mec_v16_i22_ofbcover
11/1/07
13:46
Page 1
ISSN 0962-1083
VOLUME 16, NUMBER 22, NOVEMBER 2007
Fundamental links between genes and elements:
evolutionary implications of ecological stoichiometry
P. D. Jeyasingh & L. J. Weider
4774
ORIGINAL ARTICLES
4674
4684
4699
4715
4728
4738
4747
Testing the role of genetic factors across multiple
independent invasions of the shrub Scotch broom
(Cytisus scoparius)
M. Kang, Y. M. Buckley & A. J. Lowe
Adaptive differences in gene expression in European
flounder (Platichthys flesus)
P. F. Larsen, E. E. Nielsen, T. D. Williams,
J. Hemmer-Hansen, J. K. Chipman, M. Kruhøffer,
P. Grønkjær, S. G. George, L. Dyrskjøt &
V. Loeschcke
High variation and strong phylogeographic pattern
among cpDNA haplotypes in Taxus wallichiana
(Taxaceae) in China and North Vietnam
L. M. Gao, M. Möller, X.-M. Zhang,
M. L. Hollingsworth, J. Liu, R. R. Mill,
M. Gibby & D.-Z. Li
Rangewide phylogeography in the greater horseshoe
bat inferred from microsatellites: implications for
population history, taxonomy and conservation
S. J. Rossiter, P. Benda, C. Dietz, Sh.-Y. Zhang &
G. Jones
High selfing and high inbreeding depression in
peripheral populations of Juncus atratus
S. G. Michalski & W. Durka
Genetic structure of Suillus luteus populations
in heavy metal polluted and nonpolluted habitats
L. A. H. Muller, J. Vangronsveld & J. V. Colpaert
Genetic structure and evolved malaria resistance in
Hawaiian honeycreepers
J. T. Foster, B. L. Woodworth, L. E. Eggert, P. J. Hart,
D. Palmer, D. C. Duffy & R. C. Fleischer
The population genomics of hepatitis B virus
C. Szmaragd & F. Balloux
Phylogeography, Speciation and Hybridization
4759
Co-phylogeography and comparative population
genetics of the threatened Galápagos hawk and three
4789
4808
4822
Kinship, Parentage and Behaviour
4837
Genome-wide analysis reveals differences in brain
gene expression patterns associated with caste and
reproductive status in honey bees (Apis mellifera)
C. M. Grozinger, Y. L. Fan, S. E. R. Hoover & M. L.
Winston
4849
Functional significance of genetically different
symbiotic algae Symbiodinium in a coral reef symbiosis
J. E. Loram, H. G. Trapido-Rosenthal & A. E. Douglas
4858
No relationship between individual genetic diversity
and prevalence of avian malaria in a migratory kestrel
J. Ortego, P. J. Cordero, J. M. Aparicio &
G. Calabuig
Avian Clock gene polymorphism: evidence for a
latitudinal cline in allele frequencies
A. Johnsen, A. E. Fidler, S. Kuhn, K. L. Carter,
A. Hoffmann, I. R. Barr, C. Biard, A. Charmantier,
M. Eens, P. Korsten, H. Siitari, J. Tomiuk &
B. Kempenaers
Ecological Interactions
VOLUME 16, NUMBER 22, NOVEMBER 2007, pp. 4649– 4880
Population and Conservation Genetics
4662
ectoparasite species: ecology shapes population
histories within parasite communities
N. K. Whiteman, R. T. Kimball & P. G. Parker
Genetic variation and phylogeography of free-living
mouse species (genus Mus) in the Balkans and the
Middle East
M. Macholán, M. Vysko0ilová, F. Bonhomme,
B. Kry9tufek, A. Orth & V. Vohralík
The role of tropical dry forest as a long-term barrier
to dispersal: a comparative phylogeographical
analysis of dry forest tolerant and intolerant frogs
A. J. Crawford, E. Bermingham & C. Polanía S.
Phylogeography and historical demography of the
Italian treefrog, Hyla intermedia, reveals multiple
refugia, population expansions and secondary
contacts within peninsular Italy
D. Canestrelli, R. Cimmaruta & G. Nascetti
Mechanical barriers to introgressive hybridization
revealed by mitochondrial introgression patterns in
Ohomopterus ground beetle assemblages
N. Nagata, K. Kubota, K. Yahiro & T. Sota
VOLUME 16
NUMBER 22
NOVEMBER
2007
ECOLOGY
INVITED REVIEW
4649
MOLECULAR
MOLECULAR
ECOLOGY
Ecological Genetics
4867
Information on this journal can be accessed at:
http://www.blackwellpublishing.com/mec
The journal is covered by AGRICOLA, Chemical Abstracts,
Current Awareness Biological Sciences and Current Contents.
This journal is available at Blackwell Synergy. Visit
www.blackwell-synergy.com to search the articles and
register for table of contents e-mail alerts.
Published by
Blackwell Publishing
MOLECULAR
ECOLOGY
Molecular Ecology (2007) 16, 4808–4821
doi: 10.1111/j.1365-294X.2007.03534.x
Phylogeography and historical demography of the Italian
treefrog, Hyla intermedia, reveals multiple refugia,
population expansions and secondary contacts within
peninsular Italy
Blackwell Publishing Ltd
D A N I E L E C A N E S T R E L L I , R O B E RTA C I M M A R U TA and G I U S E P P E N A S C E T T I
Dipartimento di Ecologia e Sviluppo Economico Sostenibile, Università della Tuscia, Via San Giovanni Decollato 1, 01100 Viterbo, Italy
Abstract
We investigated the geographical patterns of genetic diversity in the Italian treefrog
through sequence analysis of a mitochondrial cytochrome b gene fragment. Three main
mitochondrial lineages were identified, distributed in northern, central and southern Italy,
respectively. Their divergence appears indicative of a split time largely predating Late
Pleistocene climatic oscillations, and syntopy between them was only observed in the
geographically intermediate populations. The historical demographic reconstructions
suggest that in both northern and central Italy, an expansion occurred during the last major
glacial phase, when a vast widening of the lowland habitats followed the glaciation-induced
fall of the sea level. Instead, in southern Italy an expansion event likely followed the end
of the last glaciation, although the inference of expansion appears less reliable for the
southern clade than for the others. Within this geographical area, a sharp phylogeographic
discontinuity separated peninsular from Sicilian populations, and the overall pattern of
diversity suggests that the latter derived from a recent colonization of the island, probably
through a Late Pleistocene land bridge. Phylogenetic, phylogeographic and historical
demographic analyses thus concur in delineating a scenario of multiple refugia, with four
groups of populations which survived the last glacial–interglacial cycles in at least three
distinct refugia arranged along peninsular Italy, and have recently come into contact
following range expansions. Therefore, these results support the hypothesis that a plethora
of microevolutionary processes, rather than the prolonged stability of populations, were
mainly responsible for shaping the patterns of diversity within this major biodiversity
hotspot.
Keywords: Hyla intermedia, Italy, mitochondrial DNA, multiple refugia, phylogeography,
secondary contacts
Received 20 April 2007; revision received 16 July 2007; accepted 27 July 2007
Introduction
Quaternary climatic oscillations have played a major role
in shaping the present geographical distribution of both
species and their genetic diversity (recent reviews in Hewitt
2004a, b). Following these oscillations, most temperate
European taxa were forced into repeated cycles of retreat
within refugial ranges during pleniglacials and of expansion
Correspondence: Daniele Canestrelli Fax: +39-0761357751;
E-mail: [email protected]
during subsequent interglacial phases. A huge amount of
literature based on both palaeoecological and genetic data
indicate the three Mediterranean peninsulas of Iberia, Italy
and the Balkans as important southern Quaternary glacial
refugia (see Weiss & Ferrand 2006 and the many references
therein). Populations from these refugial ranges have often
been observed to harbour a high genetic diversity. Also,
widely distributed species often show in these areas the
largest portion of this diversity, leading to the pattern of
so-called ‘southern richness, northern purity’ (e.g. Hewitt
1996, 1999, 2000; Taberlet et al. 1998; but see also Stewart &
Lister 2001; Petit et al. 2003; Deffontaine et al. 2005; and
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
P H Y L O G E O G R A P H Y O F T H E I TA L I A N T R E E F R O G 4809
references therein). According to a widely accepted hypothesis, within the southern refugial ranges populations
would have experienced a prolonged demographic stability,
which would in turn be a major factor in explaining such
a southern richness (Hewitt 1996, 2000). Nevertheless,
more recently, growing emphasis has been placed on an
alternative scenario named ‘refugia-within-refugia’ (see
Gómez & Lunt 2006 and references therein). According to
this scenario, well supported by case studies mainly from
the Iberian Peninsula (Gómez & Lunt 2006; MartìnezSolano et al. 2006; Bella et al. 2007), multiple refugia existed
within the southern refugia, so that the main factors
shaping the present patterns of genetic diversity within
these areas would be the allopatric differentiation during
glacial phases, possibly followed by demographic expansions and consequent population admixture due to
secondary contacts during subsequent interglacials. Understanding the relative roles of these two possible scenarios
in shaping the patterns of genetic diversity within the
refugial ranges is a matter of great interest for various
reasons. In a multiple refugia scenario, the largest portion
of the overall genetic variation found in the refugial range
could be allocated to the among-population or the
among-regions levels of variation, rather than to the withinpopulation level, due to the existence of a high geographical
structuring of populations (e.g. Sanz et al. 2000; Alexandrino
et al. 2000; Paulo et al. 2002). Also, the higher intrapopulation
diversity often found in these areas could at least in part
be due to the abovementioned admixture events following
secondary contacts among allopatrically differentiated
lineages (Canestrelli et al. 2006). Furthermore, in a context
of strong population structure within the refugial range,
the process of genetic diversity loss during postglacial
range expansion could have been exacerbated by the fact
that only a subset of the overall southern diversity would
have contributed to such a process (Gómez & Lunt 2006).
Finally, to ascertain the relative roles of the two above
scenarios could also contribute to our understanding of the
patterns of diversity within the refugial ranges at community
level. A scenario of prolonged stability of populations
within a single refugium would offer opportunities for longterm co-evolution among community members. Instead, the
time available for such co-evolutionary processes to occur,
could have been much shorter in a refugia-within-refugia
scenario. In this latter case, the location of distinct glacial
refugia would not necessarily overlap among presently
interacting species or lineages within them. Furthermore,
they could have responded in either a concerted or an
independent manner to Quaternary climatic fluctuations
(Sullivan et al. 2000; Carstens et al. 2005; Steele & Storfer
2006).
Besides being part of one of the major world biodiversity hotspots (Myers et al. 2000), southern Mediterranean
peninsulas are therefore emerging as geographical areas
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
of particular interest for study of certain long-standing
issues in ecology and evolutionary biology, such as factors
underlying the uneven geographical distribution of
diversity, as well as the continuum between population
differentiation and speciation. Furthermore, these geographical regions have been indicated as the European areas that
will be most heavily impacted in future decades by climate
change-induced biodiversity depletion (e.g. Araujo et al.
2006). This reemphasizes the urgency of investigating
biodiversity patterns and the underling ecological and
evolutionary processes within these regions, also with the
aim of helping shape future conservation efforts.
As mentioned above, in the context of the Western
Palaearctic region, the vast majority of case studies published to date on these topics have focused on the Iberian
Peninsula, whereas comparatively few have investigated
taxa from other putative Quaternary refugia. With respect
to the Italian peninsula, however, several case studies have
recently suggested the occurrence of multiple refugia as a
possible explanation for the observed pattern of population genetic differentiation (Santucci et al. 1996; Nieberding
et al. 2005; Podnar et al. 2005; Canestrelli et al. 2006; Ursenbacher et al. 2006; Böhme et al. 2007; see Canestrelli 2006
for a brief review), which suggests that a refugia-withinrefugia scenario may be relevant also within this geographical
area. Among species from this area, the Italian treefrog,
Hyla intermedia, constitutes an interesting case study. It is
an endemic species distributed from the southern edge of
the Alpine massif (but see Dubey et al. 2006) to the tip of
Calabria and into Sicily, where it breeds in a variety of lentic
environments, mainly located at low altitudes (Lanza 1983).
Like several other Palaearctic treefrog species, the Italian
treefrog populations were long attributed to the European
species Hyla arborea and were only recently assigned to a
separate species, based on genetic studies (Nascetti et al.
1995). Despite a substantial morphological homogeneity
(but see Rosso et al. 2004), a recent survey of genetic variation
across the species’ range, using both the nuclear (allozymes)
and mitochondrial [polymerase chain reaction–restriction
fragment length polymorphism (PCR–RFLP)] markers,
showed the existence of two major groups of populations
within the Italian treefrog, one located north of the Northern
Apennines, the other located to the south (Canestrelli et al.
2007). In this study, we further investigate the population
genetic structure of the Italian treefrog through sequence
analysis of a mitochondrial cytochrome b gene fragment.
Our aim here is to study the historical and demographic
processes that have shaped the present patterns of genetic
diversity within the two previously identified population
groups. We are particularly interested in teasing apart
the contribution of prolonged stability vs. multiple refugia
scenarios in shaping the patterns of diversity within the
refugial ranges. Therefore, our main purpose here is to
elucidate whether the two lineages identified within the
4810 D . C A N E S T R E L L I , R . C I M M A R U TA and G . N A S C E T T I
Fig. 1 (A) Geographical location of the 27 populations sampled of the Italian treefrog, and results of the barrier analysis. Populations
are numbered as in Table 1 and presented as pie-diagrams, with slice size proportional to the frequency of the major haplotype groups
as identified by phylogenetic analyses. The barriers are numbered (roman numbers) in order of importance and their thickness is
proportional to the ratio of the observed FST value to the maximum FST value in the analysis, thus providing a basic index of their
significance. (B) Bayesian skyline plots showing the historical demographic trends for the three main mitochondrial lineages detected
within Hyla intermedia. Along the y-axis the estimated population sizes are expressed in units of Neτ, the product of the effective
population size per generation length. Solid lines are median estimates, whereas shaded areas represent confidence intervals.
Italian treefrog underwent a prolonged stability within the
respective inferred Quaternary ranges, or if a more complex
series of microevolutionary processes were involved.
Materials and methods
Sampling and laboratory procedures
The geographical location of sampling sites are shown in
Fig. 1 and listed in Table 1 together with sample size.
Details about sampling techniques were given in a previous
study (Canestrelli et al. 2007). For the present study, we
analysed a comprehensive number of 166 individuals
from 27 localities spanning the entire species range.
DNA was extracted following the standard cetyltrimethyl ammonium bromide protocol of Doyle & Doyle (1987).
Partial sequences of the mitochondrial cytochrome b gene
were obtained through PCR amplification. The generic
primers L14841 (Kocher et al. 1989) and MVZ16 (Moritz
et al. 1992) were used to carry out preliminary amplifications
and sequencing. Sequences obtained with these primers
were used to design the primers CytHYf (5′-ATCCAATTTGTCTTCATGATGAAA-3′) and CytHYr (5′-CCAAGGATATTTGGGGCAAATGTTG-3′), which were then employed
to screen all studied individuals. Amplifications were
performed in 50 µL tubes, containing MgCl2 (2.5 mm), the
reaction buffer (1×; Promega), the four dNTPs (0.2 mm each),
the two primers (0.2 µm each), the enzyme Taq polymerase
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
P H Y L O G E O G R A P H Y O F T H E I TA L I A N T R E E F R O G 4811
Locality
Latitude N
Longitude E
n
h
π
1 Corleone
2 Gibilmanna
3 Vendicari
4 Vandina
5 Gambarie
6 Pizzo
7 Fiumefreddo
8 Macchia Longa
9 Lago Remmo
10 San Giovanni Rotondo
11 Ostia
12 Latina
13 Roseto
14 San Lorenzo
15 Firenze
16 Bagno di Romagna
17 Magliano
18 Verucchio
19 Punta Alberete
20 Langhirano
21 Cremona
22 Novara
23 Torino
24 C. Ticino
25 Cavarzere
26 Bavaria
27 San Daniele
37°49′
37°55′
36°52′
38°11′
38°09′
38°44′
39°20′
39°15′
40°07′
41°43′
41°45′
41°28′
42°40′
43°34′
43°49′
43°50′
43°60′
43°59′
44°30′
44°37′
45°09′
45°29′
45°07′
46°01′
45°08′
45°34′
46°10′
13°18′
14°01′
15°08′
15°22′
15°41′
16°10′
16°07′
16°46′
15°47′
15°43′
12°20′
12°56′
13°59′
13°26′
11°28′
11°57′
12°05′
12°25′
12°16′
10°16′
10°01′
8°39′
7°34′
8°54′
12°04′
12°05′
13°05′
5
4
4
3
10
17
5
5
9
19
5
4
5
5
5
7
8
6
8
4
4
5
4
2
2
5
6
0.00
0.00
0.67
—
0.47
0.80
0.40
0.80
0.78
0.78
0.40
0.83
0.00
0.70
0.70
0.29
0.75
0.73
0.64
0.50
0.00
0.40
0.50
—
—
1.00
0.73
0.00000
0.00000
0.00110
—
0.00077
0.00247
0.00066
0.00234
0.00523
0.00542
0.00067
0.00166
0.00000
0.00166
0.00133
0.03360
0.07117
0.00155
0.00334
0.00166
0.00000
0.00066
0.00083
—
—
0.00670
0.00232
(2 U; Promega) and 2 µL of DNA template. PCR cycling
procedure was: 95 °C for 5 min followed by 35 cycles of
93 °C for 1 min, 52 °C for 45 s, 72 °C for 1 min 30 s and a single
final step at 72 °C for 10 min. Sequencing was carried out
using an ABI PRISM 377 DNA sequencer (PE Applied
Biosystems) following the ABI PRISM BigDye Terminator
Cycle Sequencing protocol. Both strands were sequenced
for all individuals analysed.
Data analysis
Sequences were aligned and checked by eye using the
software clustal_x (Thompson et al. 1997). Nucleotide
and amino-acid composition was determined using the
software mega 3.1 (Kumar et al. 2004). Haplotype (h) and
nucleotide (π) diversity (Nei 1987) were estimated for each
sampled population, using the software dnasp 4.0 (Rozas
et al. 2003).
Phylogenetic analyses were computed using the software
paup* 4.0b10 (Swofford 2003). The neighbour-joining
(NJ), maximum-parsimony (MP) and maximum-likelihood
(ML) methods were used to infer phylogenetic relationships
among the haplotypes found. Since tree-building methods
might not always be the most appropriate way to represent
genealogical relationships among haplotypes, as in cases
of shallow genetic divergence (Posada & Crandall 2001),
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
Table 1 Geographical location, sample
size (n) and estimate of genetic diversity
(for samples with n > 3), for the 27
populations studied of Hyla intermedia.
h , haplotype diversity; π, nucleotide
diversity
we also constructed a statistical parsimony network
(Templeton et al. 1992) by means of the software tcs 1.21
(Clement et al. 2000). A heuristic search was conducted for
MP and ML analyses, with tree-bisection–reconnection
branch swapping and 10 random addition sequence replicates. For MP, characters were unordered and equally
weighted. For ML and NJ analyses the best-fit model of
sequence evolution for our data was assessed using Akaike
information criterion (AIC) as implemented in the software
modeltest 3.7 (Posada & Crandall 1998). This analysis
supported the (Hasegawa–Kishino–Yano) HKY+Γ as the
best-fit substitution model for the data (Hasegawa et al.
1985), with gamma value = 0.648 and unequal base frequencies (A = 0.252, C = 0.292, G = 0.163, T = 0.293). The
robustness of topologies was assessed by the nonparametric
bootstrap procedure with 1000 pseudoreplicates. To test
the constancy of rates of molecular evolution among clades,
we compared likelihood scores obtained by enforcing and
supressing the molecular clock, using a likelihood-ratio-test
(Huelsenbeck & Crandall 1997).
The geographical structure of genetic variation was first
investigated with the method of Manni et al. (2004), as
implemented by the software barrier 2.2. This method
allows the identification of geographical areas where abrupt
changes in the genetic landscape occur. The analysis starts
connecting adjacent populations by means of a Delaunay
4812 D . C A N E S T R E L L I , R . C I M M A R U TA and G . N A S C E T T I
triangulation network (Delaunay 1934), upon which a
Voronoï tessellation is superimposed. An estimate of the
pairwise population genetic differentiation is then associated to each linked population pair, and the Monmonier
(1973) maximum difference algorithm is used to identify
genetic barriers. Genetic differentiation between populations was evaluated by estimating pairwise values of FST
with the software arlequin (Schneider et al. 2000). The
statistical significance of the estimates was assessed by
10 000 permutations. With barrier, the number of genetic
barriers to be computed is determined a priori by the user.
We continued adding boundaries until the last one starting
from a statistically significant FST value was included. As
an approximate measure of the significance of a computed
barrier, we used the ratio between the underlying FST value
and the maximum FST value in the analysis (Manni et al. 2004).
To partition the total genetic variance into its hierarchical
components among groups, among populations within
groups and within populations, we carried out an analysis
of molecular variance (amova; Excoffier et al. 1992) as
implemented in the arlequin software (Schneider et al.
2000; significances assessed by 1023 permutations). For
this analysis, groups of populations were defined according
to the geographical location of the genetic boundaries
identified by the Monmonier’s maximum differences
algorithm. Both the pairwise FST and amova tests were
performed incorporating Tamura-Nei (1993) genetic
distance, the best approximation available in arlequin
of the HKY, which, as stated above, is the best-fit model of
sequence evolution for our data.
To reconstruct backward in time the demographic history
of the main lineages detected, we used the coalescent-based
method called Bayesian skyline plot (BSP) (Drummond
et al. 2005). Compared with previous skyline plot methods
(Pybus et al. 2000; Strimmer & Pybus 2001), BSP has the
advantage of allowing inference of demographic history
from sampled sequence data, rather than from a previously
generated phylogeny. The uncertainty associated with
phylogenetic reconstruction is therefore accounted for by
this method, a feature particularly relevant when analysing
low variable data sets (Drummond et al. 2005). This
method uses a Markov chain Monte Carlo procedure to
sample the distribution of generalized skyline plots, given
the data and according to their posterior probabilities, and
combines these plots to generate estimate and credibility
intervals (confidence limits) for the effective population
size at every point backward until the time to the most
recent common ancestor (TMRCA) of the sampled sequences
is reached. The BSPs were calculated with the program
beast 1.4 (Drummond & Rambaut 2006) using the best-fit
substitution model for the data as estimated by modeltest
(HKY). This approach was also used to obtain estimates
and credibility intervals (as 95% HPD) of the TMRCA for
the main lineages. To this purpose, and to set a time scale
for the inferred demographic trends, a cytochrome b specific divergence rate of 3.6% per million years was assumed,
as derived by Babik et al. (2004) for European frogs based
on previous works of Beerli et al. (1996) and Veith et al.
(2003). Markov chain Monte Carlo tests were run for 20·106
steps and sampled every 1000 steps. Convergence of the
chains, burn-in and effective sample size of each parameter
were evaluated by means of the program tracer 1.3.
Finally, the possible occurrence of past demographic
expansions was also investigated by computing the statistics
FS (Fu 1997) and R2 (Ramos-Onsins & Rozas 2002), which in
a recent study have been shown as the most powerful test
statistics for detecting population growth (Ramos-Onsins &
Rozas 2002). The software dnasp 4.0 (Rozas et al. 2003) was
used to compute these statistics and their significance, evaluated by coalescent simulations (10 000 replicates).
Results
Sequence variation and population genetic diversity
A fragment of 608 base pairs of the cytochrome b gene was
obtained from all the 166 individuals of Hyla intermedia
analysed. Thirty-eight distinct haplotypes were found
(GenBank Accession nos EF531252–EF531289) defined
by 89 polymorphic sites, of which 68 were parsimony
informative. ML-corrected sequence divergence between
haplotypes ranged from 0.2 to 13.8%, whereas p-uncorrected
sequence divergence ranged from 0.2 to 10.4%. Seventyseven of the variable sites were in third position, three in
second position and nine in first position, with a comprehensive number of 10 amino-acid substitutions.
Estimates of intrapopulation genetic diversity are given
in Table 1 for each sampled population. Haplotype diversity (h) showed the largest possible variation, ranging
from 0 (at samples 1, 2, 13 and 21) to 1.00 (at the only sample
26), although the majority of populations showed h values
ranging from 0.40 to 0.80. A very wide variation (ranging
from 0 to 0.07117) was also observed for nucleotide
diversity (π). The highest values for this parameter were
observed at samples 16 and 17 (0.03360 and 0.07117,
respectively), the only two samples where the highly
divergent haplotypes from clades N and C+S were found
co-present (see the next section). Higher-than-average
values of π were also observed for samples 9 and 10
(0.00523 and 0.00542, respectively), the only two samples
where haplotypes from both clades C and S were found
(see the next section), as well as for sample 26 (0.00670).
Phylogenetic analyses, geographical distribution of
phylogroups and divergence time estimates
The NJ tree showing relationships between the haplotypes found, based on HKY+Γ genetic distances, is
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
P H Y L O G E O G R A P H Y O F T H E I TA L I A N T R E E F R O G 4813
Fig. 2 Phylogenetic relationships of the 38 haplotypes found among the 166 individuals of the Italian treefrog screened for sequence
variation at the mitochondrial cytochrome b gene. (A) Neighbour-joining tree based on HKY+Γ genetic distance. Bootstrap supports
> 50% over 1000 pseudoreplicates are given at nodes for neighbour-joining, maximum-parsimony and maximum-likelihood trees,
respectively. A sequence of Pseudacris regilla (from Ripplinger & Wagner 2004; GenBank Accession no. AY363197.1) was used as outgroup.
(B) Minimum-spanning haplotype networks. The diameter of the circles is proportional to haplotype frequency, and open dots represent
missing intermediate haplotypes.
presented in Fig. 2A. Essentially identical topologies
were also yielded by both the MP and ML methods (not
shown). MP analysis recovered four most parsimonious
trees, 194 steps in length (consistency index excluding
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
uninformative sites = 0.833; retention index = 0.981). The
log-likelihood score for the single ML tree was loglk =
–1696.71. Two main haplotype groups can be recognized,
whose geographical distribution fully matched that
4814 D . C A N E S T R E L L I , R . C I M M A R U TA and G . N A S C E T T I
previously reported by the preliminary mitochondrial
DNA (mtDNA) PCR–RFLP analysis of Canestrelli et al.
(2007). One group (clade N of Fig. 2) comprised 15 closely
related haplotypes and was found in all samples located
north of the Northern Apennines, whereas the second
group (clades C+S of Fig. 2) comprised 23 haplotypes and
was distributed from the Northern Apennines to Sicily
(see Fig. 1). The only sites where these major clades were
found co-present were Magliano and Bagno di Romagna
(samples 16 and 17), both located close to the northern
side of the Northern Apennines (see Fig. 1). These two
haplotype groups were supported by high bootstrap
values, and showed an average ML-corrected sequence
divergence of 12.5% (p-uncorrected distance being 9.7%).
Within the southern group, two main clades were found.
A first clade (C) was found geographically restricted to
the northern and central portions of the peninsula
(samples 9–17), whereas the second one (clade S) was
distributed in southern Italy and Sicily (samples 1–10).
Samples 9 and 10 shared haplotypes from both clades C
and S. These two clades presented an average ML-corrected
(as well as p-uncorrected) sequence divergence of 1.4%.
Within the group S, a subclade constituted by the haplotypes
hS6 and hS7 was observed, receiving moderate bootstrap
support and being geographically restricted to Sicily, a
region where no other haplotypes were found.
Using the 95% criterion suggested for the statistical parsimony method implemented by tcs, it was not possible
to connect all the haplotypes found into a single network.
Instead, two haplotype networks were generated (Fig. 2B).
One of these connected all the haplotypes found north of
the Northern Apennines, thus corresponding to the clade
N yielded by tree-building methods. It showed a star-like
structure centred on the haplotype hN1, which was also
the most frequent haplotype of this network (shared by
51.1% of the individuals analysed). The second network
connected all remaining haplotypes. Two main haplogroups were apparent in this network, fully corresponding
to the clades C and S yielded by tree-building methods,
and separated by five mutational steps of inferred haplotypes. The clade C also showed a clear star-like structure,
and the haplotype at the centre of this structure was hC1,
the most common haplotype within this group (shared by
61.7% of individuals analysed). A star-like shape was not
apparent for the clade S.
The null hypothesis of homogeneous evolutionary rates
across clades was not rejected by the likelihood ratio test
(–2 log ∆ = 27.12; not significant). According to the coalescentbased method employed and the divergence rate derived
by Babik et al. (2004), the divergence time between the
clade N and the clade C+S was estimated at 2.97·106 bp
years (95% HPD: 2.15·106–3.81·106 bp), whereas divergence time between clades C and S was estimated at
340 000 bp. (95% HPD: 157 000–542 000 bp)
Population genetic structure
A strong and significant pattern of differentiation was
observed over all populations (FST = 0.863; P < 0.01). Three
putative barriers to gene flow were inferred using the
method implemented by the software barrier (Fig. 1A).
The first barrier separated samples located in northern
Italy (samples 19–27) from all other samples. The second
barrier separated Sicilian samples (samples 1–4) from
those located in peninsular Italy. Finally, the third barrier
separated samples from central and southern Calabria
(samples 5–8) from those located farther north, thus
approximately corresponding to the Crati-Sibari plain. As
expected, high and significant values of FST were observed
in the majority of pairwise comparisons between populations located on different sides of these three barriers
(Table 2), whereas nonsignificant FST values were observed
in the majority of comparisons involving samples located
on the same side. An exception to this general pattern is
the sample from Magliano (sample 17), which showed
similar FST values when compared with samples from
both northern Italy and peninsular Italy.
The amova test was conducted using several grouping
options (Table 3). When all sampled populations were
considered and grouped according to their location relative
to the three genetic barriers detected, the among-group
component of variation was by far the most relevant
(85.95%). The same was also apparent when grouping
samples according to their position with respect to the
first barrier detected (i.e. the barrier separating samples
from northern Italy from those from the rest of the species
range). When restricting the analysis to samples south of
the Northern Apennines and grouping them according to
the two genetic barriers within this geographical area, the
largest portion of the overall variation (74.23%) was still
accounted for by the among-group level of variation,
although a non-negligible portion was also due to variation
within populations (22.37%). An almost identical pattern
of partition of the overall variation among the different
hierarchical levels was also observed when separating
populations of the southern group (samples 1–8) according
to their location with respect to the Messina strait.
Historical demography
The historical demographic trends were investigated
for clades N, C and S separately. The TMRCA (and 95%
HPD) were estimated at 130 000 (46 000–252 000), 50 000
(12 000–103 000) and 119 000 (28 000–227 000) bp for clades
N, C and S, respectively. According to the BSP (Fig. 1B),
after a phase of constant population size, the northern
group appears to have experienced a demographic expansion, commencing approximately 80 000 bp. During
this demographic expansion, the growth rate appeared to
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
1
—
0.00
0.39
0.19
0.90*
0.71*
0.91*
0.79*
0.66*
0.54*
0.97*
0.93*
1.00*
0.93*
0.94*
0.33*
0.41*
0.92*
0.98*
0.99*
1.00*
1.00*
1.00*
1.00*
0.99*
0.97*
0.99*
Population
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
—
0.33
0.11
0.90*
0.70*
0.90*
0.77*
0.63*
0.52*
0.96*
0.92*
1.00*
0.92*
0.93*
0.29*
0.37
0.91*
0.98*
0.99*
1.00*
1.00*
1.00*
1.00
0.99
0.96*
0.99*
2
—
0.00
0.86*
0.70*
0.81*
0.73*
0.63*
0.53*
0.92*
0.88*
0.96*
0.88*
0.89*
0.29*
0.37
0.89*
0.98*
0.99*
0.99*
0.99*
0.99*
0.99
0.98
0.96*
0.98*
3
—
0.86*
0.68*
0.81*
0.70*
0.59*
0.50*
0.93*
0.87*
0.96*
0.88*
0.89*
0.23
0.32*
0.88*
0.97*
0.99*
0.99*
0.99*
0.99*
0.99
0.98
0.95*
0.98*
4
—
0.06
0.60*
0.00
0.74*
0.63*
0.94*
0.92*
0.96*
0.92*
0.93*
0.50*
0.54*
0.92*
0.98*
0.99*
0.99*
0.99*
0.99*
0.99*
0.99*
0.98*
0.99*
5
—
0.33*
0.00
0.70*
0.62*
0.85*
0.83*
0.85*
0.84*
0.84*
0.56*
0.62*
0.84*
0.97*
0.98*
0.98*
0.98*
0.98*
0.98*
0.98*
0.97*
0.98*
6
—
0.30
0.62*
0.50*
0.94*
0.90*
0.97*
0.90*
0.91*
0.33*
0.40*
0.90*
0.98*
0.99*
0.99*
0.99*
0.99*
0.99*
0.99*
0.97*
0.99*
7
—
0.63*
0.55*
0.88*
0.85*
0.91*
0.85*
0.86*
0.35*
0.42*
0.86*
0.97*
0.98*
0.99*
0.99*
0.98*
0.98
0.98
0.96*
0.98*
8
—
0.01
0.08
0.06
0.08
0.18
0.09
0.05
0.41*
0.17
0.96*
0.96*
0.96*
0.96*
0.96*
0.95*
0.95*
0.94*
0.96*
9
—
0.11
0.09
0.11
0.20
0.12
0.13
0.52*
0.19*
0.95*
0.95*
0.96*
0.96*
0.95*
0.95*
0.95*
0.95*
0.95*
10
—
0.03
0.00
0.23
0.00
0.00
0.33
0.21
0.98*
0.99*
0.99*
0.99*
1.00*
0.99*
0.99
0.96*
0.98*
11
—
0.06
0.23
0.01
0.00
0.29
0.17
0.97*
0.98*
0.99*
0.99*
1.00*
0.98
0.98
0.96*
0.98*
12
—
0.37
0.00
0.00
0.33
0.26
0.98*
0.99*
1.00*
1.00*
1.00*
1.00*
0.99
0.97*
0.99*
13
—
0.25
0.00
0.32
0.24
0.97*
0.98*
0.99*
0.99*
0.99*
0.98*
0.98*
0.96*
0.98*
14
—
0.00
0.33*
0.00
0.97*
0.99*
0.99*
0.99*
0.99*
0.99*
0.98*
0.96*
0.98*
15
—
0.13
0.00
0.83*
0.78*
0.78*
0.80*
0.78*
0.72
0.73*
0.78*
0.81*
16
—
0.36*
0.40*
0.29
0.29
0.33
0.29
0.14
0.16
0.31
0.37*
17
—
0.97*
0.98*
0.99*
0.99*
0.99*
0.98*
0.98*
0.96*
0.98*
18
—
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.21
19
—
0.00
0.03
0.00
0.00
0.11
0.00
0.08
20
—
0.00
0.00
0.38
0.38
0.03
0.46
21
—
0.01
0.17
0.28
0.08
0.45*
22
—
0.11
0.21
0.04
0.42
23
—
0.00
0.00
0.35
24
—
0.00
0.38*
25
—
0.00
26
—
27
Table 2 Pairwise FST values among the 27 populations surveyed of the Italian treefrog. Populations are numbered as in Table 1. *P < 0.05 after 10 000 permutations. Values ≤ 0 were
in all cases set to 0
P H Y L O G E O G R A P H Y O F T H E I TA L I A N T R E E F R O G 4815
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
4816 D . C A N E S T R E L L I , R . C I M M A R U TA and G . N A S C E T T I
Table 3 Summary of results of the hierarchical analysis of molecular variance (amova) conducted using several grouping options
Grouping
Level of variation
Φ statistic*
% of variation*
[Northern Italy (19–27)] [Central Italy (9–18)]
[Southern Italy (5–8)] [Sicily (1–4)]
Among groups
Within groups
Within populations
Among groups
Within groups
Within populations
Among groups
Within groups
Within populations
Among groups
Within groups
Within populations
ΦCT = 0.860
Φ SC = 0.252
Φ ST = 0.895
Φ CT = 0.879
ΦSC = 0.480
Φ ST = 0.937
ΦCT = 0.742
ΦSC = 0.132
Φ ST = 0.776
Φ CT = 0.739
Φ SC = 0.134
Φ ST = 0.774
85.95
3.54
10.50
87.85
5.83
6.31
74.23
3.40
22.37
73.93
3.50
22.57
[Northern Italy (19–27)] [Peninsular Italy
and Sicily (1–18)]
[Central Italy(9–18)] [Southern Italy (5–8)]
[Sicily (1–4)]
[Southern Italy (5–8)][Sicily (1–4)]
*All P << 0.01.
increase until about 20 000 bp, when it began slowing
down. A somewhat similar trend was also observed for
the central clade, for which the expansion phase was still
in progress at the estimated TMRCA. Also with this clade,
the growth rate slowing-down phase would have begun
about 20 000 bp. Finally, the southern group appears to
have experienced a more prolonged phase of demographic
stability, followed by a recent expansion which started
approximately 30 000 bp. In order to assess whether the
strong subdivision identified at the level of the Messina
Strait (see the previous paragraph) could have affected
the inferred historical demographic trend for the southern
group, we also reran the analysis for this group after
removal of sequences drawn from the Sicilian samples
(A. Drummond, personal suggestion). However, this did
not yield appreciable modifications of the inferred
demographic trend.
Finally, a significant population growth was also inferred
for both clades N and C by the statistic FS (FS = –8.64 and
–9.81, respectively; both P < 0.01) and R2 (R2 = 0.043 and
0.037, respectively; both P < 0.01). By contrast, these statistics
did not support an inference of population expansion for
the clade S, neither when all haplotypes were included in
the analyses (FS = –2.17, R2 = 0.101; both P = not significant)
nor excluding the Sicilian haplotypes (FS = –3.07, R2 = 0.080;
both P = not significant).
Discussion
In a previous survey of genetic variation within the Italian
treefrog, Canestrelli et al. (2007) showed the existence of
two main groups of populations. The ranges of these two
groups extended on alternative sides of the Northern
Apennines. The overall pattern observed also suggested
a pre-Pleistocenic origin of the southern group through
colonization from northern areas, followed by allopatric
differentiation in distinct Quaternary ranges located at
opposite sides of the Northern Apennines and a subsequent
secondary contact. In the present study, the genetic
variation of the Italian treefrog has been further investigated, in order to shed light on the evolutionary and
historical demographic processes that have shaped the
present patterns of variation within the two major groups,
and to contribute to recent discussions about factors
shaping the patterns of diversity within Quaternary refugial
ranges in the Mediterranean peninsulas of Europe.
As expected, the data presented here are in full agreement with previous findings, indicating the Northern
Apennines as the site of the deepest phylogeographic
break within the species range. Since the possible scenario
for the origin of the two major groups has been widely
examined elsewhere (Canestrelli et al. 2007), it will be
discussed no further here.
Based on both phylogenetic and barrier analysis (see
Fig. 1A and 2), all samples collected throughout the entire
Padanovenezian plain can be assigned to the northern
group of populations. Within this geographical area, no
phylogeographic discontinuities were identified, and the
overall pattern of differentiation indicated this northern
group as substantially homogeneous. The historical demographic reconstruction carried out by means of the BSP
suggested a trend of demographic stability from the onset
of the Late Pleistocene until about 80 000 bp, when the
population starts growing. The demographic expansion
phase appears to have lasted until very recently, although
the growth rate began decreasing approximately 20 000
bp. A necessary premise to further discussion is that these
time estimates, as well as the following ones, should be
taken with caution and handled as approximate time
frames for the inferred events, for at least two reasons:
they rely on a substitution rate calibrated not directly for
the species under study, and, most importantly, several
lines of evidence suggest the need for caution when dating
historical events based on genetic divergence (Ayala 1997,
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
P H Y L O G E O G R A P H Y O F T H E I TA L I A N T R E E F R O G 4817
1999; Gibbons 1998; Welch & Bromham 2005; Ho & Larson
2006). However, it is also worth noting that in the present
case, the estimated timescale for the inferred historical
events, fits well with the known palaeogeographic evolution of the underlying geographical area. For the northern
group, the estimated time lapse for the main expansion
event (i.e. 80 000–20 000 bp) falls within the last major glacial phase. During this phase, the Adriatic Sea coastline
moved several hundreds kilometres southeast of its present
location, due to marine regression, leading to a considerable
widening of the Po plain (Corregiari et al. 1996; Amorosi
et al. 1999). According to Amorosi et al. (2004; see also
references therein) the transition to this phase was also
accompanied by the ultimate establishment of a vast
Pleistocene alluvial plain environment in this geographical
area. Since at present, the Italian treefrog populations are
mainly distributed in lowland habitats (with more than
90% of populations being located below 500 m above sea
level, e.g. Mazzotti et al. 1999), it appears highly plausible
that the species could have been favoured by such widening
of alluvial plains during the glacial phase, and that it
reached its demographic maximum during this phase,
when the area of available habitat was also at its maximum.
The evolutionary history of Italian treefrog populations
south of the Northern Apennines seems very different
from that of the northern group. The most ancient event
that can be inferred dated back to about 340 000 bp (95%
HPD: 157 000–542 000 bp), and produced the split between
two mtDNA lineages (clades C and S of Fig. 2), one distributed across the northern and central portions of the
peninsula, the other geographically restricted to southern
Italy and Sicily. The geographical distribution of the two
clades and the results of the barrier analysis (Fig. 1, see
barrier III) suggest the Crati-Sibari plain as the possible
source of such phylogeographic discontinuity. A possible
historical scenario for the origin of the two clades would
imply an allopatric fragmentation during the Middle
Pleistocene, with the Crati-Sibari plain acting as an extrinsic
barrier to gene flow. Palaeogeographic reconstructions
(Martini et al. 2001; Cucci 2004), palaeontological and
comparative biogeographic evidence (Pignatti 1984; Caloi
et al. 1989 and references therein; Bernasconi et al. 1997)
and genetic studies also concerning other amphibian
species (Santucci et al. 1996; Canestrelli et al. 2006), all
suggest the Crati-Sibari plain as a main historical barrier
to dispersal along the north–south axis of the peninsula
for many taxa, thus supporting the above scenario for the
Italian treefrog. Indeed, the Crati-Sibari plain corresponds
to a major Calabrian graben, affected by intense tectonic
activity throughout the Plio–Pleistocene when it was
repeatedly marine-flooded following the glacio-eustatic
sea-level fluctuations, the plain gradually emerging during the middle–late Pleistocene. A secondary contact between
the two lineages would have followed, as suggested by
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
their sintopy at the geographically intermediate samples
of Lago Remmo and San Giovanni Rotondo (samples 9
and 10; see Fig. 1), and the higher than average nucleotide
diversity of these populations.
The historical demographic reconstruction, as inferred
through the BSPs (Fig. 1), suggested a recent population
expansion for both clades C and S, thus also supporting
an inference of recent secondary contact between the
two lineages. The general shape of the past demographic
trend is however, different for the two clades. For clade
C, a shallower population history can be inferred (the
TMRCA being approximately 50 000 years bp) compared
with both clades N and S. Together with the star-shaped
arrangement of the haplotype network, this suggests
that a population bottleneck could have preceded the
expansion event (Slatkin & Hudson 1991; Avise 2000). The
historical trend of the expansion event for the clade C
appears similar to that observed for the clade N, suggesting
that correlated historical events could have driven the
demographic expansion trends for both clades. Although
less impressive than for the Padanovenezian plain, also
in the Tyrrhenian side of the central and north-central
portion of peninsular Italy, the width of lowland habitats
underwent wide — and concordant — fluctuations associated
with eustatic sea-level oscillations (Tortora et al. 2001;
Lambeck et al. 2004; Ferranti et al. 2006; Amorosi, personal
communication). It appears therefore plausible that also
on the central and north-central Tyrrhenian side of the
peninsula, the treefrogs could have been favoured by the
widening of lowland habitats during the last major glacial
phase. In this respect, the demographic trend observed in
southern Italy appears also of interest (clade S, see Fig. 1).
In this geographical area, according to the BSP, the demographic expansion appears of lesser magnitude than for
the northernmost groups and linked to the end of the last
glaciation, whereas during the previous climatic phases
of the Late Pleistocene the species appears to have undergone a prolonged demographic stability. Interestingly,
coastline oscillations associated with the last glacial cycle
appeared much less extensive in this geographical area,
particularly in Calabria, than in the northernmost portion
of the peninsula (Tortora et al. 2001; Lambeck et al. 2004;
Ferranti et al. 2006). In this geographical area, the moderate
population growth suggested by the BSP may have been
linked to the limited altitudinal increment of suitable
habitat which has followed the interglacial amelioration
of climatic conditions at higher altitudes. Nevertheless,
it is also worth noting that the inference of population
expansion for the southern clade, based on the BSP,
appears not supported neither by the nonstar-like shape
of the haplotype network, nor by nonsignificant values
of the test statistics FS (Fu 1997) and R2 (Ramos-Onsins &
Rozas 2002). This discordance could at least in part be due
to the strong subdivision observed among the southern
4818 D . C A N E S T R E L L I , R . C I M M A R U TA and G . N A S C E T T I
populations (see Table 2), even excluding the Sicilian
ones. In fact, strong population subdivision has been
shown to lower the power in detecting past population
growths based on both neutrality tests and the shape of
gene genealogies (see Ray et al. 2003 and references therein).
However, its specific effects on BSP reconstructions have
not been addressed yet. Based on the data at hand, we are
therefore presently unable to conclusively tease apart
between an inference of moderate population growth in
the recent past (as suggested by the BSP) or the absence of
this (as suggested by the shape of the haplotype network
and the statistics FS and R2) for the southern clade.
Among populations of the southern group, all sharing
haplotypes from clade S, a strong phylogeographic discontinuity was observed at the level of the Messina Strait
(see Fig. 1, barrier II), accounting for the largest portion
(73.9%) of the overall variation observed among this
group of samples (Table 3). Following Quaternary glacioeustatic oscillations of the sea level, the Messina Strait
repeatedly underwent phases of complete or partial
emersion, particularly during the Late-Middle and Late
Pleistocene, allowing for faunal exchanges between
southern Calabria and Sicily (Bonfiglio et al. 2002 and
references therein). As has been suggested for several
other species (e.g. Bonfiglio et al. 2002; Podnar et al. 2005),
and judging by the very low divergence between Sicilian
haplotypes (hS6, hS7) and all others observed in southern
Italy, it is thus likely that the Italian treefrog reached Sicily
from peninsular Italy by jump dispersal across a Late
Pleistocene land bridge.
The data presented and the discussion carried out so far
indicate the existence of four main groups of populations,
which have survived the last glacial–interglacial cycles in
at least three distinct refugia (i.e. northern, central and
southern Italy). The hypotheses presented to explain the
observed population structure and the location of major
phylogeographic breaks also appear well supported by
external evidence, such as palaeogeographic reconstructions and lines of concordance with other taxa. Interestingly,
such a geographical structure was not observed at the
nuclear allozyme markers (Canestrelli et al. 2007), which
depicts the Italian treefrog populations from central and
southern Italy and Sicily as a substantially homogeneous
group of populations. To explain such a discrepancy, at
least two kinds of hypotheses can be made: (i) that the
allozyme markers failed to reach an appreciable differentiation between the two population groups, because of
balancing selection (see also Hare & Avise 1998), or lack of
allozyme variation in the ancestral population and/or
insufficient time elapsed between divergence and secondary
contact to allow the allele frequencies to diverge appreciably; (ii) a male-biased pattern of dispersal has led to
extensive admixture at the nuclear genome following
secondary contacts, while evidence of the historical sep-
aration between the two lineages have been retained at
the maternally inherited mitochondrial marker. To date,
based on the data at hand, we cannot confidently choose
between the different hypotheses. However, if the latter
were the case, one could expect to see some kind of clinal
variation at the previously differentiated nuclear loci. In
order to gain better insight into the factors underlying
the observed discordance, we are presently planning a
scrutiny of the patterns of genetic variation at both autosomal and Y-linked microsatellite loci (Arens et al. 2000;
Berset-Brändli et al. 2006).
Several studies have recently suggested the occurrence
of multiple refugia as a possible explanation for the
observed patterns of divergence within the Italian peninsula (see Introduction). However, only a few have been
conducted with a sampling scheme appropriate to unravel
the location of major phylogeographic discontinuities
along this peninsula and their roles in shaping patterns
of differentiation and populations’ genetic diversity
(Santucci et al. 1996; Podnar et al. 2005; Canestrelli et al.
2006), so that in-depth cross-taxa comparisons for this
geographical area still await the accumulation of more
phylogeographic data. Nevertheless, the data at hand to
date indicates extensive type III of phylogeographic
concordance (Avise 1996), thus suggesting that shared
historical biogeographic factors have contributed to shaping
the intraspecific patterns of differentiation. The concordance is, however, not complete. For instance, within the
population genetic structure of the Italian treefrog, there
is no evidence for a historical role of the Catanzaro palaeostrait (south-central Calabria) as a barrier to dispersal,
whereas it was identified as a source of a major phylogeographic break and as a secondary contact zone for both
Bombina pachypus (Canestrelli et al. 2006) and Rana lessonae
(Santucci et al. 1996). This latter species in particular is
often syntopic with, and otherwise shows a close phylogeographic concordance with, the Italian treefrog (Santucci
et al. 1996, 2000). These findings underline the role of
species-specific natural history traits and/or stochastic
factors in shaping population structures and species’
evolutionary histories. However, due to the limited
number of detailed phylogeographies to date available
for peninsular Italy, we cannot exclude that a number of
shared evolutionary histories exist among taxa — as
observed for the Iberian Peninsula (Gómez & Lunt 2006)
— rather than a scenario in which either all or none the
phylogeographies are concordant. This clearly indicates
the need for future studies, based on the widest possible
taxonomic range of species.
Conclusions
The evolutionary history of the Italian treefrog presents some
peculiar features. Evidences for a prolonged demographic
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
P H Y L O G E O G R A P H Y O F T H E I TA L I A N T R E E F R O G 4819
stability during the glacial phase followed by interglacial
demographic expansion were only observed for the
southern group of populations. By contrast, the northern
and central groups likely expanded earlier, following
glaciation-induced sea-level dropping and the consequent
opening of lowland habitats. Also, the Italian treefrog
populations do not show the pattern of so-called ‘southern
richness, northern purity’ often observed in taxa from
southern European peninsulas, also including some anuran
amphibians from peninsular Italy (see citations above).
Nevertheless, (i) the strong phylogeographic structure
observed, (ii) the lines of concordance with other species,
and (iii) the fact that, in line with what was previously
found for several of these species, a higher genetic (nucleotide)
diversity was observed in populations in which evidence
of secondary contacts were seen, all serve to strengthen
the idea that a diversity of microevolutionary processes,
rather than prolonged population stability, mainly accounts
for the patterns of diversity within the Italian peninsula.
This finding, when compared to evidence from the other
Mediterranean peninsulas (e.g. Gómez & Lunt 2006;
Schmitt et al. 2006), appears as an emerging common
feature for these Western Palearctic biodiversity hotspots.
Acknowledgements
We are grateful to Francesca Zangari and Daniele Porretta for
useful discussions and suggestions, to Angus Davison and two
anonymous reviewers, whose suggestions greatly improved an
early version of the manuscript, to Alexei Drummond for kind
advise in the use of Beast, to Claudio Bagnoli, Paola Bellini,
Cristina Giacoma, Paolo Laghi, Christian Pastorelli, Alessandra
Rosso and Giancarlo Tedaldi for their help with sampling and/
or manuscript preparation and to Mark Eltelton who reviewed
the English. This work was funded by MIUR (Italian Ministry of
University and of Scientific and Technological Research).
References
Alexandrino J, Froufe E, Arntzen JW, Ferrand N (2000) Genetic
subdivision, glacial refugia and postglacial recolonization in
the golden-striped salamander, Chioglossa lusitanica (Amphibia:
Urodela). Molecular Ecology, 9, 771–781.
Amorosi A, Cotalongo ML, Fusco F (1999) Gladio-eustatic control
of continental-shallow marine cyclicity from Late Quaternary
deposits of the southeastern Po plain, northern Italy. Quaternary
Research, 52, 1–13.
Amorosi A, Colalongo ML, Fiornini F et al. (2004) Palaeogeographic and palaeoclimatic evolution of the Po plain from
150-ky core records. Global and Planetary Change, 40, 55–78.
Araujo MB, Thuiller W, Pearson RG (2006) Climate warming and
the decline of amphibians and reptiles in Europe. Journal of
Biogeography, 33, 1712–1728.
Arens P, Van’t Westende W, Butger R, Smulders MJM, Vosman B
(2000) Microsatellite markers for the European tree frog Hyla
arborea. Molecular Ecology, 9, 1919–1952.
Avise JC (1996) Toward a regional conservation genetics perspective:
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
phylogeography of faunas in the southeastern United States.
In: Conservation Genetics: Case Histories from Nature (eds Avise JC,
Hamrick JL). Chapman & Hall, New York.
Avise JC (2000) Phylogeography: The History and Formation of
Species. Harvard University Press, Cambridge, Massachusetts.
Ayala FJ (1997) Vagaries of the molecular clock. Proceeding of the
National Academy of Sciences, USA, 94, 7776–7783.
Ayala FJ (1999) Molecular clock mirages. Bioassays, 21, 71–75.
Babik W, Branicki W, Sandera M et al. (2004) Mitochondrial
phylogeography of the moor frog, Rana arvalis. Molecular
Ecology, 13, 1469–1480.
Beerli P, Hotz H, Uzzell T (1996) Geologically dated sea barriers
calibrate a protein clock for aegean water frogs. Evolution, 50,
1676–1687.
Bella JL, Serrano L, Orellana J, Maso PL (2007) The origin of
Chorthippus parallelus hybrid zone: chromosomal evidence of
multiple refugia for Iberian populations. Journal of Evolutionary
Biology, 20, 568–576.
Bernasconi MP, Corselli C, Carobene L (1997) A bank of the
scleractinian coral Cladocora caespitosa in the Pleistocene of the
Crati valley (Calabria, Southern Italy): growth versus environmental conditions. Bollettino della Società Paleontologica Italiana,
36, 53–61.
Berset-Brändli L, Jaquiéry J, Dubey S, Perrin N (2006) A sex-specific
marker reveals male heterogamety in European treefrogs.
Molecular Biology and Evolution, 23, 1104–1106.
Böhme MU, Fritz U, Kotenko TD et al. (2007) Phylogeography and
cryptic variation within the Lacerta viridis complex (Lacertidae,
Reptilia). Zoologica Scripta, 36, 119–131.
Bonfiglio L, Mangano G, Marra AC, Masini F, Pavia M, Petruso
D (2002) Pleistocene Calabrian and Sicilian bioprovinces.
Geobios, 35, 29–39.
Caloi L, Malatesta A, Palombo RM (1989) Biogeografia della
Calabria meridionale durante il Quaternario. Atti Accademia
Peloritana dei Pericolanti classe I, Scienze Matematiche Fisiche e
Naturali, 67, 261–278.
Canestrelli D (2006) Comparative phylogeography of Italian amphibians,
with implications for conservation. PhD Thesis. Tuscia University,
Viterbo, Italy.
Canestrelli D, Cimmaruta R, Costantini V, Nascetti G (2006)
Genetic diversity and phylogeography of the Apennine yellowbellied toad Bombina pachypus, with implications for conservation.
Molecular Ecology, 15, 3741–3754.
Canestrelli D, Verardi A, Nascetti G (2007) Genetic differentiation
and history of populations of the Italian treefrog: lack of
concordance between mitochondrial and nuclear genes. Genetica,
130, 241–255.
Carstens BC, Brunsfeld SJ, Demboski JR, Good JD, Sullivan J
(2005) Investigating the evolutionary history of the Pacific
Northwest mesic forest ecosystem: hypothesis testing within a
comparative phylogeographic framework. Evolution, 59, 1639–
1652.
Clement M, Posada D, Crandall KA (2000) tcs: a computer
program to estimate gene genealogies. Molecular Ecology, 9,
1657–1660.
Corregiari A, Roveri M, Trincardi F (1996) Late Pleistocene and
Holocene evolution of the North Adriatic Sea. Il Quaternario, 9,
697–704.
Cucci L (2004) Raised marine terraces in the northern Calabrian
Arc (Southern Italy): a ~600 Kyr-long geological record of
regional uplift. Annals of Geophysics, 47, 1391–1406.
4820 D . C A N E S T R E L L I , R . C I M M A R U TA and G . N A S C E T T I
Deffontaine V, Libois R, Kotlik P et al. (2005) Beyond the Mediterranean peninsulas: evidence of central European glacial
refugia for a temperate forest mammal species, the bank vole
(Clethrionomys glareolus). Molecular Ecology, 14, 1727–1739.
Delaunay B (1934) Sur la sphère vide. Bulletin of the Academy of
Sciences of the USSR, 7, 793–800.
Doyle JJ, Doyle JL (1987) A rapid DNA isolation procedure for small
quantities of fresh leaf tissue. Phytochemistry Bulletin, 19, 11–15.
Drummond AJ, Rambaut A (2006) BEAST, Version 1.4. University
of Oxford, Oxford, UK.
Drummond AJ, Rambaut A, Shapiro B, Pybus OG (2005) Bayesian
coalescent inference of past population dynamics from molecular
sequences. Molecular Biology Evolution, 22, 1185–1192.
Dubey S, Ursenbacher S, Fumagalli L (2006) Origin of tree frog
(Hyla spp.) populations in western Switzerland. Revue Suisse de
Zoologie, 113, 879–887.
Excoffier L, Smouse PE, Quattro JM (1992) Analysis of molecular
variance inferred from metric distances among DNA haplotypes:
Application to human mitochondrial DNA restriction data.
Genetics, 131, 479–491.
Ferranti L, Antonioli F, Mauz B et al. (2006) Markers of the last
interglacial sea-level high stand along the coast of Italy: tectonic
implications. Quaternary International, 145–146, 30–54.
Fu YX (1997) Statistical tests of neutrality of mutations against
population growth, hitchhiking and background selection.
Genetics, 147, 915–925.
Gibbons A (1998) Calibrating the mitochondrial DNA clock.
Science, 279, 38–39.
Gómez A, Lunt DH (2006) Refugia within refugia: patterns of
phylogeographic concordance in the Iberian Peninsula. In:
Phylogeography in Southern European Refugia: Evolutionary
Perspectives on the Origins and Conservation of European Biodiversity
(eds Weiss S, Ferrand N). Springer Verlag, Dordrecht, The
Netherlands.
Hare MP, Avise JC (1998) Population structure in the American
oyster as inferred by nuclear gene genealogies. Molecular Biology
and Evolution, 15, 119–128.
Hasegawa M, Kishino H, Yano T (1985) Dating of the human-ape
splitting by a molecular clock of mitochondrial DNA. Journal
of Molecular Evolution, 22, 160–174.
Hewitt GM (1996) Some genetic consequences of ice ages and
their role in divergence and speciation. Biological Journal of the
Linnean Society, 58, 247–276.
Hewitt GM (1999) Postglacial re-colonization of European biota.
Biological Journal of the Linnean Society, 68, 87–112.
Hewitt GM (2000) The genetic legacy of the Quaternary ice ages.
Nature, 405, 907–913.
Hewitt GM (2004a) Genetic consequences of climatic oscillations
in the Quaternary. Philosophical Transactions of the Royal Society
of London. Series B, Biological Sciences, 359, 183–195.
Hewitt GM (2004b) The structure of biodiversity — insights from
molecular phylogeography. Frontiers in Zoology, 1, 4.
Ho SY, Larson G (2006) Molecular clocks: when times are achangin’. Trends in Genetics, 22, 79–83.
Huelsenbeck JP, Crandall KA (1997) Phylogeny estimation and
hypothesis testing using maximum likelihood. Annual Reviews
in Ecology and Systematics, 28, 437–466.
Kocher TD, Thomas WK, Meyer A et al. (1989) Dynamics of
mitochondrial DNA evolution in animals: amplification and
sequencing with conserved primers. Proceeding of the National
Academy of Sciences, USA, 86, 6196–6200.
Kumar S, Tamura K, Nei M (2004) mega3: Integrated software for
molecular evolutionary genetics analysis and sequence alignment. Briefings in Bioinformatics, 5, 150–163.
Lambeck K, Antonioli F, Purcell A, Silenzi S (2004) Sea level
change along the Italian coast for the past 10 000 yrs. Quaternary
Science Reviews, 23, 1567–1598.
Lanza B (1983) Anfibi e rettili. In: Guide Per Il Riconoscimento Delle
Specie Animali Delle Acque Interne Italiane. CNR, Italy.
Manni F, Guérard E, Heyer E (2004) Geographic patterns of (genetic,
morphologic, linguistic) variation: how barriers can be detected
by ‘Monmonier’s algorithm’. Human Biology, 76, 173–190.
Martìnez-Solano I, Teixeira J, Buckley D, Garcìa-Parìs M (2006)
Mitochondrial DNA phylogeography of Lissotriton boscai
(Caudata, Salamandridae): evidence for old, multiple refugia
in an Iberian endemic. Molecular Ecology, 15, 3375–3388.
Martini IP, Sagri M, Colella A (2001) Neogene-Quaternary basins
of the inner Apennines and Calabria arc. In: Anatomy of an
Orogen: The Apennines and Adjacent Mediterranean Basins (eds
Martini IP, Vai GB). Kluwer Academic Publishers, Dordrecht,
The Netherlands.
Mazzotti S, Caramori G, Barbieri C (1999) Atlante degli Anfibi e
dei Rettili dell’Emilia-Romagna. Quaderni della Stazione Ecologica
del Museo Civico di Storia Naturale di Ferrara, 12, 1–121.
Monmonier MS (1973) Maximum–difference barriers: an alternative
numerical regionalization method. Geographical Analysis, 3,
245–261.
Moritz C, Schneider CJ, Wake DB (1992) Evolutionary relationships within Ensatina eschscholtzi complex confirm the ring
species interpretation. Systematic Zoology, 41, 273–291.
Myers N, Mittermeier RA, Mittermeier CG, Fonseca GAB, Kent J
(2000) Biodiversity hotspots for conservation priorities. Nature,
403, 853–858.
Nascetti G, Lanza B, Bullini L (1995) Genetic data for the specific
status of the Italian treefrog (Amphibia: Anura: Hylidae).
Amphibia-Reptilia, 16, 215–227.
Nei M (1987) Molecular Evolutionary Genetics. Columbia University
Press, New York.
Nieberding C, Libois R, Douady CJ, Morand S, Michaux JR (2005)
Phylogeography of a Nematode (Heligmosomoides polygyrus) in
the Western Palearctic region: persistence of northern cryptic
populations during ice ages? Molecular Ecology, 14, 765–779.
Paulo OS, Jordan WC, Bruford MW, Nichols RA (2002) Using
nested clade analysis to assess the history of colonization and
the persistence of populations of an Iberian lizard. Molecular
Ecology, 11, 809–819.
Petit RJ, Aguinagalde I, de Beaulieu JL et al. (2003) Glacial refugia:
hotspots but not melting pots of genetic diversity. Science, 300,
1563–1565.
Pignatti S (1984) La flora dell’Appennino meridionale: distribuzione attuale ed ipotesi sull’origine. Biogeographia, 10, 89 –100.
Podnar M, Mayer W, Tvrtkovi N (2005) Phylogeography of the
Italian wall lizard, Podarcis sicula, as revealed by mitochondrial
DNA sequences. Molecular Ecology, 14, 575–588.
Posada D, Crandall KA (1998) modeltest: testing the model of
DNA substitution. Bioinformatics, 14, 817–818.
Posada D, Crandall KA (2001) Intraspecific gene genealogies:
trees grafting into networks. Trends in Ecology & Evolution, 16,
37– 45.
Pybus OG, Rambaut A, Harvey PH (2000) An integrated framework
for the inference of viral population history from reconstructed
genealogies. Genetics, 155, 1429–1437.
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
P H Y L O G E O G R A P H Y O F T H E I TA L I A N T R E E F R O G 4821
Ramos-Onsins SE, Rozas J (2002) Statistical properties of new
neutrality tests against population growth. Molecular Biology
and Evolution, 19, 2092–2100.
Ray N, Currat M, Excoffier L (2003) Intra-deme molecular diversity
in spatially expanding populations. Molecular Biology and
Evolution, 20, 76–86.
Ripplinger J, Wagner RS (2004) Phylogeography of northern
populations of the Pacific chorus frog, Pseudacris regilla.
Northwestern Naturalist, 85, 118–125.
Rosso A, Castellano S, Giacoma C (2004) Ecogeographic analysis
of morphological and life-history variation in the Italian
treefrog. Evolutionary Ecology, 18, 303–321.
Rozas J, Sànchez-DelBarrio JC, Messeguer X, Rozas R (2003)
dnasp, DNA polymorphism analyses by the coalescent and
other methods. Bioinformatics, 19, 2496–2497.
Santucci F, Nascetti G, Bullini L (1996) Hybrid zones between two
genetically differentiated forms of the pond frog Rana lessonae
in southern Italy. Journal of Evolutionary Biology, 9, 429–450.
Santucci F, Andreani P, Nascetti G, Bullini L (2000) Genetic
identification and geographic distribution of the water frogs
Rana lessonae and Rana esculenta in Italy, Sicily and Corsica
(Amphibia, Anura). Museo Regionale Scienze Naturali di Torino,
2000, 359–367.
Sanz N, García-Marín JL, Pla C (2000) Divergence of brown trout
(Salmo trutta) within glacial refugia. Canadian Journal of Fisheries
and Aquatic Sciences, 57, 2201–2210.
Schmitt T, Habel JC, Zimmermann M, Müller P (2006) Genetic
differentiation of the marbled white butterfly, Melanargia
galathea, accounts for glacial distribution patterns and postglacial
range expansion in southeastern Europe. Molecular Ecology,
15, 1889–1901.
Schneider S, Roessli D, Excoffier L (2000) ARLEQUIN Version 2.0:
A Software for Population Genetic Data Analysis. Univiversity of
Geneva Genetic and Biometric Laboratory, Geneva, Switzerland.
Slatkin M, Hudson RR (1991) Pairwise comparisons of mitochondrial DNA sequences in stable and exponentially growing
populations. Genetics, 129, 555–562.
Steele CA, Storfer A (2006) Phylogeographic incongruence of
codistributed amphibian species based on small differences in
geographic distribution. Molecular Phylogenetics and Evolution,
doi: 10.1016/j.ympev.2006.10.010.
Stewart JR, Lister AM (2001) Cryptic northern refugia and the
origins of the modern biota. Trends in Ecology & Evolution, 16,
608–613.
Strimmer K, Pybus OG (2001) Exploring the demographic history
of DNA sequences using the generalized skyline plot. Molecular
Biology and Evolution, 18, 2298–2305.
Sullivan J, Arellano E, Rogers DS (2000) Comparative phylogeography of mesoamerican highland rodents: concerted versus
independent responses to past climatic fluctuations. American
Naturalist, 155, 755–768.
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd
Swofford DL (2003) PAUP*. Phylogenetic Analysis Using Parsimony
(*and Other Methods), Version 4. Sinauer Associates, Sunderland,
Massachusetts.
Taberlet P, Fumagalli L, Wust-Saucy A-G, Cosson J-F (1998)
Comparative phylogeography and postglacial colonization
routes in Europe. Molecular Ecology, 7, 453–464.
Tamura K, Nei M (1993) Estimation of the number of nucleotide
substitutions in the control region of mitochondrial DNA in
humans and chimpanzees. Molecular Biology and Evolution,
10, 512–526.
Templeton AR, Crandall KA, Sing CF (1992) A cladistic analysis
of phenotypic associations with haplotypes inferred from
restriction endonuclease mapping and DNA sequence data.
III. Cladogram estimation. Genetics, 132, 619–633.
Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins
DG (1997) The clustal x Windows interface: flexible strategies
for multiple sequence alignment aided by quality analysis
tools. Nucleic Acids Research, 25, 4876– 4882.
Tortora P, Bellotti P, Valeri P (2001) Late Pleistocene and
Holocene deposition along the coasts and continental shelves.
In: Anatomy of an Orogen: the Apennines and Adjacent Mediterranean
Basins (eds Martini IP, Vai GB). Kluwer Academic Publishers,
Dordrecht, The Netherlands.
Ursenbacher S, Conelli A, Golay P et al. (2006) Phylogeography
of the asp viper (Vipera aspis) inferred from mitochondrial
DNA sequence data: Evidence for multiple Mediterranean
refugial areas. Molecular Phylogenetics and Evolution, 38, 546–
552.
Veith M, Kosuch J, Vences M (2003) Climatic oscillations triggered
post-Messinian speciation of Western Palearctic brown frogs
(Amphibia, Ranidae). Molecular Phylogenetics and Evolution, 26,
310–327.
Weiss S, Ferrand N (eds) (2006) Phylogeography in Southern
European Refugia: Evolutionary Perspectives on the Origins and
Conservation of European Biodiversity. Springer Verlag, Dordrecht,
The Netherlands.
Welch JJ, Bromham L (2005) Molecular dating when rates vary.
Trends in Ecology & Evolution, 20, 320–327.
D. Canestrelli is a postdoctoral research associate, interested in the
evolutionary and ecological factors underlying the geographical
patterns of diversity, as well as biodiversity conservation. R.
Cimmaruta is a senior researcher, mainly interested in biodiversity
conservation and relationships between environmental stress and
population genetic structure. G. Nascetti is Professor in Ecology
and head of the Department of Ecology and Sustainable Economic
Development at the Tuscia University, where both D.C. and R.C.
presently work. His interests encompass co-evolution, speciation,
phylogeography and conservation genetics.