Studies on polyomavirus virus-like particles : as vaccines and vectors for immune and gene therapy

Abstract: Virus-like particles (VLPs) are similar to natural virus particles except that they lack viral genes. They have a similar cellular uptake to the natural virus from which they are derived but are non-infectious and can therefore not reproduce themselves. VLPs have been used as a model to understand viral entry, infection and cell tropism, but have also been shown to be useful in other areas e.g. as vaccines against viral infection, as well as carriers for molecules in immune and gene therapy. This thesis is based on VLPs from two related viruses, murine polyomavirus (MPyV) and murine pneumotropic virus (MPtV) and the aim has been to investigate their possible use as viral vaccines and as vectors in gene and immune therapy. Both viruses consist of only a few genes and a protein capsid surrounding them. MPyV was discovered when it was shown to induce tumors in mice, thereof the name, “polyoma”, Greek for many tumors. It is easy to grow in cell culture, and because of its oncogenic potential and its small size it has been an important research tool in molecular biology. Studies on MPyV have led to many discoveries in understanding cellular events like DNA replication, cell growth regulation, and genes involved in tumor development. The MPtV on the other hand, is non-oncogenic, is difficult to grow and has not been well studied. We have been successful in producing VLPs from both these viruses, and used them as immunogens and as carriers for tumor antigens. They should both be suitable for therapeutic use in humans, since they are of non-human origin, and humans have no pre-existing immunity against them. In the first paper, the aim was to optimize MPyV-vaccination by examining the importance of the route of administration and the VLP structure. All, even immune deficient, mice were protected against subsequent MPyV infection by changing from intraperitoneal to subcutaneous VLP vaccination. Furthermore, VLPs were more efficient, than the more linear GST-VP1 in viral protection and in the induction of an antibody response. The conclusion from this study was that, the route of administration, and the antigenic structure are important. This antibody response is of value for MPyV vaccination, but more of an obstacle when using MPyV-VLPs as antigen carriers, where the antibodies may abolish the effect of a similar second treatment. Therefore the aim of the second study was to study VLPs from MPtV, as a possible complement to e.g. MPyV-VLPs. MPtV-VLPs were successfully produced; they entered all cell types tested and did not cross-react with MPyV-VLPs, which make them suitable as complement to MPyV-VLPs in prime-boost therapy. In the third paper MPyV-VLPs carrying the oncoprotein Her2, Her21-683PyVLPs, were successfully produced and were used to vaccinate against Her2-expressing tumors. Vaccination with Her21-683PyVLPs efficiently protected mice from tumor outgrowth of the transplantable Her2 positive tumor D2F2/E2, as well as against spontaneous mammary carcinoma outgrowth in BALBneuT mice, transgenic for rat Her2. In the fourth study, Her21-683PyVLP vaccination was compared to vaccination with Her21-683PyVLPs loaded on dendritic cells (DCs), with regard to efficiency and to anti-VLP response. Vaccination with Her21-683PyVLP loaded DCs was more efficient in protecting mice against outgrowth of D2F2/E2 tumor where a lower Her21-683PyVLP dose was sufficient for full protection, compared to vaccination with Her21-683PyVLPs alone. Furthermore; vaccination with Her21-683PyVLP loaded DCs resulted in lower anti-VLP titers. In conclusion, VLPs derived from mouse polyomaviruses can be used to vaccinate against a subsequent polyoma infection. Moreover, they can be used as carriers for molecules in immune and gene therapy. We show that VLPs have substantial potential for use in cancer immune therapy, and that MPyV-VLPs and MPtV-VLPs, due to lack of cross-reactivity, should be complementary and suitable for prime-boost therapy.