The Epidemiology and Pathogenesis of Koala Retrovirus

ABSTRACT Koala retrovirus (KoRV) is a recently characterised Gammaretrovirus which shares strong homology with Gibbon ape leukaemia virus (GALV). KoRV appears to be widespread throughout wild koala populations in Australia and has been demonstrated to be endogenous in koalas in south-east Queensland. Because of the potential for KoRV to be a significant pathogen in koalas, a study was undertaken to gain a better understanding the prevalence of KoRV throughout Australia. Samples were obtained from over 700 wild koalas, ranging from Blair Athol in the north to Kangaroo Island in South Australia. This study demonstrated koala populations in Queensland and New South Wales have the highest prevalence, with 100% of 377 animals tested being KoRV provirus-positive. The prevalence declined to 73% of 89 animals tested from mainland Victoria, 28% of 80 animals from the Victorian islands, and 15% of 162 animals from Kangaroo Island. Discovery of KoRV-positive koalas on Kangaroo Island was an interesting finding because the only previous report in 2006 found all of 26 tested animals from the island to be KoRV-free. These discordant results may indicate a recent introduction of the virus to Kangaroo Island or may perhaps reflect the small sample size in the previous study. An additional interesting finding from the prevalence study was the appearance of a much higher KoRV proviral copy number in genomic DNA from koalas in Queensland compared to other states. The mean estimated proviral copy number per cell nucleus in eight Queensland koalas was 165 and the copy number appeared remarkably uniform between animals (range 139-199 copies per cell). The proviral copy number in Victorian koalas was much lower and more variable. Divided into three groups each of eight koalas based on results in three separate PCR assays, the Victorian koalas had means of 1.5, 1.53 x 10-3, and 1.29 x 10-4, respectively. The large variation in proviral load in animals from different populations suggests that while KoRV may have endogenised in Queensland koalas, it may exist in some Victorian koalas in only its exogenous form. The widespread distribution of KoRV raises the question of its mode of transmission. Attempts were made to demonstrate the horizontal transmission of KoRV in vitro with mosquitoes, tabanid flies and paralysis ticks. KoRV provirus was detected in mosquitoes sourced from captive koala pens, indicating the presence of koala blood in the insects. Attempts to demonstrate horizontal transmission of KoRV viral RNA with mosquitoes and tabanid flies using a membrane feeding device were unsuccessful. However, feeding trials using a Pasteur pipette fixed in place over the tick’s hypostome demonstrated horizontal transmission of KoRV viral RNA to an artificial feeding medium from some individual paralysis ticks. Although the paralysis tick result was an interesting finding, it is unlikely to represent a major route of transmission in nature considering the biology of these ticks. A previous study in 2005 demonstrated a significant link between level of KoRV viraemia and lymphoid tumours in captive koalas. Two trials were conducted in the current study to better assess the link between KoRV viraemia and disease. The first trial involved a total 18 captive and wild koalas. Although there was a positive correlation between level of viraemia and chlamydial copy number, there were some factors which cast doubt on the relevance of these results. In the second larger trial, 99 wild koalas were subjected to a detailed clinical exam including an assessment of the severity of and other diseases. This trial failed to demonstrate any correlation between level of KoRV viraemia and disease or chlamydial load and it was concluded that this result more accurately depicts the situation in nature. In an attempt to explain the close homology between KoRV and GALV a search was conducted for a genetically similar virus in a potential intermediate host species. DNA samples from a large range of native and introduced vertebrate species were screened with PCR in order to detect proviral sequences related to KoRV or GALV. The discovery of a partial proviral sequence in DNA from the grassland melomys, Melomys burtoni, a native Australian rodent, was an exciting development. Nucleotide sequence of approximately one third of the putative 8000 base pair genome was obtained. The sequence of this novel Gammaretovirus, designated Melomys retrovirus (MeRV), is more closely related to GALV than to KoRV, and contains open reading frames. The MeRV provirus was detected in all 17 M. burtonii individuals tested, but was not detected in any of 15 individuals of the closely related species, M. cervinipes. The finding of MeRV provirus in 100% of individuals indicates that either the virus is of unusually high prevalence for a retrovirus or is endogenous in the species. However, the finding of open reading frames in the MeRV genome suggests that the virus may be an exogenous virus, although attempts to propagate the virus in cell culture were unsuccessful. The discovery of MeRV adds to the mystery surrounding the KoRV/GALV relationship. MeRV is so closely related to both viruses that it is likely to have played a role in the process by which a retrovirus spread between koalas in Australia and gibbons in South-East Asia. It is interesting to speculate on how this process occurred, and further research on the biology of all three viruses will hopefully shed more light on this fascinating question