Murine polyomavirus VP1 virus-like particles as vectors for gene therapy and as vaccines against polyomavirus infection and tumors

Abstract: Murine polyomavirus (MPyV) is a double stranded DNA virus ubiquitous in mice. The virus genome encodes for three transforming antigens that interact with the host cell machinery in order to initiate viral production. The transforming antigens can under certain circumstances when not controlled, e.g. in immune deficient individuals, accidentally turn the host cell into a tumor cell. The viral capsid is composed of three structural proteins. One of these proteins, VP1, that normally constitutes 80 % of the total protein content in the capsid can be produced and purified in vitro. Purified VP1 proteins self assemble into empty pseudocapsids, or virus-like particles (VLPs). MPyV is closely related to two groups of viruses that are pathogenic to humans, the human polyomaviruses (BKV and JCV) and the human papillomaviruses (HPVs). The two human polyomaviruses are associated with severe and lethal diseases in immune deficient individuals. Around 100 HPV types have been identified so far and they are grouped into low-risk or high-risk types depending on their capacity to induce tumors. Examples of high-risk types are HPV-16 and -31 that cause e.g. ano-genital cancer and cancer in the head and neck region. Today there are no commonly used vaccines against these two groups of human pathogens. The MPyV system serves as an excellent animal model allowing studies of natural virus-host interactions, vaccine trials, and virus induced tumor development. The aim of this thesis was to examine the use of murine polyomavirus VP1 virus-like particles (MPyV-VLPs) as vectors for gene therapy, and as vaccines against both MPyV infection and MPyV induced tumors. To evaluate MPyV-VLPs as vectors for gene therapy, DNA was introduced alone or together with MPyV-VLPs into normal and immune deficient mice. MPyV infected mice were used as controls. DNA introduced by MPyV-VLPs, was taken up by most tissues, where it persisted for several months in a similar way to after MPyV infection, whereas DNA introduced alone only was detectable in a few organs for a few days. Furthermore, the amount of DNA/cell was approximately 100 times higher when introduced by MPyV-VLPs. MPyV-VLPs were also shown to induce antibody titers in normal and T-cell deficient mice. To study if MPyV-VLP vaccination could protect T-cell deficient mice from a forthcoming MPyV infection, different experiments were performed. The experiments included alternative vaccines such as a modified MPyV-VLP (M17-VLP) that lacked one of the extruding loops of VP1, temperature sensitive (ts-) mutants and GST-VP1. Ts-mutants are MPyVs with short mutations allowing replication at 32°C but not 39°C, whereas GST-VP1 consists of VP1 proteins in pentamer formations. In short, intra peritoneal (i.p.) vaccination with MPyV-VLPs and ts-mutants was more efficient than M17-VLP vaccination, protecting 63-67% compared to 26% of the T-cell deficient mice against MPyV infection. However, subcutaneus (s.c.) vaccination with MPyV-VLPs induced a complete protection of both normal and T-cell deficient mice against MPyV infection whereas s.c. vaccination with GST-VP1 protected normal mice, but not all (60%) T-cell deficient mice. MPyV-VLPs were also used as vaccines against MPyV induced tumors. Three different MPyV induced tumors were evaluated (SEBA, SEBB and SECA). When examined in vitro, SEBA and SEBB had high viral load and detectable VP1, whereas SECA had a low viral load and no detectable VP1. Nevertheless, MPyV-VLP vaccinated mice were completely protected against SECA, but not against SEBA or SEBB tumor outgrowth. MPyV-VLP vaccination did not protect mice against outgrowth of a non-MPyV induced tumor. It is possible that the origin of the tumor, rather than if the tumor produces virus, is of importance for immune recognition.