Introduction

Despite the increasing affordability and high throughput data of Next Generation Sequencing (NGS) technologies, microsatellite markers remain valuable tools in evolutionary biology, ecology, and conservation. These highly polymorphic, co-dominant and noncoding markers offer fine-scale resolution of genetic variation within and among populations [1] and thus, they are still widely used in population genetic research (e.g. [2, 3]).

European vipers (fam. Viperidae, genus Vipera) constitute a clade of venomous snakes distributed across Eurasia and North Africa. With up to 25 species described and multiple studies on phylogenetics, phylogeography and ecology, current knowledge at both interspecific and intraspecific level is still limited by the lack of informative nuclear DNA data [4]. This shortage is particularly evident in the limited number of population genetic studies (but see [5,6,7]), crucial for the conservation of declining viper populations. Nuclear markers were developed for three species (Vipera berus [5, 8], Vipera ursinii [9], V. aspis [10]), but their high specificity limits a broader application across the group. For instance, V. berus markers were applied in a contact zone between the three Iberian vipers, V. aspis, V. latastei and V. seoanei, to detect hybridization [11]. Although they were able to cross-amplify in these species and provided enough resolution to identify hybrids and parentals, they failed to recover intraspecific structure due to low polymorphism (authors, unpublished data).

While 12 microsatellite loci have been already developed for V. a. aspis populations (from Switzerland [10]), the Iberian endemics Vipera latastei and Vipera seoanei lack suitable nuclear markers. Vipera aspis shows high genetic and phenotypic diversity and deep genetic divergence that granted the description of four subspecies [12]. Vipera aspis is listed as Least Concern but at least two of the subspecies face significative population decline and are listed as Critically Endangered (V. a. aspis) and Endangered (V. a. francisciredi) in regional assessments. Developing a new battery of markers for other subspecies can enhance resolution to address several aspects of evolutionary history, demography, reproduction and conservation that are still understudied. Similarly, mitochondrial DNA (mtDNA) studies recovered deep evolutionary lineages and high genetic diversity within V. latastei [13,14,15]. However, nuclear markers are needed to address reproductive isolation and the extent of gene flow among divergent lineages, aiding in clarifying their taxonomic status [15]. Vipera seoanei, on the other hand, exhibits a shallow genetic structure and low genetic diversity at the mtDNA level [16], that contrasts with its high morphological and colouration polymorphism [17, 18]. The use of microsatellite loci has the potential to provide a detailed perspective on the genetic diversity and structure of populations and the factors contributing to their phenotypic variation. Vipera latastei is listed as Vulnerable, due to significant population fragmentation and decline, and V. seoanei is listed as Least Concern, although southern populations are presumably isolated and highly vulnerable to climate change [19]. Yet, genetic studies assessing population diversity and gene flow are lacking for these species.

To address these gaps, we have developed a new set of polymorphic microsatellite markers specific for each of the three Iberian vipers. We tested these markers for polymorphism in north-central Iberian populations and explored cross-amplification across the three target species. With this initiative, we aim to contribute with molecular tools with which we can: (1) characterize patterns of genetic diversity and population structure at an intraspecific level; (2) investigate gene flow and reproductive isolation in contact zones between species and lineages; and (3) perform paternity tests for various ecological and evolutionary questions, ultimately informing effective conservation strategies for each species.

Materials and methods

Genomic libraries were prepared using high molecular weight genomic DNA from 10 tissue samples of each species, collected across their distribution in the Iberian Peninsula (Online Resource 1). Tissue samples consisted of a portion of tail (< 1 mm) collected from alive (non-euthanized) and road-killed specimens encountered during fieldwork. Capture and handling of vipers for data collection was carried out following all ethical procedures (see Ethics approval).

DNA extractions were performed using EasySpin Kit, following the manufacturer’s instructions. DNA integrity and purity were assessed by agarose gel electrophoresis (2%) and spectrophotometry. Enriched microsatellite libraries were then generated from genomic DNA at a 50 ng µL − 1 concentration using a modified protocol from [20]. Sequencing was carried out using MiSeq V2 kit (500 cycles) and single Illumina indexes.

Reads from sequencing were analysed using Micro-Primers [21] to find microsatellites using optimized parameters to detect those more polymorphic. The analysis included quality trimming using Trimmomatic, merge of pairs using Flash and microsatellites finding through MISA. The obtained microsatellites were then screened and selected to design multiplexes for each species using Multiplex Manager 1.2. The following parameters were applied for markers selection: (1) to maximize the space between markers in the same dye, (2) minimize the difference in annealing temperatures of markers in the same reaction and (3) minimize complementarity of primers in the same reaction.

Using these criteria, we selected 70 potential markers (ca. 20 for each species) that were multiplexed in two reactions for each species. See Table 1 for primer details. For each marker, a third primer was used, following the M13 tailed primer method [22]. This primer was labelled with FAM, NED, VIC or PET, depending on the selected dye for each locus (Table 1). Forward primers had a 5’ tail complementary to the fluorescently labelled primers.

Table 1 Characteristics of 48 polymorphic microsatellite loci developed for V. aspis, V. latastei and V. seoanei. TD: touchdown temperature of annealing
Full size table

Multiplexes were tested for amplification and polymorphism using 26, 20 and 10 samples of V. aspis, V. latastei and V. seoanei respectively (Online Resource 1). PCRs were performed in a total volume of 10 µl with 1–2 µl template DNA, 100–150 nM of each primer, and 5 µl Multiplex PCR Master Mix (Qiagen). Touchdown PCR conditions started with an initial denaturation step of 15 min at 95 °C; first round (nine cycles) of 30 s at 95 °C, 2 min for annealing (-0.5 °C/cycle), and 1 min at 72 °C; second round (36 cycles) of 30 s at 95 °C, 1 min at annealing temperature, 30 s at 72 °C, and a final extension of 30 min at 60 °C. Multiplexes details and touchdown temperatures can be consulted in Table 1. Amplifications were performed in Biorad T100 Thermal Cyclers, and the PCR products were separated by capillary electrophoresis on an automatic sequencer ABI3500xl Genetic Analyzer (AB Applied Biosystems). Fragments were scored and manually checked by two independent persons using GENEMAPPER 5.0 (Applied Biosystems).

Individuals from the same species were pooled and three populations were defined for loci characterization. The existence of null alleles, allelic dropouts and stuttering were verified using MICROCHECKER v2.2.3. Tests of Hardy–Weinberg equilibrium (HWE) deviations for each locus and linkage disequilibrium (LD) for all pairs of loci were computed with ARLEQUIN 3.5. Basic population summary statistics such as the number of alleles per locus (NA), observed heterozygosity (HO) and expected heterozygosity (HE) were estimated with ARLEQUIN 3.5.

Moreover, the transferability of these markers was tested by cross amplification on the three vipers, using a subset of eight samples per species.

Results and discussion

From the 70 primer pairs initially tested, 25 were discarded due to no amplification (or low amplification success), no polymorphism or unreliable allelic patterns that could cause potential scoring errors. In total we obtained 16, 11, and 21 loci that amplified reliably and were polymorphic within species, for V. aspis, V. latastei and V. seoanei, respectively (Tables 1 and 2). MICROCHECKER showed no evidence of allelic dropout or stuttering, but null alleles (unamplified alleles) were diagnosed across several loci that showed significant excess of homozygous (Table 2). Significant deviation of HWE towards heterozygosity deficiency was detected for several loci in the three species (Table 2). This can be attributed to presence of null alleles and also to the effect of population substructure, since the samples analysed were obtained from different localities, and thus they may not represent a panmictic population. Significant linkage disequilibrium (LD) was also observed across multiple loci, particularly in V. seoanei (Table 2). LD, which is the non-random association of alleles at different loci, can be induced by several factors, including selection and genetic drift [23]. The range expansion experienced by V. seoanei from northwestern Iberia likely led to profound bottleneck events and resulted in decreased genetic diversity eastward [16]. Allele surfing processes during range expansion events shape patterns of genetic structure through genetic drift (e.g. [24]), and thus can be associated to increased genome-wide LD, due to the random sampling of loci during bottleneck events. Therefore, the demographic history of this species offers an explanation for the high LD proportion observed.

Table 2 Genetic parameters of the 48 evaluated loci and transferability across the three Vipera species. NA means no amplification. Most polymorphic loci (higher NA, HE, HO) are highlighted in bold. Shaded areas correspond to loci for which cross-amplification tests were not performed for a given species
Full size table

For V. aspis, the number of alleles (NA) ranged between 5 (Vasp8) to 23 (Vasp2), with a mean number of alleles per locus of 10. The majority of loci had moderate to high observed (minimum of 0.21 in Vasp1 to 0.88 in Vasp2) and expected heterozygosity (HE = 0.40 in Vasp4, 0.95 in Vasp2). For V. latastei, NA ranged between two (VLa1 and VLa11) to 17 (VLa12 and VLa16), with a mean number of alleles per locus of nine. Observed and expected heterozygosity varied between locus from very low to high (HO = 0.05 in VLa11 and 1.0 in VLa20; HE = 0.05 in VLa11 and 0.92 in VLa12). For V. seoanei, number of alleles per locus ranged from 2 (Vse22) to 9 (Vse13) with a mean of 5. Observed and expected heterozygosity also ranged from very low to high (HO = 0.05 in Vse8, and 0.86 in Vse26; HE = 0.28 in Vse15, and 0.93 in Vse13). Out of the 48 loci analysed, 36 are highly polymorphic (high NA, HE, HO) and have the potential to be informative in future studies that encompasses other populations and subspecies (Table 2).

Cross-species transferability was achieved for 31 microsatellites loci that consistently amplified and were polymorphic in the three target species and four additional loci that were transferable to only one of the species (Table 2). Cross-amplification failed for six V. aspis (Vasp 2, 5, 9, 11, 15 and 22), five V. seoanei (Vse 5, 8, 20, 22, and 23) and two V. latastei markers (VLa 11 and 12). Two V. seoanei markers (Vse 16 and 18) failed to amplify in V. latastei only. Additionally, VLa17 and Vse14 were monomorphic for at least one of the species. The cross-species amplification efficiency is usually inversely proportional to the phylogenetic distance between the species [25]. However, we did not observe a lower amplification success in V. seoanei compared to the closely-related V. aspis and V. latastei. In fact, V. seoanei markers showed high applicability potential in the other two species, and vice-versa, supporting a broader use of the set of loci generated in this study in other Vipera species.

Conclusion

Our study provides a selection of highly polymorphic microsatellite markers that can deepen current knowledge on genetic patterns, demographic dynamics, systematics, reproduction and heritability in these species, and potentially in other Vipera taxa.