In this study, we demonstrated that HERV envelope protein-coated baculoviral vectors could serve as delivery systems for multivalent DNA vaccines in mice and pigs. Co-immunization with AcHERV-based HPV16L1 and HPV18L1 bivalent DNA vaccines provided antigen type-specific humoral serum IgG and vaginal IgA immune responses and induced neutralizing antibodies to an extent comparable to the commercial bivalent VLP vaccine, Cervarix. Notably, the AcHERV-based bivalent DNA vaccines induced cell-mediated immune responses substantially greater than those induced by Cervarix. Moreover, multiple, sequential single immunizations with AcHERV-HP18L1 followed by AcHERV-HP16L1 resulted in the stepwise induction of type-specific IgG antibodies.
We found that AcHERV-based bivalent vaccinations induced antigen-type specific serum IgG () and vaginal IgA antibody responses (). Given the multiple HPV type-related pathogenicities, it is essential to develop multivalent rather than monovalent vaccines. Indeed, the currently available VLP vaccines, Cervarix and Gardasil, are bivalent (type 16, 18) and tetravalent (type 6, 11, 16, 18), respectively 
. Although the current study demonstrated that AcHERV vectors are effective for the delivery of two discrete DNA vaccines, the results obtained indicate the feasibility of applying these AcHERV vectors as a delivery system for multiple DNA vaccines.
The induction of neutralizing antibody is crucial for the prevention of genital tract infection by HPV. Like the currently marketed Cervarix, the AcHERV-based bivalent DNA vaccine induced neutralizing antibody () to levels that almost completely blocked subsequent challenges with HPV type-specific pseudoviruses (). Given the species restriction of HPV, many HPV vaccine studies have evaluated the immunogenicity by analyzing neutralizing antibodies. In this study we used pseudovirus expressing luciferase to test the cross-protection of immunized mice against other HPV types (). Recently, Christensen’s group developed a challenge model using chimeric HPV capsid/cottontail rabbit papillomavirus genome particles for the direct testing of HPV vaccines in rabbits. In the previous studies, rabbit models have been used for evaluation of chimeric virus-like particle vaccines 
, and N-terminal HPV16 L2 polypeptides 
. In future study, the cross-protection capability of the AcHERV-based DNA vaccines needs to be further evaluated using the rabbit models.
The fact that AcHERV-based bivalent vaccinations afforded complete protection against challenges with HPV pseudoviruses, a result comparable to that of commercial VLP vaccines, could reflect the effective DNA delivery functions of AcHERV vectors. In our previous study, we reported that AcHERV-HP16L1 notably increased the delivery of HPV16L1 DNA into NIH3T3 human cells lines compared to a baculovirus vector carrying HPV16L1 without HERV env
. Until now, various papillomavirus L1-based DNA vaccines have been studied in animal models such as rabbits 
, and beagle dogs 
. Although DNA vaccines have several advantages over subunit vaccines, one of their major drawbacks is their limited intracellular delivery efficiencies 
. By coating the surface of a non-replicating baculoviral vector with HERV envelope protein, we were able to increase the intracellular delivery efficiency of DNA vaccines.
Unlike Cervarix, which mainly produced humoral immune responses, inducing only a minimal cell-mediated immune response, AcHERV-based bivalent immunization induced antigen-specific, cell-mediated immune responses (). This inability of current VLP vaccines to induce a cell-mediated immune response limits the suitable target population to adolescent girls with no pre-exposure to HPV infection 
. To enhance the cell-mediated immune responses of currently available VLP vaccines, researchers have employed approaches to incorporate immune stimulators and block the effect of interleukin-10, which prevents the induction of a cell-mediated immune response 
. Although efforts have been made to generate a vaccine that can induce cell-mediated immune responses, this goal has been difficult to achieve without using a live attenuated vaccine 
. In this respect, the induction of both cell-mediated and humoral immune responses by AcHERV-based bivalent DNA vaccines supports the feasibility of developing AcHERV systems for preventive and therapeutic multivalent DNA vaccines, thus expanding the population of those who could benefit from the vaccinations to include individuals who have been previously infected with HPV.
Importantly, we observed that the AcHERV vectors retained effective DNA vaccine delivery functions after multiple injections. To date, several viral vectors have been developed as DNA vaccine delivery systems 
. One of the limitations of viral vectors for DNA vaccine or gene therapy applications is their induction of antibodies against the viral vector itself or the existence of pre-existing antibodies against the viral vectors, which nullify the effect of the viral vectors after a few doses 
. To address the feasibility of multiple dosing, we tested the immune responses to separate antigens by sequential immunization of mice with AcHERV-HP18L1 and AcHERV-HP16L1 (). The step-wise induction of anti-HPV18L1 antibody followed by anti-HPV16L1 antibody resulting from this immunization protocol suggests that an AcHERV-based vaccine can function as a multiuse delivery system. The mechanisms that allow repeated dosing of AcHERV vector systems need to be demonstrated at the molecular levels and will require further investigation. However, we speculate that the surface coating of HERV envelope protein on baculovirus nanoparticles prevents elimination by preexisting or induced antibodies and is thus the primary contributor to the persistent in vivo delivery in mice. The clinical implication of the feasibility of repeated dosing is that, once immunized with AcHERV-HPV DNA vaccines, a population could also be immunized in the future with AcHERV vectors carrying other DNA vaccines.
Further development of new vaccine systems requires demonstration of proof-of-concept in large, non-rodent animals in addition to rodent species. The demonstrated induction of humoral immune responses in pigs () indicates that AcHERV-based bivalent vaccine delivery systems may be effective in large animals. One concern in using pigs as an animal model is that AcHERV vectors are coated with HERV envelope protein, not with protein encoded by porcine endogenous retrovirus env. Given the species specificity of endogenous retrovirus, the use of primates in a future study might reveal greater efficacy of the AcHERV-based vectors, possibly reflecting enhanced activation of HERV envelope protein-mediated endocytosis by primate cells. The induction of a humoral immune response in pigs suggests that AcHERV vector-based DNA vaccine delivery systems could be applied against contagious porcine diseases. Moreover, the significant induction of humoral immune responses in pigs weighing 130 kg provides strong evidence that AcHERV systems could prove effective in other large animals, including humans, in addition to laboratory animals. Moreover, we currently construct AcHERV vectors expressing HPV16L1 and 18L1 fusion proteins. It needs to be tested whether the sequential orders of the fusion proteins affect the expression levels and immunogenicity, and whether the AcHERV vectors expressing HPV16L1 and 18L1 fusion proteins provide the immunogenicity comparable to the co-immunization of two separate AcHERV vaccines expressing HPV16L1 and HPV18L1, respectively.
In conclusion, the humoral and cell-mediated immune responses that follow immunization with AcHERV vector-based bivalent DNA vaccines suggest that AcHERV vector systems can be developed as preventive and therapeutic multivalent DNA vaccine delivery systems. Moreover, the persistent immunization efficacy after repeated dosings indicates the future clinical applications of AcHERV vectors as a platform for various DNA vaccines for the same populations.