Infectome analysis of bat kidneys from Yunnan province, China, reveals novel henipaviruses related to Hendra and Nipah viruses and prevalent bacterial and eukaryotic microbes

Abstract

Bats are natural reservoirs for a wide range of microorganisms, including many notable zoonotic pathogens. However, the composition of the infectome (i.e., the collection of viral, bacterial and eukaryotic microorganisms) within bat kidneys remains poorly understood. To address this gap, we performed meta-transcriptomic sequencing on kidney tissues from 142 bats, spanning ten species sampled at five locations in Yunnan province, China. This analysis identified 22 viral species, including 20 novel viruses, two of which represented newly discovered henipaviruses closely related to the highly pathogenic Hendra and Nipah viruses. These henipaviruses were found in the kidneys of bats inhabiting an orchard near villages, raising concerns about potential fruit contamination via bat urine and transmission risks to livestock or humans. Additionally, we identified a novel protozoan parasite, tentatively named Klossiella yunnanensis, along with two highly abundant bacterial species, one of which is a newly discovered species-Flavobacterium yunnanensis. These findings broaden our understanding of the bat kidney infectome, underscore critical zoonotic threats, and highlight the need for comprehensive, full-spectrum microbial analyses of previously understudied organs to better assess spillover risks from bat populations.

Fig 1. Bat kidney sampling and species identification.

(A) Map showing the five sampling locations in Yunnan province, China, with nearby countries (Myanmar, Laos, Vietnam, Thailand, Cambodia, and Malaysia) shown for reference. Pie charts indicate the species composition of the bats sampled at each site. The basemap shapefile used in ArcGIS was obtained from the publicly available GADM data set (https://gadm.org/download_country.html). (B) Unrooted phylogenetic tree inferred from full-length COX1 gene sequences of bat kidney samples analyzed in this study. Colors correspond to different bat species, matching the color scheme used in the pie charts. Branch lengths are scaled to the number of nucleotide substitutions per site.

Fig 2. Overview of the bat kidney infectome.

(A) Numbers of total reads (light blue) and microbial reads (orange) for each library. (B) Number of viral, bacterial, and eukaryotic microbial species detected, with color schemes corresponding to those used in panel C. (C) Heatmap illustrating the distribution and relative abundance of viral, bacterial, and eukaryotic microbes, represented as RNA abundance (RPM: reads per million non-rRNA reads) in each library. Host species and orders are labeled at the top and color-coded according to their respective categories.

Fig 3. Phylogenetic diversity of viruses identified in this study.

Phylogenetic trees of viruses from 12 virus families estimated using the maximum likelihood method based on conserved protein sequences (RdRp for RNA viruses, NS1 for Parvoviridae, and DNA polymerase for Hepadnaviridae). Colored dots on the trees, corresponding to host genera as indicated in the legend, represent viral species identified in this study. Red stars mark members of known mammal-associated viral lineages. All trees are mid-pointed rooted for clarity only with horizontal branch lengths depicting the number of amino acid substitutions per site.

Fig 4. Characterization of the novel henipavirus species examined in this study.

(A) Genome organization and sequencing coverage of two novel henipavirus species. Coverage across the full-length genome is displayed, with open reading frames (ORFs) depicted as colored arrows below the coverage plots. Regions confirmed by Sanger sequencing for Yunnan bat henipavirus 1 are marked with a red bar beneath the coverage graph. (B) Maximum likelihood phylogenetic trees estimated using amino acid sequences of each gene within the genus Henipavirus, rooted with J-virus. Color blocks indicate different species groups, and newly identified viruses are marked with solid red circles. All trees are mid-pointed rooted for clarity only with horizontal branch lengths depicting the number of amino acid substitutions per site.

Fig 5. Gene expression profiles, prevalence and identification of the two bacterial microbes.

(A) Maximum likelihood phylogenetic tree of the genus Flavobacterium, constructed using the rpoB gene. (B) Maximum likelihood phylogenetic tree of the genus Phyllobacterium, constructed using the groEL gene. (C) Top 25 expressed genes (measured in RPM) for Flavobacterium yunnanensis and Phyllobacterium calauticae in pools YNBS16 and YNBS01, respectively (left panel), compared with their expression in other positive pools (right panel).

Fig 6. Identification and characterization of a eukaryotic microbe.

(A, B) Phylogenetic trees of Klossiella, estimated using nucleotide sequences of the cox1 gene (a) and cytb gene (b). Colored dots indicate newly identified eukaryotic species, with colors corresponding to host genera. (C) Transcriptomic profiles of the Klossiella mitochondrion, represented as RPM, across positive pools. (D) Spearman’s correlation analysis showing the relationship between the total relative abundance (RPM) of totiviruses, narnaviruses, and Klossiella yunnanensis.