Isolation and Characterization of Distinct Rotavirus A in Bat and Rodent Hosts

Abstract

Rotavirus A (RVA) causes diarrheal disease in humans and various animals. Recent studies have identified bat and rodent RVAs with evidence of zoonotic transmission and genome reassortment. However, the virological properties of bat and rodent RVAs with currently identified genotypes still need to be better clarified. Here, we performed virus isolation-based screening for RVA in animal specimens and isolated RVAs (representative strains: 16-06 and MpR12) from Egyptian fruit bat and Natal multimammate mouse collected in Zambia. Whole-genome sequencing and phylogenetic analysis revealed that the genotypes of bat RVA 16-06 were identical to that of RVA BATp39 strain from the Kenyan fruit bat, which has not yet been characterized. Moreover, all segments of rodent RVA MpR12 were highly divergent and assigned to novel genotypes, but RVA MpR12 was phylogenetically closer to bat RVAs than to other rodent RVAs, indicating a unique evolutionary history. We further investigated the virological properties of the isolated RVAs. In brief, we found that 16-06 entered cells by binding to sialic acids on the cell surface, while MpR12 entered in a sialic acid-independent manner. Experimental inoculation of suckling mice with 16-06 and MpR12 revealed that these RVAs are causative agents of diarrhea. Moreover, 16-06 and MpR12 demonstrated an ability to infect and replicate in a 3D-reconstructed primary human intestinal epithelium with comparable efficiency to the human RVA. Taken together, our results detail the unique genetic and virological features of bat and rodent RVAs and demonstrate the need for further investigation of their zoonotic potential. IMPORTANCE Recent advances in nucleotide sequence detection methods have enabled the detection of RVA genomes from various animals. These studies have discovered multiple divergent RVAs and have resulted in proposals for the genetic classification of novel genotypes. However, most of these RVAs have been identified via dsRNA viral genomes and not from infectious viruses, and their virological properties, such as cell/host tropisms, transmissibility, and pathogenicity, are unclear and remain to be clarified. Here, we successfully isolated RVAs with novel genome constellations from three bats and one rodent in Zambia. In addition to whole-genome sequencing, the isolated RVAs were characterized by glycan-binding affinity, pathogenicity in mice, and infectivity to the human gut using a 3D culture of primary intestinal epithelium. Our study reveals the first virological properties of bat and rodent RVAs with high genetic diversity and unique evolutional history and provides basic knowledge to begin estimating the potential of zoonotic transmission.

Keywords: Rotavirus A; bats; genetic diversity; glycan specificity; human intestinal epithelium model; rodents.

FIG 1

Isolation of rotavirus A (RVA) from wild animals in Zambia. (A) Map of sampling sites in Zambia. Egyptian fruit bats were captured in Lusaka and Shimabala, rodents and shrews were captured in Mpulungu. (B) Schematic workflow of virus isolation-based RVA screening. MA104-T2T11D cells were inoculated with fecal suspensions and cultured in roller tubes. After a single blind passage, the culture supernatants were pooled, concentrated, and analyzed by next-generation sequencing (NGS). If RVA genomes were detected, passaged culture supernatants were screened for RVA by reverse transcription PCR (RT-PCR) with specific primers for RVA sequences identified in the NGS analysis. Figure art was created with BioRender.com. (C) MA104-T2T11D cells infected with 16-06 and MpR12 were stained for RVA (green) and nuclei (blue). Scale bars = 50 μm. (D) Negative-stain electron micrographs of 16-06 and MpR12 virions. Scale bars = 100 nm.

FIG 2

Growth kinetics of 16-06, MpR12, and Wa in different culture conditions. Monolayered cells were inoculated with 16-06 (MOI = 0.005), MpR12 (MOI = 0.1) and Wa (MOI = 0.1). Progeny virus in the supernatants was harvested at the indicated time points (hours postinfection [hpi]) and titrated by a focus assay. (A) Infected MA104-T2T11D cells were cultured in static and rotary culture conditions. (B) Viruses were infected and cultured in MA104, MA104 with trypsin (25 μg/mL or 2.5 μg/mL), or MA104-T2T11D cells. Virus titers at 48 hpi for each virus are given as means ± standard deviation (SD) of triplicate data from a representative experiment. Statistical analysis was performed using a Student’s t test (A) or one-way analysis of variance (ANOVA) with Tukey’s test (B). ***, P < 0.001; **, P < 0.01; *, P < 0.05. (C) Representative focus induced by 16-06, MpR12 and Wa in MA104 cells and MA104-T2T11D cells. Infected cells were overlaid with 0.5% agar and cultured for 72 hpi. Fixed cells were stained for RVA (green) and nuclei (blue). Scale bars = 200 μm.

FIG 3

Whole-genome characterization of the isolated RVAs. (A) Comparison of genome constellations between the isolated RVAs and related RVA strains. Identical genotypes are shown in the same color, and genotypes of undetermined segments are indicated by “x.” (B and C) Maximum-likelihood tree of VP7 (B) and VP4 (C) genes based on the sequences of isolated RVAs, bat-derived RVAs, rodent-derived RVAs, and type strains of each genotype. Phylogenetic trees were constructed by the maximum-likelihood method using models of general time-reversible with gamma rate categories and invariant sites (GTR+G+I) with bootstrap values of 1,000 replicates. Avian and raccoon RVAs were regarded as the outer groups. The isolated RVAs are indicated in red. Bat-specific genotypes and rodent- and shrew-specific genotypes are highlighted in blue and yellow, respectively. Genotypes which include bat-derived and nontypical human RVAs are colored purple. Genotypes consisting of RVAs from multiple animal species are highlighted in green.

FIG 4

Involvement of sialic acid on the infectivity of the isolated RVAs. (A and B) MA104-T2T11D cells were pretreated with neuraminidase (NA) at 200 mU/mL or reaction buffer (mock-treated) and infected with 16-06 and MpR12. The SA11 and DS-1 strains were used as positive and negative controls, respectively. (C and D) MA104-T2T11D cells were infected with RVA pretreated with N-acetylneuraminic acid (NeuAc), N-glycolylneuraminic acid (NeuGc), or reaction buffer only (mock-treated). (A and C) Cells were stained with anti-RVA antibody (green) and Hoechst 33342 for nuclei (blue). Scale bars = 200 μm. The figures shown are representative images. (B and D) The number of infected cells with RVA is expressed as a percentage of the mock-treated control. Means ± SD of triplicate data from a representative experiment are shown in the graph. Statistical analysis was performed using multiple t tests with the Holm-Sidak method for the NA assay and a one-way ANOVA with Dunnett’s test for the SA inhibition test. ***, P < 0.001; **, P < 0.01; *, P < 0.05.

FIG 5

Pathogenicity of the isolated RVAs in suckling mice. Three-day-old BALB/c mice were orally inoculated with 1.0 × 10⁵ FFU of RVA strains 16-06, MpR12, SA11, or phosphate-buffered saline (mock) by gavage (n = 7 in each group). (A) Incidence rate of diarrhea in each group was monitored from 0 to 7 days postinfection (dpi). (B) Fecal consistency in each group was scored according to the criteria described in Methods. (C) Average viral RNA copy numbers in feces from 0 to 7 dpi were calculated based on the results of qRT-PCR. Dashed line indicates detection limit of qRT-PCR. (D) Areas under the curves (AUCs) were calculated based on the diarrheal score and viral RNA copy number in feces of each mouse. (E) Viral RNA copy number of small and large intestines of infected mice at 1, 2, 3, and 5 dpi were determined using reverse transcription-quantitative PCR (qRT-PCR). (F) Infected sucking mice were euthanized at 3 dpi for histopathological examinations. Representative images of the small intestines of 16-06-, MpR12-, or SA11-infected mice and the control mice are shown. At 3 dpi, histopathological changes in the infected mice were characterized by vacuolization of the enterocytes in the villus tips. Hematoxylin and eosin staining (H&E). Scale bars = 500 μm. Areas in black squares are magnified in lower panels. (G) Neutralizing titers of mouse sera at 15 dpi were expressed as the dilution at which the number of viral foci was reduced by 50% compared to the no-serum control (FRNT₅₀). Dashed line indicates detection limit of focus reduction neutralization test (FRNT). Means ± SD of each group from representative experiments are shown in the graph. Each dot represents one value from each mouse. Statistical analysis was performed using a one-way ANOVA with Tukey’s test for AUC analysis and multiple t tests with the Holm-Sidak method for the NA assay and the neutralizing antibody titers of suckling mice. ***, P < 0.001; **, P < 0.01; *, P < 0.05.

FIG 6

Infectivity of the isolated RVAs in a human small intestinal epithelial ex vivo model, SMI-100. (A) Growth kinetics of 16-06, MpR12, and Wa in SMI-100. SMI-100 was inoculated with 1.0 × 10⁵ FFU of 16-06, MpR12, or Wa. Supernatants were collected at the indicated time points (hpi), and virus titers were measured by a focus assay. Means ± SD of triplicate data from a representative experiment are shown in the graph. (B) AUCs based on the viral titers in the culture supernatants of each SMI-100-well infected with 16-06, MpR12, or Wa. Means ± SD of triplicate data from a representative experiment are shown in the graph. Each dot represents one AUC value from each SMI-100 well. Statistical analysis was performed using a one-way ANOVA with Tukey’s test. ***, P < 0.001; **, P < 0.01; *, P < 0.05. (C) Histopathological images of vertical sections of SMI-100 at 3 dpi with H&E staining. Black arrowheads indicate acidophilic dead cells with fragmented nuclei. Scale bars = 100 μm. (D) Vertical sections of SMI-100 at 3 dpi were stained for RVA (green), nuclei (blue), and cell membrane using wheat germ agglutinin (red). Scale bars = 100 μm. Areas in white squares are magnified in lower panels. White arrowheads indicate exfoliated infected cells. Asterisks indicate the mesh membranes which support the epithelium. The figures shown are representative images.