Concise overview of knowledge on T. breyeri

The initial specimens of T. breyeri were collected in 1929 within the province of La Rioja, Argentina, which subsequently served as the foundation for the species’ formal description [8]. During the 1950s, the species remained relatively understudied, despite the collection of numerous specimens in that region. Indeed, at that time, its distribution was primarily constrained to the provinces of Catamarca and La Rioja in Argentina [9]. Subsequently, the species was also documented in the province of Córdoba [7] and within the province of Santiago del Estero, near its borders with Tucumán and Catamarca, as supported by a collection carried out in 1977 (GBIF data).

Triatoma breyeri can be readily identified by examining the morphological characteristics of both adults and nymphal stages [7, 9, 10] and even its egg has been described [11].

The taxonomic classification of T. breyeri remains a subject of ongoing debate within the scientific community. Initially, it was assigned to the "spinolai" complex, a group introduced in 1979 by Lent and Wygodzinsky to describe “a small, taxonomically isolated, and geographically limited group inhabiting semiarid regions of southern South America. This complex encompassed three species: T. spinolai (found in central and northern Chile), as well as the central and western Argentinian species T. eratyrusiformis and T. breyeri” [7]. Subsequently, with the establishment in 1940 of the Mepraia genus, T. spinolai Porter 1934 was reclassified as M. spinolai. Moreover, further complicating matters, T. breyeri was eventually placed within the Mepraia genus in 2002 as M. breyeri [12]. The spinolai complex became the Mepraia complex where all the species (the 3 above mentioned species plus M. gajardoi Frias, Henry & Gonzalez, 1998 and M. parapatrica Frias, 2010) constitute a monophyletic group [12–14]. However, some authors maintain a “breyeri” complex, which comprises of the only two species from the eastern sides of the Andes: T. breyeri and T. eratyrusiformis [15, 16] while the Mepraia s.s. species from the western side of the Andes, are maintained in the Mepraia complex. It is important to note that these taxonomic rearrangements remain contentious and continue to be a subject of dispute in the scientific community [17, 18]. In La Rioja Province, a subspecies known as T. breyeri dallasi was described [19]. Although it has not undergone formal synonymization [7], some authors have proposed such synonymy [4, 9, 20]. Nevertheless, the validation of this subspecies continues to be a subject of debate [21].

The ecosystem where T. breyeri grows has been described as a salty semidesert ecosystem around Salinas Grandes in central Argentina, ranging from 200 to 700 meters above sea level [15]. As of the present day, as revealed by the 12 distinct collecting points from the GBIF database (https://www.gbif.org/), the species’ geographic range is acknowledged to be restricted to the Andean pre-cordillera within these Argentinian provinces. These occurrences fall inside the two ecoregions Dry Chaco and High Monte as defined by the WWF (World Wide Fund for Nature) [22] (Fig 1A). The Dry Chaco ecoregion is primarily situated in the northwestern two-thirds of western Paraguay, extending eastward across the Andes into southeastern Bolivia and northwestern Argentina. The climate is dry, with an annual rainfall of 650–350 mm, and an average temperature of 28°C-12°C. The Chaco region comprises various habitats, including savannas, thorn forests, and transitional areas between these two dominant types. The High Monte ecoregion is situated along the eastern slopes of the Andes of Argentina, stretching from the province of Salta in the north to the province of Mendoza in the south. This landscape is characterized by rugged mountains and enclosed basins nestled between the Sierras Pampeanas to the east and the primary Andean mountain range to the west. The climate in this region is characterized by temperate, arid, and semi-arid conditions. Annual precipitation levels in the High Monte ecoregion vary depending on geographic location and elevation, ranging from under 100 mm to approximately 450 mm [23]. More precisely, the T. breyeri occurrences from the GBIF data base lie within several Köppen-Geiger climate classification [24]: BWh (arid, desert, hot) [2 collecting points], BWk (arid, desert, cold) [2 collecting points], BSh (arid, steppe, hot) [5 collecting points], Cwa (temperate, dry winter, hot summer) [2 collecting points], and Cwb (temperate, dry winter, warm summer) [1 collecting point] (Fig 1B).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1.

Collecting points of T. breyeri in Argentina and Bolivia, their related WWF ecoregions (A) and their related Köppen-Geiger climate classes (B). Country contours were downloaded from the GADM website, WWF ecoregions from the USGS website, and the Köppen-Geiger climatic regions from the World Bank Data Catalog. All data are classified as public.

https://doi.org/10.1371/journal.pone.0307989.g001

The ecological characteristics of T. breyeri are only partially understood. It predominantly inhabits the lower sections of branch fences, typically in proximity to the burrows of small rodents of the Caviidae and Cricetidae family [7]. It is reported that during the summer, adults can be encountered in the vicinity of houses and, occasionally, inside rooms [25–27]. However, it is important to note that they do not establish colonies within domestic environments [26, 28, 29]. Laboratory experiments conducted at a controlled temperature of 25°C (±4°C) and relative humidity of 60% (±20%) have revealed that T. breyeri exhibits some resistance to fasting and demonstrates a low population growth rate (r = 0.017) with limited dispersal capacity. The nymphal stage spans approximately 318 days (comprising 37 days for N1, 35 days for N2, 56 days for N3, 87 days for N4, and 103 days for N5). Adult individuals have an average lifespan of ≈three weeks, and the incubation period for eggs lasts ≈ 32 days. The generation time is estimated to be approximately 58 days (s.d. ≈ 4), while the R₀ value is calculated to be 2.68 (s.d. ≈0.3) [30]. The species has never been incriminated as a vector for Trypanosoma cruzi, the causative agent of Chagas disease, in humans [7], but may carry the entomopathogenous virus Triatoma virus [31].

Discovery in bolivia

In Bolivia, specimens of T. breyeri were first collected in 2010, in the small locality of Estancia-Mataral (lat. -18.5974, lon. -65.1463, alt. 1520 m), within a peri-domestic environment. Its discovery occurred during entomological surveys (i.e., active search in households by specialized technicians) conducted by the National Chagas Program of Bolivia as part of the efforts to control T. infestans the main vector of Chagas disease in Bolivia. Estancia-Mataral has no more than 60 households and lies where the small river Rio Novillero joins the Rio Grande, approximately 90 km by road, north of Sucre, the country’s administrative capital. Live specimens were transferred to the entomology laboratory at the Escuela Técnica de Salud Boliviano Japonesa de Cooperación Andina in Cochabamba, where they were reared under controlled conditions (28°C, 70% relative humidity), with monthly feeding on hens, for several years by one of us (RR). More recently, in two occasions in September 2017 a local resident of the small locality Media Luna (lat. -18.9533, lon. -65.1384) in the Rio Chico valley (S1 Fig), has reported adult specimens to the health centers of La Palma (about 50 km by road, south of Estancia Mataral), after discovering them in its peridomicile. Then, in April 2020 we collected another adult specimen in a domicile (lat. -18.8255, long. -65.1194) in Chuqui Chuqui, a small locality from the Rio Chico valley at about 15 km north of La Palma. In that instance, it was merely a unique specimen that had entered the house, lured by indoor lighting. In the same year, shortly afterward, another adult specimen was brought to the health center by an inhabitant of Chuqui Chuqui. The three collecting points of Bolivia are in close proximity. The valleys of Rio Chico and Rio Novillero converge at nearly the same point along the Rio Grande, with one valley seamlessly continuing into the other in a south-to-north direction. The straight-line distances between key locations are as follows: Mataral Estancia to Chuqui Chuqui is about 25 km, and Chuqui Chuqui to Media Luna (La Palma) is about 15 km. Considering the two nearest occurrences between Bolivia and Argentina, it becomes evident that the distance spans approximately 950 km, indicating a substantial gap between them (Fig 1A).

The three discovery points in Bolivia lies within the WWF “Bolivian Monte Dry Forest” ecoregion [22] and are located in the Köppen-Geiger class "BSh” (arid, steppe, hot) characterized by a pronounced dry season which can last several months. This climate class is also the one where most Argentinean locations are situated. The ecoregion is in the dry mountain valleys of the Andes in southern Bolivia. There, the mean monthly temperatures range from 13.5 C in July to 18.8°C in November. Yearly total precipitation is about 580 mm. Average monthly precipitation is below 10 mm in May–August, and rises to 137.4 mm in January.

Identification

Morphological identification.

Identification was conducted using unambiguous identification keys and descriptions [7, 9]. Triatoma breyeri closely resembles T. eratyrusiformis but can be distinguished by the absence of long hairs on the appendages, rounded posterior angles of the thorax, and uniform coloration along the lateral edges of the abdomen [32]. A general view of one of the Bolivian specimens from Estancia-Mataral is given in Fig 2 and S2–S4 Figs.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. General dorsal view of a T. breyeri specimen collected in Estancia-Mataral, Bolivia (voucher “Isolate A”).

https://doi.org/10.1371/journal.pone.0307989.g002

MaxEnt model of habitat suitability

The MaxEnt model was constructed using the kuenm R-package [35]. The kuenm process selected 46 statistically significant MaxEnt models. The top-performing tenth of the models were constructed employing the bioclimatic variables BIO6 (minimum temperature of the coldest month), BIO13 (precipitation of the wettest month), and BIO17 (precipitation of the driest quarter), with minimal variations attributed solely to differences in regularization parameters or feature classes. The foremost model was characterized by a regularization multiplier set at 0.6; and feature class ’l’. This model was ultimately selected as the definitive model. The computed mean AUC value of 0.93 represent an excellent performance and prediction of the model [36] (general MaxEnt outputs are in S1 Table). The representation of the habitat suitability is given in Fig 3. The Complementary Log-Log (cloglog) output format of MaxEnt was used. It gives an estimate between 0 and 1 of probability of presence of the species [37]. The median prediction value from all 12 model replicates (from bootstrapping) is represented in Fig 3. Unlike the mean, the median has a lower sensitivity to extreme values, which could be prevalent across replicates. Nevertheless, for reference, the mean, minimum and maximum predictions are available within the S8–S10 Figs respectively. To enhance the clarity of the T. breyeri habitat suitability maps, the suitability threshold chosen for visualization was the value of the mean maximum training sensitivity plus specificity (Max TSS) Cloglog. Consequently, areas with values below this threshold (0.6144 in this study) are depicted in grey in Fig 3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. MaxEnt representation of habitat suitability for T. breyeri. Median predictions.

https://doi.org/10.1371/journal.pone.0307989.g003

Habitat suitability for T. breyeri encompasses approximately the ecoregions of “Dry Chaco” and “High Monte” in Argentina and, not surprisingly, the southern part of the “Bolivian Montane Dry Forest” where it was collected in Bolivia. In this region, the prevailing Köppen-Geiger climate classes are BSh (arid, steppe, hot) and BSk (arid, steppe, cold). It is worth noting that these specific climate classes correspond to the majority of T. breyeri occurrences found in Argentina.

MaxEnt output is the Complementary Log-Log (cloglog) format which is a probability of presence. Median values of 12 model replicates (boostrap) are computed. The mean maximum sensitivity plus specificity (MaxTSS) Cloglog threshold was used to depict in grey values below it (0.63 in the study). Numbers in the map refer to Id number of Table 1. Country contours were downloaded from the GADM website where all data are classified as public.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Geographical characteristics of T. breyeri occurrences in Argentina and Bolivia.

ID numbers correspond to those in the MaxEnt representation of habitat suitability in Fig 3.

https://doi.org/10.1371/journal.pone.0307989.t001

It is worth noting that the three bioclimatic variables, BIO6, BIO13, and BIO17, do not have equal contributions to the model construction. BIO13 exhibits a relatively low contribution value of 0.83, whereas BIO17 and BIO6 contribute significantly with values of 41.5 and 57.6, respectively. This discrepancy is further evident in their permutation importance, with values of 1.47, 41.9, and 56.6 for BIO13, BIO17, and BIO6, respectively (S1 Table).

Species molecular identification and clustering analysis

The three gene fragments used here unambiguously characterize isolates A and B as T. breyeri. In the case of 16S and COI, it is unsurprising that isolate A and B clustered together with “isolate 56 Nb. seq. 1”, a specimen also originating from Mataral, Bolivia [14]. However, it’s worth noting that when analyzing the CytB tree, which included two insects from Argentina, these Argentinean insects unequivocally clustered with the Bolivian specimens.

The species is very close to T. eratyrusiformis and to a lesser extent to M. spinolai, as mentioned by morphological taxonomists [7] a correlation further supported by the COI and 16S trees of this study. Unfortunately, there was no CytB sequence for these last two species which would have made it possible to confirm this trend. Nevertheless, the data presented here do not allow definitive conclusions to be drawn on the exact phylogenetic position of T. breyeri.

Ecology

The consistent capture of T. breyeri across multiple sites since 2010 in the region of Estancia-Mataral–La Palma provides substantial evidence that the species has established a robust presence in the area. At the country level, the reporting of Triatomines at healthcare facilities is a common practice in Bolivia, with community reports playing a vital role in optimizing the scheduling of insecticide sprayings by the National Chagas Program. Despite this extensive community involvement and our collections, there have been no reports of T. breyeri from other regions of Bolivia. Currently it appears that the presence of this species is limited to the studied region comprised in the small valleys of Rio Chico and to a lesser extend of Rio Novillero.

The persistence of the species in this limited region has occurred despite the insecticide interventions carried out to control T. infestans. Notably, the insecticide applications primarily target domiciles and peri-domiciles, while T. breyeri is known to utilize sylvatic breeding sites, although it is also found in peri-domiciliary environments. The low prevalence of the species may also be attributed to other bio-ecological factors. In the region, T. breyeri may experience interspecific competition (although probably reduced) from T. guayazayana, and T. sordida, which are two other sylvatic/peridomestic species. In the domiciles (which are not its usual biotopes), it will encounter T. infestans, a very competitive species that is the most abundant in the region. Furthermore, T. breyeri exhibits a replacement rate of only one-third of that of T. infestans [30] which is not in favor of a rapid population growth.

The potential geographic distribution of the species, as indicated by the MaxEnt model, suggests that Bolivia may represent a habitat enclave within the species’ extreme geographical range. The best MaxEnt model predicting habitat suitability was constructed with a limited number of occurrences in Argentina. While this number is low, the number of occurrences was sufficient for accurate results [38]. Although the model appears accurate, it should be interpreted with caution, yet it provides valuable insights. The model points out that both temperature and precipitation variables play a role in the species habitat suitability, as previously mentioned [39]. In particular, within the selected set of climatic variables, the minimum temperature of the coldest month (BIO6) and the precipitations of the driest quarter (BIO17) play a predominant role. The model highlights the suitability of the "Bolivian Montane Dry Forest" ecoregion, characterized by Köppen-Geiger climate classes BSh and BSk. These classes are akin to the climate conditions found in the nearby "High Monte" ecoregion of Argentina, where T. breyeri is known to exist. Despite being categorized as "arid and steppe," the Bolivian region where T. breyeri is present exhibits a cooler climate, which might serve as a potential limiting factor for its distribution in the area. Considering all these bio-ecological factors, the species appears to persist in very low densities in Bolivia.

As of today, the species has not expanded its range to encompass the entire ’Bolivian Montane Dry Forest’ ecoregion. This limitation is likely due to the ecoregion serving as an ecological distribution boundary, as well as the species’ recent introduction and relatively slow development rate. However, there are occurrences in various distinct locations within its limited presence in Bolivia, possibly because the species exhibits a good flight dispersal capacity [30], allowing it to gradually colonize new sites.

The infrequent occurrence of T. breyeri in Bolivia and the substantial distance separating its known presence from its focal points of presence in Argentina strongly suggest that the species was likely introduced into the region relatively recently, and probably as a result of accidental introduction. Passive transport of the species by a vertebrate host (including humans) is a reasonable hypothesis [40]. As the species exhibits a high longevity and resist fasting [30], it can therefore withstand travel inside human belongings.

Field samplings of triatomines

Our samplings for triatomines in the Chuqui-Chuqui area, were conducted within multiple domiciles, with the explicit permission of each respective homeowner, all of whom were fully informed about the purpose and methods of our study.

Molecular identification

Two samples (isolate A from Estancia-Mataral and isolate B from Chuqui Chuqui) were processed to genomic DNA extraction, from one hind leg, using the Qiagen DNeasy Blood & Tissue kit according to the manufacturer instructions except that insect’s leg have been manually homogenized in 180μl of PBS 1X with a pestle and incubated 3 hours with 20μl of proteinase kinase and 200μl of AL buffer, vortexed multiple times during incubation. Amplification of the 16S ribosomal RNA, cytochrome c oxydase subunit 1 (COI) and cytochrome b (CytB) genes from genomic DNAs was carried out by simplex PCR with respective primers 16S_F (5’-GGG CTG GAA TGA AAG GTT GG-3′) and 16S_R (5′-GCT CCT GCA CCC ACA ACT TA-3′), COI_F (5’-ACG TTG ACA CAC GAG CAT AC-3’) and COI_R (5’-AAG TGT TGT GGG AAG AAT GTC A-3’), CytB_F (5’-TCT GCG ATC CCT TAC TTA GGA-3’) and CytB_R (5’-TCA GGT TGA ATG TGT ACT GGG G-3’) [13, 14, 41].

The PCR reaction mix using the Qiagen Taq DNA Polymerase kit, was prepared as follow: 1X buffer (containing 1.5mM of MgCl2), 0.5mM of MgCl2, 0.2mM of dNTP, 0.2μM of both forward and reverse primers and 1U of taq polymerase. The reaction mix was performed with 2μL of sample DNA at 50ng/μl for a final volume of 25μl. PCR reactions of three loci were performed using the same following cycling conditions: an initial denaturation at 94°C for 2 min, 10 cycles at 94°C for 30 sec, 65°C [-1°C/cycle] for 30 sec and 72°C for 30 sec, 20 cycles at 94°C for 30 sec, 55°C for 30 sec and 72°C for 30 sec, and a final extension at 72°C for 5 min. For sample A and CytB gene, a specific program was done with following cycling conditions: an initial denaturation at 94°C for 2 min, 30 cycles at 94°C for 30 sec, 55°C for 30 sec and 72°C for 30 sec, and a final extension at 72°C for 5 min.

The expected size of the PCR products (including the primers) of 369, 397 and 372 base pairs (bp) for 16S rRNA, COI and CytB genes respectively, were confirmed by electrophoresis on a 1.5% agarose gel stained with GelRed and visualized by UV transillumination.

Sequencing was performed by Eurofins Genomics sequencing services. Sequences of all amplicons (CytB, COI and 16S) were manually corrected from Sanger’ chromatograms and primers were removed. Each sequence was compared by BLAST searching against the database of Triatominae (taxid:70999) on the NCBI site. The hits returned were limited by a minimum query cover (qc) > 95% to remove sequences that were too short and by a minimum identity (id) depending on the quantity of sequences returned, in order to build a readable tree. The best model of nucleotide substitution according to the lowest BIC (Bayesian Information Criterion) score was estimated and a maximum likelihood tree (ML tree) was constructed with this best model. A phylogeny statistical test based on bootstrap [42] of 100 replicates was applied for each locus. Sequence alignments, best model determinations and ML trees were performed with MEGA7 [43].

Habitat suitability

Habitat suitability analysis for T. breyeri was conducted through the application of an ecological niche modeling approach employing the MaxEnt algorithm (ver. 3.4.4) [44, 45] and following established recommendations for good practice [46–48]. The analysis involved the following sequential steps: