EPIDEMIOLOGY
Prior to 2013, detection of KoRV in epidemiological studies was based on pol gene-specific primers, which are not capable of differentiating KoRV A from other KoRV variants. Work conducted in that period indicated that KoRV was ubiquitous (Simmons et al.
2012) and endogenous in Qld and northern NSW; Southern hybridisation of genomic DNA demonstrated similar intensity and banding patterns across a range of tissues within individual koalas (Hanger et al. 2000) and the virus was demonstrated in single-cell fluorescent PCR of koala sperm, together with fluorescent in situ hybridisation of whole koala sperm of a Qld koala (Tarlinton et al. 2006). However, unusual among ERVs, the virus exhibited some traits of an exogenous virus, presenting as a full-length, replication-competent genome; viral transcription was evident in south-east Qld koalas (Tarlinton et al. 2005) and there was considerable variation in both the number and pattern of KoRV insertions between individuals (Tarlinton et al. 2006). The partial prevalence and lower proviral copy number reported in koalas from Vic. and SA suggested that KoRV might comprise a rare opportunity to observe the process of retroviral endogenisation as the virus moved southward following a presumed recent introduction (Tarlinton et al. 2006).Since the discovery of KoRV variants, studies using the env gene variant-specific primers have confirmed that KoRV A env gene is ubiquitous in all koalas but the pol gene, required for replication, occurs in only about 25% of koalas in Vic. and SA (Blyton et al. 2022a). There is now some evidence that KoRV A may be exclusively endogenous, at least in Qld, as it exhibits a markedly lower level of genetic diversity and is more closely related to GALV, when compared with other variants (Chappell et al. 2017). This is supported by a recent study that found that all 39 KoRV A proviral loci identified in the sire and dam of a Qld koala were vertically transmitted to the progeny and no new proviral loci were identified (Ishida et al.
2015). This absence of new KoRV A proviral integrants, despite presence of KoRV A RNA in plasma, suggests superinfection by KoRV A is unlikely (at least in these individuals). Retroviral superinfection resistance is an interference mechanism, established after primary infection and which prevents the cell being infected by a similar type of virus using the same receptor, while susceptibility to infection by variants utilising other receptors is unaltered. KoRV variants display similar diversity of receptor use to that of FeLV and, where characterised, it is evident that KoRV variants use different receptors to KoRV A to infect cells (Shojima et al. 2013; Xu and Eiden 2015), which would allow them to circumvent the KoRV A receptor blockade expected to be induced by KoRV A infection. For example, the receptor for KoRV A is inorganic phosphate transporter 1, which is known to be a receptor for both the pathogenic retroviruses GALV and FeLV subgroup B. On the other hand, KoRV B and KoRV J use the thiamine transport protein 1 receptor, which is the same as that used by pathogenic FeLV subgroup A that is transmitted between cats. The receptors for remaining subtypes have not been identified, but they appear to be diverse, based on sequence diversity at the relevant env gene locus (Chappell et al. 2017).To what extent, or if, KoRV A has endogenised in Vic. and SA has not been examined in full. The finding of proviral copy numbers in orders of magnitude lower than what would be expected for the presence of a single copy in every cell for many animals in Vic. (Simmons et al. 2012), and the significant difference in KoRV A copy number per genomic unit between fecal and ear punch DNA samples (Wedrowicz et al. 2016), suggests that KoRV is not endogenous in these animals and likely reflects ongoing exogenous infection. Why these patterns change at the NSW/Vic. border, as appears to be the case so far, is uncertain but could possibly relate to relatively recent introduction caused by absence of the putative ancestral host in the south and/or the absence of a natural genetic or historical gradient of koala populations from north to south; most of the Vic. and all the SA populations were reintroduced from genetically bottlenecked insurance populations following harvest-induced population crashes in the early part of the 20th century.
All variants other than KoRV A are presumed not to be ERVs, based on their high genetic diversity (Chappell et al. 2017), incomplete penetrance in related family groups and a relatively low number of proviral copies per cell in infected animals (Oliveira et al. 2006; Denner and Young 2013; Shojima et al. 2013; Xu and Eiden 2015). Longitudinal studies in free-ranging (Quigley et al. 2018) and managed koalas (Joyce et al. 2022) indicate a high rate of transfer of subtypes A, B, D, I and K from mothers to young, though the route is unclear; in-utero, via milk or pap, and via generation by recombination or mutation of inherited KoRV elements, all remain possible. There is no evidence to date for sexual transmission. The possibility of transmission by an arthropod vector has been suggested and transmission of equine infectious anaemia virus by biting flies is well documented, although it is a lentivirus. Although KoRV-infected blood has been detected in ticks and mosquitoes (Simmons 2011), and fleas have been shown to transfer the related gammaretrovirus FeLV to non-viraemic blood (Vobis et al. 2003), actual transmission to a host has not been demonstrated.
The distribution of variants in zoos is not fully known. KoRV B and KoRV J are not identical but are sufficiently similar to be considered the same subgroup (Shimode et al. 2014). KoRV J was discovered in koalas in Japanese zoos (Shojima et al. 2013) and KoRV B was originally reported in koalas in Los Angeles Zoo (Xu et al. 2013) and has since been reported in two European Zoos (koalas from San Diego Zoo, of Qld origin) (Fiebig et al. 2015a, 2016) and in two zoos in NSW (koalas of NSW origin) (Maher and Higgins 2016). KoRV C and D were reported in a koala in Japan (Qld origin) (Shimode et al. 2014). KoRV B/J, C and D were not detected from DNA of 10 museum koala skin samples collected between 1870 and 1938 and the authors suggest that these variants may have arisen more recently (Tsangaras et al. 2014). KoRV E and F were discovered more recently, again in USA zoo koala populations (Qld origin) (Xu and Eiden 2015).
In free-ranging populations of koalas, prevalence of non-KoRV A subtypes varies geographically. The most comprehensive survey to date (Blyton et al. 2022a) did not detect non-KoRV A subtypes from Southern and Vic. populations. In Qld and NSW populations KoRV D was at high prevalence across almost all populations studied. Other subtypes were generally of lower prevalence and were geographically restricted and genetically diverse. This, coupled with detection of non-KoRV A subtypes only in koalas with competent endogenous KoRV A suggests the likelihood that these non-A subtypes arise sporadically in populations by mutation and recombination of KoRV A, with selection for variants able to avoid superinfection blockade by use of alternative receptors for cell entry. Overall KoRV loads, as measured by KoRV proviral pol gene, also varies geographically, with a cline from 20 to 97 copies per cell in Qld and northern NSW to <20 copies per cell in the two southern NSW populations studied and, among koalas with pol gene in Vic. and SA, 1 copy per cell. The epidemiology of KoRV A is complex. Three KoRV A env sequences have been identified in northern koalas and another, distinct, KoRV A env sequence was detected in all Vic. koalas studied but in the latter case, 70-75% of these exist in the absence of the KoRV pol gene suggesting that some KoRV A env fragments are endogenous in southern koalas, making them available for recombination and potential beneficial or detrimental genomic interference. In addition to KoRV A, several defective and recombinant versions of KoRV exist (recKoRV). The significance of these to KoRV dynamics and evolution remains uncertain.
2.