Species that are functionally equivalent but with little taxonomical relationship may display similar genetic patterns if the ecological function evolves genetically in the same way. This study investigated the levels of genetic diversity in the D-Loop gene of random samples collected from 21 bat species inhabiting El Ocote Biosphere Reserve (REBISO, for its acronym in Spanish), and whether the genetic diversity pattern could be associated with the ecological role. Genetic differences between functional groups, localities, and species were evaluated through generalized linear models using the Gaussian distribution error family for nucleotide diversity (π) and the Poisson family for haplotype diversity (h) and segregating sites (s). To study the clustering pattern of species based on nucleotide variation, genetic distances (Kimura’s two-parameter model) between functional groups were calculated, and a Principal Components Analysis on genetic diversity parameters was run. Most of the species analyzed (20) maintained genetic diversity levels ranging from medium to high in all genetic diversity estimators. According to genetic distances, the species with the same ecological function shared a greater number of nucleotide substitutions, with some exceptions. The Principal Components Analysis did not detect any genetic structure in relation to the ecological function. Our study found no association between the diversity of the D-Loop gene and ecological function; nonetheless, it confirms the importance of REBISO as a reservoir of bat species richness and genetic diversity in Mexico.
Introduction
Genetic diversity determines the evolutionary potential of a species and its ability to cope with changes in its environment (Frankham, Ballou, & Briscoe, 2002). Community genetics aims at understanding how within-species variation, species diversity, and environmental factors interact to shape community assemblages (Lamy, Laroche, David, Massol, & Jarne, 2017). Genetic diversity within communities can correlate with species diversity, both within and between trophic levels, in at least three ways (Avolio, Beaulieu, Lo, & Smith, 2012). First, intraspecific genetic diversity can influence the species diversity, structure, and functioning of communities through a genetic-feedback mechanism (see Genung et al., 2011; Whitlock, 2014). Second, genetic diversity and ecological function can be associated through parallel responses to selection pressures and eco-evolutionary dynamics over time (Hughes, Inouye, Johnson, Underwood, & Vellend, 2008; Whitlock, 2014). Third, ecological interactions (e.g., competition, depredation, reproduction) determine demographic changes (birth, death, and movement of different biological types) affecting the dynamics of both populations and communities, hence establishing levels of species diversity and genetic structure within the community (Moreira, Abdala-Roberts, Parra-Tabla, & Mooney, 2014; Vellend & Geber, 2005). When competitive interactions dominate in a community, these increase the species diversity of one group while reducing the diversity in others by occupying the available niche space (Vellend, 2008). These conditions, together with the life history of the species (i.e., social structure, mating, behavior), may contribute to define the levels of genetic diversity (Lamy et al., 2017).
Understanding the relationship between genetic diversity and ecological function is key; if genetic diversity is a structuring driver in communities and ecosystems, it deserves to be included in ecological models constructed to explain the distribution, diversity, and abundance of species. In addition, the genetic diversity contained within species may determine the responses of communities and ecosystems to anthropogenic environmental change. Gaining knowledge on genetic diversity at the community level allows anticipating and managing the potential shifts in community structure and function that may arise as correlated responses. Also, the study of genetic diversity at a community level may provide useful information to evaluate the evolutionary potential, establish models for inferring evolutionary pathways within the community, and identify potential threats for a group of species, which can be useful to support strategies aimed at the conservation of priority species, communities, and ecosystems (Crawford & Rudgers, 2013).
The ecological function of a species is likely the result of natural selection, which drives adaptive differentiation (Hughes et al., 2008; Vellend, 2006; Whitham et al., 2006). The divergence between species is reflected in genetic relationships and ecological attributes, which in turn determine taxonomic groups (Aguirre, Montaño-Centellas, Gavilanez, & Stevens, 2016). However, the same ecological function can be observed in species with no apparent taxonomic relationship (He, Lamont, Krauss, Enright, & Miller, 2008; Symstad, 2000; Vellend & Geber, 2005). Species with little taxonomic relationship but with similar ecological functions may display similar genetic diversity patterns if the evolutionary processes that drive the diversification of species and their role occur concomitantly and in the same way (Avolio et al., 2012; Hoehn, Tscharntke, Tylianakis, & Steffan-Dewenter, 2008). Alternatively, species may be unrelated because of different evolutionary processes, by differences in their original gene pool, and because intrinsic ecological processes affect differentially the genetic diversity of each species (Aguirre et al., 2016; Freeland, 2005).
In theory, communities with high species diversity show a better ecosystem functioning than those with lower levels (Hoehn et al., 2008). The composition of a community depends on the survival and reproduction of the species, and these in turn are determined by genotypic composition. If evolution via natural selection promotes the coexistence of species, the loss of genetic diversity within species could hinder this process. This may lead to the loss of species (He et al., 2008; Vellend, 2006; Vellend & Geber, 2005), with negative effects on the community, particularly when the remaining species cannot assume or replace the ecological role that is lost along with the species (Cottontail, Wellinghausen, & Kalko, 2009; Crutsinger et al., 2006; Mooney et al., 2009; Park, 2015; Vellend, 2005; Vellend & Geber, 2005).
Tropical ecosystems are characterized by the highest species richness, biomass, and productivity levels (Ricklefs, 2004; Sahu, Sagar, & Singh, 2008). The expansion of human settlements and the need to increase productive areas in tropical regions has led to accelerated deforestation and habitat fragmentation, resulting in the loss of species and natural communities (Galindo-González & Sosa, 2003; Ripperger, Tschapka, Kalko, Rodríguez-Herrera, & Mayer, 2013). Bats are one of the most abundant and diverse mammal groups in tropical forests, displaying a great variety of behavioral, morphological, and ecological attributes; in these environments, bats play a key role in pollination, insect predation, and seed dispersal (Burns & Broders, 2014; Cosson, Pons, & Masson, 1999; Fenton et al., 2001; Fenton & Ratcliffe, 2010; Gorresen & Willig, 2004; Meyer, Struebig, & Willig, 2016). Bats are classified into five functional groups (trophic guilds) in relation to feeding habits and food preference: insectivorous, frugivorous, hematophagous, nectarivorous, and carnivorous (Calonge, 2009; Duckworth, Kent, & Ramsay, 2000; Merritt, 2010; Patterson, Willig, & Stevens, 2003; Reich, Walters, & Ellsworth, 1997; Reid, 2009; Soriano, 2000). The diverse functionality of bats involves ecological relationships with a wide range of species from other biological groups, both plants and animals (Medellín, Arita, & Sánchez, 1997).
There are approximately 140 species of bats in Mexico, accounting for about 13% of the total number of bat species worldwide (Ceballos & Ehrlich, 2002; Medellín et al., 1997). In El Ocote Biosphere Reserve (REBISO, for its acronym in Spanish), one of the main remnants of tropical forest in Mexico, 48 bat species have been recorded (accounting for 34% of the total number of bat species recognized in Mexico; Navarrete, Alba, March, & Espinoza, 1996). However, their genetic diversity has not been explored. The importance of bat species richness (Espinoza et al., 1999; Hernández-Mijangos, Gálvez-Mejía, Díaz-Negrete, & Cruz-Durante, 2008; Navarrete et al., 1996; Riechers, 2004, 2009) and the role of bats in the recovery of disturbed areas (Preciado-Benítez, Gómez, Navarrete-Gutiérrez, & Horváth, 2015) have been recognized to some extent. The current lack of information coupled with the continued habitat loss and fragmentation in El Ocote (Flamenco-Sandoval, Martínez, & Masera, 2007) call for the need to conduct research on this taxonomic group characterized by a high functional diversity.
Although some studies have correlated genetic diversity with species diversity (e.g., Avolio et al., 2012; Blum et al., 2012; Csergö, Hufnagel, & Höhn, 2014; Wei & Jiang, 2012), few have been conducted in México (outside the REBISO; e.g., Simental-Rodríguez et al., 2014; Wehenkel, Bergmann, & Gregorius, 2006) and none has related genetic diversity with the ecological function of species as intended in this research. The aim of this study was to determine the genetic diversity of a random sample of bat species in a semievergreen tropical forest at REBISO and explore whether this genetic diversity is correlated with the ecological role they play in the ecosystem. Considering that natural selection acts over all genomes and leads to an increased frequency of the genotypes governing the function of an organism in the community, our expectation was to find a positive relationship, that is, similar levels of genetic diversity across species with similar ecological functions, regardless of their taxonomic affinity. Measures of genetic diversity related to ecological function in a bat community can be further trait to estimate the conservation status and get an insight on their vulnerability to environmental changes.
Methods
Study Area
REBISO is a Protected Natural Area located to the northeast of the state of Chiapas, Mexico, between 16°45′42′′ and 17°09′00′′ N, and 93°54′19′′ and 93°21′20′′ W (Figure 1). The prevailing climate in El Ocote is warm and humid with abundant summer rainfall, with a mean annual precipitation of 2,145 mm and a mean annual temperature of 23.3℃ (Secretaria de Medio Ambiente y Recursos Naturales-Comisión Natural de Áreas Naturales Protegidas, 2001). The study was conducted at localities within the core of REBISO, which are sites with semievergreen tropical forest (Veinte Casas, Emilio Rabasa, Nuevo San Juan Chamula and San Joaquín; Figure 1).
Fieldwork
Unlike conventional population genetics studies, which focus on obtaining a large sample size for different populations of a given species, our sampling design aimed at sampling the functional diversity of bats inhabiting the semievergreen tropical forest at REBISO. To this end, we focused on obtaining largest number of species with different ecological roles, thereby recording the genetic diversity associated with the ecological function of the species in the community. Two sampling sites were established; in each locality, bats were sampled over six consecutive nights with similar weather conditions and lunar phase, from January 2015 to September 2015 (4 Localities × 2 Sampling Sites × 6 Nights). In each sampling site, four mist nets (12 m × 2.5 m, 6 m × 2.5 m) were placed between the vegetation and near water bodies, at an average height of 2 m. In relatively open areas, nets were placed at 10 m height (Cosson et al., 1999; Preciado-Benítez et al., 2015). All nets remained open after sunset for seven hours (6p.m.–1a.m.). The specimens captured were identified taxonomically according to Medellín and Sánchez (2008) and Reid (2009). For each specimen, the biological characteristics (i.e., body size, sex, age, and reproductive status) and geographical location were recorded, and a tissue sample from the uropatagium was collected for genetic analysis; afterward, the specimen was released. Tissue samples were preserved in 1.5 ml vials containing 96% ethyl alcohol. Bat sampling was conducted under the collection license SGPA/DGVS/14214/15 issued by the Mexican Secretariat of Environment and Natural Resources (Secretaria de Medio Ambiente y Recursos Naturales).
DNA Extraction and Amplification
Genomic DNA was extracted through the cellular lysis method followed by purification with phenol/chloroform-isoamyl alcohol (Hamilton, Pincus, Di Fiore, & Fleicher, 1999). Genetic diversity was determined based on the control region of mitochondrial DNA (D-Loop); this gene is found in all vertebrates and is characterized by high substitution rates, allowing the comparison of the same genomic region between species to describe genetic structure at the intraspecific level (Piaggio, Navo, & Stihler, 2009). The D-Loop region was amplified with markers D-Loop-E (5′-CCTGAAGTAGGAACCAGATG-3′) and D-Loop-P (5′-CCCCACCATCAACACCCAAAGCTGA-3′; Wilkinson & Chapman, 1991). Amplifications were performed in a C1000 Touch™ thermal cycler (Bio-Rad) using a total volume of 50 µl. The amplification process consisted of an initial denaturation at 94℃ × 4 min, 35 cycles at 94℃ × 1 min (denaturation), 5℃ × 1:30 min (alignment) and 72℃ × 1 min (extension), and a final extension at 72℃ × 10 min. To evaluate the amplification of the D-Loop, the PCR products were visualized through 2% agarose gel electrophoresis with a 100 base pair (bp) control marker (Ladder, PROMEGA). Positively amplified reactions were sequenced in Macrogen Inc, Korea, through capillary electrophoresis (Sanger) sequencing using the ABI PRISM® BigDyeTM Terminator Cycle Sequencing Kits and the ABI Prism® 3730XL Analyzer.
Analysis
The sequences obtained were edited with the program Chromas Pro v. 1.5 (McCarthy, 1996) and were aligned with the program Clustal X v. 2.1 (Thompson, Gibson, Plewniak, Jeanmougin, & Higgins, 1997). The genetic diversity parameters by species were obtained with the software DnaSP v. 5 (Librado & Rozas, 2009), based on the Kimura two-parameter substitution model (Kimura, 1980). The Tajima index (D) was estimated for each species (Tajima, 1989) to identify some of the demographic evolutionary processes associated with genetic diversity (e.g., bottleneck).
Genetic diversity was associated with three potential explanatory variables: ecological function, locality, and species. To assess the influence of each of these variables on genetic diversity (i.e., number of haplotypes [h], nucleotide diversity [π], segregating sites [s]), Generalized Linear Models were built with the software R v. 3 (R Core Team, 2013). According to the nature of the genetic parameters and their distribution, we used the Poisson error distribution family for Generalized Linear Model constructed for h and s, while for π the Gaussian family was used. All models were evaluated with the Akaike Information Criterion (AIC; Akaike, 1974) using the statistical package mentioned earlier.
Genetic distances between species were obtained with the Kimura two-parameter method (Kimura, 1980), with 1,000 bootstrap replicates using the program MEGA v. 6 (Tamura, Stecher, Peterson, Filipski, & Kumar, 2013). For this purpose, sequences were tested for similarity with the software TCS v. 1.21 (Clement, Posada, & Crandall, 2000). The clustering pattern was studied from dendrograms obtained through the Neighbor-Joining method (Saitou & Nei, 1987) using the PAUP program v. 4.0 a (Swofford, 2002). In addition, an analysis of genetic distances was carried out considering the four functional group, that is, frugivores, hematophages, insectivores, and nectarivores, assigning each species to one of these groups to get the genetic relationship between functional groups. Finally, the relationship between genetic structure and functional ecology was assessed through a Principal Components Analysis (PCA) with the software Statistica v. 8 (StatSoft, 2007); to this end, a colinearity analysis was performed between the three genetic diversity parameters (nucleotide diversity, segregating sites, number of haplotypes) through Pearson’s correlations with the program R v. 3 (R Core Team, 2013).
Results
We analyzed 281 sequences from 21 bat species belonging to three taxonomic families (Phyllostomidae, Vespertilionidae, and Mormoopidae; Table 1). Of these, 11 were frugivores, four nectarivores, five insectivores, and one hematophage (Table 1). All sequences comprised 396 bp. GenBank accession numbers for these sequences are MF803983-MF804264 (Appendix). The composition of the D-Loop region showed 32.5% thymine, 12.6% cytosine, 31.6% adenine, and 23.4% guanine.
Table 1.
Taxonomic Classification, Acronyms, Common Names, and Functional Group of 21 Bat Species Inhabiting El Ocote Biosphere Reserve.
Eptesicus furinalis (Argentine brown bat, an insectivore) showed the lowest genetic variation (s = 0, π = 0.00, and h = 1). Artibeus jamaicensis (Jamaican fruit-eating bat) showed the largest number of segregating sites (s) and haplotypes (h) (s = 176, π = 39), while Micronycteris microtis (Common big-eared bat, an insectivore) showed the highest nucleotide diversity (π = 0.27) (Table 1). Haplotype diversity (Hd) was high in most species (Hd = 0.94–1.00), except for E. furinalis (Hd = 0.00) and Desmodus rotundus (Common vampire bat; Hd = 0.51).
The Tajima test (D) was performed only in 11 of the 21 species analyzed due to the insufficient sample size for the analysis. We found a significant negative relationship between nucleotide diversity (π) and nucleotide variation by sequence (Θ) in A. jamaicencis (D = −2.68, p < .001), A. lituratus (Great fruit-eating bat; D = −2.09, p < .001), and D. rotundus (D = −2.32, p < .001) (Table 2).
Table 2.
Genetic Diversity of 21 Tropical Bat Species Inhabiting El Ocote Biosphere Reserve Based on the Mitochondrial DNA Control Region (D-loop).
The genetic diversity measure, which included segregating sites (s), nucleotide diversity (π), and number haplotypes (h), was variable both across species and across localities within individual species (Figure 2(a) to (c)). Segregating sites displayed a wide variation from 0 to 160 (Figure 2(a)). Nucleotide diversity was also heterogeneous within and between both species and functional groups, ranging from 0.00 to 0.27 in E. furinalis and M. microtis, respectively (Figure 2(b)). The number of haplotypes was reduced for hematophagous, nectarivorous, and insectivorous species (Figure 2(c)). Species that could be sampled in more than one locality, like A. jamaicensis and Carollia sowelli, showed different level of s, π, and h per locality (Figure 2(a) to (c)). The best Generalized Linear Models identified by AIC was the species-by-locality interaction for all three genetic diversity parameters; however, none reached statistical significance (Table 3).
Table 3.
Generalized Linear Models to Explain the Genetic Diversity of 21 Tropical Bat Species Inhabiting El Ocote Biosphere Reserve.
The clustering analyses showed a random grouping of species based on functional groups. A visual examination reveals two small groups of insectivores, a small group of nectarivores, a large group of frugivores, two groups with two species, and finally three isolated individual species (Figure 3). Glossophaga soricina (Pallas’s long-tongued bat, a nectarivorous species) was clustered together with C. sowelli (Sowell's short-tailed bat), a frugivorous species (Figure 3). The analysis of genetic distances by functional group revealed the closest genetic relationship between frugivores and nectarivores (0.205), followed by hematophages and frugivores (0.331), frugivores and insectivores (0.347), nectarivores and hematophages (0.349), and nectarivores and insectivores (0.371); the greatest genetic distance occurred between insectivores and hematophages (0.470).
In the PCA, the first three components accounted for 100% of the genetic variation. Variables in the first and third components made a similar contribution: segregating sites (s) showed the greatest contribution (0.57, 0.43), followed by the number of haplotypes (h) (0.25). In the second component, both nucleotide diversity (π) (0.55) and the number of haplotypes (h) (0.45) accounted for virtually all the variation. The three components were positively related to the three genetic diversity variables, that is, segregating sites (s), number of haplotypes (h), and nucleotide diversity (π). The PCA showed no relationship between genetic structure and functional groups (Figure 4).
Discussion
When genetic diversity is high, the capacity of a species is better suited to respond to environmental selective pressures and stochastic events (Frankham et al., 2002). The levels of genetic diversity detected in this study suggest that 20 of the 21 species analyzed probably maintain genetic diversity levels that may contribute to their conservation if changes in the local habitat continue at the current rate (Flamenco-Sandoval et al., 2007; Frankham et al., 2002). With the exception of E. furinalis (π = 0.000), the genetic diversity of the bats studied lied within the range reported in previous studies of some species. For example, in A. jamaicensis and based on the same mitochondrial gene (D-Loop), a range of π = 0.009–0.23 was reported (Carstens, Sullivan, Dávalos, Larsen, & Pedersen, 2004; Llaven, Ruiz, García, Lesher, & Machkour, 2017; Redondo, Brina, Silva, Ditchfield, & Santos, 2008; Ruiz, Vargas-Miranda, & Zúñiga, 2013).
The lack of genetic variation observed in E. furinalis (an insectivore) in REBISO (π = 0,000, s = 0, h = 1) is worth noting; this may be associated either with a likely recent bottleneck or with sweeping selection. Both processes lead to a drastic decrease in genetic diversity levels (Kaplan, Darden, & Hudson, 1989; Perfectii, Picó, & Gómez, 2009). Glossophaga morenoi and M. microtis showed the highest genetic diversity (π = 0.24, 0.27, respectively). According to the IUCN Red List, M microtis is a generalist bat characterized by high tolerance to disturbance and high local abundance (Miller, Reid, Arroyo-Cabrales, Cuarón, & de Grammont, 2008), moreover, Téllez-Girón and Ceballos (2005) consider that this bat is a common species in undisturbed areas in Mexico, which account for the high genetic diversity levels observed in REBISO. The case of G. morenoi (Western long-tongued bat) is worth noting, since its endemism (Arita, 2005) would suggest a lower genetic diversity relative to species such as A. jamaicensis (Ortega & Steers, 2005), contrary to what we found in REBISO. However, populations of G. morenoi with high local abundance have been recently reported in protected areas in southern Mexico (Arroyo-Cabrales, Álvarez-Castañeda, Cuarón, & Grammont, 2015), which may explain the high levels of genetic diversity found in REBISO populations.
Hd was high in 19 species (Hd = 0.94–1.00), indicating a lower genetic diversity in D. rotundus and E. furinalis (Arboleda, 2008; Castillo-Cobián, 2007). Desmodus rotundus showed an intermediate value of unique haplotypes (Hd = 0.51) that may be related to its migratory habits, high flight capacity (>100 km), and high tolerance to anthropogenic environments (Burns & Broders, 2014; Castro-Castro, Muñoz-Flores, & Uieda, 2016). All these features can facilitate gene flow between populations of the REBISO localities studied and others outside the Reserve (Burns & Broders, 2014).
The Tajima test suggest that A. jamaicensis, A. lituratus, and D. rotundus are undergoing a population expansion, a condition that may result from their high tolerance to disturbed environments and their ability to use various types of habitats (Ortega & Steers, 2005; Steers & Flores, 2005; Suzán, 2005). The recent changes in land use and human activities that have transformed much of REBISO into crop and livestock areas (Flamenco-Sandoval et al., 2007) have probably favored the populations of both D. rotundus and Artibeus spp., as evidenced by their high abundance and association with crop areas (Barquez, Perez, Miller, & Diaz, 2015; Miller, Reid, Arroyo-Cabrales, Cuarón, & de Grammont, 2016).
These findings should be interpreted with caution, as the genetic diversity values for bat species reported here could be modified if further studies include a larger sample size by species (in terms of both individuals and populations) using a different molecular marker. The smaller the number of individuals captured, the lower the possibility of obtaining genetic variants; thus, a small sample size may lead to underestimate the genetic diversity of at least some species. Initially, we were interested to obtain a diversity genetic measure for several bats species for which no specific molecular markers have been developed to date. For this reason, we selected the D-Loop for its suitability to be reproducible and variable across mammal species (Freeland, 2005; Hernández-Baños, Honey-Escandón, Cortés-Rodríguez, & García, 2007). This gene allowed having a genetic diversity measure to be correlated with ecological community factors, such as species richness and locality.
We found no significant correlation between genetic diversity based on variations in the D-Loop gene and functional group. It is possible that the relationship between the genetic diversity of a species and its ecological function within a community is mediated by geographical distribution and ecological interactions.
Kimura’s two-parameter genetic distances clustered species sharing the same ecological function, with a few exceptions (G. soricina, G. morenoi, C. sowelli, and P. parnellii), indicating that the number of nucleotide substitutions in these lineages is similar between species that coincide in feeding habits. The Kimura two-parameter genetic distances obtained in this study are consistent with those reported by Simmons, Seymour, Habersetzer, and Gunnell, (2008), who mention that insectivory is the ancestral feeding habit in the Chiroptera, while frugivores and nectarivores evolved subsequently.
The close relationship between frugivores and nectarivores may be due to the fact that they occasionally share food resources during periods of resource scarcity (fruits, nectar), according to the composition of the habitat and landscape heterogeneity (Calonge, 2009; Pedro & Taddei, 1997; Vleut, Levy-Tacher, de Boer, Galindo-González, & Vázquez, 2013; Vleut, Levy-Tacher, Galindo-González, & de Boer, 2015). In addition, frugivorous and nectarivorous bats are essentially tropical (Fleming, Geiselman, & Kress, 2009), a condition that could determine a converging evolutionary history of these functional groups (Hughes et al., 2008).
Within a community, species interact in time and space regardless of their taxonomic affinity (Martínez, 1996); these interactions may lead to feedback evolutionary processes (Genung et al., 2011). As a result, the evolutionary history of each species exerts a crucial effect on the genetic structure and diversity of communities (Vellend, 2005, 2006; Vellend & Geber, 2005), in ways not yet unveiled for the bat community in the tropical forest of REBISO. One possibility is that the evolutionary processes that govern the levels of genetic diversity are independent of their role in the ecosystem (Vellend, 2005; Vellend & Geber, 2005), and are probably determined by the historical, biological and behavioral characteristics of each species (Lamy et al., 2017). These factors can explain the genetic distances observed, with two different groups of frugivores, as well as isolated individual species.
An aspect that could influence genetic diversity at the community level is intra- and interspecific competition, which interferes with the strength of natural selection and is expected to increase genetic diversity (Vellend, 2008). Another factor of importance is the life history of the species (e.g., behavior, dispersal, feeding, and reproduction), as it determines the geographical distribution, population dynamics, and genetic polymorphism (Arboleda, 2008; Burns & Broders, 2014; Hedrick, 2000). Population size, flight capacity for searching food and colonization of new sites, reproduction (i.e., polygamous vs. monogamous), and tolerance to disturbance are all characteristics that influence the competitive ability (Meyer et al., 2016) and responses to selective pressures of individual species. These characteristics are unique to each species and may facilitate evolutionary processes (genetic drift and genetic flow) with different intensity and frequency over time (Moreira et al., 2014), leading to variable levels of genetic diversity and preventing the identification of parallel ecological-evolutionary processes between functional groups (Vellend & Geber, 2005; Whitlock, 2014).
It is important to underline that the limitations of the local geographical scale used may have not captured the genetic variation contained within each individual species to an extent that would allow an accurate identification of the genetic diversity patterns associated with functional groups (Jackson & Fahrig, 2014). Furthermore, the limited resources for fieldwork in this study, as well as the rarity and geographical distribution of hematophagous bats likely resulted in that our total sample failed to capture a number of hematophagous species equivalent to those recorded for nectarivorous, frugivorous, and insectivorous bats, likely lowering the statistical power of the comparisons between functional groups.
Social structure, mating systems, and past and current environmental conditions (Burns & Broders, 2014; Hedrick, 2000) are all important aspects to consider to explain the genetic structure of communities (Hoehn et al., 2008; Vellend, 2008); therefore these aspects should be included in subsequent analysis. Likewise, knowing the demography, niche breadth, and niche overlap between both species and functional groups will allow a better assessment of the association between genetic diversity and ecological function to unveil how these interactions influence genetic diversity patterns (Hedrick, 2000; Hughes et al., 2008; Vellend, 2008; Whitlock, 2014).
Implications for Conservation
A high genetic diversity was observed in the bat assemblages inhabiting REBISO, although some species may be more vulnerable than others to changes in their habitat (Ávila-Flores & Fenton, 2005; Bilgin, Karatas, Coraman, Disotell, & Morales, 2008; Meyer et al., 2016), in particular those with limited mobility or specialized ecological requirements (Martins, Ditchfield, Meyer, & Morgante, 2007). This may be the case of E. furinalis, which showed extremely low genetic diversity values. Although our results may be limited due to the low sample size of E. furinalis (N = 3), we suggest monitoring this species in REBISO. Local human activities can severely affect the isolation of populations and population size, both of which reduce genetic diversity and increase species vulnerability (Ripperger et al., 2013). Therefore, E. furinalis should be considered as priority species warranting close monitoring in REBISO.
El Ocote Biosphere Reserve is a key biological corridor for the fauna of moist tropical forests that harbor Neartic and Neotropical species and facilitates the genetic flow between two natural areas of paramount importance in Mexico and the world: Uxpanapa, Veracruz and Chimalapas, and Oaxaca (Flamenco-Sandoval et al., 2007). This research highlights the ecological relevance of El Ocote for its high bat species richness (Mendoza-Sáenz, 2016; Riechers, 2004, 2009) and genetic diversity.
Unfortunately, the changes in land use at REBISO are the primary factor of habitat loss and transformation (Flamenco-Sandoval et al., 2007), leading to the disruption of functional connectivity and the decreased evolutionary potential and survival of species and communities at a regional scale (Frankham et al., 2002; Hedrick, 2000; Hoehn et al., 2008; Jackson & Fahrig, 2014). All of this undermines the environmental services provided by the ecosystem, with important effects on the local human populations that subsist and benefit from these resources. Therefore, maintaining the biodiversity and ecological functionality of REBISO requires preserving the species diversity and genetic diversity of tropical bats through the control of the loss of habitat caused by changes in land use and human activities. This research is the first of its kind in REBISO, contributing to conservation based on the knowledge of biodiversity, and gives rise to a number of different questions about the patterns of genetic diversity and community genetics of tropical bats.
Acknowledgments
The authors are grateful to the communities of El Ocote Biosphere Reserve and the Comisión Nacional de Áreas Naturales Protegidas (National Commission of Protected Natural Areas) for their assistance in the development of this work. The authors are also grateful to Maurilio Jiménez Hernández for his assistance in sample processing in the laboratory and Dario Navarrete Gutiérrez for his support in the drafting of the location map. María Elena Sánchez-Salazar translated the manuscript into English.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was funded by the CONACYT project PDCPN-2013/214650-Biological and Social Vulnerability to Climate Change in El Ocote Biosphere Reserve.