Annals of Botany 108: 241 –252, 2011
doi:10.1093/aob/mcr134, available online at www.aob.oxfordjournals.org
The importance of Anatolian mountains as the cradle of global diversity
in Arabis alpina, a key arctic – alpine species
Stephen W. Ansell1,*, Hans K. Stenøien3, Michael Grundmann1, Stephen J. Russell2, Marcus A. Koch4,
Harald Schneider1 and Johannes C. Vogel1
1
Department of Botany, The Natural History Museum, London SW7 5BD, UK, 2Department of Zoology, The Natural History
Museum, London SW7 5BD, UK, 3Systematics and Evolution Group, Section of Natural History, Museum of Natural History
and Archaeology, Norwegian University of Science and Technology, Trondheim, N-7491, Norway and 4Division of Biodiversity
and Plant Systematics, Institute of Plant Sciences, University of Heidelberg, Im Neuenheimer Feld 345, Heidelberg,
D-69120, Germany
* For correspondence. E-mail [email protected]
Received: 19 December 2010 Returned for revision: 1 March 2011 Accepted: 4 April 2011 Published electronically: 28 June 2011
† Background and Aims Anatolia is a biologically diverse, but phylogeographically under-explored region. It is
described as either a centre of origin and long-term Pleistocene refugium, or as a centre for genetic amalgamation, fed from distinct neighbouring refugia. These contrasting hypotheses are tested through a global phylogeographic analysis of the arctic–alpine herb, Arabis alpina.
† Methods Herbarium and field collections were used to sample comprehensively the entire global range, with
special focus on Anatolia and Levant. Sequence variation in the chloroplast DNA trnL-trnF region was examined
in 483 accessions. A haplotype genealogy was constructed and phylogeographic methods, demographic analysis
and divergence time estimations were used to identify the centres of diversity and to infer colonization history.
† Key Results Fifty-seven haplotypes were recovered, belonging to three haplogroups with non-overlapping distributions in (1) North America/Europe/northern Africa, (2) the Caucuses/Iranian Plateau/Arabian Peninsula and
(3) Ethiopia–eastern Africa. All haplogroups occur within Anatolia, and all intermediate haplotypes linking the
three haplogroups are endemic to central Anatolia and Levant, where haplotypic and nucleotide diversities
exceeded all other regions. The local pattern of haplotype distribution strongly resembles the global pattern,
and the haplotypes began to diverge approx. 2.7 Mya, coinciding with the climate cooling of the early Middle
Pleistocene.
† Conclusions The phylogeographic structure of Arabis alpina is consistent with Anatolia being the cradle of
origin for global genetic diversification. The highly structured landscape in combination with the Pleistocene
climate fluctuations has created a network of mountain refugia and the accumulation of spatially arranged genotypes. This local Pleistocene population history has subsequently left a genetic imprint at the global scale,
through four range expansions from the Anatolian diversity centre into Europe, the Near East, Arabia and
Africa. Hence this study also illustrates the importance of sampling and scaling effects when translating
global from local diversity patterns during phylogeographic analyses.
Key words: Anatolia, centre of origin, Pleistocene glaciations, chloroplast trnL-F, divergence times, alpine
plants, Arabis alpina.
IN T RO DU C T IO N
Genetic diversity is often structured within species distribution
ranges (Taberlet, 1998; Hewitt, 2000, 2004; Médail and
Diadema, 2009). Local hotspots of diversity may arise
through several means: the long-term survival of populations
during past climatic fluctuations (glacial refugia; Hewitt,
2004; Schmitt, 2007), or through the recent amalgamation of
divergent lineages (Ferris et al., 1998; Petit et al., 2003;
Walter and Epperson, 2005), or local ecological niche gradients may promote the accumulation of lineages (Ricklefs,
2007). Discriminating among the first two scenarios still
remains a challenge when interpreting phylogeographic
history, particularly outside the much studied European and
North American continents.
Anatolia is a hotspot for biological diversity in the
Mediterranean basin and is a biogeographically interesting,
but under-explored region. It is situated between the
European and Asian continents and is a meeting point for
the European and Turko-Iranian floras, which broadly
overlap in central Anatolia (Davis, 1965 –85). During the
Pleistocene (approx. 2.5 Mya) glacier development within
Anatolia was limited to the higher mountain peaks (Atalay,
1996), while the lowlands remained open, developing steppe
communities (Michaux et al., 2004; Magyari et al., 2008).
These provided suitable habitats for temperate species to
survive the last glacial maximum (Rokas et al., 2003; Dubey
et al., 2006; Fritz et al., 2009) and perhaps in earlier cycles.
Late Pleistocene sea levels around the Mediterranean and
Caspian seas were regularly lower than current levels
# The Author 2011. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved.
For Permissions, please email: [email protected]
242
Ansell et al. — Anatolian mountain plant phylogeography
(approx. 130 m, Kerey et al., 2004) and land bridges routinely
formed across the Sea of Marmara and the Bosporus Strait
(Magyari et al., 2008). This allowed for frequent intercontinental population exchange, especially for temperate
organisms (Kučera et al., 2006; Dubey et al., 2007; Gündüz
et al., 2007). In the east, the high ‘Anatolia diagonal’
(Davis, 1971) mountains (.2000 m) have limited the mixing
of European and Near-Eastern biota (Ekim and Güner, 1986;
Bilgin et al., 2006), but their topography may also have facilitated local survival and speciation, as suggested by the high
level of endemism (Davis, 1971; Çiplak, 2003). Together
this combination of geography and climatic history has promoted both the long-term local survival of biota, and repeated
uni- or bi-directional colonization(s) from nearby areas,
making Anatolia a complex but important area for understanding the diversification of Eurasian biodiversity.
Increasingly the biogeographic history of Anatolia and
adjoining areas is being explored using molecular tools.
While data are still limited, bats, rodent, turtles and wasp
groups are now represented (e.g. Michaux et al., 2004;
Bilgin et al., 2006; Dubey et al., 2007; Gündüz et al., 2007;
Flanders et al., 2009; Furman et al., 2009; Fritz et al.,
2009). Plants remain relatively poorly represented, despite
several studies focusing on trees (Bartish et al., 2006;
Kučera et al., 2006; Gömöry et al., 2007; Naydenov et al.,
2007) or herbs (Jakob et al., 2007; Font et al., 2009; Parolly
et al., 2010). Unfortunately low sampling densities have
limited the ability to characterize the location of refugia,
hybrid zones and colonization routes (Gündüz et al., 2007).
Nevertheless several phylogeographic trends emerge from
these studies, but with no clear consensus across all groups:
(1) Near East and European lineages are genetically divergent;
(2) Anatolia was the source area for European colonization; (3)
the Thracian Plain/Sea of Marmara is a barrier to European
colonization; and (4) the high Anatolian diagonal mountains
are a barrier to colonization from the Near East (Fig. 1).
Phylogeographic information from alpine species is currently
missing and would complement the existing information
from temperate species, especially in a context of the
Anatolian mountains as a refugium.
The arctic – alpine herb Arabis alpina was selected to
explore the phylogeographic history of the Anatolia mountains
and relationships with adjacent mountainous areas. Arabis
alpina is a short-lived perennial belonging to the mustard
family Brassicaceae (Hegi, 1986), and a relative of the
model organism Arabidopsis thaliana (Koch et al., 2000,
2001, 2007; Beilstein et al., 2008; Couvreur et al., 2010).
Arabis alpina itself is currently being developed as a model
for studying alpine and life-history adaptation (Wang et al.,
2009; Poncet et al., 2010), although complicated by loss of
self-incompatibility in parts of the Europe (Tedder et al.,
2011). The herb is distributed throughout the alpine habitats
in Europe, in the arctic zone of Greenland and North
America, in the high mountains of northern and eastern
Africa and Anatolia; and in the ranges extending into the
Caucasus and the Near East (Meusel et al., 1965). The
global phylogeographic structure of A. alpina was outlined
by an earlier chloroplast DNA studies, and three haplogroups
were recovered from the core areas adjacent to Anatolia:
Arctic/Europe/north Africa, the Near East and east Africa
(Koch et al., 2006; Assefa et al., 2007; Ehrich et al., 2007).
The importance of Anatolia at the centre of the global diversity
range for A. alpina was recognized by a ‘three-times out of
Asia-Minor’ hypothesis (Koch et al., 2006), but sampling
was modest (n ¼ 3). An alternative hypothesis of Anatolia as
a centre for the amalgamation of European and Near East
lineages is considered for other Brassicaceae (Kučera et al.,
2006). These two phylogeographic hypotheses depend on the
availability of suitable ecological and geographical connections between Anatolia and adjoining territories during the
Pleistocene. This is especially true for alpine species like A.
alpina, as colonizations between Anatolian and adjoining
mountain systems are limited to glacial periods when alpine
habitats are more widespread at lower elevations. As such
we recognize that two key geographic features may have hindered or aided Pleistocene colonization of Anatolia: the
Thyracian plain and the ‘Anatolian diagonal’. Consequently
we put forward two additional hypotheses: (1) that the low
laying areas around the Sea of Marmara and adjacent
Thyracian plain (,1000 m) with its Mediterranean climate
have presented an ecological barrier to European/Asian
migrations; (2) that the high mountains of the ‘Anatolian diagonal’ have allowed population exchange between Anatolia and
adjacent Caucasus and Near East mountain systems during
ecologically favourable glacial periods.
To explore these two alternative hypotheses the following
specific questions are raised. (1) Is Anatolia a centre of
global genetic diversity for A. alpina ? (2) Is the local
Anatolian diversity comprised of related haplotypes, as
expected under a scenario of in-situ survival and local
lineage accumulation? (3) Is genetic diversity geographically
partitioned around the Sea of Marmara? (4) Is there evidence
of range expansion associated with the ‘Anatolian diagonal’.
(5) In what way has the Anatolian diversity contributed to
the diversity of Europe, the Near East and eastern Africa?
These questions are addressed by a global chloroplast
DNA phylogeographic analysis of A. alpina, using the herbarium collections from the Flora of Turkey project (Davis,
1965 – 1985) as a sampling resource to place special focus
on Anatolia and associated Levant.
M AT E R I A L S A N D M E T H O D S
Leaf material from 182 individuals of Arabis alpina L., representing 135 previously unpublished localities, were sampled
from herbarium specimens deposited in the following herbaria;
Berlin-Dahlem (B), Copenhagen (C), the Natural History
Museum London (BM) and the Royal Botanical Gardens
Edinburgh (E). This sampling targeted mainly Iran, Iraq,
Lebanon, Saudi Arabia, Syria, Turkey and Yemen. To a
lesser extent, material was sampled from Armenia,
Azerbaijan, Crete, Crimea (Ukraine), Georgia, Greece,
Ethiopia, Kenya, Tanzania, Uganda and former Yugoslavian
states (Supplementary Data Table S1, available online). The
GLOBAL GAZETTER version 2.1 (http://www.fallingrain.
com/world) was used to georeference herbarium voucher
localities with the derived latitude/longitude co-ordinates
reduced to two decimal places.
Total genomic DNA was extracted from the leaf tissue using
a combination of the Retsch Tissue Lyser and Biosprint 96
Ansell et al. — Anatolian mountain plant phylogeography
243
Crimea
Thyracian plain
Pontic m
o
untains
Marmara Sea
Taurus
Western
Taurus
AntiTaurus
al
gon
ia
nD
tolia
Ana
Mt. Lebanon
range
400 km
F I G . 1. Some important Anatolian geographic features. Diagonal adapted from Davis (1971).
BioRobot (Qiagen) workstation and the Biosprint 96 plant
DNA extraction Kit (Qiagen). The trnL-trnF (trnL-F) region
including the trnL(UAA) gene and the trnL-trnF(GAA) intergenic spacer (IGS) was amplified using PCR and the ‘C’ and
‘F’ primers (Taberlet et al., 1991), according to the conditions
specified in Ansell et al. (2007). Both DNA strands were
sequenced using BigDye Terminator Kit version 1.3
(Applied Biosystems, ABI), and an ABI 3730 capillary DNA
analyser. The trnL-F IGS sequences were assembled using
SEQMAN version 6.00 (Lasergene, DNAstar).
Sequence data from published studies, especially for
European and eastern African specimens were also
utilized. The following chloroplast trnL-F IGS sequences of
Arabis alpina were recovered from GenBank: DQ060112 –
DQ060145 (Koch et al., 2006; 142 accessions world-wide,
including 122 from Europe, Greenland and Canada),
EF449508-EF449514 (Assefa et al., 2007; 66 accessions
from 11 populations in the eastern African mountains) and
EU403083– EU403084 (Ansell et al., 2008; 105 accessions
from 34 populations in the Italian mountains and European
Alps). The total dataset comprised 483 trnL-F IGS sequences
from 296 locations (Supplementary Data Table S1).
Data analysis
Sequences from the new and published studies were aligned
using Clustal X algorithm (Thompson et al., 1994) and subsequently manually corrected using MEGALIGN version
6.00 (Lasergene, DNAstar). New haplotypes types were submitted to GenBank (accession numbers JF705219– 705252).
Alignment positions 419– 435 comprised a highly variable
microsatellite C(1 – 5)T(8 – 12) (Supplementary Data Table S2),
which was excluded during haplotype discrimination. Two
haplotypes previously reported as being private to single
accessions from Croatia and Iraq (types 16 and 31; Koch
et al., 2006) where reclassified as haplotypes 13 and 04
respectively after examination of the original electropherograms. In total, 57 unique trnL-F haplotypes are recognized
for A. alpina. The relationship among haplotypes was reconstructed using statistical parsimony with a 95 % connection
threshold (Templeton, 1992) and the TCS program version
1.21 (Clement et al., 2000). All haplotypes were re-classified
to obtain a unifying scheme (Supplementary Data Table S2)
to overcome the previous adoption of both Arabic numeric
and Roman numeral haplotype naming schemes (Koch et al.,
2006; Assefa et al., 2007).
Samples were organized into ten groupings based on mountain range and observed haplotype distributions (Fig. 2A and
Supplementary Data Table S1). Haplotype richness, haplotype
diversity and nucleotide diversity was estimated for each
region.
Haplotype richness (R) with correction for the smallest
regional sample size (n ¼ 16) through rarefraction (El
Mousadik and Petit, 1996) was calculated using CONTRIB
(Petit et al., 1998). Nucleotide diversity ( p; Nei, 1987) and
genetic differentiation by analysis of molecular variance procedure (AMOVA) were calculated using ARLEQUIN version
3.10 (Excoffier et al., 2005). Differentiation was initially calculated from haplotype frequencies (FST), subsequently
incorporating haplotype pairwise genetic distances (FST), as
the number of character changes. The significance of differentiation was assessed by 1000 non-parametric permutation
tests (Excoffier et al., 1992), as performed by ARLEQUIN.
GST and NST are analogous to FST and FST (Pons and
Petit, 1996), and when the genealogical relationship of haplotypes is significantly correlated to geographic distribution of
haplotypes, estimates of NST (FST) are predicted to be significantly greater than GST (FST) (Pons and Petit, 1996).
Estimates of GST and NST were calculated using PERMUT
version 2.0 (Pons and Petit, 1996), and the significance of
difference between the two was assessed by 1000
permutations.
244
Ansell et al. — Anatolian mountain plant phylogeography
A
1
7
2
C
4
6
5
3
8
9
B
10
B11
Common haplotype (n > 25)
D
1000 km
Rare haplotype (n ≤ 12)
B8
D
G4
R3
400 km
R13
R12
B1
2
B1
3
B5
B4
B3
B2
B9
B7
B6
R2
B10
G10
B1
R1
R11
R6
0
R9
B2
9
B2
8
B27
B26
B25
R8
R7
R10
8
G8
G9
G11
G2
G6
G12
G13
G14
2
B2
3
B2
B3
G7
G3
4
B1 5
B1
B16
B17
G1
B18
B19
B2
B2 0
1
9
R4
B24
G5
R5
C
7
n = 6 per
population
4
n = 1 per
population
6
5
n = 1 per population
400 km
F I G . 2. (A) Global distribution range of Arabis alpina, and the three major chloroplast trnL-trnF IGS haplotype groups. Distribution adapted from Meusel et al.
(1965) and Jalas and Suominen (1994), with dotted lines delimiting the ten regions used for genetic analysis. (B) The genealogical relationship among the 57
unique chloroplast haplotypes, with inferred haplotype sequences shown in black. (C, D) Regional distributions of haplotypes in Asia Minor (C) and the Arabian
Peninsula and Ethiopia (D). Locations with multiple samples are indicated by black dots and connecting lines define regions: (1) Arctic and North America, (2)
western and central Europe, (3) north Africa and Macronesia, (4) south-east Europe, (5) western Anatolia, (6) central Anatolia and Levant, (7) Caucasus and
Iranian Plateau, (8) Arabian Peninsula, (9) Ethiopia, (10) east Africa. n, Sample size.
Demographic history was explored by mismatch distribution
analyses of observed haplotype pairwise differences, following
a sudden population expansion model (Rogers and
Harpending, 1992), employing 1000 bootstraps, as performed
by ARLEQUIN. Prior to estimating among haplogroup divergence time, each haplogroup was tested for deviations from
neutral expectations by Tajima’s D test (Tajima, 1989), as
implemented in DnaSP version 4.50.3 (Rozas et al., 2003),
Ansell et al. — Anatolian mountain plant phylogeography
and intraspecific recombination was tested using the SITES
software (Hey and Wakeley, 1997). Since no deviations from
neutrality and no signs of intraspecific recombination were
found, time since divergence between haplogroups was estimated by the IM program (Nielsen and Wakeley, 2001; Hey
and Nielsen, 2004). IM implements a Bayesian analysis
approach based on coalescent theory and using Markov
Chain Monte Carlo methodology to sample from a given probability distribution. It has been shown that sampling of single
haplotypes from each of many demes should yield genealogies
with similar properties as belonging to random mating populations (Wakeley, 1999, 2001; Wakeley and Aliacar,
2001).Three major haplogroups were recognized: blue (B),
green (G) and red (R) (see below). Genetically non-admixed
regions representing each haplogroup were selected to
estimate the divergence times between the genetic clusters.
The B– G-haplogroup separation was estimated by comparing
region 3 with regions 7 and 8 (Fig. 2), the B– R separation by
comparing regions 3 and 10, and the R– G-haplogroup separation by comparing regions 7 and 8 with region 10. The R –
G analysis should be treated cautiously, as the B haplogroup
is more similar to both R and G, than R and G are to each
other, violating the IM assumption that no populations
should be more genetically similar to the populations
implemented in the model than to each another.
Three separate runs were performed for each pairwise comparison, one short (run time ≥1 h) and two long (run time long
enough to achieve at least 30 million steps in chains after a
burn-in of 106 steps), with minimum effective sample size
(ESS) recorded being 138, and most ESS values being
.1000. The average B– G estimate is based on four IM
runs, the B– R estimate is based on two IM runs, while the
R– G estimate is based on four IM runs with similar prior
values but different random number seeds. Average substitution rates was set to 1.3 × 1029 per nucleotide
(Richardson et al., 2001), average generation time was arbitrarily set to 5 years (which does not influence divergence time
estimates), and historical migration rates was set to zero to
reduce the number of parameters in the model. Since no
migration is assumed, the estimated divergence times
between haplogroups should be viewed as minimum estimates.
In the initial, short IM runs, upper prior values of the mutation
and divergence time parameters were set to 100. In the two following long runs, prior values were adjusted according to
initial results, ensuring that the complete likelihood profiles
were included. Prior values for the various analyses are
shown in Supplementary Data Table S3.
R E S U LT S
Haplotype diversity and distribution
The total chloroplast trnL-F dataset comprised 483 accessions
of A. alpina sampled from 296 locations, and 71 trnL-F
types were recognized, 34 of which are newly discovered
(types 38– 71). The alignment positions 419 – 438 comprised a
complex variable microsatellite C(1 – 5)T(8 – 12) (Supplementary
Data Table S2), the size of which varied independently among
unrelated sequences. After excluding this region, 57 unique
haplotypes were recognized (Supplementary Data Table S2).
245
A network consisting of three major haplogroups was obtained
for the unique sequences and these are coloured blue (B), green
(G) and red (R) in Fig. 2.
The samples were then organized into ten groupings, reflecting mountain systems and observed haplotype distributions
(Fig. 2A and Supplementary Data Table S1). Samples from
Levant (Mt Lebanon complex) and central Anatolia were
pooled (region 6) due to a sharing of rare related green and
red haplotypes (Supplementary Data Table S1). The global
distribution of haplogroups is given in Fig. 2A, and detailed
distributions for Anatolia and eastern Africa are given in
Fig. 2C, D. The B haplogroup occurs throughout the western
Anatolian – European – Arctic parts of the species range, the
G haplogroup occurs in the eastern Anatolian – Caucasus –
Iranian Plateau – Arabian Peninsula part, but also enters the
northern Ethiopian mountains (Fig. 2D). The R haplogroup
mainly occurs in the eastern African part of the species
range (red lineage; Fig. 2B, D), but is narrowly distributed
in southern parts of central Anatolia and Levant ( pink and
light orange lineages; Fig. 2C). central Anatolia/Levant is
the only part of the species range were all three haplogroups
occur, including the three important haplotypes B1, G1 and
R1, which link the three haplogroups together (Fig. 2B, C).
Phylogeography and population genetics
The dataset was structured into ten geographic groups
(Table 1). Only four haplotypes, G6, R6, R9 and B1, were
widespread. Thirty-two of the 57 haplotypes were private to
single accessions (Appendix I), with 50 private to single
regions. Central Anatolia was the most genetically diverse
region, having the greatest haplotype diversity in absolute
terms, with 13 of 57 haplotypes, the highest haplotype richness
at R ¼ 6.378 and the highest nucleotide diversity at p ¼ 0.423
% (Table 2), exceeding the pooled global population value at
p ¼ 0.319 %. Haplotype B1 dominates throughout the
B-haplogroup range, and western Anatolia had the highest haplotype and nucleotide diversity within the B haplogroup
(Table 2). Both the haplotypic and nucleotide diversity
reduced progressively from south-east Europe towards the
Arctic (regions 4 – 1 in Fig. 2A). Haplotype G6 dominates
throughout the G-haplogroup range, with regions 7 and 8
(Fig. 2A) having similar low values of haplotype richness
and nucleotide diversity (Table 2). Haplotypes R6 and R9
are common in Ethiopia and eastern Africa, the two main
areas of the R-haplogroup range. The Ethiopian mountains
had substantially higher haplotypic and nucleotide diversity
compared with the rest of eastern Africa (Table 2), reflecting
the local mixing for G and R haplotypes (region 9; Fig 2D).
At a global scale a significant proportion of genetic
variation was partitioned among populations, and FST was
0. 438 (P ¼ 0.0001) when based solely on haplotype frequencies, and FST was 0.631 (P ¼ 0.0001) when pairwise haplotype genetic distances were also considered. GST and NST are
analogous measures to FST and FST and were estimated to
be 0.439 and 0.557, respectively, and NST was significantly
larger (P ¼ 0.05), indicating related haplotypes are geographically arranged. Again, NST (0.521) was significantly higher
than GST (0.357) for the central Anatolian diversity hotspot
and the neighbouring regions (regions 5 – 7), indicating local
.
.
1
1
.
.
1
.
.
1
.
5
.
.
16
12
.
.
.
3
.
.
.
.
.
.
.
.
.
.
.
.
1
.
.
1
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
.
haplotype diversity is geographically arranged in this critical
area (P ¼ 0.01).
Genetic differentiation among regions outside of the central
Anatolia diversity hotspot was considerable FST ¼ 0.773 (P ¼
0.0001), and this was mainly due to the differences among the
main B/G/R-haplogroup ranges FST ¼ 0.745 (P ¼ 0.0002),
rather than due to differences within B/G/R-haplogroup
ranges. This was supported by individual AMOVA tests on
the combined regions of the B-haplogroup range (FST ¼
0.041, P ¼ 0.0001), the G-haplogroup range (FST ¼ 0.006,
P ¼ 0.3450) and the R-haplogroup range (FST ¼ 0.183, P ¼
0.0001). There is little genetic differentiation among populations from south-east Europe and western Anatolia (regions
4 and 5, FST ¼ 0.095, P ¼ 0.005), which are separated by
the Sea of Marmara (Fig. 2A, C). Conversely, populations
from central Anatolia and the eastern Anatolia –Iranian
Plateaux had higher genetic differentiation (regions 6 and 7;
FST ¼ 0.333, P ¼ 0.000), despite greater geographic connectivity via the Anatolian diagonal mountain systems.
.
Demographic history and divergence
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
.
.
.
1
11
.
.
.
.
.
.
.
.
.
1
.
.
.
.
.
.
.
.
.
2
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
.
.
.
.
.
.
1
.
.
.
.
.
.
.
.
.
1
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
6
.
.
.
.
.
1
.
.
.
2
.
.
.
.
.
1
.
.
.
.
Separate mismatch analyses on the B and G haplogroups,
including the representative Anatolia diversity, did not reject
the sudden expansion model (P ¼ 0.710, P ¼ 0.660), and
both followed a uni-modal mismatch distribution (Fig. 3A,
B). The sudden expansion model was rejected for
R-haplogroup diversity (P ¼ 0.050; Fig. 3C), and was also
rejected when the African R-haplotype diversity was analysed
separately (P ¼ 0.001; Fig. 3D). Mismatch analysis on the data
derived from within Anatolia did not reject the sudden expansion model for either the G haplogroup (P ¼ 1.000), or the B
haplogroup (P ¼ 0.270). Insufficient data (n ¼ 6) prevented a
meaningful test for the Anatolia R-haplogroup diversity.
The IM analyses resulted in convergence of parameter estimates, with high ESS values, almost identical distributions of
all pairs of long-runs, and posterior estimates not near the edge
of uniform prior values. The divergence time probability distributions were estimated for all pairwise combinations of haplogroups (Fig. 4): the B– G haplogroups had the highest
probability for t ¼ 0.66 Mya (averaged over four runs, 95 %
CI between 0.25 and 3.86 Mya), the B– R haplogroups had
the highest probability for t ¼ 0.73 Mya (averaged over two
runs 95 % CI between 0.23 and 2.48 Mya), and the highest
probability for the R– G-haplogroup separation was t ¼ 2.69
Mya (averaged over four runs 95 % CI between 0.80 and
6.43 Mya). The latter included the haplo-groups with the greatest nucleotide distances, and therefore represents an estimate
for the overall age of extant haplotype diversification.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
.
.
.
.
.
.
1
.
.
.
.
.
.
.
.
.
16
44
39
.
.
1
.
2
20
118
12
7
.
26
22
65
Arctic
Western and
central Europe
North Africa and
Macronesia
South-east.Europe
Western Anatolia
Central Anatolia
Eastern Anatolia
and Iranian
Plataeux
Arabian penisula
Ethiopia
EasternAfrica
149
21
44
58
.
.
21
52
6
.
.
.
.
.
.
.
.
.
.
1
1
.
.
.
.
.
.
.
.
.
.
.
2
1
.
2
.
.
.
.
.
1
.
.
.
.
1
.
.
.
.
2
.
.
.
.
.
.
.
.
5
.
.
.
.
.
1
.
.
.
5
.
.
.
.
.
2
.
.
.
.
.
1
.
.
1
.
.
1
.
.
.
.
.
1
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
.
.
.
.
.
.
3
.
2
.
.
.
14
11
.
.
.
.
.
2
.
.
.
.
1
.
.
.
1
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
.
.
.
1
.
.
.
.
.
1
.
.
.
1
.
.
.
.
.
.
18
52
.
.
.
.
.
.
.
.
.
.
2
.
.
.
.
1
.
.
.
.
.
7
21
.
.
.
.
.
.
.
.
.
.
.
2
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
R12
R11
R10
R9
R8
R7
R6
R5
R4
R3
R2
R1
G14
G13
G12
G11
G10
G9
G8
G7
G6
G5
G4
G3
G2
G1
B30
B29
B28
B27
B26
B25
B24
B23
B22
B21
B20
B19
B18
B17
B16
B15
B14
B13
B12
B11
B10
B9
B8
B7
B6
B5
B4
B3
B2
B1
n
Region
TA B L E 1. Distribution of haplotypes among the ten geographic regions and numbers of individuals (n) used for genetic analysis
.
.
Ansell et al. — Anatolian mountain plant phylogeography
R13
246
D IS C US S IO N
This study confirms the importance of Anatolia and associated
Levant in understanding the global diversification history of
the arctic –alpine Arabis alpina. From the dense sampling in
the present study, extensive genetic diversity was recovered,
including substantial new chloroplast sequence variation not
recovered by previous studies (Koch et al., 2006; Assefa
et al., 2007; Ansell et al., 2008), despite these studies representing much of the world-wide range. In total the number
Ansell et al. — Anatolian mountain plant phylogeography
247
TA B L E 2. Distribution of haplotype diversity by sampling region
Haplogroup frequencies
n blue
¼
n green
¼
n red
¼
R
Total (n
haps)
n haps private
individual
n haps
private
regional
% abundence n haps
regionally private
n
(1) Arctic and North
America
(2) Western-central
European
(3) North Africa and
Macronesia
(4) South-eastern
European
(5) Western Anatolia
(6) Central Anatolia and
Levant
(7) Caucasus and Iranian
Plateau
(8) Arabian Peninsula
(9) Ethiopia
(10) Eastern Africa
Blue regions 1 –5
Green regions 7– 8
Red regions 9–10
Red regions 9–10 (no
green haplotypes)
World-wide (regions
1– 10)
22
22
0
0
2
1
1
4.545
0.727
–
0.013
65
65
0
0
7
3
6
20.000
2.438
–
0.060
26
26
0
0
6
4
3
11.538
3.323
–
0.077
149
149
0
0
14
8
12
13.422
2.698
–
0.083
21
44
21
8
0
30
0
6
6
13
4
6
5
10
42.857
40.909
4.048
6.378
–
–
0.110
0.423
58
0
58
0
6
4
5
10.344
1.583
–
0.034
16
44
39
283
74
83
70
0
0
0
283
0
0
0
16
13
0
0
74
13
0
0
31
39
0
0
70
70
3
7
6
31
8
10
8
2
3
2
20
6
2
1
2
3
3
28
8
9
8
12.500
20.000
13.158
20.922
9.459
86.747
100.000
2.000
4.440
3.037
–
–
–
–
484
291
117
76
57
32
–
–
16
–
74
p
Region
–
–
–
11.314
7.000
8.750
–
–
0.036
0.430
0.113
0.077
0.035
0.313
0.130
0.3187
Sample size (n), frequencies of haplotypes by haplotype clades, numbers of haplotypes, private haplotypes, haplotype diversity by rarefraction (R) for
corrected for smallest sample size at regional (n ¼ 16) and haplogroup levels (n ¼ 74), nucleotide diversity ( p).
Regions 1 –10 are shown in Fig. 2.
of haplotypes increased from 31 to 57 (Supplementary Data
Table S2) and, remarkably, on average one new haplotype
was discovered for every 8.5 samples at the global scale,
increasing to every 3.6 samples in Anatolia. All three previously described haplogroups and their global distributions
were recovered (Koch et al., 2006; Assefa et al., 2007), and
their distributions were confirmed to extend into the interior
of Anatolia. Furthermore, it was shown that the genotypic
and nucleotide diversity of Anatolia, especially central
Anatolia (including Levant), exceeded that of all other areas.
Centres of genetic diversity have often been associated with
range expansion and population amalgamation (Petit et al.,
2003; Walter and Epperson, 2005). The present results,
however, strongly indicate that Anatolia is the cradle of diversity. First, there was no narrowly delimitated range of haplogroup frequency shift, as typified from hybrid zones of
trees (Ferris et al., 1998; Bartish et al., 2006), or by grasshoppers, shrews and various bats from central Anatolia, which
have divergent European and Near East lineages (Cooper
et al., 1995; Bilgin et al., 2006, 2008; Furman et al., 2009).
Secondly, the Anatolia haplotype diversity includes the most
common European and Near East haplotypes (B1, G6), plus
all intermediate haplotypes linking the European, Near East
and eastern African lineages (Fig. 2B, C). Of the remaining
genealogically important haplotypes, only the red eastern
African diversity is missing from Anatolia, and these probably
originated in situ due to geographic isolation from the main
species range (Koch et al., 2006; Assefa et al., 2007). Hence
the prsent results indicate thatAnatolia is the centre of origin
for A. alpina, with a long history of lineage accumulation,
functioning as a major Pleistocene refugium.
Species origin
The evidence presented is consistent with previous hypotheses of (1) Anatolia being the cradle for Brassicaceae diversification (Koch and Kiefer, 2006; Franzke et al., 2009) and
more specifically, (2) that the genus Arabis originated in the
Mediterranean or Turko-Iranian areas (Meusel et al., 1965;
Hedge, 1976). A molecular phylogeny of the genus Arabis
(R. Karl et al., University of Heidelberg, Germany, unpubl.
res.) supports these conclusions and confirms the B haplogroup
is ancestral for A. alpina. Within the B-haplogroup range
western-Anatolia had the highest genetic diversity (Table 2),
suggesting that Arabis alpina may have originated in this
region. An alternative ‘out of Africa’ scenario is not supported,
as the ancestral R-haplotype (R1) and descendents (R2 – R5)
are private to the areas covering central Anatolia and Levant.
For alpine species, the availability of cold habitats are influential on species and population ranges and the accumulation
of genetic diversity (Comes and Kadereit, 2003). Using
coalescence analysis, it was estimated that divergence
between the most distantly related haplogroups (R, G) commenced around 2.69 Mya (95 % CI 0.80– 6.43 Mya). These
results must be interpreted cautiously, reflecting the single
gene approach of the present study, and the violation of IM
assumption, specifically that no other populations are more
closely related to the sampled populations than they are to
248
Ansell et al. — Anatolian mountain plant phylogeography
Frequency (absolute)
30 000
A
for the separation of Euro-Anatolian/Near Eastern lineages
of yellow-necked house mouse, the white-toothed shrew and
oak gallwasps (Michaux et al., 2004; Dubey et al., 2006;
Challis et al., 2007), suggesting the widespread involvement
of early Middle Pleistocene climate fluctuation on the diversification of Anatolian biota.
Observed
Expected
25 000
20 000
15 000
10 000
5000
0
1
2
3
4
5
6
7
Pleistocene glacial survival and the accumulation of diversity
Frequency (absolute)
Pairwise difference
5000
B
4500
4000
3500
3000
2500
2000
1500
1000
500
0
1
2
3
4
5
6
7
8
9
7
8
9
Frequency (absolute)
Pairwise difference
1400
1200
C
1000
800
600
400
200
0
1
2
3
4
5
6
Frequency (absolute)
Pairwise difference
1400
1200 D
1000
800
600
400
200
0
1
2
3
4
Pairwise difference
5
6
F I G . 3. Mismatch distribution analyses of pairwise haplotype differences,
observed values, expected values (as indicated), for European (blue) lineages
(A), Near East (green) lineages (B), Asian–African (red) lineages (C) and
African only (red) lineages (D).
each other (R – G vs. R-B or G-B comparisons). Interestingly,
our estimate is similar to a previous estimate of 2.1 Mya estimated from synonymous mutation rates (Koch et al., 2006),
and approximates to an origin at the Pliocene –Pleistocene
transition, when the climate rapidly cooled (Webb and
Bartlein, 1992) and suitable alpine growing conditions would
have become more widespread. The subsequent haplogroup
diversifications were estimated around 0.70 Mya and roughly
coincided with the onset of the Cromerian Complex of glaciation cycles (approx. 300 –850 Kya), when the Anatolian
climate was 4– 5 8C cooler than at present (Erinç, 1978;
Furman et al., 2009). Similar divergence times were derived
Anatolia’s dissected mountainous landscape has been
crucial for the long-term survival of A. alpina and its ability
to accumulate genetic diversity. There was no major
Pleistocene ice-shield equivalent to those covering the
European Alps or Scandinavia and only the higher mountain
peaks above 2200 m were glaciated (Erinç, 1978; Atalay,
1996), providing opportunities for local population survival.
Furthermore, the mountain systems provided moist conditions
at higher elevations, offsetting the drier climate that prevailed
during the glacial periods (Webb and Bartlein, 1992). Alpine
species like A. alpina would have potentially had broad
glacial distributions in Anatolia, albeit around the alpine belt
as a fragmented network of local survival centres. This is
especially likely in the more moist coastal Taurus Mountains
of southern Anatolia (Davis, 1971; Ekim and Güner, 1986).
The situation would have permitted localized altitudinal
migration responses to the Pleistocene temperature fluctuations
within a shifting alpine belt, thereby mostly retaining local
diversity structures. This is analogous to the ‘phalanx’
migration model (Ibrahim et al., 1996) more often applied to
the survival of temperate biota in the southern European
mountain refugia (Hewitt, 2004; Schmitt, 2007).
For European alpines there are extensive palaeo-ecological,
geological and fossil data to support the developed phylogeographic scenarios (Tribsch and Schönswetter, 2003; Comes
and Kadereit, 2003). Similar data are currently lacking for
Anatolian alpines, and we must rely on genetic data and
Pleistocene glacial dynamics to support our hypothesis. It is
reasonable to assume that the availability of suitable habitats
becomes limited for alpines during the warm inter-glacial
phases in Anatolia. Regular range fragmentation during the
inter-glacials (geographic isolation) may potentially restrict
genetic exchange and promote the accumulation of local genotypes among the surviving populations. This should result in
the widespread distribution of local genotypes and for related
genotypes to be spatially clustered. Haplotypes private to
single localities were scattered throughout the Anatolian
range of A. alpina (Fig. 2C), and genealogically related haplotypes (G2 –5, R2 – 4) were locally arranged in the southern
Anti-Taurus and Mount Lebanon complexes (Fig. 2C).
Notably, haplotype B26 was distributed among the southern
Taurus Mountain localities (Fig. 2C), which are also important
for their endemic Arabis species (R. Karl et al., University of
Heidelberg, Germany, unpubl. res.) and locally differentiated
lineages of Lebanese cedar and ground squirrels (Gündüz
et al., 2007; Fady et al., 2008). Furthermore, the southern
Anatolian systems were recently recognized as molecular
diversity hotspots (Médail et al., 2009), and local ‘microrefugia’ for pond turtles (Fritz et al., 2009), again consistent
with a hypothesis of multiple survival centres.
Ansell et al. — Anatolian mountain plant phylogeography
249
4·5
Blue region 3–Green region
Likelihood (×10–3)
4·0
Blue region 3–Green region
3·5
Blue region 3–Red region
3·0
Green region 7–Red region
2·5
Green region 8–Red region
2·0
1·5
1·0
0·5
0
0
1
2
3
4
5
6
7
8
9
Time since divergence (t )
F I G . 4. Results of the IM runs, comparing divergence time estimates between haplogroup pairings as indicated, based on the sampling regions outlined
in Fig. 2A, and haplotype information given in Table 1. Time since divergence is measured in millions of years since present.
Out of Anatolia
Most strikingly the internal haplogroup arrangement of
Anatolia closely resembles the wider global pattern for
A. alpina (Koch et al., 2006) (Fig. 2C). Hence, we argue
that the global haplogroup arrangement was achieved by
expansions from already internally structured Anatolian diversity. This ‘inside-out’ scenario is compatible with an existing
‘three-times out of Asia Minor’ hypothesis for A. alpina
(Koch et al., 2006), but we note that the G1-dominated
Arabian Peninsula and Caucasus/Iranian Plateau are separated
by G2- to G5-dominated intervening Anti-Taurus/Mt Lebanon
ranges (Fig. 2C). This suggests these regions were independently colonized, and thus a ‘four-times out of Anatolia’
scenario seems likely.
Two geographic features are widely acknowledged to have
influenced the expansions of plants and animals around
Anatolia: the Sea of Marmara and associated Thyracian
Plain in European Turkey, and the high ‘Anatolian diagonal’
mountains in the east (Davis, 1971; Ekim and Güner, 1986;
Rokas, et al., 2003; Bilgin et al., 2009). The hills around the
Sea of Marmara reach only 1100 m, and currently have a
Mediterranean climate and flora (Davis, 1965 – 85), but dry
steppe communities dominated this area during Pleistocene
glaciations (Michaux et al., 2004). Pleistocene land bridges
regularly formed across the Sea of Marmara (Kerey et al.,
2004; Magyari et al., 2008) and, for some temperate animal
species there is clear evidence of inter-continental gene flow
(Dubey et al., 2007; Furman et al., 2009; Stamatis et al.,
2009). In contrast, for alpine plant species, the current and
past ecological conditions around the Sea of Marmara
potentially represent a substantial geographic/ecological
barrier for inter-continental gene flow. Surprisingly, related
B-haplotypes were found distributed in both western
Anatolia and the southern Balkans and there is negligible
genetic differentiation between these two areas (FST ¼
0.095). Hence we are forced to conclude that the Sea of
Marmara region has not been a historical ecological barrier
for A. alpina. Inter-continental genetic exchange most likely
occurred during favourable glacial conditions, when the
alpine habitats were presumably at lower elevations, creating
greater ecological connectivity between the two continents.
Furthermore, the progressive loss of nucleotide diversity
from western Anatolia to northern Europe (Table 2) suggests
that gene flow has predominantly occurred from Anatolia
into Europe.
The mountains of the ‘Anatolian diagonal’ in eastern
Anatolia are substantially different (.2000 m), providing
extensive areas for species preferring cool moist climates to
grow at higher elevations. Indeed, the diagonal is characterized
by numerous plant and insect endemics (Davis, 1971; Ekim
and Güner, 1986; Çiplak, 2003), which may have persisted
locally or nearby during recent glaciations (Gündüz et al.,
2007). Importantly the diagonal mountains have a north-east/
south-west orientation, whereas most Anatolian mountain
systems have east– west orientations, and this provides a
continuous mountain chain connection between the
Mediterranean and Caspian seas. This links the southern
Taurus diversity hotspot (Médail and Diadema, 2009) to
nearby Caucasus and Near East mountain systems. Thus, we
postulate the diagonal has been a migratory corridor for
A. alpina range expansion out of Anatolia during favourable
glacial periods. This is supported by populations near to the
Mediterranean Sea harbouring haplotypes sister to those commonly detected from populations in the Caucasus and Iranian
Plateau. Mismatch tests further indicated a history of sudden
range expansion for populations encompassing the diagonal
and the adjacent eastern Anatolia. We note that for several
temperate animal species the diagonal does function as a
barrier to lineage mixing (Rokas, et al., 2003; Dubey et al.,
2006; Bilgin et al., 2009), and the present contrasting results
for an alpine species highlight the importance of individual
ecologies when interpreting mountains as barriers or corridors.
Conclusions
Despite the dominating influence of mountain systems on
the biogeographic structure of Anatolia, the phylogeographic
histories of mountain species are poorly known in this
250
Ansell et al. — Anatolian mountain plant phylogeography
biodiversity hotspot. The prsent study demonstrates the importance of the Anatolia mountains as the centre for global diversification for arctic – alpine A. alpina, that the local Anatolian
diversity pattern mirrors the global pattern, and that local
Pleistocene population history has left a genetic imprint on
the global population structure. Relatively dense and even
sampling was crucial for recovering critical genotypes in
southern Anatolia to substantiate these findings. Currently
there are few detailed studies of intra-specific genetic diversity
from other plants to reliably assess Anatolia’s Pleistocene
history as a centre of gene pool amalgamation or centre of
diversity (Kučera et al., 2006; Gömöry et al., 2007;
Naydenov et al., 2007). We were fortunate to have access to
the large herbarium legacy of the Flora Turkey project
(Davis, 1965 – 85) as a molecular resource to discriminate
between these alternative interpretations for A. alpina.
Consequently our study also illustrates the importance of
sampling and scaling effects when conducting phylogeographic analysis, especially in complex regions of lineage
and flora mixing.
S U P P L E M E NTA RY D ATA
Supplementary data are available online at www.aob.oxford
journals.org and consist of the following. Table S1:
Sampling information. Table S2: Haplotype alignments and
classifications.Table S3: Prior values used for the various IM
program analyses.
AC KN OW LED GEMEN T S
We thank Mary Gibby and Andreas Hemp for collecting
material in Yemen and eastern Africa, Cathy Smith for
geo-referencing, Bob Press for comments on the manuscript
development, Maria Albani and George Coupland for
support during fieldwork in Greece, and Julia LlewellynHughes and her team for the NHM DNA sequencing
service. We are grateful to the curators of the herbaria B,
BM, CO and E for allowing us to use leaf samples from herbaria sheets in our study. The project was financially supported
by the Department Botany at the NHM London, and by the
Swedish and Norwegian Research Councils to H.K.S.
L I T E R AT U R E CI T E D
Abbott RJ, Brochmann C. 2003. History and evolution of the arctic flora: in
the footsteps of Eric Hultén. Molecular Ecology 12: 299– 313.
Ansell SW, Schneider H, Pedersen N, Grundmann M, Russell SJ, Vogel
JC. 2007. Recombination diversifies chloroplast trnF pseudogenes in
Arabidopsis lyrata. Journal of Evolutionary Biology 29: 2400– 2411.
Ansell SW, Grundmann M, Russell SJ, Schneider H, Vogel JC. 2008.
Genetic discontinuity, breeding system change and population history
of Arabis alpina in the Italian Peninsula and adjacent Alps. Molecular
Ecology 17: 2245– 2257.
Assefa A, Ehrich D, Taberlet P, Nemomissa S, Brochmann C. 2007.
Pleistocene colonization of afro-alpine ‘sky islands’ by the arctic-alpine
Arabis alpina. Heredity 99: 133–142.
Atalay I. 1996. Palaeosoils as indicators of the climatic changes during
Quaternary period in S. Anatolia. Journal of Arid Environments 32:
23– 35.
Bartish IV, Kadereit JW, Comes HP. 2006. Late Quaternary history of
Hippophaë rhamnoides L. (Elaeagnaceae) inferred from chalcone
synthase intron (chsi) sequences and chloroplast DNA varation.
Molecular Ecology 15: 4065–4083.
Beilstein MA, Al-Shehbaz IA, Mathews S, Kellogg EA. 2008. Brassicaceae
phylogeny inferred from phytochrome A and ndhF sequence data: tribes
and trichomes revisited. Annals of Botany 95: 1307– 1327.
Bilgin R, Karataş A, Çoraman E, Pandurski I, Papadatou E, Morales JC.
2006. Molecular taxonomy and phylogeography of Miniopterus schreibersii (Kuhl, 1817) (Chiroptera: Vespertilionidae), in the Eurasian transition. Biological Journal of the Linnean Society 87: 577–582.
Bilgin R, Furman A, Çoraman E, Karataş A. 2008. Phylogeography of the
Mediterranean horseshoe bat, Rhinolophus euryale (Chiroptera:
Rhinolophidae), in southeastern Europe and Anatolia. Acta
Chiropterologica 10: 41– 49.
Bilgin R, Karatas A, Çoraman E, Disotell T, Morales JC. 2009. Regionally
and climatically restricted patterns of distribution of genetic diversity
in a migratory bat species, Miniopteris schreibersii (Chiroptera:
Vespertilionidae). BMC Evolutionary Biology 8: 209. doi:10.1186/
1471-2148-8-209.
Challis RJ, Mutun S, Nieves-Aldrey J-L, et al. 2007. Longitudinal range
expansion and cryptic eastern species in the western Palaearctic oak gallwasp, Andricus coriarius. Molecular Ecology 16: 2103–2114.
Çiplak B. 2003. Distribution of Tettigoniinae (Orthoptera, Tettigoniidae)
bush-crickerts in Turkey: the importance of the Anatolian Taurus
Mountains in biodiversity and implications for conservation.
Biodiversity and Conservation 12: 47– 64.
Clement M, Posada D, Crandall KA. 2000. TCS: a computer program to estimate gene genealogies. Molecular Ecology 9: 1657– 1660.
Comes HP, Kaderit JW. 2003. Spatial and temporal patterns in the evolution
of the flora of the European alpine system. Taxon 52: 451–462.
Cooper SJB, Ibrahim KM, Hewitt GM. 1995. Postglacial expansion and
genome subdivision in the European Grasshopper Chorthippus parallelus.
Molecular Ecology 4: 49–60.
Couvreur TLP, Franzke A, Al-Shehbaz IA, Bakker FT, Koch MA,
Mummenhoff K. 2010. Molecular phylogenetics, temporal diversification and principles of evolution in the mustard family (Brassicaceae).
Molecular Biology & Evolution 27: 55–71.
Davis PH. 1965–85. Flora of Turkey and the east Aegean islands, Vols 1– 9.
Edinburgh: University Press Edinburgh.
Davis PH. 1971. Distribution patterns in Anatolia with particular reference to
endemism. In: Davis PH, Harper PC, Hedge IC. eds. Plant life of southwest Asia. Aberdeen: University Press Aberdeen, 15– 28.
Dubey S, Zaitsev K, Cosson J-F, Abdukadier A, Vogel P. 2006. Pliocene
and Pleistocene diversification and multiple refugia in a Eurasian shrew
(Crocidura suaveolens group). Molecular Phylogenetics and Evolution
38: 635 –647.
Dubey S, Cosson J-F, Vohralı́k V, Kryštufek B, Diker E, Vogel P. 2007.
Evidence of Pleistocene bidirectional faunal exchange between Europe
and the Near East: the case of the bicoloured shrew (Crocidura leucodon,
Soricidae). Journal of Evolutionary Biology 20: 1799– 1808.
Ehrich D, Gaudel M, Assefa A, et al. 2007. Genetic consequences of
Pleistocene range shifts: contrast between the Arctic, the Alps and the
East African mountains. Molecular Ecology 16: 2542–2559.
El Mousadik A, Petit RJ. 1996. High level of genetic differentiation for
allelic richness among populations of the Argan tree [Argania spinosa
(L.) Skeels] endemic to Morocco. Theoretical and Applied Genetics 92:
832– 839.
Ekim T, Güner A. 1986. The Anatolian Diagonal: fact or fiction? Proceedings
of the Royal Society of Edinburgh 89B: 69–77.
Erinç S. 1978. Changes in the physical environment in Turkey since the end of
the last glacial. In: Brice WC. ed. The environmental history of the Near
and Middle East since the last Ice Age. London: Academic Press,
87–110.
Excoffier L, Smouse P, Quattro J. 1992. Analysis of molecular variance
inferred from metric distances among DNA haplotypes: applications to
human mitochondrial DNA restriction data. Genetics 131: 479–491.
Excoffier L, Laval G, Schneider S. 2005. ARLEQUIN (version 3.0): an intregated software package for population genetic data analysis. Evolutionary
Bioinformatics Online 1: 47– 50.
Fady B, Lefèvre F, Vendramin GG, Ambert A, Régnier C, Bariteau M.
2008. Genetic consequences of past climate and human impact on
eastern Mediterranean Cedrus libani forests: impacts for their conservation. Conservation Genetics 9: 85–95.
Ansell et al. — Anatolian mountain plant phylogeography
Ferris C, King RA, Vainola R, Hewitt GM. 1998. Chloroplast DNA recognizes three refugial sources of European oaks and suggests independent
eastern and western immigrations to Finland. Heredity 80: 584–593.
Flanders J, Jones G, Benda P, et al. 2009. Phylogeography of the greater
horseshoe bat, Rhinolophus ferrumequinum: contrasting results from
mitochondrial and microsatellite data. Molecular Ecology 18: 306–318.
Font M, Garcia-Jacas N, Vilatersana R, Roquet C, Susanna A. 2009.
Evolution and biogeography of Centaurea section Acrocentron inferred
from nuclear and plastid DNA sequence analyses. Annals of Botany
103: 985–997.
Franzke A, German D, Al-Shehbaz IA, Mummenhoff K. 2009. Arabidopsis
family ties: molecular phylogeny and age estimates in Brassicaceae.
Taxon 58: 425– 437.
Fritz U, Ayaz D, Hundsdörfer AK, et al. 2009. Mitochondrial diversity of
European pond turtles (Emys orbicularis) in Anatolia and
Ponto-Caspian region: multiple old refuges, hotspots of extant diversification and critically endangered endemics. Organisms, Diversity and
Evolution 9: 100–114.
Furman A, Çoraman E, Bilgin R, Karataş A. 2009. Molecular ecology and
phylogeography of the bent-winged bat complex (Miniopterus schreibersii) (Chiroptera: Vespertilionidae) in Asia Minor and adjacent regions.
Zoologica Scripta 38: 129– 141.
Gömöry D, Paule L, Vyšny J. 2007. Patterns of allozyme variation in western
Eurasian Fagus. Botanical Journal of the Linnean Society 154: 165– 174.
Gündüz I, Jaarola M, Tez C, Yeniyurt C, Polly PD, Searle JB. 2007.
Multigeneic and morphological differentiation of ground squirrels
(Spermophilus, Scuiridae, Rodentia) in Turkey, with a description of a
new species. Molecular Phylogenetics and Evolution 43: 916–935.
Hedge IC. 1976. A systematic and geographical survey of the old world cruciferae. In: Vaughan JG, Macleod AJ. eds. The biology and chemistry of
the Cruciferae. London: Academic Press, 1 –45.
Hegi G. 1986. Illustrierte Flora von Mittel-Europa. Pteridophyta und
Spermatophyta. Band 4. Angiosperma Dicotyledones 2, Teil
1. München: Carl Hanser, 230– 301.
Hewitt GM. 2000. The genetic legacy of Quaternary ice ages. Nature 405:
907– 913.
Hewitt GM. 2004. Genetic consequences of climatic oscillations in the
Quaternary. Proceedings of the Royal Society of London Series B 359:
183– 195.
Hey J, Nielsen R. 2004. Multilocus methods for estimating population sizes,
migration rates and divergence time, with applications to the divergence
of Drosophila pseudoobscura and D. persimilis. Genetics 167: 747– 760.
Hey J, Wakeley J. 1997. A coalescent estimator of the population recombination rate. Genetics 145: 833– 846.
Ibrahim K, Nichols RA, Hewitt GM. 1996. Spatial patterns of genetic variation generated by different forms of dispersal during range expansion.
Heredity 77: 282– 291.
Jakob S, Ihlow A, Blattner FR. 2007. Combined ecological niche modelling
and molecular biogeography revealed the evolutionary history of
Hordeum marinum (Poaceae) – niche differentiation, loss of genetic
diversity, and speciation in Mediterranean Quaternary refugia.
Molecular Ecology 16: 1713–1727.
Jalas J, Suominen J. 1994. Atlas florae Europaeae, Vol. 10. Helsinki:
University of Helsinki Press.
Kerey IE, Meric E, Kelling G, Brenner RA, Dogan AU. 2004. Black
Sea-Marmara Sea Quaternary connections: new data from the Bosporus,
Istanbul, Turkey. Paleogeography, Paleoclimatology and Paleoecology
204: 277–295.
Koch M, Kiefer C. 2006. Molecules and migration: biogeographical studies in
cruciferous plants. Plant Systematics and Evolution 259: 121–142.
Koch MA, Haubold B, Mitchell-Olds T. 2000. Comparative evolutionary
analysis of the chalcone synthase and alcohol dehydrogenase loci
among different lineages Arabidopsis, Arabis, and related genera
(Brassicaceae). Molecular Biology and Evolution 17: 1483–1498.
Koch MA, Haubold B, Mitchell-Olds T. 2001. Molecular systematics of the
cruciferae: evidence from coding plastome matK and nuclear CHS
sequences. American Journal of Botany 88: 534– 544.
Koch MA, Kiefer C, Ehrich D, Vogel J, Brochmann C, Mummenhoff K.
2006. Three times out of Asia Minor: the phylogeography of Arabis
alpina L. (Brassicaceae). Molecular Ecology 15: 825– 839.
Koch MA, Dobeš C, Kiefer C, Shmickl R, Klimes L, Lysak MA. 2007.
SuperNetwork identifies multiple events of plastid trnF (GAA)
251
pseudogene evolution in the Brassicaceae. Molecular Biology and
Evolution 24: 63–73.
Kučera J, Lihova J, Marhold K. 2006. Taxonomy and phylogeography of
Cardamine impatiens and C. petinata (Brassicaceae). Botanical Journal
of the Linnean Society 152: 169–195.
Magyari EK, Chapman JC, Gaydarska B, et al. 2008. The ‘oriental’ component of the Balkan flora: evidence of presence on the Thracian Plain
during the Weichselian late-glacial. Journal of Biogeography 35:
865–883.
Médail F, Diadema K. 2009. Glacial refugia influence plant diversity in the
Mediterranean basin. Journal of Biogeography 36: 1333–1345.
Meusel H, Jäger E, Weinert E. 1965. Vergleichende Chorologie der zentraleuropäischen Flora. Jena: Gustav Fischer Verlag, 190.
Michaux JR, Libois R, Paradis E, Filippucci M-G. 2004. Phylogeographic
history of the yellow-necked fieldmouse (Apodemus flavicollis) in Europe
and in the Near and Middle East. Molecular Phylogeography and
Evolution 32: 788–798.
Naydenov K, Senneville S, Beaulieu J, Tremblay F, Bousquet J. 2007.
Glacial vicariance in Eurasia: mitochondrial DNA evidence from Scots
pine for a complex heritage involving genetically distinct refugia at midnorthern Latitudes and in Asia Minor. BMC Evolutionary Biology 7: 233.
doi:10.1186/1471-2148-7-233.
Nei M. 1987. Molecular evolutionary genetics. New York, NY: Columbia
University Press.
Nielsen R, Wakeley J. 2001. Distinguishing migration from isolation: a
Markov chain Monte Carlo approach. Genetics 158: 885– 896.
Parolly G, Nordt B, Bleeker W, Mummenhoff K. 2010. Heldreichia Boiss.
(Brassicaceae) revisited: a morphological and molecular study. Taxon 59:
187–202.
Petit RJ, El-Mousadik A, Pons O. 1998. Identifying populations for conservation on the basis of genetic markers. Conservation Biology 12:
844–855.
Petit RJ, Aguinagalde I, De Beaulieu J-L, et al. 2003. Glacial refugia: hotspots but not melting pots of genetic diversity. Science 300: 1563–1565.
Poncet BN, Herrmann D, Gugerli F, et al. 2010. Tracking genes of ecological relevance using a genome scan in two independent regional population samples of Arabis alpina. Molecular Ecology 19: 2896– 2907.
Pons O, Petit RJ. 1996. Measuring and testing genetic differentiation with
ordered versus unordered alleles. Genetics 144: 1237– 1245.
Richardson JE, Pennington RT, Pennington TD, Hollingsworth PM. 2001.
Rapid diversification of a species-rich genus of neotropical rain forest
trees. Science 293: 2242– 2245.
Ricklefs RE. 2007. History and diversity: explorations at the intersection of
ecology and evolution. American Naturalist 170: S56–70.
Rogers AR, Harpending H. 1992. Population growth makes waves in the distribution of pairwise genetic differences. Molecular Biology and
Evolution 9: 552 –569.
Rokas A, Atkinson RJ, Webster LMI, Csoka G, Stone GN. 2003. Out of
Anatolia: longitudinal gradients in genetic diversity support an eastern
origin for a circum-Mediterranean oak gallwasp Andricus quercustozae.
Molecular Ecology 12: 2153– 2174.
Rozas J, Sánchez-DelBarrio JC, Messeguer X, Rozas R. 2003. DnaSP,
DNA polymorphism analyses by the coalescent and other methods.
Bioinformatics 19: 2496–2497.
Schmitt T. 2007. Molecular biogeography of Europe: Pleistocene cycles and
postglacial trends. Frontiers in Zoology 4: 11. doi:10.1186/
1742-9994-4-11.
Stamatis C, Suchentrunk F, Moutous KA, et al. 2009. Phyloegography of
the brown hare (Lepus europaeus) in Europe: a legacy of south-eastern
Mediterranean refugia? Journal of Biogeography 36: 515–528.
Taberlet P. 1998. Biodiversity at the intraspecific level: the comparative phylogeographic approach. Journal of Biotechnology 64: 91–100.
Taberlet P, Gielly L, Pautou G, Bouvet J. 1991. Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular
Biology 17: 1105– 1109.
Tajima F. 1989. Statistical methods to test for nucleotide mutation hypothesis
by DNA polymorphism. Genetics 123: 585–595.
Tedder A, Ansell SW, Lao X, Vogel JC, Mable BK. 2011. Sporophytic selfincompatibility genes and mating system variation in Arabis alpina.
Annals of Botany 108: in press. doi:10.1093/aob/mcr157.
Templeton AR, Crandall KA, Sing CF. 1992. A cladistic analysis of phenotypic associations with haplotypes inferred from restriction endonuclease
252
Ansell et al. — Anatolian mountain plant phylogeography
mapping and DNA sequence data. III. Cladogram estimation. Genetics
132: 619– 633.
Thompson JD, Higgins DG, Gibson TJ. 1994. CLUSTAL W: improving the
sensitivity of progressive multiple sequence alignment through sequence
weighting, position-specific gap penalties and weight matrix choice.
Nucleic Acids Research 22: 4673–4680.
Tribsch A, Schönswetter P. 2003. Patterns of endemism and comparative
phylogeography confirm palaeo-environmental evidence for Pleistocene
refugia in the Eastern Alps. Taxon 52: 477–497.
Wakeley J. 1999. Nonequilibrium migration in human history. Genetics
153: 1863–1871.
Wakeley J. 2001. The coalescent in an island model of population subdivision
with variation among demes. Theoretical Population Biology 59: 133– 144.
Wakeley J, Aliacar N. 2001. Gene genealogies in a metapopulation. Genetics
159: 893–905.
Walter R, Epperson BK. 2005. Geographic pattern of genetic diversity in
Pinus resinosa: contact zone between descendants of glacial refugia.
American Journal of Botany 92: 92– 100.
Wang R, Farrona S, Vincet C, et al. 2009. PEP1 regulates perennial flowering in Arabis alpina. Nature 459: 423– 428.
Webb T, Bartlein PJ. 1992. Global changes during the last 3 million years:
climatic controls and biotic responses. Annual Review of Ecology and
Systematics 23: 141–173.