ArticlePDF Available

A pattern of hybridisation and population genetic structure of two water frog species (Ranidae, Amphibia) in the southwestern Balkans

February 2025
Zoologica Scripta 54(4):1-22

DOI:10.1111/zsc.12723

Authors:

Petr Papežík

Comenius University Bratislava

Michal Benovics

Masaryk University

Dmitrij Dedukh

Show all 14 authorsHide

The distribution of the neutral component of genetic diversity is the interplay of historical and ongoing processes resulting in the species-specific genetic structure of populations, which can, however, be disrupted by interspecific hybridisation and introgression. In this study, we focused on two species of water frogs, Pelophylax epeiroticus and P. kurtmuelleri, which live in sympatry in the southwestern Balkans, to investigate the rate of hybridisation and population genetic structure using cytogenetic, mitochondrial (ND2) and nuclear DNA (microsatellite) markers. The overall hybridisation rate was 2.6%, with rates reaching up to 10% at specific sites. The course of gametogenesis and the occurrence of later generations of hybrids (beyond the F1 generation) indicate a sexual mode of hybrid reproduction. The bimodal structure of hybrid populations and the rarity of hybrids suggest substantial reproductive isolation between the two species; however, this isolation is unlikely attributable to differences in ecological niche occupation. In P. epeiroticus, sequence variation in the ND2 gene revealed two divergent lineages with a clear geographic pattern that corresponds

Content uploaded by Petr Papežík

Content may be subject to copyright.

Zoologica Scripta. 2025;00:1–22. wileyonlinelibrary.com/journal/zsc

Received: 23 July 2024

Revised: 6 February 2025

Accepted: 8 February 2025

DOI: 10.1111/zsc.12723

ORIGINAL ARTICLE

A pattern of hybridisation and population genetic structure

of two water frog species (Ranidae, Amphibia) in the

southwestern Balkans

PetrPapežík1

SandraAschengeschwandtnerová1

MichalBenovics1,2

DmitrijDedukh3

MarieDoležálková- Kaštánková3

LukášCholeva4,5

AdamJavorčík1

PetrosLymberakis6

SimonaPapežíková1

NikosPoulakakis6,7,8

İdrisSarı9

RadekŠanda10

JasnaVukić11

PeterMikulíček1

1Department of Zoology, Faculty of Natural Sciences, Comenius University in Bratislava, Bratislava, Slovakia

2Department of Botany and Zoology, Faculty of Science, Masaryk University, Brno, Czech Republic

3Laboratory of NonMendelian Evolution, Institute of Animal Physiology and Genetics, The Czech Academy of Sciences, Liběchov, Czech Republic

4Laboratory of Fish Genetics, Institute of Animal Physiology and Genetics, The Czech Academy of Sciences, Liběchov, Czech Republic

5Department of Biology and Ecology, Faculty of Science, University of Ostrava, Ostrava, Czech Republic

6Natural History Museum of Crete, School of Sciences and Engineering, University of Crete, Heraklion, Greece

7Department of Biology, School of Sciences and Engineering, University of Crete, Voutes University Campus, Heraklion, Greece

8Ancient DNA Lab, Institute of Molecular Biology and Biotechnology (IMBB), Foundation for Research and Technology – Hellas (FORTH),

Heraklion, Greece

9Department of Biology, Faculty of Science and Art, Erzincan Binali Yıldırım University, Erzincan, Turkey

10Department of Zoology, National Museum of the Czech Republic, Prague 1, Czech Republic

11Department of Ecology, Faculty of Science, Charles University, Prague 2, Czech Republic

Correspondence

Petr Papežík, Department of Zoology,

Faculty of Science, Comenius

University in Bratislava, Mlynská

dolina, Ilkovičova 6, Bratislava 4 842 15,

Slovakia.

Email: petr.papezik.upol@gmail.com

Funding information

Vedecká Grantová Agentúra MŠVVaŠ

SR a SAV, Grant/Award Number:

1/0014/24

Abstract

The distribution of the neutral component of genetic diversity is the interplay

of historical and ongoing processes resulting in the species- specific genetic

structure of populations, which can, however, be disrupted by interspecific

hybridisation and introgression. In this study, we focused on two species of water

frogs, Pelophylax epeiroticus and P. kurtmuelleri, which live in sympatry in the

southwestern Balkans, to investigate the rate of hybridisation and population

genetic structure using cytogenetic, mitochondrial (ND2) and nuclear DNA

(microsatellite) markers. The overall hybridisation rate was 2.6%, with rates

reaching up to 10% at specific sites. The course of gametogenesis and the

occurrence of later generations of hybrids (beyond the F1 generation) indicate a

sexual mode of hybrid reproduction. The bimodal structure of hybrid populations

and the rarity of hybrids suggest substantial reproductive isolation between the

two species; however, this isolation is unlikely attributable to differences in

ecological niche occupation. In P. epeiroticus, sequence variation in the ND2 gene

revealed two divergent lineages with a clear geographic pattern that corresponds

PAPEŽÍK etal.

INTRODUCTION

The distribution of genetic variation is spatially struc-

tured and determined by historical processes, such as

population bottlenecks, founder events or past evolu-

tionary history and ongoing ecological and evolution-

ary forces associated with genetic drift and gene flow.

Genetic drift is particularly pronounced in small popu-

lations and leads to allele fixation and a reduction of ge-

netic diversity (e.g., Allendorf etal.,2012). Conversely,

gene flow tends to homogenise the genetic structure of

populations by introducing new alleles, thus prevent-

ing genetic divergence and making populations genet-

ically similar. The usual prerequisite for gene flow is

dispersal, which is species- specific and can vary even

among phylogenetically closely related species (Bowler

& Benton,2005; Edelaar & Bolnick,2012).

The equilibrium between genetic drift and gene flow

can be disrupted by interspecific hybridisation and sub-

sequent introgression (interspecific gene flow), which

introduces novel genetic variation into populations and

thus potentially blurs species boundaries. The extent of

hybridisation and introgression and their effect on the

population genetic structure of the hybridising species

depends on the strength of the reproductive isolation

barriers (Abbott etal., 2013; Kollár etal.,2022; Taylor &

Larson, 2019; Westram et al., 2022). When species with

a different genetic structure exhibit strong reproductive

isolation and introgression is limited, the genetic struc-

ture tends to remain distinct between the species. In cases

where reproductive barriers are weak, interspecific gene

flow may proceed more freely, upsetting the equilibrium

between gene flow and drift and thus the population ge-

netic structure of individual species.

The southwestern Balkans is recognised as an area

of high endemism and exceptional genetic diversity

(Crnobrnja- Isailovic,2007; Džukić & Kalezić,2004). This

region is also home to two water frog species, the Balkan

water frog, Pelophylax kurtmuelleri (Gayda, 1940) and

the Epirus water frog, Pelophylax epeiroticus (Schneider,

Sofianidou et Kyriakopoulou- Sklavounou,1984), whose

ranges overlap vastly. Divergence of the lineages leading

to these species likely occurred in the late Miocene, ap-

proximately 8.3 Mya (Dufresnes etal.,2024). Pelophylax

kurtmuelleri is a disputable taxon representing either

a valid species (e.g., Plötner etal., 2010, 2012) or a di-

verged evolutionary lineage within P. ridibundus (e.g.,

Lymberakis etal.,2007; Speybroeck etal.,2010, 2020).

Although this species is considered a Balkan endemic,

widely distributed in the southwest of the peninsula

(e.g., Valakos et al., 2008), its haplotypes and alleles

have been recorded in other parts of Europe, either as

a result of human- mediated introduction or historical

hybridisation with other water frog species (Kolenda

etal.,2017; Litvinchuk etal.,2020; Papežík etal.,2023).

In contrast, P. epeiroticus has a smaller range extending

from southern Albania to the northwestern Peloponnese,

including the island of Corfu. At the eastern edge,

its range is bounded by the Pindus Mountain range

(Oruçi, 2008; Schneider & Haxhiu, 1992; Sofianidou

et al., 1987; Sofianidou & Schneider, 1989). Recently,

two deeply diverged mitochondrial lineages, which are

geographically structured and come into contact around

the Ambracian Gulf in Greece, were identified within

P. epeiroticus (Papežík etal.,2023). In P. kurtmuelleri, the

detected mtDNA lineages show only shallow divergence

but higher haplotype diversity compared to P. epeiroticus

(Papežík etal.,2023).

Previous studies based on allozyme markers and bio-

acoustics (mating calls of males) revealed interspecific

hybridisation between P. epeiroticus and P. kurtmuelleri.

However, the evidence comes only from two sympatric

populations in Ioannina and Sagiada (western Greece),

where the proportion of hybrids did not exceed 15%

(Schneider et al., 1984; Schneider & Sofianidou, 1986;

Sofianidou et al., 1987; Sofianidou & Schneider, 1989).

to the genetic structure in microsatellite markers. In contrast, P. kurtmuelleri

populations were not as geographically structured and showed only weak genetic

differentiation in both types of markers. Pelophylax epeiroticus was significantly

less variable at microsatellite loci compared to P. kurtmuelleri, which, together

with the high differentiation of its populations, suggests a stronger influence of

genetic drift. We can hypothesise that the differential strength of genetic drift in

the two species may lead to unequal interspecific gene flow.

KEYWORDS

gene flow, genetic variation, microsatellites, mitochondrial DNA, Pelophylax epeiroticus,

Pelophylax kurtmuelleri

14636409, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/zsc.12723 by Commenius University, Wiley Online Library on [20/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

PAPEŽÍK etal.

While specific reproductive barriers reducing the hy-

bridisation rate remain unknown, field studies have re-

vealed instances of temporal and spatial isolation of both

species during the breeding season, suggesting possible

mechanisms of prezygotic isolation (Plötner et al., 2010;

Schneider et al., 1984; Sofianidou & Schneider, 1989).

Hybrids between P. epeiroticus and P. kurtmuelleri appear

to be sexual, in contrast to asexual (hybridogenetic) hybrids

that originate from hybridisation between P. ridibundus

(a sister species/phylogenetic lineage of P. kurtmuelleri)

with other species (P. lessonae in central Europe, P. perezi

in southern France and northeastern Spain and P. bergeri

in Italy) (Guerrini etal.,1997; Hotz etal.,1985; Hotz &

Uzzell,1982). However, the presumed sexual mode of re-

production of P. kurtmuelleri × P. epeiroticus hybrids does

not correspond to the presence of triploid hybrids that

have been identified in western Greece (Kyriakopoulou-

Sklavounou & Vasara,2001; Sofianidou,1996). They were

characterised by the presence of either two genomes of

P. kurtmuelleri and one genome of P. epeiroticus, or vice

versa, and their occurrence would suggest a clonal rather

than a sexual mode of reproduction, as increased ploidy is

thought to be a consequence of hybridisation and asexual-

ity (Choleva etal.,2012).

In this study, we investigated the extent of hybridisa-

tion and population genetic structure of P. epeiroticus and

P. kurtmuelleri in the southwestern Balkans. We aimed

(1) to find the rate of interspecific hybridisation and con-

firm the mode of reproduction (sexual versus asexual) of

hybrids, (2) to compare patterns of genetic diversity and

population structure inferred from mitochondrial and mi-

crosatellite markers and (3) to compare ecological niche of

both species and evaluate its impact on their hybridisation

and population genetic structure.

MATERIALS AND METHODS

2.1

Sampling

In total, we collected 306 samples of water frogs from

14 localities in Albania and Greece, covering the entire

range of P. epeiroticus and the sympatric part of the range

of P. kurtmuelleri (Table1, TableS1, Figure1). Sampling

was random, regardless of frog phenotype, to avoid bias in

species and hybrid abundance estimates at the sampled

sites. For species identification in the field, we followed the

morphological criteria proposed by Papežík etal. (2021).

For further genetic analyses, a blood sample or toe clip

was collected from each individual. All sampled frogs were

subsequently released at the original site. Tissue samples

were preserved in 96% ethanol and stored at −25°C.

2.2

Laboratory procedures

Genomic DNA was extracted using the NucleoSpin®

Tissue kit (Macherey- Nagel, Düren, Germany) following

the manufacturer's protocol. Complete mitochondrial

DNA (mtDNA) NADH dehydrogenase 2 (ND2) gene,

1038 bp, was amplified using a combination of primers

ND2- L1 and ND2- H2 (Plötner et al., 2008). PCR was

carried out in a total volume of 10 μL with 5 μL of VWR

Red Taq 2× mix with 1.5 mM MgCl2 (VWR, Radnor, PA,

USA), 0.2 μL of each primer (10 μM), 3.6 μL of ddH2O and

1 μL of DNA. The following PCR program was used: initial

denaturation for 2 min at 94°C, followed by 35 cycles

consisting of denaturation for 30 s at 94°C, primer

annealing for 20 s at 63°C and elongation for 1 min at 72°C,

with a final elongation step for 10 min at 72°C (modified

Locality (abbreviation) Latitude (N) Longitude (E) n

Fragma Kalama (FK) 39.58 20.25 9

Igoumenitsa (IG) 39.54 20.20 41

Ioannina (IO) 39.69 20.86 44

Kalogria (KA) 38.16 21.37 19

Limanaki (LI) 38.17 21.42 32

Louros (LO) 39.15 20.77 13

Panetolio (PA) 38.59 21.47 10

Ropa Valley (RV) 39.63 19.79 15

Stjar (ST) 39.93 20.05 46

Sybanion (SY) 37.64 21.48 10

Syri i Kaltër (SK) 39.92 20.19 6

Vrisera (VR) 39.88 20.36 18

Xare (XA) 39.73 20.05 8

Zirou Lake (ZL) 39.24 20.85 35

TABLE  A list of analysed

populations of water frogs of the genus

Pelophylax. For each site, geographic

coordinates and the number of sampled

individuals (n) are provided.

PAPEŽÍK etal.

from Plötner et al., 2008). Purification of PCR products

was performed using the EPPiC Fast enzymatic clean- up

(A&A Biotechnology, Gdansk, Poland) following the

manufacturer's protocol. PCR products were commercially

sequenced from both strands at Macrogen Europe

(Amsterdam, the Netherlands) using the same primers

as for PCR. In addition to the newly obtained sequences,

178 sequences obtained in the previous study (Papežík

et al., 2023) were added to the dataset. Sequences were

aligned with the Geneious algorithm and default setting,

visually checked and translated to detect stop codons using

Geneious Prime v. 2020.2.4 (Biomatters, Auckland, New

Zealand). The newly generated sequences were deposited

in the NCBI GenBank database (Benson etal.,2013) under

accession numbers PQ570882–PQ570898.

Seventeen microsatellite loci were amplified in three

multiplex PCRs (TableS2). PCR reactions were per-

formed in a total volume of 10 μL and consisted of 5 μL of

Qiagen Microsatellite PCR Master mix (Qiagen, Hilden,

Germany), 0.2 or 0.1 μL of each primer (10 μM), 1 μL

of DNA and ddH2O. PCR program was modified from

Christiansen and Reyer(2009): 5 min of initial denatur-

ation at 95°C followed by 30 cycles of denaturation for 30 s

at 95°C, 60°C for 90 s and 72°C for 1 min, with a final ex-

tension at 60°C for 30 min. Microsatellite fragments were

run on an automated ABI 3130 genetic analyser (Applied

FIGURE  Distribution of Pelophylax epeiroticus (EPE; red) and P. kurtmuelleri (KURT; green) genomes in the southwestern Balkans

based on mtDNA and microsatellite markers. (a) Assignment of individuals to species- specific mtDNA groups. (b) Assignment of individuals

to species- specific microsatellite groups based on the Bayesian analysis in Structure. (c) The proportion of P. epeiroticus and P. kurtmuelleri

nuclear genomes in each population based on microsatellites. (d) Discriminant analysis of principal components (DAPC) indicating the

position of parental species and hybrids (HYBR) in multivariate space.

PAPEŽÍK etal.

Biosystems, Carlsbad, CA, USA) and their genotypes were

determined using Geneious Prime v. 2020.2.4.

2.3

Estimation of genotyping

errors and null alleles in microsatellite loci

Potential genotyping errors like stuttering, allelic dropout

or the presence of null alleles were tested separately for

each species using the program Micro- Checker v. 2.2

(Oosterhout etal.,2004). The test was conducted in four

P. kurtmuelleri (Igoumenitsa, Stjar, Vrisera and Zirou

Lake) and six P. epeiroticus (Igoumenitsa, Ioannina,

Kalogria, Limanaki and Louros) populations. In the other

populations, there was not a sufficient number of samples

to perform the analysis. The majority of microsatellite

loci in all populations tested were in Hardy–Weinberg

equilibrium. Only the loci Rrid169A, RlCA1b5, Re2Caga3

and Res14 in the P. kurtmuelleri genome and RICA1b6,

RlCA1b5 and Res14 in the P. epeiroticus genome showed

signs of null alleles. However, the presence of null alleles

at each locus was only detected in some populations.

None of the loci showed null alleles in all populations.

Therefore, we retained all microsatellite loci for further

population genetic analyses.

2.4

Mitochondrial DNA variability

To assign mtDNA haplotypes to a particular species, a

parsimony network algorithm of TCS (Clement etal.,2000)

implemented in PopART v. 1.7 (Leigh & Bryant, 2015)

was used. Nucleotide diversity (π), haplotype diversity

(hd), number of haplotypes (h), segregating sites (S) and

Watterson's theta per site (θW) were used to estimate the

genetic diversity using DnaSP v. 6.00 (Rozas etal.,2017).

Uncorrected p- distances were calculated using the ape

package v. 5.8 (Paradis & Schliep,2019) in the R statistical

environment v. 4.1.3 (R Core Team,2024).

2.5

Detection of hybrids

To estimate the rate of hybridisation, two Bayesian

inference methods implemented in the programs

Structure v. 2.3.4 (Pritchard etal.,2000) and NewHybrids

v. 1.1 Beta3 (Anderson & Thompson,2002), and a series of

ordination methods (PCA, PCoA, DAPC) were applied to

microsatellite genotypes.

In Structure, we calculated the parameter q for each

individual, that is, the proportion of an individual's ge-

nome originating in one of the two inferred clusters (fixed

K = 2), corresponding to parental species P. epeiroticus and

P. kurtmuelleri. This approach allowed the detection of

individuals with mixed (hybrid) ancestry. The admixture

and non- correlated allele frequency models with 106 iter-

ations, following a burn- in period of 105 iterations, were

used. The admixture model in Structure assumes that in-

dividuals may have ancestry from multiple populations or

clusters, e.g., from two parental species as is the case in

this study. In the non- correlated allele frequency model,

Structure does not assume that allele frequencies in dif-

ferent populations are correlated. We selected this model

because we did not expect that the geographically distant

populations (such as those in our study) share recent

ancestry. A series of five independent runs was carried

out with the same parameters to test the accuracy of the

results.

The program NewHybrids v. 1.1 Beta3 was used to

compute the posterior probability that an individual be-

longs to one of the defined hybrid classes (F1, F2 and

backcross hybrid classes B1) or parental species. All indi-

viduals were assigned to defined categories without any

prior population information. Jeffrey's prior for π and Θ,

and 106 iterations, following a burn- in period of 105 itera-

tions were used.

To visualise a multivariate microsatellite dataset of the

parental species and potential hybrids, principal compo-

nent analysis (PCA) and subsequent discriminant analysis

of principal components (DAPC) implemented in the ade-

genet package v. 2.1.10 (Jombart,2008) were performed.

DAPC analysis was conducted to infer genetic differences

in multivariate space with a priori- defined clusters match-

ing parental species. The optimal number of PCs retained

for DAPC was determined as k−1, where k is the num-

ber of clusters. This criterion should capture maximal

among- population variation, while is also less sensitive

to unintended interpretations of the population struc-

ture (Thia, 2023). We followed the recommended stan-

dard reporting for DAPC suggested by Miller etal.(2020).

Prior to the analysis, each individual was assigned to the

inferred cluster based on Structure results. Another ordi-

nation method used for microsatellite data was principal

coordinate analysis (PCoA) with distances proposed by

Peakall and Smouse(2006) obtained by gd.smouse() func-

tion from the PopGenReport package v. 3.1 (Adamack &

Gruber,2014).

2.6

Intraspecific population genetic

structure

To infer intraspecific population genetic structure, we

used different approaches, namely Bayesian cluster-

ing implemented in Structure, ordination methods PCA

and DAPC, analysis of molecular variance (AMOVA),

PAPEŽÍK etal.

genetic distances and the isolation- by- distance (IBD).

Analyses were performed separately for the P. epeiroticus

and P. kurtmuelleri datasets, excluding individuals that

showed mixed ancestry (putative hybrids). For analyses of

the population structure of P. epeiroticus, 10 loci were used

(RlCA1b5, GA1A19, Rrid013, Res22, Pper4.7, Pper3.22,

Rrid059A, Rrid082A, Res14, RlCA1b6); the remaining

seven loci either did not amplify at all or were amplified

only in a part of individuals. In P. kurtmuelleri, population

genetic analyses were performed with the full set of 17 loci

as well as with a restricted set of 10 loci that were simulta-

neously amplified in both species to verify the consistency

of the population genetic structure.

In Structure, the number of clusters (K) was assumed

to be between one and 10. All other parameters were set

as above. The most likely number of K was inferred using

the statistic ΔK (Evanno etal.,2005) implemented in the

pophelper package v. 2.3.1 (Francis,2017) as part of the

R statistical environment and using the thermodynamic

integration (TI) method implemented in the program

MavericK v. 1.0.5 (Verity & Nichols,2016).

PCA and DAPC were performed as described above,

however, with a priori- defined clusters matched sampled

populations within each species.

Genetic differentiation between populations of the in-

dividual species was estimated using Jost's D (Jost,2008),

Nei's GST (Nei,1973; Nei & Chesser,1983) and Hedrick's

G'ST (Hedrick,2005; Meirmans & Hedrick,2011) statistics

using the mmod package (Winter,2012), and FST statis-

tics using GenAlEx v. 6.51 (Peakall & Smouse,2012). The

presence of population genetic structure in each species

was tested by the analysis of molecular variance (AMOVA)

using FST statistics as implemented GenAlEx. The par-

tition of variance was calculated within and among in-

dividuals, among populations (localities) and among

Bayesian (Structure) inferred clusters. Significance of the

AMOVA was assessed by 10,000 permutations.

The effect of IBD among populations was tested by

the correlation between pairwise genetic distances (in-

ferred as FST, Jost's D, Nei's GST and Hedrick's G'ST) and

geographic Euclidean distances using the function man-

tel() from the vegan package v.2.7- 0 (Oksanen etal.,2024)

based on 10,000 permutations.

Parameters of genetic diversity in microsatellite loci

(NA—mean number of alleles, HO—observed hetero-

zygosity, HE—expected heterozygosity) were calculated

using GenAlEx.

All analyses except for mtDNA (DnaSP) and micro-

satellite (GenAlEx) variation and Bayesian methods

(Structure, NewHybrids) were performed in the R statisti-

cal environment v. 4.1.3 (R Core Team,2024). The graph-

ical outputs were created by the ggplot2 package v. 3.5.1

(Wickham,2016).

2.7

Ecological niche modelling

For ecological niche modelling, the occurrence data of both

species were retrieved from species distribution modelling

(SDM) datasets used in Papežík etal.(2023) and thinned

using the SpThin package v. 0.2.0 (Aiello- Lammens

etal.,2015). In P. kurtmuelleri, localities from Thrace and

northern Albania were omitted to retain only localities in

the proximity of the P. epeiroticus range. In total, 66 thinned

records for P. epeiroticus and 108 for P. kurtmuelleri were ob-

tained. Climatic variables included 19 Bioclimatic variables

downloaded from the WorldClim website (https:// www.

world clim. org/ data/ biocl im. html) (Fick & Hijmans,2017)

and 17 Envirem variables derived from temperature mini-

mum, maximum and precipitation data, according to the

methods of Title and Bemmels(2018), with a resolution of

2.5 arc- minutes.

To select variables with the highest contribution in the

current distribution of the two species, the MaxEnt model

was used as implemented in the SDMtune package v. 1.3.1

(Vignali et al., 2020). Initially, 1000 background points

were extracted using the dismo package v. 1.3- 14 (Hijmans

etal.,2011). Subsequently, the variables were ranked based

on permutation importance. Variables ranked as the most

important were examined if they displayed significant

correlations (R2 > .8) with other variables. Finally, a leave-

one- out Jackknife test was conducted among the cor-

related variables according to Elith etal.(2010). Variables

selected for further analyses are summarised in Table2.

Niche analyses were performed using the packages

ENMTools v. 1.1.2 (Warren etal.,2021) and Humboldt v.

1.0.0 (Brown & Carnaval,2019). In the Humboldt pack-

age, we calculated Equivalence and Background statistics,

followed by the Niche Divergence Test (NDT) and the

Niche Overlap Test (NOT; see Brown & Carnaval,2019 for

details). This approach emphasises the quantification of

niche similarity solely in environmental space (E- space),

which proves to be less sensitive than geographic space

(G- space) to variations in the spatial abundance of crucial

environmental variables. The results obtained from the

NDT and NOT were interpreted according to Brown and

Carnaval(2019).

To conduct further niche analyses, the ‘species.ob-

jects’ were constructed using species presence data and

selected variables within the ENMTools package (Warren

etal.,2021). Considering the ecological characteristics of

the species and projected areas, 1000 pseudo- absences

were generated with a radius of 50,000 m around each

presence point. The MaxEnt algorithm was then em-

ployed to fit niche models.

In the ENMTools package, first, a niche identity test

was performed using 25 replicates of MaxEnt mod-

els with a model formula for empirical models. Next, a

PAPEŽÍK etal.

Background test was run, comparing one species' actual

occurrence to random background occurrences of other

species, implementing an asymmetric test as proposed

by Warren et al. (2008). The ‘enmtools.species’ objects

were also utilised for conducting identity and background

tests, simplifying the interface following the approach by

Broennimann etal.(2012). Thus, the Ecospat test was per-

formed to account for environmental availability when

assessing overlaps, ensuring that overlaps are not solely

influenced by varying availability of environmental vari-

ables. All the test results provided replicate models, p-

values and plots of the outcomes. Ecological niche overlap

was assessed using indices of equivalence (D) and similar-

ity (I) proposed by Warren etal.(2008, 2010), both ranging

from 0 to 1, denoting no overlap to complete similarity be-

tween the models (Warren etal.,2019).

To compare the hypsometric distribution of species,

elevation data were obtained for the sites used in the eco-

logical niche modelling. The mean elevation for both spe-

cies was tested using the Wilcoxon rank sum test (Mann &

Whitney,1947) in the R statistical environment.

2.8

Cytogenetic methods

Interspecific hybridisation in the genus Pelophylax results

in either sexual or asexual hybrids (Plötner,2005). Asexual

hybrids reproduce by hybridogenesis, which is character-

ised by the elimination of the genome of one parental spe-

cies and the formation of gametes with the unrecombined

genome of the other parental species. The eliminated ge-

netic material is excluded during gametogenesis in the

form of so- called micronuclei, the presence of which can

be considered evidence of hybridogenetic gametogen-

esis (Chmielewska etal.,2018; Dedukh etal.,2017, 2020;

Dedukh & Krasikova,2022; Ogielska, 1994). In the pre-

sent study, gametogenesis was analysed in four F1 hybrids

resulting from experimental crossing between two pairs of

frogs (one cross of P. epeiroticus female and P. kurtmuel-

leri male and one reciprocal cross). Pelophylax epeiroticus

female and male originated from the localities of Ioannina

and Igoumenitsa, respectively, in western Greece; both

sexes of P. kurtmuelleri originated from Velipoje, Albania.

Experimental crossings were carried out at the Institute

of Animal Physiology and Genetics ASCR, Liběchov,

Czech Republic based on the protocol described in Berger

etal.(1994) and Pruvost etal.(2013).

Four F1 hybrids from two crosses were used for the

analysis of gonadal microanatomy and identification of

genome composition of germ cells by whole- mount flu-

orescent insitu hybridisation with RrS1 probe, which

is specific to pericentromeric regions of both P. kurt-

muelleri and P. epeiroticus chromosomes (Choleva

etal.,2023). Probe labelling by biotin was performed by

PCR from the genomic DNA of P. ridibundus. RrS1 re-

peat was labelled using the following primers: Forward:

5′- AAGCCGATTTTAGACAAGATTGC- 3′, Reverse:

5′- GGCCTTTGGTTACCAAATGC- 3′. Whole- mount

FISH was performed following Choleva et al. (2023).

Gonadal tissues were permeabilised in 0.5% solution

of Triton X100 in 1× PBS for 4–5 h and impregnated by

50% formamide, 10% dextran sulfate and 2× SSC (saline-

sodium citrate buffer; 20 × SSC – 3 M NaCl 300 mМ

Na3C6H5O7) for 3–4 h at 37°C. Afterwards, gonads were

transferred to a hybridisation mixture including RrS1

(20 ng/μL) probe diluted in 50% formamide, 2× SSC

and 10% dextran sulfate and salmon sperm DNA (with

10–50- fold excess compared to the probe concentration).

Then hybridisation mixture with gonads was denatured

at 72°C for 15 min and incubated at least for 24 h at room

TABLE  The most important environmental variables for Pelophylax epeiroticus and P. kurtmuelleri with the contribution and

permutation importance retrieved from the SDMTune package.

P. epeiroticus P. kurtmuelleri

Variable Contribution (%)

Permutation

Importance Variable Contribution (%)

Permutation

Importance

Bio03 35.48 32.55 Bio12 28.47 29.78

Bio16 27.62 37.25 Bio03 19.54 4.38

Bio05 14.21 0.02 PETDriestQuarter 13.61 0.71

Bio14 6.49 6.87 Bio17 12.74 17.01

Bio02 3.84 2.60 PETWettestQuarter 9.91 12.46

PETDriestQuarter 3.00 5.26 Bio09 7.43 10.46

Bio09 2.92 3.56 Bio07 4.51 15.17

PETWettestQuarter 2.75 6.50 PETWarmestQuarter 3.39 8.33

Bio08 2.62 1.07 Bio08 0.39 1.70

Continentality 1.07 4.31 — — —

PAPEŽÍK etal.

temperature (RT) followed by washing in 0.2× SSC at

50°C. After incubation in a blocking solution [4× SSC

containing 1% blocking reagent (Roche)] for 1 hour at

RT, streptavidin conjugated with Alexa 488 (Invitrogen,

San Diego, CA, USA) was used for the probe detection.

Tissues were washed in 4× SSC for 15 min and stained

with DAPI (1 μg/μL) (Sigma, St. Louis, MO, USA) in 1×

PBS at RT overnight.

Tissue fragments were mounted in a drop of

Vectashield antifade solution and examined using a

Leica TCS SP5 confocal laser scanning microscope based

on the inverted microscope Leica DMI 6000 CS (Leica

Microsystems, Wetzlar, Germany). Diode and argon lasers

were used to excite the fluorescent dyes DAPI and fluo-

rochromes Alexa488 with the usage of LAS AF software

(Leica Microsystems, Wetzlar, Germany).

RESULTS

3.1

Interspecific hybridisation

Bayesian analysis based on microsatellite genotypes and

implemented in structure identified eight individuals

out of 306 analysed with admixed nuclear genomes

(TableS1). These putative hybrids possessed mtDNA

of either P. epeiroticus or P. kurtmuelleri. Additionally,

two frogs (M3300 from Ioannina and J8066 from

Kalogria) were assigned to the P. epeiroticus cluster in

microsatellites but possessed P. kurtmuelleri- specific

mtDNA (Table3, Figure1a,b). Considering only eight

individuals with admixed nuclear genomes, the overall

hybridisation rate in studied populations was 2.6%.

(Figure1b). Hybrids were recorded in Igoumenitsa

(9.8%), Ioannina (6.8%) and Kalogria (5.2%). In three

localities, only one of the parental species was recorded

(P. epeiroticus in Fragma Kalama and Panetolio,

P. kurtmuelleri in Vrisera) (Figure1c).

NewHybrids identified individuals with admixed nu-

clear genomes as F2 hybrids (Table3). In one case (J8061),

the posterior probabilities were too low, and it was not

possible to reliably determine whether the individual be-

longed to the category of the parental species P. epeiroticus

or F2 hybrids. All other frogs were assigned to the category

of one or the other parental species.

Ordination analyses (PCoA, PCA, DAPC) of micro-

satellite genotypes (Figure1d, FigureS1) congruently

revealed two well- separated clusters corresponding to pa-

rental species P. epeiroticus and P. kurtmuelleri. Individuals

with admixed genomes inferred in Structure showed pre-

dominantly an intermediate position in multidimensional

space or showed affinity to one of the parental species

clusters (Table3, Figure1d, FigureS1).

Four hybrid tadpoles between P. epeiroticus and

P. kurtmuelleri analysed for gonadal microanatomy had

a normal distribution of germ cells throughout the go-

nads (TableS3, FigureS2A,B), consistent with previ-

ously published data (Dedukh et al., 2020). A total of

962 germ cells were analysed from the gonads of the

studied tadpoles, but no micronuclei were observed. By

the analysis of the RrS1 signal distribution, 26 signals

corresponding to the diploid chromosomal sets were

detected in individual germ cells. The analysis of mi-

totic divisions of germ cells did not reveal chromosomal

TABLE  Putative hybrids between Pelophylax epeiroticus and P. kurtmuelleri.

ID Locality mtDNA type q kurtmuelleri q epeiroticus Genotype classes

M3250 Igoumenitsa epeiroticus 0.400 0.600 F2 (1.000)

M3273 Igoumenitsa epeiroticus 0.523 0.477 F2 (1.000)

M3275 Igoumenitsa epeiroticus 0.396 0.604 F2 (1.000)

J8061 Kalogria epeiroticus 0.178 0.822 F2 (0.553)

B- epeiroticus (0.445)

M3278 Igoumenitsa kurtmuelleri 0.715 0.285 F2 (0.984)

M3303 Ioannina kurtmuelleri 0.570 0.430 F2 (1.000)

M3309 Ioannina kurtmuelleri 0.610 0.390 F2 (1.000)

M3610 Ioannina kurtmuelleri 0.575 0.425 F2 (0.994)

M3300 Ioannina kurtmuelleri 0.011 0.989 epeiroticus (0.820)

F2 (0.176)

J8066 Kalogria kurtmuelleri 0.002 0.998 epeiroticus (1.000)

Note: The parameter q indicates the proportion of an individual's genome originating in one of the two inferred clusters in the program Structure,

corresponding to parental species P. epeiroticus and P. kurtmuelleri. Genotype classes, inferred from the program NewHybrids, denote posterior probabilities

(in parentheses) that an individual is assigned to the parental species (epeiroticus, kurtmuelleri) or F1, F2 and backcross hybrid categories (B- epeiroticus,

B- kurtmuelleri).

PAPEŽÍK etal.

misalignment or lagging during metaphase or anaphase,

respectively.

3.2

Genetic structure inferred

from mtDNA

A total of 177 mtDNA sequences of P. epeiroticus formed

two well- differentiated mitochondrial lineages with

clear geographical distribution (Figure2a,b, FigureS3).

These lineages were named, according to Papežík

et al. (2023), as Northern and Southern. Within the

Southern lineage, a Central haplogroup differentiated

by four substitutions from other haplotypes, was distin-

guished. In total, 40 unique haplotypes with hd = 0.88

and π = 1.09% were recognised (Table4). The Northern

lineage is widespread in southern Albania and north-

western Greece (Figure2c) and consists of 22 haplo-

types (hd = 0.79 and π = 0.21%; Table4). The Southern

lineage (including Central haplogroup), represented

by 18 haplotypes (hd = 0.72 and π = 0.25%; Table5), is

widespread from Ioannina, the Ambracian Gulf (south-

western Greece), along the northwestern coast of the

Peloponnese up to the Kaiafa Lake (Figure4c). The

Central haplogroup, with only four recorded haplotypes

(hd = 0.87 and π = 0.17%; Table4), is distributed around

FIGURE  Mitochondrial diversity and structure of Pelophylax epeiroticus displayed in the haplotype network (a), principal component

analysis with embedded eigenvalues (PCA, b) and visualisation of mtDNA lineages on the map (c). The size of the circles in the network

corresponds to the frequency of haplotypes. The colour scheme indicates the distribution of haplotypes across sampled localities. Small

black circles are missing node haplotypes, each line connecting two haplotypes corresponds to one mutation step unless otherwise indicated

by a number along the line. In PCA, points were jittered to show the frequency of identical sequences in each haplotype. See Table1 for site

abbreviations.

PAPEŽÍK etal.

the Ambracian Gulf and in Ioannina. The uncorrected

p- distance between the Northern and Southern line-

age was 2.08% and between the Southern lineage and

Central haplogroup 0.68% (FigureS4A).

In P. kurtmuelleri, 126 mtDNA sequences formed

three weakly differentiated lineages without a clear

geographic pattern (Figure3a–c). They were named,

in concordance with Papežík et al. (2023), as Red,

Blue and Yellow lineage. The Blue lineage was

found in southern Albania, Ioannina and the north-

western coast of the Peloponnese (Figure3c) and is

represented by nine haplotypes with hd = 0.82 and

π = 0.16% (Table4). The Red lineage, distributed from

southern Albania to the Ambracian Gulf and the

Kaiafa Lake in the western Peloponnese (Figure3c),

is represented by 15 haplotypes with hd = 0.58 and

π = 0.12% (Table4). The Yellow lineage was recorded

in Ioannina, Vrisera, Zirou Lake and Corfu (Figure3c)

and consisted of three haplotypes with hd = 0.52 and

π = 0.09% (Table4). The uncorrected p- distances be-

tween the lineages reached the values 0.65%–0.77%

(FigureS4B).

TABLE  Results of niche similarity of Pelophylax epeiroticus and P. kurtmuelleri and other tests conducted in the Humboldt and

ENMTools packages.

Tests implemented

P. kurtmuelleri (Background

P. kurtmuelleri > P. epeiroticus)

P. epeiroticus (Background

P. epeiroticus > P. kurtmuelleri)

p- values p- values

Niche overlap test (NOT)a.010 .010

Niche divergence test (NDT)a.287 .010

Identity (Equivalency)a0.410 (in NOT)

0.480 (in NDT)

Background (Niche similarity)aNiche Similarity: 0.348

Similarity—analogous environments only: 0.539 (in NOT)

Niche Similarity: 0.344

Similarity—analogous environments only: 0.366 (in NDT)

Analogous climate space percentagea45.15% (in NOT)

65.96% (in NDT)

p- values of Identity, Asymmetric Background and Ecospat tests

D I rank.cor env.Denv.I

Identityb.038 .038 .120 .040 .040

Asymmetric backgroundb.038 .038 .040 .040 .040

Ecospat (identity test p- values)b.030 .010 Expansion:

1.000

Stability:

.010

Unfilling:

1.000

Note: For the Humboldt package tests, a value of 0 indicates completely different niches and a value of 1 indicates identical niches. Statistically significant

values are in bold.

aHumboldt package.

bENMTools package.

TABLE  Summary of mtDNA polymorphism for Pelophylax epeiroticus and P. kurtmuelleri.

Species/mtDNA lineage/haplogroup N S H hd ± SD π ± SD (%) θ ± SD (%)

P. epeiroticus 177 60 40 0.88 ± 0.01 1.09 ± 0.04 1.04 ± 0.26

Northern lineage 112 26 22 0.79 ± 0.03 0.21 ± 0.02 0.49 ± 0.15

Southern lineage 65 24 18 0.72 ± 0.06 0.25 ± 0.04 0.50 ± 0.16

Central haplogroup 6 4 4 0.87 ± 0.13 0.17 ± 0.04 0.17 ± 0.11

P. kurtmuelleri 126 31 27 0.77 ± 0.04 0.36 ± 0.03 0.57 ± 0.16

Blue haplogroup 27 9 9 0.82 ± 0.05 0.16 ± 0.02 0.23 ± 0.10

Red haplogroup 92 13 15 0.58 ± 0.06 0.12 ± 0.02 0.25 ± 0.09

Yellow haplogroup 7 3 3 0.52 ± 0.21 0.09 ± 0.04 0.12 ± 0.08

Abbreviations: H, number of haplotypes; hd, haplotype diversity; N, number of individuals; S, segregation sites; SD, standard deviation; θ, Watterson theta; π,

nucleotide diversity.

PAPEŽÍK etal.

3.3

Genetic structure inferred from

microsatellites

Pelophylax epeiroticus showed lower genetic diversity val-

ues (NA = 2.85 ± 0.18, HO = 0.32 ± 0.03, HE = 0.38 ± 0.03) at

microsatellite loci than P. kurtmuelleri (NA = 5.31 ± 0.27,

HO = 0.59 ± 0.03, HE = 0.68 ± 0.02; TableS4). This pat-

tern remained consistent when analysing only the ten

loci successfully amplified in both species (P. epeiroticus:

NA = 2.75 ± 0.17, HO = 0.37 ± 0.03, HE = 0.38 ± 0.03; P. kurtm-

uelleri: NA = 4.17 ± 0.27, HO = 0.51 ± 0.04, HE = 0.58 ± 0.03).

In P. epeiroticus, the ΔK and the thermodynamic inte-

gration (TI) statistics identified congruently two genetic

clusters with significant geographic structure, matching

the pattern inferred from mtDNA (Figure4a–c). The

first cluster (northern) was composed of populations in

southern Albania and northwestern Greece (Igoumenitsa,

Ioannina and Fragma Kalama), including the island

of Corfu (Figure4c). The second cluster (southern)

was represented by localities in the surrounding of the

Ambracian Gulf (Louros and Zirou Lake), in southern

continental Greece (Panetolio) and the Peloponnese

(Kalogria, Limanaki, Sybanion). The localities Louros,

Kalogria and Panetolio showed partial admixture be-

tween both clusters.

Ordination analyses (PCA and DAPC) indicated the

similarity of Louros and Zirou Lake to the northern clus-

ter, despite being weakly separated from the other lo-

calities, rather suggesting the presence of three clusters

(Figure4d, FigureS5A). In PCoA (FigureS5C), however,

Louros and Zirou Lake showed affinity to the southern

populations.

FIGURE  Mitochondrial diversity and structure of Pelophylax kurtmuelleri displayed in the haplotype network (a), principal

component analysis with embedded eigenvalues (PCA, b) and visualisation of mtDNA haplogroups on the map (c). The size of the circles

in the network corresponds to the frequency of haplotypes. The colour scheme indicates the distribution of haplotypes across sampled

localities. Small black circles are missing node haplotypes, each line connecting two haplotypes corresponds to one mutation step unless

otherwise indicated by a number along the line. In PCA, points were jittered to show the frequency of identical sequences in each haplotype.

See Table1 for site abbreviations.

PAPEŽÍK etal.

Significant structuring of genetic variability in

P. epeiroticus was also revealed by AMOVA (p = 0.001).

Overall, 28% of genetic variation was partitioned among

the Structure inferred clusters, 17% among populations,

18% among individuals within populations and 38%

within individuals.

The average value of FST statistics over all P. epeiroti-

cus populations was 0.313, minimal 0.060 (between

Igoumenitsa and Xare) and maximal 0.601 (between

Ioannina and Limanaki) (TableS5). Average values of ge-

netic distances were 0.628 for Jost's D (range 0.336–2.201),

0.330 for Nei's GST (range 0.034–1.012) and 1.895 for G'ST

(range 0.105–8.680) (Figure6a, FigureS6A,B). The lowest

genetic distances were observed between Ioannina and

Igoumenitsa, and the highest between Ioannina and

Limanaki (Figure6a, FigureS6A,B). FST statistics and

genetic distances were highly correlated with geographic

distances showing the IBD pattern (Mantel test; FST,

p = 0.0005; Jost's D, p = 0.0014; Nei's GST, p = 0.0005; G'ST,

p = 0.0017) (Figure6a, FigureS7A,C,E).

In P. kurtmuelleri, the most likely number of K in

Bayesian clustering was two based on the ΔK and TI sta-

tistics. However, the geographic pattern in the distribution

of clusters was less obvious and did not follow a north–

south orientation compared to P. epeiroticus (Figure5b,c).

The first cluster consisted of populations distributed from

FIGURE  Population structure of Pelophylax epeiroticus in the southwestern Balkans. Comparison of mtDNA (a) and microsatellite (b)

admixture proportion of each sampled individual. The distribution of average admixture per locality is shown in the map (c). Discriminant

analysis of principal components (d) with embedded number of retained PCs and DA eigenvalues indicating the position of each sampled

population in multivariate space. Localities labels are positioned in the group centroids. The colour palette used in sampled localities is

identical to Figures2 and 3. See Table1 for site abbreviations.

PAPEŽÍK etal.

northwestern Greece (Ioannina) across the Ambracian

Gulf (Zirou Lake) to the Peloponnese (Limanaki and

Sybanion, Figure5b,c). The second cluster included all

other populations distributed from southern Albania to

the Peloponnese. When Structure was performed, using

only loci that were simultaneously amplified in both spe-

cies, the most likely number of K was three. The overall

population structure was consistent with the results ob-

tained from the analysis of all loci, but the population in

the Ropa Valley in Corfu and the five individuals from

Stjar were assigned to a separate, third, cluster.

Ordination analyses provided similar outcomes as

Bayesian clustering and did not separate any specific

geographic regions. The only relatively separated popula-

tions in DAPC were the Ropa Valley in Corfu and Zirou

Lake (Figure5d). In contrast, both PCoA and PCA showed

sampled localities admixed, without any clear pattern

(FigureS5B, D). Similar to the structure results, ordina-

tion analyses performed with a limited number of loci

simultaneously amplified in both species resulted in the

same population pattern as the full dataset analysis.

AMOVA showed statistically significant population

structure (p = 0.001) but, contrarily to P. epeiroticus, 5%

of genetic variation was partitioned among the Bayesian

inferred clusters, 4% among populations, 20% among indi-

viduals and 70% within individuals.

FIGURE  Population structure of Pelophylax kurtmuelleri in the southwestern Balkans. Comparison of mtDNA (a) and microsatellite

(b) admixture proportion of each sampled individual. The distribution of average admixture per locality is shown in the map (c).

Discriminant analysis of principal components (d) with embedded number of retained PCs and DA eigenvalues indicating the position of

each sampled population in multivariate space. Localities labels are positioned in the group centroids. The colour palette used in sampled

localities is identical to Figures2 and 3. See Table1 for site abbreviations.

PAPEŽÍK etal.

FST statistics and pairwise genetic distances between

P. kurtmuelleri populations were substantially lower than

between populations of P. epeiroticus (TableS6). FST sta-

tistics reached the average value of 0.085 (minimal 0.001

between Sybanion and Kalogria, maximal 0.192 between

Louros and the Ropa Valley). Average values of genetic

distances were 0.483 for Jost's D (range 0.050–1.003), 0.098

for Nei's GST (range 0.013–0.243) and 0.795 for G'ST (range

0.079–1.975) (Figure6b, FigureS6C,D). The lowest genetic

distances were observed between Sybanion and Limanaki,

and the highest between the Ropa Valley (Corfu) and Syri i

Kaltër (Figure6b, FigureS6C,D). Despite the higher level

of genetic differentiation in more distant populations,

Mantel tests were not statistically significant (Figure6b,

FigureS7B,D,F).

3.4

Niche comparison

In total, six out of the nine most effective variables in the

current distribution of the two species were linked to pre-

cipitation, encompassing BIO12 (Annual Precipitation),

BIO16 (Precipitation of Wettest Quarter), BIO17

(Precipitation of Driest Quarter), PETDriestQuarter

(mean monthly Potential Evapotranspiration of cold-

est quarter), PETWarmestQuarter (mean monthly

Potential Evapotranspiration of warmest quarter)

and PETWettestQuarter (mean monthly Potential

Evapotranspiration of wettest quarter). The remain-

ing three variables were temperature- related, including

BIO3 (Isothermality), calculated as BIO2 (Mean Diurnal

Range)/BIO7 (Temperature Annual Range) × 100, BIO7,

computed as BIO5 (Max Temperature of Warmest

Month)—BIO6 (Min Temperature of Coldest Month), and

BIO9 (Mean Temperature of Driest Quarter).

For P. kurtmuelleri, the most influential variables

were BIO12 with permutation importance of 22.6, fol-

lowed by BIO7 with 15.9, BIO17 with 14.2, and BIO9

with 13.1. Conversely, for P. epeiroticus, the key contrib-

uting variables were BIO3 with permutation importance

of 62.2, followed by BIO16 with 11.2, BIO9 with 8, and

PETWarmestQuarter with 5.2.

The results obtained from the Humboldt package indi-

cated that the current niches of the species are not equiva-

lent. The implemented NOT test provided both significant

equivalence (0.010) and background statistics (0.010),

while the NDT test provided significant equivalency sta-

tistics (0.010) and non- significant background statistics

(0.287). Additionally, the PCA biplot (Figure7a) with ker-

nel densities showed significant overlap between analo-

gous environments of both species in multivariate space

and along both principal components, indicating rather

similarity of species niches. Next, the significance of both

D and I values in the NOT test (p = 0.010) indicates that

the occupied niches of each species were not more similar

than they would be expected by chance (p < 0.05).

The ENMTools estimation of niche identity returned

moderate to high values of D (0.626) and I (0.898), pre-

dicting that the species' niches are similar. However, when

considering similarity measurements that consider the n-

dimensional spaces of all combinations of environmental

variables used rather than just the ENMs, the values of

D (0.171) and I (0.359) reached low to moderate values,

indicating dissimilarity in their estimated responses to en-

vironments. The raster breadth of the MaxEnt model for

measuring niche the species similarities for P. epeiroticus

FIGURE  Pairwise Hedricks G'st genetic distances between populations of Pelophylax epeiroticus (a) and P. kurtmuelleri (b). Numbers

indicate genetic distances, and the colouring of each cell indicates a scale of Euclidean geographic distances. Insets show the correlation

between Hedricks G'st and Euclidean distance (Mantel test). r is the value of the correlation coefficient with the corresponding p- value. See

Table1 for site abbreviations.

PAPEŽÍK etal.

FIGURE  Density plots of the analogous environmental space (E- space) occupied by Pelophylax epeiroticus and P. kurtmuelleri

estimated in the Humboldt package (a). Niches for both species are shown in a single plot, where higher densities are shown with

darker colours and lines representing the kernel density isopleths from 1% to 100% (periphery to centre). On the top and right side,

histogram density plots for each of the two first PCs are shown; each E- space is displayed in the respective colour (red and green)

and the niche overlaps with the combination of the two colours. The hypsometric distribution of both species was compared using

the Wilcoxon rank sum test, **** = 3.1e−7 (b). Localities used in niche overlap for P. epeiroticus (c) and P. kurtmuelleri (d) plotted as

partially transparent points against merged topographic maps of Albania and Greece, indicating differences in the hypsometric species

distribution.

PAPEŽÍK etal.

was B1: 0.937 and B2: 0.398, and for P. kurtmuelleri B1:

0.987 and B2: 0.798.

Niche similarity tests yielded mixed results using tools

from the two R packages. The NOT test, focusing on anal-

ogous environments, showed moderate similarity val-

ues (D = 0.348 and I = 0.539). Similarly, the Ecospat test

produced close overlap values (D = 0.364 and I = 0.562),

suggesting that the species partially share their climatic

preferences. Details of the niche comparison tests are

available in Table5 and FiguresS8–S10.

Pelophylax epeiroticus sites lay at elevations rang-

ing from 0 to 481 m a.s.l., with a mean elevation of

78.22 m a.s.l.; P. kurtmuelleri sites lay at elevations ranging

from 0 to 1401 m a.s.l., with a mean elevation of 296 m a.s.l.

Differences in mean elevation between the species were

highly significant (Wilcoxon rank sum test, W = 2133,

p = 3.1e−7) (Figure7b).

DISCUSSION

4.1

Hybridisation pattern

Interspecific hybridisation is an important evolutionary

phenomenon, which may result in a new hybrid taxon with

sexual or asexual reproduction, introgression of genetic

material in sympatric populations or the establishment of

hybrid zones in regions where ranges of the hybridising

species meet (Allendorf etal.,2001; Rosser etal., 2024).

The extent of hybridisation depends primarily on the ge-

netic divergence between species and reproductive isola-

tion barriers that prevent hybridisation and introgression

and maintain species integrity (Abbott etal.,2013). In the

studied system of P. epeiroticus and P. kurtmuelleri, species

hybridise in sympatric and syntopic sites, but the rate of

interspecific mating seems to be relatively low, ranging

from 0% to 9.8% in specific populations.

The comparison of the hybridisation rate of P. epeiroti-

cus and P. kurtmuelleri between our and other studies is

challenging due to different authors using different mark-

ers for species and hybrid identification (allozymes, mi-

crosatellites, bioacoustic and morphological markers),

different statistical analyses for hybrid detection and

different sampling design (e.g., the number of frogs ana-

lysed). Moreover, we can assume that the rate of hybridi-

sation might depend on the specific ecological conditions

in the locality and the ratio of the parental species indi-

viduals. Hotz and Uzzell(1982) detected 6.5% of hybrids

in Igoumenitsa using allozyme markers, a rate of hybridi-

sation comparable to that found in our study at the same

(Igoumenitsa 9.8%) and two other sites (Ioannina 6.8%

and Kalogria 5.2%). However, no hybrids were detected in

other syntopic localities investigated in our study, probably

due to the limited number of samples analysed or the rela-

tive rarity of one of the parental species, and the resulting

low rate of interspecific mating. A higher percentage (14%)

of P. epeiroticus × P. kurtmuelleri hybrids was identified by

Schneider etal.(1984) in Ioannina using a combination

of morphometric and bioacoustic characters. However,

these methods may be less reliable than genetic markers

for detecting hybrids, potentially leading to an inaccurate

estimation of the number of hybrid individuals. Sagonas

etal.(2020) identified 4% of hybrids with a high posterior

probability using microsatellites and NewHybrids analy-

sis, but it is important to note that approximately 10% of

individuals were not accurately assigned to any genotypic

category (i.e., it was not possible to determine with cer-

tainty whether they were hybrids).

Not only the rate of interspecific mating but also the

types of hybrids are important for the evolutionary inter-

pretation of hybridisation, specifically for estimating the

sexual versus asexual mode of hybrid reproduction and

the survival of different hybrid genotypes. In our study,

using Bayesian analysis implemented in NewHybrids, we

detected only F2 hybrids (in one case, the assignment of

the individual was not clear, and it fell into either the F2

or a backcross hybrid category). Hotz and Uzzell(1982)

identified a low frequency of interspecific hybrids with

F1 (7% of all individuals) and backcross (1.3%) electro-

phoretic phenotypes. Sagonas et al. (2020) unambigu-

ously confirmed only the presence of F2 hybrids. These

results suggest that a hybrid reproduction mode is sexual

and not asexual (hybridogenetic). If hybrids reproduced

hybridogenetically, we would expect only F1 hybrids,

but not F2 or backcross genotypes. This is because hy-

bridogenetic hybrids transmit clonally the genome of one

parental species and backcross with the other parental

species, thus restoring F1 genotypes (Avise,2008; Tunner

& Heppich,1981).

The sexual mode of reproduction of the hybrids was

also confirmed by cytogenetic analysis of four hybrid

larvae produced by the experimental crossing of the pa-

rental species in the laboratory. Their gonads showed no

evidence of micronuclei, which contain genetic material

from genome elimination and are specific for hybridoge-

netic reproduction (Chmielewska et al., 2018; Choleva

et al., 2023; Dedukh et al., 2017, 2020; Ogielska,1994).

Induction of hybridogenesis is primarily associated with

P. ridibundus, which forms hybridogenetic hybrids with

P. lessonae, P. bergeri and P. perezi (Plötner,2005). However,

it seems that the P. epeiroticus genome is resistant to elim-

ination in P. ridibundus × P. epeiroticus hybrids (Guerrini

etal.,1997) and that the P. kurtmuelleri genome does not

induce hybridogenesis (Hotz etal.,1985).

Potential hybridogenesis of P. kurtmuelleri×P. epeiroti-

cus hybrids could be indicated by the existence of triploids,

PAPEŽÍK etal.

which have been sporadically recorded in populations with

a relatively high abundance of F1 hybrids (Kyriakopoulou-

Sklavounou & Vasara, 2001; Sofianidou, 1996). The

increase in ploidy is generally considered to be a conse-

quence of clonality (Choleva et al., 2012). However, the

existence of later- generation hybrids (F2 and backcrosses)

in natural populations evidenced in our and previous stud-

ies, as well as results of crossing experiments (Guerrini

etal.,1997; this study), are congruent with sexual rather

than hybridogenetic reproduction of hybrids. The exis-

tence of later generations of hybrids also points to the fact

that F1 hybrids of P. epeiroticus and P. kurtmuelleri, species

that diverged ca 8.3 Mya (Dufresnes etal.,2024), are not

sterile and are capable of reproduction, albeit limited.

4.2

Reproductive isolation barriers

Reproductive isolation barriers prevent hybridisation and

the production of fertile offspring (Coyne & Orr, 2004).

Uncovering the specific mechanisms of reproductive iso-

lation is difficult and requires an experimental approach,

but the analysis of hybridisation patterns allows us to indi-

cate which barriers maintain species integrity and prevent

hybridisation. In unimodal hybrid zones, intermediate

hybrid genotypes predominate; in bimodal zones, hybrids

are rare and parental genotypes are more frequent (Jiggins

& Mallet,2000). Bimodal hybrid zones are invariably cou-

pled with strong prezygotic barriers like assortative mat-

ing or fertilisation, habitat and temporal isolation, while

postzygotic incompatibilities, reducing the survival or re-

production of hybrid offspring, are similar in bimodal and

unimodal zones. Syntopic populations of P. epeiroticus and

P. kurtmuelleri show a bimodal distribution of genotypes.

Based on these findings, we can assume that hybridisation

of species is primarily prevented by prezygotic barriers.

Nevertheless, we cannot exclude that postzygotic incom-

patibilities resulting from negative epistatic interactions

between alleles from two different parental genomes may

also be applied. However, these have not yet been investi-

gated in P. epeiroticus x P. kurtmuelleri hybrids.

In previous studies, several prezygotic barriers have

been suggested (Plötner etal.,2010). Mating calls play a

key role in frog reproduction and are often species- specific

(Schneider, 2005), suggesting that differences in mating

calls between species could potentially contribute to re-

productive isolation. Because males of P. epeiroticus and

P. kurtmuelleri differ markedly in mating calls (Schneider

etal.,1984), females may preferentially choose mates of

their own species for reproduction, as previously suggested

by Hotz and Uzzell (1982). In addition to bioacoustics,

field studies suggest partial spatial and temporal isolation

between species. During the breeding season, P. epeiroticus

and P. kurtmuelleri tend to form separate groups of indi-

viduals with minimal mixing at the edges (Sofianidou &

Schneider, 1989). Male aggregations (choruses) contain

only single individuals of the foreign species and the call-

ing of P. epeiroticus males does not stimulate the calling

activity of P. kurtmuelleri males (Kordges,1988; Plötner

et al., 2010). In addition, there is a temporal difference

in breeding seasons, with P. epeiroticus usually beginning

its breeding activities several weeks before P. kurtmuel-

leri (Sofianidou & Schneider, 1989). Furthermore, the

two species show differences in habitat preferences, with

P. epeiroticus preferring shallow waters and irrigation ca-

nals, whereas P. kurtmuelleri is commonly found in larger

water bodies (Hotz & Uzzell,1982).

4.3

Niche comparison

The niche overlap and divergence tests provided insights

into the ecological dynamics between P. epeiroticus and

P. kurtmuelleri, potentially revealing other prezygotic

reproductive isolation barriers. However, our analyses

indicated slightly ambiguous results. While most analy-

ses indicated significant niche overlap between the two

species, the NOT and NDT tests (implemented in the

Humboldt package) revealed that the species occupy

distinct niches within their current distributions. The

results of the NOT and NDT tests also contradict the

hypothesis that evolutionary divergence alone explains

the niche differences (for details see Table2 in Brown

& Carnaval, 2019). Instead, asymmetrical habitat acces-

sibility, which influences each species' level of access to

environmental resources, emerged as a significant fac-

tor in niche differentiation between both species. These

Humboldt- based analyses might provide more insightful

results compared to the results of the ENMTools pack-

age (see the comparison in Brown & Carnaval, 2019).

However, other tests (Schoeners's D and I, Ecospat test)

within ENMTools imply a higher degree of niche similar-

ity based on environmental distributions. This suggests

that the observed overlap between the species' niches is

significantly greater than what would be expected by

chance within their current geographic ranges.

Our analyses also highlighted that, although the

ecological niches of the two species are neither identi-

cal nor fully divergent, a substantial overlap does exist.

Furthermore, P. kurtmuelleri appears to have a consid-

erably broader niche than P. epeiroticus. This difference

may be driven by P. kurtmuelleri's tolerance of a wider

elevation range, resulting in a more extensive ecological

niche compared to P. epeiroticus, a species primarily found

in lowland areas (Schneider & Haxhiu,1992; Sofianidou

etal.,1987; Sofianidou & Schneider,1989).

PAPEŽÍK etal.

4.4

Population genetic structure

Investigating the relationships between mitochondrial

and microsatellite genetic diversity provides valuable

insights into the evolutionary dynamics and population

structure of species. While mtDNA captures maternal lin-

eages and reflects historical demographic events, micros-

atellites, as highly variable nuclear markers, offer insights

into current population dynamics influenced by gene flow

and genetic drift.

In P. epeiroticus, we confirmed the existence of two

major and deeply diverged mtDNA lineages distributed

along the north–south direction (see Papežík etal., 2023

for details). The pattern in mtDNA aligns with the results

from microsatellites, where both Bayesian and ordina-

tion methods indicate the distinctiveness of the two main

clusters distributed in the northern and southern parts of

the species range. In P. kurtmuelleri, three mitochondrial

lineages (described in Papežík etal.,2023) with shallow

divergence were identified. The Red and Blue lineages are

restricted to the southwestern Balkans, and the Yellow lin-

eage occurs in the whole P. kurtmuelleri range. In the part

of the P. kurtmuelleri range surveyed in this study, no clear

geographic pattern in the distribution of mtDNA lineages

was detected. Microsatellite data reflected the mtDNA re-

sults and did not reveal any significant geographic pattern

as was observed in P. epeiroticus.

While P. epeiroticus and P. kurtmuelleri do not differ

significantly in haplotype diversity, P. epeiroticus exhib-

its significantly higher nucleotide diversity reflecting the

existence of two diverged lineages, the Northern and the

Southern (Papežík etal.,2023; this study). In microsatel-

lite markers, species differ significantly both in genetic

variability and genetic differentiation of populations.

Pelophylax epeiroticus showed about half lower values of

allelic diversity and heterozygosity and significantly higher

values of FST and genetic distances, which were also re-

flected in a significant IBD pattern. Significant structuring

of genetic variability in P. epeiroticus was also revealed by

AMOVA, where 28% of genetic variation was partitioned

among the Bayesian inferred clusters and 17% among pop-

ulations, while in P. kurtmuelleri it was only 5% and 4%,

respectively. Lower genetic diversity and higher genetic

differentiation indicate that populations of P. epeiroticus

are more influenced by genetic drift and the gene flow

among them is limited compared to P. kurtmuelleri. We

can hypothesise that the strong population structure of

P. epeiroticus may be related to its limited dispersal ability

in the mountainous landscape. The species is restricted

to lowlands (see Figure7b,c) and reaches elevations

above 500 m a.s.l. only in the Ioannina region (Schneider

& Haxhiu, 1992; Sofianidou et al., 1987; Sofianidou &

Schneider,1989). A similar altitudinal distribution pattern

is observed in other Balkan endemics. For example, the

water frog species Pelophylax shqipericus is restricted

to the Lake Skadar basin and coastal plains in Albania

(Haxhiu,1994; Plötner,2005; Schneider & Haxhiu,1992),

where it comes into contact with P. kurtmuelleri. Similarly,

the wall lizard Podarcis ionicus inhabits low to moderate

elevations west of the Pindus Mountains, whose range

partially overlaps with widely distributed Podarcis tauri-

cus, commonly found from sea level up to about 1500 m

(Psonis et al., 2017). Contrarily to P. epeiroticus, P. kurt-

muelleri is widespread from lowlands up to 2000 m a.s.l.

(Figure7b,d), as documented by Szabolcs et al. (2017)

or Valakos et al. (2008). We can assume that the ability

of P. kurtmuelleri to inhabit different altitudes also influ-

ences its dispersal rate and consequently its population

genetic structure, which is manifested by higher genetic

diversity, lower genetic differentiation and more intense

gene flow. Hybridisation between the two species appears

to occur across the whole altitudinal gradient inhabited

by P. epeiroticus, as the localities where we detected hy-

brids lie from sea level (Igoumenitsa, 2 m a.s.l.; Kalogria,

3 m a.s.l.) to mid- elevations (Ioannina, 472 m a.s.l.), which

are close to the upper limit of the hypsometric distribu-

tion of this species. Future investigations focusing on

landscape genetics may yield deeper insights into the dis-

tribution patterns of both water frog species as well as the

role of the Pindus Mountain range in the divergence of

P. kurtmuelleri populations.

The different population genetic structure of P. epeiroti-

cus and P. kurtmuelleri may have important conservation

and evolutionary implications. As P. epeiroticus appears to

be more affected by genetic drift, the loss of genetic vari-

ation in its populations will be more pronounced than in

P. kurtmuelleri. This assumption, together with its limited

geographic range, makes P. epeiroticus a more vulnerable

species.

The markedly different population genetic structure

of the two species may both influence and be influenced

by interspecific hybridisation. If one species has a strong

population structure with limited gene flow (in our case

P. epeiroticus) and the other species has a weak popula-

tion structure with intense gene flow (like P. kurtmuel-

leri), it can be assumed that interspecific hybridisation

(introgression) will be more likely to affect the species

with the strong population structure. First, a species with

intense gene flow is more likely to introduce its genes

into populations of the other species, i.e., P. kurtmuelleri

is more likely to introduce its genes into populations of

P. epeiroticus than vice versa. Second, since genetic drift

leads to allele fixation, we can assume that introgressed

alleles are more likely to be fixed in the species with

stronger genetic drift, i.e., in P. epeiroticus. Thus, the pop-

ulation genetic structure of studied water frogs may be

PAPEŽÍK etal.

the result of the interaction of genetic drift, gene flow

and introgression. If the level of hybridisation were high,

we would expect interspecific gene flow to significantly

disrupt the population structure of both species in their

sympatric range. However, given the low level of hybri-

disation between P. epeiroticus and P. kurtmuelleri, we ex-

pect that introgression will not have such a strong effect

on their population structure.

AUTHOR CONTRIBUTIONS

Petr Papežík and Peter Mikulíček contributed to the

study conception and design. Michal Benovics, Adam

Javorčík, Petros Lymberakis, Simona Papežíková, Nikos

Poulakakis, Radek Šanda and Jasna Vukić participated

in the collection of biological material. Genetic data col-

lection was carried out by Petr Papežík, Peter Mikulíček

and Sandra Lálová. Population genetic analyses were con-

ducted by Petr Papežík and Peter Mikulíček. Ecological

niche modelling was undertaken by Ídris Sari; crossing

experiments and cytogenetic analyses by Dmitrij Dedukh,

Marie Doležálková- Kaštánková and Lukáš Choleva. The

first draft of the manuscript was written by Petr Papežík

and Peter Mikulíček, and all authors commented on pre-

vious versions of the manuscript. All authors read and ap-

proved the final manuscript.

ACKNOWLEDGEMENTS

We thank Daniel Jablonski for providing samples from

southern Albania and for his valuable comments on the

manuscript. We are also grateful to Eva Aschenbrener,

Nuria Viñuela Rodríguez and Stamatis Zogaris for their

assistance in the field and laboratory. Additionally, we

appreciate the constructive feedback from the two anon-

ymous reviewers and the editor, which significantly

improved the quality of the manuscript. Finally, we ex-

tend our thanks to Martina and Mike Lawson for their

help with language editing. This study was supported

by the Scientific Grant Agency of the Slovak Republic

VEGA (project 1/0014/24). RŠ was supported by the

Ministry of Culture of the Czech Republic (DKRVO

2024–2028/6.I.b, National Museum, 00023272), JV by in-

stitutional resources of the Ministry of Education, Youth

and Sports of the Czech Republic. DD, MDK and LC

were supported by the Czech Science Foundation grant

(23- 07028K) and RVO (67985904). The research permits

were provided by the Directorate of Forest Management,

Ministry for Environment and Energy of the Hellenic

Republic (154073/823/7- 4- 2017, 173857/1638/18- 9-

2018, 181012/807/28- 3- 2019, 183226/1246/11- 6- 2019,

123199/3356/22- 12- 2020 and 123199/3356/01- 02- 2021),

and the National Agency of Protected Areas, Ministry of

Tourism and Environment of the Albanian Republic (No.

480/2019).

DATA AVAILABILITY STATEMENT

The newly generated sequences were deposited in the

NCBI GenBank database under accession numbers

PQ570882–PQ570898.

ORCID

Nikos Poulakakis https://orcid.

org/0000-0002-9982-7416

Peter Mikulíček https://orcid.org/0000-0002-4927-493X

REFERENCES

Abbott, R., Albach, D., Ansell, S., Arntzen, J. W., Baird, S. J. E.,

Bierne, N., & Zinner, D. (2013). Hybridization and speciation.

Journal of Evolutionary Biology, 26, 229–246. https:// doi. org/ 10.

1111/j. 1420- 9101. 2012. 02599. x

Adamack, A. T., & Gruber, B. (2014). PopGenReport: Simplifying

basic population genetic analyses in R. Methods in Ecology

and Evolution, 5, 384–387. https:// doi. org/ 10. 1111/ 2041- 210X.

12158

Aiello- Lammens, M. E., Boria, R. A., Radosavljevic, A., & Vilela, B.

(2015). spThin: An R package for spatial thinning of species oc-

currence records for use in ecological niche models. Ecography,

38, 541–545. https:// doi. org/ 10. 1111/ ecog. 01132

Allendorf, F. W., Leary, R. F., Spruell, P., & Wenburg, J. K. (2001). The

problems with hybrids: Setting conservation guidelines. Trends

in Ecology & Evolution, 16, 613–622. https:// doi. org/ 10. 1016/

S0169 - 5347(01) 02290 - X

Allendorf, F. W., Luikart, G., & Aitken, S. N. (2012). Conservation

and the genetics of populations (2nd ed.). John Willey & Sons.

Anderson, E. C., & Thompson, E. A. (2002). A model- based method

for identifying species hybrids using multilocus genetic data.

Genetics, 160, 1217–1229. https:// doi. org/ 10. 1093/ genet ics/

160.3. 1217

Avise, J. C. (2008). Clonality. The genetics, ecology, and evolution of

sexual abstinence in vertebrate animals. Oxford University Press.

Benson, D. A., Cavanaugh, M., Clark, K., Karsch- Mizrachi, I.,

Lipman, D. J., Ostell, J., & Sayers, E. W. (2013). GenBank.

Nucleic Acids Research, 41, D36–D42. https:// doi. org/ 10. 1093/

nar/ gks1195

Berger, L., Rybacki, M., & Hotz, H. (1994). Artificial fertilization of

water frogs. Amphibia- Reptilia, 15, 408–413. https:// doi. org/ 10.

1163/ 15685 3894X 00452

Bowler, D. E., & Benton, T. G. (2005). Causes and consequences of

animal dispersal strategies: Relating individual behaviour to

spatial dynamics. Biological Reviews, 80, 205–225. https:// doi.

org/ 10. 1017/ S1464 79310 4006645

Broennimann, O., Fitzpatrick, M. C., Pearman, P. B., Petitpierre, B.,

Pellissier, L., Yoccoz, N. G., & Guisan, A. (2012). Measuring eco-

logical niche overlap from occurrence and spatial environmen-

tal data. Global Ecology and Biogeography, 21, 481–497. https://

doi. org/ 10. 1111/j. 1466- 8238. 2011. 00698. x

Brown, J. L., & Carnaval, A. C. (2019). A tale of two niches:

Methods, concepts, and evolution. Frontiers of Biogeography,

11, e44158.

Chmielewska, M., Dedukh, D., Haczkiewicz, K., Rozenblut- Kościsty,

B., Kaźmierczak, M., Kolenda, K., & Ogielska, A. (2018). The

programmed DNA elimination and formation of micronuclei

in germ line cells of the natural hybridogenetic water frog

PAPEŽÍK etal.

Pelophylax esculentus. Scientific Reports, 8, 7870. https:// doi.

org/ 10. 1038/ s4159 8- 018- 26168 - z

Choleva, L., Doležálková- Kaštánková, M., Labajová, V., Sember, A.,

Altmanová, M., Lukšíková, K., Volenikova, A., Dalikova, M.,

Nguyen, P., Pustovalova, E., Fedorova, A., & Dedukh, D. (2023).

Formation of hemiclonal reproduction and hybridogenesis in

Pelophylax water frogs studied with species- specific cytogenomic

probes. In Press. https:// doi. org/ 10. 1101/ 2023. 10. 29. 564577

Choleva, L., Janko, K., De Gelas, K., Bohlen, J., Šlechtová, V., Rábová,

M., & Ráb, P. (2012). Synthesis of clonality and polyploidy in

vertebrate animals by hybridization between two sexual spe-

cies. Evolution, 66, 2191–2203. https:// doi. org/ 10. 1111/j. 1558-

5646. 2012. 01589. x

Christiansen, D., & Reyer, H.- U. (2009). From clonal to sexual hy-

brids: Genetic recombination via triploids in all- hybrid popula-

tions of water frogs. Evolution, 63, 1754–1768. https:// doi. org/

10. 1111/j. 1558- 5646. 2009. 00673. x

Clement, M., Posada, D., & Crandall, K. A. (2000). TCS: A computer

program to estimate gene genealogies. Molecular Ecology, 9,

1657–1659. https:// doi. org/ 10. 1046/j. 1365- 294x. 2000. 01020. x

Coyne, J. A., & Orr, H. A. (2004). Speciation. Sinauer Associates, Inc.

Crnobrnja- Isailovic, J. (2007). Cross- section of a refugium: Genetic

diversity of amphibian and reptile populations in the Balkans.

In S. Weiss & N. Ferrand (Eds.), Phylogeography of southern

European refugia (pp. 327–337). Springer. https:// doi. org/ 10.

1007/ 1- 4020- 4904- 8_ 13

Dedukh, D., & Krasikova, A. (2022). Delete and survive: Strategies

of programmed genetic material elimination in eukaryotes.

Biological Reviews, 97, 195–216. https:// doi. org/ 10. 1111/ brv.

12796

Dedukh, D., Litvinchuk, S., Rosanov, J., Shabanov, D., & Krasikova,

A. (2017). Mutual maintenance of di- and triploid Pelophylax

esculentus hybrids in R- E systems: Results from artificial cross-

ings experiments. BMC Evolutionary Biology, 17, 220. https://

doi. org/ 10. 1186/ s1286 2- 017- 1063- 3

Dedukh, D., Riumin, S., Chmielewska, M., Rozenblut- Kościsty,

B., Kolenda, K., Kaźmierczak, M., & Krasikova, A. (2020).

Micronuclei in germ cells of hybrid frogs from Pelophylax es-

culentus complex contain gradually eliminated chromosomes.

Scientific Reports, 10, 8720. https:// doi. org/ 10. 1038/ s4159 8- 020-

64977 - 3

Dufresnes, C., Monod- Broca, B., Bellati, A., Canestrelli, D., Ambu,

J., Wielstra, B., & Jablonski, D. (2024). Piecing the barcoding

puzzle of palearctic water frogs (Pelophylax) sheds light on

amphibian biogeography and global invasions. Global Change

Biology, 30, e17180. https:// doi. org/ 10. 1111/ gcb. 17180

Džukić, G., & Kalezić, M. L. (2004). The biodiversity of amphib-

ians and reptiles in the Balkan Peninsula. In H. I. Griffiths

(Ed.), Balkan Biodiversity (pp. 167–192). Kluwer Academic

Publishers.

Edelaar, P., & Bolnick, D. I. (2012). Non- random gene flow: An

underappreciated force in evolution and ecology. Trends in

Ecology & Evolution, 27, 659–665. https:// doi. org/ 10. 1016/j. tree.

2012. 07. 009

Elith, J., Phillips, S. J., Hastie, T., Dudík, M., Chee, Y. E., & Yates,

C. J. (2010). A statistical explanation of MaxEnt for ecologists.

Diversity and Distributions, 17, 43–57. https:// doi. org/ 10. 1111/j.

1472- 4642. 2010. 00725. x

Evanno, G., Regnaut, S., & Goudet, J. (2005). Detecting the number

of clusters of individuals using the software STRUCTURE: A

simulation study. Molecular Ecology, 14, 2611–2620. https:// doi.

org/ 10. 1111/j. 1365- 294X. 2005. 02553. x

Fick, S. E., & Hijmans, R. J. (2017). WorldClim 2: New 1- km spatial

resolution climate surfaces for global land areas. International

Journal of Climatology, 37, 4302–4315. https:// doi. org/ 10. 1002/

joc. 5086

Francis, R. M. (2017). POPHELPER: An R package and web app to

analyse and visualize population structure. Molecular Ecology

Resources, 17, 27–32. https:// doi. org/ 10. 1111/ 1755- 0998. 12509

Guerrini, F., Bucci, S., Ragghianti, M., Mancino, G., Hotz, H., Uzzell,

T., & Berger, L. (1997). Genomes of two water frog species

resist germ line exclusion in interspecies hybrid. Journal of

Experimental Zoology, 279, 163–176. https:// doi. org/ 10. 1002/

(SICI) 1097- 010X(19971 001) 279: 2< 163:: AID- JEZ7> 3.0. CO; 2- M

Haxhiu, I. (1994). The herpetofauna of Albania. Amphibia: Species

composition, distribution, habitats. Zoologische Jahrbücher.

Abteilung für Systematik, Geographie und Biologie der Tiere, 121,

321–334.

Hedrick, P. W. (2005). A standardized genetic differentiation mea-

sure. Evolution, 59, 1633–1638. https:// doi. org/ 10. 1111/j. 0014-

3820. 2005. tb018 14. x

Hijmans, R. J., Phillips, S., Leatwick, J., & Elith, J. (2011). Package

‘dismo’. http:// cran. r- proje ct. org/ web/ packa ges/ dismo/ index.

html

Hotz, H., Mancino, G., Bucci- Innocenti, S., Ragghianti, M., Berger,

L., & Uzzell, T. (1985). Rana ridibunda varies geographically in

inducing clonal gametogenesis in interspecific hybrids. Journal

of Experimental Zoology, 236, 199–210. https:// doi. org/ 10. 1002/

jez. 14023 60210

Hotz, H., & Uzzell, T. (1982). Biochemically detected sympatry of

two water frog species: Two different cases in the Adriatic

Balkans (Amphibia, Ranidae). Proceedings of the Academy of

Natural Sciences of Philadelphia, 134, 50–79.

Jiggins, C. D., & Mallet, J. (2000). Bimodal hybrid zones and spe-

ciation. Trends in Ecology & Evolution, 15, 250–255. https:// doi.

org/ 10. 1016/ S0169 - 5347(00) 01873 - 5

Jombart, T. (2008). Adegenet: A R package for the multivariate anal-

ysis of genetic markers. Bioinformatics, 24, 1403–1405. https://

doi. org/ 10. 1093/ bioin forma tics/ btn129

Jost, L. (2008). GST and its relatives do not measure differentiation.

Molecular Ecology, 17, 4015–4026. https:// doi. org/ 10. 1111/j.

1365- 294X. 2008. 03887. x

Kolenda, K., Pietras- Lebioda, A., Hofman, S., Ogielska, M., &

Pabijan, M. (2017). Preliminary genetic data suggest the oc-

currence of the Balkan water frog, Pelophylax kurtmuelleri, in

Southwestern Poland. Amphibia- Reptilia, 38, 187–196. https://

doi. org/ 10. 1163/ 15685 381- 00003103

Kollár, J., Poulíčková, A., & Dvořák, P. (2022). On the relativity of spe-

cies, or the probabilistic solution to the species problem. Molecular

Ecology, 31, 411–418. https:// doi. org/ 10. 1111/ mec. 16218

Kordges, T. (1988). Beobachtungen am Epirusfrosch (Rana epeirot-

ica), dem neuen griechischen Wasserfrosch. In R. Günther

& R. Klewen (Eds.), Beiträge zur Biologie und Bibliographie

(1960–1987) der europäischen Wasserfrösche. Jahrbuch für

Feldherptologie, Beiheft 1.

Kyriakopoulou- Sklavounou, P., & Vasara, E. (2001). Morphometric

data on triploid water frogs of the system Rana epeirotica and

Rana balcanica in Greece. In P. Lymberakis, E. Valakos, P.

Pafilis, & M. Mylonas (Eds.), Herpetologia Candiana. Societas

Europaea Herpetologica.

PAPEŽÍK etal.

Leigh, J. W., & Bryant, D. (2015). PopART: Full- feature software

for haplotype network construction. Methods in Ecology and

Evolution, 6, 1110–1116. https:// doi. org/ 10. 1111/ 2041- 210X.

12410

Litvinchuk, S. N., Ivanov, A. Y., Lukonina, S. A., & Ermakov, O. A.

(2020). A record of alien Pelophylax species and widespread

mitochondrial DNA transfer in Kaliningradskaya Oblasť (the

Baltic coast, Russia). BioInvasions Records, 9, 599–617. https://

doi. org/ 10. 3391/ bir. 2020.9. 3. 16

Lymberakis, P., Poulakakis, N., Manthalou, G., Tsigenopoulos, C.

S., Magoulas, A., & Mylonas, M. (2007). Mitochondrial phy-

logeography of Rana (Pelophylax) populations in the eastern

Mediterranean region. Molecular Phylogenetics and Evolution,

44, 115–125. https:// doi. org/ 10. 1016/j. ympev. 2007. 03. 009

Mann, H. B., & Whitney, D. R. (1947). On a test of whether one of

two random variables is stochastically larger than the other.

Annals of Mathematical Statistics, 18, 50–60. https:// doi. org/ 10.

1214/ aoms/ 11777 30491

Meirmans, P. G., & Hedrick, P. W. (2011). Assessing population struc-

ture: FST and related measures. Molecular Ecology Resources,

11, 5–18. https:// doi. org/ 10. 1111/j. 1755- 0998. 2010. 02927. x

Miller, J. M., Culingham, C. I., & Perry, R. M. (2020). The influence

of a priori grouping on inference of genetic clusters: Simulation

study and literature review of the DAPC method. Heredity, 125,

269–280. https:// doi. org/ 10. 1038/ s4143 7- 020- 0348- 2

Nei, M. (1973). Analysis of gene diversity in subdivided populations.

Proceedings of the National Academy of Sciences of the United

States of America, 70, 3321–3323. https:// doi. org/ 10. 1073/ pnas.

70. 12. 3321

Nei, M., & Chesser, R. K. (1983). Estimation of fixation indices

and gene diversities. Annals of Human Genetics, 47, 253–259.

https:// doi. org/ 10. 1111/j. 1469- 1809. 1983. tb009 93. x

Ogielska, M. (1994). Nucleus- like bodies in gonial cells of Rana es-

culenta (amphibia, Anura) tadpoles: A putative way of chromo-

some elimination. Zoologica Poloniae, 39, 461–474.

Oksanen, J., Simpson, G., Blanchet, F., Kindt, R., Legendre, P.,

Minchin, P., & Borman, T. (2024). Vegan: Community ecology

package. R Package Version 2.7–0. https:// github. com/ vegan

devs/ vegan

van Oosterhout, C., Hutchinson, W. F., Wills, D. P. M., & Shipley,

P. (2004). MicroChecker: Software for identifying and correct-

ing genotyping errors in microsatellite data. Molecular Ecology

Notes, 4, 535–538. https:// doi. org/ 10. 1111/j. 1471- 8286. 2004.

00684. x

Oruçi, S. (2008). Data on geographical distribution and habitats

of the Rana epeirotica in Albania. Natura Montenegrina, 7,

419–423.

Papežík, P., Kubala, M., Jablonski, D., Doležálková- Kaštánková, M.,

Choleva, L., Benovics, M., & Mikulíček, P. (2021). Morphological

differentiation of endemic water frogs (Ranidae: Pelophylax)

from the southwestern Balkans. Salamandra, 57, 105–123.

Papežík, P., Mikulíček, P., Benovics, M., Balogová, M., Choleva,

L., Doležálková- Kaštánková, M., & Jablonski, D. (2023).

Comparative mitochondrial phylogeography of water frogs

(Ranidae: Pelophylax spp.) from the southwestern Balkans.

Vertebrate Zoology, 73, 525–544. https:// doi. org/ 10. 3897/ vz. 73.

e95220

Paradis, E., & Schliep, K. (2019). Ape 5.0: An environment for modern

phylogenetics and evolutionary analyses in R. Bioinformatics,

35, 526–528. https:// doi. org/ 10. 1093/ bioin forma tics/ bty633

Peakall, R., & Smouse, P. E. (2006). GENALEX 6: Genetic analysis

in excel. Population genetic software for teaching and research.

Molecular Ecology Notes, 6, 288–295. https:// doi. org/ 10. 1111/j.

1471- 8286. 2005. 01155. x

Peakall, R., & Smouse, P. E. (2012). GenAlEx 6.5: Genetic analysis in

excel. Population genetic software for teaching and research- an

update. Bioinformatics, 28, 2537–2539. https:// doi. org/ 10. 1093/

bioin forma tics/ bts460

Plötner, J. (2005). Die westpaläarktischen Wasserfrösche: von

Märtyrern der Wissenschaft zur biologischen Sensation. Laurenti

Verlag.

Plötner, J., Baier, F., Akin, C., Mazepa, G., Schreiber, R., Beerli, P.,

& Uzell, T. (2012). Genetic data reveal that water frogwater

frogs of Cyprus (genus Pelophylax) are an endemic species of

Messinian origin. Zoosystematics and Evolution, 88, 261–283.

https:// doi. org/ 10. 1002/ zoos. 20120 0021

Plötner, J., Uzzell, T., Beerli, P., Akin, C., Bilgin, C. C., Haefeli, C.,

Haefeli, C., Ohst, T., Köhler, F., Schreiber, R., Guex, G.- D.,

Litvinchuk, S., Westaway, R., Reyer, H.- U., Pruvost, N., Hotz,

H., Akın Pekşen, Ç., Guex, G.- D., & Hotz, H. (2010). Genetic

divergence and evolution of reproductive isolation in eastern

Mediterranean water frogs. In M. Glaubrecht (Ed.), Evolution

in action, case studies in adaptive radiation, speciation and the

origin of biodiversity (pp. 373–403). Springer.

Plötner, J., Uzzell, T., Beerli, P., Spolsky, C., Ohst, T., Litvinchuk, S.

N., & Hotz, H. (2008). Widespread unidirectional transfer of

mitochondrial DNA: A case in western palearctic water frogs.

Journal of Evolutionary Biology, 21, 668–681. https:// doi. org/ 10.

1111/j. 1420- 9101. 2008. 01527. x

Pritchard, J. K., Stephens, M., & Donnelly, P. (2000). Inference

of population structure using multilocus genotype data.

Genetics, 155, 945–959. https:// doi. org/ 10. 1093/ genet ics/

155.2. 945

Pruvost, N. B. M., Hoffmann, A., & Reyer, H. U. (2013). Gamete pro-

duction patterns, ploidy, and population genetics reveal evolu-

tionary significant units in hybrid water frogs (Pelophylax escu-

lentus). Ecology and Evolution, 3, 2933–2946. https:// doi. org/ 10.

1002/ ece3. 687

Psonis, N., Antoniou, A., Kukushkin, O., Jablonski, D., Petrov, B.,

Crnobrnja- Isalović, J., & Poulakakis, N. (2017). Hidden di-

versity in the Podarcis tauricus (Sauria, Lacertidae) species

subgroup in the light of multilocus phylogeny and species de-

limitation. Molecular Phylogenetics and Evolution, 106, 6–17.

https:// doi. org/ 10. 1016/j. ympev. 2016. 09. 007

R Core Team. (2024). R: A language and environment for statistical

computing. R Foundation for Statistical Computing. https://

www. R- proje ct. org/

Rosser, N. F., Seixas, F., Queste, L. M., Cama, B., Mori- Pezo, R.,

Kryvokhyzha, D., Nelson, M., Waite- Hudson, R., Goringe,

M., Costa, M., Elias, M., Mendes Eleres de Figueiredo, C.,

Freitas, A. V. L., Joron, M., Kozak, K., Lamas, G., Martins, A.

R. P., McMillan, W. O., Ready, J., … Dasmahapatra, K. K. (2024).

Hybrid speciation driven by multilocus introgression of ecolog-

ical traits. Nature, 628, 811–817. https:// doi. org/ 10. 1038/ s4158

6- 024- 07263 - w

Rozas, J., Ferrer- Mata, A., Sánchez- DelBarrio, J. C., Guirao- Rico, S.,

Librado, P., Ramos- Onsins, S. E., & Sánchez- Garcia, A. (2017).

DnaSP 6: DNA sequence polymorphism analysis of large data

sets. Molecular Biology and Evolution, 34, 3299–3302. https://

doi. org/ 10. 1093/ molbev/ msx248

PAPEŽÍK etal.

Sagonas, K., Karameta, E., Kotsakiozi, P., & Poulakakis, N. (2020).

Cross- species testing of nuclear markers in Pelophylax water

frogs in Greece and examination of their power to detect ge-

netic admixture. Amphibia- Reptilia, 42, 1–7. https:// doi. org/ 10.

1163/ 15685 381- 20191229

Schneider, H. (2005). Bioakustik der Froschlurche: einheimische und

verwandte Arten. Volume 6 Supplemente der Zeitschrift für

Feldherpetologie 6. Laurenti Verlag.

Schneider, H., & Haxhiu, I. (1992). The distribution of the Epirus

frog (Rana epeirotica) in Albania. Amphibia- Reptilia, 13,

293–295.

Schneider, H., & Sofianidou, T. S. (1986). Bioacoustic study of

water frogs (Ranidae) in Greece. In Z. Roček (Ed.), Studies in

Herpetology (pp. 561–564). Charles University.

Schneider, H., Sofianidou, T. S., & Kyriakopoulou- Sklavounou, P.

(1984). Bioacoustic and morphometric studies in water frogs

(genus Rana) of Lake Ioannina in Greece, and description of

a new species (Anura, amphibia). Zeitschrift für Zoologische

Systematik Und Evolutionsforschung, 22, 349–366. https:// doi.

org/ 10. 1111/j. 1439- 0469. 1984. tb006 69. x

Sofianidou, T. S. (1996). Electrophoretic studies of hybrids of water

frogs (Rana epeirotica, R. balcanica) in the Ionian zone of

Greece. Israel Journal of Zoology, 42, 149–157. https:// doi. org/

10. 1080/ 00212 210. 1996. 10688838

Sofianidou, T. S., & Schneider, H. (1989). Distribution range of the

Epirus frog Rana epeirotica (Amphibia: Anura) and the com-

position of the water frog populations in Western Greece.

Zoologischer Anzeiger, 223, 13–25.

Sofianidou, T. S., Schneider, H., & Asimakopoulos, B. (1987).

Distribution of the water frog Rana epeirotica (Amphibia,

Anura) in Greece. In J. J. van Gelden, H. Strijbosch, & P. J.

M. Bergers (Eds.), Proceedings of the fourth ordinary general

meeting of the Societas europaea Herpetologica (pp. 365–367).

Faculty of Sciences, Nijmegen.

Speybroeck, J., Beukema, W., & Crochet, P.- A. (2010). A tenta-

tive species list of the European herpetofauna (amphibia and

Reptilia) - an update. Zootaxa, 2492, 1–27.

Speybroeck, J., Beukema, W., Dufresnes, C., Fritz, U., Jablonski,

D., Lymberakis, P., & Crochet, P.- A. (2020). Species list of

the European herpetofauna – 2020 update by the taxo-

nomic Committee of the Societas Europaea Herpetologica.

Amphibia- Reptilia, 41, 139–189. https:// doi. org/ 10. 1163/ 15685

381- bja10010

Szabolcs, M., Mizsei, E., Jablonski, D., Vági, B., Mester, B., Végvári,

Z., & Szabolcs, L. (2017). Distribution and diversity of amphib-

ians in Albania: New data and foundations of a comprehensive

database. Amphibia- Reptilia, 38, 435–448. https:// doi. org/ 10.

1163/ 15685 381- 00003126

Taylor, S. A., & Larson, E. L. (2019). Insights from genomes into the

evolutionary importance and prevalence of hybridization in na-

ture. Nature Ecology & Evolution, 3, 170–177. https:// doi. org/ 10.

1038/ s4155 9- 018- 0777- y

Thia, J. A. (2023). Guidelines for standardizing the application of

discriminant analysis of principal components to genotype

data. Molecular Ecology Resources, 23, 523–538. https:// doi. org/

10. 1111/ 1755- 0998. 13706

Title, P. O., & Bemmels, J. B. (2018). Envirem: An expanded set of bio-

climatic and topographic variables increases flexibility and im-

proves performance of ecological niche modeling. Ecography,

41, 291–307. https:// doi. org/ 10. 1111/ ecog. 02880

Tunner, H. G., & Heppich, S. (1981). Premeiotic genome exclusion

during oogenesis in the common edible frog, Rana esculenta.

Naturwissenschaften, 68, 207–208. https:// doi. org/ 10. 1007/ BF010

47207

Valakos, E. D., Pafilis, P., Sotiropoulos, K., Lymberakis, P., Maragou,

P., & Foufopoulos, J. (2008). The amphibians and reptiles of

Greece. Chimaira.

Verity, R., & Nichols, R. A. (2016). Estimating the number of sub-

populations (K) in structured populations. Genetics, 203, 1827–

1839. https:// doi. org/ 10. 1534/ genet ics. 115. 180992

Vignali, S., Barras, A. G., Arlettaz, R., & Braunisch, V. (2020).

SDMtune: An R package to tune and evaluate species distribu-

tion models. Ecology and Evolution, 10, 11488–11506. https://

doi. org/ 10. 1002/ ece3. 6786

Warren, D. L., Beaumont, L. J., Dinnage, R., & Baumgartner, B.

(2019). New methods for measuring ENM breadth and overlap

in environmental space. Ecography, 42, 444–446. https:// doi.

org/ 10. 1111/ ecog. 03900

Warren, D. L., Glor, R. E., & Turelli, M. (2008). Environmental niche

equivalency versus conservatism: Quantitative approaches to

niche evolution. Evolution, 62, 2868–2883. https:// doi. org/ 10.

1111/j. 1558- 5646. 2008. 00482. x

Warren, D. L., Glor, R. E., & Turelli, M. (2010). ENMTools: A tool-

box for comparative studies of environmental niche models.

Ecography, 33, 607–611. https:// doi. org/ 10. 1111/j. 1600- 0587.

2009. 06142. x

Warren, D. L., Matzke, N. J., Cardillo, M., Baumgartner, J. B.,

Beaumont, L. J., Turelli, M., & Dinnage, R. (2021). ENMTools

1.0: An R package for comparative ecological biogeography.

Ecography, 44, 504–511. https:// doi. org/ 10. 1111/ ecog. 05485

Westram, A. M., Stankowski, S., Surendranadh, P., & Barton, N.

(2022). What is reproductive isolation? Journal of Evolutionary

Biology, 35, 1143–1164. https:// doi. org/ 10. 1111/ jeb. 14005

Wickham, H. (2016). ggplot2: Elegant graphics for data analysis.

Springer Verlag.

Winter, D. J. (2012). MMOD: An R library for the calculation of popu-

lation differentiation statistics. Molecular Ecology Resources, 12,

1158–1160. https:// doi. org/ 10. 1111/j. 1755- 0998. 2012. 03174. x

SUPPORTING INFORMATION

Additional supporting information can be found online

in the Supporting Information section at the end of this

article.

How to cite this article: Papežík, P.,

Aschengeschwandtnerová, S., Benovics, M.,

Dedukh, D., Doležálková- Kaštánková, M., Choleva,

L., Javorčík, A., Lymberakis, P., Papežíková, S.,

Poulakakis, N., Sarı, İ., Šanda, R., Vukić, J., &

Mikulíček, P. (2025). A pattern of hybridisation and

population genetic structure of two water frog

species (Ranidae, Amphibia) in the southwestern

Balkans. Zoologica Scripta, 00, 1–22. https://doi.

org/10.1111/zsc.12723

zsc12723-sup-0001-appendixs1.pdf

Data

February 2025

Petr Papežík · Sandra Aschengeschwandtnerová · Michal Benovics · Dmitrij Dedukh · Peter Mikulicek

Download

Assessment of environmental variables for species distribution modelling: insights from the mosaic distribution of red- and yellow-bellied toads

Preprint

Full-text available

Apr 2025

Species distribution modelling can possibly be improved through the preferential use of explanatory variables that reflect the natural history features of the species being modelled. Red- and yellow-bellied toads (genus Bombina ) engage in an intricate mosaic distribution across Europe. Analysing new atlas data on these species’ mutual distribution in Hungary with principal coordinate analysis we identified their differential ecological preferences as forested, hilly and mountainous for B. variegata and open lowland for B. bombina . These locally operating parameters we consider to be good proxies for the essential species difference which resides in breeding in ephemeral puddles ( B. variegata ) versus larger permanent ponds ( B. bombina ). With two-species distribution modelling – in which the presence of one species is contrasted with the presence of the counterpart species – we obtained excellent model fit (AUC) for climate and elevation / land cover datasets alike (AUC=0.98 versus 0.95). For both models fit values dropped upon transference to surrounding countries, yet the latter model kept significantly higher predictive power (AUC=0.91) than the climate model (AUC=0.79). Swapping elevation for ‘hilliness’ as suggested in the literature had a significant negative effect on model performance. We conclude that an informed parameter selection enhances model transferability, therewith improving our understanding of species-habitat associations. Graphical abstract Highlights Red- and yellow-bellied toads engage in a yin-yang-like mosaic distribution across Europe. Species’ differential habitat characteristics were studied with principal coordinates. Species distribution modelling was with climate data and with landscape variables. Models produced with atlas data for Hungary were assessed over neighbouring countries. Transferred models with elevation and forestation performed better than climate-based ones.

A new diagnostic character to distinguish the Epirus Waterfrog, Pelophylax epeiroticus (Schneider, Sofianidou & Kyriakopoulou-Sklavounou, 1984), from the syntopic Balkan Frog, Pelophylax ridibundus kurtmuelleri (Gayda, 1940)

Article

Full-text available

Jun 2025

Hybrid speciation driven by multilocus introgression of ecological traits

Article

Full-text available

Apr 2024
NATURE

Hybridization allows adaptations to be shared among lineages and may trigger the evolution of new species1,2. However, convincing examples of homoploid hybrid speciation remain rare because it is challenging to demonstrate that hybridization was crucial in generating reproductive isolation³. Here we combine population genomic analysis with quantitative trait locus mapping of species-specific traits to examine a case of hybrid speciation in Heliconius butterflies. We show that Heliconius elevatus is a hybrid species that is sympatric with both parents and has persisted as an independently evolving lineage for at least 180,000 years. This is despite pervasive and ongoing gene flow with one parent, Heliconius pardalinus, which homogenizes 99% of their genomes. The remaining 1% introgressed from the other parent, Heliconius melpomene, and is scattered widely across the H. elevatus genome in islands of divergence from H. pardalinus. These islands contain multiple traits that are under disruptive selection, including colour pattern, wing shape, host plant preference, sex pheromones and mate choice. Collectively, these traits place H. elevatus on its own adaptive peak and permit coexistence with both parents. Our results show that speciation was driven by introgression of ecological traits, and that speciation with gene flow is possible with a multilocus genetic architecture.

Piecing the barcoding puzzle of Palearctic water frogs (Pelophylax) sheds light on amphibian biogeography and global invasions

Article

Full-text available

Mar 2024
GLOBAL CHANGE BIOL

Palearctic water frogs (genus Pelophylax) are an outstanding model in ecology and evolution, being widespread, speciose, either threatened or threatening to other species through biological invasions, and capable of siring hybrid offspring that escape the rules of sexual reproduction. Despite half a century of genetic research and hundreds of publications, the diversity, systematics and biogeography of Pelophylax still remain highly confusing, in no small part due to a lack of correspondence between studies. To provide a comprehensive overview, we gathered >13,000 sequences of barcoding genes from >1700 native and introduced localities and built multigene mitochondrial (~17 kb) and nuclear (~10 kb) phylogenies. We mapped all currently recognized taxa and their phylogeographic lineages (>40) to get a grasp on taxonomic issues, cyto‐nuclear discordances, the genetic makeup of hybridogenetic hybrids, and the origins of introduced populations. Competing hypotheses for the molecular calibration were evaluated through plausibility tests, implementing a new approach relying on predictions from the anuran speciation continuum. Based on our timetree, we propose a new biogeographic paradigm for the Palearctic since the Paleogene, notably by attributing a prominent role to the dynamics of the Paratethys, a vast paleo‐sea that extended over most of Europe. Furthermore, our results show that distinct marsh frog lineages from Eastern Europe, the Balkans, the Near East, and Central Asia (P. ridibundus ssp.) are naturally capable of inducing hybridogenesis with pool frogs (P. lessonae). We identified 14 alien lineages (mostly of P. ridibundus) over ~20 areas of invasions, especially in Western Europe, with genetic signatures disproportionally pointing to the Balkans and Anatolia as the regions of origins, in line with exporting records of the frog leg industry and the stocks of pet sellers. Pelophylax thus emerges as one of the most invasive amphibians worldwide, and deserves much higher conservation concern than currently given by the authorities fighting biological invasions.

Formation of hemiclonal reproduction and hybridogenesis in Pelophylax water frogs studied with species-specific cytogenomic probes

Preprint

Full-text available

Nov 2023

Meiosis is a conservative process in all sexual organisms which ensures fertility and is central for producing genetic diversity by recombination and random segregation of parental chromosomes. Yet unexplored mechanisms may disrupt it and cause loss of sex followed by the emergence of clonal modes of reproduction. Interspecific hybridization is the primary trigger for this process, but mechanistic basis of the transition to asexuality remains still unknown for most vertebrate animals. To study these processes in water frogs, we performed reciprocal mating between two sexual species, Pelophylax ridibundus and P. lessonae, and produced vital F1 progeny (P. esculentus). The RepeatExplorer2 analysis of low-coverage genomic data of the two parental species identified the P. lessonae-specific minisatellite marker PlesSat01-48 (44 bp), which hybridized to (peri)centromeric regions of two chromosome pairs in P. lessonae, namely the acrocentric chromosome 8 and the chromosome 10 (a carrier of nucleolar organizer region; NOR). Chromosomal mapping combining the novel hybridization probe with the previously designed marker for P. ridibundus-specific centromeric satellite DNA showed that the P. esculentus progeny do not reproduce sexually. Instead, the F1 generation of P. esculentus instantly modified its gametogenesis and established asexual reproduction via hybridogenesis. Gametogenic modifications included premeiotic elimination of one of the parental genomes and clonal propagation of the remaining genome via endoreplication followed by standard meiotic division. The origin of DNA elimination and hybridogenesis in laboratory-produced hybrids supports a hypothesis that P. esculentus arises recurrently in nature whenever parental species come into reproductive contact. Based on the observed pattern of DNA elimination in the F1 progeny we discuss the origin and evolution of population systems in water frogs and the applicability of a newly designed chromosomal probe for other Pelophylax taxa.

Comparative mitochondrial phylogeography of water frogs (Ranidae: Pelophylax spp.) from the southwestern Balkans

Article

Full-text available

May 2023

The genus Pelophylax (water frogs) includes relatively common, widely distributed, and even invasive species, but also endemic taxa with small ranges and limited knowledge concerning their ecology and evolution. Among poorly studied species belong endemics of the southwestern Balkans, namely Pelophylax shqipericus, P. epeiroticus and P. kurtmuelleri. In this study, we focused on the genetic variability of these species aiming to reveal their phylogeographic patterns and Quaternary history. We used 1,088 published and newly obtained sequences of the mitochondrial ND2 gene and a variety of analyses, including molecular phylogenetics and dating, historical demography, and species distribution modeling (SDM). We revelated the existence of two mitochondrial lineages within P. epeiroticus and P. shqipericus that diverged at ~ 0.9 Mya and ~ 0.8 Mya, respectively. Contrarily, no deeply diverged lineages were found in P. kurtmuelleri. Pelophylax kurtmuelleri also shows a close phylogenetic relationship with widely distributed P. ridibundus, suggesting that both represent one evolutionary clade called here P. ridibundus/kurtmuelleri. The estimated split between both lineag-es in the clade P. ridibundus/kurtmuelleri date back to ~ 0.6 Mya. The divergence between the ridibundus and kurtmuelleri lineages on the ND2 gene is thus lower than the divergence between the two lineages found in P. epeiroticus and P. shqipericus. According to haplotype networks, demographic analyses, and SDM, endemic water frogs survived the last glacial maximum (LGM) in Balkan in the Balkans microrefugia, and their distribution has not changed significantly or even retracted since the LGM. Haplotypes of the kurtmuelleri lineage were also found in northern parts of Europe, where haplotype diversity is however much lower than in the Balkans, suggesting the possible hypothesis of their postglacial expansion to the north.

What is reproductive isolation?

Article

Full-text available

Sep 2022
J EVOLUTION BIOL

Sean Stankowski

Reproductive isolation (RI) is a core concept in evolutionary biology. It has been the central focus of speciation research since the modern synthesis and is the basis by which biological species are defined. Despite this, the term is used in seemingly different ways, and attempts to quantify RI have used very different approaches. After showing that the field lacks a clear definition of the term, we attempt to clarify key issues, including what RI is, how it can be quantified in principle, and how it can be measured in practice. Following other definitions with a genetic focus, we propose that RI is a quantitative measure of the effect that genetic differences between populations have on gene flow. Specifically, RI compares the flow of neutral alleles in the presence of these genetic differences to the flow without any such differences. RI is thus greater than zero when genetic differences between populations reduce the flow of neutral alleles between populations. We show how RI can be quantified in a range of scenarios. A key conclusion is that RI depends strongly on circumstances—including the spatial, temporal and genomic context—making it difficult to compare across systems. After reviewing methods for estimating RI from data, we conclude that it is difficult to measure in practice. We discuss our findings in light of the goals of speciation research and encourage the use of methods for estimating RI that integrate organismal and genetic approaches. Abstract Reproductive isolation (RI) is a core concept in evolutionary biology and the basis by which biological species are defined. Despite this, the term is used in different ways and efforts to quantify RI from data have used vastly different approaches. In this paper, we attempt to clarify key issues about RI, including what it is, how it can be quantified in principle, and how it can be measured in practice.

Guidelines for standardizing the application of discriminant analysis of principal components to genotype data

Article

Full-text available

Sep 2022
MOL ECOL RESOUR

Joshua A. Thia

Despite the popularity of discriminant analysis of principal components (DAPC) for studying population structure, there has been little discussion of best practice for this method. In this work, I provide guidelines for standardizing the application of DAPC to genotype data sets. An often overlooked fact is that DAPC generates a model describing genetic differences among a set of populations defined by a researcher. Appropriate parameterization of this model is critical for obtaining biologically meaningful results. I show that the number of leading PC axes used as predictors of among‐population differences, paxes, should not exceed the k−1 biologically informative PC axes that are expected for k effective populations in a genotype data set. This k−1 criterion for paxes specification is more appropriate compared to the widely used proportional variance criterion, which often results in a choice of paxes ≫ k−1. DAPC parameterized with no more than the leading k−1 PC axes: (i) is more parsimonious; (ii) captures maximal among‐population variation on biologically relevant predictors; (iii) is less sensitive to unintended interpretations of population structure; and (iv) is more generally applicable to independent sample sets. Assessing model fit should be routine practice and aids interpretation of population structure. It is imperative that researchers articulate their study goals, that is, testing a priori expectations vs. studying de novo inferred populations, because this has implications on how their DAPC results should be interpreted. The discussion and practical recommendations in this work provide the molecular ecology community with a roadmap for using DAPC in population genetic investigations.

On the relativity of species, or the probabilistic solution to the species problem

Article

Full-text available

Oct 2021
MOL ECOL

For centuries, both scientists and philosophers have discussed the nature of species resulting in c. 35 species concepts proposed to date. However, in our opinion, none of them incorporated neither recent advances in evolutionary genomics nor dimensionality of species in befitting depth. Our attempt to do so resulted in the following conclusions. Due to the continuous nature of evolution (regardless of its rate and constancy), species are inevitably undefinable as natural discontinuous units (except those originating in saltatory speciation) whenever the time dimension is taken into consideration. Therefore, the very existence of species as a natural discontinuous entity is relative to its dimensionality. A direct consequence of the relativity of species is the duality of speciators (e.g., incipient species) meaning that, in a given time, they may be perceived as both being and not being a species. Finally, the most accurate way to reflect both the relativity of species and the duality of speciators in species delimitation is probabilistic. While the novelty of these ideas may be questionable, they still deserve more extensive attention from the biological community. Here, we hope to draw such attention by outlining one of the possible pathways towards a new kind of probabilistic species delimitation methods based on the probability of irreversible divergence of evolutionary lineages. We anticipate that our probabilistic view of speciation has the potential to facilitate some of the most serious and universal issues of current taxonomy and to ensure unity of the species‐level taxonomy across the tree of life.

Delete and survive: strategies of programmed genetic material elimination in eukaryotes

Article

Full-text available

Sep 2021

Genome stability is a crucial feature of eukaryotic organisms because its alteration drastically affects the normal development and survival of cells and the organism as a whole. Nevertheless, some organisms can selectively eliminate part of their genomes from certain cell types during specific stages of ontogenesis. This review aims to describe the phenomenon of programmed DNA elimination, which includes chromatin diminution (together with programmed genome rearrangement or DNA rearrangements), B and sex chromosome elimination, paternal genome elimination, parasitically induced genome elimination, and genome elimination in animal and plant hybrids. During programmed DNA elimination, individual chromosomal fragments, whole chromosomes, and even entire parental genomes can be selectively removed. Programmed DNA elimination occurs independently in different organisms, ranging from ciliate protozoa to mammals. Depending on the sequences destined for exclusion, programmed DNA elimination may serve as a radical mechanism of dosage compensation and inactivation of unnecessary or dangerous genetic entities. In hybrids, genome elimination results from competition between parental genomes. Despite the different consequences of DNA elimination, all genetic material destined for elimination must be first recognised, epigenetically marked, separated, and then removed and degraded.

Morphological differentiation of endemic water frogs (Pelophylax, Ranidae) from the southwestern Balkans

Article

Full-text available

Feb 2021

The southwestern Balkans is inhabited by three endemic water frog species, Pelophylax shqipericus, P. epeiroticus and P. kurtmuelleri. Although these species are genetically and bioacoustically distinct from each other, their morphological differentiation is less pronounced as is generally observed in the genus Pelophylax. In this study, we analyzed morphological data of 215 individuals of these Balkan endemics in order to evaluate their morphological variability and facilitate their identification in the field. Moreover, since P. kurtmuelleri is a disputable taxon, we compared it with closely related P. ridibundus from Central Europe. The most pronounced morphological differences expressed by the analysis of variance (ANOVA) and the discriminant analysis of the principal components (DAPC) were found between P. kurtmuelleri and the two other Balkan species. Pelophylax kurtmuelleri and P. ridibundus, as well as P. shqipericus and P. epeiroticus were only slightly differentiated by morphology, which demonstrates that the morphological characters measured do not reflect the genetic divergence of the species and are weak indicators of their taxonomic status. Morphological characters suitable for species identification include a combination of indices L/CINT, T/CINT, L/T, and DP/CINT (L - snout-vent length, CINT - length of the metatarsal tubercle, T - length of the tibia, DP - length of the first toe), and qualitative traits like the pattern and shape of the lateral spots, presence/absence of yellow pigment in the flanks and thighs, foot webbing colouration, size and shape of the metatarsal tubercle, and colouration of the male vocal sacs.

A Model-Based Method for Identifying Species Hybrids Using Multilocus Genetic Data

Article

Mar 2002
GENETICS

We present a statistical method for identifying species hybrids using data on multiple, unlinked markers. The method does not require that allele frequencies be known in the parental species nor that separate, pure samples of the parental species be available. The method is suitable for both markers with fixed allelic differences between the species and markers without fixed differences. The probability model used is one in which parentals and various classes of hybrids (F1's, F2's, and various backcrosses) form a mixture from which the sample is drawn. Using the framework of Bayesian model-based clustering allows us to compute, by Markov chain Monte Carlo, the posterior probability that each individual belongs to each of the distinct hybrid classes. We demonstrate the method on allozyme data from two species of hybridizing trout, as well as on two simulated data sets.

A pattern of hybridisation and population genetic structure of two water frog species (Ranidae, Amphibia) in the southwestern Balkans

Abstract

Supplementary resource (1)

Recommended publications

Comparative mitochondrial phylogeography of water frogs (Ranidae: Pelophylax spp.) from the southwes...

A new diagnostic character to distinguish the Epirus Waterfrog, Pelophylax epeiroticus (Schneider, S...

Inheritance patterns of male asexuality in hybrid males of a water frog Pelophylax esculentus

Hidden in plain sight: novel molecular data reveal unexpected genetic diversity among paramphistome...