INTRODUCTION
Trypanosoma cruzi is the causative agent of Chagas disease, a neglected human protozoan disease that is estimated to affect approximately six million people, spanning 21 endemic Latin American countries, with 60–80 million at risk of infection (WHO, 2015). Trypanosoma cruzi is genetically heterogenous, infecting a large number of mammal species and transmitted by haematophagous triatomine insect vectors. Nomenclature is historically complicated, but T. cruzi is currently subdivided into six subspecific groups, referred to as genetic lineages or discrete typing units (DTUs) and designated TcI to TcVI (Zingales et al. Reference Zingales, Miles, Campbell, Tibayrenc, Macedo, Teixeira, Schijman, Llewellyn, Lages-Silva, Machado, Andrade and Sturm2012). A cohort of geographically disparate bat trypanosomes, provisionally designated as TcBat, has been shown to share phylogenetically close affiliations with TcI (Marcili et al. Reference Marcili, Lima, Cavazzana, Junqueira, Veludo, Maia Da Silva, Campaner, Paiva, Nunes and Teixeira2009a ), although more detailed sampling is required to confirm this as a formal taxonomic group. In a recent review, analysing more than 400 sequences with two mitochondrial (CytB and COII) and one nuclear gene (Gpi), authors propose three significant reliable mitochondrial clades, named mtTcI, mtTcII and mtTcIII, instead of seven (Barnabé et al. Reference Barnabé, Mobarec, Jurado, Cortez and Brenière2016). Phyloepidemiology and host vector associations of T. cruzi are complex, but have been partially resolved (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005; Miles et al. Reference Miles, Llewellyn, Lewis, Yeo, Baleela, Fitzpatrick, Gaunt and Mauricio2009; Messenger et al. Reference Messenger, Miles and Bern2015; Brenière et al. Reference Brenière, Waleckx and Barnabé2016). TcI is widespread through the Americas. This is the major DTU found infecting Didelphis opossums in nature, which is believed to be its most ancestral host. TcI was reported predominating in domestic transmission cycles in northern countries of South America (for example, Colombia and Venezuela) and in Central America (Miles et al. Reference Miles, Llewellyn, Lewis, Yeo, Baleela, Fitzpatrick, Gaunt and Mauricio2009). TcIII and TcIV primarily circulate in sylvatic transmission cycles, the former especially associated with armadillos (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005) and the latter with a variety of sylvatic mammal species (Miles et al. Reference Miles, Llewellyn, Lewis, Yeo, Baleela, Fitzpatrick, Gaunt and Mauricio2009). TcIV is a secondary cause of Chagas disease in Venezuela. TcIII is rarely reported from domestic transmission cycles. In contrast, TcII, TcV and TVI predominate in domestic transmission cycles in Southern Cone countries of South America (Miles et al. Reference Miles, Llewellyn, Lewis, Yeo, Baleela, Fitzpatrick, Gaunt and Mauricio2009; Messenger et al. Reference Messenger, Miles and Bern2015; Brenière et al. Reference Brenière, Waleckx and Barnabé2016). Remarkably, TcV and TcVI are known to be natural hybrids derived from genetic exchange between TcII and TcIII in recent evolutionary history and are at present strongly associated with domestic transmission cycles (Zingales et al. Reference Zingales, Miles, Campbell, Tibayrenc, Macedo, Teixeira, Schijman, Llewellyn, Lages-Silva, Machado, Andrade and Sturm2012; Brenière et al. Reference Brenière, Waleckx and Barnabé2016).
Trypanosoma cruzi infection is considered primarily a zoonosis and as such eradication is not possible. Effective control of Chagas disease is achieved by interrupting vectorial transmission, primarily through residual insecticide-spraying to reduce domestic infestation and also by serological surveillance and interruption of transmission by blood transfusion, organ donation and congenitally (WHO, 2015).
There has been remarkable progress in controlling Chagas in some regions of the Americas. However, the Gran Chaco region, which includes territories of Argentina, Bolivia and Paraguay, is currently considered one of the most difficult regions for effective control and remains highly endemic (Hotez, Reference Hotez2014a ). The land area is vast and arid, with a low population density consisting of small widely dispersed communities of low socioeconomic status (Gürtler, Reference Gürtler2009). More than 20 ethnic groups live in marginalized conditions with minimal access to health care provisions (Gracey and King, Reference Gracey and King2009; Hotez, Reference Hotez2014b ). Indigenous communities of the Gran Chaco show consistently high seroprevalence of human T. cruzi infection, ranging from 12 to 83% with local variation (Canese and Brice, Reference Canese and Brice1977; Rojas de Arias et al. Reference Rojas de Arias, Arévalo de Guillén, Inchaustti, Samudio and Schmeda-Hischmann1993; Moretti et al. Reference Moretti, Castro, Franceschi and Basso2010; Samuels et al. Reference Samuels, Clark, Galdos-Cardenas, Wiegand, Ferrufino, Menacho, Gil, Spicer, Budde, Levy, Bozo, Gilman and Bern2013).
Although the success of vector control interventions in some areas of the Chaco has substantially reduced disease incidence, the main challenge is the long-term sustainability, and in particular, entomological surveillance. Reinfestation of treated dwellings, when the residual effect of insecticides decreases, is a common feature especially in areas with peridomestic vectors and/or reinvasion by secondary vectors from the sylvatic environment (Provecho et al. Reference Provecho, Gaspe, del Pilar Fernández, Enriquez, Weinberg and Gürtler2014; Gaspe et al. Reference Gaspe, Provecho, Piccinali and Gürtler2015). Other obstacles adversely affecting control in the Chaco include the low efficacy of pyrethroid insecticide spraying on often poorly constructed peridomestic structures in this region (Gürtler et al. Reference Gürtler, Kitron, Cécere, Segura and Cohen2007; Cécere et al. Reference Cécere, Vázquez-Prokopec, Ceballos, Boragno, Zárate, Kitron and Gürtler2013). Of further concern, is the appearance of Triatoma infestans populations resistant to pyrethroid insecticides in localities of northern Argentina and southern Bolivia (Lardeux et al. Reference Lardeux, Depickère, Duchon and Chávez2010; Gurevitz et al. Reference Gurevitz, Gaspe, Enríquez, Vassena, Alvarado-Otegui, Provecho, Cueto, Picollo, Kitron and Gürtler2012). Sylvatic populations of T. infestans have been identified in the Chaco and pose a potential risk of reinvasion (Noireau et al. Reference Noireau, Flores, Gutierrez and Dujardin1997a ; Ceballos et al. Reference Ceballos, Piccinali, Marcet, Vazquez-Prokopec, Cardinal, Schachter-Broide, Dujardin, Dotson, Kitron and Gürtler2011; Quisberth et al. Reference Quisberth, Waleckx, Monje, Chang, Noireau and Brenière2011; Rolón et al. Reference Rolón, Vega, Román, Gómez and Rojas de Arias2011) as do secondary vectors, including Triatoma sordida (Almeida et al. Reference Almeida, Vinhaes, de Almeida, Silveira and Costa2000; Damborsky et al. Reference Damborsky, Bar and Oscherov2001; Feliciangeli et al. Reference Feliciangeli, Campbell-Lendrum, Martinez, Gonzalez, Coleman and Davies2003; Lauricella et al. Reference Lauricella, Stariolo, Riarte, Segura and Gürtler2005); additionally sylvatic mammals are potential reservoirs of infection, and all of these factors may confound effective control. This study ascertains the different triatomine species present in the region, the mammal species infected and the associated circulating T. cruzi DTUs. Through a better understanding of the T. cruzi transmission dynamics we aim to improve control strategies.
MATERIALS AND METHODS
Fieldwork
Fieldwork collections were performed from 2002 to 2008, with the objective of obtaining and genotyping new isolates of T. cruzi from triatomine bugs and sylvatic mammals. Isolates obtained in previous surveys (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005; Llewellyn et al. Reference Llewellyn, Lewis, Acosta, Yeo, Carrasco, Segovia, Vargas, Torrico, Miles and Gaunt2009, Rojas de Arias et al., manuscript in preparation) were also included to generate a more detailed picture of the distribution of T. cruzi DTUs in the Chaco region.
Study area
The study area encompasses the Paraguayan Chaco (western region), Bolivian Chaco (southern region) and also three further Paraguayan localities 250 km northeast of Asunción (San Pedro, San Alfredo and Aguapey), within the Department of San Pedro. Study areas are shown in Fig. 1. In total, data were acquired from 28 localities, 24 Paraguayan and four Bolivian. Of the 24 Paraguayan localities, 21 were situated in the Paraguayan Chaco spanning three departments (Boquerón, Presidente Hayes and Alto Paraguay). Within the Department of Boquerón 12 localities were studied: Betania, Campo Loro, Campo Nuevo, Campo Salado, Cesarea, Galilea, Jerico, Campo Alegre, Casuarina, Jotoisha, Tiberia and Samaria. Within the Department of Presidente Hayes a further eight localities were included: Cerrito, Estancia Salazar, 12 de Junio, 20 de abril, Campo Largo, 10 Leguas, Fischat and Jope. Lastly, the one locality from the third department was Don Anibal ranch. These aforementioned localities have been under epidemiological surveillance by the National Program Control of Chagas since 2001. Further details regarding localization, ethnic group and estimated population size of these communities is shown in Table A1 (Appendix A). The remaining three localities lie within the Department of San Pedro (Fig. 1), in the Southeast Chaco: San Pedro, San Alfredo and Aguapey. Localities from the Bolivian Chaco region, San Antonio, Mora, Cuatro Cañadas and Gutierrez were all from Santa Cruz Department.
Fig. 1. Distribution of Trypanosoma cruzi DTUs in the study area. Map illustrating study sites in Paraguay (orange) and Bolivia (yellow) and the distribution of trypanosomes characterized. Circles and triangles represent isolates from domestic and sylvatic cycles, respectively. Colours indicate different Trypanosoma cruzi DTUs. Green, TcII; blue, TcIII; yellow, TcV and light blue, TcVI.
Indigenous communities consist of nomadic hunters, gatherers and fishermen with some groups of sedentary farmers from distinct linguistic groups (Rojas de Arias, Reference Rojas de Arias2003). Dwellings are typically of low quality, walls constructed of wattle and brick, adobe or palm trunk, soil floor with straw, and palm leaves or tin as roofing material. Villages are located in peripheral areas surrounding Mennonite towns. Domestic animals consist primarily of dogs and chickens. Further to the Northeast and in the Bolivian Chaco land use is predominantly for cattle farming and agriculture. To the east of the Paraguayan Chaco, the Department of San Pedro is characterized by more abundant vegetation and intensive agricultural land use.
Collection of triatomine bugs and sylvatic mammals
Each locality was visited only once. Most of the collection of domiciliary and peridomiciliary triatomines was made by manual searches (1·0 person hour per house) (Gürtler et al. Reference Gürtler, Chuit, Cécere, Castañera, Cohen and Segura1998). Domestic areas inspected included walls, interior side of roofs, furniture and bedding. Peridomestic searches encompassed chicken coups, wood piles and other peridomestic structures (Gurevitz et al. Reference Gurevitz, Ceballos, Gaspe, Alvarado-Otegui, Enríquez, Kitron and Gürtler2011). In Campo Loro, Estancia Salazar and Fischat localities domestic bugs were obtained by the collection of local inhabitants. Sylvatic triatomines were collected by the use of live-bait Noireau traps (Noireau et al. Reference Noireau, Flores, Gutierrez, Abad-Franch, Flores and Vargas2000), manual dissection of natural ecotopes (bird nests, fallen trees and scrub) and light traps (Vazquez-Prokopec et al. Reference Vazquez-Prokopec, Ceballos, Marcet, Cécere, Cardinal, Kitron and Gürtler2006). Light traps were left overnight and checked in the morning. Captured bugs were placed in labelled containers, the developmental stage noted and identified to the level species by trained personel according to Lent and Wygodzinsky (Reference Lent and Wygodzinsky1979). The captured nymphs were raised in laboratory conditions until they reached the adult stage to confirm the species. Sylvatic mammals were captured by the use of collapsible ‘live-traps’ including Sherman (H.B. Sherman Trap, Inc., Tallahassee, FL) and Tomahawk (Tomahawk Live Trap Co., WI), baited with a mixture of peanut butter, ripe banana and oats. Ten traps per night were set at approximately 10 m intervals on animal trails, near burrow entrances, in dense scrub or close to fallen trees. Traps were left in situ for 3 days, set at sunset and examined at dawn, where applicable local hunters were hired to collect live mammals. All specimens captured were sexed, identified to the species level (Neris et al. Reference Neris, Colmán, Ovelar, Sukigara and Ishii2002), and released unharmed after processing (see below).
Isolation and characterization of trypanosomes
Trypanosomes were obtained from animals via xenodiagnosis and from triatomines by haemoculture and xenoculture as previously described (Miles, Reference Miles and Hyde1993). Mammals were first anaesthetized by intramuascular injection using ketamine (Holliday-Scott®, 50–80 mg kg−1 body weight). Animal-handling procedures were in accordance with the American Society of Mammalogists (Sikes and Gannon, Reference Sikes and Gannon2011). To excluded the presence of mixed infection biological clones of Trypanosoma cruzi were obtained by direct culture of infected triatomine feces, onto solid medium agar plates, as described previously by Yeo et al. (Reference Yeo, Lewis, Carrasco, Acosta, Llewellyn, da Silva Valente, de Costa Valente, de Arias and Miles2007), five clones of each isolate were expanded in culture and DNA extracted [DNeasy kits (QIAGEN™)].
Genetic characterization of DTUs was undertaken using a combination of amplicon profiles from four different polymerase chain reactions (PCR) details of which are shown in Table 1. Genetic targets were the D7 divergent domain of the 24Sα rRNA (Souto et al. Reference Souto, Fernandes, Macedo, Campbell and Zingales1996), the size variable domain of 18S rRNA sequence (Brisse et al. Reference Brisse, Barnabé and Tibayrenc2000), the non-transcribed spacer of the mini-exon gene (Souto et al. Reference Souto, Fernandes, Macedo, Campbell and Zingales1996) and the PCR–restriction fragment length polymorphism (PCR–RFLP) of the intergenic region of the heat-shock protein 60 (HSP60) gene (Lewis et al. Reference Lewis, Ma, Yeo, Carrasco, Llewellyn and Miles2009). Primers and reaction conditions are described in Table B1 (Appendix B). A panel of reference strains, encompassing the known DTUs, was obtained from the London School of Hygiene and Tropical Medicine cryobank repository and consisted of X10 Clone I (DTU TcI), Esmeraldo-cl3 (DTU TcII), Arma 13 (DTU TcIII), CAN III (DTU TcIV), SC43 (DTU TcV) and CL Brener (DTU TcVI).
Table 1. DTU discrimination based on PCR amplification products (bp) a
a According to Yeo et al. (Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005) and Lewis et al. (Reference Lewis, Ma, Yeo, Carrasco, Llewellyn and Miles2009).
b Brisse et al. (Reference Brisse, Barnabé and Tibayrenc2000) reported bands of 125 bp in one strain (Saimiri 3) and of 130 bp for three strains of North American origin. Band of 117 bp was reported by Kawashita et al. (Reference Kawashita, Sanson, Fernandes, Zingales and Briones2001).
c According to Brisse et al. (Reference Brisse, Barnabé and Tibayrenc2000). They reported product a low intensity band of 300 bp in two strains (M6241 cl6 and M5631 cl5).
d According to Brisse et al. (Reference Brisse, Barnabé and Tibayrenc2000).
e According to Westenberger et al. (Reference Westenberger, Barnabé, Campbell and Sturm2005) and Lewis et al. (Reference Lewis, Ma, Yeo, Carrasco, Llewellyn and Miles2009).
f Differentiates between TcII and TcVI.
RESULTS
Triatomines
A total of 1044 triatomine bugs were included in the current study, 1037 from Paraguay and seven from the Bolivian Chaco. Triatoma infestans (n = 715) was found in both domestic and peridomestic environments (n = 245 in domestic; n = 470 in peridomestic) and T. sordida (n = 203) only in peridomestic environments (Table 2). Adults, fourth- and fifth-instar nymphs were collected from both species in both areas. In Betania, Campo Salado, Galilea both species shared the same niche in chicken coops. In the localities of Campo Nuevo, Cesarea and Samaria only T. sordida was present, while in Estancia Salazar only T. infestans was found. Although T. infestans was collected from both peridomestic and domestic areas, only domestic specimens were microscopically positive for T. cruzi. Thus, 55 (5·4%) of T. infestans were positive, including adults (n = 41), fourth-instar (n = 10) and fifth-instar (n = 4) nymphs. Positive triatomines were from the localities of Jerico (n = 31), Galilea (n = 2), Betania (n = 1), Campo Loro (n = 5), Estancia Salazar (n = 8), Jope (n = 5) and Fischat (n = 3). In the sylvatic area, 115 of adults (n = 107) and nymphs (n = 8) of Triatoma guasayana, three adults of T. sordida, and one female of Triatoma platensis were captured from the localities of Campo Loro, Don Anibal ranch and Betania respectively and they were microscopically negative. Triatomines obtained from Bolivian Chaco by the cooperation of local inhabitants included seven sylvatic adult bugs, T. sordida (one specimen) and T. guasayana (six specimens). One T. guasayana specimen from the locality of Mora was positive.
Table 2. Summary table: species captured, location, ecotopes, number of positives and DTUs observed
a, domestic; b, captured by local inhabitants; c, include samples from Yeo et al. (Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005); d, include samples from Llewellyn et al. (Reference Llewellyn, Lewis, Acosta, Yeo, Carrasco, Segovia, Vargas, Torrico, Miles and Gaunt2009); e, include samples from Rojas de Arias et al. (in preparation); ND, not determined.
Sylvatic mammals
A total of 138 mammals were included in the current study, 26 from the Bolivian Chaco and 112 from Paraguayan localities. From Bolivia animals representing six species were examined, Euphractus sexcinctus (n = 7), Dasypus novemcinctus (n = 9), Chaetophractus spp. (n = 6), Tamandua tetradactyla (anteaters, n = 1), Dasyprocta spp. (agutí, n = 2) and Nasua nasua (coati, n = 1). Eight animals (30%) were found to be infected by haemoculture and/or xenodiagnosis: five D. novemcinctus, two E. sexcinctus and one Chaetophractus spp. from Gutierrez and Mora localities.
From Paraguayan localities a total of 112 sylvatic animals of five species were captured and examined. In the Chaco localities, 84 mammals included four different armadillo species: E. sexcinctus (n = 20), D. novemcinctus (n = 5), Cabassous spp. (n = 1) and Chaetophractus spp. (n = 58). Six animals (7%) from Campo Loro and Don Anibal ranch were infected: two E. sexcinctus, two Chaetophractus spp. and two D. novemcinctus. All the specimens captured in Cerrito were negative by microscopy, xenodiagnosis and haemoculture. Twenty-eight animals were captured from San Pedro. They included D. novemcinctus (n = 14) and Didelphis albiventris (opossum, n = 14). Three animals (11%) from Colonia San Alfredo and Aguapey, one armadillo and two opossums, were infected.
Trypanosome isolates
A total of 166 T. cruzi isolates were included in this study, consisting of 63 new field isolates and a further 103 obtained from previous collections, as described below. Table 2 summarizes the origins of the isolates.
Sixty-three isolates were genotyped to DTU; ten from eight triatomines and two Didelphis spp., could not be maintained in culture and were excluded. The isolates from previous collections (Table 2) included 12 that had been genotyped, six from domestic T. infestans and six from sylvatic armadillos from the Chaco region of Paraguay (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005). Eighty-three isolates originated from domestic T. infestans (Rojas de Arias et al., in preparation). Eight isolates were from sylvatic armadillos in Bolivia (Llewellyn et al. Reference Llewellyn, Lewis, Acosta, Yeo, Carrasco, Segovia, Vargas, Torrico, Miles and Gaunt2009).
Characterization of trypanosome isolates
Amplicon sizes obtained with new trypanosome isolates were as expected for the corresponding DTU, according to previous surveys (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005; Lewis et al. Reference Lewis, Ma, Yeo, Carrasco, Llewellyn and Miles2009). Examples for PCR–RLFP HSP60/EcoRV that differentiate between TcIII, TcII and TcVI are shown in Fig. 2. Domestic T. infestans (n = 136) were infected with TcII, TcIII, TcV and TcVI. The most common DTU was TcV (64%), followed by TcVI (28%), TcII (6·6%) and TcIII (1·5%). Twenty-nine sylvatic isolates examined from different armadillo species and one T. guasayana (Bolivia) showed an amplicon profile corresponding to DTU TcIII. Five biological clones from each one of the 63 new field isolates were genotyped to detected mixed infections, but none were found.
Fig. 2. Agarose gel electrophoresis of PCR–RFLP products from HSP60/EcoRV of selected Trypanosoma cruzi isolates. Samples from sylvatic and domestic hosts in Paraguay and Bolivia: A: products without digestion, B: products after of digestion by EcoRV. Lanes: 1 and 12 contain hyperladder 4; 2: negative control; 3: TcIII from sylvatic Triatoma guasayana in Bolivia; 4–5: TcII from domestic T. infestans in Paraguay; 6–7–8: TcIII, TcII, TcVI reference strains, respectively; 9–10: TcIII from domestic T. infestans in Paraguay; 11: TcVI from domestic T. infestans in Paraguay.
Spatial distribution of DTUs
The locality with the most T. cruzi diversity was Jope, with all 4 DTUs circulating in the domestic transmission cycle, followed by Jerico, Casuarina and Fischat with three different DTUs. DTU distribution in the study areas is shown in Fig. 1 and Table 2. Within the domestic transmission cycle nine isolates of TcII were identified, all originated from T. infestans from six Chaco localities: Campo Loro (n = 1), Tiberia (n = 1), Casuarina (n = 1), Jope (n = 2), Campo Largo (n = 1) and Fischat (n = 3). DTU TcV was the most widely dispersed DTU in domestic areas, observed in 11 localities,which were Campo Loro (n = 5), Campo Largo (n = 6), Jerico (n = 26), 12 de Junio (n = 16), Casuarina (n = 18), Fischat (n = 1), Jope (n = 1), Galilea (n = 2), Campo Alegre (n = 3), Jotoisha (n = 8) and Betania (n = 1). Lastly, TcVI was identified in domestic areas of nine localities: Jerico (n = 1), Casuarina (n = 1), Tiberia (n = 1), Estancia Salazar (n = 4), 12 de Junio (n = 21), 20 de abril (n = 6), 10 Leguas (n = 2), Fischat (n = 1) and Jope (n = 1).
TcIII was predominant among sylvatic isolates and widespread in different armadillo species, including D. novemcinctus (n = 20), E. sexcinctus (n = 6) and Chaetophractus spp. (n = 3) from both Paraguay and Bolivia. TcIII was rarely observed in the domestic environment, being found in only two T. infestans specimens from the Jope locality of the Paraguayan Chaco. A single sylvatic specimen of T. guasayana from Mora locality (Bolivia) also harboured TcIII.
DISCUSSION
Triatomines and mammal species
Triatomines
Four triatomine species were found in the current study: T. infestans (n = 715), T. sordida (n = 203), T. guasayana (n = 115) and T. platensis (n = 1). Triatoma infestans was the only species found in the domestic environment and there was no evidence of domiciliation by T. guasayana and T. sordida. The latter two species were found in peridomestic and/or sylvatic areas. A single specimen of sylvatic T. platensis was captured. This presence of T. infestans in human dwellings (245 specimens) and peridomestic habitats (470 specimens) confirms that it is the primary vector of human T. cruzi infection in the Paraguayan Chaco (Rojas de Arias et al. Reference Rojas de Arias, Velázquez de Saldívar, Merlo, Monzón de Calabró, Gamarra de Galaher and Keller1990, Reference Rojas de Arias, Arévalo de Guillén, Inchaustti, Samudio and Schmeda-Hischmann1993, Reference Rojas de Arias, Rolón, Vega, Gómez, Román, Sánchez, Acosta, Villalba, Cécere, Marcet and Dotson2011; Zelada et al. Reference Zelada, Sanabria, Canese and Canese1998; Acevedo et al. Reference Acevedo, Aguilera, Miers, Velázquez and Canese2002). The overall prevalence of T. cruzi infection in captured triatomines was 5·4%, and all those infected in Paraguay were domestic T. infestans. One explanation is peridomestic T. infestans feeding on avian blood in chicken coops, and birds cannot maintain T. cruzi infection (Miles et al. Reference Miles, Feliciangeli and Rojas de Arias2003). Considering only domestic T. infestans the infection rate is 22·4%, representing a substantial risk of transmission to both humans and domestic animals. These findings accord with previous studies in the Paraguayan Chaco, which report human seroprevalence ranging from 12 to 83%, and house infestation rates from 26 to 100% (Canese and Brice, Reference Canese and Brice1977; Chapman et al. Reference Chapman, Baggaley, Godfrey-Fausset, Malpas, White, Canese and Miles1984; Rojas de Arias et al. Reference Rojas de Arias, Arévalo de Guillén, Inchaustti, Samudio and Schmeda-Hischmann1993, Reference Rojas de Arias, Rolón, Vega, Gómez, Román, Sánchez, Acosta, Villalba, Cécere, Marcet and Dotson2011; Rojas de Arias, Reference Rojas de Arias2003). Thus, the Chaco region is the most highly endemic for Chagas disease in Paraguay, primarily among native Amerindians of the low socioeconomic status.
In peridomestic environments, two triatomines species were found (T. infestans and T. sordida). Triatoma infestans in peridomestic areas poses a potential risk of re-infestation of buildings if surveillance measures are not continuous. Although peridomestic areas are also insecticide sprayed as part of the national control programme, the residual effect is lower than in the domicile because of exposure to climatic conditions. As a consequence this makes peridomestic bugs more difficult to control, requiring that spraying and surveillance to be more frequent. Triatoma sordida is widely distributed throughout Central Brazil, Eastern and Central Bolivia, the Chaco region of Paraguay and northwestern Argentina where it occurs primarily in the sylvatic environment (Lent and Wygodzinsky, Reference Lent and Wygodzinsky1979; Diotaiuti et al. Reference Diotaiuti, Pereira, Loiola, Fernandes, Schofield, Dujardin, Dias and Chiari1995). In the Bolivian Chaco, two putative cryptic species belonging to T. sordida complex, named groups 1 and 2, respectively, were recognized circulating in sympatry, using multilocus enzyme electrophoresis (Noireau et al. Reference Noireau, Gutierrez, Zegarra, Flores, Brenière, Cardozo and Dujardin1998). This species shows great capacity of adaptation to peridomestic sites, especially in association with chickens (Macchiaverna et al. Reference Macchiaverna, Gaspe, Enriquez, Tomassone, Gürtler and Cardinal2015). In the current study, all specimens (n = 203) were negative for flagellates probably because their principal food sources are avian. A primary domiciliation by T. sordida in the Chaco region was described in localities of Velasco Province, Department of Santa Cruz (Bolivia), where 16·2% of bugs were found infected by T. cruzi, although the probability of transmission to humans was considered low (Noireau et al. Reference Noireau, Brenière, Ordoñez, Cardozo, Morochi, Gutierrez, Bosseno, Garcia, Vargas, Yaksic, Dujardin, Peredo and Wisnivesky-Colli1997b ; Brenière et al. Reference Brenière, Morochi, Bosseno, Ordoñez, Gutierrez, Vargas, Yaksic and Noireau1998). Three species of sylvatic triatomine were found: T. guasayana (n = 115), T. sordida (n = 3) and T. platensis (n = 1). Their typical ecotopes were fallen trees and dense shrubs, where they were captured using Noireau traps. Triatoma guasayana was an active flyer seeking out potential hosts, with most flight activity occurring just after sunset. Both T. guasayana and T. sordida have been implicated as sylvatic vectors of T. cruzi in parts of the dry Chaco region (Wisnivesky-Colli et al. Reference Wisnivesky-Colli, Schweigmann, Pietrokovsky, Bottazi and Rabinovich1997; Vezzani et al. Reference Vezzani, Schweigmann, Pietrokovsky and Wisnivesky-Colli2001). In Paraguayan localities, both species were frequently observed near and around households, especially the adults, which have a great capacity for flight (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005; Rolón et al. Reference Rolón, Vega, Román, Gómez and Rojas de Arias2011). Because of these characteristics and their ability to colonize man made structures, they are candidate as secondary vectors. All our sylvatic T. guasayana and T. sordida captured in Paraguay were not infected with T. cruzi, apart from a single sylvatic Bolivian T. guasayana. However, infected T. guasayana and T. sordida have previously been reported in the Argentinean and Bolivian Chaco (Noireau et al. Reference Noireau, Flores, Gutierrez, Abad-Franch, Flores and Vargas2000; Bar et al. Reference Bar, Pieri Damborsky, Oscherov, Milano, Francisco, Avalos and Wisnivesky-Colli2002; Lauricella et al. Reference Lauricella, Stariolo, Riarte, Segura and Gürtler2005; Ceballos et al. Reference Ceballos, Piccinali, Berkunsky, Kitron and Gürtler2009) with average infection rates of 13·3 and 9·1%, respectively. In the Argentinean Chaco, sylvatic T. sordida have been reported with high infection rates (38·5%; Bar and Wisnivesky-Colli, Reference Bar and Wisnivesky-Colli2001; Bar et al. Reference Bar, Pieri Damborsky, Oscherov, Milano, Francisco, Avalos and Wisnivesky-Colli2002).
Here we did not find T. infestans in the sylvatic ecotope. However, sylvatic ‘dark morph’ T. infestans have been reported in the Chaco region of Bolivia, in nests of Myiopsitta monachus (parrot), in bromeliads and hollows of live trees in several localities (Noireau et al. Reference Noireau, Flores, Gutierrez and Dujardin1997a , Reference Noireau, Brenière, Ordoñez, Cardozo, Morochi, Gutierrez, Bosseno, Garcia, Vargas, Yaksic, Dujardin, Peredo and Wisnivesky-Colli b ; Brenière et al. Reference Brenière, Aliaga, Waleckx, Buitrago, Salas, Barnabé, Tibayrenc and Noireau2012; Waleckx et al. Reference Waleckx, Depickère, Salas, Aliaga, Monje, Calle, Buitrago, Noireau and Brenière2012), in the Argentinean Chaco (Ceballos et al. Reference Ceballos, Piccinali, Berkunsky, Kitron and Gürtler2009, Reference Ceballos, Piccinali, Marcet, Vazquez-Prokopec, Cardinal, Schachter-Broide, Dujardin, Dotson, Kitron and Gürtler2011) and Chile (Bacigalupo et al. Reference Bacigalupo, Segura, García, Hidalgo, Galuppo and Cattan2006). Prevalence of T. cruzi infection is markedly lower in such ‘dark morph’ forms from the Chaco region with prevalence of between 2·5 and 12·5% (Noireau et al. Reference Noireau, Flores, Gutierrez, Abad-Franch, Flores and Vargas2000; Brenière et al. Reference Brenière, Aliaga, Waleckx, Buitrago, Salas, Barnabé, Tibayrenc and Noireau2012; Waleckx et al. Reference Waleckx, Depickère, Salas, Aliaga, Monje, Calle, Buitrago, Noireau and Brenière2012) or zero (Ceballos et al. Reference Ceballos, Piccinali, Berkunsky, Kitron and Gürtler2009, Reference Ceballos, Piccinali, Marcet, Vazquez-Prokopec, Cardinal, Schachter-Broide, Dujardin, Dotson, Kitron and Gürtler2011) probably due that avian blood is the more often source of food. A few surveys have previously reported putative sylvatic populations of T. infestans in the Paraguayan Chaco, although they were also presumed attributable to dispersed peridomiciliary populations (Velázquez and González, Reference Velázquez and González1959; Usinger et al. Reference Usinger, Wygodzinsky and Ryckman1966; Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005). More recently putative sylvatic colonies were discovered using a trained dog (Rolón et al. Reference Rolón, Vega, Román, Gómez and Rojas de Arias2011), and these bugs were found 3 km from infested houses. It is significant that this species is capable of surviving in sylvatic ecotopes in at least in some regions of the Paraguayan Chaco. Further research is needed to establish the risk of reinvasion from such sylvatic populations of T. infestans.
Mammals
Eight species of mammals belonging to five different orders were captured in the study area. The overall prevalence of infection by T. cruzi was 12·3% (17/138), although this percentage varied according to the genus. Armadillos were the most common species captured in both regions of Paraguay (Chaco and San Pedro Departments) and from different localities in Bolivia. In two recent surveys performed in the humid Argentinean Chaco, marsupials and rodents together with armadillos were the most frequently captured species (Alvarado-Otegui et al. Reference Alvarado-Otegui, Ceballos, Orozco, Enriquez, Cardinal, Cura, Schijman, Kitron and Gürtler2012; Orozco et al. Reference Orozco, Enriquez, Alvarado-Otegui, Cardinal, Schijman, Kitron and Gürtler2013). The scarce number of marsupials captured and none for rodents in our study is probably related to the environment, since most of our successful collections were from the dry zone of the Chaco. The highest rate of infection was observed in D. novemcinctus (28·5%) followed by E. sexcinctus (14·8%) and Chaetophractus spp. (4·7%). In previous surveys in the same area, infection in armadillos ranged from 3 to 63%, with the highest prevalence in the Dasypus (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005; Llewellyn et al. Reference Llewellyn, Lewis, Acosta, Yeo, Carrasco, Segovia, Vargas, Torrico, Miles and Gaunt2009). Dasypus novemcinctus and the other armadillo especies are commonly hunted by the inhabitants of rural communities for food or for handicraft products, and they may be kept alive for several days before being used. Thus, infected armadillos pose a risk for bringing sylvatic T. cruzi into the domestic habitat. The triatomine vectors involved in sylvatic transmission cycles in the Chaco region remain uncertain. Members of the genus Panstrongylus were reported associated with armadillo burrows in Brazil (Grisard et al. Reference Grisard, Carvalho-Pinto, Scholz, Toma, Schlemper and Steindel2000), Venezuela (Llewellyn et al. Reference Llewellyn, Lewis, Acosta, Yeo, Carrasco, Segovia, Vargas, Torrico, Miles and Gaunt2009) and Argentina (Alvarado-Otegui et al. Reference Alvarado-Otegui, Ceballos, Orozco, Enriquez, Cardinal, Cura, Schijman, Kitron and Gürtler2012). Our finding of one infected sylvatic T. guasayana in fallen trees in Bolivia could suggest some role in sylvatic transmission. The omnivorous behaviour of some mammal species also may contribute to their acquisition of infection (Rabinovich et al. Reference Rabinovich, Schweigmann, Yohai and Wisnivesky-Colli2001). The prevalence of T. cruzi infection in Chaetophractus spp. was lower than the other armadillos. Although this species construct their own burrows, they are nomadic and rarely use the same burrow twice, and thus unlikely to become infested with triatomines. Three other mammal species: T. tetradactyla (anteater), Dasyprocta spp. (agutí) and N. nasua (coati) from Bolivian localities were not infected with T. cruzi. Natural infection of T. tetradactyla by T. cruzi has been reported in Brazil (Miles et al. Reference Miles, Póvoa, De Souza, Lainson, Shaw and Ketteridge1981; Bento et al. Reference Bento, Farias, Godoy and Araújo1992; Fernandes et al. Reference Fernandes, Mangia, Lisboa, Pinho, Morel, Zingales, Campbell and Jansen1999) and Colombia (Ramírez et al. Reference Ramírez, Duque and Guhl2011). In addition, anteaters and coati are the known hosts of T. rangeli (Miles et al. Reference Miles, Arias, Valente, Naiff, de Souza, Póvoa, Lima and Cedillos1983; Dereure et al. Reference Dereure, Barnabé, Vie, Madelenat and Raccurt2001).
Two D. albiventris of 14 captured (14·2%) from the Department of San Pedro were infected with T. cruzi, although isolates were not genotyped. These marsupials are usually found in humid areas, so the dry expanse of some Chaco zones may not be suitable for them. They are frequently observed in close proximity to human populations, and high T. cruzi infection rates have been found in Brazil (21·9 and 45·2% prevalence; Grisard et al. Reference Grisard, Carvalho-Pinto, Scholz, Toma, Schlemper and Steindel2000) and in the humid Chaco of Argentina (36 and 38% prevalence; Alvarado-Otegui et al. Reference Alvarado-Otegui, Ceballos, Orozco, Enriquez, Cardinal, Cura, Schijman, Kitron and Gürtler2012; Orozco et al. Reference Orozco, Enriquez, Alvarado-Otegui, Cardinal, Schijman, Kitron and Gürtler2013). In San Pedro Department, T. cruzi has also been found in the terrestrial opossum Monodelphis domestica (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005). Further studies are needed to understand fully the role of marsupials in transmission of T. cruzi in Paraguay.
Host–vector of T. cruzi genotypes in the Paraguayan Chaco
Ours is the most comprehensive survey of T. cruzi genotypes in the Paraguayan Chaco region, providing new insight into the transmission dynamics and dispersion among domestic and sylvatic cycles.
TcII, TcIII, TcV and TcVI were circulating in the region, with the hybrids TcV and TcVI being most frequently found, supporting earlier observations (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005; Lauthier et al. Reference Lauthier, Tomasini, Barnabé, Rumi, D'Amato, Ragone, Yeo, Lewis, Llewellyn, Basombrío, Miles, Tibayrenc and Diosque2012; Maffey et al. Reference Maffey, Cardinal, Ordóñez-Krasnowski, Lanati, Lauricella, Schijman and Gürtler2012; Pérez et al. Reference Pérez, Monje, Chang, Buitrago, Parrado, Barnabé, Noireau and Brenière2013). TcV and TcVI were predominant and the most dispersed, and found solely infecting T. infestans in the domestic cycle, also in agreement with previous surveys (Chapman et al. Reference Chapman, Baggaley, Godfrey-Fausset, Malpas, White, Canese and Miles1984; Acosta et al. Reference Acosta, Samudio, López, Vargas, Yaksic, Brenière and Rojas de Arias2001; Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005). TcV and TcVI were reported in domestic T. infestans in the Bolivian Chaco (Pérez et al. Reference Pérez, Monje, Chang, Buitrago, Parrado, Barnabé, Noireau and Brenière2013), in domestic and peridomestic triatomines (T. infestans and T. sordida) and domestic dogs and cats in the Argentinean Chaco (Maffey et al. Reference Maffey, Cardinal, Ordóñez-Krasnowski, Lanati, Lauricella, Schijman and Gürtler2012; Enriquez et al. Reference Enriquez, Cardinal, Orozco, Lanati, Schijman and Gürtler2013). Thus, TcV and TcVI constitute the largest current threat to human health, and have been associated with severe chronic manifestations of Chagas disease in the southern Cone countries (Corrales et al. Reference Corrales, Mora, Negrette, Diosque, Lacunza, Virreira, Brenière and Basombrio2009; Cura et al. Reference Cura, Lucero, Bisio, Oshiro, Formichelli, Burgos, Lejona, Brusés, Hernández, Severini, Velázquez, Duffy, Anchart, Lattes, Altcheh, Freilij, Diez, Nagel, Vigliano, Favaloro, Favaloro, Merino, Sosa-Estani and Schijman2012; Vicco et al. Reference Vicco, Bontempi, Ortiz, Solari, Bottasso and Marcipar2012; Lucero et al. Reference Lucero, Brusés, Cura, Formichelli, Juiz, Fernández, Bisio, Deluca, Besuschio, Hernández and Schijman2016). TcV and TcVI are infrequently reported in sylvatic cycles: TcV has been observed in one sylvatic D. novemcinctus and one E. sexcinctus in Paraguay (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005), in a rodent (Octodontomys spp.), three opossums, two ferrets and one skunk in Argentina (de Luca D'oro et al. Reference de Luca D'oro, Gardenal, Perret, Crisci and Montamat1993; Montamat et al. Reference Montamat, De Luca d'Oro, Perret and Rivas1992), and in two sylvatic triatomines (Triatoma spp.) from the Bolivian Chaco (M. Llewellyn, unpublished data). There is one record of TcVI in a D. marsupialis in the Northeast La Paz (the Jungas and Alto Beni regions) in Bolivia (Valette et al. Reference Valette, Brenière, Le Pont and Desjeux1988). It has been suggested that the domestic predominance of TcV and TcVI may be due to their recent anthropogenic origin and rapid clonal dissemination with T. infestans and human migration (Lewis et al. Reference Lewis, Llewellyn, Yeo, Acosta, Gaunt and Miles2011). The occurrence of sylvatic TcV and TcVI in other regions, such as the Atlantic forest, remains to be fully explored.
TcII was found only in domestic T. infestans, in agreement with previous surveys in the Paraguayan Chaco, where it is also associated within human infections (Acosta et al. Reference Acosta, Samudio, López, Vargas, Yaksic, Brenière and Rojas de Arias2001; Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005), although in lower frequency than the TcV and TcVI hybrids. TcII has been detected in single triatomines carrying mixed infection with TcVI (Yeo et al. Reference Yeo, Lewis, Carrasco, Acosta, Llewellyn, da Silva Valente, de Costa Valente, de Arias and Miles2007), and the presence of TcII may have been underestimated as discriminatory markers have not been applied. Like TcV and TcVI, TcII rarely been reported in sylvatic cycles, although this may reflect limited research. Recently, TcII was reported infecting one sylvatic T. infestans in the Bolivian Chaco (Waleckx et al. Reference Waleckx, Depickère, Salas, Aliaga, Monje, Calle, Buitrago, Noireau and Brenière2012). Likewise, this DTU has been reported from one monkey (Acosta et al. Reference Acosta, Miret, López and Schinini2016) and one E. sexcinctus, in Paraguay (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005) and from sylvatic mammals in Brazil (Fernandes et al. Reference Fernandes, Mangia, Lisboa, Pinho, Morel, Zingales, Campbell and Jansen1999; Bhattacharyya et al. Reference Bhattacharyya, Mills, Jansen and Miles2015; Lisboa et al. Reference Lisboa, Monteiro, Martins, Xavier, Lima and Jansen2015). Sylvatic TcII reservoirs are of particular interest as it is considered to be ancient (Westenberger et al. Reference Westenberger, Barnabé, Campbell and Sturm2005; de Freitas et al. Reference de Freitas, Augusto-Pinto, Pimenta, Bastos-Rodrigues, Gonçalves, Teixeira, Chiari, Junqueira, Fernandes, Macedo, Machado and Pena2006).
A striking predominance of TcIII was apparent in sylvatic isolates. Twenty-nine sylvatic armadillos from Paraguay (both regions) and Bolivia, one sylvatic T. guasayana (from Bolivia) and two domestic T. infestans (from Paraguay) harboured TcIII. TcIII is frequently and widely found in sylvatic habitats with armadillos, particularly the genus Dasypus (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005; Llewellyn et al. Reference Llewellyn, Lewis, Acosta, Yeo, Carrasco, Segovia, Vargas, Torrico, Miles and Gaunt2009; Morocoima et al. Reference Morocoima, Carrasco, Boadas, Chique, Herrera and Urdaneta-Morales2012). Armadillos infected with TcIII were also reported in Colombia (Saravia et al. Reference Saravia, Holguín, Cibulskis and D'Alessandro1987), Venezuela (Llewellyn et al. Reference Llewellyn, Lewis, Acosta, Yeo, Carrasco, Segovia, Vargas, Torrico, Miles and Gaunt2009; Morocoima et al. Reference Morocoima, Carrasco, Boadas, Chique, Herrera and Urdaneta-Morales2012), Bolivia (Llewellyn et al. Reference Llewellyn, Lewis, Acosta, Yeo, Carrasco, Segovia, Vargas, Torrico, Miles and Gaunt2009), Brazil (Lisboa et al. Reference Lisboa, Xavier, Herrera and Jansen2009; Marcili et al. Reference Marcili, Lima, Valente, Valente, Batista, Junqueira, Souza, da Rosa, Campaner, Lewis, Llewellyn, Miles and Teixeira2009b ) and Argentina (Alvarado-Otegui et al. Reference Alvarado-Otegui, Ceballos, Orozco, Enriquez, Cardinal, Cura, Schijman, Kitron and Gürtler2012; Orozco et al. Reference Orozco, Enriquez, Alvarado-Otegui, Cardinal, Schijman, Kitron and Gürtler2013). In San Pedro (Paraguay), armadillos and one specimen of M. domestica (opossum) were infected previously with TcIII (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005). One sylvatic T. guasayana from the Bolivian Chaco carried TcIII, presumably acquired by feeding on an armadillo; this is the first report of TcIII in T. guasayana in Bolivia. Triatoma guasayana is frequently found near houses, attracted by light and CO2, may therefore introduce TcIII into the domestic cycle. This DTU has also been isolated from terrestrial sylvatic triatomines collected, such as P. geniculatus, T. rubrovaria, T. brasiliensis and T. vitticeps in Brazil (Póvoa et al. Reference Póvoa, de Souza, Naiff, Arias, Naiff, Biancardi and Miles1984; Martins et al. Reference Martins, Marcili, Castanho, Therezo, de Oliveira, Suzuki, Teixeira, da Rosa and Sperança2008; Santos-Mallet et al. Reference Santos-Mallet, Silva, Gomes, Oliveira, Santos, Sousa, Pinheiro, Junqueira and Gonçalves2008; Lisboa et al. Reference Lisboa, Xavier, Herrera and Jansen2009) and from Panstrongylus spp. associated with a burrow of D. novemcinctus in Venezuela (Llewellyn et al. Reference Llewellyn, Lewis, Acosta, Yeo, Carrasco, Segovia, Vargas, Torrico, Miles and Gaunt2009).
Two domestic T. infestans from the Chaco region of Paraguay harboured TcIII. Previously in the same region TcIII isolates were obtained from domestic dogs (Chapman et al. Reference Chapman, Baggaley, Godfrey-Fausset, Malpas, White, Canese and Miles1984) and from sylvatic armadillos (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005), suggesting leaky separation between domestic and sylvatic cycles. Dogs are commonly used for hunting of armadillos in the Chaco region, and dogs may thus introduce TcIII into the domestic transmission cycles, but TcIII has so far not been isolated from human cases of Chagas disease in the Chaco region. In contrast, in the eastern region of Paraguay (Cordillera and Paraguarí Departments), using amplification products of the 24Sα rRNA, mini-exon and hybridization, TcIII was reported in human cases and domestic T. infestans (del Puerto et al. Reference del Puerto, Sánchez, Nara, Meza, Paredes, Ferreira and Russomando2010) as well as from domestic and peridomestic T. sordida from Concepción Department (Sánchez et al. Reference Sánchez, Russomando and Guillén2012); this interesting and surprising finding merits follow-up analyses.
Notably, TcI was absent from this survey. TcI has predominantly been found associated with arboreal marsupials, especially the Didelphis opossum throughout the Americas but also with rodents and other sylvatic mammals (Yeo et al. Reference Yeo, Acosta, Llewellyn, Sánchez, Adamson, Miles, López, González, Patterson, Gaunt, de Arias and Miles2005; Messenger et al. Reference Messenger, Miles and Bern2015). Records of the presence and distribution of TcI in Paraguay are scarce. It has been identified in the direct analysis of feces of domestic T. infestans from the Chaco and eastern region (Cura et al. Reference Cura, Mejía-Jaramillo, Duffy, Burgos, Rodriguero, Cardinal, Kjos, Gurgel-Gonçalves, Blanchet, De Pablos, Tomasini, da Silva, Russomando, Cuba, Aznar, Abate, Levin, Osuna, Gürtler, Diosque, Solari, Triana-Chávez and Schijman2010), in samples from domestic T. sordida in Concepción Department (Sánchez et al. Reference Sánchez, Russomando and Guillén2012), and in one human case from the Chaco region in a mixed infection with TcII (Risso et al. Reference Risso, Sartor, Burgos, Briceño, Rodríguez, Guhl, Chavez, Espinoza, Monteón, Russomando, Schijman, Bottasso and Leguizamón2011). Unfortunately, isolates were not obtained from the two infected opossums from San Pedro Department. TcIV was not found among our many Chaco region isolates, but it has been reported by direct analysis of the intestinal contents of domestic and peridomestic T. sordida captured in Concepción Department (eastern region) (Sánchez et al. Reference Sánchez, Russomando and Guillén2012), although additional analyses are required to confirm this observation.
Biological clones analysed in this study produced similar profiles to the original isolates with the combination of PCR techniques used in this study. The cloning technique on solid media has proven to be useful for discriminating mixed infections in T. cruzi reservoirs (Yeo et al. Reference Yeo, Lewis, Carrasco, Acosta, Llewellyn, da Silva Valente, de Costa Valente, de Arias and Miles2007), especially when a variety of DTUs are circulating sympatrically in the same area.
In summary, the distribution and the high prevalence of TcII, TcV and TcVI in domestic transmission cycles shows the remarkable diversity of T. cruzi in the Chaco region of Paraguay. In eight localities more than one T. cruzi DTU was present in the domestic transmission cycle showing the great capacity of T. infestans in indigenous communities to harbour a variety of T. cruzi populations. Furthermore, there is increasing evidence of interaction between domestic and sylvatic transmission cycles. Especially, TcIII in the Jope locality was found in both transmission cycles, suggesting introduction of TcIII into the domestic cycle. TcIII is known to be highly virulent in mice (Morocoima et al. Reference Morocoima, Carrasco, Boadas, Chique, Herrera and Urdaneta-Morales2012) and may therefore prove to be an agent of severe human Chagas disease. The abundance and aggressive nature of T. guasayana also carries a risk, if it should adapt to colonization of human dwellings.
The data generated here provide a regional baseline for future research and an indication of potential risks for human health. High-resolution analyses, including comparative genomics, will give further insight into T. cruzi transmission dynamics, interactions between sylvatic and transmission and molecular genetics, to inform the much needed improved control of Chagas disease in the Gran Chaco region.
ACKNOWLEDGEMENTS
Special thanks to those who gave assistance in the fieldwork in Paraguay, Elvio Benitez, Pedro Recalde, Victoria Bogado (+), Mercedes Bogado, César Zelaya, Cecilio Lezcano and to Carlos Chura from Bolivia, for his help in collecting samples from Bolivia. Thanks to Dr Antonieta Rojas de Arias, Dr Celeste Vega and Dr Mirian Rolón from the Centro para el Desarrollo de la Investigación Científica (CEDIC)/Díaz Gill Medicina Laboratorial/Fundación Moisés Bertoni, Asunción, Paraguay for the T. cruzi isolates used in this study.
FINANCIAL SUPPORT
This survey was supported by Wellcome Trust funding, grant number 066806.
APPENDIX A
Table A1. Localities surveyed in the Chaco (Paraguay and Bolivia) and San Pedro Departments
APPENDIX B
Table B1. Primers used and reaction conditions for each one of the PCR reactions performed
Open access
