Experimental crosses detect reproductive isolation among populations of Triatoma pallidipennis (Hemiptera: Reduviidae: Triatominae)

Abstract

Chagas disease is one of the most important vector-borne diseases in Mexico. Triatoma pallidipennis (Stål) is one of the most epidemiologically important vector species. Despite being classified as a single species, various studies (molecular, morphometric, and biological) on populations across its distribution suggested it is composed of a group of cryptic species. This study examined reproductive isolation among 5 populations of T. pallidipennis originating from the western, southern, and central regions of Mexico to help clarify the potential existence of a cryptic species complex of T. pallidipennis in Mexico. A generation of hybrids was analyzed for fertility and fecundity. Fertility rates varied from 50% to 100% in the parental crosses and from 20% to 100% in the F1 × F1 crosses. Fecundity ranged from 1.4 to 3.2 eggs/female/day in the parental crosses, which decreased to 0.9–2.9 in the F1 × F1 crosses (except in Jalisco × Morelos). The fertility of the eggs ranged from 61.4% to 85.4% in the parental crosses, dropping to 44% to 90.1% in some F1 × F1 crosses, indicating partial reproductive isolation among these populations.

Graphical Abstract

Open in new tabDownload slide

Neglected tropical diseases and socially determined diseases result from complex health issues. Among them, Chagas disease, an infectious ailment caused by the protozoan parasite Trypanosoma cruzi, can transform an infection into a life-threatening condition when a missed or late diagnosis combines with an absent or incomplete treatment (WHO 2024). It is believed approximately 75 million people are at risk of T. cruzi infection, with an estimated 6–7 million people worldwide harboring this infection, leading to approximately 12,000 deaths every year (WHO 2024). Despite its increasing global presence, Chagas disease is primarily found in endemic areas of 21 continental Latin American countries, where transmission is closely related to the presence of vector (WHO 2024). The main vectors belong to the subfamily Triatominae, with over 155 species of insects (Hemiptera) primarily feeding through hematophagy (de Oliveira et al. 2023).

In Mexico, the government estimates up to 900,000 people might be infected with T. cruzi (SS 2022). However, a national seroprevalence of 3.38% was recently estimated by academic researchers, representing 4.06 million infected people in the country (Bravo-Ramírez et al. 2023). Complementarily, it is believed that more than 88% of the population is exposed to vector infection, given that 19 out of 34 described triatomine species in Mexico invade human dwellings and 29 species can harbor T. cruzi (SS 2022, Bravo-Ramírez et al. 2023). Thirteen of 34 T. cruzi vector species recorded in Mexico are major T. cruzi transmitters to humans. Among them, T. dimidiata sensu lato (Latreille), T. barberi (Usinger), and T. pallidipennis (Stål) are the 3 most abundant triatomine species in Mexico, with frequently domiciliated populations widely distributed in the country (SS 2022).

Triatoma pallidipennis is found in 9 out of 32 Mexican states in the western, northwestern, central, and southern regions (Fig. 1) (Meraz-Medina et al. 2024). The infestation indices (percentage of human dwelling with the presence of triatomines) reached 66.3% in some areas, with reported infection percentages in triatomines ranging from 51.4% to 64.5% (Martínez-Ibarra et al. 2011, Rodríguez-Bataz et al. 2011, Pineda-Rodríguez et al. 2018, García-Mares et al. 2022, Salgado 2023). Triatoma pallidipennis has traditionally been considered as a single species (Salazar-Schettino et al. 2019). However, some studies have started questioning this due to differences observed among T. pallidipennis populations from geographically distant areas using the genetic marker Cyt B and antennal phenotype studies (Martínez-Hernández et al. 2010). Significant variations in biology and behavior have been recorded between T. pallidipennis populations from different geographical areas of Mexico (Martínez-Ibarra et al. 2012, 2014). For example, a population from Michoacan had longer egg-to-adult development time and produced more eggs per female per day than populations from Morelos and Oaxaca. The Morelos population, in contrast, required fewer blood meals to molt from the first-instar nymph to adult than the Michoacan and Oaxaca populations (Martínez-Ibarra et al. 2014). Also, 2 T. pallidipennis populations (from Jalisco and Oaxaca) had longer egg-to-adult development time than those from Guerrero, Michoacan, Morelos, and Puebla (Martínez-Ibarra et al. 2012, 2014). A higher mortality rate was observed in the first group compared to the second. These differences, regardless of geographic distance between populations, were mainly attributed to local environmental conditions; however, researchers also pointed to potential unexplored genetic differences as an alternative explanation (Martínez-Ibarra et al. 2012, 2014, Meraz-Medina et al. 2024).

Molecular analysis of 40 populations of T. pallidipennis from 8 states of western, central, and southern Mexico identified 4 haplogroups, where all studied populations were included, suggesting a cryptic species (Cruz and Arellano 2022). The authors recognized the need for complementary studies (morphological, ecological, crossing experiments) on T. pallidipennis populations to provide new information in addition to existing genetic data before formally describing these new species (Cruz and Arellano 2022). In a follow-up analysis, morphometrical characteristics did not support the previously proposed haplogroups, and environmental variable analyses barely validated the existence of 2 haplogroups (Cruz et al. 2023).

Experimental crosses play a significant role in addressing systematic issues in Triatominae (Alevi et al. 2018, 2021). Reproductive isolation, even when observed experimentally, is among the best criteria for evaluating the taxonomic status of populations that are morphologically or genetically related (Villacís et al. 2020, Vicente et al. 2022). The degrees of reproductive incompatibility can be estimated by identifying the pre- and post-zygotic reproductive barriers (i.e., isolation mechanisms) present when 2 taxa (e.g., species, subspecies, and populations) are experimentally crossed (Ravazi et al. 2021, dos Reis et al. 2022, Martínez-Ibarra et al. 2023a, 2023b). Multiple biological parameters, such as the number of breeding pairs, female fecundity, egg fertility, mortality rates, offspring survival to adulthood, and offspring fertility, can be used to study the impact of isolation mechanisms on the crossbreeding of 2 taxa (dos Reis et al. 2022, Martínez-Ibarra et al. 2023a, 2023b).

Thus, we conducted experimental crossbreeding to characterize the potential presence of pre- and post-zygotic reproductive barriers in an effort to decipher the possible existence of a cryptic species complex of T. pallidipennis in Mexico.

Materials and Methods

The genetic and reproductive compatibility of T. pallidipennis was examined (Cruz and Arellano 2022, Cruz et al. 2023, Martínez-Ibarra et al. 2023a, 2023b) using reciprocal crossing experiments with 5 different T. pallidipennis populations in Mexico. These populations were selected from various areas in the western and southern Pacific coastal states of Mexico, as well as central Mexico, where Chagas disease prevalence is greatest among the human community (Salazar-Schettino et al. 2019, Meraz-Medina et al. 2024).

Biological Material

Insects were obtained from colonies established from allopatric populations. These represented 5 populations of T. pallidipennis from which past studies (Martínez-Hernández et al. 2010, Martínez-Ibarra et al. 2012, 2014, Cruz and Arellano 2022, Meraz-Medina et al. 2024) detected genetic and behavioral differences. Adopting the framework proposed by Cruz and Arellano (2022) on the existence of 4 haplogroups of T. pallidipennis, we chose at least 1 location from each group. Initially, a minimum of 40 individuals were collected from each population to establish each colony. The selected locations were: Jilotlan de los Dolores, in Jalisco (19°22ʹN, 103°01ʹW), San Juan Cieneguilla, Oaxaca (17°51ʹN, 98°17ʹW), Taretan, Michoacan (19°20ʹN,101°55ʹW), Santa Catarina, Guerrero (17°45ʹN, 99°05ʹW), and Amilcingo, Morelos (18°74ʹN, 98°76ʹW) (Fig. 1). Specimens were collected from animal shelters near human dwellings along the Mexican Pacific coast and in central Mexico. Identification of the specimens was based on the most commonly used keys (Lent and Wygodzinsky 1979) and exhibited the typical morphological characteristics of T. pallidipennis, with slight variations in size and color (from orange to dark yellow) of the connexivum (Fig. 2).

Crossing Experiments

We used the fifth generation of each population for the experimental and control crosses in this study, as in previous similar studies (Martínez-Ibarra et al. 2023a, 2023b). This was done to ensure the purity of the colonies with regard to the exclusion of previous interfertility events. Ten pairs from each cross-combination were housed in plastic jars (5 cm diameter × 10 cm height) (Table 1). To simplify comparisons later with articles advocating the existence of 4 haplogroups (Cruz and Arellano 2022, Cruz et al. 2023), we named our studied populations in accordance with their states of origin: Jalisco (haplogroup IV), Michoacan (haplogroup III), Guerrero (haplogroup II), Oaxaca, and Morelos (both in haplogroup I). The crossbreeding was executed as follows: ♀ Jalisco × ♂ Oaxaca (and its reciprocal cross), ♀ Jalisco × ♂ Michoacan (and its reciprocal cross), ♀ Jalisco × ♂ Guerrero (and its reciprocal cross), ♀ Jalisco × ♂ Morelos (and its reciprocal cross), ♀ Michoacan × ♂ Guerrero (and its reciprocal cross), ♀ Michoacan × ♂ Oaxaca (and its reciprocal cross), ♀ Michoacan × ♂ Morelos (and its reciprocal cross), ♀ Guerrero × ♂ Oaxaca (and its reciprocal cross), ♀ Guerrero × ♂ Morelos, and ♀ Morelos × ♂ Oaxaca (and its reciprocal cross). Crosses of the 5 parental lineages involved in the study were used as controls (Table 1). Specimens were kept in incubators at 27°C ± 1°C and 70% ± 5% relative humidity, with a 12:12 h light-dark cycle, fed biweekly on New Zealand rabbits (Oryctolagus cuniculus L.) (Meraz-Medina et al. 2024). Rabbits were held sustained under laboratory conditions and were managed and anesthetized according to the Norma Oficial Mexicana NOM-062-ZOO-1999 (SAGARPA 1999).

Pairings were monitored daily to note each copulation event for direct observation of the copula or by observation of spermatophore elimination once the spermatophore received by the female is expelled after copulation and after the separation of the couple (Chiang and Chiang 2021). The average egg production of each cross-combination was calculated considering only successful crosses. Egg fertility was evaluated by collecting eggs from each pairing over 30 days, with an estimated egg production of 1.7–2.9 eggs/female/day, and incubating them under conditions previously outlined (Martínez-Ibarra et al. 2012, 2014).

Twenty first-instar nymphs were acquired from each of the 10 crossed combinations and placed in plastic jars, with 10 in each jar until 200 specimens were reached. These were all fed on rabbit blood and cultivated in a controlled environment. Uncontrolled mating was prevented by sexing the fifth instars, separating the males from females, and holding these in separate plastic jars until reaching the adult stage (Martínez-Ibarra et al. 2023a, 2023b). The first 100 individuals reaching adulthood from each cross-combination were chosen for inclusion in the results. In cases where crossings were unsuccessful (yielding no offspring), those inviable eggs were observed under a stereoscopic microscope to verify embryo formation. Complementarily, F1 individuals were paired with unmated individuals from their parental lineage. If the females then laid a significant number of fertile eggs, that hatched into F1 nymphs, their fertility was confirmed (Martínez-Ibarra et al. 2023a, 2023b).

Crosses Between F1 Individuals

We followed our standardized methodology for crossing triatomines, as detailed in 2 previous studies (Martínez-Ibarra et al. 2023a, 2023b). Specifically, to determine the fecundity of F1 offspring resulting from crosses among the 5 study populations as well as in the controls, we paired 10 females and 10 males of each type. The order in the experimental and control crosses was 10 F1 females × 10 sibling F1 males. We then selected 20 F2 nymphs from each of the crosses in each set in plastic jars, with the expectation of obtaining at least 10 adults, given mortality rates. Raised on rabbit blood, these nymphs matured into adults. As soon as first-instar nymphs were obtained, F1 fertility was proven. Any individuals involved in F1 × F1 crosses that did not yield any F2 offspring were backcrossed them with members of a parental lineage to verify their fecundity (Southwood and Henderson 2000). If this resulted in no offspring, a backcross was performed with members of the other parental lineage. Fecundity was confirmed in both instances when females produced numerous fertile eggs that hatched into first-instar nymphs (Martínez-Ibarra et al. 2023a, 2023b).

Statistical Analyses

The Holm–Sidak method was utilized to compare the average number of eggs laid per female. A Chi-square test was employed to compare frequencies. Statistical analysis was conducted using Sigma Stat 3.1 software (Version 3.1 for Windows, Systat Software Inc., San Jose, CA). The results were deemed statistically significant when P < 0.05.

Results

A 100% success rate (100% of couples with offspring) was observed in most parental crosses. Only 3 crosses (in both directions) exhibited significantly (χ² = from 69.9 to 76.7, df = 1, P < 0.01) lower success rates (50%–60%): Jalisco × Guerrero, Michoacan × Guerrero, and Michoacan × Morelos (Table 1). Those crosses with 100% success did not differ when each cross was compared with its reciprocal. Control crosses from the 5 populations studied also had a 100% success rate. Evidence of copulation events was observed in all sets of experimental and control crosses. Specimens involved in unsuccessful crosses were proven to be fertile when crossed with virgin specimens from the same parental cohort. A similar pattern was observed between the average egg production rate and the success rate of parental crosses. The 3 cross-combinations with lower success rates also had lower egg production. The 7 cross-combinations of crosses with 100% success in offspring production had an average egg production rate of 2.4–3.2 eggs/female/day. In contrast, the crosses with a success rate of 50%–60% (Jalisco × Guerrero, Michoacan × Guerrero, and Michoacan × Morelos) had a significantly (t = 3.28–7.41, df = 9, P < 0.01) lower average egg production rate of 1.4–1.9 eggs/female/day (Table 1). This average egg production rate was also significantly (t = 4.11–6.61, df = 8, P < 0.01) lower when comparing with involved controls (Jalisco, Michoacan, and Guerrero), whose average egg production rate was 2.7–3.0 eggs/female/day (Table 1).

In contrast, the percentage of fertile eggs displayed an irregular pattern. Only 3 crosses (Jalisco × Oaxaca, × Morelos, and Michoacan × Oaxaca) achieved a significantly (χ² = 61.27–77.23, df = 1, P < 0.05) higher (>80%) percentage of fertile eggs than the remaining 7 crosses, which had fertile egg percentages ranging from 51.8%–78.4% (Table 1). No significant (P > 0.05) differences were recorded when the 3 crosses with higher percentage of fertile eggs were compared with the 4 controls involved (Jalisco, Michoacan, Morelos, and Oaxaca) (Table 1). Those 3 crosses with higher percentage of fertile eggs also exhibited a higher success rate and mean egg production.

An interesting phenomenon was observed in the success rate of F1 × F1 crosses. Unlike parental crosses, which mostly revealed success rates of 100%, most F1 × F1 crosses had success ranging from 48% to 56%. Statistical analysis showed significant (χ² = 77.34–85.13, df = 1, P < 0.01) differences when F1 × F1 crosses were compared (Table 2). Similar differences (χ² = 72.41–81.17, df = 1, P < 0.01) were recorded between most F1 × F1 experimental crosses when compared with F1 × F1 control crosses (Table 2). Based on this variation in successful rates, 4 distinct groups of cross-sets emerged: The first group encompassed the crosses that maintained a high success rate, such as 3 of the 4 crosses involving the population from Jalisco (× Oaxaca, × Michoacan, and × Morelos) as well as the cross of Michoacan × Oaxaca (in both directions) (Table 1). A second group was formed by those cross-combinations that had a consistently low success rate (~50%), such as 2 cross-combinations involving the population from Guerrero (× Jalisco and × Michoacan). The third group was formed by those crosses with initially high success rates (100%) in parental crosses but which decreased by around 50% in F1 × F1 crosses; these were the inter-crosses between Guerrero, Oaxaca, and Morelos (Table 1). Finally, the cross-combination of Morelos × Michoacan exhibited a low success rate in the parental crosses (50%), which further decreased (to 20%) in F1 × F1 crosses (Table 1). In all cases, copulation events were observed. Specimens involved in unsuccessful crosses proved to be fertile when crossed with virgin specimens from the same parental cohort.

The mean egg production of F1 × F1 crosses varied from 0.9 to 1.7 eggs/female/day (mean = 1.3 ± 0.2) in most cross-combinations, significantly (t = from 3.28 to 6.53, df = 9, P < 0.01) lower than the mean egg production of involved F1 × F1 control crosses (2.1–3.1 eggs/female/day, mean = 2.5 ± 0.2) (Table 2). In contrast, most cross-combinations involving Jalisco (× Michoacan, × Oaxaca, and × Morelos, in both directions), as well as those of Michoacan × Oaxaca, and all control crosses, produced the highest mean number of eggs, ranging from 2.4 to 2.9 eggs/female/day (mean = 2.7 ± 0.1) (t = from 3.33 to 7.18, df = 9, P < 0.01) (Table 2).

The percentage of fertile eggs in F1 × F1 crosses exhibited an irregular pattern, similar to the percentage of fertile eggs resulting from parental crosses. Four cross-combinations, in both directions, had a percentage of fertile eggs exceeding 77%, significantly (χ² = 59.41–61.88, df = 1, P < 0.01) higher than the remained crosses. Three of these involved the population from Morelos (crossed with Jalisco, Michoacan, and Guerrero), in addition to the Michoacan × Guerrero cross. No significant (P > 0.05) differences were recorded when the percentages of fertile eggs in F1 × F1 crosses were compared with their control crosses (Table 2). Three cross-combinations, also in both directions, produced percentages of fertile eggs ranging from 48.8% to 61.2%. These involved the population from Jalisco crossed with Oaxaca and Michoacan, and Michoacan crossed with Oaxaca. Lastly, striking differences were observed in the percentages of fertile eggs between some crosses and their inverse: ♀ Jalisco ♂ Guerrero yielded 72.4% fertile eggs, while the inverse produced 81.7% (χ² = 43.22, df = 1, P < 0.05); ♀ Oaxaca ♂ Morelos resulted in 70.2% fertility compared to its inverse at 87.3% (χ² = 59.02, df = 1, P < 0.05); and ♀ Oaxaca ♂ Guerrero yielded only 44% fertile eggs, while its inverse yielded 90.2% (χ² = 72.07, df = 1, P < 0.01) (Table 2).

The percentages of F2 nymphs from each cross-combination and controls that reached the adult stage varied between 73% and 90%. No morphological abnormalities were observed in the F2 offspring (data not shown).

Discussion

The study of divergence and speciation within a population can be analyzed based on the phenomenon of hybridization, which is highly valuable for analyzing such processes. Although the outcomes of experimental interbreeding among certain groups (e.g., species and subspecies) may be slightly biased due to the compulsion of specimens to interbreed (for instance, by breaking ecological and geographical barriers), these studies nonetheless contribute to our understanding of a group’s systematics (Villacís et al. 2020, Martínez-Ibarra et al. 2023a, 2023b).

Integrative taxonomy involves the integration of various analyses to characterize a taxon through accumulation or congruence, thus contributing to the establishment of the appropriate status (e.g., species, subspecies, population) of that taxon (Dayrat 2005, Vicente et al. 2022). Even though T. pallidipennis is considered a single species (Lent and Wygodzinsky 1979, Salazar-Schettino et al. 2019), significant differences have been observed when comparing populations, including those registered in the current study (Martínez-Hernández et al. 2010, Martínez-Ibarra et al. 2012, 2014, Cruz and Arellano 2022, Cruz et al. 2023, Meraz-Medina et al. 2024).

The results of the current study indicated varying degrees of reproductive isolation among the studied populations from the 4 proposed haplogroups (Cruz and Arellano 2022). We observed a low degree of isolation between the populations of Jalisco (haplogroup IV) with respect to Oaxaca and Morelos (both belonging to haplogroup I) and Michoacan (haplogroup III). We observed a low degree of isolation between the populations of Jalisco (haplogroup IV) with respect to Oaxaca and Morelos (both belonging to haplogroup I) and Michoacán (haplogroup III). There appeared to be no influence from the geographical distance between the Jalisco population and those from Michoacan (~120 km), Morelos (~700 km), and Oaxaca (~800 km). There appear to be no pre-zygotic isolation mechanisms acting between Jalisco and the other 3 populations. In contrasts, a post-zygotic mechanism seemed to be in effect when F1 × F1 specimens from Jalisco × Oaxaca were crossed, resulting in a reduction in the percentage of fertile eggs to less than 60%. In the set of crosses of Jalisco × Michoacan, the percentage of fertile eggs decreased by approximately 20% in F1 × F1 crosses compared to parental crosses and controls. Given copulation events were evident, and embryo formation was confirmed among all studied pairs of T. pallidipennis, we suggest a post-zygotic mechanism could be acting through the elimination of hybrids during the zygote or embryo stage (hybrid inviability).

Our results were inconsistent with those of previous genetic distance research, in which the population from Jalisco was genetically distant (belonging to different haplogroups) and strongly reproductively isolated from Michoacan, Oaxaca, and Morelos (Cruz and Arellano 2022). Likewise, our results do not align with those of a morphometric study (Cruz et al. 2023) featuring head and pronotum features of the same T. pallidipennis populations previously genetically analyzed by the same research group (Cruz and Arellano 2022), in which the Jalisco population noticeably differed from the remaining studied populations. Comparable inconsistent results arose in a study in which the T. pallidipennis population in Jalisco showed notable differences compared to Michoacan, Oaxaca, and Morelos when comparing some biological parameters (life cycle, fecundity, and fertility) (Martínez-Ibarra et al. 2012). This lack of concordance among different studies prevents us from concluding the actual status of the Jalisco population concerning the Oaxaca and Morelos populations of T. pallidipennis. This lack of conformity was also observed in comparisons between 2 genetically separate populations of T. longipennis (Usinger) from northern and western Mexico (approximately 900 km apart), a species close to T. pallidipennis phylogenetically, demonstrating low isolation (Martínez-Ibarra et al. 2023a). Contrarily, our data suggest that the Jalisco population was notably isolated from the Guerrero population (haplogroup III) (Cruz and Arellano 2022), apparently exhibiting a pre-zygotic mechanism (gametic isolation) in some intercrossing pairs (which proved to be fertile when backcrossed) and a post-zygotic mechanism (hybrid breakdown) in the intercrossing pairs involved a female from Jalisco (Singh 2022). This is in line with the genetic, morphometric, and biological data reported by Cruz and Arellano (2022) and Cruz et al. (2023), alongside data on egg-to-adult development and the number of blood meals required to molt to adulthood (Martínez-Ibarra et al. 2014). For T. pallidipennis, any conclusion about relations between certain populations was unattainable. However, various studies suggest potential geographical isolation of the Jalisco population, which may lead to allopatric speciation of this group, providing evidence for incipient speciation (Singh 2022).

The population from Michoacan (haplogroup III) showed a low degree of isolation from that of Oaxaca (haplogroup I) (Cruz and Arellano 2022) despite the significant geographic distance between both populations (~600 km). It appeared that a post-zygotic mechanism (hybrid inviability) functioned when F1 × F1 specimens from Michoacan × Oaxaca were crossed. This led to a reduction in the percentages of fertile eggs from 80% to 60% in the F1 × F1 crosses relative to the parental crosses. This finding conflicts with a related study that portrayed the Michoacan population as being genetically distant and significantly reproductively isolated from Oaxaca (Cruz and Arellano 2022). In contrast, a morphometric study (Cruz et al. 2023) examining populations of T. pallidipennis from the same or geographically close localities in Michoacan and Oaxaca did not support the existence of the previously purported haplogroups (I and III) (Cruz and Arellano 2022), thereby casting these populations as similar. This was further reinforced by similarities in certain biological parameters (e.g., the number of blood meals required to molt to the adult stage and fertility) (Martínez-Ibarra et al. 2012).

However, a high degree of reproductive isolation was observed among the populations from Michoacan (III), those from Guerrero (II), and Morelos (I) (Cruz and Arellano 2022). This was particularly notable in the F1 × F1 crosses × Morelos, where the success rate was as low as 20%, and fertility was 1 egg/♀/day. These findings seem to align with the proposition of 3 distinct haplogroups (Cruz and Arellano 2022). Yet, this contrasts with the results from the morphometric study by Cruz et al. (2023), that failed supporting which did not support the existence of these 3 populations belonging to 3 separate haplogroups (I, II, and III), leading them to suggest that these populations belong to the same species (T. pallidipennis). Similarities in certain biological data (the number of blood meals needed to reach the adult stage, fecundity, and fertility) bolster the argument for the existence of only 1 species (Martínez-Ibarra et al. 2012, 2014). Such divergence between different studies impedes reaching a decisive conclusion about the true status of the population from Michoacan concerning Guerrero, Oaxaca, and Morelos populations of T. pallidipennis.

A low degree of isolation was observed between the parental crosses of the haplogroup I populations in Oaxaca and Morelos (Cruz and Arellano 2022), aligning with the relatively short geographical distance (~150 km) between chosen locations in both states. However, the F1 × F1 crosses displayed a substantial reduction (almost 50%) in the success rate and a decrease in female fecundity. Additionally, fertility was reduced when crosses were executed directly (as opposed to inversely), that is, involving a female from Oaxaca. These data suggest the actuation of a post-zygotic mechanism (hybrid breakdown) (Singh 2022). These observations are consistent with variations in biological parameters, such as development time from egg to adult and fecundity (Martínez-Ibarra et al. 2012). In contrast, genetic and morphometric studies suggest a close relationship between these populations (Cruz and Arellano 2022, Cruz et al. 2023). Taken together, our results suggest that both populations could have originated from a recent diversification from a common ancestor (Pinotti et al. 2021), akin to the proposal for the T. longipennis populations in northern and western Mexico (Martínez-Ibarra et al. 2023a.

A low degree of isolation was observed between the parental crosses of the populations from Guerrero (II) and both groups from haplogroup I (Oaxaca and Morelos), which aligns with the short geographic distance (~200 km) between the selected localities from these states (Cruz and Arellano 2022). Contrarily, the F1 × F1 crosses demonstrated a significant reduction (by almost 50%) in the success rate and a decrease in female fecundity. Still, fertility increased when a female from Morelos was involved but decreased when the involved female was from Oaxaca. All those data suggest the actuation of a post-zygotic mechanism (hybrid breakdown) (Singh 2022). Contrastingly, some biological parameters such as development time, the number of blood meals, and fertility were similar among the 3 populations (Martínez-Ibarra et al. 2012, 2014). These results align with those of Cruz et al. (2023), who found no morphometric differences among these populations to consider them different species. While most studies support the notion of these 3 species belonging to a single species, the existence of a genetic study, where the classification into 4 haplogroups was proposed (Cruz and Arellano 2022), necessitates further studies to corroborate each proposal.

Our results indicated varying degrees of reproductive isolation among the 5 studied populations without a discernible pattern. Taking all comparisons between populations into consideration does not distinctly categorize them. However, they can be divided into 3 groups: (i) Jalisco, Oaxaca, and Michoacan; (ii) Morelos; (iii) Guerrero. It appears that the Guerrero population is the most isolated with respect to the other 4.

By employing integrative taxonomy data, including literature-sourced data, we can discern several conclusions. Current data support the proposed existence of substantial differences among the 5 studied populations—genetic, biological, and reproductive isolation differences—suggesting a division of T. pallidipennis into distinct groups (Martínez-Ibarra et al. 2012, 2014, 2018, Cruz and Arellano 2022). However, these cannot be classified as full species from a biological standpoint as per the biological species concept: “A species is a group of interbreeding natural populations that is reproductively isolated from other such groups” (Mayr and Ashlock 1991).

Our findings do not align with the previously proposed group structures of Cruz and Arellano (2022): Group I, containing Morelos and Oaxaca; Group II, Guerrero; Group III, Michoacan; and Group IV, Jalisco, with the sole exception being Guerrero. Our analysis also was inconsistent with the conclusions of a morphometric study by Cruz et al. (2023), which portrayed the population from Jalisco as significantly distinct and warranted separation from the remaining 4. A hypothesis to explain the inconsistencies recorded is the influence of the molecular marker used to study a population, since some differences may arise depending on the marker used, such as when the origin of an invasive population of T. infestans (Klug) discovered in western Mexico was determined using the 16S, ITS-2, Cyt B, and cox I (Martínez-Hernández et al. 2022a, 2022b). An alternative explanation is the different origin in all cases (except Taretan, Michoacan), of the specimens studied in our study compared to those of Cruz and Arellano (2022) and Cruz et al. (2023). That is because notable genetic differences have been recorded even between geographically close (<20 km) populations of T. pallidipennis, such as Amacuzac and Puente de Ixtla in Morelos (Cruz and Arellano 2022).

Considering previous studies (Martínez-Ibarra et al. 2012, 2014, 2018, Cruz and Arellano 2022, Cruz et al. 2023), we propose certain hypotheses to elucidate the relationships among the studied populations. The T. pallidipennis populations from Jalisco and Michoacan are believed to be lineages of a metapopulation, defined as “an inclusive population made up of connected subpopulations extended through time” (Pavan et al. 2021). The populations from Morelos, Guerrero, and Oaxaca—the latter representing the species’ extremity of southern distribution (Salazar-Schettino et al. 2019)—are postulated to be undergoing a divergence process caused by scant genetic flow with geographically distant (>600 km) T. pallidipennis populations in western Mexico (Jalisco and Michoacan), known as isolation by distance.

The diversification hypothesis from a common ancestor might explain why the population from Oaxaca shares numerous characteristics with those from western Mexico (Pinotti et al. 2021). The population from Oaxaca appears to be undergoing allopatric speciation, given its geographical isolation from other T. pallidipennis populations, situated within the Mixtecan Sierra region’s higher altitude, where it remains predominantly sylvatic (Ramsey et al. 2000). This is starkly contrasted with other studied populations (Jalisco, Michoacan, Morelos, and Guerrero) of T. pallidipennis, commonly found in peridomestic and domestic areas (Salazar-Schettino et al. 2019).

The populations from Morelos and Guerrero are apparently undergoing an incipient speciation process, as certain factors severely limit the gene flow between them and the remaining populations, such as isolation by distance. When a population accumulates mutations independently, it can develop a degree of genetic divergence, potentially becoming genetically isolated. Complete allopatric speciation may occur if populations of incipient species develop pre- or post-zygotic barriers to reproduction (Pavan et al. 2021). A similar conclusion was reached regarding the populations of T. longipennis from Guadalupe y Calvo, Chihuahua, Mexico, the extreme northern distribution of this species (Salazar-Schettino et al. 2019), with respect to populations of this species from central and western Mexico (Martínez-Ibarra et al. 2023a).

Our data are limited by the small number of studied populations of T. pallidipennis, as well as incomplete collections in the distribution areas of this species, not including states such as Colima (on the Pacific coast) and 3 centrally located states in Mexico—Guanajuato, Puebla, and Estado de Mexico. Despite these limitations, our findings on the same (or geographically nearby) previously studied populations of T. pallidipennis (Cruz and Arellano 2022, Cruz et al. 2023) aid in deciphering the possible existence of a cryptic species complex of T. pallidipennis in Mexico. The potential presence of this new complex of cryptic species would open new questions regarding its evolutionary position within the genus, and even more so within the Phyllosoma complex (Cruz and Arellano 2022, Cruz et al. 2023). Therefore, future phylogenetic reconstructions maybe should consider the presence of the new cryptic lineages and clarify their relationship with the rest of the species (Cruz and Arellano 2022, Cruz et al. 2023).

Further studies involving a broader collection and additional populations of T. pallidipennis are now essential to thoroughly define the variability among their populations.

Acknowledgments

We thank Alejandro Martínez-Pérez for his technical advice. We thank inhabitants of studied localities for the donation of specimens.

References Cited