Diversity of phlebotomine sand flies and molecular detection of trypanosomatids in Brumadinho, Minas Gerais, Brazil

This study aimed to describe the sand fly fauna and detect trypanosomatids in these insects from Casa Branca, state of Minas Gerais, Brazil, an endemic area of both visceral (VL) and tegumentary leishmaniasis (TL). Sand flies were collected bimonthly from May 2013 to July 2014, using automatic light traps exposed for three consecutive nights in peridomiciliary areas of nine houses with previous reports of VL and TL. ITS1-PCR and DNA sequencing were performed for trypanosomatids identification. A total of 16,771 sand flies were collected belonging to 23 species. The most abundant species was Nyssomyia whitmani (Antunes & Coutinho, 1939) (70.9%), followed by Lutzomyia longipalpis (Lutz & Neiva, 1912) (15.2%) and Migonemyia migonei (França, 1920) (9.1%). Leishmania amazonensis DNA was detected in Ny. whitmani (four pools) and Le. braziliensis DNA was detected in Psychodopygus lloydi (one pool). In seven pools of Ny. whitmani and in one pool of Lu. longipalpis positive for Leishmania DNA, the parasite species was not determined due to the low quality of the sequences. Moreover, DNA of Herpetomonas spp. was detected in Ny. whitmani (two pools) and Cortelezzii complex (one pool). DNA of Crithidia spp. was detected in Ny. whitmani and Ps. lloydi (both one pool). Our results suggest that Ny. whitmani may be involved in the transmission of Le. amazonensis in the study area. The molecular detection of Le. amazonensis suggests the presence of this species in a sylvatic cycle between vertebrate and invertebrate hosts in the region of Casa Branca. Our data also reveal the occurrence of other non-Leishmania trypanosomatids in sand flies in Casa Branca District.


Introduction
Phlebotomine sand flies (Diptera: Psychodidae: Phlebotominae) are known to be natural hosts of etiological agents such as bacteria, viruses and protozoa [1,2]. Among these agents, are species of the genus Leishmania, which are transmitted during blood feeding by female sand flies [1]. Several sand flies species have been considered proven or suspected vectors of Leishmania parasites in Brazil, suggesting that multiple vectors may be involved in parasite cycles, making the epidemiological picture more complex than currently recognized [3,4].
Intense urbanization over the last decades has caused changes in natural habitats, especially those occupied by phlebotomine sand flies. Some species are able to withstand several environmental modifications that may occur in natural breeding sites and successfully adapt to urban environments [1]. In some regions, the presence of vector species in the human environment may be associated with the emergence of autochthonous cases of leishmaniases, as occurs in the state of Minas Gerais, where both VL and TL are widely spread [32,33]. In this context, the municipality of Brumadinho has reported an increasing number of VL and TL-cases over the last years, and some of them have been reported in the district of Casa Branca (Source: Municipal Secretary of Health of Brumadinho, MG). In this area, the presence of Leishmania spp. in wild and synanthropic mammals have been recorded [34]. Given this situation, this study aimed to describe the sand fly fauna, detect and identify by molecular assay the presence of trypanosomatids in females sand flies from Casa Branca.

Ethic statements
The collections were carried out with the authorization of the Sistema de Autorização e Informação em Biodiversidade-SISBIO (license number 15237-2). The owners of the residences verbally gave permission to the installation of the traps during the study. The sand flies were deposited in the Coleção de Flebotomíneos of the Instituto René Rachou (Fiocruz/COLFLEB).
Study area, collection and identification of phlebotomine sand flies. The study was conducted in the district of Casa Branca (20˚6'2.58"S; 44˚2'59.45" W), northern region of the municipality of Brumadinho (Fig 1). Casa Branca is bordered by the Parque Estadual Serra do Rola Moça (PESRM) conservation unit, the third largest urban park in Brazil, covering a transition area between Atlantic Forest and Cerrado. About eight kilometers west of Casa Branca is a village called Córrego do Feijão, where, in January 2019, an ore tailing dam ruptured, causing great environmental and social impact in all areas of Brumadinho municipality, as well as the death of 270 people.
Sand flies were collected from nine sampling points in Casa Branca from May 2013 to July 2014. Bimonthly, two automatic light traps (model HP) [35] were set at each sampling point, in the peridomiciliary environment for three consecutive nights. Collection sites with previous report of human or canine leishmaniasis were selected. Collected sand flies were taken to the laboratory for further screening by sex, clarification process and dissection. All sand flies collected in July/2013, November/2013, March/2014 and July/2014 were mounted in Canada balsam on a glass slide for species identification. The female sand flies from May/2013, September/2013, January/2014 and May/2014 were dissected by removing the last three segments of the abdomen and head, mounted on a glass slide in Berlese liquid. The remainder of the body was individually stored dry at -20˚C in a 1.5 mL tube for further DNA extraction.

PLOS ONE
Species identification was determined using the updated version of the classification proposed by Galati in 2003 [36]. The females of Evandromyia sallesi and Evandromyia cortelezzii, were identified as belonging to the "Cortelezzii complex" since they are undistinguishable by their morphology [37]. The abbreviation of the genera followed the proposal of Marcondes [38].
DNA extraction and identification of trypanosomatids. Females sand flies without blood in their abdomen were processed individually or grouped into pools of up to 20 individuals of the same species, date and site of collection. Females were subjected to DNA extraction using the Gentra Puregene1 kit (Qiagen, Valencia, CA, USA) following the manufacturer's protocol. To control cross-contamination during DNA extraction, males sand flies were used as negative controls in all procedures [39,40]. In addition, all instruments and the entire work area were previously treated with DNAZap (Ambion, Life Technologies, Inc.).
The extracted DNA was used to investigate the presence of trypanosomatids by amplifying the internal transcribed spacer 1 region (ITS1), following the conditions described previously Step Ladder provided as molecular length size standard. The PCR-positive products were digested using HaeIII enzyme following the conditions previously described [41]. The amplicons were purified with QIAquick1 PCR Purification Kit (Qiagen, Valencia, CA, USA) as described by the manufacturer. Once purified, the products were used for DNA sequencing following the conditions described by Rêgo et al. [43]. Geneious software (v. 9.1.3) was used to check the electropherograms, align sequences and to perform comparison with sequences deposited in the GenBank database through BLAST alignment algorithm, considering the two most similar organisms according to identity above 90% for genus level and 98% for species level. These sequences are available in Gen-Bank (Accession numbers MN638721-MN638737).
The positive rate of each sand fly species was determined considering only one specimen per positive pool.
Statistical analyses. To evaluate species abundance the index of species abundance "ISA", and the standardized index of species abundance "SISA" were calculated [44]. The values of SISA range from 0 to 1, with values closer to 1 representing the most abundant species. The Shannon index (H) and evenness index (J) [45] were used to determine species diversity and evenness of species abundance, respectively.
Evandromyia was the richest genera with eight species collected, followed by Pintomyia with six species (Table 1). May 2013 showed the greatest species abundance (7,054 sand flies) while July 2014 showed the lowest (266 sand flies). The highest species richness was observed in May 2014 with a total of 18 species (Table 1).

Molecular identification of trypanosomatids
Of a total of 4,913 females, 47 were individually tested and 4,866 grouped in 311 pools. Table 2 presents information about the species collected and their distributions for molecular analysis. No sample individually analyzed was PCR positive for trypanosomatids.  Of the 311 pools tested, 13 (4.1%) were PCR-positive for Leishmania: Lu. longipalpis (1), Ny. whitmani (11) and Ps. lloydi (1) ( Table 3). The positive rate obtained for Leishmania was 4.18% (13/311). Four pools of Ny. whitmani were positive for Le. amazonensis (4/311 = 1.2%), however, in other seven pools of this sand fly (7/311 = 2.2%) it was not possible to determinate the Leishmania species, due to low quality of the sequences, which contained a high number of ambiguous sites. Therefore, sequences were left as undetermined (Leishmania sp.). One pool of Ps. lloydi was positive for Le. braziliensis (1/311 = 0.3%) and Lutzomyia longipalpis was positive for Leishmania sp. (1/311 = 0.3%) ( Table 3).
A total of five pools (5/311 = 1.6%) presented single amplicons of inconsistent size (400-500 bp), distinct to that predicted for Leishmania parasites (300-350 bp). These results led us to investigate the identity of these fragments and trypanosomatid species were identified. Nyssomyia whitmani and Cortelezzii complex (both one pool) were positive for Herpetomonas spp., while Psychodopygus lloydi (one pool) and Ny. whitmani (two pools) were positive for Crithidia spp. (Table 3). Sequence similarity to GenBank sequences ranged between 90-97%. The five non-Leishmania sequences found in our study, Herpetomonas spp. and Crithidia spp., are sufficiently different between them and possibly represent three and two, respectively, different MOTUs (Molecular Operational Taxonomic Units).

Discussion
The sand fly diversity found in Casa Branca may be explained by the proximity to the PESRM conservation unit. The residences selected are also close to forest environments and presented breeding sites for phlebotomine sand flies, such as vegetation (fruit trees) and animal shelters (hennery, pigsty and dog kennel). The highest standardized index of species abundance (SISA)

PLOS ONE
Phlebotomine sand flies and trypanosomatids in Minas Gerais in this study was found for Ny. whitmani (0.74). Despite the quite diversity of the study area (H ' = 0.995), the species abundances were not similar, in which Ny. whitmani was a dominant species. This likely explains the low value for the evenness index (J ' = 0.309), since the great abundance of just one species tends to impact the uniformity of the other ones.
Here, Ny. whitmani was found in high density close to residences. This species frequently has been found in endemic areas for TL in southeastern Brazil, and there is no doubt about their role as vector of Le. braziliensis [46]. Furthermore, Ny. whitmani has also been found harboring Le. infantum DNA in the state of Minas Gerais [47,48]. The presence Le. amazonensis DNA within Ny. whitmani has been previously reported in the states of Tocantins and Maranhão [49,50]. Changes in the natural environment may triggered a new epidemiological profile of leishmaniases that could involve Leishmania amazonensis, an alternative vector (i.e. Ny. whitmani) and new reservoirs (domestic and wild). Fonteles et al. [51] reports the capacity of Ny. whitmani to sustain infections with Le. amazonensis, suggesting their role as vector of this parasite in the state of Maranhão. In this context, as we shown here, the same may be occurring in Casa Branca where this sand fly was found harboring Le. amazonensis DNA.
Lutzomyia longipalpis, the main vector of Le. infantum in Brazil, was the second most abundant species in Casa Branca (15.22%) and was also found to be frequent close to the residences. This finding reinforces the adaptation of this species to human dwellings and peridomiciliary sites [52]. Lu. longipalpis is able to sustain infections with several Leishmania species [53][54][55][56] and their role as the main vector of Le. infantum in Latin America is undeniable [2,4]. The finding of Leishmania DNA within Lu. longipalpis from Casa Branca draw attention specially about cases of both human and canine visceral leishmaniasis reports.
Migonemyia migonei was the third most captured species in the study area with 9.14% and is also extremely anthropophilic species and frequently found inhabiting peridomiciliary sites

PLOS ONE
Phlebotomine sand flies and trypanosomatids in Minas Gerais and feeding in domestic animals [57]. Recent studies have added information about the vector capacity of Mg. migonei. Guimarães et al. have reported their capacity to support Le. infantum development and late-stage infection [58] and moreover, natural infections by Le. infantum have been reported especially in the absence of the main vector, Lu. longipalpis [59][60][61]. Migonemyia migonei has also been implicated in the cycle of Le. braziliensis [62][63][64] and their capacity to support infections with this parasite have been confirmed by Alexandre et al [65]. The genus Evandromyia presented the highest number of species and although there are no proven vector species within this genus, natural infection or Leishmania DNA has been detected in some species [66,67]. In studies conducted in Minas Gerais, Carvalho et al. [68] detected Le. infantum DNA in Ev. cortelezzii and Saraiva et al. [69] found females of Ev. sallesi naturally infected by the same species of Leishmania. These findings may indicate that these species of the Cortelezzii complex (females of Ev. cortelezzii and Ev. sallesi do not differ morphologically) hold some epidemiological significance.
The species Ps. lloydi has been associated with the transmission of Le. braziliensis in wild areas of Minas Gerais [70,71] and was also found in this state with DNA of Le. infantum [72]. Psychodopygus lloydi was positive for DNA of Leishmania braziliensis, suggesting that this sand fly could maintain a wild cycle of TL in the District of Casa Branca.
The finding of Herpetomonas and Crithidia DNA within sand fly species from Casa Branca need to be investigated. Possibly, more than one species within each genera were detected, however, due to the small length of the amplicon it was not possible to confirm it. On the other hand, the molecular detections presented here, may represented different populations (different MOTUs), reinforcing the importance of further studies on these trypanosomatids. Although Herpetomonas seems not to be of epidemiological importance, it has been reported in Egyptian rat [73] and even in immunodepressed humans [18]. In their more common hosts, the Diptera, promastigotes live in the digestive tract, preferentially in the rectum, attached either to endothelian cells or as free-swimmers among fewer opisthomastigotes [74]. Crithidia-related parasites have been involved in an atypical manifestation similar to VL in Brazil [19]. The report of Crithidia causing clinical manifestations similar to VL and the detection of this parasites within sand flies from Brazil, as we found here and previously reported by Machado et al. [49] draw attention and need to be investigated.
Although in some positive pools were not possible to determinate the Leishmania species, due to the low quality of the sequences, it is noteworthy that there was no information about the sand fly fauna and putative vectors in Casa Branca. Here we provide consistent data that should be used for further investigations, such as trying to isolate Le. amazonensis from Ny. whitmani to confirm their role as a vector. Moreover, it is extremely important to investigate the presence of trypanosomatids in Casa Branca and their association with the sand fly fauna.

Conclusion
This is the first study of a sand fly fauna in the municipality of Brumadinho, Minas Gerais. Our findings show that the Casa Branca locality has a diverse sand fly fauna with species that have been previously reported in the state of Minas Gerais. The Ny. whitmani species is the probable vector of Le. braziliensis in Casa Branca and may also be involved in the transmission of Le. amazonensis. The knowledge of the interactions between sand flies and trypanosomatids reported in this study shows that the infection may be occurring in the peridomiciliary environment in the study area. In addition, these results help to understand the dynamics of the leishmaniases transmission cycle in Casa Branca providing support for disease control actions in the region. The presence of non-Leishmania trypanosomatids raises an issue that has been neglected and is of great importance, the circulation of these parasites within phlebotomine sand flies.