Content uploaded by Adam Boratyński
Author content
All content in this area was uploaded by Adam Boratyński on Apr 10, 2014
Content may be subject to copyright.
ORIGINAL ARTICLE
High level of genetic differentiation of Juniperus phoenicea
(Cupressaceae) in the Mediterranean region: geographic
implications
Adam Boratyn
´ski ÆAndrzej Lewandowski Æ
Krystyna Boratyn
´ska ÆJose M. Montserrat Æ
Angel Romo
Received: 11 January 2008 / Accepted: 31 October 2008 / Published online: 6 January 2009
ÓSpringer-Verlag 2009
Abstract Fourteen natural populations of Juniperus
phoenicea L. from the quite entire species range have been
compared using isoenzyme polymorphism. Among 17 loci,
5(Got1,6Pgd3,Pgi2,Pgm2 and Shdh2) appeared to be
differentiated sufficiently to provide useful information for
discrimination between the subspecies phoenicea and
turbinata (Guss.) Nyman. Two distinct groups of popula-
tions were detected using the Nei’s genetic distance
unweighted pair group method with arithmetic mean
(UPGMA) and discrimination analyses, one including the
inland populations of the eastern Iberian Peninsula and
southern France (subsp. phoenicea), and the second from
the Mediterranean and Atlantic shores, and from the Atlas
mountains in Africa (subsp. turbinata). The high level of
differences confirms a long period of isolation, probably
during the whole Pleistocene. The population from the
Aegean Sea shore differed from the other Mediterranean
shore plus Atlas mountain population. It also suggests
spatial isolation between them, at least during the last
Glaciation.
Keywords Biogeography Plant variation
Cupressaceae Genetic diversity Glacial survival
Isoenzyme loci Mediterranean plant Pleistocene refugia
Introduction
Juniperus phoenicea L. is a small, 8–12 m tall monoecious,
or rarely dioecious, tree. Its range covers the whole
Mediterranean region, from the Canary Islands, the Atlas
mountains in Africa and the Atlantic coast of Portugal in
the West, to Jordan and Saudi Arabia in the East (Fig. 1).
The center of the species’ occurrence covers the western
part of the Mediterranean, mostly the Iberian Peninsula and
North West Africa (Jalas and Suominen 1973; Que
´zel and
Pesson 1980; Que
´zel and Barbero 1981; Browicz and
Zielin
´ski 1982; Kerfoot and Lavranos 1984; Greuter et al.
1984; do Amaral Franco 1986; Christensen 1997; Charco
2001; Farjon 2005). These regions are areas of refugia of
the tertiary flora (Reinig, after Kornas
´and Medwecka-
Kornas
´2002, p. 433; Carrio
´n2002; Benito Garzo
´n et al.
2007) in the Pleistocene, mostly the Iberomoroccan but
including the Macaronesian, Korsardynian and Appeninan
ones (Reinig, after Kostrowicki 1999, p. 70; Comes 2004).
J. phoenicea is a variable species and two subspecies,
the type subsp. phoenicea and the maritime subsp. turbi-
nata (Guss.) Nyman were distinguished on the basis of
taxonomic studies, which have been summarized by Farjon
(2005). The existence of two subspecies of J. phoenicea
was confirmed by biometrical studies (Lebreton 1983;
Mazur et al. 2003). However, the morphological differ-
ences between the subspecies phoenicea and turbinata used
to determine herbarium samples revealed a much more
frequent presence and a larger range of the type subspecies
(Farjon 2005, p. 340) than those described by Lebreton and
A. Boratyn
´ski (&)A. Lewandowski K. Boratyn
´ska
Polish Academy of Sciences, Institute of Dendrology,
5 Parkowa Street, 62-035 Ko
´rnik, Poland
e-mail: borata@man.poznan.pl
J. M. Montserrat
Institute of Culture of Barcelona, Botanic Garden,
C/Font i Quer 2, 08038 Barcelona, Spain
A. Romo
Consejo Superior de Investigaciones Cientı
´ficas,
Institute of Botany, Passeig del Migdia s/n,
08038 Barcelona, Spain
123
Plant Syst Evol (2009) 277:163–172
DOI 10.1007/s00606-008-0122-z
Rivera (1989). Nevertheless, the biochemical diversity was
correlated with the diameter of the cones (Lebreton and
Rivera 1989), and confirmed the subdivisions of subsp.
phoenicea (including eu-mediterranea P. Lebreton and
S. Thivend, see Farjon 2005, p. 337) and subsp. turbinata;
significant differences between them were also found at the
level of random amplified polymorphic DNA (RAPD)
(Adams et al. 2002).
Distinguishing the two subspecies of J. phoenicea
within the area of the Iberomoroccan refugial region (do
Amaral Franco 1986) and the results of investigation of the
species variation (Lebreton and Thivend 1981; Lebreton
1983; Lebreton and Rivera 1989; Adams et al. 1996;
Lewandowski et al. 2000; Cavaleiro et al. 2001; Adams
et al. 2003; Mazur et al. 2003) seem to indicate the exis-
tence of at least two centers, isolated against gene flow for
a sufficiently long period of time, where J. phoenicea was
able to survive during the Pleistocene climate cooling on
the Iberian Peninsula. This ancient isolation was the cause
of early divergence of the ancestor and for the formation of
morphological, genetic and biochemical differences that
permit distinction of the two contemporary subspecies, and
this was confirmed by the isozyme variation (Lewandowski
et al. 2000). The genetic variation of J. phoenicea with
RAPD, tested on several individuals in the West and
Central Mediterranean (Adams et al. 2002), and inter-
simple sequence repeat (ISSR) markers between four
populations from the West and one from the East of the
Mediterranean region (Meloni et al. 2006) have revealed
differences between samples from the West and the East.
Most studies that have investigated interspecies bio-
diversity at the gene level have focused on how genetic
diversity is structured within and among populations
(Meister et al. 2006; Mehes et al. 2007; Myers et al. 2007).
Isozymes, among other markers, are generally considered
to be neutral and therefore suitable indicators of historical
processes through drift and migration. Comparatively little
work has been done on Mediterranean conifer forest tree
species; however, the available data indicated that the level
of genetic differentiation within them is about 45%
higher than those of other conifer species worldwide
(Fady-Welterlen 2005). The high level of differentiation
found in widely distributed Mediterranean conifers could
be explained by habitat fragmentation that occurred at
some point during their evolutionary history. In the
Mediterranean, despite low evolutionary rates in trees, they
have accumulated much intraspecific biodiversity thanks to
the great antiquity of their populations, dating back from
the Tertiary (Petit et al. 2005). At the same time, J. phoe-
nicea has not been planted on a large scale, its seeds have
not been transported for long distances, and the pattern of
geographic differentiation of the species has not been as
disturbed as those of more important forest trees, making it
a good subject for studying the patterns of geographical
variation in the Mediterranean region.
The present study is based on the hypothesis that
J. phoenicea has an infraspecific differentiation at the
isoenzyme level in the West Mediterranean region that
corresponds to the species differentiation described on the
basis of biochemical characters, and the West and East
Mediterranean populations are different as a result of
repeated long-lasting spatial isolation during the Pleisto-
cene. The aim of the study was to verify the genetic
variation of the Phoenician Juniper and differentiation
between its well-recognized subspecies using isozymes as
genetic markers.
Materials and methods
Plant material
Cones were sampled from 14 natural populations of
J. phoenicea in north-western Africa, southern Europe and
Fig. 1 Area of distribution of
Juniperus phoenicea (compiled
from Jalas and Suominen 1973;
Que
´zel and Pesson 1980;
Browicz and Zielin
´ski 1982;
Boratynski et al. 1992, Charco
2001, Farjon 2005) and
localization of compared
populations (for acronyms
follow Table 1)
164 A. Boratyn
´ski et al.
123
south-western Asia (Table 1), covering an extensive area
of the species distribution (Fig. 1). Each population, except
of those from Kassandra in Greece, was represented by
24–30 individuals, and 391 individuals were tested in total.
Nine of the sample zones represent subsp. turbinata and
five represent subsp. phoenicea. The latter were collected
from the area determined by Lebreton and Rivera (1989,
p. 39, Figs. 3, 4) as regions corresponding to the distribu-
tion of the type subspecies, delimited mostly on the
prodelphinidine and procyanidine content. The individuals
of all samples were collected randomly, without distinction
at the subspecies level.
Genetic analyses
Before electrophoresis, seeds were kept on moistened filter
paper in Petri dishes at room temperature for 3 days. Seven
macrogametophytes from each of the studied trees were
used to define their genotypes. The following enzymes
were investigated in this study; the abbreviation, Enzyme
Commission codes, and the number of loci analyzed are
given in parentheses. Alcohol dehydrogenase (ADH; EC
1.1.1.1; 1); fluorescent esterase (FLE; EC 3.1.1.1; 2); gluta-
mate dehydrogenase (GDH; EC 1.4.1.2; 1); glutamate
oxaloacetate transaminase (GOT; EC 2.6.1.1; 2); isocitrate
dehydrogenase (IDH; EC 1.1.1.42; 1); malate dehydroge-
nase (MDH; EC 1.1.1.37; 2); menadione reductase (MNR;
EC 1.6.99.2; 1), 6-phosphogluconate dehydrogenase
(6PGD; EC 1.1.1.44; 1); phosphoglucose isomerase (PGI;
EC 5.3.1.9; 1); phosphoglucomutase (PGM; EC 2.7.5.1; 2);
shikimate dehydrogenase (SHDH; EC 1.1.1.25; 2); and
superoxide dismutase (SOD; EC 1.15.1; 1). Details con-
cerning the electrophoresis, staining procedure and
inheritance of individual isozymes have been described by
Lewandowski et al. (2000). Alleles at each locus were
numbered according to the electrophoretic mobility of
allozymes. The most anodally migrating band was num-
bered 1, the next 2, and so on. Alleles occurring in a
population with a frequency of less than 5% are referred to
as rare, and those occurring in only one population are
referred to as unique.
Statistical treatment
Allozyme frequencies, expected (Nei 1978) and observed
heterozygosities, gene diversity statistics (Nei 1978) and
unbiased genetic distance measures (Nei 1972) were cal-
culated using PopGen software (Yeh et al. 1999). The
populations were clustered based on Nei’s genetic distance
coefficient, using the unweighted pair group method
(UPGMA) (Sneath and Sokal 1973). The percentage data
were arcsine transformed before statistical analysis (Watala
2002). The statistical difference between populations was
tested using analysis of discrimination (Tabachnik and
Fidell 1996; Sokal and Rohlf 2003). STATISTICA 7.0 for
Windows software (StatSoft) was used in the statistical
analyses.
Table 1 Sampled populations of Juniperus phoenicea (boldface, samples of J. phoenicea subsp. phoenicea)
Acronym Locality Subspecies Number of individuals Longitude Latitude Altitude (m)
MOR_1 Morocco, coast of Atlantic about 10–15 km S of Tanger turbinata 28 5°590W35°320N20
MOR_2 Morocco, High Atlas, Tizi-n-Tagalm SE of Miledt turbinata 30 4°340W32°370N 1800
MOR_3 Morocco, High Atlas, between Agouti and Azilal turbinata 30 6°290W31°380N 1900
PORT_1 Portugal, Algarve, Cabo de Pontal turbinata 30 8°550W37°100N25
SP_1 Spain, Huelva, Playa de Matalascan
˜as turbinata 29 6°340W37°000N20
SP_2 Spain, Teruel, Sierra de Nogueruela E of Rubielos
de Mora
phoenicea 30 0°400W40°240N 1100
SP_3 Spain, Teruel, Sierra de Valdancha, near Portella
de Morella
phoenicea 29 0°130W40°340N 1100
SP_4 Spain, Zaragoza, Montes de la Retuerta de Pina W
of Bujaraloz
phoenicea 30 0°190W41°290N 400
AND_1 Andorra, Coll de Jau near San Julia de Loria phoenicea 30 1°280E42°270N 1200
FRA_1 France, Narbonne, near St. Pierre s Mere phoenicea 28 3°100E43°100N50
IT_1 Italy, Sabaudia turbinata 28 13°020E41°150N10
GR_1 Greece, Peloponnese, Ionian Coast N of Kiparissia turbinata 29 21°320E37°280N15
GR_2 Greece, Thessaloniki, Kassandra, Akros Paliourion turbinata 16 23°400E39°550N20
TU_1 Turkey, Mugla, Marmaris Peninsula, about 30 km W
of Marmaris
turbinata 24 27°500E36°490N 700
High level of genetic differentiation of Juniperus phoenicea 165
123
Results
Genetic diversity
Seventeen loci were studied in each of the populations
investigated. All of them were polymorphic in at least one
population, showing the presence of three (in loci Gdh,
Got2,Idh and Pgm1) and up to seven (in loci Pgi2 and
Shdh2) alleles per locus. In total, 76 alleles were found; 9
were rare and 11 were unique. A high level of genetic
variation was found for these populations. On average,
72% of loci were polymorphic and the expected hetero-
zygosity was 0.187. The mean and effective number of
alleles per locus were 2.11 and 1.32, respectively. Small
differences between populations were detected. However,
levels of genetic diversity between the two subspecies of
J. phoenicea were not statistically significant (Table 2).
Values of the fixation index were negative, exhibiting a
slight excess of heterozygotes in the nine populations and
positive in the other five populations.
Differences between the subspecies phoenicea
and turbinata
Our results revealed substantial differences between two
subspecies of J. phoenicea in allele frequencies at a
number of allozyme loci. Among the 17 loci analyzed, 5
(Got1,6Pgd3,Pgi2,Pgm2 and Shdh2) appeared to be
differentiated sufficiently to provide useful information for
discrimination of one subspecies from another. At these
loci, certain alleles occurred in substantial frequency in
populations of one subspecies but were absent or nearly so
from populations of the other (Fig. 2). At Got1, the third
allele is the major allele in all the populations of J. phoe-
nicea subsp. turbinata (with a frequency of 0.933–1.000),
while it occurred with a lower frequency (0.071–0.383) in
populations of J. phoenicea subsp. phoenicea. Similarly, in
the locus 6Pgd3, the frequencies of the fourth allele of
J. phoenicea subsp. turbinata varied between 0.531 and
1.0, and between 0 and 0.017 in J. phoenicea subsp.
phoenicea. The larger differences were found in the fre-
quencies of the sixth allele at the Pgi2 locus, which has not
been reported for populations of J. phoenicea subsp.
turbinata, while it occurred in frequencies of 0.983–1.0
within subsp. phoenicea. The frequencies of the second
allele of locus Pgm2 were 0.896–1.0 in J. phoenicea subsp.
turbinata, and 0–0.050 in subsp. phoenicea. Inversely, the
frequencies of the first allele of locus Shdh2 were low and
varied from 0 to 0.138 in J. phoenicea subsp. turbinata,
and from 0.948 to 1.0 in subsp. phoenicea.
The level of genetic differentiation was high
(F
ST
=0.43), indicating that about 57% of the genetic
Table 2 Genetic variability in the investigated populations (standard errors in parentheses) of J. phoenicea subsp. phoenicea and subsp.
turbinata
Sample P Na Ne He F
J. phoenicea subsp. turbinata
MOR_3 76 2.00 (0.71) 1.34 (0.35) 0.210 (0.186) 0.000
MOR_2 100 2.82 (0.73) 1.41 (0.47) 0.239 (0.187) 0.013
MOR_1 71 2.06 (0.90) 1.38 (0.52) 0.209 (0.216) -0.079
SP_1 65 2.06 (1.03) 1.31 (0.40) 0.183 (0.203) -0.072
PORT_1 59 1.94 (0.97) 1.26 (0.42) 0.148 (0.200) 0.007
IT_1 82 2.23 (1.03) 1.41 (0.58) 0.211 (0.227) -0.041
GR_1 41 1.64 (0.93) 1.21 (0.32) 0.128 (0.185) -0.036
GR_2 71 2.18 (0.95) 1.44 (0.46) 0.248 (0.229) -0.090
TU_1 76 1.88 (0.70) 1.30 (0.42) 0.179 (0.203) -0.110
Mean 71.2 (5.2) 2.09 (0.16) 1.34 (0.09) 0.195 (0.060)
J. phoenicea subsp. phoenicea
SP_2 76 2.23 (0.97) 1.31 (0.40) 0.183 (0.205) 0.042
SP_3 71 2.12 (0.99) 1.26 (0.32) 0.171 (0.177) 0.084
SP_4 59 1.88 (0.93) 1.26 (0.34) 0.158 (0.189) -0.017
AND_1 71 2.12 (0.93) 1.30 (0.46) 0.171 (0.197) -0.076
FRA_1 88 2.35 (0.86) 1.29 (0.39) 0.176 (0.187) -0.046
Mean 73.0 (4.5) 2.14 (0.17) 1.28 (0.07) 0.172 (0.040)
Mean for all populations 71.9 (3.6) 2.11 (0.12) 1.32 (0.07) 0.187 (0.050)
166 A. Boratyn
´ski et al.
123
variation is within populations and 43% is between popu-
lations. This value, however, is much higher for the five
loci mentioned above, differentiating between J. phoenicea
subsp. phoenicea and subsp. turbinata, and ranges from
0.51 for Shdh2 to 0.80 for 6Pgd3. Genetic differentiation
within J. phoenicea subsp. phoenicea and subsp. turbinata
is much lower, with values of F
ST
=0.059 and 0.202,
respectively.
The genetic differences between the subspecies are
made clear by the calculated values of the genetic distance.
The values of genetic distance among populations ranged
between 0.027 and 0.248 (average 0.072) for J. phoenicea
subsp. turbinata and between 0.004 and 0.036 (average
0.016) for subsp. phoenicea, and the average value of
the genetic distance between subspecies was 0.369. The
dendrogram constructed on the basis of the genetic dis-
tances illustrates differences among the populations that
form the two most distant groups (Fig. 3). The group of
type subspecies, J. phoenicea subsp. phoenicea, is slightly
differentiated, while the differences among samples of
subsp. turbinata are at a higher level.
The analysis of the discriminant function on the allele
frequencies, in particular individuals and populations,
indicated that among 17 loci, 14 varied significantly
Fig. 2 Geographic pattern of interpopulation variation of J. phoenicea revealed in frequency of five most differentiating alleles, the third of
Got1, fourth of 6Pgd3, sixth of 6Pgi2, second of Pgm2 and first of Shdh2 (for sample acronyms follow Table 1)
Fig. 3 Dendrogram constructed on the genetic distances among
compared samples of J. phoenicea (for sample acronyms follow
Table 1)
High level of genetic differentiation of Juniperus phoenicea 167
123
between the populations at level P=0.01, and 1 at level
P=0.05 (Table 3). The alleles of 6Pgd3,Pgi2,Pgm1,
Sod1 and Idh1 were among the most frequent contributors
with respect to the overall diversity.
The distribution of samples in the space of the two first
canonical variables, U1 and U2, which cover about 95% of
the total variation, confirms the high degree of separation
between samples of J. phoenicea subsp. phoenicea and
subsp. turbinata. The individuals of subspecies form two
very well-separated groups (Fig. 4); no individual sampled
in populations of J. phoenicea subsp. phoenicea (samples
SP_2, SP_3, SP_4, AND_1 and FRA_1, acronyms as in
Table 1) fell into the group of individuals representing
subsp. turbinata, and vice versa (Fig. 4). The discriminant
variable U1, which accounts for more than 89% of the total
variation, is determined mostly by the frequency of alleles
Pgi2,6Pgd3 and Pgm2, then by Got1 and Shdh2, while U2,
which includes only 5.5% of variation, is determined
mostly by the frequency of allele 6Pgd3.
Differentiation of subspecies turbinata
Discrimination analysis revealed differentiation among
samples of J. phoenicea subsp. turbinata. The third group
includes 23 from 24 individuals of population TU_1, 10
from 16 individuals of population GR_2 and 2 from 29
individuals of population GR_1, and is separated from
J. phoenicea subsp. turbinata by variable U2 (Fig. 4). The
other individuals of the three above-mentioned samples are
placed among typical J. phoenicea subsp. turbinata.
Similar differentiation of population GR_1, GR_2 and
TU_1 has been found in the analysis of the genetic dis-
tances among individuals (data not shown).
Discussion
Genetic diversity
Intrapopulation variation
The level of intrapopulation isozyme variation in J. phoe-
nicea was similar to those reported earlier for other
conifers in Europe (Mu
¨ller-Starck et al. 1992). The wide
continuous ranges, outcrossing, high population density,
large effective size of populations, wind pollination and
seed dispersal are recognized as factors that allow for the
maintenance of a large amount of genetic variability in
conifers in boreal areas (Loveless and Hamrick 1984).
However, these factors do not apply to all Mediterranean
species, including J. phoenicea; its range is sufficiently
large, but it is fragmented and discontinuous (Que
´zel and
Pesson 1980; Browicz and Zielin
´ski 1982; Boratynski et al.
1992; Que
´zel et al. 1992; Charco 2001). Populations of
J. phoenicea can have a pioneer character, especially on
the dunes along the coasts and on exposed cliffs, but
individuals in rocky sites can be more than 1,000 years old
(Mandin 2005). J. phoenicea is a wind-pollinated tree that
can have both male and female strobili, but frequently with
a predominance of micro- or macrostrobili, or even func-
tionally dioecious (Jordano 1991). Finally, unlike the
majority of coniferous trees, the species is ornitochorous
(Jordano 1993).
Table 3 Discriminant power testing for frequencies of isoenzyme
loci in the J. phoenicea populations
Isoenzyme loci Partial Wilks’ lambda FP
ADH 0.9125 2.6637 0.00140
FLE-1 0.9523 1.3897 0.16154
FLE-2 0.8731 4.0347 0.00000
GDH 0.8462 5.0490 0.00000
GOT-1 0.7771 7.9655 0.00000
GOT-2 0.9732 0.7639 0.69844
IDH 0.6914 12.3960 0.00000
MDH-1 0.9364 1.8864 0.03036
MDH-2 0.9214 2.3705 0.00468
MEN-1 0.8020 6.8546 0.00000
6PGD-3 0.2173 100.0466 0.00000
PGI-2 0.5310 24.5250 0.00000
PGM-1 0.6727 13.5124 0.00000
PGM-2 0.7274 10.4077 0.00000
SHDH-1 0.7740 8.1080 0.00000
SHDH-2 0.7516 9.1767 0.00000
SOD-1 0.6853 12.7524 0.00000
Fig. 4 Result of discriminant analysis based on the allele frequencies
in 14 populations of Juniperus phoenicea plotted along the two first
discriminant variables, U1 and U2, which accounted for 95% of the
total variation (for sample acronyms follow Table 1)
168 A. Boratyn
´ski et al.
123
The high level of intrapopulation genetic variability in
J. phoenicea is thought to be associated with repeated long
periods of survival of separate populations within the
territories around the refugial areas under diverse environ-
mental conditions, probably in the maritime mountain
regions during the Pleistocene. The species had to share its
range, benefiting from each warmer period during glacia-
tions (e.g. see Elenga et al. 2000; van Andel 2002), which
made possible the connection of the local populations and
the consequent exchange of genetic information, and a
wider dispersion of the seeds by birds (Jordano 1993;
Garcia 2001; Bonet and Pausas 2004). The Late Glaciation
periods stand out particularly, where J. phoenicea domi-
nated or co-dominated the local plant landscapes of the
southern Iberian Peninsula (Uzquiano and Arnaz 1997).
And finally, the reasonably high level of intrapopulation
genetic diversity of many Mediterranean conifers may have
been retained during the Last Glaciation, or even through
the whole Pleistocene, because the effective size of isolated
populations probably never dropped below the critical
threshold (Fady-Welterlen 2005), as occurred in the relict
populations of trees in the subtropical mountain zones
(Ge et al. 1998; Aguirre-Planter et al. 2000).
Interpopulation variation
The high level of genetic differentiation between popula-
tions is consistent with earlier reports based on isozymes
(Lewandowski et al. 2000), RAPD (Adams et al. 2002) and
morphological characteristics (Mazur et al. 2003). The
level of interpopulation genetic differentiation among
populations of J. phoenicea appeared to be much higher
than that observed among populations of coniferous species
in the boreal zone. For example, the overall degree of
genetic differentiation for Picea abies throughout Europe
has been estimated as 0.052 (Lagercrantz and Ryman
1990), while it is 0.43 for J. phoenicea. However, the level
of genetic differentiation of J. phoenicea is similar to that
of other Mediterranean gymnosperms; for example Pinus
brutia (Kara et al. 1997) and Cupressus sempervirens
(Korol et al. 1997). It is supposed that the high level of
differentiation found in widely distributed Mediterranean
conifers could be explained by habitat fragmentation dur-
ing their Pleistocene history. They probably achieved a
higher level of differentiation by recolonization of the
current territories from more glacial refugia, which were
isolated during longer periods of time than those from
temperate areas (Fady-Welterlen 2005).
Differences between subspecies
The two groups of samples recognized in the isozyme
differentiation study partly confirmed division of the
species into groups of individuals containing various
amounts of prodelphinidine and procyanidine described by
Lebreton and Thivend (1981) and Lebreton and Rivera
(1989). Populations in our study, sampled as J. phoenicea
subsp. phoenicea (samples SP_2, SP_3, SP_4, AND_1 and
FRA_1), all appeared significantly different from the
others, and their distribution falls within the area of the
subspecies determined by Lebreton and Rivera (1989,
p. 39, Figs. 3, 4). Our results also indicate large genetic
differences between the two subspecies of Juniperus
phoenicea (Figs. 2,3,4), which suggest species rather than
subspecies level of taxa identified as J. phoenicea subsp.
phoenicea and subsp. turbinata. In spite of that, there is no
great morphological difference between these taxa (do
Amaral Franco 1986; Mazur et al. 2003; Farjon 2005).
Nevertheless, the five loci (Got1,6Pgd3,Pgi2,Pgm2 and
Shdh2) can be used as markers to distinguish the subspe-
cies at the population and individual levels.
Significant differences between subspecies of J. phoe-
nicea have been described earlier (Lebreton and Rivera
1989; Adams et al. 1996) in the contents of prodelphinidine
and procyanidine and have been used as the basis to dis-
tinguish J. phoenicea subsp. eu-mediterranea (Lebreton
1983; Lebreton and Thivend 1981, Lebreton and Rivera
1989), regarded ultimately as a synonym of J. phoenicea
subsp. phoenicea (Farjon 2005). The biochemical differ-
ences have been also confirmed in the larger cone diameter
in subsp. eu-mediterranea and a greater number of seeds in
the cone in subsp. phoenicea (Lebreton and Rivera 1989),
the latter coinciding with the results of biometrical studies
on the morphological characters of cones and ultimate
shootlets with leaves (Mazur et al. 2003).
The genetic variation of J. phoenicea with RAPD
markers was tested on several individuals in the Medi-
terranean, confirming differences between subsp. phoenicea
and turbinata (Adams et al. 2002). The latter subspecies,
however, revealed differentiation, interpreted partly as
confirmation of taxonomic division of three varieties and
two groups without taxonomic status (Adams et al. 2002,
p. 225, Fig. 1). The study by Adams et al. (l.c.) used indi-
viduals; nevertheless, the large differences between three
individuals collected as J. phoenicea subsp. phoenicea and
19 others should be interpreted as a confirmation of the
taxonomic position of the type subspecies of J. phoenicea,
as shown here (Fig. 4).
The taxonomic differences, and especially the pheno-
logical isolation, are reasons for the lack of cross-
pollination between individuals of J. phoenicea subsp.
phoenicea and subsp. turbinata (Arista et al. 1997). The
reproductive isolation between them has contributed to the
present level of differences and will lead to their increase
in the future (Ferris et al. 1999; Rieseberg and Wendel
2004; Rieseberg et al. 2004; Hewitt 2004; Petit et al. 2005).
High level of genetic differentiation of Juniperus phoenicea 169
123
Variation of subsp. phoenicea
The area of distribution of the two subspecies of J. phoe-
nicea was different from that proposed by Farjon (2005).
The type subsp. phoenicea occurs in the eastern part of the
Iberian Peninsula and in the southern part of France
(Fig. 2). These regions have been recognized by Lebreton
and Rivera (1989, p. 39, Figs. 3, 4) on the basis of pro-
delphinidine and procyanidine contents, as a range of
subsp. phoenicea. Our finding confirms the low level of
geographic differentiation of J. phoenicea subsp. phoeni-
cea (compare our Fig. 2with Figs. 3, 4 in Lebreton and
Rivera 1989, p. 39, and Figs. 1, 2 in Adams et al. 2002,
p. 225 and 227). This suggests the local character is con-
served during the Pleistocene in areas isolated from other
subspecies.
Variation of subsp. turbinata
The geographic range of J. phoenicea subsp. turbinata
covers a much larger area than that of subsp. phoenicea.
The distribution of samples of subsp. turbinata used in our
study covers most of the species range. Generally, the
genetic differences between populations were low, except
for several individuals from the three populations in the
eastern part of the Mediterranean region (Fig. 4). It can be
(1) a trace of isolation of those eastern populations dating
back to pre-Holocene period, or (2) the result of faster rate
of divergence than in the western populations.
The high RAPD differentiation of individuals not
included into subsp. phoenicea in Adams et al. (2002,
p. 225, Fig. 1) study is indicative of the high level of
variation of subsp. turbinata. This was confirmed, in part,
by the differentiation of J. phoenicea subsp. turbinata at
the isoenzyme level found in this study (Fig. 4). The dif-
ferences between western and eastern populations of subsp.
turbinata were also mentioned by Meloni et al. (2006).
ISSR markers were used to describe the species’ variation
among five populations, and differences have been found
between four western and one eastern populations (Meloni
et al. 2006), as in our finding. It should be stressed that
samples in the latter study came from the area of distri-
bution of J. phoenicea subsp. turbinata (after Lebreton and
Rivera 1989, not after Farjon 2005).
The frequency of genetically different individuals within
samples GR_1, GR_2 and TU_1 increases from West to
East, indicating the existence of another center of species
variation further to the East. We have not collected suffi-
cient material from populations far to the East to be able to
identify them properly, as it was mentioned above, but the
nature of the variation confirms the possibility of the
existence of another Pleistocene refugium of the species in
the eastern part of the Mediterranean region.
Possible palaeobotanic determination of present
variation
Unfortunately, the pollen of Juniperus has not been
determined at the species level (e.g. Huntley 1988; Elenga
et al. 2000; Carrio
´n2002; Eastwood 2004; Tzedakis 2004;
Gonza
´lez-Sampe
´riz et al. 2005) and much of the data
concerning the occurrence of Junipers during the Late
Glaciation and early Holocene cannot be used directly. The
survival of J. phoenicea during Late Glacial Maximum in
the western Mediterranean region was confirmed by mac-
roscopic palaeo-findings in South East Spain (Uzquiano
and Arnaz 1997).
The reconstruction of palaeo-climates (van Andel 2002;
Carrio
´n2002; Eastwood 2004; Gonza
´lez-Sampe
´riz et al.
2005) also supported the possibility of the persistence of
Mediterranean species of the genus Juniperus, at least
along the coastal regions of the southern part of present-
day Mediterranean Europe. Suitable thermal conditions
existed on the Iberian Peninsula, and in the southernmost
parts of the Italian and Peloponnesian Peninsulas, but our
results did not confirm such a differentiation of J. phoe-
nicea (see Fig. 2). The species, in spite of its contemporary
broad range (Fig. 1), probably had two main centers, from
which it has migrated across Mediterranean Europe and
North Africa, reaching in the East as far as West Anatolia,
Sinai and Arabian Peninsula.
The recognition of high differences between the two
groups of populations suggests a long period of isolation,
which indicates their origin from another Pleistocene
refugium or, better, refugial areas. Ecologically, J. phoe-
nicea is a pioneer, light demanding and relatively resistant
to a dry climate (Zohary 1973; Que
´zel and Pesson 1980;
Que
´zel and Barbero 1981; Auclair 1996; Charco 2001;
Tzedakis 2004; Petit et al. 2005). For this reason it prob-
ably did not occur inside the refugium, but on peripheral
sites. The positive reaction of Juniperus to the arid periods
during the late Pleistocene/Holocene and its suppression by
more humidity demanding, predominantly broad-leaved
trees have been reported for the Iberian Penisula (Uzquiano
and Arnaz 1997; Carrio
´n et al. 2001a,b,2003,2004). Arid
and sufficiently warm environments have been recognized
as present along the Mediterranean coast and shelf of the
Iberian Peninsula, even during the Late Glacial Maximum
(van Andel 2002; Carrio
´n2002), and probably also during
the entire Pleistocene (Fady-Welterlen 2005).
Taking into account the high level of differences
between samples representing J. phoenicea subsp. phoe-
nicea and all the other, and the rather low level of
differences among them (Figs. 3,4), it should be stated that
the isolation between refugial areas of J. phoenicea subsp.
phoenicea and of subsp. turbinata was over a long distance
and covered a long time-span. The centers of formation of
170 A. Boratyn
´ski et al.
123
both subspecies proposed by Lebreton and Rivera (1989)
seem very probable, with isolation from the end of
Messinian period (about 5.3 million years ago), as it was
suggested lately for Juniperus thurifera L. (Terrab et al.
2008) and Abies spp. (Terrab et al. 2007).
The differences between western and eastern popula-
tions representing J. phoenicea subsp. turbinata also
suggest an isolation longer than the Holocene, but signifi-
cantly shorter than between J. phoenicea subsp. phoenicea
and subsp. turbinata. Our data confirm suggestion con-
cerning the possibility of survival through the Last
Glaciation by the taxon in the East Mediterranean Pleis-
tocene refugium. Unfortunately, neither the data reported
by Adams et al. (2002) and Meloni et al. (2006), nor our
ownes, revealed the total geographic variation of J. phoe-
nicea, which will hopefully be confirmed in further studies
of more complete material from the eastern part of its
range.
Acknowledgments The authors are grateful to Samuel Pyke for
correcting the English of a previous version of the manuscript. The
study was sponsored, in part, by the Polish Committee for Scientific
Research, grant no. 2P04C 030 26. The majority of the plant material
was collected thanks to cooperation between CSIC (Spanish Research
Council) and PAS (Polish Academy of Sciences).
References
Adams RP, Barrero AF, Lara A (1996) Comparisons of the leaf
essential oils of Juniperus phoenicea,J. phoenicea subsp.
eu-mediterranea Lebr. et Thiv. and J. phoenicea var. turbinata
(Guss.) Parl. J Essential Oil Res 8:367–371
Adams RP, Pandey N, Rezzi S, Casanova J (2002) Geographic
variation in the random amplified polymorphic DNAs (RAPDs)
of Juniperus phoenicea,J. p. var. canariensis,J. p. subsp.
eu-mediterranea and J. p. var. turbinata. Biochem Syst Ecol
30:223–229
Adams RP, Schwarzbach A, Pandey RN (2003) The concordance of
terpenoid, ISSR and RAPD markers and ITS sequence data sets
among genotypes: an example from Juniperus. Biochem Syst
Ecol 31:375–387
Aguirre-Planter E, Furnier GR, Eguiarte LE (2000) Low levels of
genetic variation within and high levels of genetic variation
among populations of species Abies from Mexico and Guate-
mela. Am J Bot 87:362–371
Arista M, Ortiz PL, Talavera S (1997) Reproductive isolation of two
sympatric subspecies of Juniperus phoenicea (Cupressaceae) in
southern Spain. Plant Syst Evol 208:225–237
Auclair L (1996) L’appropration communautaire des forets dans le
Haut Atlas marocain. Cah Sci Humain 32:177–194
Benito Garzo
´nM,Sa
´nchez de Dios R, Sa
´inz Ollero H (2007)
Predictive modelling of tree species distributions on the Iberian
Peninsula during the Last Glacial Maximum and Mid-Holocene.
Ecography 30:120–134
Bonet A, Pausas JG (2004) Species richness and cover along a
60-year chronosequence in old-fields of southeastern Spain.
Plant Ecol 174:257–270
Boratyn
´ski A, Browicz K, Zielin
´ski J (1992) Chorology of trees and
shrubs in Greece. Sorus, Poznan
´/Ko
´rnik
Browicz K, Zielin
´ski J (1982) Chorology of trees and shrubs in south-
west Asia and adjacent regions. Polish Scientific Publishers,
Warszawa–Poznan
´
Carrio
´n JS (2002) Patterns and processes of Late Quaternary
environmental change in a montane region of southwestern
Europe. Q Sci Rev 21:2047–2066
Carrio
´n JS, Andrade A, Bennett KD, Munuera M (2001a) Crossing
forests thresholds. Inertia and collapse in Holocene sequence
from south-central Spain. Holocene 11:635–653
Carrio
´n JS, Munuera M, Dupre
´M, Andrade A (2001b) Abrupt
vegetation changes in the Segura mountains of southern Spain.
J Ecol 89:783–797
Carrio
´n JS, Yll EI, Walker MJ, Legaz AJ, Chain C, Lo
´pez A (2003)
Glacial refugia of temperate, Mediterranean and Ibero-North
African flora in south-eastern Spain: new evidence from cave
pollen at two Neanderthal sites. Global Ecol Biogeogr
12(2):119–129
Carrio
´n JS, Yll EI, Willis KJ, Sanchez P (2004) Holocene forest
history of the eastern plateaux in the Segura Mountains
(Murcia, southeastern Spain). Rev Paleobot Palynol 132(3–
4):219–236
Cavaleiro C, Rezzi S, Salgueiro L, Bighelli A, Casanova J, da Cunha
AP (2001) Infraspecific chemical variability of the leaf essential
oil in Juniperus phoenicea var. turbinata from Portugal.
Biochem Syst Ecol 29(11):1175–1183
Charco J (2001) Guı
´a de los a
´rboles y arbustos del norte de A
´frica.
Agencia Espan
˜ola de Cooperacio
´n Internacional, Madrid
Christensen KI (1997) Juniperus L. In: Strid A, Tan K (eds) Flora
hellenica, vol 1. Koeltz Scientific Books, Ko
¨nigstein, pp 10–14
Comes HP (2004) The Mediterranean region—a hotspot for plant
biogeographic research. New Phytol 164:11–14
do Amaral Franco J (1986) Juniperus L. In: Castroviejo S, Laı
´nz M,
Lo
´pez Gonza
´lez G, Montserrat P, Mun
˜oz Garmendia F, Paiva J,
Villar L (eds) Flora iberica, vol 1. Real Jardı
´n Bota
´nico, C.S.I.C.,
Madrid, pp 181–188
Eastwood WJ (2004) East Mediterranean vegetation and climate
change. In: Griffiths HI, Krys
ˇtufek B, Reed JM (eds) Balkan
biodiversity. Kluwer, Dortrecht, pp 25–48
Elenga H, Peyron O, Bonnefille R, Jolly D, Cheddadi R, Guiot J,
Andrieu V, Bottema S, Buchet G, de Beaulieu J-L, Hamilton
AC, Maley J, Marchant R, Perez-Obiol R, Reille M, Riollet G,
Scott L, Straka H, Taylor D, Van Campo E, Vincens A, Laarif
F, Jonson H (2000) Pollen-based biome reconstruction for
southern Europe and Africa 18,000 yrBP. J Biogeogr 27:621–
634
Fady-Welterlen B (2005) Is there really more biodiversity in
Mediterranean forest ecosystems? Taxon 54:905–910
Farjon A (2005) A monograph of Cupressaceae and Sciadopitys.
Royal Botanic Gardens, Kew
Ferris C, King RA, Hewitt GM (1999) Isolation within species and the
history of glacial refugia. In: Hollingswoth PM, Bateman RM,
Gornall RJ (eds) Molecular systematics and plant evolution.
Taylor & Francis, London, pp 20–33
Garcia D (2001) Effects of seed dispersal on Juniperus communis
recruitment on a Mediterranean mountain. J Vegetat Sci
12(6):839–848
Ge S, Hong DZ, Wang HQ, Liu YZ, Yhang CM (1998) Population
genetic structure and conservation of an endangered conifer
Cathaya argyrophylla (Pinaceae). Int J Plant Sci 159:351–357
Gonza
´lez-Sampe
´riz P, Valero-Garce
´s BL, Carrio
´n JS, Pen
˜a-Monne
´
JL, Garcı
´a-Ruiz JM, Martı
´-Bono C (2005) Glacial and Lategla-
cial vegetation in northeastern Spain: new data and a review.
Quart Int 140–141:4–20
Greuter W, Burdet HM, Long G (1984) Med.-Checklist 1. Conserva-
toire et Jardin Botaniques de la Ville de Gene
`ve et Med.-Checklist
Trust of OPTIMA, Gene
`ve
High level of genetic differentiation of Juniperus phoenicea 171
123
Hewitt GM (2004) Genetic consequences of climatic oscillations in
the Quaternary. Philos Trans R Soc Lond, Ser B Biol Sci
359(1442):183–195
Huntley B (1988) Glacial and holocene vegetation history, Europe.
In: Huntley B, Webb TIII (eds) Vegetation history, handbook of
vegetation science, vol 7. Kluwer, Dortrecht, pp 341–383
Jalas J, Suominen J (eds) (1973) Atlas florae europaeae 2. The
Committee for Mapping the Flora of Europe and Societas
Biologica Fennica Vanamo, Helsinki
Jordano P (1991) Gender variation and expression of monoecy in
Juniperus phoenicea (L.) (Cupressaceae). Bot Gaz 152(4):476–
485
Jordano P (1993) Geographic ecology and variation of plant-seed
disperser interaction: southern Spanish junipers and frugivorous
thrushes. Vegetatio 107/108:85–104
Kara N, Korol L, Isik K, Schiller G (1997) Diversity in Pinus brutia
Ten.: altitudinal variation. Silvae Genet 46(2/3):155–161
Kerfoot O, Lavranos JJ (1984) Studies on the flora of Arabia. X.
Juniperus phoenicea L. and J. excelsa M. Bieb. Notes Roy Bot
Garden, Edinburgh 41:483–498
Kornas
´J, Medwecka-Kornas
´A (2002) Geografia ros
´lin (Plant geog-
raphy). Wydawnictwo Naukowe PWN, Warszawa (in Polish)
Korol L, Kara N, Isik K, Schiller G (1997) Differentiation among and
within natural and planted Cupressus sempervirens L. Eastern
Mediterranean populations. Silvae Genet 46(2/3):151–155
Kostrowicki AS (1999) Geografia biosfery. Wydawnictwo Naukowe
PWN, Warszawa (in Polish)
Lagercrantz U, Ryman N (1990) Genetic structure of Norway spruce
(Picea abies): concordance of morphological and allozymatic
variation. Evolution 44:38–53
Lebreton P (1983) Nouvelles donne
´es sur la distribution au Portugal
et en Espagne des sous-espe
`ces du genevrier de Phenicie
(Juniperus phoenicea L.). Agron Lusit 42:55–62
Lebreton P, Rivera D (1989) Analyse du taxon Juniperus phoenicea
L. sur des bases biochimiques et biome
´triques. Nat Mosnpel, Ser
Bot 53:17–41
Lebreton P, Thivend S (1981) Sur une sus-espe
`ce du gene
´vrier de
Phe
´nicie Juniperus phoenicea L., de
´finie a
`partir de crite
`res
biochimiques. Nat Mosnpel, Ser Bot 47:1–12
Lewandowski A, Samoc
´ko J, Boratyn
´ski A, Mejnartowicz L (2000)
Inheritance and linkage of allozymes in Juniperus phoenicea L.
(Cupressaceae). Acta Soc Bot Pol 69:201–205
Loveless MD, Hamrick JL (1984) Ecological determinants of genetic
structure in plant populations. Ann Rev Ecol System 15:65–95
Mandin J-P (2005) De
´couverte de tre
`s vieux gene
´vriers de Phe
´nicie
(Juniperus phoenicea L.) dans les gorges de l’Arde
`che (France).
J Bot Soc Bot France 29:53–62
Mazur M, Boratyn
´ska K, Marcysiak K, Go
´mez D, Tomaszewski D,
Didukh J, Boratyn
´ski A (2003) Morphological variability of
Juniperus phoenicea (Cupressaceae) from three distant localities
on Iberian Peninsula. Acta Soc Bot Pol 72:71–78
Mehes MS, Nkongolo KK, Michael P (2007) Genetic analysis of
Pinus strobus and Pinus monticola populations from Canada
using ISSR and RAPD markers: development of genome-
specific SCAR markers. Plant Syst Evol 267:47–63
Meister J, Hubaishan MA, Kilian N, Oberpiler C (2006) Temporal
and spatial diversification of the shrub Justicia areysiana Deflers
(Acanthaceae) endemic to the monsoon affected coastal moun-
tains of the southern Arabian Peninsula. Plant Syst Evol
262:153–171
Meloni M, Perini D, Filigheddu R, Binelli G (2006) Genetic variation
in five Mediterranean populations of Juniperus phoenicea as
revealed by inter-simple sequence repeat (ISSR) markers. Ann
Bot 97:299–304
Mu
¨ller-Starck G, Baradat P, Bergmann F (1992) Genetic variation
within European tree species. New Forests 6:23–47
Myers ER, Chung MY, Chung MG (2007) Genetic diversity and
spatial genetic structure of Pinus strobus (Pinaceae) across an
island landscape inferred from allozyme and cpDNA markers.
Plant Syst Evol 264:15–30
Nei M (1972) Genetic distance between populations. Am Nat
106:283–292
Nei M (1978) Estimation of average heterozygosity and genetic
distance from a small number of individuals. Genetics 89:583–
590
Petit R, Hampe A, Cheddadi R (2005) Climate changes and tree
phylogeography in the Mediterranean. Taxon 54:877–885
Que
´zel P, Barbero M (1981) Contribution a l’etude des formations
pre-steppiques a Genevriers au Maroc. Bol Soc Broter 53:1137–
1160
Que
´zel P, Pesson P (1980) Biogeographie et ecologie des coniferes
sur le poutrour mediterraneen. Actualites d’ecologie forestiere.
Gauthier-Villars, Paris, pp 205–255
Que
´zel P, Barbero M, Benabid A, Rivas-Martines S (1992) Contri-
butions a
`l’e
´tude des groupements forestiers et pre-forestiers du
Maroc oriental. Studia Bot 10:57–90
Rieseberg LH, Wendel J (2004) Plant speciation—rise of the poor
cousins. New Phytol 161:3–8
Rieseberg LH, Church SA, Mojran CL (2004) Integration of popu-
lations and differentiation of species. New Phytol 161:59–69
Sneath HA, Sokal R (1973) Numerical taxonomy. The principles and
practice of numerical classification. W.H. Freeman, San
Francisco
Sokal RR, Rohlf TJ (2003) Biometry: the principles and practice of
statistics in biological research. Eighth printing, W.H. Freeman,
San Francisco
Tabachnik BG, Fidell LS (1996) Using multivariate statistics. Harper
Collins College Publishers, New York
Terrab A, Talavera S, Arista M, Paun O, Stuessey FT, Tremetsberger
K (2007) Genetic diversity and geographic structure at chloro-
plast microsatellites (cpSSR) in endangered West Mediterranean
firs (Abies spp., Pinaceae). Taxon 56:409–416
Terrab A, Scho
¨nswetter P, Talavera S, Vela E, Stuessey TF (2008)
Range-wide phylogeography of Juniperus thurifera L., a
presumptive keystone species of western Mediterranean vegeta-
tion during cold stages of the Pleistocene. Mol Phylogen Evol
48:94–102
Tzedakis PC (2004) The Balkans as prime glacial refugial territory of
European temperate trees. In: Griffiths HI, Krys
ˇtufek B, Reed
JM (eds) Balkan biodiversity. Kluwer, Dortrecht, pp 49–68
Uzquiano P, Arnaz AM (1997) Consideraciones paleoambientales del
Tardiglacial y Holoceno inicial en el Levante espan
˜ol: macr-
orestos vegetales de El Tossal de la Roca (Vall d’Alcala,
Alicante). Anal Jard Bot Madrid 55:125–133
van Andel TH (2002) The climate and landscape of the middle part of
Weichselian Glaciation in Europe: the stage 3 project. Quater
Res 57:2–8
Watała C (2002) Biostatystyka - wykorzystanie metod statystycznych
w pracy badawczej w naukach biometrycznych. a-Medica Press,
Bielsko-Biała
Yeh FC, Yang RC, Boyle T (1999) Popgene version 1.31. University
of Alberta, Canada
Zohary M (1973) Geobotanic foundation of the Middle East. Gustav
Fischer Verlag and Sweets and Zeitlinger, Stuttgart
172 A. Boratyn
´ski et al.
123
