Molecular characterization of Trypanosoma cruzi samples derived from Triatoma vitticeps and Panstrongylus geniculatus of the Atlantic rainforest, southeast Brazil

Background: In rural areas of Espírito Santo state, southeast Brazil, triatomine species attracted by light frequently invade residences. The aim of this study was to investigate the Trypanosoma cruzi discrete typing units (DTUs) harbored by these triatomines. Methods: Triatomine’s intestinal contents were examined, inoculated in mice, and the positive samples were cultivated. Flagellates obtained from infected mice hemoculture were submitted to DNA extraction using a salting-out method and to TcSC5D gene amplification. The amplified samples were sequenced, and polymorphism was analyzed for DTU identification. Results: Three hundred and ninety-four triatomines were identified: Triatoma vitticeps (90.03%), Panstrongylus geniculatus (8.89%), Panstrongylus megistus (0.54%), Panstrongylus diasi (0.27%), and Triatoma tibiamaculata (0.27%). Among the specimens, 251/394 (67.65%) presented flagellated forms similar to T. cruzi. After triatomine intestinal content inoculation into mice, 134 mice presented T. cruzi-like trypomastigotes from Tr. vitticeps and P. geniculatus and 89 samples were positive in hemoculture. Sixty-two samples were analyzed for the TcSC5D gene and TcI, TcII, TcIII, and TcIV DTUs were identified. Conclusions: We observed T. cruzi DTU diversity in Tr. vitticeps and P. geniculatus, which showed the predominance of TcII and occurrence of TcI, TcIII and TcIV. Triatomines presented high T. cruzi infection rates. Since little is known regarding the possible mammalian hosts that maintain the T. cruzi cycle, further studies are necessary to obtain a better understanding of the parasite transmission cycle in this region.


Introduction
Trypanosoma cruzi is transmitted by blood sucking insects termed triatomines. Triatomines belong to the subfamily Triatominae, of which over 130 species are considered as potential T. cruzi vectors [21,31]. In Brazil, 52 species and three genera: Triatoma, Panstrongylus, and Rhodnius, are registered as causing parasite transmission. Triatomine distribution is related to the overlap of their habitat with human colonization in areas with natural Brazilian Cerrado and Caatinga biomes, whereas Atlantic rainforest fragment areas generally do not present foci of triatomine domiciliation [20].
According to the Instituto de Pesquisas da Mata Atlântica [27], during the last century, the Espírito Santo state (ES) Atlantic rainforest fragment was devastated by wood exploitation and agricultural activities, with only 8.15% of the natural vegetal cover still remaining. Seven triatomine species are currently reported to occur in this region:  (Pinto, 1926), and Triatoma vitticeps (Stal, 1859). Although these species present sylvatic habits [21,56], adults are captured by human dwellers inside residences in rural areas, mostly from the ES state mountainous region [30], and Tr. vitticeps have been found colonizing human dwellings [57,58]. In this context, Tr. vitticeps constitutes the main captured species, representing high T. cruzi natural infection rates [17,56,60].
Serological inquiry in ES state showed a low prevalence (0.019%) of Chagas disease (CD) [4,61], consistent with the ineffectiveness of Tr. vitticeps to act as a vector through the contaminative route, as adult specimens eliminate their excreta late after their blood meal [57]. Nevertheless, as adult specimens are constantly invading residences, the potential contact with humans includes the risk of parasite transmission by other routes, such as oral transmission. No studies have yet been performed regarding the T. cruzi population circulating in triatomine species from ES state, reinforcing the idea that little information is available related to the T. cruzi populations in non-endemic areas [64]. Accordingly, the aim of this study was to investigate the T. cruzi populations that are present in the triatomine species that invaded residences in rural areas of ES state, which will provide a better understanding of the transmission dynamics of the parasite in the area.

Materials and methods
Study area ES state is located along the southeastern coast of Brazil, encompassing 78 municipalities in a 46.095 km 2 area. The municipalities are grouped into four macro regions (Metropolitana, Norte, Central, and South) and ten micro regions (Metropolitana, Central Serrana, Sudoeste Serrana, Litoral Sul, Central Sul, Caparaó, Rio Doce, Centro-Oeste, Nordeste, and Noroeste). The state covers the Atlantic rainforest central corridor, which comprises one of the main dense forest areas inside this biome [30].

Triatomine capture
ES health agents were trained and instructed to visit rural area residents, teaching them to recognize, capture and preserve the triatomines that invaded their residences, as well as to handle the insects in an approved manner to avoid accidents. Upon capture of a triatomine by an individual in their residence, the insect was taken to a health center nearby or the municipality health agents were contacted to retrieve the triatomines from the residence.
The triatomine species from different localities within ES state collected by residents were delivered by municipality health agents to Núcleo de Entomologia e Malacologia from Secretaria de Estado da Saúde do Espírito Santo (NEMES -SESA/ES), located at Departamento de Patologia, Centro Ciências da Saúde, Universidade Federal do Espírito Santo (CCS/UFES), between June 2010 and May 2012. The collection of all the triatomines received was catalogued.

Triatomine identification, examination, and intestinal content inoculation in mice
Triatomine identification was performed according to Lent and Wygodzinsky [31]. The insect digestive tract was dissected using forceps and scissors. The intestinal contents were diluted in saline solution (0.85%) and examined using an optical microscope to examine flagellate forms similar to T. cruzi. Upon positive results, the intestinal content was inoculated into Swiss albino mice by the intraperitoneal route. To confirm parasitemia by T. cruzi, the mice were examined every 15 days to observe the presence of trypomastigote bloodstream forms. The animals were examined up to the 60th day. In case of positivity in blood or negativity after the 60-day period, the animals were sacrificed using methods recommended by the Universidade Federal do Espírito Santo Animal Ethics Committee (pentobarbital sodium 40 mg/kg; intraperitoneal route).

Hemoculture examination
Blood from positive mice was collected by cardiac puncture under anesthesia (acepromazine (2%) in 9:1 ketamine hydrochloride (10%)) for subsequent hemoculture in MacNeal, Novy, and Nicole (NNN) medium with Brain Heart Infusion (BHI) overlay, and incubated at 28°C. The hemocultures were examined under an optical microscope every 15 days until the 60th day. Subsequently, if the hemocultures did not present T. cruzi epimastigote forms, they were discarded. Positive hemocultures were washed with 1 mL of phosphate buffer solution (pH 7.2) and centrifuged at 448 · g for 10 min. The supernatant was discarded, and the pellet was stored at À20°C until DNA extraction.

Trypanosoma cruzi molecular characterization
Trypanosoma cruzi DNA extraction was performed using a salting out method [2]. For T. cruzi genotyping, the DNA was submitted to polymerase chain reaction (PCR) for the TcSC5D marker using the following primers: TcSC5Dfwd 5 0 -GGACGTGGCGTTTGATTTAT-3 0 and TcSC5D-rev 5 0 -TCCCATCTTCTTCGTTGACT-3 0 [12], as previously described. In all PCR reactions, we used positive controls and water as a negative control, to validate the reaction.

Ethics statement
This study was approved by the Ethical Animal Use Committee of the Federal University of Espírito Santo (CEUA/UFES), protocol number: 016/2010. DNA extraction was performed for 89 cultures in which the presence of T. cruzi was evident by microscopy, and PCR amplification products for the TcSC5D gene were obtained for 62 samples. These were genotyped as representing four T. cruzi DTUs circulating in Tr. vitticeps (58 samples) and P. geniculatus (four samples). Of these, 50 samples (80.65%) were genotyped as DTU TcII, five (8.06%) as TcIII, five (8.06%) as TcIV, and two (3.22%) as TcI (Tables 2 and 3). This result indicates that in ES state, triatomines demonstrate high diversity of T. cruzi DTUs circulating in the sylvatic environment (Fig. 1).

Discussion
Although only three cases of autochthonous CD have been reported in ES state since 2007, according to the Espírito Santo State Health Department (SESA/ES), triatomines frequently invade residences in rural areas of ES state resulting in the dwelling inhabitants being in contact with this parasite. However, only studies addressing T. cruzi infection in triatomines have been mostly reported to date. In contrast, the current report comprises a recent study demonstrating that together with the high rates of T. cruzi infection, triatomines represent a source of considerable T. cruzi genotype diversity circulating in a given biome.
In this study, we identified five different triatomine species that occur in ES state: Tr. vitticeps, P. geniculatus, P. megistus, P. diasi, and Tr. tibiamaculata, representing the same species that were found by Leite et al. [30] during an earlier study regarding triatomines in ES state. The occurrence of C. pilosa and R. domesticus species has been described previously [21,25], but these species were not identified in the current study. Although Tr. vitticeps is considered a sylvatic species, it is a secondary vector of CD transmission by the vectorial contaminative route, and has been found frequently in the intradomiciliar environment. This triatomine species exhibits a high T. cruzi infection rate, which has previously been reported in ES as well as in Rio de Janeiro and Minas Gerais states [22,55,58,59]. In the present study, among the 67.65% of triatomines infected with T. cruzi, 64.96% were from Tr. vitticeps. This high T. cruzi infection rate demonstrates a relationship between this triatomine and sylvatic mammals, indicating the capacity of this species to maintain the sylvatic cycle in ES state [57,58]. Panstrongylus geniculatus and P. megistus were also found to be infected with T. cruzi, although with a T. cruzi infection rate lower than that of Tr. vitticeps. Panstrongylus geniculatus is widely distributed in Latin America [29]. This species is correlated with the enzootic cycle and the transmission of CD in Venezuela and Colombia [9,11,18,23]. In Brazil, the species is found in Bahia and in the Amazon region [19,51,54,67,68]. Notably, natural T. cruzi infection by P. geniculatus has previously been identified only in the Amazon region [51]; the current finding of 33 P. geniculatus specimens from ES state is the first report in the Atlantic rainforest region. Panstrongylus megistus is one of the most important triatomine species domiciliated in Cerrado and Caatinga biomes and can be found from northeast to south Brazil as well as in Bolivia, Paraguay, Uruguay, and Argentina [47,63]. A study performed by Paula et al. [48] in Minas Gerais state demonstrated that P. megistus exhibited a T. cruzi infection rate of 8.3% and that from among a total of 147 specimens identified in Brasilia Federal District, only three (1.45%) presented . .
The dots represent the same base position. Trypanosoma cruzi marinkellei (KC881179) was used as the outgroup.
flagellate forms [36]. These results reinforce the low T. cruzi infection rate exhibited by P. megistus, which can be justified by its correlation with chicken houses in the peridomicile [5,48,69]. However, although P. megistus has been shown to represents an effective CD vector and to be present in domiciles in Brazilian Cerrado and Caatinga biomes, in the Atlantic rainforest it is predominantly sylvatic, as was also observed by Barbosa et al. [3] in Santa Catarina state. Trypanosoma cruzi is a genetically diverse parasite: six T. cruzi populations are recognized, denoted as TcI to TcVI, as well as a seventh DTU termed TcBat [38,71,72]. Notably, in Brazil, the existence of a heterogeneous T. cruzi population supporting the occurrence of the six T. cruzi DTUs has been reported [6,14,43,65]. In the present study, the T. cruzi genotypes circulating in triatomines from the Atlantic rainforest were found to be genetically heterogeneous, with the predominance of DTU TcII and DTUs TcI, TcIII, and TcIV were also presented. This DTU diversity was observed in Guarapari municipality, ES state, in which the DTUs TcI, TcII, TcIII and TcIV were identified infecting triatomines [15,16].
DTU TcII is found in the southern cone of South America [72], although its occurrence has been reported in the Amazon region, Colombia, Mexico, and the United States [26,32,37,50]. According to Miles et al. [44], TcII is rarely found in the sylvatic cycle; however, in the Atlantic rainforest, it has been reported in primates and has been isolated from the armadillo (Euphractus sexcinctus) as well [33,34,70]. Notably, whereas all CD vectors in ES state are in sylvatic areas, the occurrence of DTU TcII was predominant in this environment. As T. cruzi is known to circulate in more than 100 mammal species [28], this genotype likely already existed in this environment prior to triatomine domiciliation. DTU TcI has a wide distribution on the American continent and is responsible for CD cases in Central and Northern South America. It is correlated to the arboreal transmission cycle in nature and opossums from Didelphis sp. comprise the main reservoir [9,72,42,7,68,24]. Although TcI is very commonly found in nature, in ES state only two samples were identified in Tr. vitticeps, which contrasts with a previous report of infection of this triatomine by DTU TcI at a high percentage [53].
In turn, DTU TcIII can be found in different regions in Brazil; this genotype has been found to be responsible for cases of CD in North, Northeast, Southeast, and South Brazilian regions [1,35,40,45]. It is correlated to the terrestrial transmission cycle, with its main vector being P. geniculatus species [39]. In the present study, we described the occurrence of DTU TcIII in P. geniculatus and in Tr. vitticeps. In comparison, according to Cardinal et al. [8] and Martins et al. [41], TcIII was described as infecting triatomines from T. rubrovaria and T. infestans, respectively. The demonstration of the infection of additional triatomine species by this genotype reinforces the notion that the correlations between single species and genotypes remain poorly understood. Finally, DTU TcIV is encountered from the southern United States to northern South America and, together with TcI, is responsible for CD cases in the Amazon region [10,24,46,50,52]. The reporting of TcIV in Tr. vitticeps in ES state shows that its distribution is more extensive than has been reported. This finding also reinforces the Amazon enclave theory proposed by Costa [13], which states that a thousand years ago, the Atlantic rainforest and the Amazon were connected. In support of this theory, Pinto et al. [49] and Silva et al. [62] reported the appearance of distinct species of sandflies and Anopheles mosquitoes from the Amazon in the ES Atlantic rainforest. Our report of T. cruzi populations that occurred both in the Amazon and in the Atlantic rainforest thus serves to further support this theory.
This study is the first report of T. cruzi DTUs in different municipalities from ES state, as prior studies were limited to investigating only T. cruzi infection rates in triatomines. Our results demonstrate that T. cruzi DTU distribution is much broader than has previously been reported; for example, TcII, which was associated with the domestic transmission cycle, was found to be widely distributed in nature. Furthermore, it was observed that the small Atlantic rainforest fragment studied herein presented marked diversity of circulating DTUs and that one triatomine species, Tr. vitticeps, hosted four different T. cruzi genotypes. For future studies, it is necessary to understand the ecological scenario that maintains DTU diversity in this biome.