HPV Pseudovirus-Based Cutaneous Vaccination

Human papillomavirus (HPV)-based gene transfer vectors, also known as pseudoviruses (PsV), have been generated in the Laboratory of Cellular Oncology, National Cancer Institute, National Institutes of Health and are currently being evaluated as vehicles for gene therapy and genetic vaccination applications. In a recently established mouse model of genital tract HPV infection intranasal or cervico-vaginal immunizations with HPV PsV that encapsidate a model antigen encoded by the packaged plasmid efficiently induced systemic and mucosal antigen-specific humoral and cellular immune responses. However, to date, HPV PsV infection of mouse skin appears to be relatively inefficient. Transcutaneous immunization (TCI) has emerged as an attractive and promising strategy to induce systemic and mucosal B and T cell responses to antigens. However, the strategy has largely focused on protein-based approaches, which, although they can induce CD8+ T cell responses by cross-priming, may be less efficient at generating robust CD8+ T cell responses in comparison to genetic immunization with a vector such as the HPV PsV, which exhibits remarkable tropism for wounded keratinocytes. The first aim is to establish a reliable and efficient TCI strategy using PsV of cutaneous and genital/mucosal HPV types in a murine model. HPV PsV infection will be tracked and measured in tissues of living animals at multiple timepoints and the cellular distribution of HPV PsV within the infected tissues will be investigated. Next HPV PsV encapsidating a model antigen will be employed to address the question whether potent antigen-specific B and T cell responses can be efficiently induced in mice by HPV PsV-mediated genetic vaccination of the skin. Antigen- specific antibodies will be measured in blood and mucosal secretions. Local and systemic antigen-specific CD4+ and CD8+ T cell responses in blood and tissue samples (skin, lymph nodes, intestine, lungs, spleen, genital tract) will be determined. The induced T cells will be analyzed regarding their phenotype, the preferential trafficking to distinct tissue sites and their localization within tissues. Furthermore it will be evaluated whether differences in immune responses between transcutaneous prime and boost immunizations occur and whether TCI can be used to boost intravaginal or intranasal priming and vice versa. The strategy that is most effective at inducing cellular immune responses after TCI with HPV PsV will be further evaluated in an established transgenic (tg) mouse model for HPV-associated epithelial tumors, in which the oncoprotein E7 of high-risk genital HPV 16 is expressed as transgene under the basal epithelium-specific keratin 14 (K14) promoter. These tg K14E7 mice develop skin cancer, and skin grafts transferred from tg K14E7 mice to syngeneic, non-tg, E7-nave animals mimic HPV- associated anogenital dysplasia and cancer, in which expression of the oncoprotein E7 is retained in the epithelial cells. The effect of HPV PsV TCI on induction of tg skin transplant regression in the tg K14E7 mouse model will be addressed with HPV PsV expressing a modified HPV 16 E7, which for safety reasons lacks transforming (oncogenic) properties. The induction of T cell responses and trafficking of these cells will be determined and rejection of transplanted grafts will be documented.

Infection with papillomaviruses (PV) can cause benign warts (papillomas) on skin and mucosae of humans and animals, but also epithelial malignancies, especially anogenital cancer, some oropharyngeal carcinomas and, in genetically predisposed individuals, cutaneous squamous cell cancers. Control and clearance of these viruses are thought to be mediated by the cellular immune system, however experimental proof for the key cellular effector(s) that regulate PV infection and lesion development is lacking. The recently identified mouse papillomavirus (MusPV1, also designated MmuPV1) represents an excellent opportunity to study the nature of cutaneous PV in the natural host, which is the most extensively characterized non-human species and for which biological, genetic, and immunological markers and reagents are widely available. We have first established an improved in vivo murine skin infection model using PV pseudovirions (PsV), that are composed of the viral capsid and encompass a plasmid-encoded antigen instead of the viral genome. These PsV are currently evaluated as vehicle for gene delivery applications, and we have identified some PsV types that are superior in infecting murine skin. Using this improved skin infection technique we were able to demonstrate that synthetic DNA of MusPV1 induced papilloma outgrowth in immunodeficient mice and drove the production of native MusPV1 virions, thus allowing in vitro and in vivo studies independent of the naturally occurring viral source. Extracts of these papillomas could be serially passaged without loss of biological potency and confirmed that infection is virion-dependent, since transmission was prevented by passive transfer of an immune serum raised against the MusPV1 major capsid protein, L1. We also revealed surprising region-specific differences in the susceptibility to MusPV1 infection within the animal, which are likely attributable to intracellular factors. We also investigated MusPV1 infection and papilloma formation in immunocompetent mice. Our study showed that in an immunocompetent setting, MusPV1 generally caused asymptomatic skin infections, but no lesion outgrowth, except in a highly susceptible strain when using extremely high amounts of virions for inoculation. Visible papillomas were consistently observed after profound immunosuppression in some, but not other, strains of mice. By selective removal of defined cellular immune populations and employing genetically modified mice, we could show that T-cells are pivotal for controlling MusPV1 infection and disease. We further showed surprising differences in the protective capacities of T-cell subsets that are required for protection in different strains of immunocompetent mice. This implies that unanticipated effector mechanisms can control virus infection and pathogenesis in specific genetic backgrounds. The findings may help to explain the wide of range of pathologic outcomes of infection by a specific human PV type in immunocompetent people. The MusPV1 infection mouse model further provides an excellent opportunity to study the impact of skin PV infection in an immunosuppressed and in an immunocompetent setting.