Introduction

Genetic tools are broadly utilized to gain understanding of wild populations and their conservation needs; e.g., examining metapopulation dynamics (Rowe et al. 2003), tracking illegal collection and trade (Coetzer et al. 2017), and characterizing genetic variation and inbreeding (Luís et al. 2007; Engelhardt et al. 2017). Although development of novel techniques continues, 2nd generation methodologies (microsatellites) are still broadly applied (Vashistha et al. 2020) as these are overall cheap to obtain data from, are preferred for small sample sizes, and can be processed and analyzed rapidly. Especially within conservation science, where funding is limited and results tend to be time sensitive, microsatellites are an important go-to genetic tool.

Among Iguaninae, Iguana is the widest ranging genus, with mainland populations occurring from central Mexico through to central Brazil, and widespread presence on Caribbean islands (Belezian, Colombian, Honduran, Mexican, Nicaraguan, and Tobago and Trinidad islands, as well as in the Southern and Lesser Antilles; Iguana Taxonomy Working Group 2016; Hedges et al. 2019). Taxonomically, this genus consists of two well defined lineages, I. delicatissima and the I. iguana complex. Within the latter, hypotheses of species boundaries are unresolved, though are currently under investigation, see Table 1 for nomenclature and geographic ranges (Stephen et al. 2013; Breuil et al. 2019, 2020, submitted, van den Burg et al. in prep.).

Table 1 Range of extent for currently named Iguana species and lineages, including sample sizes as in IguanaBase v1 (van den Burg et al. 2021a)
Full size table

As is true for many Iguanids (IUCN 2020), insular lineages within Iguana are highly endangered. The main threat to these species is the continuous spread and recurrent incursion of non-native iguanas, both through natural events and, since at least the 19th Century (Breuil 2003, 2013), anthropogenic assistance, which is expected to continue in the absence of stricter biosecurity measures (Falcón et al. 2012; Bock et al. 2018; van den Burg et al. 2018a, 2020a, 2020b, 2021b; Breuil et al. 2019, 2020; Knapp et al. 2020). These invasive, non-native iguanas have profoundly impacted native insular iguana populations throughout the Lesser Antilles, as the former have larger clutch and body sizes and show higher aggressiveness than native iguanas (van den Burg et al. 2018a; van Wagensveld and van den Burg 2018; Breuil et al. 2019). These incursions from different nonnative sources have resulted in region-wide hybridization, best documented for the Critically Endangered I. delicatissima (Vuillaume et al. 2015; van den Burg et al. 2018a, 2018b; Pounder et al. 2020), and to a lesser extent for the distinct lineages within the I. iguana complex (Breuil et al. 2019). Given this lengthy, continuous period of hybridization and subsequent within-region movement of non-native iguanas (Censky et al. 1998), morphologically distinguishing between native and (hybridized) non-native iguanas is often not straightforward (Breuil 2013; Vuillaume et al. 2015). This status quo clearly demonstrates the urgent need for robust genetic tools to distinguish native from non-native (or hybrid) animals. Additionally, the increased activity in illegal trade of iguanas from Lesser Antillean islands (Noseworthy 2017; van den Burg and Weissgold 2020) further highlights the need for island-level identification.

Within Iguana, microsatellite research has focused on a wide range of topics; invasion biology, population dynamics, as well as conservation (Valette et al. 2013; Vuillaume et al. 2015; Martin Judson et al. 2018; van den Burg et al. 2018b; De Jesús Villanueva et al. 2021). However, different genetic laboratories have not standardized their data sets and employ different loci sets (Table 2), limiting the vital ability to compare results, efficiently build upon prior work, and improve our understanding of these vulnerable populations. This knowledge gap of variation from across the genus’ range limits our ability to meaningfully assess population structure within lineages, and hinders the essential identification of Iguana hybrids and conservation of endangered lineages. From the perspective of hybridization issues, the research community currently lacks an understanding of allelic variation among known Iguana species and clades because data from native samples are almost entirely limited to island populations. The one exception is a 2015 study by Vuillaume et al. that included a single sample from each of two mainland localities with known origin. Hence, a genus-wide dataset is urgently needed in order to provide a foundation upon which all stakeholders can build, and use to guide conservation management decisions.

Table 2 Research using any subset of the 25 microsatellite loci identified by Valette et al. (2013)
Full size table

Materials and methods

The microsatellite loci set that has been most implemented among Iguana studies is that of Valette et al. (2013). We compiled data generated from all or a subset of these 25 loci (Valette et al. 2013; Vuillaume et al. 2015; Breuil et al. 2019, 2020, submitted). Furthermore, we collected data from new mainland (23) and insular (5) localities for a total of 53 native iguanas, to represent a wide-spread Iguana dataset (Fig. 1, Table 1). We also included six morphologically identified hybrids between I. delicatissima and non-native iguanas; two samples from St. Barths, and Basse Terre and Grande Terre (Guadeloupe) (Vuillaume et al. 2015).

Fig. 1
figure 1

Map showing sample localities and current taxonomic status with Iguana: Inset shows the Lesser Antilles. Further geographic details for each lineage can be found in Table 1. ‘aff.’ indicates the uncertain taxonomic status, which is undergoing assessment (van den Burg et al. in prep)

Full size image

Laboratory procedures were performed at Labofarm-GenIndexe (France) where PCR settings and primers outlined by Valette et al. (2013) were used to amplify 17 loci (named as IgdL1-25 or L1-25 in previous studies; Valette et al. 2013; Vuillaume et al. 2015). Subsequently, PCR products were ran with Thermo ROX500 and analyzed using an ABI 3130 XL; alleles were next scored using Genemapper, while samples IGU039–43 were used to standardize allele sizes with aforementioned studies (see van den Burg et al. 2021a).

Data analyses

After sample processing, we retained specimens with ≥ 80% scored loci for data analyses (for sample sizes used in analyses see Table 3 and figure captions). Considering our sample-set characteristics, only one population, from Ilet Chancel (29 samples), was suitable to test the presence of null alleles. Following Dabrowski et al (2014), we implemented both Genepop and micro-checker software to assess the presence of null alleles (Raymond and Rousset 1995; van Oosterhout et al. 2004). The remaining sample-set included groups not amenable to this question as they were comprised of; (1) wide geographic sampling within lineages (producing a Wahlund Effect), or (2) low sample size of breeding populations (non-representative of existing variation), or (3) impacted by recent bottlenecks (strong effects of genetic drift).

Table 3 Mean allelic richness from two microsatellite loci sets for each taxonomic unit or population, including number of samples sequenced for > 80% of loci
Full size table

Data were split into two sets, as previous studies have used different loci (Table 2). Our 12-loci dataset includes all native Iguana locations and the six Iguana-delicatissima hybrids, and our 17-loci dataset includes all native locations from the I. iguana complex. For these datasets we then calculated relative allelic richness using the GenPopReport package within the R environment (R Core Team 2019).

Allele variation across all samples was assessed and visualized in color-coded tables to identify genetic variation within and between Iguana lineages (see Results). Within these tables we also include allelic data from the non-native and hybrid iguanas from Vuillaume et al. (2015) and Breuil et al. (2019).

Genetic differentiation and clustering were assessed using STRUCTURE as implemented within GenoDive 3 (Pritchard et al. 2000; Meirmans 2020). For the 12-loci dataset, we applied this methodology to assess its utility to successfully assign native pure and hybrid specimens, as well as to assess the genetic structure among I. delicatissima populations. For the 17-loci dataset, we assessed genetic structuring within the I. iguana complex, for which we standardized sample sizes among lineages by reducing the sample size of I. melanoderma and I. insularis to those of the other lineages by randomly sampling while maintaining sample size-island proportions. STRUCTURE runs used identical parameters including an admixture model, correlated allele frequencies, and without linkage. Runs composed of 10 repeats for K ranges 1–10 using 100 K burnin and 500 K MCMC repeats. Subsequently, we assessed and visualized delta-K estimates for different K values using Structure Harvester (Evanno et al. 2005; Earl and vonHoldt 2012), and employed distruct to create final structure barplots (Rosenberg 2004). Lastly, within R through the loadingplot() command in the adegenet package, we assessed which loci and alleles most contribute to differentiation of I. delicatissima and non-delicatissima iguanas through a discriminant principal component analysis (DAPC; Jombart 2008).

Results

For the 12-loci dataset, mean allelic richness among I. delicatissima populations ranges between 1.03–1.16, and between 1.05–1.43 within the I. iguana complex. For the 17-loci dataset, mean allelic richness shows higher values with a similar pattern among the I. iguana lineages, 1.06–1.98 (Table 3). Within the complex, I. melanoderma and I. insularis show lower mean allelic richness compared to taxonomic units that mostly have a mainland range, though Iguana aff. iguana Clade I is here represented only by island populations.

Null-allele detection tests show inconsistent results with micro-checker suggesting presence for five loci (L3, L5, L12, L18, L24), whereas Genepop only for L20. We retain all loci in further discussion given that no single locus was flagged by both methods. We recommend these analyses be performed on definable populations that have been adequately sampled and meet the a priori assumption of Hardy–Weinberg equilibrium.

Hybridization

Figure 2 provides an overview of results for hybridization purposes between native I. delicatissima and native non-delicatissima, as well as six hybrids. Structure results are visualized for K = 2, which clusters I. delicatissima and non-delicatissima separately, except for the six known hybrids (Fig. 2a). Surprisingly, assignment of Dominica and St. Eustatius samples to I. delicatissima was < 75% (Fig. 2a), likely due to their low sample sizes and alleles with proportionally higher representation in non-delicatissima samples; all samples in these two populations were fixed for unique I. delicatissima alleles. Visual inspection of patterns of allele variation and size range overlap between I. delicatissima and non-delicatissima reveals four loci with a complete lack of allele variation overlap, three of which include data from all I. delicatissima locations: L8, L18, and L25 (Fig. 2b). An additional two loci (L5 and L24) also indicate a high degree of usefulness for hybridization studies, but show a single allele overlap. Lastly, DAPC analysis identified a set of six loci with the highest contribution to differentiating I. delicatissima and non-delicatissima: L5, L8, L14, L18, L20 and L25 (Fig. 2c).

Fig. 2
figure 2

Genetic differentiation results to study hybridization in Iguana delicatissima, using 12 loci dataset. a Structure plot for pure I. delicatissima (67), pure non-delicatissima (117), and hybrid (6) samples, see Table 3; for K = 2. Asterisks indicate individuals from St. Eustatius and Dominica. b Overview of allele variation between I. delicatissima and non-delicatissima individuals, excluding hybrids. C DAPC loading contribution of loci and alleles for clustering between I. delicatissima – non-delicatissima, excluding hybrids

Full size image

Iguana delicatissima

Within I. delicatissima, delta-K identified four clusters with highest support (Fig. 3b): (1) Ilet Chancel and St. Barths, (2) Dominica and St. Eustatius, (3) Petite Terre and La Désirade, and, (4) Guadeloupe. The first two groups contain islands not in close geographic proximity. For delta-K of 2 (but see Janes et al. 2017), assignment followed that of Ilet Chancel and St. Barths vs. all other islands. Allelic variation is highest for larger islands or those with high population sizes; Dominica, Guadeloupe, Petite Terre and Ilet Chancel (Table 3, Fig. 3c).

Fig. 3
figure 3

Genetic variation among Iguana delicatissima island populations. a Structure and b delta K plot for population differentiation among pure I. delicatissima individuals (67; Table 3), using 12 loci dataset; K = 4. c Allele variation among seven island populations for 20 loci; islands assigned to identical Structure cluster (a) have equal-tone colors

Full size image

Iguana iguana complex

Genetic differentiation and allele variation within the I. iguana complex are presented in Fig. 4. Optimal delta-K support indicated four clusters, with sub-structural support for seven clusters (Fig. 4a-c). Allele variation patterns between mainland (generally non-native) and insular lineages highlight loci of interest for identifying the genetic origin of non-native iguanas within those insular lineages; L3, L5, L6, L8, L9, L14 L16, L17, L19, L20, L23–25 (Fig. 4d). Loci that appear to be useful for characterizing insular groups, differs per lineage; I. melanoderma, L6, L14, L20, L25; I. insularis, L6, L13, L14, L17, L20; I. aff. iguana Clade I, L2, L5, L16, L17, L20, L23–25.

Fig. 4
figure 4

Genetic variation within Iguana iguana complex for 17 microsatellite loci. a + b Structure and c delta K plots for population differentiation among seven taxonomic units; optimal K = 4, suboptimal K = 7. d Allele variation among seven taxonomic units. Sample sizes as in Table 3, except for I. insularis insularis (10), and I. melanoderma (10)

Full size image

Discussion

Here we present a nuclear microsatellite database for Iguana that includes all current taxonomic groups with broad geographic representation across its range, demonstrating clear population genetic structure within I. delicatissima and the I. iguana complex. Furthermore, our results identify key sets of loci and private alleles useful in the study of conservation-hybridization issues and for narrowing down the geographic origin of commercially available pet iguanas and illegally transported individuals. This database fills an important gap given that nuclear loci sequence in prior work show low variability within the I. iguana complex (Stephen et al. 2013), prohibiting the fine scale variation at biparentally inherited loci required to address such issues. Overall, our work demonstrates the successful amplification of these markers and their utility within the Iguana genus. Importantly, the benefits of this database will increase dramatically with further input from the Iguana research community, which we promote through free availability on figshare (van den Burg et al. 2021a).

Assignment of loci and private alleles to address research questions is dependent on sampling scheme, both its geographical width and density, as well as sample size per population. Given the geographic range of Iguana, the coverage of the presented database is wide, though mainland sampling density is still low, with few samples per location. Notably, the database greatly increases coverage and lineage inclusion compared to Valette et al. (2013), which assigned 19 loci as useful for I. delicatissima hybridization studies based on one I. delicatissima (Ilet Chancel, Martinique), and one introgressed (I. delicatissima – I. iguana) population (Grande-Terre, Guadeloupe). Here, including data from almost all I. delicatissima populations (only lacking Anguilla), we identify 5–8 loci of specific use for identification of hybrids between I. delicatissima and the I. iguana complex (Fig. 2b, c). Although these loci are fewer in number, their identification is based on a significant data increase from throughout the genus’ range; allowing conservationists to target fewer, highly informative loci and lower research costs. Furthermore, the need to standardize allele sizes across different studies and laboratories is exemplified in our finding that the three samples from St. Eustatius rescored in this study had different allele size for all five loci compared to results in van den Burg et al. (2018b), shifting higher or lower by 5–22 bps, depending on the locus. Similarly, sample reanalyzes in the same laboratory can indicate allele size deviations, as data of several loci presented by Vuillaume et al. (2015) have since been updated (2–16 bps; Breuil et al. 2019). Lastly, with ongoing conservation priorities aimed at insular lineages, the characterization of their allele diversity will carry on, though we address the importance of characterizing new mainland localities. This continuous data collection highlights the need for a “live” dynamic database that provides a crucial, up-to-date, Iguana conservation tool.

The identification and continued usefulness of microsatellites in research on wild populations lies in their rapid mutation rate (Schlötterer 2000), demonstrating their use in studies of recently separated populations (Kuehn et al. 2007), but highlighting their poor informative value for long evolutionary timelines (Carsten et al. 2013). Indeed, this high mutation rate prevents the determination of shared alleles as synapomorphic or homoplasious, when comparing highly divergent lineages. This is evident in our database for shared alleles (e.g., L2) between I. iguana lineages and two I. delicatissima populations (St. Eustatius and Dominica) (Fig. 2a). Genetic samples from these I. delicatissima populations were obtained prior to first documented hybridization on both islands (van den Burg et al. 2018b, 2020a), though historic hybridization might have occurred. Another explanation for shared L2 alleles is independent mutation as might be expected after a long period of evolutionary divergence.

The Critically Endangered, Lesser Antillean iguana has seen a > 75% decline from its pre-Colombian range due to extirpations subsequent to human settlement, and more recently, hybridization with non-native iguanas of the I. iguana complex (van den Burg et al. 2018a). Given the extremely low mtDNA variation throughout the range of I. delicatissima (Martin et al. 2015), microsatellites are better suited to guide research and conservation (Vuillaume et al. 2015; Martin Judson et al. 2018; van den Burg et al. 2018b). Small populations will lose diversity quickly and randomly diverge due to founder effects and/or drift over short time periods (Wright 1931), which has been demonstrated in Iguanids as well (Colosimo et al. 2014; Welch et al. 2017). Thus, to maintain long term genetic variation within I. delicatissima it is essential that all island populations are genotyped. Our database builds on Vuillaume et al. (2015) by inclusion of two additional island populations, St. Eustatius and Dominica. Genetic clustering among currently sampled populations is not correlated with geographic proximity, given two identified clusters: St. Eustatius–Dominica, and St. Barths–Ilet Chancel (Fig. 3a). Their genetic similarity could originate from within-region dispersal facilitated by hurricanes, Amerindians or post-Colombian colonizers (Censky et al. 1998; Bryan et al. 2007; Bochaton et al. 2016). Although this database represents the most inclusive to date, we call for sampling and genotyping of the Anguilla population and those on isolated islets around St. Barths, as well as additional genotyping of samples from Dominica and La Désirade. These genetic data can furthermore help identify the remaining pure animals on Guadeloupe and Martinique, which are in need of urgent translocation to secure their genetic diversity. Lastly, these data and identified key loci will aid other conservation management decisions as identified by conservation action plans (e.g., translocations (Scott et al. 2020), captive breeding, and reintroductions (Knapp et al. 2014; Angin 2017; ANT/ATE/STENAPA 2018; Breuil 2021)).

We provide an initial assessment of allele variation among recently described lineages within the I. iguana complex. Optimal support was found for four genetic clusters, which grouped currently known lineages as (1) I. rhinolopha, (2) I. aff. iguana Clade IIA, (3) I. insularis, (4) I. iguana and I. melanoderma, and with an admixed assignment for I. aff. iguana Clade I (Fig. 4a, c). A suboptimal 7-cluster division distinguished I. iguana and I. melanoderma, as well as uniquely assigned I. aff. iguana Clade I (Fig. 4b, c). Allelic diversity indicated higher variation in mainland lineages and I. aff. iguana Clade I compared to solely insular lineages, with especially low diversity in I. i. sanctaluciae (Table 3). For vulnerable island populations (I. melanoderma, I. insularis, I. aff. iguana Clade I), we find three loci sets (4–8 loci) for origin assessment (see Results text for specifics). For the current dataset, we note the skewed sampling effort towards recently described insular lineages (Breuil et al. 20192020, submitted), and highlight the need for improved sampling of the mainland lineages. Although threats within the I. iguana complex are less studied compared to those in I. delicatissima, it is well-known that mainland populations can be locally threatened due to overharvesting, either for food or to supplement the pet trade (Stephen et al. 2011). More recently, hybridization between non-native iguanas and native island populations has been described, as well as illegal trade of I. insularis and I. melanoderma (Noseworthy 2017; Breuil et al. 2019; van den Burg and Weissgold 2020). Overall, the identified allelic variation (Fig. 4D) will aid in the discovery of non-native and hybrid iguanas and the geographic origin of these as well as of iguanas in the global commercial hobbyist trade.

In conclusion, this manuscript provides a reference for Iguana microsatellite loci in order to allow standardization and comparison among studies, aiding broader assessment of research and conservation hypotheses. Importantly, this work provides an essential tool for conservation of the Critically Endangered I. delicatissima, and taxa within the I. iguana complex. To ensure future comparability, we strongly urge that future studies utilize: (1) the same allele size ladder, (2) positive controls from localities with fixed alleles, and/or (3) samples scored in prior studies.