Introgressive hybridization in a Spiny-Tailed Iguana, Ctenosaura pectinata, and its implications for taxonomy and conservation

Introgression, the transmission of genetic material of one taxon into another through hybridization, can have various evolutionary outcomes. Previous studies have detected signs of introgression between western populations of the Mexican endemic and threatened spiny-tailed iguana, Ctenosaura pectinata. However, the extent of this phenomenon along the geographic distribution of the species is unknown. Here, we use multilocus data together with detailed geographic sampling to (1) define genotypic clusters within C. pectinata; (2) evaluate geographic concordance between maternally and biparentally inherited markers; (3) examine levels of introgression between genotypic clusters, and (4) suggest taxonomic modifications in light of this information. Applying clustering methods to genotypes of 341 individuals from 49 localities of C. pectinata and the closely related C. acanthura, we inferred the existence of five genotypic clusters. Contact zones between genotypic clusters with signatures of interbreeding were detected, showing different levels of geographic discordance with mtDNA lineages. In northern localities, mtDNA and microsatellites exhibit concordant distributions, supporting the resurrection of C. brachylopha. Similar concordance is observed along the distribution of C. acanthura, confirming its unique taxonomic identity. Genetic and geographic concordance is also observed for populations within southwestern Mexico, where the recognition of a new species awaits in depth taxonomic revision. In contrast, in western localities a striking pattern of discordance was detected where up to six mtDNA lineages co-occur with only two genotypic clusters. Given that the type specimen originated from this area, we suggest that individuals from western Mexico keep the name C. pectinata. Our results have profound implications for conservation, management, and forensics of Mexican iguanas.

). This network constitutes 244 an update from the one produced in 2011 (Zarza, Reynoso & Emerson, 2011). The new 245 haplotypes connected to haplotypes in the North A clade, and did not alter the previously 246 observed patterns. In the SAMOVA analyses, a change of less than 1% in FCT was observed at 247 K=10 (Table A and   Manuscript to be reviewed 299 analyses (Nuc 3a and Nuc 3b in Fig. 2 E). However, several individuals of Nuc 3a and 3b 300 showed admixed ancestry, indicating weak geographic structure (Fig. 2 E). The division between 301 Nuc 3a and 3b was not detected with SAMOVA (Fig. S2). Two other clusters were identified 302 with the South-SS analyses, one equivalent to *Nuc 4 and the other comprising individuals 303 identified as C. acanthura and equivalent to *Nuc 5 (Fig. S2). Individuals forming these two 304 clusters were consistently assigned among runs and in accordance with the assignment observed 305 when analyzing the entire dataset.

306
Given the weak geographic structure observed between Nuc 3a and Nuc 3b and the lack 307 of support for such subdivision with SAMOVA, we take a conservative approach and consider 308 these as forming only one genotypic cluster (equivalent to *Nuc 3 and Nuc 3). Both SAMOVA 309 and STRUCTURE support the distinction between *Nuc 4 (Nuc 4) and *Nuc 5 (Nuc 5, in the 310 South-SS analyses). Taking into account the results of SAMOVA and STRUCTURE we 311 recognize a total of five microsatellite genotypic clusters within the entire distribution of 312 Ctenosaura pectinata + C. acanthura ( Fig. 1 and Fig. 2).
316 However, the presence of introgression is supported by the hybrid indices calculated with 317 NewHybrids between SAMOVA genotype clusters (Table 4) (Table 4). Almost 50% of the individuals forming these clusters could not be 341 assigned to any category. Among these, 83 individuals showed a posterior probability <0.2 of 342 belonging to any of the pure classes, thus they might be hybrids of some sort. F ST values between 343 these genotypic clusters are the lowest observed in the pairwise comparisons (Table 3).  Table 3). Assignment of individuals to pure and hybrid classes also shows that contact 403 zones have different hybrid compositions. A higher proportion of individuals were assigned to a 404 pure class when analyzing *Nuc 1 and *Nuc 2 (89%) than when analyzing *Nuc 2 and *Nuc 3 405 (36%). This is also observed in the STRUCTURE plots which reveal Nuc 2 and Nuc 3 admixed 406 individuals more frequently than admixed Nuc 1 and Nuc 2.

407
Thus Ctenosaura pectinata constitutes an excellent system to better understand the   Fig. 1 and Fig. 2).

421
The observed concordance in the geographic distribution of nuclear and mtDNA might be 422 the result of stochastic coalescent processes, which is particularly true in taxa with low dispersal 423 rates, as is the case for iguanas (Irwin, 2002).  Manuscript to be reviewed Figure 1 Geographic distribution of mtDNA lineages and genotypic clusters within Ctenosaura pectinata and C. acanthura.
Lines represent the geographical limits of the mtDNA lineages. Colors and lineage names follow the scheme shown in the haplotype network (Fig. S1)