Occurrence of Bartonella genotypes in bats and associated Streblidae flies from Maranhão state, northeastern Brazil

Braz J Vet Parasitol 2020; 29(4): e014420 | https://doi.org/10.1590/S1984-29612020088 This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Occurrence of Bartonella genotypes in bats and associated Streblidae flies from Maranhão state, northeastern Brazil

Bats play an important role in the maintenance of ecosystem stability. Moreover, this mammal group provides important ecosystem services through pollinating flowers, dispersing seeds and consuming insects. Likewise, bats are recognized as reservoirs or carriers for many zoonotic pathogens (Mühldorfer, 2013).
Bartonella spp. have been reported in over 60 bat species worldwide. In addition, the diversification of bartonellae in bats seems to have followed the diversification of bats, with clustering of bartonellae restricted to single bat families (McKee et al., 2017). Moreover, studies have highlighted the role of bats as reservoirs for zoonotic Bartonella species (Veikkolainen et al., 2014;Bai et al., 2018).
Although the prevalence and genetic diversity of Bartonella have previously been assessed in bats and bat flies around the world (Morse et al., 2012;Bai et al., 2015), several biological aspects of this important bacterial group remain poorly assessed in Brazil. So far, Bartonella DNA has been detected in bats sampled in Brazil with prevalence ranging from 5.28% (17/322) to 24.51% (51/208) (Ikeda et al., 2017;Ferreira et al., 2018;André et al., 2019). In addition, Bartonella DNA has been amplified from flies in the family Streblidae (19.8% [40/202]) collected from bats in Brazil (Amaral et al., 2018). Thus, the current study aimed to verify the occurrence and the phylogenetic positioning of Bartonella in bats and associated flies sampled in northeastern Brazil. Additionally, flies collected from bats were molecularly characterized.

Material and Methods
Between September and July 2019, 29 bats belonging to five species were trapped in São Luís Island, state of Maranhão, northeastern Brazil (Table 1)

Pteronotus personatus
Subsequently, EDTA-blood samples were subjected to DNA extraction using the InstaGene TM Matrix (Bio-Rad). Additionally, 15 Streblidae flies were individually subjected to DNA extraction using the lllustra Tissue and Cells Genomic Prep Mini Spin kit (GE Healthcare Life Sciences), in accordance with the manufacturer's instructions. To confirm the presence of amplifiable DNA, the DNA samples obtained from the bat blood samples and flies were initially subjected to conventional PCR assays targeting the endogenous mammals-gapdh and insects-cox-1 (~600 bp) genes, respectively (Birkenheuer et al., 2003;Folmer et al., 1994). Endogenous gene-PCR positive DNA samples were subsequently subjected to a previously described broad-range qPCR assay based on the nuoG Bartonella gene (André et al., 2016). Finally, the positive DNA samples in the above mentioned screening assay were subjected to conventional PCR assays targeting the gltA (750 bp) and rpoB (825 bp) genes, as previously described (Norman et al., 1995;Renesto et al., 2001).
Thereafter, the amplicons obtained, including insects-associated cox-1, were purified using the EXOSAP-IT® system (Applied Biosystems). Purified amplified DNA fragments were subjected to sequence confirmation in an automatic sequencer (ABI Prism 310 Genetic Analyzer; Applied Biosystems/ Perkin Elmer).
The Bartonella species were identified through BLASTn analysis using Megablast (NCBI, 2020). The phylogenetic analysis was performed using the Bayesian inference method, through MrBayes in XSEDE (3.2.7.a) and was performed in the CIPRES Science Gateway.

Results
All the bat-blood and fly DNA samples subjected to PCR assays targeting the endogenous mammals-gapdh and insects-cox-1 genes, respectively, were positive. The BLASTn results from seven out of 13 cox-1 fly sequences (two DNA samples showed weak band intensity, which precluded sequencing) shared identities ranging from 88.94% to 92.91% with Trichobius parasiticus (MH282310; sampled in Costa Rica). The other five cox-1 fly sequences shared identities ranging from 97.28% to 97.79% with Trichobius joblingi (MH282259; sampled in Panama). Lastly, one sequence was 99.04% similar to Trichobius yunkeri (KY882244; sampled in Mexico).
None of the 29 bat-blood DNA samples were positive in qPCR assays for Bartonella spp. targeting the nuoG gene. On the other hand, three out of 15 DNA samples (20%) from Streblidae flies were positive for Bartonella ( Table 1). The amplified sequences shared identities ranging from 97.2% to 100% with sequences previously detected in Carollia perspicillata (MH234352) or bat flies (Trichobius joblingi; KJ816691) from Costa Rica; and in bats (Desmodus rotundus; MK578352) from Brazil (Table 1).
Moreover, and in agreement with phylogenetic analyses, the sequences clustered with other sequences detected in bats and bat flies from the countries mentioned above, and they were supported by high posterior probability values (98%) in the Bayesian inference analysis (Figure 1 and 2). The sequences amplified in the current study were deposited in GenBank under accession numbers MT275628 and MT275629 for the gltA gene and MT275630 for the rpoB gene.

Discussion
Bats and associated ectoparasites have been distinguished as important sources of new Bartonella species/ genotypes (McKee et al., 2017;Sándor et al., 2018). Here, none of the 29 bat blood samples were positive in qPCR assays for Bartonella. In a recent report, André et al. (2019) found high prevalence (24.5% [51/208]) of Bartonella in vampire bat liver samples from 15 different states in Brazil. However, the three bats caught in the state of Maranhão were negative in real-time PCR assays for Bartonella spp. (André et al., 2019), thus corroborating the results found in the present study.
On the other hand, low prevalence (5.28% [17/322]) of bartonellae was previously reported in non-hematophagous bats from Brazil that were sampled in the states of São Paulo, Pará, Tocantins and Mato Grosso (Ikeda et al., 2017). These differences in the prevalence of Bartonella in bats may be attributed to distinct factors, such as sample type (e.g., blood, spleen and heart), bat guilds, assay type (e.g. culturing, qPCR assays or conventional PCR assays), number of samples analyzed and distribution of sampled animals.
However, three flies (20%) that were identified as Trichobius spp. through cox-1 sequencing, were found to be positive for Bartonella. Similar prevalence of Bartonella (19.8% [40/202]) had previously been reported in Streblidae flies sampled in northeastern Nova Iguaçu, Rio de Janeiro, Brazil (Amaral et al., 2018). On the other hand, high prevalence of Bartonella spp. was reported in bat flies from western Africa (66%)  and Costa Rica (up to 100%) (Judson et al., 2015). Additional studies aiming to evaluate the factors affecting Bartonella prevalence in bats and associated bat flies are needed.
The presence of amplifiable DNA in the bat blood samples was strikingly evident and was confirmed through an endogenous control PCR assay targeting the mammals-gapdh gene in all the bat samples analyzed. This excluded the possibility of false negative results due to PCR inhibitors. Figure 1. Phylogenetic tree constructed with 800-bp Bartonella rpoB sequences, using Bayesian method and GTR+G+I evolutionary model. Numbers at nodes correspond to posterior probability over 50%. Brucella abortus sequence was used as outgroup. The sequence amplified in the current study is highlighted in red. The blue color highlighted sequences belonging to the cluster in which the amplified sequences grouped with.

Figure 2.
Phylogenetic tree constructed with 700-bp Bartonella gltA sequences, using Bayesian method and TIM+G+I evolutionary model. Numbers at nodes correspond to posterior probability over 50%. Ochrobactrum anthropi and Brucella abortus sequences were used as outgroup. The sequences amplified in the current study are highlighted in red. The blue color highlighted sequences belonging to the cluster in which the amplified sequences grouped with.
Even though none of the bat-blood DNA samples analyzed were positive for Bartonella, three out of the 15 flies were positive in the screening assay. This result can possibly be partially explained by the higher concentration of parasitized red blood cells in the fly's midgut, compared with the erythrocyte levels in the bats' bloodstream.
Interestingly, the gltA and rpoB genotypes detected in Streblidae flies in the present study were shown to be phylogenetically related to those previously detected in bats and bat flies from Brazil and Costa Rica. This highlights the fact that bats and bat fly-related Bartonella genotypes are widespread in Latin America. Further studies are required to elucidate the role of bat flies in the transmission of Bartonella among bat species.

Conclusions
The present study showed the occurrence of new Bartonella genotypes in Trichobius spp. from northeastern Brazil, despite the absence of bartonellae DNA in the bats that were sampled.