Papillomaviruses (PVs) infect the epithelia of animals and man, where they generally induce benign proliferation at the site of infection. However, there is a strong association between malignant progression of human genital lesions and certain human papillomavirus (HPV) types, most frequently HPV 16. Our research is primarily concerned with development of vaccines and other infection inhibition strategies against HPV and the elucidation of the PV life cycle. We have developed a simple and efficient strategy for generating high titers of infectious papillomavirus particles that transduce encapsidated marker plasmids, i.e. pseudovirions. We have exploited this technology in our basic virologic and translational research efforts. We have used our pseudovirus technology to develop the first cervicovaginal challenge model for HPVs. We found that the infection of the female mouse genital tract, even of monolayer endocervical cells, requires exposure the basement membrane to the virus. The capsids bind avidly to the basement membrane but not to the apical surfaces of intact columnar or stratified squamous epithelia. Using a quantitative assay based on whole tissue fluorecence imaging after infection with red fluorescent protein (RFP) expressing pseudovirions, we determined that nonoxynol-9 (N-9) and Conceptrol, an over the counter spermacide that containing N-9, dramatically potentiate in vivo papillomavirus infection, presumably due to it's ability to permeablize the epithelial layers and thereby expose the basement membrane to the virus. Interesting, no infection was detected if nonoxynol-9 was formulated in carrageenan rather than its normal gelling agent. Carrageenan is an algal polysaccharide widely used in processed food and cosmetics and is the main gelling agent in some over-the-counter lubricants. The results suggest that women using N-9 spermacides may be at increased risk of acquiring genital HPV infection and that this risk might be mitigated by reformulation of N-9 in a carrageenan based gel. Two NCI-sponsored clinical trial of carrageenan as a microbicide to prevent genital HPV infection in young women will soon commence, and we will collaborate on technical aspects of the trials. Our mouse cervicovaginal studies suggested that other interventions that compromise the integrity of the genital epithelium might potentiate HPV infection. Acquisition of ecto- and endocervical cells for cytology (Pap) screening disrupts the epithelium by design. Therefore we sought to determine whether the cytology specimen (Pap smear) collection procedure renders the cervix more susceptible to HPV infection in a rhesus macaque model. We found that the Pap smear collection procedure greatly potentiated infection by RFP expressing HPV16 pseudovirus. However, use of a carrageenan gel rather than Surgilube as the lubricant used for the internal digital exam after specimen collection largely abrogated the infection enhancing effect. These findings suggest that cytology screening in women might lead to a transient enhancement of susceptibility to HPV infection and that use of a carrageenan-based gel during the examine might mitigate this enhancement. The pseudovirus-based mouse cerviovaginal challenge model is also being used to explore the basic features of papillomavirus infection and the mechanisms of antibody-mediated neutralization of the virus in the female reproductive tract. We have determined that, in a process unique to papillomaviruses, the initial steps in infection occur after initial binding to heperan sulfate proteoglycans (HSPG) on the basement membrane, prior cell surface binding. Specifically the virions undergo a conformational change that exposes the N-terminus of the L2 minor capsid protein to cleavage by furin, a cellular protease that we detect in abundance at the sites of infection in vivo. This cleavage in turns leads to exposure of highly conserved L2 cross-neutralizing epitopes that are immediately downstream of the cleavage site. Blocking HSPG interactions or furin cleavage prevented cerviovaginal infection. In a process that takes several hours, the virions transfer from the basement membrane to the surface of keratinocytes invading the site of trauma, and the virions are then internalized. We have further determined that antibodies induced by L1 VLP vaccines (of type that are how commercially available) and L2 vaccines (which we developing as a more broadly cross type-neutralizing alternative) block in vivo infection by distinct mechanisms. L1 VLP antibodies block binding to the basement membrane, but allow binding to the keratinocytes. However, the L2 cross-neutralizing epitopes are not exposed and the virions are not internalized. L2 antibodies that broadly cross-neutralize HPV types permit the initial virion binding to the basement membrane HSPG. However, the virions are not detected in the tissue at later time points, suggesting that the L2 antibodies prevent stable transfer to the keratinocytes. These studies may well provide the most detailed mechanistic understanding of how a virus, in a living mammalian host, infects its relevant tissue and how vaccine induced neutralizing antibodies prevent infection in vivo. In vitro neutralizing assays we developed for L1 VLP vaccine analysis have proven to be insensitive for L2. Based on understanding of the in vivo infectious process, we have recently developed novel in vitro neutralizing assay that are 1000-fold more sensitive measure of L2 antibody activity. Our development of a method to induce efficient HPV pseudovirus infection of the female genital tract after transient disruption with N-9 has proven to be the key to our recent development of an effective intravaginal vaccination strategy. In patent pending studies, we have found that intravaginal pseudovirus vaccination of N-9 treated mice induces strong systemic and mucosal T and B cell responses to target antigens transduced by the pseudovirions. Systemic responses rival those induced by previously optimized Ad5 vectors. Intravaginal responses are remarkably strong, with up to 80% of all intravaginal CD8 T cells staining tetramer positive for the targeted antigen. Most of the induced T appear to be intraepithelial and high level of effector memory CD8 T cells are maintained in the vaginal tract 100 days after vaccination. Intravaginal pseudovirus vaccination is a promising approach for focusing immune responses to the female genital tract and so might increase the effectiveness of vaccines directed against HSV and HIV infections and against HPV induced neoplasia. This concept is now being testing in an SIV/rhesus macaque intravaginal challenge model in collaboration with Dr. Franchini. To more generally evaluate the potential of HPV pseudoviruses as gene transfer vehicles, we have conducted a broad infection tropism survey. In patent pending studies, we demonstrated that intact murine epithelium at all sites, whether simple, columnar, or squamous, was highly resistant to both virion binding and infection, whereas disrupted epithelium was susceptible. In contrast, virtually all human-derived epithelial cell lines in the NC1-60 panel were highly susceptible to infection in vitro. The remarkable specificity of HPV pseudovirus binding and infection suggests that they may be useful in tumor diagnostic or cytotoxic gene therapy applications. In proof of concept studies, we documented highly specific binding and infection, and dramatic imaging, of human ovarian tumor nodules implanted in nude mouse peritoneum after intraperitoneal injection of RFP-expressing pseudovirus. In a preliminarly study using a mouse model of ovarian metastases, intraperitoneal injection of Herpes TK-expressing HPV psuedovirions followed by ganciclovir treatment, increased survival of tumor bearing mice.
Showing the most recent 10 out of 55 publications
