Phylogeny of Neurergus crocatus and Neurergus strauchii in Turkey based on morphological and molecular Data

Full text

PHYLOGENY OF NEURERGUS CROCATUS AND NEURERGUS
STRAUCHII IN TURKEY BASED ON MORPHOLOGICAL AND
MOLECULAR DATA
NURHAYAT O
¨ZDEMIR
1,3
,NAZAN U
¨ZU
¨M
2
,AZIZ AVCI
2
,AND KURTULUS¸ OLGUN
2
1
Department of Biology, Rize University, 53100, Rize, Turkey
2
Department of Biology, Adnan Menderes University, 09100, Aydın, Turkey
ABSTRACT: We described the phylogeny of the salamandrid genus Neurergus in Turkey using statistical
analyses on morphometric characters and a molecular analysis based on 12S and 16S rRNA genes. Two
different species, Neurergus crocatus and Neurergus strauchii (N. s. strauchii and N. s. barani), are reported
from new localities. Morphologically, the only significant differences between N. s. strauchii and N. s. barani
were eye width, head width, and forelimb length in females; and eye width, head length, head width and
inter-nostril distance in males in terms of 12 external morphometric measurements, with N. s. barani having
larger values. According to PCA, species segregated along morphological axes and all the variables (except
inter-limb and eye width in males and except longest toe forelimb, head width, eye width and inter-orbital
distance in females) were important in separating species along these axes. We obtained a total of 833
basepairs (bp) of two mitochondrial genes (478 bp of 12S rRNA and 355 bp of 16S rRNA) from Neurergus
crocatus (n510), Neurergus strauchii strauchii (n59) and N. s. barani (n59). Neurergus crocatus differed
from N. s. strauchii and N. s. barani with sequence divergences of 4.5%–5.1% and 5.4%–5.5%, respectively.
However, among three haplotypes of N. s. strauchii, sequence divergence was very low (0.24% to 0.96%).
The nucleotide difference between these two subspecies ranged from 0.48% to 1.2%. As a result, N. s. barani
specimens in this study were not strongly differentiated from N. s. strauchii, suggesting that their
distributions are either connected or only recently separated or that N. s. barani does not represent a distinct
genetic unit.
Key words: Neurergus crocatus;Neurergus strauchii barani;Neurergus strauchii strauchii; Phyloge-
ography; Turkey
NEWTS of the genus Neurergus (Salaman-
dridae) are confined to Turkey and the
Middle East. Four species have been de-
scribed (Leviton et al., 1992; Schmidtler,
1994; Sparreboom et al., 2000): Neurergus
crocatus Cope, 1862 from northern Iraq,
southeast Turkey and northwest Iran; Neur-
ergus strauchii (Steindachner, 1887) from
western area of the Van Lake in eastern
Turkey and from central Turkey (N. s. barani);
Neurergus microspilotus (Nesterov, 1916)
from the border area between Iraq and Iran;
and Neurergus kaiseri (Schmidt, 1952) from
the surroundings of Shah-Bazan of Luristan
Province, Iran (Fig. 1). All species of Neur-
ergus can be easily distinguished on morpho-
logical and ecological grounds such as breed-
ing nonbreeding habitats (Schmidtler, 1994;
Schmidtler and Schmidtler, 1970, 1975).
Knowledge of many aspects of the natural
history and biology of Neurergus is fragmen-
tary. Morphological evidence suggests that
newts belonging to this genus are closely
related to newts of the genus Triturus
(Schmidtler, 1994). A recent molecular study
(Weisrock et al., 2006) have validated earlier
studies which suggested that Mertensiella,
Euproctus and Triturus were polyphyletic
assemblages, and the results indicated strong
support for a large clade containing all species
of the genera Calotriton, Cynops, Euproctus,
Neurergus, Pachytriton, Paramesotriton, and
Triturus. Within this clade, Neurergus is most
closely related to Ommatotriton vittatus and
monophyly of Neurergus was strongly sup-
ported. Data based on mitochondrial DNA
and nuclear coded allozymes (Steinfartz et al.,
2002) indicated monophyly of Neurergus
within this group and allowed us to distinguish
two subgroups, one comprising N. s. strauchii
and N. s. barani and another represented by
N. crocatus, N. microspilotus and N. kaiseri.A
relatively complete phylogeny study of Sala-
mandridae by Steinfartz et al. (2007) stated
that a clade comprising all the ‘‘newts’’ can be
separated from the ‘‘true salamanders’’ and
Salamandrina clades. Within the ‘‘newts’’
3
CORRESPONDENCE: e-mail, nurhayat61@yahoo.com
Herpetologica, 65(3), 2009, 280–291
E2009 by The Herpetologists’ League, Inc.
280

well-supported clades are: Tylototriton-Pleur-
odeles, the ‘‘New World newts’’ (Notophthal-
mus-Taricha) and the ‘‘modern Eurasian
newts’’ (Cynops,Pachytriton,Paramesotriton
5the ‘‘modern Asian newts,’’ Calotriton,
Euproctus,Neurergus and Triturus species).
In Turkey, the genus Neurergus is repre-
sented by two species, N. crocatus Cope, 1862
(Baran and O
¨z, 1986) and N. strauchii
(Steindachner, 1887). Of the two described
subspecies, N. s. strauchii (Steindachner,
1887) is known from the Lake Van area
(Schmidtler and Schmidtler, 1970), and the
subspecies N. s. barani O
¨z, 1994 was de-
scribed from Pu¨tu¨ rge, Malatya (O
¨z, 1994).
Molecular phylogeny of the genus Neurergus,
including N. s. barani from Pu¨tu¨ rge (Malatya)
and N. s. strauchii from Su¨ru¨ m (Van), was
studied by Steinfartz et al. (2002). Based on
the 12S and 16S rRNA genes, allozymes and
three plasma protein loci of all known species
of Neurergus, Steinfartz et al. (2002) conclud-
ed that separation within the N. crocatus clade
(comprising N. crocatus, N. microspilotus and
N. kaiseri) occured around 5 mya, while N. s.
strauchii split from N. s. barani approximately
3 mya (Steinfartz et al., 2002).
Pasmans et al. (2006) collected N. s. barani
from a new locality in Malatya and N. s.
strauchii from four new localities in Elazıg˘,
Bitlis, Diyarbakır (Fig. 1, Table 1). Those
findings bridged the gap of approximately
300 km between known N. s. strauchii and N.
s. barani populations. Here we present a
molecular study of 829 bp from mitochondrial
12S and 16S rRNA genes, and morphological
analyses of N. crocatus,N. s. strauchii and N.
s. barani to examine the differentiation of
species and subspecies collected from new
localities.
Our results provide important data for the
phylogeny of N. crocatus and N. strauchii in
Turkey based on morphological and molecular
data. This is the first work reporting new
localities of N. s. strauchii,N. s. barani and N.
crocatus from Turkey. With the addition of
recent locations found during this study, the
number of N. s. strauchii locations increase to
nine, N. s. barani increases to five with the
previously published work (Pasmans et al.,
2006), and we also record a location for N.
crocatus from S¸ırnak, Turkey (there was a
record from S¸ ırnak reported by Weisrock et
al.’s (2006) study and published in GenBank
but then removed by database staff because of
poor sequence quality). Our aims were (1) to
investigate the morphological and genetic
differences of N. crocatus and N. strauchii,
and (2) to assess the level of molecular and
morphological differentiation among N. s.
strauchii and N.s. barani subspecies except
the differences mentioned before (O
¨z, 1994;
Pasmans et al., 2006).
MATERIALS AND METHODS
Samples and Morphometric Characters
Two different species, Neurergus crocatus
and Neurergus strauchii (N. s. strauchii and
N. s. barani), from Turkey were used in this
study. Neurergus crocatus were collected from
Bas¸aran Village, Beytu¨s¸s¸ebap, S¸ ırnak (37u299
N, 43u079E) at an altitude of 1135 m above
sea level (34 males and 19 females). The other
species, N. strauchii, is known as two different
subspecies in Turkey. The specimens of N. s.
strauchii were collected from breeding pop-
ulations near each other in Bitlis Province,
one in Yolyazı(Yam) Village (38u239N, 42u
059E) at an altitude of 1483 m above sea level
(24 males and 13 females) and the other in
Yalnızc¸amlar Village (38u289N, 42u119E) at
an altitude of 1777 m asl (2 females). The
specimens of the other subspecies, N. s.
barani, were collected from Tepehan (about
30 km away from type locality), Malatya (38u
FIG. 1.—Sampling sites of Neurergus in this study (12:
Bas¸aran Ko
¨yu¨/S¸ırnak; N. crocatus, 13: Yalnızc¸ amlar Ko
¨yu¨/
Bitlis; N. s. strauchii, 14: Yam Ko
¨yu¨ /Bitlis; N. s. strauchii,
15: Tepehan/Malatya; N. s. barani) and the localities (1–
11) described in Pasmans et al. (2006) (circles refer to our
localities, triangles refer to Pasmans et al.’s localities).
September 2009] HERPETOLOGICA 281

TABLE 1.—Samples used in this study with mitochondrial haplotypes. Specimen codes identify each sample sequenced and its locality (Fig. 1)
Specimen
code
Fig. 1
GenBank Accession Nos.
No. Taxa Locality 12S 16S Haplotype Reference
1Neurergus crocatus 12 37u299N; 43u079E (Turkey) EF029940 EF029968 Nc1-T This study
2Neurergus crocatus 12 37u299N; 43u079E (Turkey) EF029941 EF029969 Nc1-T This study
3Neurergus crocatus 12 37u299N; 43u079E (Turkey) EF029942 EF029970 Nc1-T This study
4Neurergus crocatus 12 37u299N; 43u079E (Turkey) EF029943 EF029971 Nc1-T This study
5Neurergus crocatus 12 37u299N; 43u079E (Turkey) EF029944 EF029972 Nc2-T This study
6Neurergus crocatus 12 37u299N; 43u079E (Turkey) EF029945 EF029973 Nc1-T This study
7Neurergus crocatus 12 37u299N; 43u079E (Turkey) EF029946 EF029974 Nc1-T This study
8Neurergus crocatus 12 37u299N; 43u079E (Turkey) EF029947 EF029975 Nc1-T This study
9Neurergus crocatus 12 37u299N; 43u079E (Turkey) EF029948 EF029976 Nc1-T This study
10 Neurergus crocatus 12 37u299N; 43u079E (Turkey) EF029949 EF029977 Nc2-T This study
11 Neurergus crocatus Aqrah, N. Iran (Iran) AY147246 AY147247 Nc3-S Steinfartz et al., 2002
12 Neurergus s. strauchii 14 38u239N; 42u059E (Turkey) EF0299950 EF0299978 Nss1-T This study
13 Neurergus s. strauchii 14 38u239N; 42u059E (Turkey) EF0299951 EF0299979 Nss2-T This study
14 Neurergus s. strauchii 14 38u239N; 42u059E (Turkey) EF0299952 EF0299980 Nss3-T This study
15 Neurergus s. strauchii 14 38u239N; 42u059E (Turkey) EF0299953 EF0299981 Nss3-T This study
16 Neurergus s. strauchii 14 38u239N; 42u059E (Turkey) EF0299954 EF0299982 Nss1-T This study
17 Neurergus s. strauchii 14 38u239N; 42u059E (Turkey) EF0299955 EF0299983 Nss1-T This study
18 Neurergus s. strauchii 14 38u239N; 42u059E (Turkey) EF0299956 EF0299984 Nss2-T This study
19 Neurergus s. strauchii 13 38u289N; 42u119E (Turkey) EF0299957 EF0299985 Nss2-T This study
20 Neurergus s. strauchii 13 38u289N; 42u119E (Turkey) EF0299958 EF0299986 Nss2-T This study
21 Neurergus s. strauchii 8Su¨ru¨m, near Lake Van (Turkey) AY147242 AY147243 Nss4-S Steinfartz et al., 2002
22 Neurergus s. strauchii 338u349N; 39u449E (Turkey) DQ131203 DQ131186 Nss5-P Pasmans et al., 2006
23 Neurergus s. strauchii 838u249N; 42u059E (Turkey) DQ131202 DQ131185 Nss6-P Pasmans et al., 2006
24 Neurergus s. strauchii 438u449N; 40u329E (Turkey) DQ131206 DQ131189 Nss7-P Pasmans et al., 2006
25 Neurergus s. strauchii 11 38u419N; 41u119E (Turkey) DQ131201 DQ131184 Nss8-P Pasmans et al., 2006
26 Neurergus s. strauchii 10 38u369N; 40u019E (Turkey) DQ131207 DQ131190 Nss9-P Pasmans et al., 2006
27 Neurergus s. barani 15 38u059N; 38u479E (Turkey) EF029959 EF029987 Nsb1-T This study
28 Neurergus s. barani 15 38u059N; 38u479E (Turkey) EF029960 EF029988 Nsb1-T This study
29 Neurergus s. barani 15 38u059N; 38u479E (Turkey) EF029961 EF029989 Nsb1-T This study
30 Neurergus s. barani 15 38u059N; 38u479E (Turkey) EF029962 EF029990 Nsb1-T This study
282 HERPETOLOGICA [Vol. 65, No. 3

059N, 38u479E) at an altitude of 1213 m asl
(31 males and 11 females). Sample locations
are shown by Fig. 1.
Newts (n5134) were caught by dipnetting
and with hands during the breeding season
(April and May 2004). We determined sex by
examining the cloacae: the male cloaca is
turgid and shaped like male cloacae of
Triturus, Mesotriton and Ommatotriton dur-
ing the breeding season, whereas the female
cloaca is not significantly different during
breeding and nonbreeding seasons.
We took the following measurements with
dial calipers with 0.02 mm precision: total body
length, body length, tail length, longest toe
forelimb, forelimb length, hindlimb length,
interlimb, head length, head width, inter-nostril
distance, eye width, and inter-orbital distance
according to Brede et al. (2000), O
¨z(1994)and
Tas¸kın and Olgun (2003).
Statistical Analyses
The morphological data are normally dis-
tributed (P.0.05, Kolmogorov–Smirnov
test), we therefore used raw data for the
statistical analyses without any transformation.
First, Pearson’s correlation coefficient was used
to check the pattern of relationships between
morphological characteristics. Analysis of co-
variance (ANCOVA) was performed to test for
differences among sexes and species in terms of
morphological characteristics by including body
length as a covariate to remove the effect of
body size variation. To discover or to reduce the
dimensionality of the data set and to identify
new meaningful underlying variables, we ana-
lyzed morphological characters by principal
component analysis (PCA) using residuals
based on a correlation matrix after testing for
sampling adequacy using Keiser-Meyer-Olkin
(KMO) analysis and Bartlett’s test. A discrim-
inant function analysis using Mahalanobis
distance method was used to estimate functions
for discriminating groups and also visualize how
the two functions discriminate between groups
by plotting the individual scores for the
discriminant functions.
DNA Extraction, Sequencing and
Sequence Alignment
Two mitochondrial genes (16S rRNA and
12S rRNA) were sequenced from 10 Neur-
TABLE 1.—Continued.
Specimen
code
Fig. 1
GenBank Accession Nos.
No. Taxa Locality 12S 16S Haplotype Reference
31 Neurergus s. barani 15 38u059N; 38u479E (Turkey) EF029963 EF029991 Nsb1-T This study
32 Neurergus s. barani 15 38u059N; 38u479E (Turkey) EF029964 EF029992 Nsb1-T This study
33 Neurergus s. barani 15 38u059N; 38u479E (Turkey) EF029965 EF029993 Nsb1-T This study
34 Neurergus s. barani 15 38u059N; 38u479E (Turkey) EF029966 EF029994 Nsb1-T This study
35 Neurergus s. barani 15 38u059N; 38u479E (Turkey) EF029967 EF029995 Nsb1-T This study
36 Neurergus s. barani 1Pu¨tu¨ rge SE Malatya (Turkey) AY147244 AY147245 Nsb2-S Steinfartz et al., 2002
37 Neurergus s. barani 138u159N; 38u379E (Turkey) DQ131212 DQ131195 Nsb3-P Pasmans et al., 2006
38 Calotriton asper France AY147258 AY147259 Casper Steinfartz et al., 2002
39 Neurergus kaiseri Zagros mountains (Iran) AY147250 AY147251 Nkaiseri Steinfartz et al., 2002
40 Neurergus microspilotus Quri-Qaleh SE Paveh (Iran) AY147248 AY147249 Nmicros Steinfartz et al., 2002
September 2009] HERPETOLOGICA 283

ergus crocatus and 18 N. strauchii from
Turkey (Table 1). Total genomic DNA was
extracted using a Qiagen
TM
kit for tissue
following the manufacturer’s directions. Prim-
ers used in both amplification and sequencing
were L1091 (59- AAA AAG CTT CAA ACT
GGG ATT AGA TAC CCC ACT AT- 39) and
H1478 (59- TGA CTG CAG AGG GTG ACG
GGC GGT GTG T- 39) (Kocher et al., 1989)
for the 12S rRNA gene, and 16Sar (59- CGC
CTG TTT ATC AAA AAC AT- 39) and 16Sbr
(59- CCG GTC TGA ACT CAG ATC ACG T-
39) (Palumbi, 1996) for the 16S rRNA gene.
PCR amplification was performed in a final
volume of 50 ml containing 10xPCR buffer, 1.5
mM MgCl
2
, each dNTP at 2 mM, each primer
at 1 uM, 1 ml genomic DNA, 1 unit Taq
polymerase. All amplifications began with an
initial 2 min denaturation at 94 C. Thermo-
cycling parameters were; 94 C for 15 s, 50 C
15 s and 72 C for 5 min for 35 cycles.
Purification of PCR products and sequencing
was performed by Macrogen Inc (Seoul,
Korea). The sequences have been deposited
in GenBank (Accession numbers EF029940–
EF029995).
Phylogenetic Analyses
DNA sequences were aligned using Clustal
X (Thompson et al., 1997) and subsequently
adjusted by eye. All the 12S rRNA and 16S
rRNA sequences had the same length and
therefore no gaps were postulated. We
obtained a final alignment of 833 bp for 12S
and 16S in phylogenetic analyses after adding
other sequences used in other studies on
Neurergus (Table 1; Pasmans et al., 2006;
Steifartz et al., 2002). Although some gaps
were postulated in order to resolve length, all
positions could be unambiguously aligned and
included in the analysis. We defined Calo-
triton asper (AY147258 for 12S rRNA and
AY147259 for 16S rRNA gene) as the out-
group for phylogenetic reconstruction based
on the study of Pasmans et al. (2006). Pairwise
distances were estimated using an uncorrect-
ed distance model.
Three methods of phylogenetic analysis
were employed for the combined data set.
These were: maximum likelihood (ML),
Bayesian analysis, and maximum parsimony
(MP). Modeltest 3.6 (Posada and Crandall,
1998) was used to select the most appropriate
model of sequence evolution for the ML and
Bayesian analyses of the combined dataset,
under the Akaike Information Criterion. The
TIM+I model had the best likelihood score
(2lnL 51872.2559) and Akaike Information
Criterion (AIC 53824.5117). Parameter
estimates were: base frequencies 0.3632 (A),
0.2237 (C), 0.1878 (G), 0.2253 (T); Ti/Tv ratio
is 4.2139. Also ML analysis, performed by
RaxML (Stamatakis et al., 2008) to test
partitioning by gene, results in a tree with a
significantly higher likelihood (-lnL 5
1848.2111).
To test for deviations from neutrality,
Tajima’s D (Tajima, 1989) test was used as
implemented in Arlequin 3.11 (Excoffier et
al., 2005).
Bayesian analyses were conducted with
MrBayes 3.0b4 (Huelsenbeck and Ronquist,
2001) for a given model of sequence evolu-
tion. Four Markov chains were used in each
replicate, and the chain was sampled every
100 generations. Analyses were allowed to run
for 2 million generations, discarding 200,000
generations as burn-in and produced summa-
ry statistics for trees (sumt) sampled during
analysis.
Both ML and MP analyses were performed
using PAUP*4.0b10 software (Swofford,
2000) and included heuristic searches involv-
ing tree bisection and reconnection (TBR)
branch swapping with 100 random stepwise
additions of taxa. Reliability of the MP and
ML trees was assessed by bootstrap analysis
involving 100 replications.
In addition, sequences were connected
under the 95% probability of parsimony
criterion using the software TCS (version
1.18, Clement et al., 2000). The resulting
network represents reconstructed gene gene-
alogy of the haplotypes.
RESULTS
Morphological Data
The means and standard errors of the
morphometric characters of N. crocatus, N.
s. strauchii and N. s. barani are shown in
Table 2. There were no significant differences
between N. s. strauchii and N. s. barani except
eyewidth (ANCOVA: P,0.05, F55.43, df 5
284 HERPETOLOGICA [Vol. 65, No. 3

1), headwidth (F59.13, df 51), forelimb
length (F58.60, df 51) in females, and
eyewidth (F518.96, df 51), head length (F
510.29, df 51), headwidth (F55.04, df 5
1) and inter-nostril distance (F54.33, df 51)
in males (ANCOVA: P,0.05) in terms of 12
external morphometric measurements, with
N. s. barani having wider eyes and larger
heads. Pearson’s correlation coefficient
ranged from 0.116 (between eye width and
hindlimb length) to 0.952 (between total body
length and tail length). We performed PCA
and discriminant analyses separately for males
and females. The sampling adequacy of the
model and the Bartlett test of sphericity were
high for males (KMO 50.89; Chi-square 5
513.28, df 555, P,0.0001) and females
(KMO 50.83; Chi-square 5283.38, df 555,
P,0.0001), after leaving out total body
length character because of high correlation
with tail length. According to PCA on the
correlation matrix, two components were
extracted, explaining 61.1% of all morphomet-
ric variation for males (Table 3). All of the
variables load well on the first component
except inter limb and eye width on the second
component. Two components were extracted
for females, with the first component explain-
ing 38.9% and the second component 22.9%
of all morphometric variation (Table 3). High
loadings on the first component were ob-
served almost for all the characters except
TABLE 2.—Means and standard errors (SE) of morphological characteristics, including Neurergus crocatus, Neurergus
strauchii strauchii and Neurergus strauchii barani.
Characteristic
N. crocatus R
n519 N. crocatus =
n534 N. s. strauchii R
n515 N. s. strauchii =
n524 N. s. barani R
n511 N. s. barani =
n531
mean SE mean SE mean SE mean SE mean SE mean SE
Total body length 149.83 1.98 130.95 1.18 165.99 3.53 148.48 2.56 164.24 1.85 149.54 1.12
Body length 71.53 0.72 62.72 0.59 79.64 0.91 69.88 0.91 78.09 0.93 69.66 0.49
Tail length 78.30 1.42 68.23 4.16 86.36 2.06 78.60 1.86 86.15 2.22 79.89 0.81
Longest toe forelimb 4.88 0.15 4.49 0.07 5.63 0.11 5.23 0.16 5.38 0.11 5.04 0.11
Forelimb length 23.97 0.37 21.99 0.22 24.39 0.37 23.29 0.39 25.59 0.46 23.66 0.22
Hindlimb length 25.22 0.39 24.20 0.23 27.61 0.49 26.41 0.43 27.99 0.54 25.89 0.26
Interlimb 37.76 0.50 31.57 0.41 42.48 0.78 36.91 0.55 42.38 0.42 36.78 0.29
Head length 16.84 0.19 15.73 0.12 18.16 0.42 16.89 0.23 18.51 0.30 17.51 0.18
Head width 13.45 0.17 12.21 0.13 14.06 0.26 13.23 0.20 14.68 0.22 13.58 0.14
Inter-nostril distance 3.52 0.09 3.31 0.07 3.56 0.09 3.42 0.06 3.61 0.08 3.61 0.06
Eye width 4.12 0.11 3.89 0.07 4.01 0.11 3.79 0.06 4.38 0.08 4.15 0.06
Inter-orbital distance 7.62 0.12 6.96 0.09 7.83 0.11 7.72 0.12 7.82 0.13 7.72 0.08
TABLE 3.—Factor loadings on the principal components extracted from the correlation matrix of 12 characteristics for
males and females of N. crocatus, N. strauchii strauchii and N. s. barani.
Characteristic
Males Females
PC1 PC2 PC1 PC2
Tail length 0.605 0.369 0.550 0.215
Body length 0.660 0.593 0.871 0.302
Longest toe forelimb 0.795 0.047 0.298 0.693
Forelimb length 0.828 0.232 0.690 0.433
Hindlimb length 0.791 0.130 0.781 0.106
Interlimb 0.217 0.790 0.761 0.196
Head length 0.595 0.476 0.788 0.069
Head width 0.670 0.461 0.575 0.634
Inter-nostril distance 0.485 0.292 0.425 0.307
Eye width 0.086 0.768 -0.052 0.876
Inter-orbital distance 0.569 0.513 0.570 0.630
Variance explained (%) 37.834 23.313 38.869 22.913
September 2009] HERPETOLOGICA 285

longest toe forelimb, head width, eye width
and inter-orbital distance that loaded highly
on the second component.
Females of Neurergus crocatus, N. strauchii
strauchii, N. s. barani were entered into
discriminant analysis as three groups to
elucidate the degree of discrimination among
the groups. The resulting discriminant analy-
sis showed significant differences among three
groups (Wilks’ Lambda 50.380, 0.759 for the
first and second functions, respectively, P,
0.001). To display the variations among the
groups graphically, the individual discriminant
scores were plotted for the canonical discrim-
inant functions (F1-F2). Although some indi-
vidual scores overlapped for the three groups
in canonical space (Fig. 2), the separation of
N. crocatus can be easily observed, but the N.
s. strauchii population does not form an
independent group from N. s. barani.Asa
result of the analyses, 67.4% of the original
grouped cases of females were correctly
classified.
For males, the resulting discriminant anal-
ysis showed significant differences among the
three groups (Wilks’ Lambda 50.290, 0.777
for the first and second functions, respective-
ly, P,0.001). The individual discriminant
scores were plotted for the canonical discrim-
inant functions (F1-F2) and seen in Fig. 3.
Although Neurergus crocatus individuals were
rather well separated, some individual scores
overlapped for N. s. strauchii and N. s. barani.
As a result of the analyses, 83.3% of the
original grouped cases of males were correctly
classified.
Molecular Data
A total of 833 bp (478 bp of the 12S rRNA
and 355 of the 16S rRNA genes) was obtained
from Neurergus crocatus (n510), Neurergus
strauchii strauchii (n59) and N. s. barani (n
59). No genetic variation was observed
within the N. s. barani specimens collected
in this study. Only two haplotypes were
identified for N. crocatus and three haplotypes
for N. s. strauchii (Table 1).
A total of 833 bp were included in the
phylogenetic analysis. Of these, 41 were
variable and 55 were parsimony informative.
ML, MP, and Bayesian analyses produced
trees that differed in some minor arrange-
ments of individual samples. In all cases, these
differerences had low support values of less
than 70%. Consequently, it was considered
that there were no major topological conflicts
between all trees, the majority rule consensus
tree of the Bayesian analyses is shown in
Fig. 4 with bootstrap and posterior probability
supports.
The sequence divergences between haplo-
types are given in Table 4. Among two
haplotypes of N. crocatus, sequence diver-
gence is 0.12% for this study; the haplotypes
are closely related to the sequences of N.
crocatus (0.84–0.96%) from Aqrah, Iraq
(Steinfartz et al., 2002) as seen in Fig. 4. In
addition, N. crocatus differs from N. s.
strauchii and N. s. barani with sequence
divergences of 4.5–5.1% and 5.4–5.5%, re-
spectively. However, among three haplotypes
of N. s. strauchii, sequence divergence was
very low, with a range from 0.24% to 0.96%
FIG. 2.—Plot of individual females of N. crocatus, N.
strauchii strauchii and N. s. barani belonging to three
groups on two canonical discriminant functions.
FIG. 3.—Plot of individual males of N. crocatus, N.
strauchii strauchii and N. s. barani belonging to three
groups on two canonical discriminant functions.
286 HERPETOLOGICA [Vol. 65, No. 3

(Table 4). The only haplotype, Nsb1-T, found
in our study is different from the haplotype
Nsb2-S in Steinfartz et al. (2002)’s study with
1.8% and from Nsb3-P haplotype in Pasmans
et al. (2006)’s with 1.33%.
All the N. s. strauchii individuals sampled
for this study and others (Pasmans et al., 2006;
Steinfartz et al., 2002) formed a single clade
(bootstrap value 100), including N. s. barani
samples. The samples of N. s. barani from
Malatya, Turkey (Nsb2-S, Nsb3-P; Pasmans et
al., 2006) represent an independent lineage
separated from N. s. strauchii clade including
N. s. barani specimen collected in this study.
The other three species of Neurergus
included in the analyses (N. kaiseri, N.
microspilotus, N. crocatus) form a well
supported group (bootstrap values 73, 100
and 77 in the ML, MP and Bayesian analyses,
respectively).
For 833 bp sequences, haplotypes connect-
ed by #26 substitutions have at least a
probability of 0.95 of being parsimoniously
connected and a network was generated using
TCS (Clement et al., 2000). All the haplotypes
could be joined in a single network (Fig. 5)
and the results are in line with the other
phylogenetic analyses.
DıSCUSSıON
While phylogenetic and statistical analyses
confirmed the species status of N. crocatus
and N. strauchii species, the status of the
subspecies N. s. strauchii and N. s. barani was
not supported by our analyses. The most
obvious difference between N. s. strauchii and
N. s. barani is the difference in the number of
yellow spots found on adults of both subspe-
cies (Pasmans et al., 2006). In our data, there
were a few significant morphological differ-
ences between N. s. strauchii and N. s. barani
(eye width, head width, forelimb length in
females; eye width, head length, head width
and inter-nostril distance in males); in both
cases measurements were larger in N. s.
barani than N. s. strauchii.
However, PCA and discriminant analyses
showed that while N. crocatus is clearly
separated from N. strauchii,N. s. strauchii
and N. s. barani specimens are partially
separated from each other with some overlap
that was easily seen both in females and
TABLE 4.—Uncorrected genetic distance among haplotypes described in Table 1.
1 234567 8 9 101112131415161718
1. Nc1-T —
2. Nc2-T 0.0012 —
3. Nss1-T 0.0457 0.0469 —
4. Nss2-T 0.0505 0.0517 0.0072 —
5. Nss3-T 0.0505 0.0517 0.0096 0.0024 —
6. Nsb1-T 0.0541 0.0553 0.0120 0.0048 0.0072 —
7. Casper 0.0761 0.0773 0.0725 0.0737 0.0737 0.0761 —
8. Nkaiseri 0.0312 0.0324 0.0493 0.0493 0.0493 0.0541 0.0641 —
9. Nmicros 0.0252 0.0264 0.0421 0.0421 0.0445 0.0457 0.0689 0.0228 —
10. Ncs3-S 0.0084 0.0096 0.0457 0.0505 0.0505 0.0541 0.0700 0.0312 0.0252 —
11. Nsb2-S 0.0506 0.0518 0.0132 0.0132 0.0156 0.0180 0.0750 0.0507 0.0434 0.0494 —
12. Nsb3-P 0.0472 0.0484 0.0085 0.0084 0.0109 0.0133 0.0729 0.0484 0.0411 0.0472 0.0024 —
13. Nss4-S 0.0505 0.0517 0.0108 0.0036 0.0060 0.0084 0.0725 0.0470 0.0409 0.0481 0.0120 0.0097 —
14. Nss5-P 0.0507 0.0519 0.0096 0.0024 0.0048 0.0024 0.0741 0.0495 0.0410 0.0495 0.0133 0.0109 0.0036 —
15. Nss6-P 0.0496 0.0508 0.0072 0.0000 0.0024 0.0048 0.0717 0.0472 0.0399 0.0484 0.0109 0.0084 0.0012 0.0024 —
16. Nss7-P 0.0507 0.0519 0.0084 0.0012 0.0036 0.0060 0.0729 0.0483 0.0410 0.0495 0.0121 0.0096 0.0024 0.0036 0.0012 —
17. Nss8-P 0.0495 0.0507 0.0072 0.0000 0.0024 0.0048 0.0717 0.0471 0.0398 0.0483 0.0109 0.0084 0.0012 0.0024 0.0000 0.0012 —
18. Nss9-P 0.0483 0.0495 0.0084 0.0012 0.0036 0.0060 0.0705 0.0483 0.0410 0.0471 0.0121 0.0096 0.0024 0.0036 0.0012 0.0024 0.0012 —
September 2009] HERPETOLOGICA 287

especially in males (Figs. 2, 3). This finding is
also in line with Pasmans et al.’s (2006) study
in that they did not find a main phenotypic
difference in terms of spots on newts.
However, they reported a gradual increase in
the number of spots in adult newts eastwards
that is not restricted to one locality or
population. This difference means that N. s.
strauchii specimens in the western part of
their range (closer to the range of N. s. barani)
have fewer yellow spots than those to the east.
Moreover, Pasmans et al. (2006) could not
find any differences in belly patterns between
subspecies and concluded that this morpho-
logical character, initially suggested by O
¨z
(1994), is of low taxonomic value and should
not be taken into consideration for the
differentiation between both subspecies of
N. strauchii.
As a result of sequencing 478 bp of 12S
rRNA and 355 bp of 16S rRNA genes in our
study, only two haplotypes were identified for
N. crocatus, three haplotypes for N. s.
strauchii, and only one haplotype for N. s.
barani. Pasmans et al. (2006) reported one
haplotype (Nsb3-P) analyzing four N. s. barani
specimens from Malatya (Fig. 1). The other
N. s. barani haplotype (Nsb2-S) belongs to the
specimen studied by Steinfartz et al. (2002)
from the same locality. There are four
substitutions (including 2 deletions) among
the sequences of Pasmans et al. (2006) and
Steinfartz et al. (2002). However, seven
substitutions were observed when comparing
our N. s. barani haplotype (Nsb1-T) with the
Nsb3-P haplotype of Pasmans et al.’s (2006)
study. Nsb3-P haplotype from Malatya dif-
fered from N. s. strauchii haplotypes by 0.48%
to 1.20% (Pasmans et al., 2006). In our study,
sequence divergence among three haplotypes
of N. s. strauchii (Nss1-T, Nss2-T and Nss3-T)
was very low, ranging from 0.24% to 0.96%
(Table 4). The only N. s. barani haplotype
(Nsb1-T) is different from Nsb2-S (Steinfartz
et al., 2002) by 1.18% and from Nsb3-P
(Pasmans et al., 2006) by 1.33%.
Although the 829 bp of 12S and 16S rRNA
genes of five N. s. barani specimens from Malatya
(Nsb2-S, Nsb3-P) formed a single clade different
from N. s. strauchii in Pasmans et al.’s (2006)
FIG. 4.—Majority rule consensus tree of the Bayesian analysis of Neurergus including new haplotypes for the studies
of Pasmans et al. (2006), Steinfartz et al. (2002) and this study. Haplotype names refer to the taxa mentioned in Table 1.
Bootstrap values of ML, MP analysis and posterior probability values of Bayesian analysis are shown on each node of the
tree (ML/MP/Bayesian posterior probability values, respectively).
288 HERPETOLOGICA [Vol. 65, No. 3

study, the differentiation of the two subspecies
was not supported by our phylogenetic analyses
in which the only haplotype of N. s. barani
(Nsb1-T) was nested within N. s. strauchii clade.
In contrast, in the study of Pasmans et al. (2006),
Nsb2-S and Nsb3-P haplotypes are partof the N.
strauchii assemblage but at the same time
represent an independent lineage separated
from all other samples of N. strauchii.While
the nucleotide difference between these two
subspecies ranged from 1–1.4% in the study of
Pasmans et al. (2006), it ranged from 0.48% to
1.2% in our study.
The above results indicate a striking lack of
genetic variation both within populations and
among haplotypes of N. strauchii. Such
reduced levels of genetic diversity could result
from colonization events that gave rise to
these populations (Wade et al., 1994). Alter-
natively, more recent population bottlenecks,
possibly because of habitat restriction during
Pleistocene glaciations, could have reduced
genetic diversity to these levels (Riberon et al.,
2001). Computations of neutral evolution of
the sequences (Tajima, 1989) resulted in a
negative and not significant (Tajima’s D 5
20.4583, P.0.05). This indicates neutral
evolving sequences with an excess of rare
polymorphisms due to bottlenecks or popula-
tion fusion (Haubold et al., 2002).
We propose two alternate hypotheses: (1)
the N. s. strauchii and N. s. barani populations
are still connected, and (2) the populations
recently became discontinuous. Under (i) N. s.
strauchii populations are expected in the west
of the Fırat River or, these two subspecies
exchange genes at regions of peripatry that is
in contrast with the suggestion of Pasmans et
al. (2006). However a gradual increase in the
number of spots in adult newts eastwards
confirmed this hypothesis. As it is known, all
species except the pond-breeding N. kaiseri
are stream breeders (Schmidtler and Schmid-
tler, 1970). Neurergus strauchii inhabit moun-
tain streams during their breeding season and
inhabit nearby terrestrial habitats during other
parts of the year. Schmidtler and Schmidtler
(1970) found N. strauchii hibernating at 25 m
away from the stream and about 5 m higher in
a heap of stones. As Bogaerts et al. (2006)
state, more information is needed on how far
newts migrate from streams; in wide streams
(wider than 2 m) it is difficult to detect newts,
and it is possible that in those streams newts
are present but not detected.
If, in contrast, hypothesis (2) applies, we
predict the River Fırat acts as a natural barrier
to gene flow between both subspecies in line
with Pasmans et al’s (2006) suggestion. The
results also agree with Steinfartz et al. (2002)
who found a relatively high level of genetic
differentiation at both mitochondrial (12S and
16S rRNA) and nuclear levels between N. s.
barani and easternmost populations of N. s.
staruchii. The ecological adaptation to high
mountain brooks may have trapped them in a
restricted area for millions of years (Steinfartz
et al., 2002). Steinfartz et al. (2002) estimated
the time since separation of species for
mitochondrial genes and protein loci for
Neurergus. They reported congruent esti-
mates only for the separation of the N.
strauchii-clade from the N. crocatus clade at
FIG. 5.—Single most parsimonious network of relation-
ships among the partial 12S rRNA and 16S rRNA gene
sequences for Neurergus strauchii. Circles along the
branches indicate nucleotide changes, the square on top
indicates the presumed ancestral haplotype.
September 2009] HERPETOLOGICA 289

8.5–13.9 mya and the separation of N. s.
strauchii and N. s. barani 3mya.Other
estimates differ greatly between mitochondrial
DNA and allozymes. It must be kept in mind
that mtDNA is strictly a marker of historical
process in females; should male and female
history differ in a species, then the marker
would not reflect the history of the species as a
whole but that of the female portion (Hurst and
Jiggins, 2005). The lack of any significant
sequence divergence between these subspecies
indicates that, at present, they may not repre-
sent distinct evolutionary lineages and incom-
plete lineage sorting can also be the explanation
for the pattern observed in our data.
In this study, it is clear that N. s. barani
specimens are not strongly differentiated from
N. s. strauchii, suggesting that their distribu-
tions are either connected or only recently
separated. However, further sampling around
the River Fırat for closing the gap between
known N. s. strauchii and N. s. barani
populations is needed to better understand
the genetic variation and taxonomic subdivi-
sions within species and contact zones.
Acknowledgments.—We thank A. Osman Beldu¨z from
Karadeniz Technical University’s Biology Department.
LITERATURE CITED
BARAN, I., AND M. O
¨Z. 1986. On the occurrence of
Neurergus crocatus and N. strauchii in Southeast
Anatolia. Zoology in the Middle East 1:96–104.
BOGAERTS, S., F. PASMANS,AND T. WOELTIES. 2006.
Ecology and conservation aspects of Neurergus strau-
chii (Amphibia: Salamandridae). Pp. 15–18. In Pro-
ceedings of the 13th Congress of the Societas Europae
Herpetologica. Bonn, Germany.
BREDE, E. G., R. S. THORPE,J.W.ARNTZEN,AND T. E. S.
LANGTON. 2000. A morphometric study of a hybrid newt
population (Triturus cristatus/T. carnifex). Beam Brook
Nurseries, Surrey, UK. Biological Journal of the
Linnean Society 70:685–695.
CLEMENT, M., D. POSADA,AND K. A. CRANDALL. 2000.
TCS, a computer program to estimate gene genealogies.
Molecular Ecology 9:1657–1659.
EXCOFFıER, L., G. LAVAL,AND S. SCHNEıDER. 2005.
Arlequin ver. 3.0: An integrated software package for
population genetics data analysis. Evolutionary Bioin-
formatics Online 1:47–50.
HAUBOLD, B., J. KROYMANN,A.RATZKA,T.MıTCHELL-OLDS,
AND T. WıEHE. 2002. Recombination and gene conver-
sion in a 170 kb genomic region of Arabidopsis thaliana.
Genetics 161:1269–1278.
HUELSENBECK, J. P., AND F. RONQUIST. 2001. MRBAYES:
Bayesian Inference of Phylogenetic Trees. Bioinfor-
matics 17:754–755.
HURST, G. D. D., AND F. M. JIGGINS. 2005. Problems with
mitochondrial DNA as a marker in population,
phylogeographic and phylogenetic studies: the effects
of inherited symbionts. Prooceedings of the Royal
Society 272:1525–1534.
KOCHER, T. D., W. K. THOMAS,A.MEYER,S.V.EDWARDS,
S. PAABO,F.X.VILLABLANCA,AND A. C. WILSON. 1989.
Dynamics of Mitochodrial DNA Evolution in Animals:
Amplification and Sequencing with Conserved Primers,
Proceedings of the National Academy of Sciences, USA,
86:6196–6200.
LEVITON, A. E., S. C. ANDERSON,K.ADLER,AND A.
MINTON. 1992. Handbook to Middle East Amphibians
and Reptiles. Oxford (Ohio) Society for the Study of
Amphibians and Reptiles.
O
¨Z, M. 1994. A new form of Neurergus strauchii (Urodela:
Salamandridae) from Turkey. Turkish Journal of
Zoology 18:115–117.
PALUMBI, S. R. 1996. The polymerase chain reaction.
Pp. 205–247. In Molecular Systematics. 2nd ed., D.
Hillis, C. Moritz, and B. K. Mable (Eds.), Sinauer
Associates, Sunderland, Massachusetts, USA.
PASMANS, F., S. BOGAERTS,T.WOELTIES,AND S. CARRANZA.
2006. Biogeography of Neurergus strauchii barani O
¨z,
1994 and N. s. strauchii (Steindachner, 1887) (Am-
phibia: Salamandridae) assessed using morphological
and molecular data. Amphibia-Reptilia 27:281–288.
POSADA, D., AND K. A. CRANDALL. 1998. Modeltest: Testing
the Model of DNA Substitution. Bioinformatics 14:
817–818.
RIBERON, A., E. SOTIRIOU,C.MIAUD,F.ANDREONE,AND F.
TABERLET. 2001. Lack of genetic diversity in Salaman-
dra lanzai revealed by cytochrome b gene sequences.
Copeia 1:229–232.
SCHMIDTLER, J. J., AND J. F. SCHMIDTLER. 1970. Morpho-
logie, Biologie und Verwandtschafts- beziehungen von
Neurergus strauchii aus der Turkei. Senckenbergiana
Biologica 51:42–53.
SCHMIDTLER, J. J., AND J. F. SCHMIDTLER. 1975. Untersu-
chunge an westpersischen Bergbachmolchen der Gat-
tung Neurergus. Salamandra 11:84–98.
SCHMIDTLER, J. F. 1994. Eine U
¨bersicht neuerer Un-
tersuchungen und Beobachtungen an der vorderasia-
tischen Molchgattung Neurergus. Abhandlungen mid
Berichte des Museums fur Natur- and Heimatskunde
zu Magdeburg, Magdeburg 17:193–198.
SPARREBOOM, M., S. STEINFARTZ,AND G. SCHULTSCHIK.
2000. Courtship behavior of Neurergus. Amphibia-
Reptilia 21:1–11.
STAMATAKıS, A., P. HOOVER,AND J. ROUGEMONT. 2008. A
Rapid Bootstrap Algorithm for the RAxML Web-
Servers. Systematic Biology 75, 5, 758–771.
STAMATAKıS, S., U. W. HWANG,D.TAUTZ,M.O
¨Z,AND
M. VEıTH. 2002. Molecular phylogeny of the salaman-
drid genus Neurergus: evidence for an intrageneric
switch of reproductive biology. Amphibia-Reptilia 23:
419–431.
STEINFARTZ, S., S. VICARIO,J.W.ARTZEN,AND A. CACCONE.
2007. A Bayesian approach on molecules and behav-
iour: reconsidering evolutionary patterns in Triturus
newts (Amphibia: Salamandridae). Journal of Experi-
mental Zoology, part B: Molecular and Developmental
Evolution 308B:139–162.
290 HERPETOLOGICA [Vol. 65, No. 3

SWOFFORD, D. L. 2000. PAUP*Phylognetic Analysis Using
Parsimony (*and Other Methods), version 4. Sinauer
Associates, Sutherland, Massachusetts, USA.
TAJıMA, F. 1989. Statistical methods to test for nucleotide
mutation hypothesis by DNA polymorphism. Genetics
123:585–595.
TAS¸KIN, N., AND K. OLGUN. 2003. Morphological compar-
ison of two populations of the Smooth Newt, Triturus
vulgaris, from western Anatolia. Zoology in the Middle
East 28:61–66.
THOMPSON, J. D., T. J. GIBSON,F.PLEWNIAK,F.
JEANMOUGIN,AND D. G. HIGGINS. 1997. The ClustalX
Windows Interface: Flexible Strategies for Multiple
Sequence Alignment Aided by Quality Analysis Tools.
Nucleic Acids Research 24:4876–4882.
WADE, M. J., M. L. MCKNIGHT,AND H. B. SHAFFER. 1994.
The effects of skinstructured colonization on nuclear
and cytoplasmic genetic diversity. Evolution
48:1114–1120.
WEISROCK, D. W., T. J. PAPENFUSS,J.R.MACEY,S.N.
LITVINCHUK,R.POLYMENI,I.H.UGURTAS,E.ZHAO,H.
JOWKAR,AND A. LARSON. 2006. A molecular assessment
of phylogenetic relationships and lineage accumulation
rates within the family Salamandridae (Amphibia,
Caudata). Molecular Phylogenetics and Evolution
41:368–383.
.Accepted: 1 August 2009
.Associate Editor: Frank Burbrink
September 2009] HERPETOLOGICA 291