We have recently shown that the NS3-based genetic immunogens should contain also hepatitis C virus (HCV) nonstructural (NS) 4A to utilize fully the immunogenicity of NS3. The next step was to try to enhance immunogenicity by modifying translation or mRNA synthesis. To enhance translation efficiency, a synthetic NS3/4A-based DNA (coNS3/4A-DNA) vaccine was generated in which the codon usage was optimized (co) for human cells. In a second approach, expression of the wild-type (wt) NS3/4A gene was enhanced by mRNA amplification using the Semliki forest virus (SFV) replicon (wtNS3/4A-SFV). Transient tranfections of human HepG2 cells showed that the coNS3/4A gene gave 11-fold higher levels of NS3 as compared to the wtNS3/4A gene when using the CMV promoter. We have previously shown that the presence of NS4A enhances the expression by SFV. Both codon optimization and mRNA amplification resulted in an improved immunogenicity as evidenced by higher levels of NS3-specific antibodies. This improved immunogenicity also resulted in a more rapid priming of cytotoxic T lymphocytes (CTLs). Since HCV is a noncytolytic virus, the functionality of the primed CTL responses was evaluated by an in vivo challenge with NS3/4A-expressing syngeneic tumor cells. The priming of a tumor protective immunity required an endogenous production of the immunogen and CD8 þ CTLs, but was independent of B and CD4 þ T cells. This model confirmed the more rapid in vivo activation of an NS3/4A-specific tumor-inhibiting immunity by codon optimization and mRNA amplification. Finally, therapeutic vaccination with the coNS3/4A gene using gene gun 6-12 days after injection of tumors significantly reduced the tumor growth in vivo. Codon optimization and mRNA amplification effectively enhances the overall immunogenicity of NS3/4A. Thus, either, or both, of these approaches should be utilized in an NS3/4A-based HCV genetic vaccine.

Background: The hepatitis C virus (HCV) mutates within human leucocyte antigen (HLA) class I restricted immunodominant epitopes of the non-structural (NS) 3/4A protease to escape cytotoxic T lymphocyte (CTL) recognition and promote viral persistence. However, variability is not unlimited, and sometimes almost absent, and factors that restrict viral variability have not been defined experimentally. Aims: We wished to explore whether the variability of the immunodominant CTL epitope at residues 1073-1081 of the NS3 protease was limited by viral fitness. Patients: Venous blood was obtained from six patients (four HLA-A2+) with chronic HCV infection and from one HLA-A2+ patient with acute HCV infection. Methods: NS3/4A genes were amplified from serum, cloned in a eukaryotic expression plasmid, sequenced, and expressed. CTL recognition of naturally occurring and artificially introduced escape mutations in HLA-A2-restricted NS3 epitopes were determined using CTLs from human blood and genetically immunised HLA-A2-transgenic mice. HCV replicons were used to test the effect of escape mutations on HCV protease activity and RNA replication. Results: Sequence analysis of NS3/4A confirmed low genetic variability. The major viral species had functional proteases with 1073-1081 epitopes that were generally recognised by cross reactive human and murine HLA-A2 restricted CTLs. Introduction of mutations at five positions of the 1073-1081 epitope prevented CTL recognition but three of these reduced protease activity and RNA replication. Conclusions: Viral fitness can indeed limit the variability of HCV within immunological epitopes. This helps to explain why certain immunological escape variants never appear as a major viral species in infected humans.

The hepatitis C virus (HCV) protease and helicase encompasses the nonstructural (NS) 3 protein and the cofactor NS4A, which targets the NS3/4A-complex to intracellular membranes. We here evaluate the importance of NS4A in NS3-based genetic immunogens. A full-length genotype 1 NS3/4A gene was cloned into a eucaryotic expression vector in the form of NS3/4A and NS3 alone. Transient transfections revealed that the inclusion of NS4A increased the expression levels of NS3. Subsequently, immunization with the NS3/4A gene primed 10- to 100-fold higher levels of NS3-specific antibodies as compared to immunization with the NS3 gene. Humoral responses primed by the NS3/4A gene had a higher IgG2a/IgG1 ratio (>20) as compared to the NS3 gene (3.0), suggesting a T helper 1-skewed response. Low dose i.m. (10 microg) immunization with the NS3/4A gene inhibited the growth of NS3/4A-expressing tumor cells in vivo, whereas the NS3 gene alone or NS3 protein did not. We then evaluated the efficiency of the NS3/4A gene administered by the gene gun, at the same doses used for humans, in priming cytotoxic T lymphocyte (CTL) responses. Three to four 4 microg doses of the NS3/4A gene primed CTL at a precursor frequency of 2-4%, which inhibited the growth of NS3/4A-expressing tumor cells in vivo. Thus, NS4A enhances the expression levels and immunogenicity of NS3, and an NS3/4A gene delivered transdermally could be a therapeutic vaccine candidate.