Redundancy analysis: disentangling patterns of isolation by distance, environment, and barrier
We implemented a set of redundancy analyses (RDA) to disentangle the relative contribution of alternative hypotheses for explaining the genetic differentiation in A. cephalotes : IBD, IBE, and IBB. RDA is a canonical extension of PCA, in which the principal components are constrained to be linear combinations of a set of predictors (Bradburd, Ralph, & Coop, 2013; Legendre & Fortin, 2010; Meirmans, 2015). The goal of RDA here was to identify the best ordination model that describes genetic differentiation (James et al., 2011) in order to better understand how spatial heterogeneity across the Andes in the Colombian Pacific and Andean regions affects patterns of gene flow inA. cephalotes .
For the microsatellite data, a conventional RDA analysis was performed using population allele frequencies as a dependent matrix since the Euclidean distance estimates involved in the analysis are directly related to F ST values (as long as the frequencies are not scaled) (De Queiroz et al., 2017; Noguerales et al., 2016). ThemtCOI data were analyzed through a distance-based RDA, making direct use of a Φ ST-based genetic distance matrix. In both cases, space (geographic distances matrix), environment (climate variables), and topography (barriers) were considered explanatory variables. These explanatory variables were grouped into three classes according to their resulting patterns of isolation: 1) IBD, with variables representing the geographic distance between populations (space); 2) IBE, with variables determining environmental differences between populations; and 3) IBB, based on the western range as a geographic barrier that splits populations in an allopatric manner. Since space was the only explanatory variable initially expressed as a distance matrix, it was transformed into a vector format (Oksanen et al., 2019) by PCA, using the ‘pcnm’ function in the package VEGAN. Only the best explanatory PCNM components were retained. The significance of predictors was assessed using multivariateF -statistics with 10000 permutations using the ‘anova.cca ’ function included in the package VEGAN. All explanatory variables were scaled using the ‘scale’ function in the package VEGAN. The allele frequencies used were not scaled in order to keep their inter-population relation with the Φ ST.
Spatial explanatory variables were tested through an IBD analysis based on RDA, followed by sequential elimination of environmental variables, in which only variables with | r | < 0.80 were retained (Dormann et al. 2013; Supplementary figure SF2). The remaining variables were tested through a model selection approach, comparing all possible combinations or marginal tests to be finally included in the IBE analysis. Each marginal test was compared to the null model (intercept; AIC: 1.41). These combinations of environmental variables were used to identify the best model based on the Akaike information criteria (AIC). We selected the best model for significant predictors with the ‘ordistep’ function, using the package VEGAN to prevent overfitting. IBB was tested using the classification of sampled populations represented by a dummy variable (see sampling). Once all three models of genetic isolation were defined and tested, both IBE and IBB were tested while controlling for spatial autocorrelation through a partial RDA (conditional test). Finally, the relative contribution of IBD, IBE, and IBB to the explanation of gene flow patterns in A. cephalotes was evaluated from the final complete model, including all mechanisms of genetic isolation, using a variation partitioning analysis with the function ‘varpart ’ implemented in the package VEGAN. This analysis allowed us to disentangle total genetic variance according to the underlying mechanisms of isolation imposed by space, environment, and barriers, as well as their relative contributions when considered together.

RESULTS

Genetic diversity

All microsatellite loci followed expectations from the HWE or LD tests. Most loci were highly variable, with the number of alleles ranging from 2 to 20 per locus (Table 2, Supplementary table ST2). The Pacific region showed the highest allelic richness, largely due to 17 private alleles, while the Andean 1 region had the lowest allelic richness, with four private alleles. Most genetic diversity parameters were significantly higher for the Pacific region, except for the Heestimates (Figure 2A, Supplementary table ST4). An excess of heterozygotes was detected in the PPY population (IAM model: T= 3.40,P < 0.01; SSM model: T= 1.92, P = 0.03), while the allele frequencies differed from the L-shaped distribution. A putative bottleneck was supported by the low number of alleles per locus detected in this population (range: 1 to 6). Inbreeding coefficients did not differ significantly from zero, suggesting random mating in all populations (Table 2).
A 307 bp mtCOI gene fragment was successfully sequenced in 146 individuals. Ten nucleotide sites (3.3%) were polymorphic and defined 10 different haplotypes that differ in 1-3 positions (0.4% sequence divergence). The two most common haplotypes, Hap1 (frequency= 0.51) and Hap4 (frequency = 0.27), were found in almost all sampling locations. One haplotype was specific to the Pacific (Hap3), one to the Andean 1 (Hap6), and three to the Andean 2 (Hap8, Hap9, Hap10; Figure 2C, Supplementary Table ST5) regions. Overall, haplotype and nucleotide diversities were Hd= 66% (range: 0.0 – 76%) and π = 0.3% (range: 0.0 – 0.4%), respectively. In contrast to the microsatellite data, the Andean 1 region presented the highest diversity (Table 2, Figure 2B). However, the differences in mitochondrial diversity across regions were not significant (Figure 2B, Supplementary Table ST4).

Population structure for nuclear and mitochondrial data

Two regional models were used in the hierarchical AMOVA, following IBB and IBE scenarios (Table 3). For the microsatellite data, the results were similar under IBB and IBE scenarios, showing substantial differentiation among regions and populations within regions. Moreover, genetic differentiation was higher among regions when compared with that among populations, and was slightly higher under the IBE model when compared to the IBB model (Table 3).
Pairwise F ST estimates between populations ranged from 0.01 to 0.28 (mean = 0.10) and were significant following Bonferroni corrections, except for the BVT-QBD comparison in the Pacific region (Figure 3A). Pairwise comparisons between regions under the IBE model were significant, showing the highest differentiation values for comparisons involving the Andean 1 region (Pacific-Andean 1: 0.04; Pacific-Andean 2: 0.02; and Andean 1-Andean 2: 0.10), consistent with the DAPC analysis (Figure 4A). However, regional comparisons under the IBB model were also significant (F ST = 0.04).
Hierarchical AMOVA for mtCOI data indicated that populations within regions harbored 34% of the genetic variance, but no region effect was detected under the IBB and IBE models (Table 3). The pairwiseΦ ST values estimated among populations ranged from 0.00 to 0.86, and 56% of these comparisons were significant (Figure 3B). Moreover, DAPC clustering analysis also failed to group populations by regions (Supplementary figure SF3), supporting the AMOVA results.

Clustering analyses

Evidence for hierarchical population structure in microsatellite data was also found from the STRUCTURE runs. The ΔK ad hocstatistic by Evanno et al. (2005) indicated an optimal K = 3 [Pr(X / K : 2) = -5061.73] as the uppermost hierarchical level (Supplementary figure SF1). These three genetic clusters correspond to the individuals belonging to the three regions originally considered under the IBE model, except for the PPY population from the Andean 2 region. The first group (blue in Figure 4B) clustered most samples from the Pacific region. The second group (purple in Figure 4B) clustered most samples from the Andean 1 region, as well as samples from Andean 2 region (PPY). The third group (orange in Figure 4B) clustered most samples from Andean 2 region, apart from the PPY population. This pattern suggests a hierarchical north-south structure in the Andean region rather than a climate-related structure, as tested through the IBE-based AMOVA (Figure 4C), and suggests that IBD is influencing regional differentiation for the Andean regions rather than, or in addition to, IBE.
To further investigate the role of IBD in the genetic structure ofA. cephalotes , we performed a new AMOVA by moving PPY from the Andean 2 to the Andean 1 region. However, the results were similar to those produced under the original IBE scenario (results not shown). We also ran another STRUCTURE analysis, this time only including samples from the Andean 1 region and the PPY population (southern Andean cluster). This analysis revealed two genetic groups (K = 2), in which PPY formed a sole cluster (Figure 4B).
The results for the IBE model were supported by the DAPC analysis, which clustered defined regions based on environmental variables. The first two principal components of the DAPC analysis explained 77.1% of the variance in allele frequencies (50 PCs retained; Figure 4A). In this case, the first principal component clustered the Pacific and Andean regions, while the second principal component reflected the differentiation between Andean 1 and Andean 2 regions (Figure 4A).

RDA analysis – Alternative scenarios of genetic differentiation: IBD, IBE, and IBB

Significant IBD was detected from the RDA analysis for microsatellite data, in which the geographic distance expressed as the first axis of the PCNM analysis (PCNM1) contributed to the genetic structure ofA. cephalotes , based on model selection. This suggests that distance is an important determinant of genetic structure in A. cephalotes (Figure 5). These results agree with those obtained from STRUCTURE, suggesting that genetic differentiation in the Andean regions can also be explained by an IBD pattern (north to south) and not exclusively by IBE across regions (Figure 4B). The IBD results followed different patterns between nuclear and mitochondrial markers since IBD was not detected from the mtCOI data.
Although alternative models used for regional differentiation suggested the stronger influence of IBE, based on its higher regional genetic differentiation when compared to IBB or IBD, all of the models were significant (Table 3). This was also observed from the RDA results, where two environmental variables (temperature and precipitation) and the barrier variable (Andes classification) were significantly associated with genetic divergence (Table 4, Figure 5). Moreover, a significant contribution of alternative models of genetic differentiation was only observed from the microsatellite data (Supplementary table ST6). The optimal model, including all mechanisms of genetic isolation as tested for IBD, IBE, and IBB, accumulated up to 33% of explained variation, which remained marginally significant after accounting for IBD in conditional tests (Figure 5).
Variation partitioning for the full model, including all alternative models of genetic isolation, showed that IBD and IBE explained the highest proportion of genetic variation relative to IBB (Figure 5). Interaction effects between major mechanisms of genetic isolation were low for the triple interaction (IBD + IBE + IBB = 2%) and were not significant for paired interactions (Figure 5). The higher fraction of unexplained variation (residuals) might reflect the occurrence of genetic drift within populations, which is not associated with the explanatory variables.

DISCUSSION

The complex landscape across the Andean uplift is an important barrier, isolating populations and increasing genetic divergence on both sides of the mountains (Antonelli et al., 2009; Hoorn et al., 2010; Salgado-Roa et al., 2018). The western range of the Andes in Colombia is one of the most biodiverse ecosystems in the Neotropics (Kattan et al., 2004; Salgado-Roa et al., 2018) and is the most complex landscape within the distribution of the leaf-cutting ant A. cephalotes . By integrating hierarchical population structure with models of isolation by distance (IBD), environment (IBE), and barrier (IBB), we explored the role of the western mountain range in the distribution of genetic variation of A. cephalotes . Here, we demonstrated that the environmental heterogeneity imposed by the Andean uplift has highly influenced the population structure of A. cephalotes .

Genetic diversity

Our results indicate that the populations of A. cephalotes are highly variable at nuclear markers, with significant genetic differentiation at both region and population levels. In contrast, populations are highly differentiated for the mtCOI gene, but no region effect was detected with this marker. Ten mtCOI haplotypes were detected in 146 nests analyzed, showing low nucleotide and moderate haplotype diversities, with most populations dominated by the same haplotype (Hap1). As expected considering the common biogeographical history, similar mtCOI patterns have been detected in the leaf-cutting ant A. colombica (Helmkampf, Gadau, & Feldhaar, 2008), where six mtCOI haplotypes and relatively low levels of nucleotide diversity (0.1% population-wide) were detected across 20 colonies, with most specimens sharing the same haplotype. However, these results were obtained from a small sample size in a small area. Despite the fact that several other genetic studies have been reported in Attini (e.g., Acromyrmex ) (Cantagalli, Mangolin, & Ruvolo-Takasusuki, 2013; Diehl, Cavalli-molina, & Mellender de Araujo, 2002; Diehl, de Araújo, & Cavalli-Molina, 2001; Pinheiro dos Reis, Fernandes Salomão, de Oliveira Campos, & Garcia Tavares, 2014), the nature of the markers and sampling employed prevents us from drawing meaningful comparisons with these studies regarding diversity estimates.

Spatial differentiation at regional and local levels

We detected significant genetic differentiation across regions from the nuclear data, as evidenced from the AMOVA analyses and clustering methods, suggesting the potential role of the Andean landscape in restricting gene flow in A. cephalotes . We tested alternative scenarios causing regional differentiation, considering IBE as a model of divergence, as well as the geographic barrier imposed by the western range under an IBB model.
Regional differentiation from microsatellite data was higher under the IBE than the IBB model, suggesting the more prominent role of climate compared to the geographic barrier in restricting gene flow in A. cephalotes . On the other hand, the STRUCTURE results support a regional differentiation with mixed effects between IBE and IBD, evidencing a north-south spatial pattern for Andean populations rather than a regional climate model. However, this IBD pattern was likely due to a single population. PPY was initially classified within the Andean 2 region, but clustered mostly (Q > 0.95) with the geographically closer Andean 1 region. Nevertheless, this pattern occurs only in the Andean region, suggesting that population differentiation in this particular case results from dispersal distance rather than environmental conditions or a geographic barrier imposed by the western range. Medium to long dispersal distances have been reported forA. cephalotes , with queens flying up to 50 km during nuptial flights (Cherrett, 1968; Helms, 2018). This distance is at least one order of magnitude smaller than the inter-population distance used in this study, but it is also probably an underestimate since it was originally measured on a small island (Helms, 2018). Moreover, only a small number of individuals are required in order to homogenize population allele frequencies, and gene flow can also be conveyed stepwise. Although the genetic variance at the region level suggested IBE, all of the models of hierarchical differentiation were significant for the RDA analysis, suggesting the likelihood that all three leading causes of genetic isolation are contributing to the observed patterns of gene flow.
Significant nuclear differentiation at the population level suggests that other features besides proximity, environment, and major dispersal barriers act to restrict gene flow. The markedly increased variance in the PPY population is consistent with a recent population expansion signal. Whether the bottleneck signal is a product of a founder effect (Barton & Turelli, 2004; Matute, 2013; Parisod, Trippi, & Galland, 2005) or a recent change in effective population size cannot be determined with the current data. However, the proximity and clustering of PPY to Andean 1 populations rather than the Andean 2 region in the STRUCTURE results, suggests a recent founder effect or an admixture between the two Andean regions.
In contrast to the nuclear data, there was no regional differentiation in the mtCOI data, but populations were significantly differentiated within the regions in both the IBE and IBB models. Muñoz-Valencia et al. (2021) used mtCOI data to study the genetic structure of A. cephalotes from a larger geographic range covering most of the distribution of the species in South and Central America. These authors found significant genetic differentiation, both at the regional scale and among populations within major regions. Together, these results show that the eastern range of the Andes appears to be a major dispersal barrier driving regional differentiation, while the western range forms a relatively homogenous biogeographic area.

Disentangling the patterns of IBD, and IBE, IBB

Our RDA framework provides strong evidence for restricted gene flow inA. cephalotes across the western mountain range of the Colombian Andes by all of the mechanisms tested. Furthermore, we found that the explained genetic variation was maximized when all spatial and environmental variables associated with these mechanisms were included in the model. These multifactorial patterns are often expected in complex and heterogeneous environments (Sexton et al., 2014; Shafer & Wolf, 2013), such as in the Colombian Andes (Kattan et al., 2004; Pérez-Escobar et al., 2017; Salgado-Roa et al., 2018). Regions classified according to climate variation and geographic barriers follow IBE but also include IBB, supporting the higher differentiation estimate (F CT) obtained under the IBE model and the significance when both models are included in the RDA (IBE + IBB). Migration through low elevation passes in the western mountain range (Hernández-Camacho, 1992; Kattan et al., 2004) would allow gene flow between regions, explaining why such a major geographic barrier did not have a more pronounced effect. Spatial and environmental variables presumably associated with restriction to gene flow often suffer from spatial autocorrelation, challenging the differentiation of their underlying causes (Crispo et al., 2006; Edwards et al., 2012; Wang et al., 2013). Our RDA framework accounted for this issue by demonstrating strong IBE even after controlling for IBD; while IBB was not significant, even when ignoring IBD. Temperature and precipitation therefore appear to be the leading causes of IBE in A. cephalotes .
The significant effect of the environment on the genetic structure ofA. cephalotes populations indicates the potential of this species to adapt to local environmental conditions. This may occur when processes such as sex-biased dispersal (Edelaar & Bolnick, 2012) or selection against migrants (Hendry, 2004; Weber, Bradburd, Stuart, Stutz, & Bolnick, 2016; Wright, 1943) decrease the rate of gene flow, as has been observed in several other species (De Queiroz et al., 2017; Lee & Mitchell-Olds, 2011; Wang et al., 2013). Moreover, geographic distance and barriers in between can restrict gene flow by reducing dispersal efficiency, which is more associated with local genetic drift (Clémencet, Viginier, & Doums, 2005; Cross, Naugle, Carlson, & Schwartz, 2016; De Queiroz et al., 2017; Noguerales et al., 2016; Smith et al., 2018), leading to combined patterns of IBD and IBB. These complex patterns have been detected, for example, in the Amazonian common sardine Triportheus albus (De Queiroz et al., 2017). However, disentangling the individual environmental drivers was not possible due to strong correlations across environmental variables (Crispo et al., 2006; Edwards et al., 2012; Wang et al., 2013). The underlying causes of genetic isolation of populations have been rarely explored in leaf-cutting ants (Branstetter et al., 2017). In A. sexdens rubipilosa , geographic separation of populations alone did not explain population divergence (Cantagalli et al., 2013), and significant IBD and biome fragmentation imposed physical barriers to gene flow inA. robusta (Pinheiro dos Reis et al., 2014). All of these results together strongly suggest that the isolating mechanisms studied are not mutually exclusive (Crispo et al., 2006; Edwards et al., 2012; Wang et al., 2013) and are important factors in the evolution of Neotropical species.
Agricultural activities have significantly contributed to the colonization processes of leaf-cutting ants (de Carvalho Cabral, 2015). Conversion of their natural habitat into cultivated lands may favor population growth and dispersal of leaf-cutting ants (Montoya-Lerma, Giraldo-Echeverri, Armbrecht, Farji-Brener, & Calle, 2012; Schowalter & Ring, 2017). Queens of Atta laevigata have been shown to prefer nesting in open areas rather than closed forests (Vasconcelos 1997). Our exceptional study population PPY is located in a large city (Popayán). It is possible that the city infrastructure and constant human interference, rather than natural dispersal associated with nuptial flights, have influenced its genetic structure. This pattern is more compatible with human-mediated dispersal (Zheng, Yang, Zeng, Vargo, & Xu, 2018), where environmental changes associated with human expansion appear to promote population growth in leaf-cutting ants (Montoya-Lerma et al., 2012; Siqueira et al., 2017). However, PPY is most likely an exception to a general large-scale migration process inA. cephalotes .
The Andean uplift appears to modulate the population structure ofA. cephalotes in a more complex manner than previously thought. The eastern mountain range of the Andean uplift in Colombia plays a major role as a geographic barrier to historical gene flow, restricting the dispersion of A. cephalotes from north to south (Muñoz-Valencia et al., 2021). By exploring a finer scale across the western mountain range and incorporating neutral genetic markers with environmental variables, we clearly show that the observed genetic differentiation in A. cephalotes is mainly affected by the combined multifactorial effect of different isolating mechanisms, mediated by the landscape complexity. Together, these results suggest that studies directed at exploring historical gene flow across the Andes should be interpreted with caution, since the complexity and history of this landscape can dramatically influence results at different scales.

ACKNOWLEDGEMENTS

We thank Kirsi Kähkönen, the ESB (Evolution, Sociality, and Behavior) research group, and everyone in the Molecular Ecology and Systematics laboratory (University of Helsinki) for helping to develop the DNA microsatellite loci. We also thank Sandra Milena Valencia Giraldo, Andrea López-Peña, and Glever Alexander Vélez-Martínez for helping with the DNA extraction. This work was funded by the Vice-Rectoría de Investigaciones, Universidad del Valle, Cali, Colombia (grant number: CI71067); and COLCIENCIAS National Program of PhD (grant number: 617-2013).

REFERENCES

Antonelli, A., Quijada-Mascareñas, A., Crawford, A. J., Bates, J. M., Velazco, P. M., & Wüster, W. (2009). Molecular studies and phylogeography of Amazonian tetrapods and their relation to geological and climatic models. In C. Hoorn & F. Wesselingh (Eds.),Amazonia, landscape and species evolution: A look into the past(pp. 386–404). London, UK: Blackwell Publishing Ltd.
Barton, N. H., & Turelli, M. (2004). Effects of genetic drift on variance components under a general model of epistasis.Evolution , 58 (10), 2111–2132. doi: 10.1111/j.0014-3820.2004.tb01591.x
Bradburd, G. S., Ralph, P. L., & Coop, G. M. (2013). Disentangling the effects of geographic and ecological isolation on genetic differentiation. Evolution , 67 (11), 1–25. doi: 10.1111/evo.12193.Disentangling
Branstetter, M. G., Jesovnik, A., Sosa-Calvo, J., Lloyd, M. W., Faircloth, B. C., Brady, S. G., & Schultz, T. R. (2017). Dry habitats were crucibles of domestication in the evolution of agriculture in ants.Proceedings of the Royal Society B , 284 , 20170095.
Cadena, C. D., Pedraza, C. A., & Brumfield, R. T. (2016). Climate, habitat associations and the potential distributions of Neotropical birds: Implications for diversification across the Andes. Revista de La Academia Colombiana de Ciencias Exactas, Físicas y Naturales ,40 (155), 275. doi: 10.18257/raccefyn.280
Cantagalli, L. B., Mangolin, C. A., & Ruvolo-Takasusuki, M. C. C. (2013). Population genetics of Atta sexdens rubropilosa (Hymenoptera: Formicidae). Acta Biológica Colombiana , 18 (1), 179–190.
Carro, B., Quintela, M., Ruiz, J. M., & Barreiro, R. (2019). Wave exposure as a driver of isolation by environment in the marine gastropod Nucella lapillus. Hydrobiologia , 839 , 51–69. doi: 10.1007/s10750-019-03993-5
Chambers, J. M., Freeny, A., & Heiberger, R. M. (1992). Analysis of variance; designed experiments. In J. M. Chambers & T. J. Hastie (Eds.), Statistical models . doi: 10.1007/978-3-642-50096-1
Chapuis, M. P., & Estoup, A. (2007). Microsatellite null alleles and estimation of population differentiation. Molecular Biology and Evolution , 24 (3), 621–631. doi: 10.1093/molbev/msl191
Chen, D., & Chen, H. W. (2013). Using the Köppen classification to quantify climate variation and change: An example for 1901-2010.Environmental Development , 6 (1), 69–79. doi: 10.1016/j.envdev.2013.03.007
Cherrett, J. (1968). A flight record for queens of Atta cephalotes L. (Hym. Formicidae). Entomologist ’s Monthly Magazine , 104 , 255–256.
Clémencet, J., Viginier, B., & Doums, C. (2005). Hierarchical analysis of population genetic structure in the monogynous ant Cataglyphis cursor using microsatellite and mitochondrial DNA markers. Molecular Ecology , 14 (12), 3735–3744. doi: 10.1111/j.1365-294X.2005.02706.x
Crispo, E., Bentzen, P., Reznick, D. N., Kinnison, M. T., & Hendry, A. P. (2006). The relative influence of natural selection and geography on gene flow in guppies. Molecular Ecology , 15 (1), 49–62. doi: 10.1111/j.1365-294X.2005.02764.x
Cross, T. B., Naugle, D. E., Carlson, J. C., & Schwartz, M. K. (2016). Hierarchical population structure in greater sage-grouse provides insight into management boundary delineation. Conservation Genetics , 17 (6), 1417–1433. doi: 10.1007/s10592-016-0872-z
De-Silva, D. L., Mota, L. L., Chazot, N., Mallarino, R., Silva-Brandão, K. L., Gómez-Piñerez, L. M., … Elias, M. (2017). North Andean origin and diversification of the largest ithomiine butterfly genus.Scientific Reports , 7 , 1–17. doi: 10.1038/srep45966
de Carvalho Cabral, D. (2015). Into the bowels of tropical earth: Leaf-cutting ants and the colonial making of agrarian Brazil.Journal of Historical Geography , 50 , 92–105. doi: 10.1016/j.jhg.2015.06.014
De Queiroz, L. J., Torrente-Vilara, G., Quilodran, C., da Costa Doria, C. R., & Montoya-Burgos, J. I. (2017). Multifactorial genetic divergence processes drive the onset of speciation in an Amazonian fish.PLoS ONE , 12 (12), 1–27. doi: 10.1371/journal.pone.0189349
Della Lucia, T. M., Gandra, L. C., & Guedes, R. N. (2014). Managing leaf-cutting ants: Peculiarities, trends and challenges. Pest Management Science , 70 (1), 14–23. doi: 10.1002/ps.3660
Diehl, E., Cavalli-molina, S., & Mellender de Araujo, A. (2002). Isoenzyme variation in the leaf-cutting ants Acromyrmex heyeri and Acromyrmex striatus (Hymenoptera, Formicidae). Genetics and Molecular Biology , 25 (2), 173–178.
Diehl, E., de Araújo, A. M., & Cavalli-Molina, S. (2001). Genetic variability and social structure of Colonies in Acromyrmex heyeri and A. striatus (Hymenoptera: Formicidae). Brazilian Journal of Biology = Revista Brasleira de Biologia , 61 (4), 667–678. doi: 10.1590/S1519-69842001000400017
Dormann, C. F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carré, G., … Lautenbach, S. (2013). Collinearity: A review of methods to deal with it and a simulation study evaluating their performance.Ecography , 36 (1), 027–046. doi: 10.1111/j.1600-0587.2012.07348.x
Earl, D. A., & vonHoldt, B. M. (2012). STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conservation Genetics Resources , 4 (2), 359–361. doi: 10.1007/s12686-011-9548-7.
Edelaar, P., & Bolnick, D. I. (2012). Non-random gene flow: An underappreciated force in evolution and ecology. Trends in Ecology and Evolution , 27 (12), 659–665. doi: 10.1016/j.tree.2012.07.009
Edwards, D. L., Keogh, J. S., & Knowles, L. L. (2012). Effects of vicariant barriers, habitat stability, population isolation and environmental features on species divergence in the south-western Australian coastal reptile community. Molecular Ecology ,21 (15), 3809–3822. doi: 10.1111/j.1365-294X.2012.05637.x
Evanno, G., Regnaut, S., & Goudet, J. (2005). Detecting the number of clusters of individuals using the software STRUCTURE: A simulation study. Molecular Ecology , 14 (8), 2611–2620. doi: 10.1111/j.1365-294X.2005.02553.x
Excoffier, L., & Lischer, H. E. L. (2010). Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows. Molecular Ecology Resources , 10 (3), 564–567. doi: 10.1111/j.1755-0998.2010.02847.x
Excoffier, L., Smouse, P. E., & Quattro, J. M. (1992). Analysis of molecular variance inferred from metric distances among DNA haplotypes: Application to human mitochondiral DNA restriction data.Genetics , 131 , 479–491. doi: 10.1007/s00424-009-0730-7
Fernández, F., Castro-Huertas, V., & Serna, F. (2015). Hormigas cortadoras de hojas de Colombia: Acromyrmex & Atta (Hymenoptera: Formicidae). In Fauna de Colombia (Monografía). Bogotá: Instituto de Ciencias Naturales, Universidad Nacional de Colombia.
Goudet, J. (1995). FSTAT (Version 1.2): A Computer program to Calculate F-Statistics. Journal of Heredity , pp. 485–486.
Goudet, J. (2001). FSTAT, a program to estimate and test gene diversities and fixation indices (version 2.9.3) . Retrieved from http://www.unil.ch/izea/softwares/fstat.html
Haffer, J. (2008). Hypotheses to explain the origin of species in Amazonia. Brazilian Journal of Biology , 68 (4 SUPPL.), 917–947. doi: 10.1590/S1519-69842008000500003
Hakala, S. M., Seppä, P., & Helanterä, H. (2019). Evolution of dispersal in ants (Hymeno ptera: Formicidae): a review on the dispersal strategies of sessile superorganisms Sanja. Myrmecological News ,29 (March), 35–55. doi: 10.25849/myrmecol.news
Helmkampf, M., Gadau, J., & Feldhaar, H. (2008). Population- and sociogenetic structure of the leaf-cutter ant Atta colombica (Formicidae, Myrmicinae). Insectes Sociaux , 55 (4), 434–442. doi: 10.1007/s00040-008-1024-3
Helms, J. (2018). The flight ecology of ants (Hymenoptera: Formicidae).Myrmecological News , 26 , 19–30. doi: 10.25849/myrmecol.news
Hendry, A. P. (2004). Selection against migrants contributes to the rapid evolution of ecologically dependent reproductive isolation.Evolutionary Ecology Research , 6 (8), 1219–1236.
Hermelin, M. (2015). Landscapes and landforms of Colombia (M. Hermelin, Ed.). Springer International Publishing.
Hernández-Camacho, J. (1992). Caracterización geográfica de Colombia. In G. Halffter (Ed.), La diversidad biológica de Iberoamérica I(Programa I, pp. 45–52). Acta Zoológica Mexicana.
Hölldobler, B., & Wilson, E. O. (2011). The leafcutter ants: civilization by instinct (1st ed.). Nueva York, NY: W. W. Norton & Company.
Hoorn, C., Wesselingh, F. P., Ter Steege, H., Bermudez, M. A., Mora, A., Sevink, J., … Antonelli, A. (2010). Amazonia through time: Andean uplift, climate change, landscape evolution, and biodiversity.Science , 330 , 927–931. doi: 10.1126/science.1194585
James, P. M. A., Coltman, D. W., Murray, B. W., Hamelin, R. C., & Sperling, F. A. H. (2011). Spatial genetic structure of a symbiotic beetle-fungal system: Toward multi-taxa integrated landscape genetics.PLoS ONE , 6 (10), 1–11. doi: 10.1371/journal.pone.0025359
Jombart, T., Devillard, S., & Balloux, F. (2010). Discriminant analysis of principal components: a new method for the analysis of genetically structured populations. BMC Genetics , 11 (94), 1–15. doi: 10.1371/journal.pcbi.1000455
Kattan, G. H., Franco, P., Rojas, V., & Morales, G. (2004). Biological diversification in a complex region: A spatial analysis of faunistic diversity and biogeography of the Andes of Colombia. Journal of Biogeography , 31 (11), 1829–1839. doi: 10.1111/j.1365-2699.2004.01109.x
Kopelman, N. M., Mayzel, J., Jakobsson, M., Rosenberg, N. A., & Mayrose, I. (2015). CLUMPAK: a program for identifying clustering modes and packaging population structure inferences across K. Molecular Ecology Resources , 15 (5), 1179–1191. doi: 10.1111/1755-0998.12387.CLUMPAK
Köppen, W. (1884). Die Wärmezonen der Erde, nach der Dauer der heissen, gemässigten und kalten Zeit und nach der Wirkung der Wárme auf die organische Welt betrachtet (The thermal zones of the earth according to the duration of hot, moderate and cold periods and to the impac.Meteorologische Zeitschrift , 1 , 215–226. doi: 10.1127/0941-2948/2011/105
Kronauer, D. J. C., Hölldobler, B., & Gadau, J. (2004). Phylogenetics of the new world honey ants (genus Myrmecocystus) estimated from mitochondrial DNA sequences. Molecular Phylogenetics and Evolution , 32 (1), 416–421. doi: 10.1016/j.ympev.2004.03.011
Lagomarsino, L. P., Condamine, F. L., Antonelli, A., Mulch, A., & Davis, C. C. (2016). The abiotic and biotic drivers of rapid diversification in Andean bellflowers (Campanulaceae). New Phytologist , 210 , 1430–1442. doi: 10.1111/nph.13920
Lee, C.-R., & Mitchell-Olds, T. (2011). Quantifying effects of environmental and geographical factors on patterns of genetic differentiation. Molecular Ecology , 20 (22), 4631–4642. doi: 10.5061/dryad.6rs51
Legendre, P., & Fortin, M.-J. (2010). Comparison of the Mantel test and alternative approaches for detecting complex multivariate relationships in the spatial analysis of genetic data. Molecular Ecology Resources , 10 (5), 831–844. doi: 10.1111/j.1755-0998.2010.02866.x
Librado, P., & Rozas, J. (2009). DnaSP v5: A software for comprehensive analysis of DNA polymorphism data. Bioinformatics , 25 (11), 1451–1452. doi: 10.1093/bioinformatics/btp187
Luebert, F., & Weigend, M. (2014). Phylogenetic insights into Andean plant diversification. Frontiers in Ecology and Evolution ,2 (27), 1–17. doi: 10.3389/fevo.2014.00027
Manel, S., & Holderegger, R. (2013). Ten years of landscape genetics.Trends in Ecology and Evolution , 28 (10), 614–621. doi: 10.1016/j.tree.2013.05.012
Manel, S., Schwartz, M. K., Luikart, G., & Taberlet, P. (2003). Landscape genetics: combining landscape ecology and population genetics.Trends in Ecology and Evolution , 18 (4), 189–197. doi: 10.1016/S0169-5347(03)00008-9
Matute, D. R. (2013). The role of founder effects on the evolution of reproductive isolation. Journal of Evolutionary Biology ,26 (11), 2299–2311. doi: 10.1111/jeb.12246
Meirmans, P. G. (2006). Using the AMOVA framework to estimate a standardized genetic differentiation measure. Evolution ,60 (11), 2399–2402.
Meirmans, P. G. (2015). Seven common mistakes in population genetics and how to avoid them. Molecular Ecology , 24 , 3223–3231. Retrieved from http://dx.doi.org/10.1111/mec.13243
Meirmans, P. G., & Hedrick, P. W. (2011). Assessing population structure: FSTand related measures. Molecular Ecology Resources ,11 (1), 5–18. doi: 10.1111/j.1755-0998.2010.02927.x
Montoya-Lerma, J., Giraldo-Echeverri, C., Armbrecht, I., Farji-Brener, A., & Calle, Z. (2012). Leaf-cutting ants revisited: Towards rational management and control. International Journal of Pest Management ,58 (3), 225–247. doi: 10.1080/09670874.2012.663946
Moser, J. C. (1967). Mating activities of Atta texana (Hymenoptera, Fromicidae). Insectes Sociaux , XIV (3), 295–312.
Mueller, U. G., Ishak, H. D., Bruschi, S. M., Smith, C. C., Herman, J. J., Solomon, S. E., … Bacci, M. J. (2017). Biogeography of mutualistic fungi cultivated by leafcutter ants. Molecular Ecology , (November), 1–17. doi: 10.1111/mec.14431
Muñoz-Valencia, V., Kähkönen, K., Montoya-Lerma, J., & Díaz, F. (2020). Characterization of a New Set of Microsatellite Markers Suggests Polygyny and Polyandry in Atta cephalotes (Hymenoptera: Formicidae).Journal of Economic Entomology , 113 (6), 3021–3027. doi: 10.1093/jee/toaa200
Muñoz-Valencia, V., Vélez-Matínez, G. A., Montoya-Lerma, J., & Díaz, F. (2021). Role of the Andean uplift as an asymmetrical barrier to gene flow in the neotropical leaf-cutting ant Atta cephalotes.Biotropica , (October), 1–14. doi: 10.1111/btp.13050
Noguerales, V., Cordero, P. J., & Ortego, J. (2016). Hierarchical genetic structure shaped by topography in a narrow-endemic montane grasshopper. BMC Evolutionary Biology , 16 (96), 1–15. doi: 10.1186/s12862-016-0663-7
Oksanen, J., Blanchet, F. G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., … Wagner, H. (2019). Vegan: Community Ecology Package. R Package Version 2.2-0. In Community ecology package . Retrieved from https://cran.r-project.org/web/packages/vegan/vegan.pdf
Parisod, C., Trippi, C., & Galland, N. (2005). Genetic variability and founder effect in the pitcher plant Sarracenia purpurea (Sarraceniaceae) in populations introduced into Switzerland: from inbreeding to invasion.Annals of Botany , 95 (2), 277–286. doi: 10.1093/aob/mci023
Peakall, R., & Smouse, P. (2012). GenAlEx 6.5: Genetic analysis in Excel. Population genetic software for teaching and research.Bioinformatics , 1 , 6–8.
Peel, M. C., Finlayson, B. L., & McMahon, T. A. (2007). Update world map of the Köppen-Geiger climate classification. Hydrology and Earth System Sciences , 11 , 1633–1644. doi: 10.1002/ppp.421
Pérez-Escobar, O. A., Gottschling, M., Chomicki, G., Condamine, F. L., Klitgård, B. B., Pansarin, E., & Gerlach, G. (2017). Andean mountain building did not preclude dispersal of lowland epiphytic orchids in the Neotropics. Scientific Reports , 7 (1), 1–10. doi: 10.1038/s41598-017-04261-z
Pinheiro dos Reis, E., Fernandes Salomão, T. M., de Oliveira Campos, L. A., & Garcia Tavares, M. (2014). Genetic diversity and structure of Atta robusta (Hymenoptera, Formicidae, Attini), an endangered species endemic to the Restinga ecoregion. Genetics and Molecular Biology , 37 (3), 581–586. doi: 10.1590/S1415-47572014000400015
Pritchard, J. K., Stephens, M., & Donnelly, P. (2000). Inference of population structure using multilocus genotype data. Genetics ,155 (2), 945–959. doi: 10.1111/j.1471-8286.2007.01758.x
Ramos-Onsins, S. E., & Rozas, J. (2002). Statistical properties of new neutrality tests against population growth. Molecular Biology and Evolution , 19 (12), 2092–2100. doi: Doi 10.1093/Molbev/Msl052
Rousset, F. (2008). GENEPOP’007: A complete re-implementation of the GENEPOP software for Windows and Linux. Molecular Ecology Resources , 8 (1), 103–106. doi: 10.1111/j.1471-8286.2007.01931.x
Rull, V. (2011). Neotropical biodiversity: timing and potential drivers.Trends in Ecology and Evolution , 26 (10), 508–513. doi: 10.1016/j.tree.2011.05.011
Salgado-Roa, F. C., Pardo-Diaz, C., Lasso, E., Arias, C. F., Solferini, V. N., & Salazar, C. (2018). Gene flow and Andean uplift shape the diversification of Gasteracantha cancriformis (Araneae: Araneidae) in Northern South America. Ecology and Evolution , 8 (14), 7131–7142. doi: 10.1002/ece3.4237
Schowalter, T. D., & Ring, D. R. (2017). Biology and management of the Texas leafcutting ant (Hymenoptera: Formicidae). Journal of Integrated Pest Management , 8 (1), 1–8. doi: 10.1093/jipm/pmx013
Sexton, J. P., Hangartner, S. B., & Hoffmann, A. A. (2014). Genetic isolation by environment or distance: which pattern of gene flow is most common? Evolution , 68 (1), 1–15. doi: 10.1111/evo.12258
Shafer, A. B. A., & Wolf, J. B. W. (2013). Widespread evidence for incipient ecological speciation: a meta-analysis of isolation-by-ecology. Ecology Letters , 16 (7), 940–950. doi: 10.1111/ele.12120
Simon, C., Frati, F., Beckenbach, A., Crespi, B., Liu, H., & Flook, P. (1994). Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Annals of the Entomological Society of America ,87 (6), 651–701. doi: 10.1093/aesa/87.6.651
Siqueira, F. F. S., Ribeiro-Neto, J. D., Tabarelli, M., Andersen, A. N., Wirth, R., & Leal, I. R. (2017). Leaf-cutting ant populations profit from human disturbances in tropical dry forest in Brazil. Journal of Tropical Ecology , 33 (5), 337–344. doi: 10.1017/S0266467417000311
Slatkin, M. (1993). Isolation by distance in equilibrium and non-equilibrium populations. Evolution , 47 (1), 264–279. Retrieved from http://www.jstor.org/stable/2410134
Smith, C. C., Weber, J. N., Mikheyev, A. S., Roces, F., Bollazzi, M., Kellner, K., … Mueller, U. G. (2018). Landscape genomics of an obligate mutualism: discordant population structures between a leafcutter-ant and its fungal cultivars. BioRxiv , 1–35. doi: 10.1101/458950
Sobel, J. M. (2014). Ecogeographic isolation and speciation in the genus Mimulus. The American Naturalist , 184 (5), 565–579. doi: 10.1086/678235
Turchetto-Zolet, A., Pinheiro, F., Salgueiro, F., & Palma-Silva, C. (2013). Phylogeographical patterns shed light on evolutionary process in South America. Molecular Ecology , 22 , 1193–1213. doi: 10.1111/mec.12164
VanOosterhout, C., Hutchinson, W. F., Wills, D. P. M., & Shipley, P. (2004). MICRO-CHECKER: Software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes , 4 (3), 535–538. doi: 10.1111/j.1471-8286.2004.00684.x
Vasconcelos, H. L. (1997). Foraging activity of an Amazonian leaf-cutting ant: Responses to changes in the availability of woody plants and to previous plant damage. Oecologia , 112 (3), 370–378. doi: 10.1007/s004420050322
Wang, I. J., & Bradburd, G. S. (2014). Isolation by environment.Molecular Ecology , 23 (23), 5649–5662. doi: 10.1111/mec.12938
Wang, I. J., Glor, R. E., & Losos, J. B. (2013). Quantifying the roles of ecology and geography in spatial genetic divergence. Ecology Letters , 16 (2), 175–182. doi: 10.1111/ele.12025
Wasko, A. P., Martins, C., Oliveira, C., & Foresti, F. (2003). Non-destructive genetic sampling in fish. An improved method for DNA extraction from fish fins and scales. Hereditas , 138 , 161–165.
Weber, J. N., Bradburd, G. S., Stuart, Y. E., Stutz, W. E., & Bolnick, D. I. (2016). Partitioning the effects of isolation by distance, environment, and physical barriers on genomic divergence between parapatric threespine stickleback. Evolution ,712 , 342–356. doi: 10.1111/evo.13545
Wright, S. (1943). Isolation by distance. Genetics , 28 , 114–138.
Zheng, C., Yang, F., Zeng, L., Vargo, E. L., & Xu, Y. (2018). Genetic diversity and colony structure of Tapinoma melanocephalum on the islands and mainland of South China. Ecology and Evolution , 8 (11), 5427–5440. doi: 10.1002/ece3.4065

DATA ACCESSIBILITY STATEMENT

The data that support the findings of this study will be openly available in DataDryad after manuscript is accepted for publication.

AUTHORS’ CONTRIBUTIONS

VMV participated in the conceptualization, ant sampling, data analysis, original draft, reviewing, and editing of the manuscript. JML participated in conceptualization, original draft, reviewing, and editing of the manuscript. PS participated in data analysis, reviewing, and editing of the manuscript. FD participated in conceptualization, data analysis, original draft, reviewing, and editing of the manuscript. All authors have read and approved the final version of the manuscript.

TABLES AND FIGURES

Table 1. Sampled populations of A. cephalotes. Locations with their regional classifications (Pacific, Andean 1, and Andean 2 regions from Colombia) and the number of nests are indicated for each population.