Genotype and Phenotype Characterization of Rhinolophus sp. Sarbecoviruses from Vietnam: Implications for Coronavirus Emergence

Abstract

Bats are a major reservoir of zoonotic viruses, including coronaviruses. Since the emergence of SARS-CoV in 2002/2003 in Asia, important efforts have been made to describe the diversity of Coronaviridae circulating in bats worldwide, leading to the discovery of the precursors of epidemic and pandemic sarbecoviruses in horseshoe bats. We investigated the viral communities infecting horseshoe bats living in Northern Vietnam, and report here the first identification of sarbecoviruses in Rhinolophus thomasi and Rhinolophus siamensis bats. Phylogenetic characterization of seven strains of Vietnamese sarbecoviruses identified at least three clusters of viruses. Recombination and cross-species transmission between bats seemed to constitute major drivers of virus evolution. Vietnamese sarbecoviruses were mainly enteric, therefore constituting a risk of spillover for guano collectors or people visiting caves. To evaluate the zoonotic potential of these viruses, we analyzed in silico and in vitro the ability of their RBDs to bind to mammalian ACE2s and concluded that these viruses are likely restricted to their bat hosts. The workflow applied here to characterize the spillover potential of novel sarbecoviruses is of major interest for each time a new virus is discovered, in order to concentrate surveillance efforts on high-risk interfaces.

Keywords: Sarbecovirus; horseshoe bat; spillover; virus evolution.

Figure 1

Bat sampling in Northern Vietnam, 2021. Maps were constructed with Quantum GIS software (https://qgis.org/en/site/ accessed on 6 September 2023). Sampling performed during February–March 2021 is presented by a black dot (Mission 1), and sampling performed during July 2021 is presented by a red dot (Mission 2). The size of the pie charts is proportional to the number of bats sampled in each location. Bats were captured using mist nets and hand nets. Each captured bat was weighted using a Pesola spring scale, then the bats were individually kept in porous humidified cotton bags and placed in a cool dry place for sampling within six hours. The following basic set of samples was collected from each bat, and included saliva (oropharyngeal swab), feces (fresh fecal sample) or rectal swab, blood (serum; rbc/wbc pellet) and urine (free catch method or urogenital swab). After sampling, bats were released at the collection site.

Figure 2

Genomic characterization of Vietnamese sarbecoviruses. (A) Phylogenetic analysis of ORF1a, ORF1b, spike, ORF6, ORF7a and nucleoprotein genes of Vietnamese sarbecoviruses. Strains are represented by symbols and colored according to their position in the different clades. (B) Recombination analysis of Vietnamese sarbecoviruses. Recombinant fragments and corresponding breakpoints are indicated below each backbone and colored according to the donor genome sequence. Breakpoint coordinates correspond to the position in the alignment available at 10.6084/m9.figshare.23821428 (accessed on 6 September 2023). SARS-CoV-2 genome (NC_045512) served as reference for protein domains.

Figure 3

Heatmap of nucleotide identities observed at the whole genome (A) and at the spike (B) levels between Vietnamese and 221 representative Sarbecovirus genomes. Aligned sequences were sorted by similarity. Low-to-high nucleotide identities are represented from blue to yellow. SARS-CoV- and SARS-CoV-2-related sequences are highlighted by white squares, and bat-specific clusters are highlighted by red squares. Identity matrices were calculated with CLC Main Workbench 21.0.4 (Qiagen) and the heatmap was created using GraphPad Prism 9.5.1 (GraphPad software). AS = Asian isolates; AF/EU = African/European isolates; Rh. = Rhinolophus. The complete identity matrices are presented in Table S3.

Figure 4

Phylogenetic analysis of Vietnamese and 221 representative sarbecoviruses. Trees were inferred for the RdRP and spike genes, and midpoint-rooted for clarity. Bootstraps above 50 are represented by black dots. Sarbecoviruses from Rhinolophus sp. bat species are represented by colored squares while Chaerephon plicatus and Aselliscus stoliczkanus sarbecoviruses are represented by dashed lines. Complete phylogenetic reconstructions are available in Figures S3 (RdRP) and S4 (spike).

Figure 5

Estimation of the binding free energy of representative sarbecovirus RBDs for different animal ACE2 receptors. Scale was set up based on the binding free energy of human SARS-CoV-2 with human ACE2 (−10.0; white). Strong predicted binding free energy (below −10.0) is represented by shades of green while weak binding free energy (above −10.0) is represented by shades of red. “R” represents ACE2 sequence that was manually reconstructed from public and private datasets (see Section 2).

Figure 6

ACE2-dependent entry assay of Rhinolophus sp. sarbecovirus pseudotypes. Results of a single experiment performed in triplicate are expressed in relative luminescence units (RLUs) produced by the firefly luciferase expressed from the lentiviral vector. Error bars indicate SD. Statistical significance was assessed using one-way ANOVA and was compared with the RLU signal of an empty vector as a reference: ns: non-significant; ****: p < 0.0001.