Introduction
More than 30 species of triatomines have been collected in Mexico, including six Meccus species (Carabarin-Lima et al., Reference Carabarin-Lima, González-Vázquez, Rodríguez-Morales, Baylón-Pacheco, Rosales-Encina, Reyes-López and Arce-Fonseca2013). Together with the four other Meccus species (Hemiptera: Reduviidae), Meccus pallidipennis (Stål), Meccus longipennis (Usinger) and Meccus picturatus (Usinger), are thought to be the dominant Chagas disease vector species in Mexico, accounting for 74% of the vectorial transmission of Trypanosoma cruzi (Trypanosomatida: Trypanosomatidae), the causative agent of Chagas disease (Ibarra-Cerdeña et al., Reference Ibarra-Cerdeña, Sánchez-Cordero, Townsend-Peterson and Ramsey2009). Stål placed the first representatives of the group in the genus Meccus. However, in the middle of the 20th century, all the five described species were moved to the Triatoma genus, which currently includes additional Triatominae species (Lent & Wygodzinsky, Reference Lent and Wygodzinsky1979). Some years later, Carcavallo et al. (Reference Carcavallo, Jurberg, Lent, Noireau and Galvão2000) revalidated the genus Meccus, taking into account morphological characteristics such as the structure and shape of the testicles. Recent molecular evidence (Martínez et al., Reference Martínez, Alejandre-Aguilar, Hortelano-Moncada and Espinoza2005, Reference Martínez, Villalobos, Ceballos, de la Torre, Laclette, Alejandre-Aguilar and Espinoza2006; Bargues et al., Reference Bargues, Klisiowicz, González-Candelas, Ramsey, Monroy, Ponce, Salazar-Schettino, Panzera, Abad-Franch, Souza, Schofield, Dujardin, Guhl and Mas-Coma2008, Reference Bargues, Zuriaga and Mas-Coma2014; Espinoza et al., Reference Espinoza, Martínez-Ibarra, Villalobos, de la Torre, Laclette and Martínez2013) has supported that revalidation, and all groups in this study will be considered members of the genus Meccus. Furthermore, another discussion involving these species focuses on their proper taxonomic range. More than 60 years ago, most species except for the sixth, then un-described member Meccus bassolsae (Alejandre-Aguilar, Nogueda-Torres, Cortez-Jiménez, Jurberg, Galvão, Carcavallo) were ranked as subspecies of Meccus phyllosomus (Mazzotti & Osorio, 1942). However, Lent & Wygodzinsky (Reference Lent and Wygodzinsky1979) reinstated the other five as bona fide species based entirely on morphological characters. Since then, an argument about their proper taxonomic rank has persisted. Mayr & Diamond (Reference Mayr and Diamond2001) stated that ‘subspecies are local populations that are recognizably different from each other but, nevertheless, are considered to belong to the same species, because they are observed to interbreed in nature or because it is inferred that they are likely to interbreed’. This definition fits with some studies on the reproductive behaviour of M. longipennis, M. picturatus, and M. pallidipennis (Martínez-Ibarra et al., Reference Martínez-Ibarra, Salazar-Schettino, Nogueda-Torres, Vences, Tapia-González and Espinoza-Gutiérrez2009). Because recent biological, morphological, and molecular evidence (Martínez et al., Reference Martínez, Villalobos, Ceballos, de la Torre, Laclette, Alejandre-Aguilar and Espinoza2006; Bargues et al., Reference Bargues, Klisiowicz, González-Candelas, Ramsey, Monroy, Ponce, Salazar-Schettino, Panzera, Abad-Franch, Souza, Schofield, Dujardin, Guhl and Mas-Coma2008, Reference Bargues, Zuriaga and Mas-Coma2014; Martínez-Ibarra et al., Reference Martínez-Ibarra, Ventura-Rodríguez, Meillon, Barajas-Martínez, Alejandre-Aguilar, Lupercio-Coronel, Rocha-Chávez and Nogueda-Torres2008b , Espinoza et al., Reference Espinoza, Martínez-Ibarra, Villalobos, de la Torre, Laclette and Martínez2013) has supported the rank of subspecies for M. pallidipennis, M. longipennis, and M. picturatus, they will be considered subspecies in this study. For many years, arguments about the differences between Meccus phyllosomus pallidipennis, Meccus phyllosomus longipennis, and Meccus phyllosomus picturatus have continued. These three groups are important vectors of T. cruzi in Mexico, however they have shown important differences among them in studied biological parameters related to percentage of egg eclosion, fecundity, egg-to-adult development time, feeding, and defecation patterns (Martínez-Ibarra et al., Reference Martínez-Ibarra, Nogueda-Torres, García-Benavídez, Vargas-Llamas, Bustos-Saldaña and Montañez-Valdez2012, Reference Martínez-Ibarra, Nogueda-Torres, Licón-Trillo, Villagrán-Herrera, de Diego-Cabrera, Montañez-Valdez and Rocha-Chávez2013, Reference Martínez-Ibarra, Nogueda-Torres, del Toro-González, Ventura-Anacleto and Montañez-Valdez2015d ). Those differences lead to important differences in their capacity as vectors of T. cruzi to human beings and, as a consequence, on their epidemiological importance in Mexico. Therefore, previously studied biological parameters have shown that M. p. pallidipennis is a more effective transmitter of T. cruzi than M. p. longipennis and M. p. picturatus, whereas this last species is the least effective of these three.
These three subspecies have been inter-crossed with each other under laboratory and wild conditions, and fertile hybrids have been obtained (Martínez-Ibarra et al., Reference Martínez-Ibarra, Ventura-Rodríguez, Meillon, Barajas-Martínez, Alejandre-Aguilar, Lupercio-Coronel, Rocha-Chávez and Nogueda-Torres2008b , Reference Martínez-Ibarra, Salazar-Schettino, Nogueda-Torres, Vences, Tapia-González and Espinoza-Gutiérrez2009). Some triatomine laboratory hybrids have displayed intermediate characteristics or have shown outstanding biological parameters that may confer higher fitness than their parental species (Almeida et al., Reference Almeida, Oliveira and Galvão2012; Chávez-Contreras et al., Reference Chávez-Contreras, De la Torre-Álvarez, Cárdenas-Barón, Aguilar-López, Jiménez-Íñiguez and Martínez-Ibarra2013, Martínez-Ibarra et al., Reference Martínez-Ibarra, Nogueda-Torres, García-Lino, Arroyo-Reyes, Salazar-Montaño, Hernández-Navarro, Díaz-Sánchez, del Toro-Arreola and Rocha-Chávez2015b , Reference Martínez-Ibarra, Nogueda-Torres, Salazar-Montaño, García-Lino, Arroyo-Reyes and Hernández-Navarro c ). Wild hybrids have also been associated with resistance to insecticides or have shown higher entomological indices than ‘pure’ species collected (Martínez-Ibarra et al., Reference Martínez-Ibarra, Grant-Guillén, Morales-Corona, Haro-Rodriguez, Ventura-Rodríguez, Nogueda-Torres and Bustos-Saldaña2008a ; Mas-Coma & Bargues, Reference Mas-Coma and Bargues2009). In contrast, some hybrids have had reduced fitness or viability compared with their parental lines (Herrera-Aguilar et al., Reference Herrera-Aguilar, Be-Barragán, Ramírez-Sierra, Tripet, Dorn and Dumonteil2009).
Anthropogenic change and landscape heterogeneity may modulate T. cruzi transmission risk. These new ecological scenarios might facilitate endemic disease emergence, and create new suitable environments for integration and mating between species, which may potentially result in natural hybrids. Because the consequences of this natural hybridization are unknown, the need for further evaluation of hybrid fitness is imperative (Correia et al., Reference Correia, Almeida and Lima-Neiva2013).
In Mexico, the geographical distribution of the subspecies of M. phyllosomus currently in the domestic environment would be the result of the colonization of the sylvatic populations present in the natural surrounding environment (Breniere et al., Reference Breniere, Bosseno, Magallón-Gastélum, Castillo-Ruvalcaba, Soto-Gutiérrez, Montaño-Luna, Tejeda-Basulto, Mathieu-Daudé, Walter and Lozano-Kasten2007). Some subspecies of M. phyllosomus have been sympatrically collected in different areas of Mexico because of anthropogenic and environmental changes. In western and central Mexico, M. longipennis, M. p. picturatus, and M. p. pallidipennis have been repeatedly sympatrically collected (including some hybrids) colonizing agro-pastoral environments recently created by deforestation. The most frequently colonized micro-habitats have been chicken roosts, pigsties and stone fences used habitually as borders for fields by rural inhabitants (Espinoza-Gómez et al., Reference Espinoza-Gómez, Maldonado-Rodríguez, Coll-Cárdenas, Hernández-Suárez and Fernández-Salas2002; Magallón-Gastélum et al., Reference Magallón-Gastélum, Lozano-Kasten, Bosseno, Cárdenas-Contreras, Ouaissi and Breniere2004, Reference Magallón-Gastélum, Lozano-Kasten, Gutiérrez, Flores-Pérez, Sánchez, Espinoza, Bosseno and Breniere2006; López-Cárdenas et al., Reference López-Cárdenas, González-Bravo, Salazar-Schettino, Gallaga-Solórzano, Ramírez-Barba, Martínez-Méndez, Sánchez-Cordero, Townsend-Peterson and Ramsey2005; Martínez-Ibarra et al., Reference Martínez-Ibarra, Bárcenas-Ortega, Nogueda-Torres, Alejandre-Aguilar, Rodríguez, Magallón-Gastélum, López-Martínez and Romero-Nápoles2001, Reference Martínez-Ibarra, Grant-Guillén, Morales-Corona, Haro-Rodriguez, Ventura-Rodríguez, Nogueda-Torres and Bustos-Saldaña2008a , Reference Martínez-Ibarra, Salazar-Schettino, Nogueda-Torres, Vences, Tapia-González and Espinoza-Gutiérrez2009; Bosseno et al., Reference Bosseno, Barnabé, Ramírez-Sierra, Kegne, Guerrero, Lozano-Kasten, Magallón-Gastélum and Breniere2009). In southern Mexico Meccus phyllosomus mazzottii, M. p. pallidipennis and M. p. phyllosomus have been sympatrically collected in stone and wood fences, in firewood piles and inside brick and cement houses (Ramsey et al., Reference Ramsey, Ordóñez, Cruz-Celis, Alvear, Chávez, López, Pintor, Gama and Carrillo2000; Rodríguez-Bataz et al., Reference Rodríguez-Bataz, Nogueda-Torres, Rosario-Cruz, Martínez-Ibarra and Rosas-Acevedo2011). All those studied subspecies of M. phyllosomus have been reported as feeding primarily on the abundant human beings, dogs, and rodents (Bosseno et al., Reference Bosseno, Santos-García, Baunaure, Magallón-Gastélum, Soto-Gutiérrez, Lozano-Kasten, Dumonteil and Breniere2006; Mota et al., Reference Mota, Chacón, Gutiérrez, Sánchez-Cordero, Wirtz, Ordóñez, Panzera and Ramsey2007; Rabinovich et al., Reference Rabinovich, Kitron, Obed, Yoshioka, Gottdenker and Chávez2011; Ibáñez-Cervantes et al., Reference Ibáñez-Cervantes, Martínez-Ibarra, Nogueda-Torres, López-Orduña, Alonso, Perea, Maldonado and León-Avila2013). These findings fit with the hypothesis that rather than innate preferences for host species, host use by kissing bugs is influenced by the habitats they colonize. Therefore, it has been established that host accessibility is a major factor that shapes the blood-foraging patterns of kissing bugs (Rabinovich et al., Reference Rabinovich, Kitron, Obed, Yoshioka, Gottdenker and Chávez2011).
This study comparing the biological characteristics of M. p. pallidipennis, M. p. longipennis, and M. p. picturatus hybrids and their parental lines, was conducted as a first step in the assessment of the epidemiological importance of these distinct groups.
Material and methods
Biological material
The individuals used in crossing experiments were obtained from the third generation of previously established Triatominae colonies that originated in two non-overlapping areas. A laboratory colony of M. p. pallidipennis that was established in 2012 from 41 specimens collected in Amilcingo (18°50′N, 98°49′W), Morelos, Mexico was used. A colony of M. p. longipennis established in 2012 from 27 specimens from El Jocuixtle, Durango (23°25′N, 105°34′W) and a colony of M. p. picturatus established in 2012 from 26 specimens from Juan Gil Preciado (19°36′N, 105°02′W) also were used. When initially collected, founders of each colony were identified following the Lent & Wygodzinsky (Reference Lent and Wygodzinsky1979) keys, the revalidation of Meccus also was taken into account (Carcavallo et al., Reference Carcavallo, Jurberg, Lent, Noireau and Galvão2000; Bargues et al., Reference Bargues, Zuriaga and Mas-Coma2014) and the specimens corresponded to the typical morphological characteristics of each species. Colonies were maintained at 27 ± 1°C and 75 ± 5% relative humidity (RH) and under a 12/12 h (light/dark) regimen, similar to the laboratory conditions used in previously published studies that investigated the biology of the three subspecies (referred to as M. pallidipennis, M. longipennis, and M. picturatus) (Martínez-Ibarra et al., Reference Martínez-Ibarra, Nogueda-Torres, García-Benavídez, Vargas-Llamas, Bustos-Saldaña and Montañez-Valdez2012, Reference Martínez-Ibarra, Nogueda-Torres, Licón-Trillo, Villagrán-Herrera, de Diego-Cabrera, Montañez-Valdez and Rocha-Chávez2013, Reference Martínez-Ibarra, Nogueda-Torres, del Toro-González, Ventura-Anacleto and Montañez-Valdez2015a ). Specimens were fed on immobilized and anesthetized (using 0.25 ml kg−1 of ketamine that was injected intramuscularly) New Zealand rabbits for a 1-h period on a fortnightly basis. Rabbits were maintained under laboratory conditions (of space, food, water, and cleanliness) and were handled and anaesthetized following Norma Oficial Mexicana NOM-062-ZOO-1999, Especificaciones técnicas para la producción, cuidado y uso de los animales de laboratorio (Technical guidelines for production, care, and use of laboratory animals) regulations (SAGARPA, 1999). Observance of the NOM-062-ZOO-1999 was fulfilled by the head of the Committee of Ethical Behaviour of the Centro Universitario del Sur.
Crossing experiments
To conduct the reciprocal experimental crosses in this study, 10 pairs from the following sets were placed in plastic jars (5 cm diameter × 10 cm height): (1) M. p. pallidipennis female and M. p. longipennis male, (2) M. p. longipennis female and M. p. pallidipennis male, (3) M. p. longipennis female and M. p. picturatus male, (4) M. p. picturatus female and M. p. longipennis male, (5) M. p. pallidipennis female and M. p. picturatus male, and (6) M. p. picturatus female and M. p. pallidipennis male. The three parental lineages involved in the study also were used as controls: (7) M. p. pallidipennis female and M. p. pallidipennis male, (8) M. p. longipennis female and M. p. longipennis male, and (9) M. p. picturatus female and M. p. picturatus male. Offspring of interspecific crosses were considered hybrids based on the definition of a hybrid ‘as the product of the crossing of individuals belonging to two unlike natural populations, principally different species’ (Mayr & Ashlock, Reference Mayr and Ashlock1991).
Differences among subspecies
Specimens were maintained as previously described. To record fecundity, all crosses were checked daily for spermatophore elimination and copulation events. To assess egg fertility, eggs from each cross were collected for 30 days and incubated under previously described laboratory conditions.
Eggs from couples in each cohort were grouped by the date of oviposition for 1 week to initiate a cohort of 200 eggs each. Following eclosion, groups of first instar nymphs were separated by cohort individually into plastic containers (5.5 cm diameter × 10.5 cm height) with a centre support of absorbent cardboard. Three days after eclosion and every 2 weeks after, each cohort of nymphs were fed individually on New Zealand rabbits as previously described. The bugs were maintained in a dark incubator at 27 ± 1°C and 75 ± 5% RH under a 12/12 h (light/dark) regimen, and were checked daily for ecdysis or death. From the insects that completed development into adults, 10 adult couples from each cohort were placed in individual containers (5 cm diameter × 10 cm height), and were maintained as previously described to determine oviposition patterns.
In order to estimate biological fitness (individual health), all biological parameters of each hybrid cohort were compared with those obtained from the reciprocal cross and with those from the two parental lines.
Statistical analyses
Variables with a normal distribution were compared using Student's t-test or an analysis of variance. A non-parametric Kruskal–Wallis test was used to compare the developmental cycle periods and the number of blood meals to moult in the cohorts. Moreover, the Holm-Sidak method was used to compare the number of eggs laid per female. Pairwise comparisons were performed for comparisons among the studied subspecies using Dunn's method, and the chi-square test was used to compare frequencies. Sigma Stat 3.1 software (version 3.1 for Windows, Systat Software Inc., San Jose, CA) was used for statistical analysis. Results were considered to be statistically different when P < 0.05.
Results
Differences among subspecies
The egg eclosion rate was variable, mostly over 69%, except for M. p. longipennis × M. p. picturatus cohorts and M. p. longipennis parents. Non-significant differences (X 2 = 0.06, df = 2, P > 0.81) were recorded when the M. p. longipennis × M. p. pallidipennis cohort egg eclosion rate and that of its reciprocal cross were compared. Similarly, non-significant differences (P > 0.05) were recorded when each set of crosses was compared with its reciprocal cross. On the other hand, both M. p. longipennis × M. p. pallidipennis cohorts from reciprocal crosses had significantly (X 2 = 38.87, df = 2, P < 0.001) higher egg eclosion rates than M. p. longipennis parental line, but significantly (X 2 = 36.05, df = 2, P < 0.001) lower egg eclosion rates than M. p. pallidipennis. Significantly (X 2 = 40.41, df = 2, P < 0.001) lower egg eclosion rates also were detected when both sets of M. p. longipennis × M. p. picturatus cohorts were compared with the M. p. longipennis and M. p. picturatus parental lines (X 2 = 43.37, df = 2, P < 0.0000001). Likewise, significantly lower (X 2 = 37.80, df = 2, P < 0.001) rates were detected when M. p. pallidipennis × M. p. picturatus cohorts were compared with the M. p. pallidipennis parental line, but no significant differences (P > 0.05) were found when the hybrid was compared with the M. p. picturatus parental line (table 1). The average incubation period was approximately 19 days for all cohorts (data not shown).
Percentage of egg eclosion, time of development (mean ± SD) from egg to adult, blood meals to moult (mean ± SD), percentage of accumulative mortality, percentage of obtained females and number of laid eggs (mean ± SD) of Meccus phyllosomus longipennis, Meccus phyllosomus picturatus, Meccus. phyllosomus pallidipennis and their hybrids.
LoPa = M. p. longipennis × M. p. pallidipennis, LoPi = M. p. longipennis × M. p. picturatus, PaPi = M. p. pallidipennis × M. p. picturatus, Pa = M. p. pallidipennis, Lo = M. p. longipennis, Pi = M. p. picturatus.
Means in rows followed by the same letters are not significantly different (P < 0.05).
1 ♀M. p. pallidipennis.
2 ♀M. p. longipennis.
3 ♀M. p. picturatus.
The average egg-to-adult development time was highly variable, but was generally longer in the hybrid cohorts. Non-significant differences (Q = 1.66, df = 2, P > 0.05) were recorded when the M. p. longipennis × M. p. pallidipennis cohorts and reciprocal crosses were compared (table 1). Furthermore, non-significant differences (P > 0.05) were found when each set of crosses was compared with its reciprocal cross. Significantly longer average egg-to-adult development times (Q = 6.03, 4.99, df = 2, P < 0.05) were recorded for M. p. longipennis × M. p. pallidipennis cohorts compared with the M. p. longipennis and M. p. pallidipennis parental crosses. Similarly, significantly longer average egg-to-adult development times (Q = 6.71, 5.56, df = 2, P < 0.05) were found when M. p. longipennis × M. p. picturatus cohorts were compared with M. p. longipennis and M. p. picturatus parental crosses. However, non-significant differences (Q = 2.80, 2.66, df = 2, P > 0.05) were recorded when M. p. pallidipennis × M. p. picturatus cohorts and the reciprocal crosses were compared with the M. p. picturatus and M. p. pallidipennis parental cohorts (table 1).
The average number of blood meals to moult to the next instar did not differ significantly (Q = 1.12, df = 2, P > 0.05) when the hybrid cohorts from each cross and the reciprocal crosses were compared. Interestingly, hybrid cohorts from crosses involving M. p. longipennis (M. p. longipennis × M. p. pallidipennis and M. p. longipennis × M. p. picturatus) required a significantly lower total number of meals (Q = 6.29, 4.25, df = 2, P < 0.05) to moult through each nymphal instar until the adult stage as compared with the M. p. longipennis parental cohort. In contrast, non-significant differences (Q = 2.46, 1.34, df = 2, P > 0.05) were recorded between the M. p. pallidipennis and M. p. picturatus parental lines. Regarding the M. p. pallidipennis × M. p. picturatus cohorts, they were only significantly (Q = 6.71, 6.66, df = 2, P < 0.05) lower in blood meals to moult to the next instar compared with the M. p. pallidipennis parental cohort (table 1).
Accumulative mortality was similar (X 2 = from 0.001 to 0.01, df = 2, P > 0.05) when each hybrid cohort from reciprocal crosses was compared. The percentage mortality of M. p. longipennis × M. p. pallidipennis cohorts was significantly (X 2 = 4.22, 4.24, df = 2, P < 0.05) lower than the M. p. pallidipennis parental line, but those hybrid cohorts were similar (X 2 = 0.001, df = 2, P > 0.05) to the M. p. longipennis parental line. Additionally, M. p. longipennis × M. p. picturatus cohorts had significantly higher mortality percentages (X 2 = 7.75, 8.1, df = 2, P < 0.05) than the M. p. longipennis parental line, and did not differ significantly (X 2 = 2.60, 2.69, df = 2, P > 0.05) from the M. p. picturatus parental line. Lastly, M. p. pallidipennis × M. p. picturatus cohorts (X 2 = 1.61, 1.7, df = 2, P > 0.05) were similar to the M. p. pallidipennis parental line, but showed significantly lower mortality percentages (X 2 = 5.9, 6.3, df = 2, P < 0.05) than the M. p. picturatus parental line (table 1).
Non-significant differences (X 2 = from 0.1 to 1.34, df = 2, P > 0.05) were found when the percentages of obtained females at the end of the life cycles of all cohorts were compared (table 1).
The mean number of eggs laid per female was similar (t = 0.322; df = 2, P > 0.05) for each hybrid cohort and its reciprocal. Moreover, M. p. longipennis × M. p. pallidipennis and M. p. pallidipennis × M. p. picturatus cohorts shown significantly (t = 4.75, 5.69, df = 2, P = 0.003, 0.002) lower mean number of eggs when compared with the M. p. pallidipennis parental cohort). In contrast, the differences between M. p. longipennis × M. p. pallidipennis and M. p. longipennis were not significant (t = 0.435, df = 2, P > 0.05) and the same was true for those of M. p. pallidipennis × M. p. picturatus as compared with M. p. picturatus (t = 1.17, df = 2, P > 0.05). Lastly, when M. p. longipennis × M. p. picturatus cohorts were compared with both parental cohorts, no significant differences (t = 0.233, 0.245, df = 2, P > 0.940, 0.915) were detected (table 1).
Discussion
The recorded egg eclosion rates of M. p. longipennis × M. p. pallidipennis and M. p. pallidipennis × M. p. picturatus cohorts are comparable with those of various species of triatomines (e.g. T. ryckmani Zeledón and Ponce, T. juazeirensis Costa and Felix, and T. patagonica Del Ponte) (Zeledón et al., Reference Zeledón, Cordero, Marroquín and Seixas-Lorosa2010; Lima-Neiva et al., Reference Lima-Neiva, Gumiel, Lima, Monte-Gonçalves, Provance, Almeida and Costa2012; Nattero et al., Reference Nattero, Rodríguez and Crocco2013), as well as to that of M. p. longipennis identified in a previous study (Martínez-Ibarra et al., Reference Martínez-Ibarra, Nogueda-Torres, Licón-Trillo, Villagrán-Herrera, de Diego-Cabrera, Montañez-Valdez and Rocha-Chávez2013). Moreover, the average incubation period was approximately 19 days, reflecting the favourable maintenance conditions for the development of these subspecies and hybrids. In the descendant cohorts of crosses where M. p. pallidipennis was involved (M. p. longipennis × M. p. pallidipennis and M. p. pallidipennis × M. p. picturatus), egg eclosion rates were lower than that recorded for M. p. pallidipennis, which indicates reduced hybrid fitness. In the case of M. p. longipennis × M. p. picturatus cohorts, reduced hybrid fitness also was recorded, constituting a postzygotic barrier that maintains the reproductive isolation of parental genotypes, which is similar to that observed with T. dimidiata (Latreille) hybrids in Yucatán, México (Herrera-Aguilar et al., Reference Herrera-Aguilar, Be-Barragán, Ramírez-Sierra, Tripet, Dorn and Dumonteil2009).
The average egg-to-adult development time for the three parental cohorts and the M. p. pallidipennis × M. p. picturatus cohorts was between five and five and a half months. However, the development time varied from seven to seven and a half months for the M. p. longipennis × M. p. pallidipennis and M. p. longipennis × M. p. picturatus cohorts. The longer average egg-to-adult development time for M. p. longipennis × M. p. pallidipennis and M. p. longipennis × M. p. picturatus cohorts in comparison with the three parental lines reflects the lower fitness of the hybrid cohorts.
Specimens of the M. p. longipennis × M. p. pallidipennis, M. p. longipennis × M. p. picturatus, and M. p. pallidipennis × M. p. picturatus cohorts required a lower number of blood meals to moult than the M. p. longipennis and M. p. pallidipennis parental lines, which suggests higher fitness. This would be an advantage for the hybrids since every triatomine may be at risk each time it leaves its shelter to find a host.
Accumulative mortality of the M. p. pallidipennis cohort was lower than M. p. longipennis × M. p. pallidipennis and M. p. pallidipennis × M. p. picturatus cohorts, but the M. p. picturatus cohort had lower mortality than the M. p. longipennis × M. p. picturatus cohort. This parameter also suggests a lack of hybrid fitness. As reported for M. p. longipennis and M. p. pallidipennis (Martínez-Ibarra et al., Reference Martínez-Ibarra, Nogueda-Torres, García-Benavídez, Vargas-Llamas, Bustos-Saldaña and Montañez-Valdez2012, Reference Martínez-Ibarra, Nogueda-Torres, Licón-Trillo, Villagrán-Herrera, de Diego-Cabrera, Montañez-Valdez and Rocha-Chávez2013), death in the youngest nymphs seemed to be caused by the feeding incapacity of insects because dead triatomines were generally found without substantial intestinal content. On the other hand, death of older nymphs appeared to occur during moulting.
The percentage of obtained females at the end of the life cycles was similar among all cohorts. However, the M. p. pallidipennis cohort laid about four times as many eggs compared with the M. p. longipennis × M. p. pallidipennis and M. p. pallidipennis × M. p. picturatus cohorts. More females laying many eggs indicate a greater possibility of having a larger population of triatomines, which might result in a highly successful population. The results presented here indicate that, for the experimental conditions used, in four of the five studied parameters (with the exception of the percentage of obtained females), at least one of the parental cohorts involved in each set of crosses had better fitness results than their hybrid descendants. Maybe under different experimental conditions (e.g., meal source, feeding frequency) hybrids may have an advantage over parental cohorts because of fitness plasticity, such as has been recorded in different cohorts of Rhodnius prolixus Stål (Rodríguez & Rabinovich, Reference Rodríguez and Rabinovich1980; Sulbaran & Chaves, Reference Sulbaran and Chaves2006).
Various studies have established that M. p. longipennis, M. p. picturatus, and M. p. pallidipennis subspecies are almost genetically identical (Martínez et al., Reference Martínez, Alejandre-Aguilar, Hortelano-Moncada and Espinoza2005, Reference Martínez, Villalobos, Ceballos, de la Torre, Laclette, Alejandre-Aguilar and Espinoza2006; Bargues et al., Reference Bargues, Klisiowicz, González-Candelas, Ramsey, Monroy, Ponce, Salazar-Schettino, Panzera, Abad-Franch, Souza, Schofield, Dujardin, Guhl and Mas-Coma2008, Martínez-Hernández et al., Reference Martínez-Hernández, Martínez-Ibarra, Catalá, Villalobos, de la Torre, Laclette, Alejandre-Aguilar and Espinoza2010; Espinoza et al., Reference Espinoza, Martínez-Ibarra, Villalobos, de la Torre, Laclette and Martínez2013), and pairwise comparisons of ITS-2 sequences indicated identical M. p. longipennis (=T. longipennis) and M. p. picturatus (=T. picturata) sequences and only two nucleotide differences between M. p. longipennis, M. p. picturatus, and M. p. pallidipennis (=T. longipennis). In contrast, other studies of the biological parameters and morphological characteristics have shown important differences among the three subspecies (Martínez-Hernández et al., Reference Martínez-Hernández, Martínez-Ibarra, Catalá, Villalobos, de la Torre, Laclette, Alejandre-Aguilar and Espinoza2010, Martínez-Ibarra et al., Reference Martínez-Ibarra, Nogueda-Torres, García-Benavídez, Vargas-Llamas, Bustos-Saldaña and Montañez-Valdez2012, Reference Martínez-Ibarra, Nogueda-Torres, Licón-Trillo, Villagrán-Herrera, de Diego-Cabrera, Montañez-Valdez and Rocha-Chávez2013, Reference Martínez-Ibarra, Nogueda-Torres, Salazar-Schettino, Cabrera-Bravo, Vences-Blanco and Rocha-Chávez2015d ; De la Rúa et al., Reference De la Rúa, Bustamante and Menes2014; Rivas et al., Reference Rivas, Sánchez-Espíndola, Camacho, Ramírez-Moreno, Rocha-Gómez and Alejandre-Aguilar2014). Our results are consistent with those last cited studies, supporting the proposal that the three subspecies are different enough from one another to be considered subspecies, not a single one.
It has been previously established that hybrid fertility and fitness are key parameters in determining the long-term outcome of the mixture of two different natural populations (Herrera-Aguilar et al., Reference Herrera-Aguilar, Be-Barragán, Ramírez-Sierra, Tripet, Dorn and Dumonteil2009; Mendoça et al., Reference Mendoça, Chaboli-Alevi, De Oliveira-Medeiros, Damieli-Nascimento, Vilela de Azeredo-Oliveira and Da Rosa2014). A lack of hybrid fitness leads to the maintenance of reproductive isolation of parental genotypes (Herrera-Aguilar et al., Reference Herrera-Aguilar, Be-Barragán, Ramírez-Sierra, Tripet, Dorn and Dumonteil2009), and our results fit with this statement. Moreover, based on our results, it can be proposed that an incipient speciation process by distance (Mayr & Ashlock, Reference Mayr and Ashlock1991) is currently developing among non-overlapping populations of each subspecies (M. p. longipennis, M. p. picturatus, and M. p. pallidipennis). This hypothesis is supported by important differences that were detected in the results of morphometric antennal analyses and molecular analyses using ITS-2 when different populations of M. p. pallidipennis and of M. p. longipennis (by species) from non-overlapping areas were compared (Martínez-Hernández et al., Reference Martínez-Hernández, Martínez-Ibarra, Catalá, Villalobos, de la Torre, Laclette, Alejandre-Aguilar and Espinoza2010; Martínez-Martínez et al., Reference Martínez-Martínez, Martínez-Hernández, Martínez-Ibarra and Espinoza-Gutiérrez2010). The results from our study and these two latter studies suggest that genetic exchange might not impede or delay the definitive divergence processes needed to reach the species level.
Our hypothesis is also supported by other considerations. For instance, the distribution areas of these three subspecies are based only on specific recent studies (López-Cárdenas et al., Reference López-Cárdenas, González-Bravo, Salazar-Schettino, Gallaga-Solórzano, Ramírez-Barba, Martínez-Méndez, Sánchez-Cordero, Townsend-Peterson and Ramsey2005; Martínez-Ibarra et al., Reference Martínez-Ibarra, Grant-Guillén, Morales-Corona, Haro-Rodriguez, Ventura-Rodríguez, Nogueda-Torres and Bustos-Saldaña2008a ; Benítez-Alva et al., Reference Benítez-Alva, Huerta and Téllez-Rendón2012) avoiding taking compilations of mixed old and new data into account. Therefore, the distribution is accurately delimited, with a few overlapping areas (Mayr & Ashlock, Reference Mayr and Ashlock1991) in western Mexico and some natural hybrids have been recorded (Martínez-Ibarra et al., Reference Martínez-Ibarra, Salazar-Schettino, Nogueda-Torres, Vences, Tapia-González and Espinoza-Gutiérrez2009). Those recorded distribution areas match mosaic hybrid zone models (Hewitt, Reference Hewitt, Otte and Endler1989), which are areas involving many independent contacts between ‘entities’ (subspecies in our case), each with a potentially unique evolutionary trajectory (Harrison & Rand, Reference Harrison, Rand, Otte and Endler1989). The existence of a heterogeneous environment, as recorded in previous studies (Martínez-Ibarra et al., Reference Martínez-Ibarra, Grant-Guillén, Morales-Corona, Haro-Rodriguez, Ventura-Rodríguez, Nogueda-Torres and Bustos-Saldaña2008a, Reference Martínez-Ibarra, Valencia-Navarro, León-Saucedo, Ibáñez-Cervantes, Bustos-Saldaña, Montañez-Valdez, Cervantes-Díaz and Nogueda-Torres2011; Benítez-Alva et al., Reference Benítez-Alva, Huerta and Téllez-Rendón2012), impedes fusion by favouring alternative types in different areas. Both the heterogeneity of the environment in which they occur and the complex internal structure of these hybrid zones promote the maintenance of diversity (species diversity or allelic diversity) (Harrison & Rand, Reference Harrison, Rand, Otte and Endler1989).
In summary, the results provided in this study show that different subspecies of M. p. longipennis, M. p. picturatus, M. p. pallidipennis, and their hybrids present differences in their biological parameters, and as a consequence, their biological fitness. This is likely due to the intrinsic variation as different groups. Evidence from our study, in conjunction with those from previous field studies, indicates that an incipient process of speciation is occurring among these three subspecies, with previously reported important differences in their capacity as vectors of T. cruzi to humans in Mexico (Martínez-Ibarra et al., Reference Martínez-Ibarra, Nogueda-Torres, García-Benavídez, Vargas-Llamas, Bustos-Saldaña and Montañez-Valdez2012, Reference Martínez-Ibarra, Nogueda-Torres, Licón-Trillo, Villagrán-Herrera, de Diego-Cabrera, Montañez-Valdez and Rocha-Chávez2013, Reference Martínez-Ibarra, Nogueda-Torres, del Toro-González, Ventura-Anacleto and Montañez-Valdez2015a ).
Acknowledgement
This study was founded by the project SA/CIP/009/2015 (Universidad de Guadalajara).
