Pathogenesis of virus-associated tumours
Table of Contents
Research
Our research interest is focused on the pathogenesis and prevention of malignancies caused by viruses. After the successful developments of vaccines against hepatitis B and human Papilloma virus infection, the Epstein-Barr virus (EBV) and the Hepatitis C virus represent the most attractive candidates for vaccine development. We concentrate on oncogenic γ-herpesviruses, with a current emphasis on EBV-associated diseases.
We study the molecular mechanisms that lead to neoplastic transformation and use the gathered information to guide the development of an effective yet safe preventative vaccine. Common to these approaches is the use of a genetic system that allows construction of virus recombinants (see panel 1); these mutants give us precious pieces of information about the function of various genetic elements during the virus natural history of infection but can also be utilised as potential vaccines.
Genetic analysis of EBV functions. The large size of the viral genome precludes the use of conventional cloning techniques to construct variants. We instead generate viral mutants by homologous recombination in E.coli. To this aim, we have inserted a bacterial artificial chromosome (BAC) into the EBV genome, thereby allowing replication in a bacterial environment. Furthermore, the BAC element carries phenotypic expression markers such as the green fluorescent protein that allows live tracking of infected cells. After modification in E.coli, the viral genome can be transferred back into eukaryotic cells from which viruses are produced. These particles carry the mutant virus and can therefore be used to assay the function of the missing genetic element during the various phases of infection.
EBV is aetiologically linked with lymphomas and carcinomas. Whilst we know in some details how EBV transforms primary B lymphocytes, the molecular events that underlie transformation of primary epithelial cells remain enshrouded in mystery. One reason for that was the relative inefficiency of primary epithelial cell infection; we have recently used modified EBVs to circumvent these limitations (see panel 2). This system will allow dissection of the mechanisms that ultimately lead to neoplastic transformation.
The recombinant EBV is stably transfected in 293 cells and can be induced to produce infectious viruses. These EBVs infect various cell lines and primary cells as depicted here for primary epithelial cells. The infected epithelial cells express the GFP protein but also the viral oncogene LMP1
A viral vaccine must be safe yet efficient. Current vaccines make frequent use of attenuated viral strains. However, due to EBV’s oncogenic potential, this approach is unlikely to be applicable to EBV. Other vaccines consist of peptides or of single viral proteins, e.g. the EBV gp350 glycoprotein. Such approach is unlikely to be efficient with herpesviruses whose control requires a potent cellular immunity that cannot be elicited by a single soluble protein. Indeed, vaccination trials with gp350 failed to prevent infection with wild type EBV, although it seemed to reduce the frequency of infectious mononucleosis, a common complication of EBV primary infection. We have chosen an intermediate alternative between these two extremes. The study of a mutant that lacks packaging signals (terminal repeats) previously showed that it produces virions devoid of EBV DNA. These so-called virus-like particles (VLP) or light particles (LP) retain all structure proteins and can elicit a T cell response (see panel 3)
The terminal repeat-null EBV mutant produces VLP/LP. These particles are devoid of DNA as shown in the Gardella gel analysis combined with a Southern blot performed with an EBV-specific probe (gp350) (A, top). In contrast, a western blot with an antibody against EBV BNRF1 that is present in the mature virion shows that the mutant produces virions (A, bottom). This assertion was confirmed by electron microscopy that identified mature particles devoid of infectious DNA (B, top). DNA-packaged wild type particles are shown in the bottom panel as a positive control. Crucially, these VLP/LP are indistinguishable from their wild type counterparts in terms of their ability to elicit a T cell response as shown in this IFN-γ release assay (C).
