Bats host several zoonotic pathogens. Island biogeography and epidemiologic theory predict small remote islands have lower infection diversity. Molecular studies of urine and feces from three species at 10 sites from three islands suggest multiple pathogenic Leptospira, but not coronavirus, paramyxovirus, or Histoplasma, circulate in isolated Pacific Fijian bat populations.
Bats play important ecologic roles worldwide (Fenton and Simmons 2014) and are hosts of diverse zoonotic pathogens. In Australasia, Oceania, and Southeast Asia, multiple viral families with proven or probable zoonotic potential have been isolated from bat guano (e.g., coronaviruses) and urine (e.g., paramyxoviruses) historically (Mackenzie et al. 2003; Rahman et al. 2010; Drexler et al. 2011). Alpha- and β-coronaviruses share a common ancestry in bats and cause human diseases, including coronavirus disease 2019, severe acute respiratory syndrome, and Middle East respiratory syndrome. Additionally, zoonotic paramyxoviruses have caused severe and fatal disease in both intermediate hosts and humans, including Hendra virus with horses as an intermediate host and Nipah and Menangle viruses with swine as an intermediate host (Philbey et al. 1998; Clayton et al. 2013; Anderson and Marsh 2015). Other pathogens of zoonotic concern recovered from bats and their guano include Leptospira spp. from urine and Histoplasma spp. from guano (Monchy et al. 1998; Lagadec et al. 2012).
There are six bat species in Fiji: Chaerephon bregullae (Fijian free-tailed bat), Notopteris macdonaldi (Fijian blossom bat), Pteropus samoensis (Samoan flying fox), Pteropus tonganus (Pacific flying fox), Mirimiri acrodont (Fijian monkey-faced bat), and Emballonura semicaudata (Pacific sheath-tailed bat; Palmeirim et al. 2007). These species are declining because of anthropogenic activities, including guano mining, hunting and consumption of bats, cave tourism, and human-wildlife conflicts around farming (Palmeirim et al. 2007; Mickleburgh et al. 2009). Because of human encroachment into bat habitats, the risk of human-bat interactions increases. Leptospira is an important zoonotic disease in Fiji, but previous studies regarding risk factors of human leptospirosis in Fiji have not investigated bats (Lau et al. 2016). Paramyxoviruses, coronaviruses, and Leptospira have been isolated from bats in Australia and Southeast Asia. Fiji's geographic isolation and small bat population sizes may limit the ability of infections to be introduced and persist, but given Fiji's proximity to these regions and the known human-bat interactions, our objective was to investigate the presence of several pathogens in Fijian bats. We used molecular techniques to investigate the presence of coronaviruses, paramyxoviruses, Histoplasma, and Leptospira in the urine and feces of Fijian bats.
During August–October 2018, we collected 176 samples (78 pooled urine and 98 pooled fecal samples) by free-catch methods onto plastic drop sheets under 10 arboreal and cave roosting sites (Fig. 1a and Supplementary Material Appendix 1 (jwdi-59-01-24_s01.docx)). Animal ethics were approved by the Fijian government (RA 35/18). Samples were added to 2 mL of RNAlater (Sigma, St. Louis, Missouri, USA) in 5-mL cryovials with sterile swabs and pipettes, stored and transported at 4 C where possible, and taken to the Hopkirk Research Institute (Palmerston North, New Zealand). Nucleic acids were extracted from samples to amplify pathogen markers ( Supplementary Material Appendix 2 (jwdi-59-01-24_s01.docx)). Briefly, nucleic acids were extracted from approximately 200 mg of feces or 200 µL of urine and amplified by reverse transcription PCR for RNA or PCR for DNA (for full details, see Supplementary Material Methods 1 (jwdi-59-01-24_s01.docx)).
To confirm the host species (because samples were collected by free-catch methods), DNA from feces was tested by PCR using LCO1490 and HCO2198 primers designed for amplifying the cytochrome c oxidase subunit I gene (Folmer et al. 1994). All PCR products were checked by gel electrophoresis on 1% agarose gels run at 70 volts for 60 min. Positive control material was synthesized with primer binding sites for the primers used in coronavirus, paramyxovirus, and Histoplasma sp. amplifications (see Supplementary Material Methods 2 (jwdi-59-01-24_s01.docx)). Cloning kits (see Supplementary Material Methods 1 (jwdi-59-01-24_s01.docx)) were used to clone copies of the synthetically generated DNA, which was amplified with previously published primers (see Supplementary Material Table S2 (jwdi-59-01-24_s01.docx)). The Leptospirapositive controls were available from our laboratory Leptospira cultures.
Successful PCR products were cleaned by gel excision and sequenced in both orientations at the Massey Genome Services (Palmerston North, New Zealand) with the ABI 3730 capillary electrophoresis DNA analyzer (Applied Biosystems, Foster City, California, USA). Sequences were assembled in Geneious 10.2.6 (Kearse et al. 2012) by de novo assembly to align reads and generate consensus sequences used for phylogenetic analyses. Representative Leptospira glmU sequences were selected (Wilkinson et al. 2021), downloaded from the National Center for Biotechnology Information (NCBI 2019) nucleotide database, and aligned with our study sequences by MAFFT 7 software (Katoh and Standley 2013). A phylogenetic tree was generated by PhyML 3.0 (Guindon et al. 2010), with parameters determined by smart model selection (Lefort et al. 2017) and branch support by a Shimodaira-Hasegawa–like approximate likelihood ratio test (Anisimova and Gascuel 2006). Tree annotations and edits were produced in iTOL v5.3 (Letunic and Bork 2006).
No samples were positive for coronaviruses (n=98, feces), paramyxoviruses (n=78, urine), or Histoplasma sp. (n=98, feces). We discovered a high prevalence (n=38, 49%, 95% confidence interval 37–60) of 78 urine samples Leptospira sp.–positive, with high genetic diversity (Fig. 1b). All were new sequences with little similarity (88.8–96.8%) with known alleles. All Leptospira sequences have been deposited in GenBank (accessions OL694103–OL694136; Supplementary Material Appendix 5 (jwdi-59-01-24_s01.docx)). Phylogenies showed clustering with Leptospira kirschneri and Leptospira interrogans (Fig. 2).
The Pacific flying fox was detected at five sites (Taveuni arboreal, Viti Levu arboreal 1 and 4, and Vanua Levu arboreal 1 and 2; see Supplementary Material Appendix 1 (jwdi-59-01-24_s01.docx)). A different bat species was detected at Vanua Levu Cave. The closest basic local alignment search tool (BLAST) match for the sequence was to Chaerephon plicatus, at 95.6% similarity, and may represent the Fijian mastiff bat Chaerephon bregullae, which is not represented in the NCBI sequence database. No Leptospira-positive samples were found at Vanua Levu Cave, and no host DNA could be isolated at the remaining sites. All bat cytochrome oxidase c subunit 1 sequences have been deposited in GenBank (accessions OL693647–OL693655; Supplementary Material Appendix 5 (jwdi-59-01-24_s01.docx)).
We found no molecular evidence of coronaviruses, paramyxoviruses, or Histoplasma in the samples tested. Study limitations include relatively small sample sizes, intermittent shedding of the organisms leading to seasonal excretion (Drexler et al. 2011), nucleic acid degradation in tropical conditions, and false negative tests. Zoonotic pathogens have been isolated from Chiroptera species worldwide, including regions neighboring the Pacific, so additional studies, including serology for viral infections, are warranted before it can be concluded that these pathogens are absent from these bat populations. However, geographic isolation and small population sizes may limit the ability for these infections to be introduced and perpetuate if they require critical community sizes larger than these populations.
Leptospira may be able to persist in small populations by chronic infection and shedding (Buhnerkempe et al. 2017). Bacterial isolation with fresh samples will be needed to characterize fully the bacteria we detected by molecular methods. Surveillance for and characterization of Leptospira among Fijian bats, other wildlife, domestic animals, and humans will be useful; genetic characterization of Leptospira human cases should be able to identify whether bats are the source, given the distinct clusters of the Fiji bat Leptospira.
D.T.S.H. was supported by the Royal Society Te Apārangi Rutherford Discovery Fellowship (RDF-MAU1701) and The Percival Carmine Chair in Epidemiology and Public Health. J.M. was supported by the Massey University Foundation, Wildlife Disease Association, and School of Veterinary Science Travel Fund Massey University.
Supplementary material for this article is online at http://dx.doi.org/10.7589/JWD-D-22-00038.
© Wildlife Disease Association 2023