A combination of long term fragmentation and glacial persistence drove the evolutionary history of the Italian wall lizard Podarcis siculus

Abstract

Background: The current distribution of genetic diversity is the result of a vast array of microevolutionary processes, including short-term demographic and ecological mechanisms and long-term allopatric isolation in response to Quaternary climatic fluctuations. We investigated past processes that drove the population differentiation and spatial genetic distribution of the Italian wall lizard Podarcis siculus by means of sequences of mitochondrial cytb (n = 277 from 115 localities) and nuclear mc1r and β-fibint7genes (n = 262 and n = 91, respectively) from all its distribution range. The pattern emerging from the genetic data was compared with current and past (last glacial maximum) species distribution modeling (SDM).

Results: We identified seven deeply divergent parapatric clades which presumably remained isolated in different refugia scattered mainly throughout the Tyrrhenian coast. Conversely, the Adriatic coast showed only two haplogroups with low genetic variability. These results appear to agree with the SDM prediction at the last glacial maximum (LGM) indicating a narrow area of habitat suitability along the Tyrrhenian coast and much lower suitability along the Adriatic one. However, the considerable land exposure of the Adriatic coastline favored a glacial colonization of the Balkan Peninsula.

Conclusions: Our population-level historical demography showed a common trend consistent with glacial expansions and regional persistence during the last glacial maximum. This complex genetic signature appears to be inconsistent with the expectation of the expansion-contraction model and post-LGM (re)colonizations from southern refugia. Hence it is one of an increasing number of cases in which these assumptions are not met, indicating that long-term fragmentation and pre-LGM events such as glacial persistence were more prominent in shaping genetic variation in this temperate species.

Keywords: Evolutionary history; Glacial expansion; Italian wall lizard; Niche modelling; Phylogeography; Podarcis siculus.

Fig. 1

Map of the study area showing the principal geographic features mentioned in the main text. The area in red corresponds to the geographic distribution of Podarcis siculus while the 115 new sampled locations are shown in black dots

Fig. 2

a Phylogenetic haplotype recostruction of the cytb data set obtained with MrBayes. At each node, MrBayes and BEAST posterior probability and TMRCA with 95% highest posterior density (HPD) intervals are respectively reported. The seven clade such as its nested haplogroups are depicted with different colors. b Geographic distribution of the mtDNA haplotypes in Podarcis siculus. Pie diagrams shows the haplotype frequencies at each sampled location and are colored according to the main haplogroups identified by the phylogeny. c Statistical parsimony networks of the seven mtDNA clades identified by the phylogeny. Each circle size is proportional to their frequencies and each filled rectangles representing one substitution. The different colors within each network depict the principal identified haplogroups. Circum-Sicilian and Corsica-Sardinia haplotypes are reported in bold with different colors (violet: Aeolian haplotypes; blue: Aegadian haplotypes, green: Sardinia and Corse haplotypes, yellow; Isola Bella haplotypes)

Fig. 3

Statistical parsimony network connecting the 61 mc1r haplotypes (a) and the 68 β-fibint7 haplotypes (b). Each haplotype is represented by circles with size proportional to their frequencies. The colors depicts the identified mitochondrial clades

Fig. 4

Extended Bayesian skyline plots inferred with BEAST, illustrating effective population sizes (Ne) through time. The grey lines represents the median population size, and the colored lines represent 95% higher posterior probability of (a) T clade, (b) A2 clade, (c) S3 clade and (d) S1 clade. The grey rectangles corresponds to the Würm, Riss and Mindel glacial periods while dashed lines indicate approximately the last glacial maximum

Fig. 5

Estimates of spatial diffusion of Podarcis siculus using mtDNA data set at three time points during the Pleistocene (a) 2.3 Mya, (b) 1.3 Mya, (c) 0.1 Mya. The polygons are colored according to each clade and indicate the uncertainty (80% HPDs) surrounding geographic locations of internal nodes of the MCC tree

Fig. 6

Species distribution model for Podarcis siculus indicating the present-day conditions and the last glacial maximum (LGM) based on the Model for Interdisciplinary Research on Climate (MIROC) and Community Climate System Model (CCSM)