yes Northern richness and cryptic refugia: phylogeography of the Italian smooth newt Lissotriton vulgaris meridionalis [1/1, 1 ed.]

Citation preview

bs_bs_banner

Biological Journal of the Linnean Society, 2014, ••, ••–••. With 4 figures

Northern richness and cryptic refugia: phylogeography of the Italian smooth newt Lissotriton vulgaris meridionalis MICHELA MAURA1, DANIELE SALVI2*, MARCO A. BOLOGNA1, GIUSEPPE NASCETTI3 and DANIELE CANESTRELLI3 1

Dipartimento di Scienze, Università degli studi Roma Tre, Viale G. Marconi 446, 00146 Rome, Italy CIBIO, Centro de Investigação em Biodiversidade e Recursos Genéticos, InBIO, Universidade do Porto, Campus Agrário de Vairão, 4485-661 Vairão, Portugal 3 Dipartimento di Scienze Ecologiche e Biologiche, Università della Tuscia, Viale dell’Università s.n.c., I-01100 Viterbo, Italy 2

Received 13 March 2014; revised 13 May 2014; accepted for publication 14 May 2014

Recent phylogeographical studies have re-evaluated the role of refugia in central and northern Europe for glacial persistence and postglacial assembly of temperate biota. Yet, on a regional scale within Mediterranean peninsulas, putative ‘northern’ refugia’s contribution to the current structure of biodiversity still needs to be fully appreciated. To this end, we investigated the phylogeographical structure and the evolutionary history of the Italian smooth newt, Lissotriton vulgaris meridionalis, through phylogeographical, molecular dating and historical demographic analyses. We found ten differentiated mitochondrial lineages with a clear geographical association, mainly distributed in northern Italy. The most ancient divergence among these lineages was estimated at the Early Pleistocene and was followed by a series of splits throughout the Middle Pleistocene. No haplogroup turned out to be derived from another one, each one occupying terminal positions within the phylogenetic network topologies. These results suggest an unprecedented scenario involving long-term survival of distinct evolutionary lineages in multiple northern Mediterranean refugia. This scenario mirrors on a smaller geographical scale what has been previously observed in the literature concerning northern European environments; it also sheds more light on how northern Italy has contributed to temperate species’ long-term survival and to the assembly of regional biota. © 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••.

ADDITIONAL KEYWORDS: genetic diversity – Italian Peninsula – mtDNA – multiple refugia – temperate species.

INTRODUCTION In recent years Western Palaearctic species’ diversity of responses to past climatic oscillations has been investigated in depth (Stewart, 2009; Hewitt, 2011a, b). As a consequence, within southern Mediterranean peninsulas the paradigmatic scenario of range contraction during glacial epochs and postglacial (re)colonization of central and northern Europe (Hewitt, 1999; Habel, Schmitt & Müller, 2005) has been enriched by a plethora of different scenarios. *Corresponding author. E-mail: [email protected]

These encompass persistence in central and northern Europe during glacial epochs (Stewart & Lister, 2001; Stewart et al., 2010; Schmitt & Varga, 2012; Salvi et al., 2013), survival in multiple glacial/interglacial refugia within the Mediterranean peninsulas, with or without subsequent secondary admixture among diverging lineages (Schmitt et al., 2006; Gómez & Lunt, 2007; Krystufek et al., 2007; Canestrelli, Cimmaruta & Nascetti, 2008; Canestrelli et al., 2010), and the use of microrefugia, nunataks and coastal lowlands as glacial refugia (Rull, 2009; Bisconti et al., 2011; Porretta et al., 2011; Schneeweiss & Schönswetter, 2011; Salvi et al., 2014).

© 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

1

2

M. MAURA ET AL.

Even at the smaller geographical scale of the Italian Peninsula, a similarly wide diversity of responses to Plio-Pleistocene climatic oscillations is emerging among temperate species (e.g. Canestrelli et al., 2010; Vega et al., 2010; Hewitt, 2011b; Canestrelli, Sacco & Nascetti, 2012a; Canestrelli et al., 2012b; and references therein). The southern part of the Italian Peninsula has been identified as a major hotspot of genetic diversity, as a glacial/interglacial refugium and as a source for later (re)colonizations of northern areas, for most species studied to date (e.g. Canestrelli et al., 2006, 2008, 2010; Magri, 2008). Moreover, phylogeographical patterns indicating multiple refugia in distinct mountain districts, coastal refugia and/or microrefugia have been found in most of these species (Canestrelli, Cimmaruta & Nascetti, 2007; Magri, 2008; Canestrelli et al., 2010, 2012a, b; Vega et al., 2010). On the other hand, an even more complicated picture is emerging as regards the northern part of Italy. Indeed, while this area has long been acknowledged as a site of postglacial (re)colonization, recent works have also identified it as a putative refugium (Canestrelli et al., 2007, 2012b; Crottini et al., 2007; Canestrelli & Nascetti, 2008; Salvi et al., 2013), revealing that many of the evolutionary and historical demographic processes leading to the assembly of the present-day regional biota have still to be fully appreciated. These findings mirror what has previously been observed in the literature concerning the northern European environments (Stewart & Lister, 2001; Stewart et al., 2010; Schmitt & Varga, 2012); similarly, they suggest the need for more studies to achieve a deeper understanding of how northern Italy has contributed to temperate species’ long-term survival and to the current structure of the Italian and European biota. In this study we analysed the phylogeographical structure of the Italian smooth newt, Lissotriton vulgaris meridionalis (Boulenger, 1882), within its putative Pleistocene refugium in Italy (Babik, Branicki & Crnobrnja–Isaloviç, 2005). Lissotriton vulgaris is a temperate species widespread in Europe and south-western Asia. A previous phylogeographical study of L. vulgaris carried out at the level of the whole species range showed a high genetic fragmentation of populations into several divergent lineages that survived Pleistocene climatic oscillations in multiple refugia located in southern Europe, central Europe and south-western Asia (Babik, Branicki & Crnobrnja–Isaloviç, 2005). Among these lineages the Italian populations were identified as belonging to an independent and anciently differentiated lineage (∼1.9 Mya; Babik et al., 2005). Here we used phylogeographical, molecular dating and historical demographic analyses, together with

a denser sampling scheme in northern Italy, with the aim of investigating the evolutionary history of L. v. meridionalis, and to shed more light on how this area has contributed to temperate species’ long-term survival and to the assembly of regional biota.

MATERIAL AND METHODS SAMPLING

AND LABORATORY PROCEDURES

We sampled 82 L. v. meridionalis individuals from 24 localities throughout its range (see Table 1 and Fig. 1B). Newts were anaesthetized by submersion in a 0.1% solution of MS222 (3-aminobenzoic acid ethyl ester) and tissue samples were collected from tail tips and stored in 96% ethanol until subsequent analyses. Afterwards, all individuals were released at the respective collection site. DNA extraction was performed by following the standard cetyltrimethyl ammonium bromide (CTAB) protocol (Doyle & Doyle, 1987). Two mtDNA fragments were amplified and sequenced for all individuals, one comprising the partial NADH dehydrogenase subunit 4 gene and the flanking tRNAHis gene (hereafter ND4), and the other comprising the NADH dehydrogenase subunit 2 gene (hereafter ND2). Preliminary amplifications and sequencing of the ND4 fragment were performed using the primers ND4 and LEU (Arévalo, Davis & Sites, 1994), and then the internal primers ND4vulgF1 (ATCCGAATTTCTATAATCMTTACCC) and ND4vulgR1 (CTTCTTGGTAGGTAGAGAGGGT TTA) were designed and used for PCR amplification and sequencing of all individuals. Preliminary amplifications and sequencing of the ND2 fragment were performed using the primers L3780 and H5018 (Babik et al., 2005); the internal forward primer LVND2F1 (AATCAGCAACAAAATACTTTTTAACG) was then designed and used with the H5018 for amplification and sequencing of all individuals. Amplifications were performed in a 25-μL volume containing MgCl2 (2.5 mM), reaction buffer (5×, Promega), the four dNTPs (0.2 mM each), the two primers (0.2 μM each), the enzyme Taq polymerase (1 U, Promega) and 2 μL of genomic DNA. PCR reactions were carried out with an initial step at 95 °C for 5 min followed by 30 (ND4) or 38 (ND2) cycles of: 94 °C for 1 min, 55 °C (ND4) or 56 °C (ND2) for 1 min, 72 °C for 1 min, and a single final step at 72 °C for 10 min. Purification and sequencing of the PCR products were carried out by Macrogen Inc. (http://www.macrogen.com) using an ABI PRISM 3700 sequencing system. All sequences were deposited in GenBank (accession numbers: KM262086–KM262179).

© 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

PHYLOGEOGRAPHY OF L. VULGARIS MERIDIONALIS

3

Table 1. Geographical location, number of individuals (N) and haplotype distribution of the 24 populations sampled of Lissotriton vulgaris meridionalis Sample

Locality

N

Latitude

Longitude

Haplotypes (N)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Arcade Bo’ de Pavei Collesel di San Rocco Crocetta del Montello Campiglia dei Berici Porto Caleri Castelleone Cesano Maderno Donega San Geminiano Mulino di Pianoro Vernio* Terra del sole Camaldoli Senigallia Pisa* Greve in Chianti Roccalbegna Orbetello Alviano Palo Laziale Navegna Rocca di mezzo Doganella Circeo Sepino

1 1 2 1 1 3 4 3 8 3 7 1 4 1 2 3 4 5 4 3 6 4 3 6 1 5

45°46′ 45°49′ 45°50′ 45°49′ 45°19′ 45°5′ 45°16′ 45°36′ 44°56′ 44°43′ 44°21′ 44°02′ 44°11′ 43°47′ 43°41′ 43°43′ 43°35′ 42°44′ 42°26′ 42°37′ 41°56′ 42°9′ 42°9′ 41°45′ 41°20′ 41°23′

12°12′ 12°10′ 12°9′ 12°2′ 11°32′ 12°19′ 9°45′ 9°7′ 9°14′ 10°26′ 11°19′ 11°09′ 11°57′ 11°49′ 13°12′ 10°24′ 11°18′ 11°30′ 11°13′ 12°14′ 12°5′ 13°2′ 13°38′ 12°47′ 13°2′ 14°34′

M6b (1) M5c (1) M5a (1), M5b (1) M6d (1) M4c (1) M4a (2), M4d (1) M6c (1), M6e (1), M6a (1), M9a (1) M6a (2), M2a (1) M9a (2), M9b (2), M9c (2), M9d (1), M9e (1) M10a (1), M8f (1), M9a (1) M8a (3), M8b (3), M8c (1) M8a (1) M8e (3), M4b (1) M8g (1) M8d (1), M3h (1) M1a (1), M1f (1), M3d (1) M1a (3), M1b (1) M3a (3), M3c (1), M3g (1) M3i (1), M3l (3) M3a (1), M3b (1), M8d (1) M3a (2), M3e (3), M3f (1) M1c (2), M1d (1), M1g (1) M1e (2), M7c (1) M7b (1), M7d (3), M7e (1), M3a (1) M7f (1) M7a (5)

*Samples from GenBank (see text for accession numbers).

A

M8e M8f M9d

M8d

M10a M8b

M1e

M5b M4c M5a

M5c

M9c M9b

M8a

M4a

M8c M8g

M6b

M4b

M9a

M9e M1d

B

M1c

M6c M6d

M6a

M6e M3e

M3i

M3l

malv10 M3a

M3d

M1a

M3b

M3f M7c M7b

M1b M1f

M7a

M1g

M3c M3h M3g M7f

M7d M7e

M2a

Figure 1. A, median-joining network showing the phylogenetic relationships among the 49 haplotypes (numbered as in Table 1) found within Lissotriton vulgaris meridionalis. Circle size is proportional to haplotype frequency; open dots represent missing intermediate haplotypes. B, geographical distribution of the ten haplogroups found across the 26 studied populations shown as pie diagrams. White dotted lines show boundaries of the subspecies’ range. Inset: geographical location of the study area within the Western Palaearctic region. © 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

4

M. MAURA ET AL.

NUCLEOTIDE

VARIATION, PHYLOGENETIC ANALYSES AND MOLECULAR DATING

Electropherograms were visually checked using FinchTv 1.4.0 (Geospiza Inc.) and aligned using Clustal X 2.0 (Larkin et al., 2007). Previously published sequences of ND4 and ND2 fragments of four additional individuals sampled in central Italy (locality numbers 12 and 16 in Table 1) were added to our analyses (Babik et al., 2005, GenBank accession numbers: AY951639, AY951495, AY951454–56 and AY951609–11). We discarded the hypothesis of pseudogenes occurring in our mitochondrial sequence dataset by confirming the absence of stop codons in protein-coding fragments, and the overall similarity with the reference mitochondrial genome of L. vulgaris (GenBank accession number: EU880339). Nucleotide variation and corrected sequence divergence (Tamura & Nei, 1993) between and within the main haplogroups were estimated using the software MEGA 5 (Tamura et al., 2011). The appropriate model of nucleotide substitution for our dataset was chosen among 88 distinct models using the Akaike Information Criterion (AIC; Akaike, 1973) implemented in jModelTest 0.1.1 (Posada, 2008). TIM2+Γ (Posada, 2003) was the best fit model for ND4 and ND2 fragments analysed both separately and combined (with the gamma distribution shape parameter = 0.141). The genealogical relationships among haplotypes were inferred by phylogenetic networks as well as by maximum-likelihood (ML) and Bayesian (BA) phylogenetic trees. Phylogenetic networks were inferred using two distinct algorithms: the statistical parsimony procedure implemented in TCS 1.21 (Clement, Posada & Crandall, 2000) and the median-joining procedure (Bandelt, Forster & Röhl, 1999) implemented in NETWORK 4.6.1 (Fluxusengineering). However, both methods recovered topologies with several missing intermediate haplotypes connecting terminal haplotypes. In such cases, the median-joining method shows relatively higher performance among network-building procedures (Cassens, Mardulyn & Milinkovitch, 2005). Therefore, only the median-joining network is shown. ML analysis was performed with PhyML 3.0 (Guindon et al., 2010) using the SPR&NNI algorithm and the model of sequence evolution suggested by jModelTest. The robustness of the inferred ML tree was assessed by the non-parametric bootstrap method with 1000 replicates (BP). BA analyses were performed in MrBayes 3.2.0 (Ronquist et al., 2012). Two independent runs were conducted using the model of evolution selected by JModeltest, with random starting trees, run length of 107 generations, sampling every 100 generations. Branching reliability was estimated as values of

Bayesian posterior probabilities (BPP) of sampled trees (burn-in = 25%). Time estimates for the most recent common ancestors (TMRCAs) of the mtDNA haplogroups were obtained by using the distance-based least squares (LS) method (Xia & Yang, 2011) implemented in the software DAMBE 5.3.8 (Xia & Xie, 2001). A likelihood ratio test performed with this software did not reject the molecular clock hypothesis for our dataset. The LS analysis was performed specifying the ML tree topology calculated in PhyML and using the related subspecies L. v. vulgaris as an outgroup (GenBank accession numbers: AY951562 and AY951396). We set the root of the tree as far back as 1.9 Mya, corresponding to the divergence time between L. v. vulgaris and L. v. meridionalis as estimated by Babik et al. (2005). We used the ‘softbound’ option and ‘MLCompositeTN93’ genetic distance, as suggested by Xia & Yang (2011). Standard deviations of the time estimates were obtained by means of 1000 bootstrap re-samplings.

POPULATION

GENETIC STRUCTURE AND

HISTORICAL DEMOGRAPHY

We investigated the geographical structure of genetic variation following two distinct approaches. In both cases populations with one individual were excluded from the dataset. First, we searched for groups of populations that are geographically homogeneous but maximally differentiated from each other by using the spatial analysis of molecular variance implemented in SAMOVA 1.0 (Dupanloup, Schneider & Excoffier, 2002). We performed this analysis setting the number of groups (K) from two to 12, and selecting the best clustering option as the one returning the highest and significant value of FCT (i.e. the among-group variance component). To verify the consistency of the results among runs, we replicated the analysis five times for each K value with 1000 independent annealing processes. Second, we investigated to what extent the geographical structure of genetic variation was accounted for patterns of isolation-by-distance, relative to genetic differentiation among groups; this was done by carrying out partial Mantel tests as implemented in the IBD web service 3.23 (Jensen, Bohonak & Kelley, 2005). To perform these analyses we computed three distance matrices among population pairs: (1) the corrected mean genetic distance (TrN+Γ model, Γ = 0.141) calculated with MEGA; (2) the geographical distance, calculated by GEOGRAPHIC DISTANCE MATRIX GENERATOR 1.2.3 (Ersts, 2012) and subsequently log-transformed following suggestions by Rousset (1997); and (3) a binary matrix (the indicator matrix) in which the value ‘0’ indicates that

© 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

PHYLOGEOGRAPHY OF L. VULGARIS MERIDIONALIS two populations are within the same group, and ‘1’ in the opposite case. We assessed the significance of correlations and partial correlations between these matrices, by using Mantel and partial Mantel tests of 1000 bootstrap replicates. Past demographic changes of each haplogroup found (with N > 9) were investigated by using two different methods. First, a mismatch distribution analysis (Rogers & Harpending, 1992) was performed with the software ARLEQUIN 3.5.1.2 (Excoffier, Laval & Schneider, 2005). In this analysis the observed distribution of nucleotide differences between haplotype pairs (mismatch distribution) is compared with that expected under both a demographic expansion model (Rogers & Harpending, 1992) and a sudden spatial expansion model (Excoffier, 2004). We used the sum of squared deviations between estimated and observed mismatch distributions as goodness-of-fit statistics; its significance was assessed using 1000 bootstrap replicates. Second, the FS (Fu, 1997) and R2 (Ramos-Onsins & Rozas, 2002) statistics were also used to infer past demographic changes. Both these neutrality tests have been shown to outperform most of the other statistics commonly used with the same aim (see Ramos-Onsins & Rozas, 2002), and, in particular, FS was shown to perform better at large sample sizes while R2 does at small sample sizes (Ramos-Onsins & Rozas, 2002). High, negative and significant values of FS and small, positive and significant values of R2 are indicative of past demographic expansions (Fu, 1997; Ramos-Onsins & Rozas, 2002). The significance of the FS and R2 values was assessed through 1000 coalescent simulations, carried out under the hypothesis of population equilibrium and selective neutrality. Moreover, for the estimated FS values the 2% cut-off criterion was used to assess the 5% nominal level of significance (Fu, 1997). FS and R2 statistics and coalescent simulations were performed with the software DnaSP 5 (Librado & Rozas, 2009).

NUCLEOTIDE

RESULTS VARIATION, PHYLOGENETIC

ANALYSES

5

Phylogenetic analyses based on ML and BA trees and network reconstructions were congruent and revealed ten reciprocally monophyletic and statistically supported haplogroups (BP > 75, BPP > 0.98; Figs 1A, 2 and Supplementary Fig. S1) whose geographical distribution is shown in Figure 1B. Seven haplogroups were distributed in northern Italy: M9 (samples 7, 9 and 10), M10 (sample 10) and M2 (sample 8) mainly in the central–western part of the Po plain; M4 (samples 5, 6 and 13) and M5 (samples 2 and 3) in the Venetian plain; M6 (samples 1, 4, 7 and 8) along the pre-Alpine area; and M8 on both sides of the northern Apennines (samples 10–15, 20). Haplogroups M1, M3 and M7 occurred in the remaining of the species’ range along the central Italian peninsula: M3 was mostly distributed along the Tyrrhenian coast (samples 15, 16, 18–21 and 24) reaching the Adriatic side in sample 15; M7 in the southern part of the subspecies range (samples 23–26); and M1 close to both the northern (samples 16 and 17) and central Apennines (samples 22, 23), showing indeed an apparently fragmented distribution. Co-presence was observed in nine localities (7, 8, 10, 13, 15, 16, 20, 23 and 24) and occurred among most of the haplogroups. Average sequence divergence among haplogroups ranged from 0.007 (between M9 and M10; SD 0.002) to 0.022 (between M2 and M5; SD 0.005) whereas within groups average genetic distance ranged from 0.001 (M8 and M9; SD 0.001) to 0.004 (M5; SD 0.001) (see Table 2). TMRCA estimates for the mtDNA haplogroups are shown as a chronogram in Figure 2 and in Supplementary Table S1. The TMRCA for the entire ingroup was estimated to have occurred in the late Early Pleistocene (1.312 ± 0.290 Mya); the subsequent splits among the haplogroups fell within the late Early Pleistocene and the Middle Pleistocene. Finally, the TMRCAs of each haplogroup ranged from the early Middle Pleistocene (i.e. 0.444 ± 0.131 Mya of M1) to the early Late Pleistocene (i.e. 0.166 ± 0.070 Mya of M8).

POPULATION

GENETIC STRUCTURE AND

AND MOLECULAR DATING

HISTORICAL DEMOGRAPHY

For all individuals analysed the ND4 fragment was 644 bp in length, comprising 587 bp of the (3′) NADH dehydrogenase subunit 4 gene and 57 bp of the tRNAHis gene; the ND2 fragment was 651 bp. The combined dataset (overall 1295 bp) included 109 variable positions, of which 61 were parsimonyinformative. We did not find indels or stop codons within the coding region of either the ND2 or the ND4 fragments. A total of 49 haplotypes were found in the combined fragment. Geographical distribution of haplotypes is presented in Figure 1B and Table 1.

The spatial analysis of molecular variance showed that FCT values progressively increased (from 0.301 to 0.596) and FSC values progressively decreased (from 0.619 to 0.153) when K increases from K = 2 to 12 (see Fig. 3A). After K = 9 population structures were no longer informative as one population at a time is removed from groups (see Fig. 3B). At K = 9 the groups recovered almost mirror the geographical distribution of the haplogroups identified in the phylogenetic analyses (M1, M3–M9) plus an additional group composed of samples 16 and 17.

© 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

6

M. MAURA ET AL.

M2 1

M8

95

0.99

M10

0.92 0.98

M9

76

0.90 70

1 95

M7

1 84

M4

1 96

M5 1 96

1 86

M6

1 75

M3

1 98

M1 L. v. vulgaris

1.9

1.3

1

0.9

Early Pleistocene

0.15

0.5

Middle Pleistocene

0 Late Pleistocene

Figure 2. Chronogram showing the estimates of time to the most recent common ancestors (TMRCAs) for the ten mtDNA lineages of Lissotriton vulgaris meridionalis. The calibration point (1.9) and the ranges of the main historical epochs on the scale bar are reported in million years. Nodal support values from Bayesian (BA) and maximum-likelihood (ML) phylogenetic analyses are reported in correspondence of the main nodes of the chronogram (above: BA posterior probabilities > 0.90; below: ML bootstrap values > 70).

The Mantel test performed between genetic and log geographical distances suggested the occurrence of a weak but significant pattern of isolation-by-distance (R2 = 0.11, P < 0.001), which is no longer statistically supported when performing the partial Mantel test between these two variables controlling for groups. On the other hand, the Mantel test between genetic distances and groups (R2 = 0.40) as well as the partial Mantel test between these two variables controlling for log geographical distances (R2 = 0.33) showed significant correlations (P < 0.001). Results from mismatch distribution analyses and values of the demographic summary statistics for the five haplogroups tested are reported in Figure 4. The observed mismatch distributions for all the haplogroups were not significantly different from those

expected from a pure demographic and spatial expansion (Fig. 4). For M7, M9 and M1 the spatial expansion model fitted better than the pure demographic expansion model, whilst for M3 the reverse was observed. FS and R2 values were not statistically significant in all the haplogroups, the only exception being M3 where a small and statistically significant R2 value was estimated (R2 = 0.0795; P < 0.01).

DISCUSSION The phylogeographical pattern shown by L. v. meridionalis is unprecedented among co-distributed taxa, especially as regards northern Italy. Indeed, this region has been identified as an area of postglacial (re)colonization for some species (Canestrelli et al.,

© 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

– 0.001 (0.001) 0.007 (0.002) M1 M2 M3 M4 M5 M6 M7 M8 M9 M10

0.004 0.018 0.014 0.015 0.016 0.016 0.014 0.013 0.012 0.012

(0.001) (0.005) (0.003) (0.004) (0.004) (0.004) (0.004) (0.004) (0.003) (0.003)

– 0.017 0.017 0.022 0.02 0.017 0.017 0.018 0.02

(0.004) (0.004) (0.005) (0.005) (0.004) (0.005) (0.005) (0.006)

0.002 0.007 0.009 0.008 0.01 0.011 0.012 0.014

(0.001) (0.002) (0.002) (0.002) (0.003) (0.003) (0.003) (0.003)

0.003 0.009 0.008 0.011 0.012 0.013 0.014

(0.001) (0.003) (0.002) (0.003) (0.003) (0.004) (0.004)

0.004 0.01 0.014 0.014 0.015 0.015

(0.001) (0.003) (0.004) (0.004) (0.004) (0.004)

0.002 0.012 0.013 0.014 0.016

(0.001) (0.003) (0.003) (0.004) (0.004)

0.002 0.011 0.012 0.014

(0.001) (0.003) (0.004) (0.004)

0.001 (0.001) 0.008 (0.002) 0.009 (0.003)

M10 M9 M8 M7 M6 M5 M4 M3 M2 M1

Table 2. Average genetic divergence among (below the diagonal) and within (along the diagonal) haplogroups based on Tamura & Nei (1993) distances; standard deviations (1000 bootstrap replicates) are reported in parentheses

PHYLOGEOGRAPHY OF L. VULGARIS MERIDIONALIS

7

2008; Magri, 2008) and as a glacial refugium for other species showing one endemic lineage in this area (Canestrelli et al., 2007, 2012b; Crottini et al., 2007; Canestrelli & Nascetti, 2008; Salvi et al., 2013), although in both cases shallow genetic structures were usually found. By contrast, we found ten differentiated and geographically restricted lineages mainly distributed in the northern part of Italy, strongly conforming to the expectations under a scenario of multiple glacial refugia. Multiple refugia across Italy have been observed in a growing number of species (Canestrelli et al., 2010, 2012a, b; Vega et al., 2010; Hewitt, 2011b; Salvi et al., 2013). Most of these refugia were located in the southern part of the species’ range, from where postglacial (re)colonizations of northern environments would have occurred (e.g. Bombina pachypus, Rana italica, Talpa romana, Fagus sylvatica; Canestrelli et al., 2006, 2008, 2010; Magri, 2008). However, our data can firmly discard this latter scenario of southern refugia and postglacial northward expansion for L. v. meridionalis. The topology of the phylogenetic network, with haplogroups occupying terminal positions and intermediate missing nodes, did not support a derivation of one haplogroup from another. This result, together with the estimated divergence times among haplogroups and their geographical distribution in close contiguity, definitely refutes the hypothesis of a recent (re)colonization of a part of the subspecies range from refugial areas located elsewhere (e.g. from southern refugia to northern areas). Furthermore, the occurrence of closely related lineages around the subspecies range (i.e. the nominal subspecies L. v. vulgaris in the north/north-east of the Alpine arch and the closely related species L. italicus in southern Italy) also does not support a scenario of recent (re)colonization from neighbouring areas out of its current range. The close geographical contiguity among haplogroups and their co-occurrence in such areas of contiguity (Fig. 1B) suggest recent secondary contacts among lineages after short-distance expansions. However, a larger sample size as well as the analysis of both nuclear and mitochondrial genetic diversity are necessary to achieve firm conclusions on the demographic history of these lineages. The genetic variance observed within L. v. meridionalis populations and its geographical distribution were mainly explained by genetic differentiation among haplogroups rather than by a pattern of isolation-by-distance; this suggests that historical isolation has probably played a major role in shaping L. v. meridionalis’ genetic structure, relative to more recent micro-evolutionary processes. The most ancient of these divergences was roughly estimated to have occurred during the late Early Pleistocene

© 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

8

M. MAURA ET AL.

0.700

0.700

0.600

A

0.600

0.500

0.500

0.400

0.400

0.300

0.300

0.200

0.200

0.100

0.100

0.000

0.000

2 3 4 5 6 7 8 9 10 11 12 Fct 0.301 0.347 0.413 0.452 0.489 0.506 0.519 0.544 0.570 0.584 0.596 Fsc 0.619 0.530 0.462 0.418 0.353 0.331 0.303 0.267 0.216 0.186 0.153

K

B Number of groups (K )

Group composition

2

(17, 22) (3, 6, 7, 8, 9, 10, 11, 13, 15, 16, 18, 19, 20, 21, 23, 24, 26)

3

(16, 17, 22, 23) (3, 6, 7, 8, 9, 10, 11, 13, 15, 18, 19, 20, 21, 24, 26)

4

(16, 17, 22, 23) (24, 26) (9, 10, 11, 13) (3, 6, 7, 8, 15, 18, 19, 20, 21)

5

(16, 17, 22, 23) (24, 26) (9, 10) (11, 13) (3, 6, 7, 8, 15, 18, 19, 20, 21)

6

(16, 17, 22, 23) (24, 26) (9, 10) (11, 13) (3, 6, 7, 8) (15, 18, 19, 20, 21)

7

(16, 17, 22, 23) (24, 26) (9) (10, 11, 13) (3, 6) (7, 8) (15, 18, 19, 20, 21)

8

(16 17) (22 (16, (22, 23) (24 (24, 26) (9 (9, 10) (11 (11, 13 13, 15) (3 (3, 6) (7 (7, 8) (18 (18, 19 19, 20 20, 21)

9

(16, 17) (22, 23) (24, 26) (9, 10) (11, 13, 15) (3) (6) (7, 8) (18, 19, 20, 21)

10

(16, 17) (22, 23) (24, 26) (9, 10) (11, 13) (3) (6) (7, 8) (19) (15, 18, 20, 21)

11

(16, 17) (22, 23) (24) (26) (9, 10) (11, 13) (3) (6) (7, 8) (19) (15, 18, 20, 21)

12

(16, 17) (22, 23) (24) (26) (9, 10) (11, 13) (3) (6) (7, 8) (19) (18, 21) (15, 20)

Figure 3. A, SAMOVA (spatial analysis of molecular variance) results. Fixation indices FCT and FSC (P < 0.001) relative to each pre-defined value of K from 2 to 12. B, group composition for each value of K.

(1.312 ± 0.290 Mya). Thereafter, L. v. meridionalis underwent a series of splits during the Middle Pleistocene. The transition between the Early and Middle Pleistocene (1.2–0.5 Mya; Head & Gibbard, 2005), also known as the ‘mid-Pleistocene revolution’ (Berger & Jansen, 1994), was characterized by an increase of amplitude in the rhythm of climatic oscillations, with profound effects on species and palaeo-landscapes, especially in the northern hemisphere biota (see Head & Gibbard, 2005; Hewitt, 2011a, and references therein). As largely documented by fossil, pollen and phylogeographical studies in the Western Palaearctic, species underwent cycles of severe reductions and expansions in range distribution, and/or extinctions

over large parts of their ranges (Hewitt, 2000, 2011a, b, and references therein). It is likely that the high number of splits occurring in L. v. meridionalis during the Early–Middle Pleistocene transition was a consequence of multiple fragmentation events primed by these climate changes. Time estimates for the MRCAs of each haplogroup ranged from the Middle Pleistocene to the early Late Pleistocene, pre-dating in all cases the last glacial stage. In turn, this suggests that these lineages would have had independent evolutionary histories over several glacial/interglacial cycles. According to the current geographical distribution of these lineages, L. v. meridionalis could have survived in refugia

© 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

PHYLOGEOGRAPHY OF L. VULGARIS MERIDIONALIS 0.18 0.16

M3

0.14 0.12 0.1

9

Fs = -2.650

R2 = 0.0795** SSDDE = 0.0045 SSDDS = 0.0050

0.08 0.06 0.04 0.02 0 0

1

2

3

4

5

6

7

8

9

0.35

M7

0.3

Fs = -0.465

R2 = 0.1459 SSDDE = 0.0373 SSDDS = 0.0243

0.25 0.2 0.15 0.1 0.05 0 0

1

2

3

4

5

6

0.35

M8

0.3

Fs = -2.163

R2 = 0.1225 SSDDE = 0.0025 SSDDS = 0.0025

0.25 0.2 0.15 0.1 0.05 0 0

1

2

3

4

5

6

0.3

M9

0.25

Fs = -0.733

R2 = 0.1514 SSDDE = 0.0036 SSDDS = 0.0025

0.2 0.15 0.1 0.05 0 0

1

2

3

4

5

6

0.2 0.18 0.16 0.14 0.12

M1

Fs = -0.342

R2 = 0.1219 SSDDE = 0.0169 SSDDS = 0.0157

0.1 0.08 0.06 0.04 0.02 0 0

1

2

3

4

5

6

7

8

9

10

Figure 4. Mismatch distribution analyses and values of the demographic summary statistics for M1, M3, M7, M8 and M9 haplogroups (N > 9) identified by phylogenetic and SAMOVA analyses. Bars (coloured according to Figs 1A and 2) show the observed distributions of pairwise nucleotide differences; ▲ with a solid line shows the expected distribution under a model of a pure demographic expansion; □ with a dotted line shows the expected distribution under a model of spatial expansion. SSD, sum of square deviations for the goodness-of-fit between the observed mismatch distribution and the mismatch distribution expected under both a pure demographic expansion (SSDDE) and a spatial expansion model (SSDSE). FS, Fu’s statistic. R2, Ramos-Onsin and Roza’s test (P < 0.01). © 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

10

M. MAURA ET AL.

located in central Italy (M1, M3 and M7) and especially in northern Italy (M2, M4–M6, M8–M10). During the last glacial periods, particularly the Last Glacial Maximum (LGM), the Alpine glaciers reached the lowland edges of the Po and Venetian plains (Orombelli, Tanzi & Ravazzi, 2004; Hughes & Woodward, 2008). The northern lowlands were dominated by a steppe vegetation (i.e. Poaceae, Artemisia and Chenopodiaceae), with colder and more arid climate conditions than at present (Cremaschi, 1992; Elenga et al., 2000; Prentice et al., 2000; Miola et al., 2003; Mozzi et al., 2003). Furthermore, palaeogeographical, palaeobotanic and palaeoclimatic studies provided evidence for less uniform climatic and vegetation patterns throughout this area, pinpointing a gradient of increasing moisture availability from west to east in the Alpine forelands, and from north to south in the Venetian plain (Florineth & Schlüchter, 2000; Ravazzi et al., 2004). Wet environments have been identified in the southern and distal portion of the Venetian plain (Miola et al., 2006; Fontana, Mozzi & Bondesan, 2008) as well as in the western Po plain (Tropeano & Cerchio, 1984). Moreover, reduced stands of coniferous trees (i.e. Pinus sylvestris, P. mugo, Picea abies and Larix decidua) and even temperate trees (i.e. Corylus avellana, Quercus deciduous, Tilia spp., Ulmus spp., Fraxinus excelsior, Carpinus spp., Abies alba and Fagus sylvatica) would have survived during the LGM in north-eastern Italy and in the central Po plain in small, environmentally favourable sites (Monegato et al., 2007; Amorosi et al., 2008; Kaltenrieder et al., 2009; Pini, Ravazzi & Donegana, 2009; Monegato et al., 2011; Ravazzi et al., 2012). This palaeoenvironmental scenario for northern Italy, with scattered pockets of (micro)environmentally suitable habitats dispersed in a matrix of unsuitable environments for temperate species, echoes the well-known pattern of northern ‘cryptic’ refugia previously discovered in central-eastern and northern Europe (Stewart & Lister, 2001; Willis & van Andel, 2004; Stewart et al., 2010; Schmitt & Varga, 2012). Such a scenario of glacial survival in several refugia located in distinct humid areas throughout the Po and Venetian plains (e.g. the foothills of the glaciated Alps) appears particularly suitable for an amphibian such as L. v. meridionalis, strictly linked to freshwater habitats for its reproduction, with phylopatric habits and with low dispersal abilities (Smith & Green, 2005; Razzetti, Lapini & Bernini, 2007). The long-term persistence of temperate species in a refugium in northern Italy has been suggested for many amphibians and reptiles, based on the occurrence of one divergent phylogeographical lineage endemic to this region (e.g. Hyla intermedia, Pelobates fuscus, Pelophylax lessonae, Triturus carnifex,

Podarcis muralis; Canestrelli et al., 2007, 2012b; Crottini et al., 2007; Canestrelli & Nascetti, 2008; Salvi et al., 2013). A precise identification of the geographical location of these putative refugia is not achievable with the genetic and fossil data to date available for these organisms. However, the occurrence of several lineages of L. v. meridionalis endemic to restricted areas in the west (M9), north-west (M2) and north-east (M4 and M5) of Italy may provide a clue for further phylogeographical and palaeoenvironmental studies aimed to identify and characterize putative glacial refugia in these areas. Finally, the deep phylogeographical structure found in L. v. meridionalis mirrors the patterns found within other lineages of L. vulgaris (e.g. L. v. vulgaris and L. v. graecus; Babik et al., 2005) as well as in other small body newts, such as L. italicus (Canestrelli et al., 2012a), endemic to southern Italy, and L. boscai (Martínez-Solano et al., 2006), endemic to the Iberian Peninsula. Thus, it seems that these newts have been particularly prone to retain the genetic imprints of Pleistocene climatic oscillations, providing very useful models for microevolutionary and phylogeographical studies. However, note that the phylogeographical inferences made in these studies, including ours, are based on mtDNA data alone, which may have some limitations (Ballard & Whitlock, 2004; Toews & Brelsford, 2012; but see also Zink & Barrowclough, 2008). In our study we can safely rule out the most common of these drawbacks, such as the occurrence of sex-biased dispersal of L. v. meridionalis (see, for example, Kovar et al., 2009) or the presence of pseudogenes (see Methods). On the other hand, we cannot rule out the hypothesis that some forms of selection may have biased the mitochondrial phylogeographical pattern of L. v. meridionalis, although these processes have been reported in a minor number of phylogeographical studies (Toews & Brelsford, 2012). Further studies using a wide range of nuclear loci and additional samples in areas of contact between lineages are currently in progress to investigate this issue in detail. In conclusion, our study indicates that since the Early Pleistocene the Italian smooth newt L. v. meridionalis has persisted throughout several glacial–interglacial cycles in the northern part of its range, undergoing events of multiple fragmentation into separate refugia. On the whole, this unprecedented scenario further extends the array of responses of Italian temperate species to the Pleistocene climatic oscillations, and suggests that we are still far from a full appreciation of how the current Italian biota has been assembled. Finally, further studies on multiple species in peri-montane areas of the Alpine arch and northern Apennines are particu-

© 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

PHYLOGEOGRAPHY OF L. VULGARIS MERIDIONALIS larly needed for an appraisal of the temporal and spatial scales at which northern Mediterranean refugia have taken place.

ACKNOWLEDGEMENTS We are grateful to Alessandra Perilli for help in the laboratory, to Catarina Pinho, Iñigo Martínez-Solano and Giovanni Monegato for discussions on an early draft of the manuscript, and four anonymous reviewers for their helpful comments. We thank Alessio Capoccia, Anna Loy, David Fiacchini, Edoardo Razzetti, Lucio Bonato, Francesco Ficetola, Niki Morganti, Francesco P. Caputo, Paolo Cipriani, Daniela Lucente and Valeria Pasqualini for help with sample collection. We also thank Martin Bennett for his review of the English text. Newts were captured under permits from the Italian Ministry of Environment (DPN-2009-0005106). This study was funded by the Agenzia Regionale dei Parchi (ARP-Lazio), by the PRIN projects 20085YJMTC (to M.A.B.) and 2012FRHYRA (to D.C.) from the ‘Ministero dell’Istruzione, dell’Università e della Ricerca’ (MIUR, Italy). D.S. was supported by a post-doctoral grant (SFRH/BPD/66592/2009) of the Fundação para a Ciência e Tecnologia (FCT, Portugal) under the Programa Operacional Potencial Humano (funds from the European Social Fund and Portuguese Ministério da Educação e Ciência) and by the project ‘Genomics and Evolutionary Biology’ cofinanced by North Portugal Regional Operational Programme 2007/2013 ON.2-O Novo Norte (European Regional Development Fund).

REFERENCES Akaike H. 1973. Information theory and an extension of the maximum likelihood principle. In: Petrov BN, Csaki F, eds. Second international symposium on information theory. Budapest: Akademiai Kiado, 267–281. Amorosi A, Dinelli E, Rossi V, Vaiani SC, Sacchetto M. 2008. Late Quaternary palaeoenvironmental evolution of the Adriatic coastal plain and the onset of Po River Delta. Palaeogeography, Palaeoclimatology, Palaeoecology 268: 80–90. Arévalo E, Davis SK, Sites JW. 1994. Mitochondrial DNA sequence divergence and phylogenetic relationships among eight chromosome races of the Sceloporus grammicus complex (Phrynosomatidae) in central Mexico. Systematic Biology 43: 387–418. Babik W, Branicki W, Crnobrnja–Isaloviç J. 2005. Phylogeography of two European newt species – discordance between mtDNA and morphology. Molecular Ecology 14: 2475–2491. Ballard JWO, Whitlock MC. 2004. The incomplete natural history of mitochondria. Molecular Ecology 13: 729–744.

11

Bandelt HL, Forster P, Röhl A. 1999. Median–joining networks for inferring intraspecific phylogenies. Molecular Biology and Evolution 16: 37–48. Berger WH, Jansen E. 1994. Mid-Pleistocene climate shift: the Nansen connection. In: Johannessen OM, Muerch RD, Overland JE, eds. The polar oceans and their role in shaping the global environment. AGU Geophysical Monograph 85: 295–311. Bisconti R, Canestrelli D, Colangelo P, Nascetti G. 2011. Multiple lines of evidence for demographic and range expansion of a temperate species (Hyla sarda) during the last glaciation. Molecular Ecology 20: 5313–5327. Canestrelli D, Aloise G, Cecchetti S, Nascetti G. 2010. Birth of a hotspot of intraspecific genetic diversity: notes from the underground. Molecular Ecology 9: 5432– 5451. Canestrelli D, Cimmaruta R, Costantini V, Nascetti G. 2006. Genetic diversity and phylogeography of the Apennine yellow-bellied toad Bombina pachypus, with implications for conservation. Molecular Ecology 15: 3741–3754. Canestrelli D, Cimmaruta R, Nascetti G. 2007. Phylogeography and historical demography of the Italian treefrog Hyla intermedia reveals multiple refugia, population expansions and secondary contacts within peninsular Italy. Molecular Ecology 16: 4808–4821. Canestrelli D, Cimmaruta R, Nascetti G. 2008. Population genetic structure and diversity of the Apennine endemicstream frog, Rana italica – insights on the Pleistocene evolutionary history of the Italian peninsular biota. Molecular Ecology 17: 3856–3872. Canestrelli D, Nascetti G. 2008. Phylogeography of the pool frog Rana (Pelophylax) lessonae in the Italian peninsula and Sicily: multiple refugia, glacial expansions and nuclear– mitochondrial discordance. Journal of Biogeography 35: 1923–1936. Canestrelli D, Sacco F, Nascetti G. 2012a. On glacial refugia, genetic diversity and microevolutionary processes: deep phylogeographic structure in the endemic newt Lissotriton italicus. Biological Journal of the Linnean Society 105: 42–55. Canestrelli D, Salvi D, Maura M, Bologna MA, Nascetti G. 2012b. One species, three Pleistocene evolutionary histories: phylogeography of the Italian crested newt, Triturus carnifex. PLoS ONE 7: e41754. Cassens I, Mardulyn P, Milinkovitch M. 2005. Evaluating intraspecific ‘network’ construction methods using simulated sequence data: do existing algorithms outperform the global maximum parsimony approach? Systematic Biology 54: 363–372. Clement M, Posada D, Crandall KA. 2000. Tcs: a computer program to estimate gene genealogies. Molecular Ecology 9: 1657–1660. Cremaschi M. 1992. I mutamenti del clima nel Quaternario: le linee generali. La penisola italiana nel Quaternario: aspetti geologici e geoarcheologici. In: Guidi A, Piperno M, eds. Italia Preistorica. Bari: Editori Laterza, 4–58. Crottini A, Andreone F, Kosuch J, Borkin LJ, Litvinchuk SN, Eggert C, Veith M. 2007. Fossorial but widespread: the

© 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

12

M. MAURA ET AL.

phylogeography of the common spadefoot toad (Pelobates fuscus), and the role of the Po Valley as a major source of genetic variability. Molecular Ecology 16: 2734–2754. Doyle JJ, Doyle JL. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11–15. Dupanloup I, Schneider S, Excoffier L. 2002. A simulated annealing approach to define the genetic structure of populations. Molecular Ecology 11: 2571–2581. Elenga H, Peyron O, Bonnefille R, Jolly D, Cheddadi R, Guiot J, Andrieu V, Bottema S, Buchet G, de Beaulieu JL, Hamilton AC, Maley J, Marchant R, Perez-Obiol R, Reille M, Riollet G, Scott L, Straka H, Taylor D, VanCampo E, Vincens A, Laarif F, Jonson H. 2000. Pollen-based biome reconstruction for southern Europe and Africa 18,000 yr BP. Journal of Biogeography 27: 621–634. Ersts PJ. 2012. Geographic distance matrix generator, version 1.2.3. Available at: http://biodiversityinformatics.amnh.org/ open_source/gdmg Excoffier L. 2004. Patterns of DNA sequence diversity and genetic structure after a range expansion: lessons from the infinite-island model. Molecular Ecology 13: 853–864. Excoffier LG, Laval G, Schneider S. 2005. Arlequin (version 3.0): an integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online 1: 47–50. Florineth D, Schlüchter C. 2000. Alpine evidence for atmospheric circulation patterns in Europe during the Last Glacial Maximum. Quaternary Research 54: 294–308. Fontana A, Mozzi P, Bondesan A. 2008. Alluvial megafans in the Veneto-Friuli Plain: evidence of aggrading and erosive phases during Late Pleistocene and Holocene. Quaternary International 189: 71–89. Fu YX. 1997. Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection. Genetics 147: 915–925. Gómez A, Lunt DH. 2007. Refugia within refugia: patterns of phylogeographic concordance in the Iberian Peninsula. In: Weiss S, Ferrand N, eds. Phylogeography in southern European refugia: evolutionary perspectives on the origins and conservation of European biodiversity. Dordrecht: Springer, 155–188. Guindon S, Dufayard JF, Lefort V, Anisimova M, Hordijk W, Gascuel O. 2010. New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 30. Systematic Biology 59: 307–321. Habel JC, Schmitt T, Müller P. 2005. The fourth paradigm pattern of postglacial range expansion of European terrestrial species: the phylogeography of the Marbled White butterfly (Satyrinae, Lepidoptera). Journal of Biogeography 32: 1489–1497. Head MJ, Gibbard PL. 2005. Early–middle Pleistocene transitions: the land–ocean evidence. London: Geological Society, Special Publications. Hewitt GM. 1999. Postglacial re-colonization of European biota. Biological Journal of the Linnean Society 68: 78–112.

Hewitt GM. 2000. The genetic legacy of the Quaternary ice ages. Nature 405: 907–913. Hewitt GM. 2011a. Quaternary phylogeography: the roots to hybrid zones. Genetica 139: 617–638. Hewitt GM. 2011b. Mediterranean peninsulas – the evolution of hotspots. In: Zachos FE, Habel JC, eds. Biodiversity hotspots. Heidelberg: Springer, 123–147. Hughes PD, Woodward JC. 2008. Timing of glaciation in the Mediterranean mountains during the last cold stage. Journal of Quaternary Science 23: 575–588. Jensen JL, Bohonak AJ, Kelley ST. 2005. Isolation by distance, web service. BMC Genetics 6: 13. Kaltenrieder P, Belis CA, Hofstetter S, Ammann B, Ravazzi C, Tinner W. 2009. Environmental and climatic conditions at a potential glacial refugial site of tree species near the Southern Alpine glaciers. New insights from multiproxy sedimentary studies at Lago della Costa (Euganean Hills, Northeastern Italy). Quaternary Science Reviews 28: 2647–2662. Kovar R, Brabec M, Vita R, Bocek R. 2009. Spring migration distances of some Central European amphibian species. Amphibia-Reptilia 30: 367–378. Krystufek B, Buzan EV, Hutchinson WF, Hänfling B. 2007. Phylogeography of the rare Balkan endemic Martino’s vole, Dinaromys bogdanovi, reveals strong differentiation within the western Balkan Peninsula. Molecular Ecology 16: 1221–1232. Larkin MA, Blackshields G, Brown NP, Chenna R, McGettiga PA, McWilliam H, Valentin F, Wallace IM, Wilm A, Lopez R, Thompson JD, Gibson TJ, Higgins DG. 2007. Clustal W and Clustal X version 20. Bioinformatics 23: 2947–2948. Librado P, Rozas J. 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25: 1451–1452. Magri D. 2008. Patterns of post-glacial spread and the extent of glacial refugia of European beech (Fagus sylvatica). Journal of Biogeography 35: 450–463. Martínez-Solano I, Teixeira J, Buckley D, García-París M. 2006. Mitochondrial DNA phylogeography of Lissotriton boscai (Caudata, Salamandridae): evidence for old, multiple refugia in an Iberian endemic. Molecular Ecology 15: 3375– 3388. Miola A, Albanese D, Valentini G, Corain L. 2003. Pollen data for a biostratigraphy of LGM in the Venetian Po Plain. Il Quaternario 16: 21–26. Miola A, Bondesan A, Corain L, Favaretto S, Mozzi P, Piovan S, Sostizzo I. 2006. Wetlands in the Venetian Po Plain (northeastern Italy) during the Last Glacial Maximum: interplay between vegetation, hydrology and sedimentary environment. Review of Palaeobotany and Palynology 141: 53–81. Monegato G, Monegato I, Pini R, Ravazzi C, Reimer PJ, Wick L. 2011. Correlating Alpine glaciation with Adriatic sea-level changes through lake and alluvial stratigraphy. Journal of Quaternary Science 26: 791–804. Monegato G, Ravazzi C, Donegana M, Pini R, Calderoni G, Wick L. 2007. Evidence of a twofold glacial advance

© 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

PHYLOGEOGRAPHY OF L. VULGARIS MERIDIONALIS during the Last Glacial Maximum in the Tagliamento end moraine system (eastern Alps). Quaternary Research 68: 284–302. Mozzi P, Bini C, Zilocchi L, Becattini R, Mariotti Lippi M. 2003. Stratigraphy, palaeopedology and palynology of Late Pleistocene and Holocene deposits in the landward sector of the lagoon of Venice (Italy), in relation to the Caranto level. Il Quaternario – Italian Journal of Quaternary Sciences 16: 193–210. Orombelli G, Tanzi G, Ravazzi C. 2004. Glacier extent over the Italian Alps during the LGM. In: Antonioli F, Vai GB, eds. CLIMEX litho-palaeoenvironmental maps of Italy during the last two climatic extremes: explanatory notes. Bologna: Museo Geologico G. Cappellini, 23–24. Pini R, Ravazzi C, Donegana M. 2009. Pollen stratigraphy, vegetation and climate history of the last 215 ka in the Azzano Decimo core (plain of Friuli, north-eastern Italy). Quaternary Science Reviews 28: 1268–1290. Porretta D, Canestrelli D, Urbanelli S, Bellini R, Schaffner F, Petric D, Nascetti G. 2011. Southern crossroads of the Western Palaearctic during the Late Pleistocene and their imprints on current patterns of genetic diversity: insights from the mosquito Aedes caspius. Journal of Biogeography 38: 20–30. Posada D. 2003. Using Modeltest and PAUP to select a model of nucleotide substitution. In: Baxevanis AD, Davison DB, Page RDM, Petsko GA, Stein LD, Stormo GD, eds. Current protocols in bioinformatics. New York: Wiley, 6.5.1– 6.5.14. Posada D. 2008. jModelTest: phylogenetic model averaging. Molecular Biology and Evolution 25: 1253–1256. Prentice IC, Jolly D, BIOME 6000 participants. 2000. Mid-Holocene and glacial-maximum vegetation geography of the northern continents and Africa. Journal of Biogeography 27: 507–519. Ramos-Onsins SE, Rozas J. 2002. Statistical properties of new neutrality tests against population growth. Molecular Biology and Evolution 19: 2092–2100. Ravazzi C, Deaddis M, De Amicis M, Marchetti M, Vezzoli G, Zanchi A. 2012. The last 40 ka evolution of the Central Po Plain between the Adda and Serio rivers. Géomorphologie: Relief, Processus, Environnement 2: 131– 154. Ravazzi C, Orombelli G, Tanzi G, CLIMEX Group. 2004. An outline of the flora and vegetation of the Adriatic basin (northern Italy and eastern side of the Apennine) during the Last Glacial Maximum. In: Antonioli F, Vai GB, eds. CLIMEX litho-palaeoenvironmental maps of Italy during the last two climatic extremes: explanatory notes. Bologna: Museo Geologico G. Cappellini, 15–19. Razzetti E, Lapini L, Bernini F. 2007. Lissotriton vulgaris. In: Lanza B, Andreone F, Bologna MA, Corti C, Razzetti E, eds. Fauna d’Italia AMPHIBIA. Bologna: Edizioni Calderini de Il Sole 24 ORE Editoria Specializzata S.r.l., 246–254. Rogers AR, Harpending HC. 1992. Population growth makes waves in the distribution of pairwise genetic differences. Molecular Biology and Evolution 9: 552–569.

13

Ronquist F, Teslenko M, van der Mark P, Ayres D, Darling A, Höhna S, Larget B, Liu L, Suchard MA, Huelsenbeck JP. 2012. MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Systematic Biology 61: 539–542. Rousset F. 1997. Genetic differentiation and estimation of gene flow from F-statistics under isolation by distance. Genetics 145: 1219–1228. Rull V. 2009. Microrefugia. Journal of Biogeography 36: 481–484. Salvi D, Harris DJ, Kaliontzopoulou A, Carretero MA, Pinho C. 2013. Persistence across Pleistocene Ice Ages in Mediterranean and extra-Mediterranean refugia: phylogeographic insights from the common wall lizard. BMC Evolutionary Biology 13: 147. Salvi D, Schembri PJ, Sciberras A, Harris DJ. 2014. Evolutionary history of the Maltese wall lizard Podarcis filfolensis: insights on the ‘expansion–contraction’ model of the Pleistocene biogeography. Molecular Ecology 23: 1167– 1187. Schmitt T, Habel JC, Zimmermann M, Müller P. 2006. Genetic differentiation of the marbled white butterfly, Melanargia galathea, accounts for glacial distribution patterns and postglacial range expansion in southeastern Europe. Molecular Ecology 15: 1889–1901. Schmitt T, Varga Z. 2012. Extra-Mediterranean refugia: the rule and not the exception? Frontiers in Zoology 9: 22. Schneeweiss GM, Schönswetter P. 2011. A re-appraisal of nunatak survival in arctic–alpine phylogeography. Molecular Ecology 20: 190–192. Smith MA, Green DM. 2005. Are all amphibian populations metapopulations? Dispersal and the metapopulation paradigm in amphibian ecology. Ecography 28: 110–128. Stewart JR. 2009. The evolutionary consequence of the individualistic response to climate change. Journal of Evolutionary Biology 22: 2363–2375. Stewart JR, Lister AM. 2001. Cryptic northern refugia and the origins of modern biota. Trends in Ecology & Evolution 16: 608–613. Stewart JR, Lister AM, Barnes I, Dalén L. 2010. Refugia revisited: individualistic responses of species in space and time. Proceedings of the Royal Society B 77: 661–671. Tamura K, Nei M. 1993. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Molecular Biology and Evolution 10: 512–526. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S. 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Molecular Biology and Evolution 28: 2731–2739. Toews DPL, Brelsford A. 2012. The biogeography of mitochondrial and nuclear discordance in animals. Molecular Ecology 21: 3907–3930. Tropeano D, Cerchio E. 1984. L’orizzonte torboso würmiano nel sottosuolo della pianura piemontese meridionale. Osservazioni preliminari. Bollettino dell’Associazione Mineraria Subalpina 21: 199–221.

© 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

14

M. MAURA ET AL.

Vega R, Amori G, Aloise G, Cellini S, Loy A, Searle JB. 2010. Genetic and morphological variation in a Mediterranean glacial refugium: evidence from Italian pygmy shrews, Sorex minutus (Mammalia: Soricomorpha). Biological Journal of the Linnean Society 100: 774–787. Willis KJ, van Andel TH. 2004. Trees or no trees? The environments of central and eastern Europe during the Last Glaciation. Quaternary Science Reviews 23: 2369–2387.

Xia X, Xie Z. 2001. DAMBE: data analysis in molecular biology and evolution. Journal of Heredity 92: 371–373. Xia X, Yang Q. 2011. A distance-based least-square method for dating speciation events. Molecular Phylogenetics and Evolution 59: 342–353. Zink RM, Barrowclough GF. 2008. Mitochondrial DNA under siege in avian phylogeography. Molecular Ecology 17: 2107–2121.

SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s web-site: Figure S1. Phylogenetic relationships among the 49 haplotypes found in Lissotriton vulgaris meridionalis based on Bayesian inference (BA) and maximum likelihood (ML) analyses. Bayesian posterior probabilities (> 0.90) are given above the nodes of the BA tree; bootstrap values (> 70) for ML are given below the nodes. Table S1. Times to the most recent common ancestor (TMRCA) of clades within Lissotriton vulgaris meridionalis (in Myr) estimated in DAMBE. Standard deviations are shown in parenthesis.

© 2014 The Linnean Society of London, Biological Journal of the Linnean Society, 2014, ••, ••–••

Supplementary Figure S1. Phylogenetic relationships among the 49 haplotypes found in Lissotriton vulgaris meridionalis based on Bayesian inference (BA) and Maximum likelihood (ML) analyses. Bayesian posterior probabilities (>0.90) are given above the nodes of the BA tree; bootstrap values (>70) for ML are given below the nodes. M10a M10 1 M9d 0.92 0.0040

M9c M9e M9b 89

0.98 76

0.99

M9a M8d M8e M8g M8f M8b M8a M8c M7f 1 M7d 86 M7e M7b 0.90 M7c M7a 74

M9

0.90

1 95

1 95

M7

M6b

M6a

1 96

0.90 70

M8

M6

M6d M6c M6e M5c

1 96

M5a M5b M4c

1

1 84

1 86

M4a M4d

M4

M4b

M3d M3a M3e

1 75

M3f M3g

1 0.99

88 M1a

70

73

77 M1c 1 77

M3

M3h 1

M3l M3i

77

1 98

M5

M3b M3c M2a M1f M1g M1b

M1d M1e

Lissotriton vulgaris vulgaris

M2

M1

Supplementary Table S1. Times to the the most recent common ancestor (TMRCA) of clades within Lissotriton vulgaris meridionalis (in million years) estimated in DAMBE. Standard deviations are shown in parenthesis.

Clade Ingroup M1, M3-M10 M3-M7, M8-M10 M7, M3-M6 M3, M4-M6 M4, M5, M6 M8, M9-M10 M9, M10 M1 M5 M3 M7 M4 M6 M9 M8

TMRCA 1.312 (± 0.290) 1.106 (± 0.219) 0.974 (± 0.196) 0.878 (± 0.190) 0.736 (± 0.176) 0.696 (± 0.179) 0.645 (± 0.178) 0.539 (± 0.208) 0.444 (± 0.131) 0.316 (± 0.146) 0.304 (± 0.097) 0.272 (± 0.087) 0.271 (± 0.104) 0.239 (± 0.115) 0.228 (± 0.114) 0.166 (± 0.070)