Characteristics of Triatomine infestation and natural Trypanosoma cruzi infection in the State of Rio Grande do Norte , Brazil

Introduction: Natural and artifi cial ecotope infestation by the kissing bug triatomines and their colonization and infection by Trypanosoma cruzi, the Chagas disease agent, were evaluated in nine municipalities of the State of Rio Grande do Norte, Brazil. Methods: Following identifi cation, triatomine intestinal contents were analyzed by direct microscopic examination, xenoculture, and polymerase chain reaction (PCR) for parasite detection. Trypanosoma cruzi isolates were genotyped using three different markers. Results: Of 842 triatomines captured, 65% were Triatoma brasiliensis, 17.8% Triatoma pseudomaculata, 12.5% Panstrongylus lutzi, and 4.7% Rhodnius nasutus. Triatoma brasiliensis and P. lutzi adults were found in the intradomicile. T. brasiliensis, T. pseudomaculata, and R. nasutus nymphs and adults were found in the peridomicile and wild environment. Intradomiciliary and peridomiciliary infestation indexes were 5.6% and 33.7%, respectively. In the peridomicile, chicken coops were the most infested ecotope. The T. cruzi triatomine infection rate was 30.2%, of which PCR detected 29%. P. lutzi (78.1%), T. brasiliensis (24.5%), and T. pseudomaculata (22.7%) were the most infected species. TcII and III genotypes were detected in T. brasiliensis and TcIII in P. lutzi. Conclusions: T. brasiliensis was found in all environments and most ecotopes with high T. cruzi infection rates. High infection rates were also detected in T. pseudomaculata and P. lutzi, suggesting their role in the interchange between the wild and peridomestic transmission cycles. The combination of PCR, microscopic examination, and xenoculture contributed to improving T. cruzi infection evaluation in triatomine bugs. The TcII and TcIII genotypes were predominant in the study area.

Trypanosoma cruzi is the etiological agent of Chagas disease, and its main vectors belong to the genera Panstrongylus, Rhodnius, and Triatoma.Among these blood-sucking reduviid bugs of the subfamily Triatominae, 70 of the over 148 Triatominae species described (1) are naturally infected by T. cruzi (2) .Infection is maintained primarily within three overlapping cycles: domiciliary, peridomiciliary, and sylvatic (3) .Vector transmission remains the most important route of parasite to human transmission owing to the natural distribution of T. cruzi in the triatomine species adapted to domestic or peridomestic environments; this adaptation to human dwellings strongly determines human infection rates (4) .
The Brazilian Northeast, one of the poorest and most underdeveloped regions within Brazil, is considered the most important region therein for American trypanosomiasis, where native species like Triatoma brasiliensis Neiva, 1911, Triatoma pseudomaculata Corrêa and Espínola, 1964, Panstrongylus lutzi Neiva & Pinto, 1926, and Rhodnius nasutus Stal, 1859 are widespread (5) (6) .Triatoma brasiliensis is distributed in nine Northeast states, Tocantins, and Minas Gerais (7) .Considered the main vector of T. cruzi in the Northeastern semiarid regions, it colonizes both sylvatic and domestic environments but is more frequently captured in peridomestic areas (5) (8) (9) (10) (11) .Triatomine control remains problematic as this region is the center of dispersion and has higher T. brasiliensis concentrations (12) (13) (14) ; it is further aggravated by local vector control activity discontinuation and wild and domestic environmental overlap (15) (16) (17) .Triatoma pseudomaculata also demonstrates high domiciliation capacity,

METHODS
thereby increasing its colonization rate in several states (5) (18) ; such native triatomines that sporadically invade or reinvade human dwellings further complicate vector control consolidation (18) .
Triatomines can be infected with several flagellates and T. cruzi infection level evaluation in wild, peridomiciliary, and domiciliary environments has relevance for control programs aimed at reducing human infections (19) (20) .Trypanosoma cruzi has been detected in the excreta or intestinal contents of triatomines by direct microscopic examination (DME) (19) (21) , necessitating the ability to distinguish this protozoan from other trypanosomatids.Although DME is reliable and relatively inexpensive, disadvantages related to sensitivity, specifi city (22) (23) (24) , and poor performance on dead insects (25) exist.Xenoculture is used for quality control to confi rm negative intestinal content results ascertained by DME, whereas positive cultures allow for T. cruzi strain isolation (26) pursuant to e.g.genetic studies.For detecting T. cruzi in reduviid bug feces or urine and mammalian blood samples, polymerase chain reaction (PCR) is generally superior (21) (22) (27) and can also genotype T. cruzi using different targets.
Such methodology has increased the rate of positivity of infection in fi eld studies, which is especially important in areas where there is scarce information on vector infection following control and surveillance programs.A recent seroepidemiological survey showed high seroprevalence for municipalities in the west and central mesoregions of the State of Rio Grande do Norte (RN) (28) , and genotyping studies identified T. cruzi I (TcI) in T. brasiliensis and TcIII (formerly called TcIIc) in armadillos (29) , TcI and TcII in humans, TcII and TcIII in T. brasiliensis, and TcIII in P. lutzi (15) .
The purpose of this study was to evaluate the occurrence of triatomine infestation of natural and artifi cial ecotopes, and the colonization and T. cruzi infection in triatomines collected from different environments in the west and central mesoregions of the State RN, Brazil.Diagnostic method efficacy and reproducibility were evaluated and the T. cruzi populations isolated from positive triatomines were genotyped to establish their genetic groups.

Study area
State of Rio Grande do Norte, located in northeastern Brazil, is divided into 167 municipalities distributed in four mesoregions: West, Central, Agreste, and East.About 90% of the territory represents arid and semiarid climates, where the predominant biome is the Caatinga, characterized as containing thorny shrubs, irregular structure, and partially uncovered soil.This study was conducted in the municipalities of Apodi, Caraúbas, Governador Dix-Sept Rosado, Lucrécia, Mossoró, Severiano Melo, São Miguel, Caicó, and Serra Negra do Norte in the West and Central mesoregions (Figure 1), which were selected in reference to the seroepidemiological survey conducted between 2007 and 2009 (28) .

Triatomine collection and identifi cation
Triatomines were captured in rural areas of the nine municipalities in intradomicile, peridomicile, and wild environments from March 2009 to August 2012.The intradomiciliary environment was surveyed in 250 domiciliary units (DU) together with 187 artificial structures in the peridomicile such as chicken coops (n = 110), corrals (n = 25), pigsties (n = 2), piles of tiles (n = 44), old stone fence structures (n = 2), and dry carnauba palm tree (Copernicia prunifera) straw (n = 4).Entomological indicators were used to calculate the colonization and species peridomiciliary and domiciliary infestation indices (30) .
In Serra Negra do Norte, wild triatomines capture was conducted at the Seridó Ecological Station (ESEC-Seridó)/ Instituto Chico Mendes de Conservação da Biodiversidade (ICMBio), created by Decree 87222 of 05/31/1982, Law 6902 04.27.1982 as a Conservation Unit.Some adult insects were captured in station lodgings and were classifi ed as an undefi ned ecotope.Insects were captured both day and night by the principal author with the assistance of the respective Municipal Health Secretariat technicians via manual searches using tweezers and a fl ashlight in all environments and without the use of insect dislodging substances, and were individually identifi ed as described (32) .

Natural infection of triatomine bugs
Direct microscopic examination.Collected triatomines were examined individually and their intestinal content was removed under aseptic conditions and placed in a well of a 24-well plate containing 500µL sterile saline solution.After homogenization, 5µL suspension was used to identify trypanosomatid forms via DME at 400× magnifi cation (26) and smear stained by Giemsa (1,000×).
PCR with species-specific primers.Deoxyribonucleic acid (DNA) was extracted via phenol-chloroform (34) using 200µL diluted insect intestinal content solution (v/v) in 0.2M guanidine-HCl 6M/ethylenediaminetetraacetic acid (EDTA) (35) maintained for 5-7 days at room temperature and stored at 4ºC until DNA extraction.PCR amplifi cations were performed in duplicate, as described (34) , using specifi c primers to identify T. cruzi (36) .PCR assays were performed in a DNA clean chamber to avoid contamination and positive and negative controls were used to monitor each step.as described (37) .As positive controls, Trypanosoma cruzi cell cultures of the Colombian strains (T.cruzi I), JG (T.cruzi II), and T. rangeli were used.
All PCR products were analyzed by 6% polyacrylamide gel electrophoresis and visualized by silver staining (48) .

Statistical analysis
To quantify the concordance between the results of different methods, generalized Kappa (hat k) coeffi cients were estimated and an approximate 95% confi dence interval (CI 95% ) for k [lower confi dence limit (LCL) and upper confi dence limit (UCL)] (49) was calculated and classifi ed accordingly (50) .
Intestinal contents of all infected triatomines and 100 uninfected specimens were submitted to multiplex PCR using primers for T. rangeli and produced no overlapping data, thus confi rming that T. cruzi-specifi c amplifi cation.

DISCUSSION
This general assessment of triatomine occurrence demonstrated that the species T. brasiliensis, T. pseudomaculata, P. lutzi, and R. nasutus continue to exist in artifi cial environments in RN State municipalities, where they have been registered since the 1950s (7) (8) (14) (15) (51) (52) and where high seroprevalence of human T. cruzi infection has been estimated (28) .The presence of other triatomine species has also been reported (5) (7) (8) (52) , but these were not found in this study.
Triatoma brasiliensis and Triatoma pseudomaculata were found in most municipalities as expected, being the main species caught in the semiarid RN region.R. nasutus was captured only in Apodi, Caraúbas, and Caicó municipalities and P. lutzi was found only in Serra Negra do Norte and in Caraúbas, as described (51) .In agreement with our studies, T. brasiliensis was the most frequently identifi ed species, followed by T. pseudomaculata and R. nasutus, as observed in several studies (9) (53) .We identified T. brasiliensis in the wild environment, peridomicile, and intradomicile with high T. cruzi infection rates.This species has been reported as semi-domestic, since it is an autochthonous species capable of colonizing domiciles and the peridomicile (12) .It is highly dispersed and frequent owing to its eurytopic characteristics, is not hygrophilous, is able to withstand very high temperatures, and thus is acclimatized to the vast expanse of the Northeastern region (13) .
Despite the absence of intradomicile colonies, frequent adult T. brasiliensis invasions occur because of the proximity of the wild environment with DUs.Residential lighting attracts these insects (54) , potentially explaining their localized capture.Peridomicile colonization was observed by T. brasiliensis and T. pseudomaculata nymphs, most frequently in chicken coops where they develop dense colonies, facilitating intradomiciliary invasion.T. brasiliensis was also found in piles of tiles, corrals, and associated with domestic and synanthropic animals, according to municipality.This species strongly associates with native rodents, especially Galea spp., which thrive around rural dwellings in this state (55) .The epidemiological profi le similarities among the municipalities studied suggest the need for interventions to prevent parasite transmission in this environment to domestic animals and humans.
Degradation of the natural wild vector habitat allows the bugs to move close to human habitations; such coexistence increases the probability of human infection (56) .In the wild, T. brasiliensis was mainly found colonizing rocky outcrops and stone fences.Its occurrence in Northeastern Brazil is associated with rock formation distribution (6) .However, this species can also be found in shrubby cacti co-occupied by native rodents in Ceará (57) .
Triatoma pseudomaculata is widely distributed in the wild and is considered to be diffi cult to control (5) .T. pseudomaculata is primarily associated with Caatinga and areas of the Cerrado in Brazil (58) , and is an arboricolous species (59) .Here, the natural habitat of T. pseudomaculata was identifi ed as bird's nests, corroborating previous fi ndings (32) .
In the study area, the natural R. nasutus habitat was the Carnauba palm, considered the major ecotope of this species (9) (60) .Colonization foci were observed in the peridomicile in dry carnauba straw used by the rural population to manufacture household items in Apodi.This can be explained by the proximity of palm trees with DUs that attract insects as previously described or by passive insect transport by residents living in carnauba extraction areas, where storing dry leaves in the peridomicile is common.In the State of Ceará, R. nasutus has been frequently found colonizing the peridomicile owing to the use of carnauba straw for chicken coop roof construction and household goods manufacture (9) (61) .
Panstrongylus lutzi was the most infected species in the intradomicile and wild environment.The former is worrying because of the T. cruzi transmission risk to domestic animals and humans and potential introduction of a new parasite genetic group into the transmission cycles, since this area is currently exclusively infected with DTU III (15) .P. lutzi has restricted distribution in areas of the semiarid northeast although it has wide geographical distribution and a high T. cruzi infection rate in the State of Pernambuco.Its domiciliation has occurred in the States of Ceará and Pernambuco (62) (63) .P. lutzi is found in hollow Auxemma oncocalyx trunks in Ceará (64) and in armadillo burrows in the semiarid Caatinga in Bahia (59) .However, the P. lutzi natural habitat was not identifi ed despite extensive searching in the wild environment of the Seridó Ecological Station where most specimens were captured; thus, further studies are required (15) (51) .
Currently, vector transmission is considered residual by a few native and peridomestic species such as T. brasiliensis and T. pseudomaculata.There is also a risk of progressive domiciliation of certain species previously considered sylvatic such as P. lutzi and the possibility of human infection directly related to the parasite enzootic cycle (62) .P. lutzi holds relevance toward maintaining the peridomestic and domestic T. cruzi transmission cycles, the risk of invasion and eventual colonization therein, and the consequent parasite transmission to domestic animals and humans.These results highlight the increasing epidemiological importance of P. lutzi and indicate the continuing necessity of maintaining epidemiological surveillance against T. cruzi transmission in the study area.
Here, we reported a slight agreement among T. cruzi detection methods used to evaluate infection in triatomines, and the importance of its association to fi eld studies.T. cruzi infection varied among triatomine species according to method with PCR showing high positivity whereas detection by DME in intestinal contents was low.However, higher infection rates were reported with fl agellates morphologically similar to T. cruzi in Pernambuco (63) , and similar or higher triatomine infection rates have been reported between methods (19) (23) (24) (25) .High PCR positivity in relation to DME has been previously observed, indicating its utility for epidemiological studies (19) (23) (24) (65) .PCR can also directly detect T. cruzi vector infection thus improving triatomine evaluation and should be used to assess infection rates in dead insects (25) owing to its higher relative degree of precision.Thus, PCR represents the best tool for parasite detection, confi rming the majority of infections ascertained by DME and xenoculture.However, these remain useful as the combination of methods can contribute to monitoring T. cruzi in triatomines and enhance confi dence regarding triatomine positivity.
The Trypanosoma cruzi DTUs identifi ed in 15 isolates from triatomine bugs were TcII and TcIII in T. brasiliensis and TcIII in P. lutzi.Previous fi ndings showed these DTUs circulating in these species in the semiarid zone of RN (15) (17) and highlight the need to understand T. cruzi population distribution in this area.Naturally heterogeneous T. cruzi populations involving TcI, TcII, and TcIII circulate among humans and triatomines in three different municipalities without domicile colonization (17) , and high genetic similarities exist among T. cruzi populations circulating in different hosts, localities, and environments (16) .TcII was detected in the peridomicile in Caicó and in the wild in Serra Negra do Norte, indicating its participation in sylvatic and peridomestic cycles.The original primary host of TcII appears to be primates.Whereas its ecological niche has yet to be determined, isolates have been described from opossums in the Atlantic forest and from sylvatic primates, suggesting that such primates might be the primary original mammalian host (66) .TcII has also been isolated from the armadillo Euphractus sexcintus in the Paraguayan Chaco (67) .This DTU has been shown as the primary cause of severe acute and chronic Chagas disease in the Atlantic Forest and central region of Brazil and represents the etiological disease agent in São Felipe in the State of Bahia, where the domestic vector is P. megistus (67) .
TcIII was isolated from T. brasiliensis in the wild environment and from P. lutzi captured in the domiciliary and wild environments of two different localities.TcIII is a poorly understood T. cruzi genetic group predominantly identifi ed among wild cycles of parasite transmission infecting terrestrial mammals and triatomine vectors, but is also a potentially important emergent human disease agent (68) .A few triatomine species have been described in sylvatic TcIII transmission such as Panstrongylus geniculatus Latreille, 1811, and Triatoma rubrovaria Blanchard, 1843, both mainly sylvatic vectors frequently associated with terrestrial ecotopes (69) .TcIII is found in a broad range of terrestrial mammals and its transmission may occur inside triatomine-infested burrows by both vectorial and oral routes (29) (70) .Overlapping geographic areas of TcIII and TcI isolates occur across South America, with shared wild mammals and vectors in terrestrial ecotopes (29) .TcIII in domestic transmission cycles, while intermittent (68) , implies a role as a human disease agent.Furthermore, TcIII may be under-reported in both domestic and sylvatic transmission cycles because some typing methodologies fail to distinguish between TcIV and TcIII (71) .The TcII and TcIII identifi ed in this study corroborate previous fi ndings and reinforce the need for constant epidemiological surveillance of T. brasiliensis and P. lutzi to prevent TcIII spread to the domestic cycle (15) .TcI was not detected in triatomines in the studied area.However, most studies conducted in northeastern Brazil have isolated this DTU from T. brasiliensis and T. pseudomaculata, with most naturally infected T. brasiliensis isolates being TcI (72) .TcI has also been identifi ed in T. pseudomaculata and R. nasutus in the peridomicile and natural environments of fi ve State of Ceará peri-urban and urban localities (73) .Therein, high TcI and TcII infection rates were detected in peridomestic T. brasiliensis and T. pseudomaculata, respectively (74) .TcI has been most frequently identifi ed in triatomines of the State of Mato Grosso do Sul (75) .The majority of isolates from humans, reservoirs, and vectors from Amazonia correspond to TcI (29) it has been detected in humans from State of RN (15) (16) , with recent isolates from patients with various clinical forms of Chagas disease (17) .T. cruzi genotypes isolated from triatomines can vary according to vector-specifi c physiology and ecological habitat (43) , highlighting the necessary to accurately identify TcI in triatomines.
Our results suggest that combining methodologies contributed to increased parasite detection and to identifying infection foci to precisely determine triatomine distribution.Despite low parasite detection via xenoculture and DME, these methods allow the isolation of T. cruzi for further study.Owing to high peridomestic ecotope infestation and high T. brasiliensis and T. pseudomaculata T. cruzi infection, we recommend effi cient entomological surveillance programs to detect possible colonization.This study contributes to our knowledge of T. cruzi diagnosis and identifi cation in fi eld-collected triatomines, further demonstrating that T. cruzi II and III predominate in the study area and, with future research, are essential for vector control and human infection prevention.

ForFIGURE 1 -
FIGURE 1 -Map of the State of Rio Grande do Norte highlighting the West and Central mesoregions.The study area showing the surveyed municipalities is shown in dark grey.

FIGURE 2 -
FIGURE 2 -Rate of natural infection of triatomine species by Trypanosoma cruzi as determined using different methods and in three distinct environments.A) Infection rate of triatomine species by detection method and global infection index.B) Infection rate of triatomines in different capture environments and total number of infected specimens.T. brasiliensis: Triatoma brasiliensis; T. pseudomaculata: Triatoma pseudomaculata; P. lutzi: Panstrongylus lutzi; R. nasutus: Rhodnius nasutus; PCR: polymerase chain reaction.