Sequences of Previously Unknown Rhabdoviruses Detected in Bat Samples from the Republic of the Congo

Introduction

The negative sense RNA virus family Rhabdoviridae (order: Mononegavirales) currently contains 191 recognized species in 30 genera, with both the number of species and genera continuously increasing. The species from five of the known genera are vertebrate host specific (Lyssavirus, Novirhabdovirus, Perhabdovirus, Sprivivirus, and Tupavirus), whereas the species within six other genera infect only plants (Cytorhabdovirus, Dichorhavirus, Alphanucleorhabdovirus, Betanucleorhabdovirus, Gammanucleorhabdovirus, and Varicosavirus). Four genera are considered to (only) contain viruses of invertebrate hosts such as insects or nematodes (Almendravirus, Alphanemrhavirus, Caligrhavirus, and Sigmavirus); however, the remaining 15 genera contain rhabdoviruses that are known to or may be suspected to be vertebrate viruses with arthropod vectors (Arurhavirus, Barhavirus, Curiovirus, Ephemerovirus, Hapavirus, Ledantevirus, Lostrhavirus, Mousrhavirus, Ohlsrhavirus, Sawgrhavirus, Sripuvirus, Sunrhavirus, Tibrovirus, Vesiculovirus, and Zarhavirus) (Walker et al. 2018, Kuhn et al. 2020).

Although the biology of many of the rhabdoviruses discovered to date remains largely unknown, some of them, such as rabies virus and other lyssaviruses are well known as pathogens causing fatal disease in humans and other animals (Shope 1982, Banyard et al. 2014, Hayman et al. 2016). One reservoir and known source of zoonotic infections with rhabdoviruses are bats, which are known to harbor and shed a rich set of not only rhabdoviruses, but viruses in general, often without themselves displaying any overt clinical signs of infection (Murphy et al. 1970, Iwasaki et al. 2004, Ghedin et al. 2013, Luis et al. 2013, Blasdell et al. 2014, Binger et al. 2015). Considering the proposed role of bats as reservoirs, we undertook testing of bats in the highly biodiverse Congo Basin for the presence of rhabdovirus RNA to further asses rhabdovirus diversity.

Materials and Methods

Bat “capture and release” specimen collection was conducted by mist-netting at heights of 1–3 meters along trails and streams in lowland forest. Nonlethal sample collection for the Emerging Pandemic Threats (EPT) program's PREDICT project was approved by the Institutional Animal Care and Use Committee (IACUC) of the University of California Davis (UC Davis) and the government of the Republic of the Congo (permit number 018/MRSIT/DGRST/DMAST issued by the Republic of the Congo Ministry of Scientific Research and Technical Innovation). Some samples analyzed in this study were collected as part of a faunal inventory, commissioned in association with a proposed iron ore mine feasibility study (Bates et al. 2013). As part of that study, a limited number of bats (n = 17) were sacrificed as voucher specimens and samples were also collected for the PREDICT project. Bat species were identified based on field guides and verified by Paul Bates of the Harrison Zoological Museum, Sevenoaks, Kent, United Kingdom (Kingdon 2005, Monadjem et al. 2010).

Oral and rectal swabs were collected from live and voucher animals using sterile micro-tipped polyester swabs (Puritan™ Medical Products Company, LLC), placed in 1.5 mL NucliSens^® lysis buffer (bio-Mérieux, Inc.) in 2.0-mL cryotubes. Tissue samples from voucher animals were collected in 2.0-mL cryotubes without medium. All were stored at ambient temperature for up to 3 h before being frozen in liquid nitrogen and later transferred to −80°C until further processing.

RNA was extracted either manually using Trizol^®, or using a Qiagen Viral RNA Mini Kit and stored at −80°C. Afterward RNA was reverse transcribed into cDNA using a Maxima H Minus First Strand cDNA Synthesis Kit (Thermo Scientific) and stored at −20°C until analysis. A semi-nested consensus touchdown PCR, amplifying ∼260 nucleotides of the conserved DNA polymerase (L) gene, was conducted (Supplement Data S1). PCR products were examined by gel electrophoresis and products excised, cloned, and Sanger sequenced at the UC Davis DNA sequencing facility. Obtained sequences were edited in Geneious and compared with the GenBank database (BLAST).

A phylogenetic tree was constructed based on the amino acid sequences of the DNA polymerase gene, including the PCR amplicons and 48 published rhabdovirus L-gene sequences; cytorhabdoviruses served as outgroup. First multiple sequence alignments were made in Geneious (Muscle Alignment), Bayesian phylogeny was inferred using MRBAYES (version 3.2), with default parameters and four chains of 1,000,000 generations, and tree sampling every 1000 steps during the process to monitor phylogenetic convergence. The final average standard deviation of split frequencies was <0.03. The first 10% of the trees were discarded, and the remaining trees combined using TreeAnnotator (version 2.5.1) and displayed with FIGTREE (1.4.4).

Results

Samples were collected between July 2012 and March 2014 in the Lékoumou and Sangha departments of the Republic of the Congo and included 259 samples from 193 bats belonging to at least 23 different species (Supplementary Data S2 and S3). The collected samples were primarily rectal (140) and oral swabs (93), but also included lung (20), brain (4), liver (1), and feces (1). The female/male ratio was 111/78, with four individuals undetermined. Rhabdovirus RNA was detected in the oral swab samples of four adult animals, accounting for an overall 2.1% apparent prevalence of rhabdovirus RNA in the bats sampled. All rhabdovirus RNA positive samples were collected within 1 week in the same area during the dry season (Supplement Data S2 and S3). Two of these bats were Hipposideros ruber, one a Triaenops afer and one a Neoromicia brunnea. All four obtained sequences were different from each other, sharing no more the 67% nucleotide identities. Closest matches upon protein BLAST shared between 70% and 80% identities with mostly other isolates of unclassified rhabdoviruses isolated from bats. The closest classified matches belonged to Vesicolovirus (ZB12028, ZB12058, and ZB12064) or Perhabdovirus (ZB12070) genera. Upon phylogenetic analysis, the detected sequences largely branch off at the base of classified genera branches, two at the base of the Caligrhavirus branch, and one at the base of the Sigmavirus branch, whereas one showed no close connection with any established genera (Fig. 1). Novel sequences are stored in GenBank with accession numbers MG956843–MG956846 (Supplement Data S4).

OPEN IN VIEWER

FIG. 1.

Maximum likelihood phylogenetic tree of rhabdoviruses based on the L gene. The four novel sequences (dark boxes), previously known sequences of 4 other bat (light gray boxes) and 44 non-bat rhabdoviruses are indicated. Virus name and GenBank accession numbers are shown and, genus and host association according to the ICTV indicated. Bootstrap support indicated at nodes.

Discussion

We have detected RNA indicative of four rhabdoviruses in samples collected from bats native to the Congo Basin. None of them matches any available sequence from the 191 previously described rhabdovirus species. Full genome sequencing was not possible, which is a limitation of the study. Given the brevity of the RNA sequences amplified definite phylogenetic and taxonomic placement was, therefore, not possible; however, comparison with the GenBank database and the clustering upon phylogenetic analysis suggest that all of the respective viruses may represent new species. Considering the apparent basal or unclear placement in the phylogenetic tree some or all of them might potentially not be associated with any of the 30 established rhabdovirus genera.

Since all sequences were amplified from oral swabs from insectivorous bats, it is possible that some or all of the detected sequences may either belong to viruses with arthropod vectors or viruses primarily infecting insects. However, virus containing saliva of bats and other animals is a known source of zoonotic lyssavirus transmission. Considering the protein similarity to other rhabdovirus sequences isolated from bats, and similarities to other better known rhabdoviruses, we believe that the detected sequences may belong to vector- or nonvector-borne vertebrate rhabdoviruses.

Although we do not have any information with regard to the biology or possible zoonotic potential of the viruses of which we detected RNA, it is known that not only lyssaviruses but also other types of rhabdoviruses can or are suspected to cause human disease (Gurav et al. 2010, Grard et al. 2012). We consequently believe that further research into, and surveillance of rhabdoviruses, particularly in bats, is warranted to better understand rhabdovirus evolution and to evaluate and mitigate potential risks bat rhabdoviruses (other than lyssaviruses) may pose.