Development of a B-cell-based vaccine is difficult due to the high genetic variability of HCV [6
]. A potent vaccine should induce a strong and long-term protective T cell response [6
]. We constructed an rAd5-based and rTTV-based HCV vaccine expressing the HCV structural gene CE1E2 from HCV genotype 1b. We identified the antigenic composition of the vaccine vectors and evaluated their immunogenicity via different delivery routes and regimens. Our results indicate that those vaccines were able to elicit a CMI response against the HCV core, E1, and E2 antigens in mice. The response remained robust for an extended period (16 weeks post-vaccination). Induction of cross-protection by rAd5-based HCV CE1E2-induced responses were confirmed in vivo using a heterologous surrogate challenge assay based on a recombinant HCV (JFH1,2a) vaccinia virus.
Since it was discovered that T cells are important for clearance of HCV, a great deal of focus has been on developing vaccines that will induce T cell responses to HCV proteins [6
]. A number of approaches, including the use of virus-like particles (VLPs) and defective or attenuated viral vectors with or without a prime-boost strategy, have been used to generate T cell responses against HCV antigens [6
]. The use of HCV VLPs produced in insect cells has proven successful for inducing effective HCV-specific immune responses [29
]. Interestingly, in these VLP studies, responses to the structural proteins were mainly specific to T cells with little or no neutralising antibody detected, and all responses resulted in modification of HCV infection and early control of replication following virus challenge [29
]. In addition, a metaanalysis of HCV vaccine efficacy in chimpanzees indicated the importance of structural proteins that may activate T cell responses and thus mediate viral clearance [30
]. We chose the structural proteins (1-746 amino acids [aa]) of the HCV isolates dominant in China (1b) as immunogens. This choice was also based on the fact that a similar structural gene (1-746 aa) can be assembled as a VLP when expressed in insect cells [29
]. Western blot analysis indicated that the core, E1, and E2 proteins were expressed efficiently in cells infected with rAd5-CE1E2 vaccine.
The viral vectors used for HCV T cell vaccines, including adenovirus and vaccinia virus vectors, are common to a number of strategies used for other infectious agents (such as HIV and TB) [6
]. A greater number of rAd5-based HCV vaccine candidates have successfully induced HCV-specific immune responses [23
]. Interestingly, an adenovirus vector-based minigene vaccine encompassing the four domains of the HCV NS3, NS4, NS4A, and NS5B proteins that contain multiple class I/II restricted epitopes also induced strong and broad HCV specific T cell responses in HLA-A2 transgenic mice and may prove promising as a tool for inducing cross-reactive responses [31
]. An extremely encouraging study reported that an rAd-based T cell vaccine expressing the NS3-NS4-NS5A-NS5B antigens elicited non-sterile, yet protective, immunity in four of five challenged chimpanzees [25
]. Protection in this study was correlated with T cell responses, in particular with CD8+T cell-mediated immunity. Based on these reports, we selected rAd as a vector to carry the immunogen. In addition, an attenuated recombinant vaccinia virus (Tian Tan strain) was selected as a vector [33
]. The data indicated that an rAd5-based HCV vaccine can elicit multi-antigen, robust, and long-lasting IFN-γ-producing CD8+T cell-mediated immunity in mice, with cross-protection. In addition, a heterologous rAd5/rTTV regimen elicited the strongest CMI and E1/E2-specific humoral immune response, compared to a homologous rAd5 regimen. These data are in accordance with previous studies of the T cell responses elicited by rAd-based HIV or HCV vaccines [20
]. Despite the current controversies concerning the use of rAd-based vaccines for HIV-1, we demonstrated here that an rAd5-CE1E2-based T cell vaccine for HCV has significant cross-protective efficacy in our surrogate challenge model. The data in this proof-of-concept study have important implications for the application of novel T cell-based HCV vaccines.
The majority of rAd5-HCV vaccines have been tested in animal model via the i.m. or i.p. route. There are limited data in the literature comparing the immunogenicity and protection elicited by various rAd5-based HCV vaccine delivery routes and regimens. Thus, we assessed the humoral and cellular immune responses and cross-protection elicited in mice immunised via different delivery routes (i.m., i.n., i.d.) and regimens. The immune effects of each delivery route differed. Compared to the i.n. route, one injection of rAd-CE1E2 induced a stronger cellular immune response to the HCV structural gene when administered via i.m. or i.d. Similarly, priming via i.n. induced a lower IFN-γ T cell response than did i.m priming. These results demonstrate that the priming route may be an important determinant of immune effects. Of the two-injection groups, an rTTV-CE1E2 boost following rAd-CE1E2 priming induced the strongest T cell responses to the HCV core, E1, and E2 proteins. Similar to the cellular responses, the heterologous regimen induced the strongest antibody response to E1 and E2, while the homologous i.n.-primed group resulted in the lowest antibody levels. Curiously, for the anti-core antibody, the titre in the i.n.-primed homologous group was much higher than in the other groups. No significant neutralising antibody response was observed among the groups based on an HCVpp assay (data not shown). Both the antibody and CMI responses could be boosted further in our model system, because even the pre-existing anti-Ad5 IgG antibody levels in the immunised groups were high (Table ).
The majority of vaccine studies have used a homologous virus challenge, i.e., the same genotype as the virus contained in the vaccine. In this study, we used a heterologous challenge model developed in our lab, which used a different subtype (2a) virus expressing an HCV-polyprotein with an approximately 18% genetic difference from the vaccine-based sequence (1b). We demonstrated in this model that vaccination of mice with an rAd5-based HCV CE1E2 vaccine strongly reduced the titre of rHCV-JFH1, and was able to fully protect immunised mice, although the i.n.-immunised and i.n.-primed regimens induced the weakest T cell responses. These results indicate that other mechanisms, such as antibody (core), are involved in protective immunity to HCV, not only the level of IFN-γ-producing T cells detected by ELISPOT or ICS. It remains possible, however, that other aspects of the CMI response may contribute to cross-protection of HCV vaccines [34
]. In addition, the potential role of antibody-dependent cell-mediated virus inhibition (ADCVI) [35
] should also be explored in a future study. Development of an immunocompetent small animal model would advance the field enormously [36
], although care should always be taken in extrapolating data to humans, as many immune response studies in mice have not translated well to humans and other primates.