Diversity of fecal viromes and zoonotic risk assessment in captive wild felids using viral metagenomics.

Emerging viral diseases-particularly zoonotic pathogens-affect the health and conservation of endangered felids, including Panthera tigris altaica (Amur tiger) and Panthera pardus (leopard). To address this challenge, we employed a viromics approach to investigate the diversity of the fecal virome in wild felids and assess its zoonotic potential. Using in-depth metagenomic sequencing and analysis of fecal samples from captive wild felids housed in a wildlife institution, this study characterized the enteric virome and evaluated associated risks. A total of 18 viral families and 48 viral genera were identified. The DNA virus community exhibited stability in abundance and composition, dominated by the phyla Heunggongvirae and Bamfordvirae. Within Heunggongvirae, the class Caudoviricetes was the core component, with its abundance aligning with the intestinal bacterial community, suggesting a potential role of these bacteriophages in regulating microbial ecology. Additionally, sequences of the family Poxviridae, homologous to Variola virus (VARV), were detected. In contrast, the RNA virus community displayed higher diversity and variability, with the order Ortervirales as the predominant group. Sequences highly homologous to feline leukemia virus (FeLV) were repeatedly identified, suggesting potential latent infections. The detection of sequences related to rare environmental viruses, such as Casadabanvirus, highlights the potential risk of cross-species virus transmission under captive conditions. Stability analysis revealed that dominant DNA virus groups exhibited low abundance variability across samples. In contrast, unclassified RNA viral taxa showed higher abundance variability. KEGG functional annotation mapped DNA viral contigs primarily to microbial metabolic modules. Conversely, RNA assemblies extensively mapped to eukaryotic pathways (e.g., arachidonic acid and energy metabolism); due to the total nucleic acid extraction methodology, these mappings primarily reflect co-extracted host transcriptomic background rather than viral-encoded functions, providing an indirect snapshot of the concurrent enteric microenvironment. These baseline data delineate the virome structure in captive environments and provide practical targets for zoological biosecurity and proactive veterinary surveillance.