Gammaretrovirus replication in human cells: implications for experimental xenografts and risk of zoonosis

I have explored the infection of human cells by the gammaretroviruses XMLV (xenotropic murine leukaemia virus) and FeLV (feline leukaemia virus). For XMLV the main aim was to assess the risks and consequences of contamination of human-mouse xenografts. For FeLV, the aim was to explore the susceptibility of human cells in vitro to assess the risks and identify the barriers to zoonotic infection in vivo. Xenografting of human cells to mice is used commonly in many disciplines of biomedical science. Infection of xenograft-derived cell lines has been reported as a common observation but it was unclear how many of these lines were infected due to in vitro cross-contamination rather than de novo infection in vivo. I conducted a prospective study and demonstrated that more than 40% xenografts passaged through BALB/c nude mice acquired XMLV. Xenografts passaged through NSG, another commonly used mouse strain for engraftment studies, did not yield replication-competent XMLV, although there was some evidence of activation of related replication-defective viruses. The source of XMLV in BALB/c mice appeared to be Bxv1, a locus encoding a replication competent virus which I showed was absent from NSG mice. I also showed that de novo isolated XMLV replicates to high copy number in MCF7 breast cancer and induces subtle changes in growth properties. Transcriptome analysis suggested that up regulation of EGR1 (early growth response gene 1) may be responsible for these growth effects. I also analysed susceptibility to XMLV infection in human cells and showed that while primary PBMCs are highly resistant, many cell lines can be infected. The Raji Burkitt’s lymphoma cell line was found to be highly susceptible to XMLV and displayed a much larger transcriptional response compared to MCF7, with marked up regulation of a series of markers of innate immunity. I examined the susceptibility of human cells to infection with FeLV B, the viral subtype which appears to be the most likely zoonotic agent. Cell lines of human origin were found to vary in susceptibility with no clear relationship to cell lineage. While some cell lines were completely susceptible for FeLV B, most showed limited virus replication and G to A hypermutation that correlated with the expression of APOBEC family members. Primary PBMCs and some leukaemia cell lines showed profound resistance to FeLV B infection at an early stage of replication and accumulated proviral DNA with only a few mutations. The mediator of this early block has not yet been identified but appears likely to be important in preventing cross-species spread of gammaretroviruses to the human population