Discussion

With economic globalisation and environmental changes, the scope of human activities continues to expand, which increases the possibility of pathogens jumping from rodent hosts to humans and domestic animals. The risk of zoonotic EID is elevated in tropical forest areas experiencing land-use change and wildlife biodiversity (mammal species richness), based on the prediction of demographic, environmental, and biological correlates associated with EID events [39]. Traditional virus detection methods rely on cell culture based on visible cytopathic effects and molecular detection of known viral genome sequences. However, the application of NGS technology enables the sequencing of pathogenic genomes in one sample, including unknown pathogens. Hainan Island is an important trade hub and tropical monsoon area with large animal species diversity. Therefore, virome and phylogenetic analyses of rodent-borne viruses in Hainan Island can provide insights into the prevention and control of rodent-borne diseases in China.
The virome data obtained in this study revealed that the risk of rodent-borne pathogens on Hainan Island should not be underestimated. In this study, we collected swab samples from six species and three genera of Muridae, covering nine districts of Hainan Province, to analyse the viral diversity carried by rodents living next to humans and living in natural reserves. The novel CoVs, PestVs, AstVs, ParVs, and PVs identified in this study increase our knowledge of the viral classification of each viral family. Many rodent-borne viruses detected in this study were not only found on Hainan Island but also in the southwestern (Yunnan, Hunan) and northern (Jilin, Ningxia, Inner Mongolia, Xinjiang) regions of China [20], Thailand, Lao PDR, Cambodia, and other parts of the world [18]. Moreover, we also identified widely distributed respiratory tract or enteric RNA viruses, such as AstroV, pestivirus, papillomavirus, picornaviruses, rotavirus, and picobirnavirus, which is consistent with a previous study [20]. However, the abundance of these viruses in the metagenomic analysis in the present study was different from that previously reported in rodent lung samples [18]. Our data was not abundant in viruses that are transmitted through blood or bodily fluids, such as phleboviruses, hepaciviruses, Hantaan viruses, and arteriviruses. This phenomenon may be due to the different tissue tropisms of these viruses. The results of the viral positivity rate suggest that these viruses infect hosts primarily through the respiratory tract, such as CoVs and PestVs. Moreover, some viruses were positive in anal swabs but negative in throat swabs. These results may be attributed to insufficient collection of virus-infected exfoliated host cells during the collection of swabs or to the different replication characteristics of faecal/oral- or respiratory-transmitted viruses, which may be detected at different locations in different cycles [19]. Some viruses, such as AstroV-HMU-5, were detected in both throat and anal swabs, indicating that the host may carry the virus for a long time and may transmit it via the faecal-oral route (food- or water-borne routes) [40]. Additionally, PV-HMU-5 was detected with PV-HMU-6 in the same host, and throat and anal swabs showed PV positivity in multiple hosts, indicating that this host group carried the PVs for a long time, which increases the probability of the virus becoming a potential pathogen.
The murine hepatitis virus in murine coronaviruses was first isolated in 1949 [41], and a variant named sialodacryoadenitis coronavirus was detected in rats in 1970 [42]. Coronavirus recombination events are highly common, such as feline CoV type I and canine coronavirus recombination events [43]. Human CoVs recombine to produce three genotypes [44] and SARS-CoV constantly undergoes evolutionary recombination [45]. The results of the phylogenetic analysis of the RdRp and S protein strains of CoVs were consistent. The four CoVs identified in this study shared high RdRp and S protein identities with known murine CoVs and clustered with murine CoVs under the subgenus Embecovirus in the evolutionary tree. CoV-HMU-4, CoV-HMU-2, CoV-HMU-3, and mouse hepatitis virus showed a relatively close distance in the evolutionary tree, especially CoV-HMU-3 and mouse hepatitis virus, under the same branch. Based on the rodent hosts of known and newly discovered CoVs, our results suggest that the same virus can be detected in different regions and rodents, indicating that rodents carrying CoVs may not exhibir geographic and host specificity. The high similarity of CoVs observed in our study suggests that each lineage shares a common ancestor. Additionally, beta-CoVs were detected in the throat and anal swabs of rodents and only a few clues were found in rodent lungs, suggesting that these viruses primarily infect the upper respiratory tract whereas the absence of alpha-CoVs in the two types of samples indicates that alpha-CoVs may not be the main cause of lower respiratory tract infections in rodents.
Pestiviruses may cause clinical illnesses, such as acute diarrhoea, acute haemorrhagic syndrome, wasting disease, and transplacental infection, leading to fetal death [46], and posing serious health and economic burden. PestV genomic sequences have been detected in bat and rat samples using NGS [8, 47]. In the present study, pairwise alignment revealed that the RdRp of PestV-HMU-1 and PestV-HMU-2 showed low aa identities with known pathogens, suggesting that they may be novel species, and rodents have not been previously reported to carry these viruses. PestV-HMU-1 was carried by the Leopoldamys edwardsi collected from the Huanjinjiaoling reserve and PestV-HMU-2 was carried by the Rattus tanezumi collected from the Chengmai. Moreover, PestV-HMU-1 was also present with ParV-HMU-2 in Leop oldamys edwardsi from the Huanjinjiaoling reserve of Hainan. The abundance and diversity of viruses are high in the central mountainous areas of Hainan.
Multiple infectious ParVs have been detected in animals; porcine ParV causes reproductive failure in pigs [48] whereas canine and feline panleukopenia ParVs are pathogenic to dogs and cats, respectively [49]. ParV infection can also cause serious harm to humans; toad virus, tusavirus, and cutavirus are mainly detected in children with diarrhoea [50]. ParV-HMU-1 and ParV-HMU-2, carried by Rattus andamanensi and Leopoldamys edwardsicollected from the two nature reserves in Baisha, showed low identities with known pathogens. Overall, the virome reported in this study is highly diverse. These results extend our knowledge of viral taxonomy and host range and show that in inaccessible areas, there are still highly diverse viruses that have evolved independently in their unique wildlife hosts. If these viruses cross the host barrier, they are highly likely to cause zoonosis.
Our results provide a profile of the composition of the rodent-borne virome and a baseline for rodent-borne viruses on Hainan Island, China. We detected PestVs and bocaviruses with low identities that were not previously reported to be carried by rodents. The associated viral genome sequences obtained in our study were corroborated, allowing us to further understand the phylogenetic relationship of the viruses on Hainan Island to other regions and obtain data on the diversity and independent evolutionary profiling of rodent-borne pathogens. This study suggests that mammals still harbour a large number of uncharacterised viruses, which require further investigation. This study provides basic data for the prevention and control of rodent-borne infectious diseases, evaluates the potential pathogenicity of newly discovered microorganisms, and provides a baseline for future virome studies and identification of potential pathogens in EIDs.