Content uploaded by Nina Sajna
Author content
All content in this area was uploaded by Nina Sajna
Content may be subject to copyright.
*e-mail: nina.sajna@uni-mb.si
POPULATION GENETICS OF THE NARROW ENDEMIC
HLADNIKIA PASTINACIFOLIA RCHB. (APIACEAE)
INDICATES SURVIVAL IN SITU DURING THE PLEISTOCENE
NINA ŠAJNA1*, TATJANA KAVAR2, JELKA ŠUŠTAR-VOZLIÈ2, and MITJA KALIGARIÈ1
1University of Maribor, Biology Department, Faculty of Natural Sciences and
Mathematics, Koroška c. 160, SI – 2000 Maribor, Slovenia
2Agricultural Institute of Slovenia, Field Crop and Seed Production Department,
Hacquetova 17, SI-1000 Ljubljana, Slovenia
Received July 15, 2011; revision accepted April 6, 2012
Hladnikia pastinacifolia Rchb., a narrow endemic, has an extremely restricted distribution in Trnovski gozd
(Slovenia), despite the presence of many sites with suitable habitats. We compared the morphological traits of
plants from different populations and habitats. The overall pattern showed that the smallest plants, with low fruit
number, are found on Èaven (locus classicus or type locality); the largest individuals, with high fruit number, grow
in the Golobnica gorge. As judged by plant size and seed set, the optimal habitats are screes. We used RAPD mark-
ers to estimate genetic variation between and within populations, as well as between and within the northern and
the southern parts of the distribution area. Hladnikia showed only a low level of RAPD variability. AMOVA parti-
tioned the majority of genetic diversity within selected populations. The low genetic differentiation between popu-
lations and their genetic depauperation indicates survival in situ, since the Trnovski gozd plateau most likely was
a nunatak region in the southern Prealps during Pleistocene glaciations. Later range expansion of extant popula-
tions was limited by poor seed dispersal. We also analyzed the cpDNA trnL-F intergenic spacer to check whether
the sequence is useful for studying the phylogenetic relationships of Hladnikia within the family Apiaceae
(Umbelliferae). Our results support the assertion that H. pastinacifolia is an old taxon.
KKeeyy wwoorrddss::Apioideae, Pleistocene, RAPD, cpDNA, nunatak, plant genetics, Hladnikia pastinacifo-
lia, relict species, endemites.
ACTA BIOLOGICA CRACOVIENSIA Series Botanica 54/1: 1–13, 2012
DOI: 10.2478/v10182-012-0009-8
PL ISSN 0001-5296 © Polish Academy of Sciences and Jagiellonian University, Cracow 2012
INTRODUCTION
Rare species are characterized by low abundance,
restricted distribution area and/or small geographi-
cal range (Gaston, 1997). The Trnovski gozd karst
plateau in western Slovenia belongs to the southern
Prealps and is one of the places where rare species,
many of them endemic, are common. Another rea-
son for its overall high species diversity is its bio-
geographical position as a meeting zone between
Submediterranean, Dinaric and Alpine biomes.
One of the most remarkable species among the
Pleistocene survivors found there is Hladnikia
pastinacifolia Rchb., the only representative of this
endemic genus, with an extremely narrow distribu-
tion in a 4 km2area. The species is not a habitat
specialist, however: it can be found growing in stony
grassland, rock crevices and screes. Hladnikia is
regarded as an ancient paleoendemic genus contain-
ing a single Tertiary relict species (Mayer, 1960;
Pawlowski, 1970), since paleoendemics are rem-
nants of previously widespread taxa and their cur-
rent distribution is sometimes reduced to small
refugia (Kruckenberg and Rabinowitz, 1985).
Rare narrow endemic species occurring in a few
small populations have to cope with random genetic
drift, inbreeding, a stronger founder effect, and a
greater potential for demographic bottlenecks that
result in low genetic variability (Kunin and Gaston,
1997). Genetic diversity analyses of narrow
endemics have generated considerable interest,
especially because depauperated genetic variability
is an important factor for conservation planning
(Oiki et al., 2001; Jimenez et al., 2002; Cole, 2003;
Torres et al., 2003; Gaudeul et al., 2004;
Vilatersana et al., 2007; Mameli et al., 2008).
Studies of rare species often involve comparisons
with common congeners (Ayres and Ryan, 1999;
Gitzendanner and Soltis, 2000; Kim and Chang,
2005). Various genetic markers have been used for
these analyses, most often allozymes, RAPDs, AFLPs
and microsatellites (SSRs).
The numerous synonyms of Hladnikia show
that has been variously considered to belong to the
genera Falcaria Fabr., Carum L., Oenanthe L. or
Prionitis Adans. (Sušnik, 1964; Hegi, 1975).
Reichenbach (1831) recognized it as the distinct
genus Hladnikia with one species H. pastinacifolia.
The botanist W.D.J. Koch disagreed with the place-
ment of this species in the genus Hladnikia. He
included another monotypic endemic species from
Trnovski gozd, Athamanta golaka Hacq. instead,
and changed its name to Hladnikia golaka (now
Grafia golaka (Hacq.) Rchb.) (Wraber, 1990). Despite
the disagreement, Reichenbach's name was retained.
Recently the taxonomic status of Hladnikia was stud-
ied using nrDNA ITS sequence data (Ajani et al.,
2008). Those results placed H. pastinacifolia within
the tribe Careae, as more closely related to the
Falcaria group than to the Carum group, genera to
which Hladnikia were thought to belong based on its
morphology and fruit anatomy.
Hladnikia pastinacifolia was discovered in
1819 (Fleischmann, 1844), but no population stud-
ies of it have been made. In this work our first task
was to compare plants from different populations
and habitats morphologically. Our second was to
determine the levels of genetic diversity within and
between populations by RAPD analysis. The third
was to analyze the cpDNA trnL-F intergenic spacer
in order to use the data to elucidate the phylogenet-
ic relationships of Hladnikia within the family
Apiaceae (Umbelliferae).
MATERIALS AND METHODS
STUDY SPECIES
Hladnikia pastinacifolia is a monocarpic herba-
ceous perennial that develops into a flowering plant
during several vegetation periods. In the early life
stages it forms rosettes. The leaf area gradually
increases with age, and the leaf shape develops
from simple to lobate. What triggers flowering is yet
unknown. The flowers are insect-pollinated. Seeds
form from the end of August through September.
We still lack information about the breeding sys-
tem. The existence of numerous and varied pollina-
tors, as well as protandry, support the idea that
outcrossing operates. The pollen:ovule ratios for
flowers have been calculated at ~4000:2 (Šajna N,
unpublished data), suggesting low-efficiency polli-
nators which might also be palynophagous
(Cruden, 1976). The fruit is a schizocarp, bearing
two 4 mm mericarps with an underdeveloped
embryo. The seeds have no specialized dispersal
adaptations.
Hladnikia pastinacifolia is an extremely rare
and strictly protected species; this limited our
research. We followed the strict restrictions on sam-
pling quantities issued by the Republic of Slovenia's
Institute for Nature Conservation, which often
allowed only small samples. We employed mainly
non-destructive research methods.
STUDY SITE
Trnovski gozd is a large limestone plateau in a
mountainous area up to 1500 m a.s.l. Together with
the South Julian Alps it forms an orographic barri-
er between the Mediterranean and moderate conti-
nental climatic regions. This area receives one of the
highest amounts of precipitation (above 3000 mm
on more than 120 precipitation days) in Slovenia
(Melik, 1960). Most of the area is covered with nat-
ural forest stands. Traditional farmland is distrib-
uted sparsely. The geographic range of H. pastinaci-
folia is limited to the southern slopes of the
Trnovski gozd plateau and two isolated locations
9 km away on the northern slopes. The entire dis-
tribution area is included in the Natura 2000
Network as a Site of Community Interest. We chose
Šajna et al.
2
TABLE 1. Field name, location (coordinates), elevation, habitat types (1 – rock crevices; 2 – stony pasture; 3 – scree),
and estimated size of five studied populations of Hladnikia pastinacifolia
five populations for the study: three at the southern
edge, and both known locations from the northern
edge (Fig. 1). The location at Predmeja is a small
stony pasture with a man-made stone wall. The
habitat is threatened by encroaching woody vegeta-
tion (Šajna et al., 2011). Location 2 was previously
described as only a secondary occurrence at the
road verge (Èušin, 2004) near the Golobnica gorge.
When we made a detailed search in the field, howev-
er, we observed that plants were also present in rub-
ble and rock crevices of the eastern rocky walls with-
in the gorge (Šajna et al., 2009), which should there-
fore be classified as a primary habitat location.
Location 3 is the type locality (locus classicus): stony
grassland with co-occurring smaller screes. On the
northern edge is a larger population stand on a small
shelf overgrown with vegetation belonging to the
association Primulo carniolicae-Caricetum firmae
Dakskobler 2006 on the peak of Mt. Poldanovec
(Dakskobler, 2006). The second northern location is
a small rock formation within a beech forest where
H. pastinacifolia is found on steep walls; the top of
the rock is covered by a Pinus mugo stand.
MORPHOMETRIC ANALYSES
In 2006, flowering plants from populations 1–4 (Fig.
1, Tab. 1) were randomly chosen for morphometric
analysis. Population 5 was excluded because very
few plants flowered that season. All measurement
and scoring of selected plant traits were done non-
destructively. The following ten plant traits were
measured from a total of 103 specimens: plant
height, leaf length, leaf width, rosette diameter, cen-
tral umbel diameter, peduncle height, number of
umbellets of central umbel, number of fruit in cen-
tral umbel, number of umbels per plant, and tap
root diameter.
We compared plants from different populations
or habitats using ANOVA followed by the Tukey
HSD. The morphological results were log-trans-
formed (10log) when necessary to obtain homogene-
ity of variance (Levene's test) and to achieve a nor-
mal distribution (Shapiro-Wilk's test). To a signifi-
cant degree the plants of the various populations
were characterized by heterogeneity of variance of
leaf width, number of umbellets of central umbel,
and number of umbels per plant, so these were not
included in ANOVA.
For the same reason we did not use the follow-
ing traits for comparison of plants growing in the
three types of habitats: rosette diameter, number of
umbellets of central umbel, number of fruit in cen-
tral umbel, and tap root diameter. Additionally, we
performed discriminant analysis to maximize
between-habitat variation in relation to within-habi-
tat variation.
RAPD ANALYSIS
For genetic analyses, leaf samples of 20 individual
plants from each of five populations representing
the entire distribution area were collected (Fig. 1).
Total genomic DNA was extracted following a modi-
Population genetics of endemic Hladnikia pastinacifolia Rchb.
3
FFiigg.. 11.. (aa) Map of Slovenia with location of Trnovski gozd karst plateau (arrow), (bb) Trnovski gozd (A) between Julian
Alps (D) and Dinaric Mts. (B – Hrušica, C – Nanos). Square indicates distribution area of Hladnikia pastinacifolia,
(cc) Detailed geomorphology and locations of the 5 studied populations: 1–3 in southern part, 4 and 5 in northern part
of distribution area (locality numbers correspond to numbers in Table 1).
fied CTAB protocol (Šuštar-VozliÈand Javornik,
1999).
RAPD-PCR amplifications were performed in a
25 μl volume with a GeneAmp PCR System 9700
thermocycler (Perkin Elmer, U.S.A.). Each reaction
contained 1x·PCR buffer, 200 μM dNTPs, 1.2 pmol
of each primer (Tab. 3), 2.5 mM MgCl2, 0.7 U Taq
DNA polymerase (Promega, U.S.A.) and 20 ng tem-
plate DNA. The PCR profile consisted of initial
denaturation at 94°C for 5 min, followed by 41
cycles of strand denaturation at 94°C for 1 min,
primer annealing at 37.5°C for 1 min 40 s, DNA
extension at 72°C for 2 min, and final extension at
72°C for 10 min. PCR products were separated elec-
trophoretically in ethidium bromide-stained 1.4%
agarose-TBE gels. A negative control without DNA
was included to check for contamination.
Reproducible RAPD bands were scored as binary
presence/absence data. Statistical analysis was done
with GENALEX 6 (Peakall and Smouse, 2006).
Because of the high frequency of ghost bands in
the RAPD profiles of some samples, we omitted
those samples from the final analysis. We believe
that the presence of these bands was a consequence
of a fungus infection, and had we included those
results the genetic diversity would have been falsely
increased. Consequently, 64 individuals were
included in AMOVA.
cpDNA trnL-F SEQUENCES
A fragment of chloroplast DNA (trnL-F) was ampli-
fied using universal primers "c" and "f" from Taberlet
et al. (1991). PCR amplification was performed in a
20 μl reaction mixture with a GeneAmp PCR System
9700 thermocycler (Perkin Elmer, U.S.A.). Each
reaction contained 1x·PCR buffer, 200 μM dNTPs,
10 pmol of each primer, 2.5 mM MgCl2, 0.5 U DNA
polymerase (Biotools, Spain) and 50 ng template
DNA. The PCR profile consisted of initial denatura-
tion at 80°C for 5 min, followed by 35 cycles of
strand denaturation at 94°C for 1 min, primer
annealing at 50°C for 1 min, DNA extension at 72°C
for 2 min, and final extension at 72°C for 10 min.
Purified PCR fragments were sequenced using the
ABI Prism BigDye Terminator Cycle Sequencing
Ready Reaction Kit (PE Applied Biosystems) on the
ABI PRISM 310 DNA Sequencer (PE Applied
Biosystems).
A search for similar sequences in the NCBI
nucleotide collection (nr/nt) was performed with
BLASTN programs (Altschul et al., 1997). Selected
sequences were retrieved from GenBank and
aligned with the sequence of H. pastinacifolia with
ClustalX (Thompson et al., 1997) and refined man-
ually. Phylogenetic analyses using the neighbor-join-
ing (NJ), maximum parsimony (MP) and maximum
likelihood (ML) methods were conducted using
MEGA ver. 3.1 (Kumar et al., 2004) or the Phylip
package (Felsenstein, 2004). Genetic distances were
calculated using the Kimura two-parameter method
(Kimura, 1980), where the transversion/transition
ratio was 2:1. Gaps were either excluded or includ-
ed in the dataset. When they were included, scoring
of gaps was the same as for transitions (one base
indel) or transversions (more than two base indels).
RESULTS
MORPHOLOGICAL DIFFERENCES BETWEEN
POPULATIONS AND HABITATS
The investigated plants differed significantly
between populations. There was a general difference
between plants from Golobnica and those from the
other populations. The plants from Golobnica dif-
fered significantly in vegetative traits (plant height,
leaf length, rosette diameter) and in reproductive
structures (peduncle height, umbel diameter), which
were bigger (Tab. 2). However, the traits associated
Šajna et al.
4
TABLE 2. Values (mean ±SD) for each studied trait of four Hladnikia pastinacifolia populations. [Values with differ-
ent letters differ significantly from each other (Tukey HSD test at P<0.05)]
*Trait was 10log-transformed for ANOVA. **P <0.01
with reproductive success (number of fruit on main
umbel) did not differ from the other populations
except in those from the type locality, Èaven, where
the smallest individuals were found, with only two-
thirds the number of fruit.
The plants differed significantly between habi-
tats (AMOVA) in plant height, leaf length, diameter of
central umbel, and peduncle height. Plants from
stony grassland differed most, with the smallest
inflorescences as well as smallest overall size. The
plants found growing in screes were biggest. The big-
ger the plant, the higher the probability that the
plant would form more prostrate branching of the
stem, resulting in more inflorescences of the second
or even third degree, which we observed only on
plants growing in screes. The plants from rock
crevices were morphologically similar to plants
from the other habitat types; the biggest differences
were between plants from screes and those from
stony pastures (Fig. 2; discriminant analysis:
F16,176=8.1507, P< 0.05).
RAPD ANALYSIS
First, 44 10-mer primers (Operon Technologies
Ltd., Alameda, U.S.A.) were screened for RAPD pro-
files with three H. pastinacifolia samples in order to
find the effective primers for RAPD analysis. Then
the ten primers giving reproducible bands were
selected for further analysis (Tab. 3). The average
number of bands was 3.9 per primer, and fragment
size ranged from 250 bp to 1000 bp (Fig. 3). Sixty-
four individuals were included in the statistical
analyses to estimate variation within and between
populations (11–15 for each population).
Population genetics of endemic Hladnikia pastinacifolia Rchb.
5
FFiigg.. 22..Discriminant analysis of morphometric data
grouped by habitat types where Hladnikia pastinacifolia
is found. Function 1 discriminates mostly between plants
from screes and those from stony pastures.
FFiigg.. 33.. Example of amplified RAPD fragments (primer OPU 19). Lanes are marked with numbers corresponding to pop-
ulations: 1–21 – Èaven; 22–34 – Zeleni rob; 35–54 – Poldanovec; 55–76 – Predmeja; 77–97 – Golobnica. Nc – negative
control; M – molecular weight scale (1 kb DNA ladder).
TABLE 3: RAPD primers used in this study, and number
of amplified and scored bands
Differences between individuals were not pro-
nounced, since many individuals from different pop-
ulations shared the same RAPD profile. The data
confirmed low levels of differentiation between pop-
ulations. AMOVA indicated that the majority of the
variation pertained to differences within populations
(90%), and only 10% was attributable to variation
between populations. When we grouped populations
into two regions (northern or southern location of
the population), AMOVA did not show any differ-
ences between regions.
Principal component analysis of RAPD data did
not show genetic differentiation of populations (Fig.
4). The first three factors accounted for 66% of the
total variation in the data set (32%, 19% and 15% by
the first, second and third factors respectively).
cpDNA trnL-F
Ten individuals (two from each of five populations)
were sequenced. All samples generated an identical
840 bp sequence, which included part of the tRNA-
Leu (trnL) gene and the trnL-F intergenic spacer. No
length variation or differences in the proportion of
nucleotide were found. The sequence has been
deposited in GenBank under Acc. No. EU514464
and used for a BLAST search against the NCBI nr
database. Sequences of the first 100 BLAST hits were
used for construction of the NJ tree (Appendix 1).
Further analyses were limited to the dataset of 21
sequences from the group of the closest relatives of
H. pastinacifolia. We randomly chose one sequence
of each genus, and the sequence of Apiopetalum
velutinum Baill. was selected as out-group.
Phylograms were constructed by various methods
(NJ, MP, ML), with gaps included or excluded. In
some cases H. pastinacifolia clustered within the
Selineae; in other cases the relationship between
Apium graveolens L., H. pastinacifolia and Selineae
remained unresolved. However, the majority of trees
had the topology presented in Figure 5, which is,
according to the bootstrap values, moderately sup-
ported. Bootstrap values ranged from 50% to 95%
(average 71%), and only 10 clades were resolved
having values ≥50%. The genus Hladnikia formed a
single clade and was clustered in the group com-
posed of the Selineae (e.g., Angelica L., Cymopterus
Raf., Zizia W.D.J. Koch, Aletes Coult. & Rose,
Lomatium Raf.), Apiaceae incertae sedis (e.g.,
Tilingia Regel & Tiling) and Apium clade. These all
belong to the apioid superclade (Plunkett and
Downie, 1999). Hladnikia is also closely related to
Scandiceae (e.g., Daucus L., Myrrhis P. Mill.,
Osmorhiza Raf.) (Fig. 5).
DISCUSSION
MORPHOLOGICAL DIFFERENCES
BETWEEN POPULATIONS AND HABITATS
The plants of all the studied populations exhibited
low genetic diversity but pronounced morphological
differences, indicating high phenotypic plasticity of
vegetative traits as well as traits of reproductive
structures. The smallest plants, with low fruit num-
ber, were those from Èaven (locus classicus); the
largest plants, with high fruit number, grew in the
Golobnica gorge. This can be explained by the habi-
tats of these two locations. At Èaven the plants grew
in stony pastures; those plants were significantly
smaller (shorter plants and peduncles, smaller
leaves). Judged by plant size and seed set, the screes
in the Golobnica gorge seem to be the optimal habitat.
Although it has no special habitat type prefer-
ence except for preferring some degree of distur-
bance, and despite its high phenotypic plasticity, H.
pastinacifolia has a very limited distribution. We
suggest that it is restricted by traits related to dis-
persal and persistence in disturbed habitats. For
example, the monocarpic life cycle of H. pastinaci-
folia is characteristic for species with transient
occurrence. Also, scree slopes and similar disturbed
habitats such as river banks are more or less per-
manent features of the landscape but they are not
common and are often widely separated; dispersal
within such habitats is strongly linked to the direc-
tion of gravel movement. Rock crevices represent a
secondary habitat; this chasmophytic habitat seems
to have become a local refuge for H. pastinacifolia
as it presented stabler microclimatic conditions.
Many chasmophytic species are believed to originate
Šajna et al.
6
FFiigg.. 44.. Ordination of the investigated populations based on
RAPDs along the first three axes extracted by principal
component analysis.
from before the Pleistocene (Davis, 1950; Rune-
mark, 1969).
GENETIC DIVERSITY
The analysis of genetic diversity (RAPD) included
plants from the entire distribution area of the species.
We expected to find at least some genetic divergence
between the southern and the northern populations,
since they are divided by a continuous dense forest
stand. West Dinaric fir-beech forest prevails. The
stands form the geographical variant Omphalodo-
Fagetum var. geogr. Saxifraga cuneifolia, further
divided into two geographical subvariants: western –
subvar. geogr. Anemone trifolia, and central-eastern
– subvar. geogr. Omphalodes verna (Surina, 2002).
Especially in the northeastern part of the Trnovski
gozd plateau, where locations 4 and 5 (Fig. 1) are sit-
uated, natural black pine stands can be found in less
favorable habitats such as steep rocky slopes
(Dakskobler, 1999). Spruce forests are restricted to
freezing ravines and cold, moist, shaded sites (Surina
and Vreš, 2009). It is believed that this area has been
at least partly forested since the end of the Pleistocene
glaciations. The forest was not intensively exploited
before the 16th century, and since then it has been
managed in several ways including selective cutting
(16th–18th cent.) and clearcutting in some parts in
the north (Surina, 2001).
Against expectations, RAPD analysis showed
that all populations express a high level of
monomorphic bands and that many individuals
from different populations share the same RAPD
profile. Recently some authors have challenged the
value of RAPD markers, citing low reproducibility of
markers and suggesting that AFLP markers are
preferable. Even though RAPDs show different
genetic similarity ranges, as do AFLPs, they often
show similar overall results (Ćwiklińska et al.,
2010).
The majority of the genetic variation in H. pasti-
nacifolia was partitioned within populations, and
did not differentiate the southern or northern parts
of the distribution area. Higher genetic diversity
within than between populations has been described
as a characteristic of perennial and outcrossing
species (Despres et al., 2002).
The lack of differences between the northern
and the southern populations can be explained in
two ways: as the result of continual gene flow among
populations (via pollen or seed dispersal), or as the
consequence of the relatively recent establishment of
extant populations from a common, genetically
depauperated founder.
In regard to the possibility of gene flow between
populations, we observed that H. pastinacifolia
inflorescences frequently were visited by various
insects (mainly Hymenoptera, Coleoptera and
Population genetics of endemic Hladnikia pastinacifolia Rchb.
7
FFiigg.. 55.. Neighbor-joining tree representing the relationships between Hladnikia pastinacifolia and its closest relatives
from the Apiaceae family. Bootstrap values higher than 50 are given at the nodes.
Diptera), suggesting cross-pollination. Some pollina-
tors can fly distances from 2.6 km up to even 9.9
km, but mostly within 1 km (Walther-Hellwig and
Frankl, 2000; Kraus et al., 2009). Even with the pos-
sibility of selfing, mating would occur only between
a small proportion of flowering individuals each sea-
son, thereby creating "temporal subpopulations" and
delaying or reducing inbreeding depression (Lopez-
Pujol et al., 2002). On the other hand, seed disper-
sal is limited because the seeds lack adaptations for
specialized dispersal. We did not observe myrmeco-
chory or epizoochory. There is only a small potential
for endozoochory, since the seeds have a strong,
unpleasant odor and taste (this is probably why we
seldom observed damaged seeding umbels). Among
the dispersal vectors we cannot neglect the human
presence. All localities of the investigated popula-
tions experience very frequent visits as trekking or
climbing destinations or as hunting spots. Wind dis-
persal seems the most promising vector. Most seeds
remain near the maternal plant in rainy autumns,
but we observed the breaking off of the entire umbel
when dry conditions prevailed (personal observa-
tions in 2004–2009). A similar dispersal strategy
was noted for the thermophilous Peucedanum are-
narium subsp. arenarium (Šera et al., 2005).
However, dry autumn conditions are rather the
exception in this region; poor seed dispersal may be
the reason for the clustered distribution of seedlings
we observed on the microscale (Šajna N., unpub-
lished data). The type of pollination and limited seed
dispersal would indicate that gene flow, if any, is
very limited. However, limited gene flow would sug-
gest stronger genetic differentiation between the
northern and southern populations, which we did
not find using RAPDs.
Besides the lack of genetic differentiation
between the southern and the northern parts of the
distribution area, the mostly identical RAPD profiles
of specimens from all populations suggest that these
genetically homogeneous populations are the result
of severe bottlenecks which dramatically reduced or
eliminated some populations, whatever the time of
colonization (before or after glaciations). Recent
establishment of the extant H. pastinacifolia popu-
lations founded by seeds from a few nearby popula-
tions could explain the low interpopulation differen-
tiation. Its phylogeny indicates the isolated position
of H. pastinacifolia (Ajani et al., 2008) and does not
reveal any closely related species that could result
from allopatric speciation following the end of the
Pleistocene glaciations. Sušnik (1964) determined
the chromosome number in H. pastinacifolia
(x=11, the most frequent number in Apiaceae) and
diploid ploidy level, which are both consistent with
paleoendemics (Kruckenberg and Rabinowitz,
1985). We can therefore still assume that H. pasti-
nacifolia is an old taxon which lost its genetic vari-
ability at some point in the past. We lean toward the
idea that the loss of genetic variation occurred dur-
ing the Quaternary glaciations when the species
became restricted to safe sites where it survived in
situ. In the Quaternary a smaller local glacier exist-
ed in Trnovski gozd; the highest exposed peaks in
the northern part were not covered by the glacier
and were nunataks (Fig. 6; Perko 2001). The influ-
ence of the Adriatic Sea was reduced because the sea
level was 120 m lower and the northern shoreline
was at least 200 km south of the present line
(Correggiari et al., 1996). The recent distribution of
H. pastinacifolia represents only a slightly
increased range in the southern part and two frag-
mented locations in the northern part of that distri-
bution, separated by forest and restricted to steep
overhangs (Fig. 6). As many narrow endemics found
south of the Alps did not spread after the end of the
glaciations and remained within refugia (Vogel et al.,
1999), this seems to be the case for H. pastinacifo-
lia as well. It can be additionally explained by poor
seed dispersal. Paleoendemics frequently do not
have the genetic variability that would allow them to
increase their distribution area following a contrac-
Šajna et al.
8
FFiigg.. 66.. Map showing the Trnovski gozd high karst plateau
during the Quaternary (adapted after Perko, 2001). White
dots indicate recent distribution of Hladnikia pastinacifo-
lia (numbers of localities correspond to numbers in Table
1; white dots without numbers all belong to continuous
population 3). White crosshatched area represents the
nunatak region, light grey area is surface shaped by
Pleistocene glaciers, and hatched areas indicate valleys,
basins and karst depressions filled with periglacial debris
and gravel.
tion of the distribution. Erinus alpinus
(Scrophulariaceae) is an interesting case of post-
glacial range expansion. Erinus is a monotypic
genus common in subalpine habitats of
Southwestern and Central Europe, which survived
the Pleistocene glaciations in southern refugia
peripheral to the Alps, as well as in situ on nunataks
in the northern Prealps (Stehlik et al., 2002). When
Stehlik et al. (2002) studied genetic diversity at
cpDNA level in 12 populations throughout the dis-
tribution area ranging from southern France to the
Swiss-Austrian border, including nunataks like
Mount Rigi, they found no variation. They identified
three phylogeographic groups with AFLPs. One
group was represented by a single central Swiss
population on Mount Rigi. Individuals from Rigi had
a significantly low number of AFLP fragments, sup-
porting in situ survival in a nunatak region.
Compared to the spatial scale of that study, our
study describes a small-scale local situation. The
populations we studied were close enough to each
other to be regarded as one population. From this
point of view the lack of cpDNA diversity is not sur-
prising, especially when we consider the existence of
a local glacier with a local nunatak where H. pasti-
nacifolia might have survived in situ. In theory the
cost of survival in nunataks is isolation, causing
lower genetic diversity through inbreeding, as well
as population shrinking accompanied by random
genetic drift or recurrent bottlenecks also contribut-
ing to genetic depauperation. We can suggest that
these processes of survival in situ or a combination
of them are the cause of the loss of most genetic
diversity as indicated by RAPD markers in H. pasti-
nacifolia. The extant populations were founded by a
single lineage starting from a nunatak population. In
populations that have been rare for a very long peri-
od, natural selection can reduce deleterious alleles
and increase species fitness (Falconer and Mackay,
1996). If populations were experiencing different
selection pressures, this would result in higher
genetic variability between populations, but in the
habitats of H. pastinacifolia the ecological charac-
teristics are similar, exerting similar selection pres-
sure. This could help maintain low genetic differen-
tiation among the populations.
Modern modeling studies support the existence
of refugia in the southeastern Prealps (Tribsch and
Schönswetter, 2003) without reference to the exis-
tence of local nunataks. Phylogenetic data indicate
that potential locations of refugia may have been cli-
matically stable for long periods of time, some even
since the Tertiary (Medail and Diadema, 2009). As
mentioned, Trnovski gozd has an important geo-
graphic position. The high level of endemism in this
region is associated with the persistence of flora
since the Tertiary, as Trnovski gozd is regarded as a
sanctuary for species common before and at the
time of the Pleistocene glaciations (Wraber, 1990).
Trnovski gozd also seems to be a stable region of
atypical climate, since even today it is a sanctuary
for many alpine and arctic plant and animal species,
despite its southern location. However, it has never
before been considered a sanctuary in the sense of a
small nunatak region (peripheral nunatak as
defined by Schönswetter et al., 2004).
Among the many molecular phylogeographic
studies, seldom has nunatak survival been used to
explain observed genetic patterns in temperate
mountain ranges (Schneeweiss and Schönswetter,
2011); a rare exception is a study by Stehlik et al.
(2002). We believe there are many similarities
between the E. alpinus population they studied on
Mount Rigi and the H. pastinacifolia populations in
Trnovski gozd. The species most likely to retain a
molecular signature of survival in situ on nunataks
are those lacking the potential for rapid range
expansion and recolonization, which can be identi-
fied by the current distribution pattern and/or by
species ecology (Westergaard et al. 2011; Lohse et
al., 2011). The dispersal characteristics of H. pasti-
nacifolia, its distribution and its germination char-
acteristics (Šajna N., unpublished data) match the
description of a nunatak survivor well.
PHYLOGENY ACCORDING TO cpDNA trnL-F
Hladnikia pastinacifolia occupies an isolated posi-
tion within the Apiaceae family. It seems to have
shared a common ancestor with other species from
the apioid superclade, although statistical confi-
dence for such topology is not high and other expla-
nations cannot be ruled out. The topology of the NJ
tree shows that H. pastinacifolia forms a single
clade which is sister to a clade comprising species
from the genera of western and eastern North
America, and from East Asia. Within this clade the
North American species and Asian species are sister
clades. Dystaenia is monophyletic and unites with
Cnidium as a clade of the East Asian endemics
D. ibukiensis, endemic to Japan, and D. takesimana,
endemic to Ullung Island (Pfosser et al., 2006). The
North American species are defined as herbaceous
perennial apioid genera endemic to North America
(north of Mexico; Sun and Downie, 2004). Some are
polyphyletic (Cymopterus, Lomatium, Pteryxia;
Sun and Downie, 2004), some are monophyletic
(Polytaenia, Thaspium, Zizia; Sun et al., 2004), and
some are monotypic (Neoparrya, Shoshonea,
Harbouria). Hladnikia and closely related endemics
to some extent represent a typical disjunction pat-
tern for Northern Hemisphere Tertiary relicts: east-
ern North America, western North America, East
Asia, and Southeastern Europe. As said above, the
statistical certainty of these conclusions is low. We
must also bear in mind the sparseness of the data in
Population genetics of endemic Hladnikia pastinacifolia Rchb.
9
GenBank, among which members of the tribes
Scandiceae and Selinae are over-represented. At this
point the usefulness of the information obtained
from the cpDNA sequence for phylogeny is open to
question.
A single sequence was obtained from all ana-
lyzed samples of H. pastinacifolia. The level of vari-
ation within the species was generally very low or
zero at the cpDNA trnL-F locus. H. pastinacifolia is
an endemic and has an extremely narrow geograph-
ic distribution; there was also almost no variation
detected at the nuclear level between the RAPD pro-
files of samples from different populations.
ACKNOWLEDGEMENTS
We thank Elizabeta Komatar for laboratory assis-
tance, the Republic of Slovenia Institute of Nature
Conservation for issuing the sampling permit, and
the anonymous reviewers for their very helpful com-
ments and suggestions. Nina Šajna gratefully
acknowledges funding provided by the Society for
the Advancement of Plant Sciences in Vienna
(Austria).
REFERENCES
AJANI Y, AJANI A, CORDES MJ, WATSON MF, and DOWNIE RS.
2008. Phylogenetic analysis of nrDNA ITS sequences
reveals relationships within five groups of Iranian
Apiaceae subfamily Apioideae. Taxon 57: 383–401.
ALTSCHUL SF, MADDEN TL, SCHAFFER AA, ZHANG J, ZHANG Z,
MILLER W, and LIPMAN DJ. 1997. Gapped BLAST and
PSI-BLAST: A new generation of protein database search
programs. Nucleic Acids Research 25: 3389–3402.
AYRES DR, and RYAN FJ. 1999. Genetic diversity and con-
struction of the narrow endemic Wyethia reticulata and
its congener W. bolanderi (Asteraceae) using RAPD and
allozyme techniques. American Journal of Botany 86:
344–353.
COLE TC. 2003. Genetic variation in rare and common plants.
Annual Review of Ecology, Evolution, and Systematics
34: 213–237.
CORREGGIARI A, ROVERI M, and TRINCARDI F. 1996. Late
Pleistocene and Holocene evolution of the north Adriatic
sea. Quaternario 9: 697–704.
CRUDEN RW. 1976. Variation in pollen-ovule ratios and nectar
secretion – Preliminary evidence of ecotypic adaptation.
Annals of the Missouri Botanical Garden 63: 277–289.
ÈUŠIN B. 2004. Hladnikia pastinacifolia Rchb. –
rebrinèevolistna hladnikija, hladnikovka. In: Èušin B
[ed.], Natura 2000 v Sloveniji – Rastline. ZRC SAZU,
Ljubljana.
ĆWIKLIŃSKA A, BRODA Z, BOCIANOWSKI J, and DOBRZYCKA A. 2010.
The usefulness of RAPD and AFLP markers for deter-
mining genetic similarity in rye (Secale L.) species and
subspecies. Acta Biologica Cracoviensia Series
Botanica 52: 19–25.
DAKSKOBLER I. 1999. Contribution to the knowledge of the
association Fraxino orni-Pinetum nigrae Martin-Bosse
1967. Wissenschaftliche Mitteilungen aus dem
Niederösterreichischen Landesmuseum 12: 25–52.
DAKSKOBLER I. 2006. Calcareous open sedge swards and stony
grasslands (Seslerietalia caeruleae) on the northern
edge of the Trnovski gozd plateau (the Dinaric
Mountains, western Slovenia). Hacquetia 5: 73–112.
DAVIS PH. 1950. Cliff vegetation in the eastern Mediterranean.
Journal of Ecology 39: 63–93.
DESPRES L, LORIOT S, and GAUDEUL M. 2002. Geographic pat-
tern of genetic variation in the European globeflower
Trollius europaeus L. (Ranunculaceae) inferred from
amplified fragment length polymorphism markers.
Molecular Ecology 11: 2337–2347.
DOWNIE SR, WATSON MF, SPALIK K, and KATZ-DOWNIE DS. 2000.
Molecular systematics of Old World Apioideae
(Apiaceae): relationships among some members of tribe
Peucedaneae sensu lato, the placement of several island-
endemic species, and resolution within the apioid super-
clade. Canadian Journal of Botany 78: 506–528.
FALCONER DS, and MACKAY TFC. 1996. Introduction to
Quantitative Genetics. Longman, Harlow, Essex.
FELSENSTEIN J. 2004. PHYLIP (Phylogeny Inference Package)
version 3.6. Distributed by the author. Department of
Genome Sciences, University of Washington, Seattle.
FLEISCHMANN A. 1844. Übersicht der Flora Krain's. Laibach.
GASTON KJ. 1997. What is rarity? In: Kunin WE, Gaston KJ
[eds.], The Biology of Rarity, Causes and
Consequences of Rare-Common Differences.
Chapman & Hall, London.
GAUDEUL M, TILL-BOTTRAUD I, BARJON F, and MANEL S. 2004.
Genetic diversity and differentiation in Eryngium
alpinum L. (Apiaceae): comparison of AFLP and
microsatellite markers. Heredity 92: 508–518.
GITZENDANNER MA, and SOLTIS PA. 2000. Patterns of genetic
variation in rare and widespread congeners. American
Journal of Botany 87: 783–792.
HEGI G. 1975. Illustrierte Flora von Mitteleuropa 5(2). Paul
Parey, Berlin, Hamburg.
JIMENEZ JF, SANCHEZ-GOMEZ P, GUEMES J, WERNER O, and
ROSSELLO JA. 2002. Genetic variability in a narrow
endemic snapdragon (Antirrhinum subbaeticum,
Scrophulariaceae) using RAPD markers. Heredity 89:
387–393.
KIM H, and CHANG CS. 2005. The genetic variation in geo-
graphically peripheral populations of Bupleurum
euphorbioides (Apiaceae) and comparisons of genetic
variation with a congener, B. longiradiatum. Integrative
Biosciences 9: 27–35.
KIMURA M. 1980. A simple method for estimating evolutionary
rates of base substitutions through comparative studies
of nucleotide sequences. Journal of Molecular Evolution
16: 111–120.
KRAUS FB, WOLF S, and MORITZ RFA. 2009. Male flight distance
and population substructure in the bumblebee Bombus
terrestris. Journal of Animal Ecology 78:247–252.
KRUCKENBERG AR, and RABINOWITZ D. 1985. Biological aspects
of endemism in higher plants. Annual Review of
Ecology, Evolution, and Systematics 16: 447–479.
KUMAR S, TAMURA K, and NEI M. 2004. MEGA3: Integrated soft-
ware for molecular evolutionary genetics analysis and
Šajna et al.
10
sequence alignment. Briefings in Bioinformatics 5:
150–163.
KUNIN WE, and GASTON KJ. 1997. The Biology of Rarity.
Causes and Consequences of Rare-Common
Differences. Chapman & Hall, London.
LOHSE K, NICHOLLS JA, and STONE GN. 2011. Inferring the
colonisation of a mountain range – refugia vs. nunatak
survival in high alpine ground beetles. Molecular
Ecology 20: 394–408.
LÓPEZ-PUJOL J, BOSCH M, SIMON J, and Blanché C. 2002.
Allozyme variation and population structure of the very
narrow endemic Seseli farrenyi (Apiaceae). Botanical
Journal of the Linnean Society 138: 305–314.
MAMELI G, FILIGHEDDU R, BINELLI G, and MELONI M. 2008. The
genetic structure of the remnant populations of
Centaurea horrida in Sardinia and associated islands.
Annals of Botany – London 101: 633–640.
MAYER E. 1960. Endemiène cvetnice obmoèja jugovzhodnih
apneniških Alp, njihovega predgorja in ilirskega prehod-
nega ozemlja. In: Lazar J [ed.], Zbornik ob 150. letnici
botaniènega vrta v Ljubljani, SAZU, Ljubljana.
MEDAIL F, and DIADEMA K. 2009. Glacial refugia influence plant
diversity patterns in the Mediterranean Basin. Journal of
Biogeography 36: 1333–1345.
MELIK A. 1960. Slovenija – Slovensko Primorje 2(4).
Slovenska matica, Ljubljana.
OIKI S, KAWAHARA T, INOUE K, OHARA M, and MAKI M. 2001.
Random amplified polymorphic DNA (RAPD) variation
among populations of the insular endemic plant
Campanula microdonta (Campanulaceae). Annals of
Botany – London 87: 661–667.
PAWLOWSKI B. 1970. Remarques sur l'endemisme dans la flore
des Alpes et des Carpates. Vegetatio 21: 181–243.
PEAKALL R, and SMOUSE PE. 2006. GENALEX 6: genetic analy-
sis in Excel. Population genetic software for teaching and
research. Molecular Ecology Notes 6: 288–295.
PERKO D. 2001. Analiza Površja Slovenije s Stometrskim
Digitalnim Modelom Reliefa. ZRC SAZU, Ljubljana.
PFOSSER M, JAKUBOWSKY G, SCHLÜTER PM, FER T, KAT H,
STUESSY TF, and SUN B-Y. 2006. Evolution of
Dystaenia takesimana (Apiaceae), endemic to Ullung
Island, Korea. Plant Systematics and Evolution 256:
159–170.
PLUNKETT GM, and DOWNIE SR. 1999. Major lineages within
Apiaceae subfamily Apioideae: a comparison of chloro-
plast restriction site and DNA sequence data. American
Journal of Botany 86:1014–1026.
REICHENBACH HGL. 1830–1833. Flora Germanica Excursoria.
Lipsiae.
RUNEMARK H. 1969. Reproductive drift, a neglected principle in
reproductive biology. Botaniska Notiser 122: 90–129.
ŠAJNA N, KUŠAR P, SLANA NOVAK L, and NOVAK T. 2009. Notes
on thermo- and hygropreference in Leiobunum roseum
C.L. KOCH, 1839 (Opiliones: Sclerosomatidae) in a habi-
tat of Hladnikia pastinacifolia Reichenbach, 1831
(Spermatophyta: Apiaceae). Contributions to Natural
History 12: 1111–1123.
ŠAJNA N, KUŠAR P, SLANA NOVAK L, and NOVAK T. 2011. Benefits
of low-intensive grazing: co-occurrence of umbelliferous
plant (Hladnikia pastinacifolia Rchb.) and opilionid
species (Phalangium opilio L.) in dry, calcareous grass-
land. Polish Journal of Ecology 59: 777–786.
ŠERÁ B, FERÁKOVÁ V, KOCIANOVÁ E, VÁGENKNECHT V, MAJZLAN O,
and DÚBRAVCOVÁ Z 2005. Peucedanum arenarium subsp.
arenarium – a critically endangered species of the Slovak
flora (morphometry, distribution, biology). Biologia 60:
17–23.
SCHNEEWEISS GM, and SCHÖNSWETTER P. 2011. A re-appraisal
of nunatak survival in arctic-alpine phylogeography.
Molecular Ecology 12: 190–192.
SCHÖNSWETTER P, TRIBSCH A, STEHLIK I, and NIKLFELD H. 2004.
Glacial history of high alpine Ranunculus glacialis
(Ranunculaceae) in the European Alps in a comparative
phylogeographical context. Biological Journal of the
Linnean Society 81: 183–195.
STEHLIK I, SCHNELLER JJ, and BACHMANN K. 2002. Immigration
and in situ glacial survival of the low-alpine Erinus alpi-
nus (Scrophulariaceae). Biological Journal of the
Linnean Society 77: 87–103.
SUN F-J, and DOWNIE SR 2004. A molecular systematic inves-
tigation of Cymopterus and its allies (Apiaceae) based on
phylogenetic analyses of nuclear (ITS) and plastid (rps16
intron) DNA sequences. South African Journal of
Botany 70: 407–416.
SUN F-J, DOWNIE SR, and HARTMAN LR. 2004. An ITS-based
phylogenetic analysis of the perennial endemic Apiaceae
subfamily Apioideae of western North America.
Systematic Botany 29: 419–431.
SURINA B. 2001. Fitocenološke raziskave jelovo-bukovega
gozda (Omphalodo-Fagetum s. lat.) v zahodnem delu
ilirske florne province. Master thesis. University of
Ljubljana, Biotechnical Faculty, Ljubljana.
SURINA B. 2002. Phytogeographical differentiation in the
Dinaric fir-beech forest (Omphalodo-Fagetum s.lat.) of
the western part of the Illyrian floral province. Acta
Botanica Croatica 61: 145–178.
SURINA B, and VREŠ B. 2009. The association Drepanoclado
uncinati-Heliospermetum pusilli (Arabidetalia
caeruleae, Thlaspietea rotundifolii) in the Trnovski
gozd plateau (Slovenia, NW Dinaric Mts). Hacquetia 8:
31–40.
SUŠNIK F. 1964. Taksonomska in horološka problematika tak-
sona Hladnikia pastinacifolia Rchb. Ph.D. dissertation
University of Ljubljana, Ljubljana.
ŠUŠTAR-VOZLIÈJ, and JAVORNIK B. 1999. Genetic relationship
in cultivars of hop, Humulus lupulus L., determined by
RAPD analysis. Plant Breeding 118: 175–181.
TABERLET P, GIELLY L, PATTOU G, and BOUVET J. 1991.
Universal primers for amplification of three non-coding
regions of chloroplast DNA. Plant Molecular Biology 17:
1105–1109.
THOMPSON JD, GIBSON TJ, PLEWNIAK F, JEANMOUGIN F, and
HIGGINS DG. 1997. The ClustalX windows interface: flex-
ible strategies for multiple sequence alignment aided by
quality analysis tools. Nucleic Acids Research 25:
4876–4882.
TORRES E, IRIONDO JM, and PÉREZ C. 2003. Genetic structure
of an endangered plant, Antirrhinum microphyllum
(Scrophulariaceae): allozyme and RAPD analysis.
American Journal of Botany 90: 85–92.
TRIBSCH A, and SCHÖNSWETTER P. 2003. Patterns of endemism
and comparative phylogeography confirm palaeo-envi-
ronmental evidence for Pleistocene refugia in the Eastern
Alps. Taxon 52: 477–497.
Population genetics of endemic Hladnikia pastinacifolia Rchb.
11
VILATERSANA R, SUSANNA A, and BROCHMANN C. 2007. Genetic
variation in Femeniasia balearica (Compositae,
Cardue), an endemic and endangered monotypic genus
from the Balearic Islands (Spain). Botanical Journal of
the Linnean Society 153: 97–107.
VOGEL JC, RUMSEY FJ, SCHNELLER JJ, BARRETT JA, and GIBBY M.
1999. Where are the glacial refugia in Europe? Evidence
from pteridophytes. Biological Journal of the Linnean
Society 66: 23–37.
WALTHER-HELLWIG K, and FRANKL R. 2000. Foraging distances
of Bombus muscorum, Bombus lapidarius, and Bombus
terrestris (Hymenoptera, Apidae). Journal of Insect
Behavior 13: 239–246.
WEISING K, NYBOM H, WOLLF K, and KAHL G. 2005. DNA
Fingerprinting in Plants. Principles, Methods, and
Applications. CRC Press, Boca Raton.
WESTERGAARD KB, ALSOS IG, POPP M, ENGELSKJØN T, FLATBERG KI,
and BROCHMANN C. 2011. Glacial survival may matter
after all: nunatak signatures in the rare European popu-
lations of two west-arctic species. Molecular Ecology 20:
376–393.
WRABER T. 1990. Èaven, ein botanisch beruhmter Berg in
Slowenien. Carinthia 180: 195–210.
Šajna et al.
12
APPENDIX 1
Population genetics of endemic Hladnikia pastinacifolia Rchb.
