Revisiting the Hybridization Processes in the Triatoma brasiliensis Complex (Hemiptera, Triatominae): Reproductive Isolation between Triatoma petrocchiae and T. b. brasiliensis and T. lenti

Simple Summary Although all triatomines are potential vectors of Chagas disease, there are species with greater or lesser vectorial importance. Therefore, the correct identification of triatomine species is essential for vector control programs. In general, triatomines are identified by external morphological characters. However, some species are very similar or even morphologically identical, making the use of complementary analyses for the correct identification of species important. For this reason, this study focused on the use of experimental crosses to assess the specific status of species of the Triatoma brasiliensis species complex. The crosses did not result in hybrids, demonstrating that there are pre-zygotic reproductive barriers installed between T. petrocchiae and the other species of the T. brasiliensis complex (which confirms the specific status of the species according to the biological species concept). On the basis of the results above, we demonstrated that T. petrocchiae is reproductively isolated from T. b. brasiliensis and T. lenti. Furthermore, we suggest that T. petrocchiae is the species most derived from the T. brasiliensis complex. Abstract Triatoma petrocchiae is a species morphologically similar to T. b. brasiliensis (which resulted in a synonymization event); despite this similarity, genetic, morphological, and experimental crossbreeding studies confirmed the specific status of T. petrocchiae. Considering that both species have been reported living in sympatry and that, for a long time, most species of the T. brasiliensis complex were considered only chromatic variants of T. b. brasiliensis, we carried out experimental crosses between T. b. brasiliensis and T. petrocchiae (to confirm whether these species are reproductively isolated) and between T. lenti and T. petrocchiae (to assess whether T. petrocchiae also presents prezygotic isolation with the other species of the T. brasiliensis complex). Reciprocal experimental crosses were conducted, and weekly, the eggs were collected, counted, and separated in new containers to assess the hatch rate. Neither cross resulted in hybrids, demonstrating that there are pre-zygotic reproductive barriers installed between T. petrocchiae and the other species of the T. brasiliensis complex. On the basis of the results above, we demonstrated that T. petrocchiae is reproductively isolated from T. b. brasiliensis and T. lenti. Furthermore, we suggest that T. petrocchiae is the species most derived from the T. brasiliensis complex.

The genus Triatoma Laporte, 1832 is the most representative (81 species) and the most morphologically diversified [6,7]. This genus is paraphyletic [8,9], and species are grouped into complexes and subcomplexes [9][10][11][12]. The T. brasiliensis complex is a grouping of endemic species from Brazil [13] composed of six species and two subspecies that share a common ancestry: T. b. brasiliensis Neiva, 1911, T. b. macromelasoma Galvão, 1965 Lent, 1941 [14-18]. The last taxon grouped in this complex was T. petrocchiae [18], a species reported in the states of Bahia, Ceará, Pernambuco, Paraíba, and Rio Grande do Norte [13,19]; however, the potential distribution map published by Caranha et al. [20] suggests that this species could also be found in the states of Piauí, Alagoas, and Sergipe, where the species has not been recorded to date.
The species T. brasiliensis sensu stricto is currently divided into two subspecies (T. b. brasiliensis and T. b. macromelasoma), which can be differentiated by morphological characters: T. b. brasiliensis presents a pronotum with 1 + 1 brownish-yellow areas extending from the posterior portion of the anterior lobe to the posterior lobe, femora with broad brownish-yellow rings, and membrane of hemelytra with a lumen of cells that are not darkened; T. b. macromelasoma presents a pronotum with 1 + 1 narrow brownish-yellow stripes on the submedian carinae, not attaining its apex, legs with an incomplete brownishyellow ring on the apical half of the femora, and hemelytra with membrane cells that are darkened on the central portion [21]. In addition to the phenotypic divergences, these species have a different geographic distribution: while T. b. brasiliensis has been noted in the states of Ceará, Maranhão, Paraíba, Piauí, and Rio Grande do Norte, T. b. macromelasoma is endemic to Pernambuco [13,19].
Triatoma petrocchiae is a species morphologically similar to T. b. brasiliensis, which led Lucena [22] to propose the synonymization of species, considering T. petrocchiae only as a chromatic variant of T. b. brasiliensis. However, Espínola [23] carried out experimental crosses between T. b. brasiliensis and T. petrocchiae from Paulo Afonso, Bahia, Brazil, and observed that these species did not produce viable hybrids. On the basis of this, Lent and Wygodzinsky [24] revalidated the specific status of T. petrocchiae from morphological data (the status was corroborated with genetic analyses using allozyme electrophoresis [25]).
The interspecific crosses performed by Espínola [23] were proposed because chromatic variations were observed in the populations of T. b. brasiliensis from Paulo Afonso, Bahia. The authors indicated that there are similarities in the coloration between T. b. brasiliensis and T. petrocchiae and, above all, these species share the same ecological niche. Considering that these species have been reported living in sympatry [26,27] and that, in 1971, most species of the T. brasiliensis complex were still considered only chromatic variants of T. b. brasiliensis, there is a need to confirm whether T. b. brasiliensis and T. petrocchiae are really reproductively isolated (mainly because all other species in this complex are capable of producing hybrids [15,[28][29][30][31][32]). On the basis of the assumptions above, we carried out experimental crosses between T. b. brasiliensis and T. petrocchiae (to corroborate the results of Espínola [23]) and between T. lenti and T. petrocchiae (to assess whether T. petrocchiae also presents prezygotic isolation with the other species of the T. brasiliensis complex).

Materials and Methods
Reciprocal experimental crosses were conducted between T. b. brasiliensis (from Currais Novos (Pedra do Sino), Rio Grande do Norte, Brazil, collected in wild ecotopes (geographic coordinates: 6 • 17 06.8 S 36 • 29 51.9 W)) and T. petrocchiae (from Caicó, Rio Grande do Norte, Brazil, collected in wild ecotopes (geographic coordinates: 6 • 27 47.8 S 37 • 09 11.3 W)) and between T. lenti (from Macaúbas, Bahia, Brazil, collected in peridomiciliary ecotopes (geographic coordinates: 13 • 11 25.7 S 42 • 31 56.3 W)) and T. petrocchiae ( Figure 1). The insects used in the experiment came from colonies kept in the Triatominae insectary of the School of Pharmaceutical Sciences, São Paulo State University (UNESP), Araraquara, São Paulo, Brazil. The experimental crosses were conducted in the Triatominae insectary, according to the experiments of Mendonça et al. [30], Neves et al. [33], and Pinotti et al. [32]: the insects were sexed as 5th instar nymphs [34], and males and females were kept separately until they reached the adult stage to guarantee the virginity of the insects used in the crosses. For the experimental crosses, three couples from each set were placed in plastic jars (diameter 5 cm × height 10 cm) (each couple in a jar) and kept at room temperature (average of 24 • C [35]) and an average relative humidity of 63% [35]). Weekly, the couples were fed on duck blood, and the eggs were collected, counted, and separated into new containers to assess the hatch rate.

Results and Discussion
The experimental crosses between T. petrocchiae and T. b. brasiliensis did not result in hybrids (Table 1), demonstrating that there are pre-zygotic reproductive barriers installed between these species (confirming the specific status of T. petrocchiae according to the biological species concept [36,37]). These results obtained for the cross between T. b. brasiliensis and T. petrocchiae from Rio Grande do Sul (the state where the species were also collected in the same rock outcrop spot [38]) agree with those obtained by Espínola [23] when

Results and Discussion
The experimental crosses between T. petrocchiae and T. b. brasiliensis did not result in hybrids (Table 1), demonstrating that there are pre-zygotic reproductive barriers installed between these species (confirming the specific status of T. petrocchiae according to the biological species concept [36,37]). These results obtained for the cross between T. b. brasiliensis and T. petrocchiae from Rio Grande do Sul (the state where the species were also collected in the same rock outcrop spot [38]) agree with those obtained by Espínola [23] when they crossed specimens from Paulo Afonso, Bahia.
Whereas in 1971, the current species T. melanica, T. sherlocki, and T. juazeirensis were considered only phenotypic variants and/or subspecies of T. b. brasiliensis [14,[39][40][41][42], we conducted a survey of the literature on the triatomine already noted in Paulo Afonso, Bahia to confirm which species Espínola [22] had crossed with T. petrocchiae, and we observed that only T. b. brasiliensis and T. petrocchiae were the species of the T. brasiliensis complex notified for the municipality [43]. Furthermore, to ensure that the T. brasiliensis complex triatomines collected in Paulo Afonso, Bahia, were correctly identified as T. b. brasiliensis and T. petrocchiae, we evaluated some specimens collected in this municipality that were deposited in the entomologic collections of the Faculty of Public Health of the University of Sao Paulo, Brazil (Figure 2). On the basis of this information, we confirmed that the specimens used in the experiments of Espínola [23] were T. b. brasiliensis. Although Espínola [23] suggested that T. petrocchiae and T. b. brasiliensis share the same ecological niche, Liloso et al. [27] recently demonstrated that while T. b. brasiliensis is mainly associated with rodents, the food sources of T. petrocchiae were strongly associated with reptiles of the Tropidurus and Hemidactylus genera; this suggests that T. petrocchiae is the single member within this complex that is associated with reptiles, indicating a distinct niche occupation related to the trophic resources. These results point to the Although Espínola [23] suggested that T. petrocchiae and T. b. brasiliensis share the same ecological niche, Liloso et al. [27] recently demonstrated that while T. b. brasiliensis is mainly associated with rodents, the food sources of T. petrocchiae were strongly associated with reptiles of the Tropidurus and Hemidactylus genera; this suggests that T. petrocchiae is the single member within this complex that is associated with reptiles, indicating a distinct niche occupation related to the trophic resources. These results point to the possible presence of a prezygotic reproductive barrier due to ecological isolation between T. petrocchiae and members of the T. brasiliensis complex. However, other possible prezygotic barriers cannot be ruled out, such as mechanical isolation, as the morphological analysis of the external female genitalia evidenced some unique characteristics for T. petrocchiae [44].
Experimental crosses between T. petrocchiae and T. lenti also did not result in hybrids (Table 1). Unlike T. b. brasiliensis, which cohabits rock outcrops with T. petrocchiae [27,38], there are no reports of T. petrocchiae and T. lenti living in sympatry (on the contrary, they inhabit municipalities in the state of Bahia that are at least 800 km away [43]). This result is in accordance with what was proposed by Oliveira et al. [18], which suggests that T. petrocchiae is the most distant species from the T. brasiliensis complex. The genomic incompatibility resulting in the inability to produce hybrids with T. lenti points to the hypothesis that T. petrocchiae was possibly the first species to be derived from the common ancestor of the T. brasiliensis complex (since all other species in the complex produce hybrids [15,[28][29][30][31][32]).
The prezygotic isolation observed between T. petrocchiae and species of the T. brasiliensis complex was only observed when members of this complex (T. b. brasiliensis) were crossed with other subcomplexes, such as T. sordida [45], T. infestans [45], and T. vitticeps subcomplexes [33] (which are species phylogenetically distant from T. b. brasiliensis [8,9]). The reproductive barrier that possibly prevents hybrids between other species of the T. brasiliensis complex is based on post-zygotic reproductive isolation due to hybrid collapse (as noted by Mendonça et al. [30] and Alevi et al. [31]). It was believed that T. melanica was the most differentiated form of the complex [37]; however, according to the results of experimental crosses and the high genetic distance observed between T. petrocchiae and members of the T. brasiliensis complex [46], the most differentiated species from a genetic point of view is T. petrocchiae.
Before concluding, it is worth mentioning that the low number of eggs produced by crosses can be a limiting factor for the research. Furthermore, it is important that new crosses between T. petrocchiae and all members of the T. brasiliensis complex be carried out to confirm that this species is indeed reproductively isolated from all other species in the complex or whether prezygotic isolation is restricted to T. b. brasiliensis and T. lenti (as noted by Espínola et al. [23] and in the present manuscript).

Conclusions
On the basis of the results above, we demonstrated that T. petrocchiae is reproductively isolated from T. b. brasiliensis and T. lenti (confirming the specific status of T. petrocchiae). Furthermore, we demonstrated that these species have prezygotic reproductive isolation and suggest that T. petrocchiae is the species most derived from the T. brasiliensis complex.