epitopes that should be involved in vivo in protection against HCV. To perform this study we used five different adenoviruses one control (Axw1) without HCV gene insertions, one containing the Lac Z gene, one encoding the whole core region (AxCA39), one the NS3 and NS4A regions (AxCANS3/4A), and one encoding the regions comprised from NS3 to NS5A (AxCA3269). We demonstrated that adherent human mononuclear cells can be efficiently infected by recombinant adenoviruses and that the infection induces HCV gene expression. Co-culturing cells infected with Ad vectors with autologous PBMC from patients with chronic HCV infection it was possible to successfully stimulate HCV-specific T lymphocytes . All adenoviruses carrying different HCV genes were able to induce significant and reproducible levels of CTL activity. Responses were generally multi-specific in individual patients, although different profiles of antigen recognition were detected in different subjects. In particular, non-structural proteins were always able to stimulate CTL responses and multiple sequences were recognized within these regions. The CTL response induced in 6 patients with chronic HCV infection by Ad stimulation was compared to the response induced in the same individuals by peptide stimulation. In some cases PBMC stimulation with peptide mixtures determined a polyclonal and multi-specific CTL response, directed simultaneously against multiple sequences. In contrast, the CTL response stimulated by Ad infected cells was directed against a more limited number of viral sequences that were not always the same recognized following peptide stimulation indicating that Adenovirus stimulation can allow epitope selection. By the adenovirus stimulation approach, we have defined a number of immunodominant epitopes; some of them (E2 402-412, NS3 1073-1081, NS4 1406-1415 and NS4 1671-1680) were previously identified by different experimental strategies. Therefore, our study extends these results confirming that these epitopes are immunodominant in the context of the HLA-A2 haplotype. Furthermore, our results define additional immunodominant CTL epitopes (NS4 1992-2000 and NS5 2145-2154) for HLA-A2 positive individuals. This experimental system based on CTL induction with endogenously synthesized HCV proteins and target cell sensitisation with synthetic peptides gives the opportunity to draw a detailed map of the immunodominant epitopes recognized by HCV-specific CTL in association with all different HLA molecules and to define reagents exploitable for the design of vaccines effective in populations with different HLA backgrounds.