ArticlePDF Available

Development of microsatellite markers for horse-chestnut (Aesculus hippocastanum), their polymorphism in natural Greek populations, and cross-amplification in related species

Authors:
  • Institute of Dendrology, Polish Academy of Sciences, University of Zielona Góra, Department of Biological Sciences, Poland
  • WWF Hungary

Abstract and Figures

New nuclear microsatellite markers (SSRs) were developed for Aesculus hippocastanum, a relict tree species from the Balkan Peninsula. The development of microsatellites was done using the Illumina MiSeq PE300 platform. Out of a set of 500 SSRs designed, a subset of 13 loci was tested using 290 individuals from seven natural populations. Twelve species-specific loci were polymorphic. The number of alleles per locus ranged from 2 to 17 and expected heterozygosity from 0.089 to 0.800 with a mean value of 0.484. The population of Kalampaka had the lowest value of allelic richness (2.63) and gene diversity in comparison to the remaining populations. STRUCTURE analysis confirmed isolation of population Mariolata from the southern edge of the species range and genetic similarity among populations from the Pindos Mts. Additionally , the utility of new SSRs in 29 individuals from nine other Aesculus taxa was tested. Eleven markers gave polymorphic products for all tested species. For 24 individuals, a high-quality product was obtained for each marker. Results confirmed the utility of specific markers for future population genetics studies.
Content may be subject to copyright.
2021, vol. 85, 105–116
https://doi.org/10.12657/denbio.085.010
Łukasz Walas*, Grzegorz Iszkuło, Zoltan Barina, Monika Dering
Development of microsatellite markers
forhorse-chestnut (Aesculus hippocastanum),
theirpolymorphism in natural Greek populations,
and cross-amplication in related species
Received: 15 March 2021; Accepted: 22 April 2021
Abstract: New nuclear microsatellite markers (SSRs) were developed for Aesculus hippocastanum, a relict
tree species from the Balkan Peninsula. The development of microsatellites was done using the Illumina
MiSeq PE300 platform. Out of a set of 500 SSRs designed, a subset of 13 loci was tested using 290 individ-
uals from seven natural populations. Twelve species-specic loci were polymorphic. The number of alleles
per locus ranged from 2 to 17 and expected heterozygosity from 0.089 to 0.800 with a mean value of 0.484.
The population of Kalampaka had the lowest value of allelic richness (2.63) and gene diversity in compari-
son to the remaining populations. STRUCTURE analysis conrmed isolation of population Mariolata from
the southern edge of the species range and genetic similarity among populations from the Pindos Mts. Ad-
ditionally, the utility of new SSRs in 29 individuals from nine other Aesculus taxa was tested. Eleven markers
gave polymorphic products for all tested species. For 24 individuals, a high-quality product was obtained
for each marker. Results conrmed the utility of specic markers for future population genetics studies.
Keywords: Tertiary relict, endemic species, cross-amplication, polymorphism
Addresses: Ł. Walas, G. Iszkuło, M. Dering, Institute of Dendrology, Polish Academy of Sciences,
Parkowa 5, 62-035 Kórnik, Poland, e-mail: lukaswalas@man.poznan;
ŁW https://orcid.org/0000-0002-4060-9801
G. Iszkuło, Institute of Biological Sciences, University of Zielona Góra, Prof. Z. Szafrana 1,
65-516 Zielona Góra, Poland, e-mail: iszkulo@man.poznan.pl; https://orcid.org/0000-0003-2067-729X
Z. Barina, Department of Botany, Hungarian Natural History Museum, H-1431 Budapest, Pf. 137,
e-mail: barina.zoltan@nhmus.hu; https://orcid.org/0000-0003-3117-7186
M. Dering, Faculty of Forestry, Poznań University of Life Sciences, Wojska Polskiego 85,
60-637 Poznań, Poland, e-mail: mdering@man.poznan.pl; https://orcid.org/0000-0001-7017-5541
*Corresponding author
Introduction
Global climate changes are projected to affect
the whole biosphere, by disturbing the function-
ing of ecosystems and shifting species distributions
(Peñuelas et al., 2013; Pecl et al., 2017). Many en-
demic and relict taxa may not be able to withstand
these changes, and their extinction could cause a
signicant decline in biodiversity (Casazza et al.,
2014). A study shows that stable climatic conditions
support high endemism (Harrison & Noss, 2017), a
factor that may no longer be valid in some regions
soon. The Mediterranean hotspot of diversity is one
of the most important areas in terms of plant diver-
sity in Europe about 25,000 species occur in this
region and half of them are endemics (Myers et al.,
106 Łukasz Walas et al.
2000). Predictions for the future climate in the Med-
iterranean area show an increase in temperatures, re-
duction in rainfalls, and higher variance in seasonal
patterns of precipitation (Giorgi & Lionello, 2008),
which can pose a serious threat to many species, par-
ticularly endemics (Orsenigo et al., 2018). Consid-
ering the rapid pace of the current climate changes,
more studies are needed to support setting up strate-
gies for the efcient conservation of endemic trees to
prevent their loss and support their survival.
Horse-chestnut (Aesculus hippocastanum L.) is a
Tertiary relict and endemic species that occurs in the
mountainous regions of the Balkan Peninsula. This
is the only representative of the genus Aesculus in Eu-
rope. The species is well-known because of its orna-
mental value, which was the reason for its great pop-
ularity in urban ora across Europe and worldwide
(Lack, 2000). But despite that, natural populations
of this species are currently endangered because of
climate changes, human activities, as well as the
spread of diseases and pests (Thalmann et al., 2003;
Steele et al., 2010; Jagiełło et al., 2017; Walas et al.,
2018). The International Union for Conservation
of Nature (IUCN) recommends urgent genetic re-
search on natural stands of horse-chestnut (Allen &
Khela, 2017). Analysis of genetic diversity and struc-
ture can help to understand the interplay between
demographic processes and selection which occur
in the natural populations of this plant. Knowledge
about the level and the spatial patterns of genetic
diversity is crucial for planning an efcient in situ
conservation strategy. Genetic markers, which allow
describing population genetic structure with high
accuracy, are a vital help in prioritizing the conser-
vation actions. The level of genetic diversity informs
on how the species reacted to environmental factors
in the past and what we can expect in the future in
terms of its adaptive capacity. For that reason, data
on genetic diversity is a prerequisite for designing
a sound conservation strategy for horse-chestnut.
However, genetic information must be generated us-
ing reliable methods that deliver good quality data
to properly conclude on species genetic resources
conserved in natural stands.
Despite fast technological progress, microsatel-
lite markers (SSRs) remain one of the most impor-
tant molecular ecology tools because of their easy
allele detection, high polymorphism, and relatively
low costs (Vieira et al., 2016; Mora et al., 2017). For
small populations with low migration rates, as in the
case of most endemic species, even a small number
of microsatellites can provide signicant informa-
tion on microevolutionary processes, their rate, and
directions (Selkoe & Toonen, 2006). Due to the lack
of specic microsatellites for A. hippocastanum, the
markers developed for the closely related Japanese
horse-chestnut, A. turbinata Blume were used for the
assessment of the population genetic structure (Mi-
nami et al., 1998; Walas et al., 2019). Many SSRs
show a high rate of transferability between close taxa;
this has also been proven for several tree species and
genera (Vignes et al., 2006; Ravishankar et al., 2011;
Boratyński et al., 2014). However, cross-amplica-
tion can give low-quality products, which may cause
scoring errors. Additionally, reaction failures that
may indicate the existence of interspecies sequence
differences in the anking DNA regions, which are
targets for primers, may lead to a substantial null
allele frequency, and in consequence, affect estima-
tors of intra- and interpopulation diversity, leading
to biased inferences. A locus that is polymorphic in
one species may be monomorphic in another, or the
products may be non-specic (Sugai et al., 2016;
Godoy et al., 2019). Therefore, the development
of new, polymorphic markers specic to the A. hip-
pocastanum genome is necessary for accurate charac-
terisation of the population structure and patterns
of genetic diversity in remnant natural stands of the
species.
In our work, we aimed to characterise a set of new
polymorphic markers designed for A. hippocastanum
that could be used in diversity studies with conser-
vation perspectives. Three multiplex reactions for
12 novel SSRs were optimised and tested in a study
of the population genetic structure of natural pop-
ulations of the species from Greece. Additionally,
we used STRUCTURE analysis to check, whether
the size of the loci set might affect the detection of
the population structure. For this purpose, we used
species-specic and cross-amplied loci used in the
previous study (Walas et al., 2019) to enlarge the
marker set and evaluate its resolution power. Finally,
we aimed to deliver information about the potential
utility of these newly designed SSRs in other Aesculus
taxa.
Materials and methods
The development of novel microsatellites was
performed at AllGenetics (www.allgenetics.eu). One
sample was used to generate a library with the Nex-
tera XT DNA Library Preparation Kit (Illumina). The
library was enriched with fragments with microsatel-
lite motifs by hybridisation to four groups of oligore-
peats (AG, AC, ATCT, and ACG) and was sequenced
in the Illumina MiSeq PE300 platform (Macrogen
Inc.). The library produced 6,990,226 sequences.
Reads were processed in Geneious 10.2.3 (Biomat-
ters Ltd). Primer design was carried out in Primer 3
software (Koressaar & Remm, 2007; Untergasser et
al., 2012), implemented in Geneious 10.2.3. Finally,
500 primer pairs were developed during the proce-
dure (Table S1). These primer pairs are located at
Development of microsatellite markers forhorse-chestnut (Aesculus hippocastanum)... 107
the anking regions of the microsatellite motifs. For-
ty primer pairs were randomly chosen for the next
step of the procedure. These pairs were multiplexed
in the sets of three to ve markers, based on their
features, and tested in eight individuals. In the last
step, 10 primers were tested in three individuals, to
optimize the nal PCR reaction.
A total number of 13 primer pairs – 10 tested in
AllGenetics and three additional (AH_037, AH_051,
AH_054, AH_101, AH_129, AH_222, AH_257,
AH_269, AH_359, AH_375, AH_419, AH_447, and
AH_485) were organised into three sets according to
their properties and expected amplicon sizes (Tables
1 and 2). These markers were tested on 290 individ-
uals of A. hippocastanum originating from seven nat-
ural populations from Greece (Table 3) and used in
the previous study (Walas et al., 2019). DNA was
extracted from leaves according to the protocol de-
scribed by Dumolin et al. (1995). PCR reactions
were conducted in a volume of 10 μL, containing 1
× reaction buffer, 0.1 μg of BSA (Bovine Serum Al-
bumin), 1.5 mM MgCl2, 2 μM of dNTP mix, 0.5 U
of SilverTaq polymerase (Syngen, Poland), 0.05 μM
of each starter and 100 ng of DNA. Reactions were
conducted using the following protocol: initial dena-
turation at 95°C for 12 min, followed by 30 cycles of
denaturation at 95°C for 30 s, annealing at 57°C for
90 s, elongation at 72°C for 30 s; 8 cycles of 95°C
for 30 s, 53°C for 90 s, 72°C for 30s and nal elon-
gation at 68°C for 15 min. Products of amplication
were analysed using 3130 Genetic Analyser (Applied
Biosystems, Foster City, California, USA) with inter-
nal size standard GeneScan LIZ-500. Genotypes were
scored using GENEMAPPER v. 4.0 (Applied Biosys-
tems). Amplication and genotyping for eight loci
developed for A. turbinata (AT3D6, AT6D8, AT7D1,
AT5D2, AT6D11, AT6D2, AT7D8, and AT6D12)
were conducted during the previous study (Walas et
al., 2019).
New markers were also tested on 29 individu-
als representing nine taxa from the genus Aesculus
collected from the Adam Mickiewicz University Bo-
tanical Garden in Poznań and from Kórnik Arbore-
tum of the Institute of Dendrology PAS. The collected
individuals were as follows: A. ×carnea Hayne (3 in-
dividuals), A. chinensis Bunge (2 individuals), A. ava
Sol. (4 individuals), A. glabra Willd. (7 individuals),
A. ×hybrida DC. (1 individual), A. ×neglecta Lindl.
(4 individuals), A. parviora Walter (3 individuals),
A. pavia L. (3 individuals), and A. turbinata Blume (2
individuals).
Basic diversity estimates such as an average num-
ber of alleles (Na) and an effective number of alleles
(Ne) were estimated using GENEALEX 6.4 (Peakall
& Smouse, 2006). FSTAT v. 2.9.3 (Goudet, 2001)
was used to calculate allelic richness (Ar), while IN-
EST v. 2.0 (Chybicki, 2016) was applied for calcu-
lating genotyping error rate (b), observed (Ho) and
expected (He) heterozygosity, as well as inbreeding
coefcient (FIS). Wright’s xation index (FST) was
estimated in FREENA, with and without Excluding
Null Alleles (ENA) correction (Chapuis & Estoup,
2007). The latter software was also used to estimate
the frequency of the null alleles (Null). A test of
Table 2. Arrangement of 13 microsatellite markers into
multiplex reactions
Loci Multiplex Dye N
AH_051 I PET 14
AH_054 I VIC 7
AH_269 IFAM 2
AH_359 IFAM 7
AH_447 INED 17
AH_101 II NED 2
AH_129 II PET 5
AH_037 II FAM 6
AH_419 II VIC 6
AH_222 III FAM 3
AH_257 III NED 7
AH_375 III PET 6
AH_485 III VIC 7
N – number of alleles.
Table 1. Species-specic primers for 13 microsatellites tested in the Aesculus hippocastanum
Loci Forward Primer Reverse Primer Size range (bp) Motif
AH_037 ATTGAACGCAGCTCTCACCT GTGAACTAAGACCGTGCCGT 68–98 (CTCTC)5
AH_051 TCGGGATGTCGTTATTAGGC CCAGAAATTCCGCACTAAGC 280–312 (AT)7
AH_054 AGCCTCCTCTCGACGTACAA CGCCGAAGCTGTTGATTATT 138–162 (TCG)7
AH_101 CGCACTCGCAAGTTTGAGTA TGGTCATCACAGTGGCTGAT 202–205 (TAG)5
AH_129 CCTGGAAATCTCAACCAGGA CTTTGGAAGGAGGCATCTCA 104–114 (AT)5
AH_222 AGAACTGCTTCCAGACCACG CCTGCAAACAGAATCGGAAT 150–162 (TTC)5
AH_257 AATGGTTACTGACGTGGCCT CGCGTGTCTTAAACGAGTCA 80–98 (CT)5
AH_269 TTGTCACGAACATGTCCACC ATGGTTCACCTTGAACTGGG 216–232 (TG)5
AH_359 CTCCTCGACCAACAACACTG AGGACATTTCGAGGATGACG 92–128 (CTCCCT)5
AH_375 TTACAATCCTCCAACCCTGC ATATCACCCAGCACCTCCCT 233–257 (CAA)5
AH_419 GTTCTCCCTTGGCCTTGACT GGTAAAGCGAGGTACGCAAA 111–129 (CAG)8
AH_447 CGGGTAGTAGCACAGCAACA TTGGAACCAATTCAGAACCG 105–139 (TA)7
AH_485 CATGATCGGTTGAACAGTGC TGTCGAGCAAACACATCAGG 82–106 (GTG)6
108 Łukasz Walas et al.
Hardy-Weinberg Equilibrium (HWE) was performed
using the hw.test function in the package “pegas” and
visualised with the function levelplot from the package
“lattice” in the R environment (Sarkar, 2008; Paradis,
2010; R Core Team, 2013). Genotyping linkage dis-
equilibrium (LD) for each pair of loci was calculated
with GENEPOP on the Web application (Raymond,
1995; Rousset, 2008) using likelihood ratio statistics
and default Markov chain parameters. For HWE and
LD tests, the p-value was corrected using Bonferroni
correction. Analysis of Molecular Variance was con-
ducted in GENEALEX (Peakall & Smouse, 2006).
This software was also applied for the calculation of
the Codom-Genotypic Genetic Distance between all
tested species. Obtained pairwise distances were vis-
ualized with the function pcoa in the package “ape”
in the R environment (R Core Team, 2013; Paradis &
Schliep, 2019).
Individuals were divided into genetic clusters
using a non-spatial Bayesian clustering model im-
plemented in STRUCTURE 2.3.4 (Pritchard et al.,
2000). The procedure included 10 independent runs
with 105 of burn-in and 106 MCMC iterations with the
maximum number of clusters set to K=8, correlated
allele frequencies within populations assumed, and
mixed ancestry of individuals allowed. Three analy-
ses were performed: 1) for species-specic loci only
(a set of 12 loci), 2) for cross-amplied loci (from
A. turbinata; a set of 8 loci) and 3) for combined sets
of loci. To estimate the best-supported number of
clusters, Evanno’s delta K method implemented in
CLUMPAK (Kopelman et al., 2015) was used.
Reserve Selection analysis from DIVA-GIS soft-
ware (Hijmans et al., 2011) was used to indicate the
populations with the priority of conservation. This
method estimates the minimum number of geo-
graphical units (such as regions or populations) nec-
essary to conserve all genetic diversity of the targeted
species. We used alleles obtained for species-specic
loci with “Rarity” option, which takes into account
the frequency of the alleles. The results of the analy-
sis were visualized in QGIS 3.10.6. (QGIS Develop-
ment Team, 2012).
Results
Twelve of the tested loci were polymorphic, and
the number of alleles ranged between 2 and 17. Locus
AH_269 showed the same prole of microsatellite
peaks for all tested individuals and thus it was ex-
cluded from further analysis due to lack of polymor-
phism (Table 4, Table S2). The remaining loci showed
good interpretable and reproducible polymorphic pat-
terns without visible errors, such as stutter bands or
split peaks (Fig. S1). The efciency of amplication
in A. hippocastanum was high as we obtained 0–3.1%
of missing data, with the average value reaching only
0.9%. The highest missing values were observed for
loci AH_419 (3.1%) and AH_375 (2.8%). Additional-
ly, genotyping error (0.07%) was very low in compar-
ison with the typical range noted for microsatellites
(Wang, 2018) and was lower than the error rate in
cross-amplied markers, for which mistyping was
0.11% (Table S3). Wright’s xation index was slight-
ly lower for specic markers than for cross-amplied
loci (values of FST with ENA correction were 0.113
and 0.116, respectively). Expected heterozygosity
ranged between 0.089 (locus AH_222) and 0.800 (lo-
cus AH_447), with a mean value of 0.484. The aver-
age frequency of null alleles was 0.062, with values
>0.1 noted in loci AH_037 (0.111), AH_419 (0.113)
and AH_257 (0.118). These markers should be used
with caution, preferably with the correction methods
applied in FREENA or another software that accounts
for null alleles (Chapuis & Estoup, 2007). According
to the analysis performed for a set of 12 loci, some
pairs of loci were in signicant linkage disequilibri-
um (Table S4). Specically, loci AH_051 and AH_257
were associated with four other loci, AH_485 with
three loci, while AH_054 and AH_375 were associat-
ed with two loci. However, the number of pairs with
signicant linkage disequilibrium differed in each
population analysed separately. In two populations
(Kalampaka and Karitsa) all loci were independent.
Loci AH_257 and AH_375 did not conform to HWE
in ve populations and locus AH_447 did not con-
form in four populations (Fig. 1).
Table 3. Location of the populations of Aesculus hippocastanum used in this study
Population Voucher Latitude Longitude Altitude Region N
Ondria KOR 51217 40°20'N 21°05'E 1463 Pindos Mts 50
KOR 51218
Kalampaka No voucher 39°48'N 21°16'E 1371 Pindos Mts 23
Dasos Nanitsa KOR 51216 39°42'N 21°21'E 1029 Pindos Mts 93
Vaeni No voucher 39°12'N 21°42'E 1089 Pindos Mts 32
Mariolata KOR 51230 38°37'N 22°26'E 1239 Parnassus Massif 42
KOR 51219
Karitsa KOR 51280 39°48'N 22°45'E 705 Ossa Massif 24
Perivoli KOR 51226 39°58'N 21°11'E 915 Pindos Mts 26
N – number of individuals sampled
Development of microsatellite markers forhorse-chestnut (Aesculus hippocastanum)... 109
The average number of alleles in a population
(Na) ranged from 2.67 in Kalampaka to 4.92 in Ka-
ritsa I, whereas the average effective number of al-
leles (Ne) varied from 1.74 in Kalampaka to 2.26 in
Vaeni (Table 5). In the population of Kalampaka, loci
AH_222 and AH_375 were monomorphic. Average
allelic richness (AR) was 3.71 and was similar in
all populations except for Kalampaka, where it was
much lower (2.63). Genetic diversity (He) ranged
from 0.345 in Kalampaka to 0.495 in Perivoli, while
observed heterozygosity (Ho) varied from 0.326 in
Kalampaka to 0.394 in Ondria. The average number
of private alleles was 3.14. Totally, 22 private alleles
were detected which is much more than in the pre-
vious study with SSRs loci designed for A. turbinata
in which 13 private alleles were observed (Walas et
al., 2019). AMOVA showed that 13% of molecular
variance occurs among populations, 16% among in-
dividuals, and 71% within individuals (Table S5).
The new SSRs were successfully amplied in all
tested taxa and gave polymorphic and high-quality
products. For one individual of A. chinensis, we did
not obtain the products for AH_037, AH_101, and
AH_419 loci. Primer AH_051 did not give the prod-
uct in one individual of A. ava, one of A. glabra, one
of A. ×neglecta, and one of A. parviora. Additionally,
18 alleles not present in A. hippocastanum were ob-
served in other species (Table S2). Eight alleles were
detected in A. chinensis – of which two were present-
ed only in this species (one for AH_051 and one
for AH_054 markers). Seven alleles not detected in
A. hippocastanum were noted in A. ava, eight in A.
glabra (two alleles for AH_129 were unique for this
species), eight in A. ×hybrida (one allele of AH_222
was detected only in this taxa), eleven in A. ×neglec-
ta, two for A. parviora, one in A. turbinata and ve
in A. pavia. All alleles detected in A. ×carnea were
presented also in A. hippocastanum. Interestingly, lo-
cus AG_375 was monomorphic for all species except
for A. hippocastanum. Despite the results, the useful-
ness of these SSRs for initial species identication
should be veried with a greater number of individu-
als. Codom-Genotypic Genetic Distance between all
tested species showed results in accordance with the
current taxonomy (Fig. 2, Table S6).
Analysis of genetic structure based on 20 loci
(12 specic for A. hippocastanum and 8 specic for A.
turbinata) made with STRUCTURE (Pritchard et al.,
2000) dened seven genetic clusters, which clearly
Fig. 1. Results of the test of Hardy-Weinberg equilibrium
(HWE) according to loci and populations (p-values
were corrected using Bonferroni correction). Signi-
cant departures from HWE are indicated by grey color.
“M“ indicates that locus is monomorphic in a given
population
Population numbers: 1 – Ondria, 2 – Kalampaka, 3 – Dasos Nanit-
sa, 4 – Vaeni, 5 – Mariolata, 6 – Karitsa, 7 – Perivoli.
Table 4. Variability of newly designed SSRs markers specic for Aesculus hippocastanum
Loci Na Ne Null AR HOHEFIS FST FST ENA b
AH_037 4.14 2.11 0.111 4.204 0.329 0.570 0.424 0.233 0.202 0.01%
AH_051 6.43 2.82 0.034 8.211 0.573 0.733 0.218 0.203 0.200 0.04%
AH_054 4.71 2.39 0.004 5.172 0.627 0.638 0.018 0.141 0.139 0.59%
AH_101 2.00 1.45 0.075 2.140 0.190 0.325 0.415 0.179 0.172 0.01%
AH_129 2.57 1.25 0.034 2.377 0.148 0.154 0.035 0.140 0.137 0.01%
AH_222 2.00 1.10 0.008 1.937 0.086 0.089 0.032 0.012 0.011 0.01%
AH_257 4.71 2.74 0.118 5.333 0.429 0.681 0.370 0.046 0.056 0.01%
AH_359 4.43 2.54 0.053 4.501 0.516 0.648 0.204 0.107 0.101 0.01%
AH_375 3.57 1.42 0.083 4.054 0.202 0.332 0.390 0.042 0.064 0.03%
AH_419 3.29 1.65 0.113 3.782 0.224 0.435 0.484 0.044 0.054 0.02%
AH_447 7.43 3.74 0.068 9.357 0.668 0.800 0.165 0.137 0.135 0.02%
AH_485 3.57 1.64 0.048 4.857 0.363 0.405 0.103 0.106 0.087 0.02%
Average 4.08 2.07 0.062 4.660 0.363 0.484 0.238 0.116 0.113 0.07%
110 Łukasz Walas et al.
Table 5. Parameters of genetic diversity of the studied populations of Aesculus hippocastanum for the loci designed in
this study
Population N Na Ne Null AR APHOHEFIS b
Ondria 50 4.42 2.22 0.068 3.90 3 0.394 0.472 0.030 0.16%
Kalampaka 23 2.67 1.74 0.036 2.63 0 0.326 0.345 0.025 0.09%
Dasos Nanitsa 93 4.92 1.92 0.078 3.88 6 0.346 0.427 0.024 0.23%
Vaeni 32 4.25 2.26 0.074 3.96 2 0.342 0.433 0.124 0.08%
Mariolata 42 4.42 2.08 0.050 3.87 3 0.371 0.411 0.042 0.05%
Karitsa 24 3.92 2.07 0.042 3.83 40.392 0.431 0.054 0.42%
Perivoli 26 3.92 2.20 0.086 3.89 40.385 0.495 0.065 0.48%
Average 4.07 2.07 0.062 3.71 3.14 0.370 0.430 0.050 0.22%
N – number of individuals, Na – the average number of alleles, Ne – effective number of alleles, Null – frequency of null alleles, AR – allel-
ic richness, AP – number of private alleles, Ho – observed heterozygosity, HE – expected heterozygosity, FIS – inbreeding coefcient,
b – genotyping error rate.
Fig. 2. Results of Principal Coordinate Analysis according to Codom-Genotypic Genetic Distance for tested taxa from
genus Aesculus
Fig 3. Individuals grouped by population and genetic clustering as a result of the STRUCTURE analysis
A – clusters for all loci for K=7; B – the best K for all loci; C – clusters for loci developed for Aesculus turbinata for K=3; D – the best K
for loci developed for A. turbinata; E – clusters for loci developed for A.hippocastanum for K=5; F – the best K for loci developed for
A. hippocastanum.
Development of microsatellite markers forhorse-chestnut (Aesculus hippocastanum)... 111
matched to the populations investigated (Fig. 3).
Analyses for both sets of loci separately showed K=3
and K=5 as the most probable for cross-amplied
loci and species-specic loci, respectively. However,
based on the species-specic loci set, K=3 and K=5
were almost equally probable. UPGMA clustering
(Fig. 4) showed isolation of Mariolata and similari-
ty between populations from the Pindos Mountains,
which was also conrmed by values of the pairwise
FST (Table S7). Reserve selection analysis pointed at
Dasos Nanitsa, Karitsa and Mariolata as sites that
should be protected with high priority (Fig. 5).
Discussion
The new markers, specic for A. hippocastanum,
showed lower allelic richness than cross-amplied
loci from A. turbinata (Table 4, Table S3). The ex-
pected and observed heterozygosity estimated with
specic markers were also lower. This result can be
Fig. 4. UPGMA tree based on Nei’s genetic distances for
natural populations of Aesculushippocastanum
Fig. 5. Location of the analyzed populations of Aesculus hippocastanum (red dots). The size of the dots indicates the conser-
vation priority according to the allelic richness values. Dashed areas show Sites of Community Importance (SCI) for
the protection of ora and fauna in Greece (source: geoadata.gov.gr)
112 Łukasz Walas et al.
related to the length of the microsatellite motifs be-
cause markers with dinucleotide motifs usually have
a higher mutation rate and a higher number of alleles
than markers with longer motifs (Zurn et al., 2020).
All cross-amplied loci have a dinucleotide motif,
whereas as many as eight new markers have a longer
motif. However, species-specic loci with the high-
est values of allelic richness (AH_051, AH_257, and
AH_447) have a dinucleotide motif (Table 1 and 4).
Consequently, because the heterozygosity is partly a
function of the number of alleles (Zurn et al., 2020),
we found a lower genetic diversity value for specic
loci. However, FIS and the frequency of null alleles
were at a similar level (Tables 4 and S3).
Although a small number of individuals per taxon
was used in our study, a simple test of Codom-Gen-
otypic Genetic Distance between species performed
in GENEALEX (Peakall & Smouse, 2006) showed re-
sults consistent with the current systematic position
of individual taxa (Fig. 2, Table S6). Species from
section Pavia (A. glabra, A. ava, A. ×neglecta, and A.
pavia) and A. ×hybrida, which is a hybrid between
A. ava and A. pavia, were included in one group.
The second group was formed by representatives of
section Aesculus (A. hippocastanum and A. turbinata),
section Macrothyrsus (A. parviora), and A. ×carnea,
which is a hybrid between A. hippocastanum and A.
pavia. A close relationship between sections Aesculus
and Macrothyrsus was previously reported in a phy-
logenetic study (Harris & Xiang, 2009). Aesculus chin-
ensis, a representative of section Calothyrsus, was dif-
ferent from the other taxa which actually reects its
current systematic position (Harris & Xiang, 2009)
(Fig. 2). In view of this, our results indicate that pol-
ymorphic microsatellites developed for one Aesculus
species can be successfully used for other species
from this genus in a situation of lack of the spe-
cies-specic markers (Bačovský et al., 2017; Walas et
al., 2019). In addition, we provide a list of nearly 500
pairs of primers that can be potentially applied in A.
hippocastanum and other taxa (Table S1).
Analysis conducted in STRUCTURE, based on
20 loci, dened seven clusters of horse-chestnut,
whereas for cross-amplied loci and species-specif-
ic loci it showed three and ve groups, respective-
ly. It is commonly known that an increased number
of used markers may help in better recognition of
the genetic structure and more correctly describes
diversity indices. Inspection of barplots generated
by STRUCTURE with species-specic and cross-am-
plied loci revealed some differences in population
structure, especially in populations from the Pindos
Mts. Based on species-specic SSRs, Kalampaka
stand was dened as more distinct from the remain-
ing populations from the northern Pindos Mts., i.e.
Dasos Nanitsa and Ondria. The distinctiveness of
Kalampaka may reect the particular demographic
history of this genetically depleted population (Table
5). Both marker sets are convergent in this aspect
and show low allelic and gene diversity of Kalam-
paka in contrast to other populations (Walas et al.,
2019) that may suggest drift-induced differentiation
detected by STRUCTURE analysis. Despite that, an-
other population from this area, Perivoli, was con-
stantly different from the neighbouring stands and
showed genetic afnities to the marginally located
population from Karitsa, irrespectively of the mark-
ers set (Fig. 3). Similarly, population Mariolata from
the southern edge of the species range was always
located in the distinct cluster using either specic or
cross-amplied markers, which conrms that both
marker sets are largely convergent in the detection
of the genetic structure. The major discrepancy re-
lates to the rate of the detected admixture which
was higher for specic loci than for non-specic loci.
Accordingly, Ondria was the most admixed popula-
tion because only 30% of the individuals from that
stand possessed the genome with 95% of the mem-
bership to a single cluster. On the contrary, most of
the individuals from Perivoli and Karitsa reached
this high level of membership (65.38% and 70.83%,
respectively). The most probable explanation for
this situation is the presence of a lower allelic diver-
sity discovered with the set of the species-specic
loci that in turn, resulted in a lower resolution and
individual genealogy of each locus that may have in-
terfered with the results. The genotypic error rate
that can overestimate the level of the admixture was
too low in our case to have such an impact (Reeves
et al., 2016).
UPGMA clustering (Fig. 4) based on Nei genetic
distances revealed the pattern which is partly congru-
ent with the Bayesian inferences and demonstrated
the distinctiveness of the marginal population Ma-
riolata. It conrmed also the similarity between On-
dria and Kalampaka populations. Additionally, Vae-
ni seemed to be genetically closer to Dasos Nanitsa
than to Karitsa or Perivoli, which was conrmed also
by the value of the pairwise FST (Table S7) but disa-
grees with the STRUCTURE results. The main con-
clusion that can be drawn from the cluster analysis is
the general genetic similarity of the populations from
the Pindos Mts. except for Perivoli that was also de-
tected by the Bayesian approach (Fig. 3). In pairwise
comparisons, the highest genetic differentiation was
noted between Kalampaka and Karitsa that originate
from distinct mountain ranges (Table S7). Generally,
the pairwise differentiation values were high which
agrees with the Bayesian inferences about signicant
genetic structure in A. hippocastanum.
The genetic separation of the populations from
the Pindos Mts., the Ossa Massif, and the Parnassus
Massif as revealed by different methods used in this
study, reects spatial isolation of different mountain
Development of microsatellite markers forhorse-chestnut (Aesculus hippocastanum)... 113
ranges that currently provide suitable habitats for
horse-chestnut. This pattern of differentiation is
commonly reported for species inhabiting mountains
(Noguerales et al., 2016). Horse-chestnut occurs in
a topographically complex landscape, frequently in
inaccessible mountainous habitats, which reduces
the opportunity of gene ow at the landscape level.
This particularly refers to Perivoli that despite its lo-
cation in Pindos Mts. lacks signicant similarity to
other stands from this mountain range which likely
demonstrates limited gene ow that was also sug-
gested in the previous estimation of migration rate
(Walas et al., 2019). Additionally, the reproduction
mode and species biology limit genetic connectivity
among populations separated geographically.
An efcient allocation of the conservation efforts
in the case of the endemic and relic species requires
broad knowledge not only about their ecology but
also their genetic structure, its attributes, and fac-
tors. Based on such collection of the data, efcient
Conservation Units may be set to protect not only
the biological objects (populations and species) but
also to support the maintenance of the evolutionary
processes (Médail & Baumel, 2018). Microsatellite
markers can be successfully used in conservation
prioritization (Mattioni, 2017; Médail & Baumel,
2018). The reserve selection analysis, conducted in
DIVA-GIS software (Hijmans et al., 2001), recog-
nised Dasos Nanitsa and Karitsa as populations with
the highest priority for conservation (Fig. 5). Unfor-
tunately, Dasos Nanitsa, as well as other populations
from the northern Pindos, are not recognised as the
Sites of Community Importance (SCI) designed for
the protection of ora and fauna in Greece. It is even
more worrying as the horse-chestnut has not been
included in the Red List of Greece (Phitos et al.,
2009). Thus, a large part of the gene pool of this spe-
cies may be lost as a result of a lack of formal protec-
tion. Populations from the edge of the species range,
like Mariolata and Karitsa, which also harbour a high
genetic diversity can be endangered with extinction
because of the projected climatic changes (Walas et
al., 2019). Their peripheral geographical location
that implies isolation and ecological marginality as
revealed in ecological niche modelling, may induce
population reduction and decline (Walas et al., 2019;
Schueler et al., 2014). Presented results are focused
only on the Greek range while getting the deep in-
sight into the species genetic resources would re-
quire the inclusion of populations from Albania and
North Macedonia into the investigation.
Microsatellite analysis is easy to perform and
is highly informative; therefore it is a widely used
tool in population and conservation genetics studies
(Madesis et al., 2013; Vieira et al., 2016). Although
many new methods and approaches have appeared in
molecular biology due to technological advancement,
SSRs remain useful, and even comparable with SNPs
(Emanuelli et al., 2013; Filippi et al., 2015; Hodel
et al., 2017; Zurn et al., 2020). If we think about a
genetic tool that is easy and repeatedly used in rou-
tine analysis aiming to support conservationists and
practitioners in their decisions, microsatellite mark-
ers full these requirements in terms of the techno-
logical and biological aspects. Integration of genetic
methods into conservation management, though it
is still not a routine procedure (Taylor et al., 2017),
becomes a more and more important approach that
may facilitate the development of an effective strat-
egy for the conservation of living organisms (McMa-
hon et al., 2014).
Acknowledgements
This study was nanced by the Polish National
Science Centre (2017/27/N/NZ8/02781) and partly
the statutory works of the Institute of Dendrology,
Polish Academy of Sciences. We thank M. Łuczak for
laboratory support.
References
Allen DJ & Khela S (2017) Aesculus hippocasta-
num (errata version published in 2018). The
IUCN Red List of Threatened Species 2017: e.
T202914A122961065.
Bačovský V, Vyhnánek T, Hanáček P, Mertelík J &
Šafránková I (2017) Genetic diversity of chest-
nut tree in relation to susceptibility to leaf miner
(Cameraria ohridella Deschka & Dimič). Trees 31:
753–763. doi:10.1007/s00468-016-1506-2.
Boratyński A, Wachowiak W, Dering M, Boratyńska
K, Sękiewicz K, Sobierajska K, Jasińska AK, Klim-
ko M, Montserrat JM, Romo A, Ok T & Didukh
Y (2014) The biogeography and genetic relation-
ships of Juniperus oxycedrus and related taxa from
the Mediterranean and Macaronesian regions. Bo-
tanical Journal of the Linnean Society 174: 637–
653. doi:10.1111/boj.12147.
Casazza G, Giordani P, Benesperi R, Foggi B, Viciani
D, Filigheddu R, Farris E, Bagella S, Pisanu S &
Mariotti MG (2014) Climate change hastens the
urgency of conservation for range-restricted plant
species in the central-northern Mediterranean
region. Biological Conservation 179: 129–138.
doi:10.1016/j.biocon.2014.09.015.
Chapuis M-P & Estoup A (2007) Microsatellite null
alleles and estimation of population differentia-
tion. Molecular Biology and Evolution 24: 621–
631. doi:10.1093/molbev/msl191.
Chybicki I (2016) INEST 2.0 [Computer Software].
https://www.ukw.edu.pl/pracownicy/strona/igor_
chybicki/software_ukw/.
114 Łukasz Walas et al.
Dumolin S, Demesure B & Petit RJ (1995) Inher-
itance of chloroplast and mitochondrial genomes
in pedunculate oak investigated with an efcient
PCR method. Theoretical and Applied Genetics
91: 1253–1256. doi:10.1007/BF00220937.
Emanuelli F, Lorenzi S, Grzeskowiak L, Catalano V,
Stefanini M, Troggio M, Myles S, Martinez-Zap-
ater JM, Zyprian E, Moreira FM & Grando MS
(2013) Genetic diversity and population structure
assessed by SSR and SNP markers in a large ger-
mplasm collection of grape. BMC Plant Biology
13: 39. doi:10.1186/1471-2229-13-39.
Filippi CV, Aguirre N, Rivas JG, Zubrzycki J, Pueb-
la A, Cordes D, Moreno MV, Fusari CM, Alvarez
D, Heinz RA, Hopp HE, Paniego NB & Lia VV
(2015) Population structure and genetic diver-
sity characterization of a sunower association
mapping population using SSR and SNP mark-
ers. BMC Plant Biology 15: 1–12. doi:10.1186/
s12870-014-0360-x.
Giorgi F & Lionello P (2008) Climate change pro-
jections for the Mediterranean region. Global
and Planetary Change 63: 90–104. doi:10.1016/j.
gloplacha.2007.09.005.
Godoy FMR, Paggi GM & Palma-Silva C (2019)
Cross-amplication of nuclear microsatellite
markers in Aechmea distichantha Lem. (Brome-
liaceae). Brazilian Journal of Botany 42: 353–359.
doi:10.1007/s40415-019-00528-z.
Goudet J (2001) FSTAT, a program to estimate and
test gene diversity and xation indices (version
2.9. 3). http://www2. unil. ch/popgen/softwares/
fstat. htm.
Harris AJ & Xiang QY (2009) Estimating ancestral
distributions of lineages with uncertain sister
groups: a statistical approach to Dispersal–Vicari-
ance Analysis and a case using Aesculus L. (Sapin-
daceae) including fossils. Journal of Systematics
and Evolution 47: 349–368. doi:10.1111/j.1759-
6831.2009.00044.x.
Harrison S & Noss R (2017) Endemism hotspots are
linked to stable climatic refugia. Annals of Botany
119: 207–214. doi:10.1093/aob/mcw248.
Hijmans RJ, Guarino L, Cruz M & Rojas E (2001)
Computer tools for spatial analysis of plant ge-
netic resources data: 1. DIVA-GIS. Plant Genetic
Resources Newsletter 127: 15–19.
Hodel RGJ, Chen S, Payton AC, McDaniel SF, Soltis
P & Soltis DE (2017) Adding loci improves phy-
logeographic resolution in red mangroves despite
increased missing data: comparing microsatellites
and RAD-Seq and investigating loci ltering. Sci-
entic Reports 7: e17598. doi:10.1038/s41598-
017-16810-7.
Jagiełło R, Baraniak E, Karolewski P, Łakomy P,
Behnke-Borowczyk J, Walczak U & Giertych MJ
(2017) Ecophysiological aspects of the interac-
tion between Cameraria ohridella and Guignardia
aesculi on Aesculus hippocastanum. Dendrobiology
78: 146–156. doi:10.12657/denbio.078.014.
Kopelman NM, Mayzel J, Jakobsson M, Rosenberg
NA & Mayrose I (2015) Clumpak: a program
for identifying clustering modes and packag-
ing population structure inferences across K.
Molecular Ecology Resources 15: 1179–1191.
doi:10.1111/1755-0998.12387.
Koressaar T & Remm M (2007) Enhancements and
modications of primer design program Primer3.
Bioinformatics 23: 1289–1291. doi:10.1093/bio-
informatics/btm091.
Lack HW (2000) Lilac and horse-chestnut: discovery
and rediscovery. Curtis’s Botanical Magazine 17:
109–141. doi:10.1111/1467-8748.00255.
Madesis P, Ganopoulos I & Tsaftaris A (2013) Mi-
crosatellites: evolution and contribution: Micro-
satellites (ed. by SK Kantartzi) Humana Press,
Totowa, NJ, pp. 1–13. doi:10.1007/978-1-62703-
389-3_1.
Mattioni C, Martin MA, Chiocchini F, Cherubini M,
Gaudet M, Pollegioni P, Velichkov I, Jarman R,
Chambers FM, Paule L, Damian VL, Crainic GC
& Villani F (2017) Landscape genetics structure
of European sweet chestnut (Castanea sativa Mill):
indications for conservation priorities. Tree Ge-
netics & Genomes 13: e39. doi:10.1007/s11295-
017-1123-2.
McMahon BJ, Teeling EC & Höglund J (2014) How
and why should we implement genomics into
conservation? Evolutionary Applications 2014 7:
999–1007. doi:10.1111/eva.12193.
Médail F & Baumel A (2018) Using phylogeography
to dene conservation priorities: The case of nar-
row endemic plants in the Mediterranean Basin
hotspot. Biological Conservation 224: 258–266.
doi:10.1016/j.biocon.2018.05.028.
Minami E, Isagi Y, Kaneko Y & Kawaguchi (1998)
Polymorphic microsatellite markers in Japanese
horse chestnut Aesculus turbinata Blume. Molecu-
lar Ecology 7: 1616–1617.
Mora F, Arriagada O, Ballesta P & Ruiz E (2017)
Genetic diversity and population structure of a
drought-tolerant species of Eucalyptus, using mi-
crosatellite markers. Journal of Plant Biochemis-
try and Biotechnology 26: 274–281. doi:10.1007/
s13562-016-0389-z.
Myers N, Mittermeier RA, Mittermeier CG, Da Fon-
seca GA & Kent J (2000) Biodiversity hotspots
for conservation priorities. Nature 403: 853–858.
doi:10.1038/35002501.
Noguerales V, Cordero PJ & Ortego J (2016) Hierar-
chical genetic structure shaped by topography in a
narrow-endemic montane grasshopper. BMC Evo-
lutionary Biology 16: 96. doi:10.1186/s12862-
016-0663-7.
Development of microsatellite markers forhorse-chestnut (Aesculus hippocastanum)... 115
Orsenigo S, Montagnani C, Fenu G, Gargano D, Peru-
zzi L, Abeli T, Alessandrini A, Bacchetta G, Barto-
lucci F, Bovio M, Brullo C, Brullo S, Carta A, Cas-
tello M, Cogoni D, Conti F, Domina G, Foggi B,
Gennai M, Gigante D, Iberite M, Lasen C, Magrini
S, Perrino EV, Prosser F, Santangelo A, Selvaggi
A, Stinca A, Vagge I, Villani M, Wagensommer
RP, Wilhalm T, Tartaglini N, Duprè E, Blasi C &
Rossi G (2018) Red Listing plants under full na-
tional responsibility: Extinction risk and threats
in the vascular ora endemic to Italy. Biological
Conservation 224: 213–222. doi:10.1016/j.bio-
con.2018.05.030.
Paradis E (2010) pegas: an R package for population
genetics with an integrated–modular approach.
Bioinformatics 26: 419–420. doi:10.1093/bioin-
formatics/btp696.
Paradis E & Schliep K (2019) ape 5.0: an environ-
ment for modern phylogenetics and evolutionary
analyses in R. Bioinformatics 35: 526–528.
Peakall ROD, Smouse PE (2006) GENALEX 6:
genetic analysis in Excel. Population genetic
software for teaching and research. Molecular
Ecology Resources 6: 288–295. doi:10.1111/
j.1471-8286.2005.01155.x.
Pecl GT, Araújo MB, Bell JD, Blanchard J, Bonebrake
TC, Chen I-C, Clark TD, Colwell RK, Danielsen F,
Evengård B, Falconi L, Ferrier S, Frusher S, Garcia
RA, Grifs RB, Hobday AJ, Janion-Scheepers C,
Jarzyna MA, Jennings S, Lenoir J, Linnetved HI,
Martin VY, McCormack PC, McDonald J, Mitchell
NJ, Mustonen T, Pandol JM, Pettorelli N, Popo-
va E, Robinson SA, Scheffers BR, Shaw JD, Sorte
CJB, Strugnell JM, Sunday JM, Tuanmu M-N,
Vergés A, Villanueva C, Wernberg T, Wapstra E
& Williams SE (2017) Biodiversity redistribution
under climate change: Impacts on ecosystems and
human well-being. Science 355. doi:10.1126/sci-
ence.aai9214.
Peñuelas J, Sardans J, Estiarte M, Ogaya R, Carnic-
er J, Coll M, Barbeta A, Rivas-Ubach A, Llusià J,
Garbulsky M (2013) Evidence of current impact
of climate change on life: a walk from genes to the
biosphere. Global Change Biology 19: 2303–2338.
doi:10.1111/gcb.12143.
Phitos D, Constantinidis T & Kamari G (2009) The
red data book of rare and threatened plants of
Greece. Hellenic Botanical Society, Patras.
Pritchard JK, Stephens M & Donnelly P (2000) Infer-
ence of population structure using multilocus gen-
otype data. Genetics 155: 945–959. doi:10.1093/
genetics/155.2.945.
QGIS Development Team (2012) QGIS Geograph-
ic Information System. Open Source Geospatial
Foundation Project. Available at: http://qgis.os-
geo.org.
R Core Team (2013) R: A language and environment
for statistical computing. R Foundation for Sta-
tistical Computing, Vienna, Austria. Available at:
http://www.R-project.org/.
Ravishankar KV, Mani BH-R, Anand L & Dinesh MR
(2011) Development of new microsatellite mark-
ers from Mango (Mangifera indica) and cross-spe-
cies amplication. American Journal of Botany 98:
e96–99. doi:10.3732/ajb.1000263.
Raymond M (1995) GENEPOP (version 1.2): popu-
lation genetics software for exact tests and ecu-
menicism. Journal of Heredity 86: 248–249.
Reeves PA, Bowker CL, Fettig CE, Tembrock LR &
Richards CM (2016) Effect of error and missing
data on population structure inference using mi-
crosatellite data. bioRxiv doi:10.1101/080630.
Rousset F (2008) GENEPOP’007: a complete
re-implementation of the GENEPOP software
for Windows and Linux. Molecular Ecology
Resources 8: 103–106. doi:10.1111/j.1471-
8286.2007.01931.x.
Sarkar D (2008) Lattice: Multivariate data visuali-
zation with R. Springer, New York. http://lmd-
vr.r-forge.r-project.org.
Schueler S, Falk W, Koskela J, Lefèvre F, Bozzano M,
Hubert J, Kraigher H, Longauer R & Olrik, DC
(2014) Vulnerability of dynamic genetic conser-
vation units of forest trees in Europe to climate
change. Global Change Biology 20: 1498–1511.
doi:10.1111/gcb.12476.
Selkoe KA & Toonen RJ (2006) Microsatellites for
ecologists: a practical guide to using and evaluat-
ing microsatellite markers. Ecology Letters 9: 615–
629. doi:10.1111/j.1461-0248.2006.00889.x.
Steele H, Laue BE, MacAskill GA, Hendry SJ &
Green S (2010) Analysis of the natural infection
of European horse chestnut (Aesculus hippocast-
anum) by Pseudomonas syringae pv. aesculi. Plant
Pathology 59: 1005–1013. doi:10.1111/j.1365-
3059.2010.02354.x.
Sugai K, Watanabe K, Kato H & Sugawara T (2016)
Development of SSR markers for Psychotria hom-
alosperma (Rubiaceae) and cross-amplication in
four other species. Applications in Plant Sciences
4: 1500133. doi:10.3732/apps.1500133.
Taylor HR, Dussex N & vanHeezik Y (2017) Bridg-
ing the conservation genetics gap by identifying
barriers to implementation for conservation prac-
titioners. Global Ecology and Conservation 10:
231–242. doi:10.1016/j.gecco.2017.04.001.
Thalmann C, Freise J, Heitland W & Bacher S (2003)
Effects of defoliation by horse chestnut leafminer
(Cameraria ohridella) on reproduction in Aesculus
hippocastanum. Trees 17: 383–388. doi:10.1007/
s00468-003-0249-z.
Untergasser A, Cutcutache I, Koressaar T, Ye J, Fair-
cloth BC, Remm M & Rozen SG (2012) Prim-
116 Łukasz Walas et al.
er3–new capabilities and interfaces. Nucleic Ac-
ids Research 40: e115. doi:10.1093/nar/gks596.
Vieira MLC, Santini L, Diniz AL & Munhoz C de F
(2016) Microsatellite markers: what they mean
and why they are so useful. Genetics and Molecu-
lar Biology 39: 312–328. doi:10.1590/1678-4685-
GMB-2016-0027.
Vignes H, Hossaert‐Mckey M, Beaune D, Fevre D,
Anstett M-C, Borges RM, Kjellberg F & Chevallier
MH (2006) Development and characterization of
microsatellite markers for a monoecious Ficus spe-
cies, Ficus insipida, and cross-species amplication
among different sections of Ficus. Molecular Ecol-
ogy Resources 6: 792–795. doi:10.1111/j.1471-
8286.2006.01347.x.
Walas Ł, Dering M, Ganatsas P, Pietras M, Pers-Kam-
czyc E & Iszkuło G (2018) The present status and
potential distribution of relict populations of Aes-
culus hippocastanum L. in Greece and the diverse in-
festation by Cameraria ohridella Deschka & Dimić.
Plant Biosystems 152: 1048–1058. doi:10.1080/1
1263504.2017.1415991.
Walas Ł, Ganatsas P, Iszkuło G, Thomas PA & Dering
M (2019) Spatial genetic structure and diversity
of natural populations of Aesculus hippocastanum L.
in Greece. PloS One 14: e0226225. doi:10.1371/
journal.pone.0226225.
Wang J (2018) Estimating genotyping errors from
genotype and reconstructed pedigree data.
Methods in Ecology and Evolution 9: 109–120.
doi:10.1111/2041-210X.12859.
Zurn JD, Nyberg A, Montanari S, Postman J, Neale
D & Bassil N (2020) A new SSR ngerprinting
set and its comparison to existing SSR-and SNP-
based genotyping platforms to manage Pyrus ger-
mplasm resources. Tree Genetics & Genomes 16:
1–10. doi:10.1007/s11295-020-01467-7.
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Horse-chestnut (Aesculus hippocastanum L.) is an endemic and relict species from the Mediterranean biodiversity hotspot and a popular ornamental tree. Knowledge about the evolutionary history of this species remains scarce. Here, we ask what historical and ecological factors shaped the pattern of genetic diversity and differentiation of this species. We genotyped 717 individuals from nine natural populations using microsatellite markers. The influence of distance, topography and habitat variables on spatial genetic structure was tested within the approaches of isolation-by-distance and isolation-by-ecology. Species niche modeling was used to project the species theoretical range through time and space. The species showed high genetic diversity and moderate differentiation for which topography , progressive range contraction through the species' history and long-term persistence in stable climatic refugia are likely responsible. A strong geographic component was revealed among five genetic clusters that are connected with very limited gene flow. The environmental variables were a significant factor in the spatial genetic structure. Modeling results indicated that future reduction of the species range may affect its survival. The possible impact of climate changes and high need of in situ conservation are discussed.
Article
Full-text available
Aesculus hippocastanum is a well-known species, which is popular because of its ornamental value. However, data on the demographic structure and potential distribution of A. hippocastanum are limited. The invasion of Cameraria ohridella into Europe has harmed those trees growing in artificial sites, but the presence of this insect in natural stands has been little studied. Here we aimed to investigate the demographic structure infestation level of natural populations of horse-chestnut. Additionally, Maxent modelling was used to predict the potential range of A. hippocastanum, based on the localities available in the literature. Field data analysis indicated that natural populations of A. hippocastanum are mostly found nearby mountain streams. The populations showed a diverse height structure and large numbers of seedlings, which indicate high population dynamics. The level of infestation by C. ohridella varied greatly and correlated with altitude. Secondary infestation might explain this infestation variability in some natural populations. Other hypotheses, such as environmental resistance factors or different genetic variability, are also discussed. By spatial distribution modelling, we found that the precipitation of the coldest and warmest quarters, as well as altitude, are important factors influencing the potential distribution of A. hippocastanum.
Article
Full-text available
The widespread adoption of RAD-Seq data in phylogeography means genealogical relationships previously evaluated using relatively few genetic markers can now be addressed with thousands of loci. One challenge, however, is that RAD-Seq generates complete genotypes for only a small subset of loci or individuals. Simulations indicate that loci with missing data can produce biased estimates of key population genetic parameters, although the influence of such biases in empirical studies is not well understood. Here we compare microsatellite data (8 loci) and RAD-Seq data (six datasets ranging from 239 to 25,198 loci) from red mangroves (Rhizophora mangle) in Florida to evaluate how different levels of data filtering influence phylogeographic inferences. For all datasets, we calculated population genetic statistics and evaluated population structure, and for RAD-Seq datasets, we additionally examined population structure using coalescence. We found higher F ST using microsatellites, but that RAD-Seq-based estimates approached those based on microsatellites as more loci with more missing data were included. Analyses of RAD-Seq datasets resolved the classic Gulf-Atlantic coastal phylogeographic break, which was not significant in the microsatellite analyses. Applying multiple levels of filtering to RAD-Seq datasets can provide a more complete picture of potential biases in the data and elucidate subtle phylogeographic patterns.
Article
Full-text available
Popular in the urban environment is the horse chestnut, Aesculus hippocastanum which is suffering mainly due to the feeding of the horse chestnut leaf miner (Cameraria ohridella). The harmfulness of this pest is well recognized. Not much attention was put in discovering the interaction of this insect with the fungal pathogen Guignardia aesuli, the agent of leaf blotch. Host plant mediation in this particular insect-plant pathogen interaction is crucial for understanding the complexity of the horse chestnut’s current and future situation. Recognising the response of the host plant for separated and simultaneous colonisation by insect and fungus was the aim of this study. Leaf damage dynamics and phenolic compounds content (total soluble phenolic compounds – TPh, and condensed tannins – CT), and stem volume increment (SVI) of the horse chestnut saplings was considered and their relationship identified. The main hypothesis was that insect feeding and fungal infection when separated elicit a similar pattern in defence response of the host but this defence response is different when they both coexist on the same plant. Basing on crown projection area photographs sequence, foliage damage dynamics was assessed (Richard’s growth model) and protocol developed. Measurements of stem volume were performed sequentially to indicate potential growth response. Through this study, it was identified that the content of phenolic compounds in leaves was higher when both pests colonized saplings in comparison with those where saplings were infested by one biotic factor. It is also documented that foliage damage dynamics was higher when only the fungal pathogen attacked plants than when it was infected by both pests. A trade-off was identified between growth and secondary metabolism. Leaf damage affected stem volume increment only in the late summer, when a high level of defoliation was observed. Simultaneous infestation by fungal and insect agents made unfavourable conditions rather for the former. How this interaction affects the latter is not covered by our results and still remains undiscovered.
Working Paper
Full-text available
Missing data and genotyping errors are common in microsatellite data sets. We used simulated data to quantify the effect of these data aberrations on the accuracy of population structure inference. Data sets with complex, randomly-generated, population histories were simulated under the coalescent. Models describing the characteristic patterns of missing data and genotyping error in real microsatellite data sets were used to modify the simulated data sets. Accuracy of ordination, tree-based, and model-based methods of inference was evaluated before and after data set modifications. The ability to recover correct population clusters decreased as missing data increased. The rate of decrease was similar among analytical procedures, thus no single analytical approach was preferable. For every 1% of a data matrix that contained missing genotypes, 2–4% fewer correct clusters were found. For every 1% of a matrix that contained erroneous genotypes, 1–2% fewer correct clusters were found using ordination and tree-based methods. Model-based procedures that minimize the deviation from Hardy-Weinberg equilibrium in order to assign individuals to clusters performed better as genotyping error increased. We attribute this surprising result to the inbreeding-like nature of microsatellite genotyping error, wherein heterozygous genotypes are mischaracterized as homozygous. We show that genotyping error elevates estimates of the level of genetic admixture. Overall, missing data negatively impact population structure inference more than typical genotyping errors.
Article
Full-text available
Sweet chestnut is a tree of great economic (fruit and wood production), ecological, and cultural importance in Europe. A large-scale landscape genetic analysis of natural populations of sweet chestnut across Europe is applied to (1) evaluate the geographic patterns of genetic diversity, (2) identify spatial coincidences between genetic discontinuities and geographic barriers, and (3) propose certain chestnut populations as reservoirs of genetic diversity for conservation and breeding programs. Six polymorphic microsatellite markers were used for genotyping 1608 wild trees sampled in 73 European sites. The Geostatistical IDW technique (ArcGIS 9.3) was used to produce maps of genetic diversity parameters (He, Ar, PAr) and a synthetic map of the population membership (Q value) to the different gene pools. Genetic barriers were investigated using BARRIER 2.2 software and their locations were overlaid on a Digital Elevation Model (GTOPO30). The DIVA-GIS software was used to propose priority areas for conser
Article
Rare and vulnerable narrow endemic species represent distinct evolutionary units emerging from various temporal processes, and the preservation of such species is a key issue in biological conservation. Phylogeography has proven to be a relevant tool for distinguishing evolutionary units within species resulting from contrasted biogeographical events, and it can be leveraged to obtain historical and evolutionary perspectives. Yet, despite its usefulness, it is curiously underutilized in plant conservation genetics. Here we provide a comprehensive review of the available case studies on the structure of genetic diversity in the Mediterranean narrow endemic plants (MNEs) of the Mediterranean Basin hotspot. The use of genetic diversity structure for phylogeographical inference and for defining conservation units was examined in eighty-four studies dealing with eighty-three distinct taxa, most of which are perennial herbs occupying a narrow ecological niche. In addition, some 91.5% of the analyzed MNEs are located in the north-western part of the Mediterranean region, and this results in a geographical coverage that is heavily biased. Half of the studied species have moderate to high genetic diversity, and genetic differentiation is geographically structured in 56% of the case studies, indicating that MNEs are not “evolutionary dead-ends,” but rather represent species that have a strong evolutionary legacy. Taken at face value, this would imply conservation planning at the population level. However, it was only a minority of the studies that used these genetic structures to define conservation units. The main insight of the present review is that phylogeography is generally overlooked in conservation genetics. In fact, the design of conservation units has not often been the main goal of these studies, which more commonly is simply to enhance the scope of genetic diversity analyses of rare plants. Nevertheless, the strong phylogeographic structure revealed by several studies of MNEs underlines the relevance of phylogeography. We argue that comparative phylogeography across several co-occurring taxa could greatly improve the proactive conservation planning for threatened endemic plants within biodiversity hotspots.
Article
After more than fifteen years of existence, the R package ape has continuously grown its contents, and has been used by a growing community of users. The release of version 5.0 has marked a leap towards a modern software for evolutionary analyses. Efforts have been put to improve efficiency, flexibility, support for 'big data' (R's long vectors), ease of use, and quality check before a new release. These changes will hopefully make ape a useful software for the study of biodiversity and evolution in a context of increasing data quantity. Availability: ape is distributed through the Comprehensive R Archive Network: http://cran.r-project.org/package=apeFurther information may be found athttp://ape-package.ird.fr/.
Article
Genotyping errors are rules rather than exceptions in reality, and are found in virtually all but very small datasets. These errors, even when occurring at an extremely low rate, can derail many genetic analyses such as parentage/sibship assignments and linkage/association studies. 2.Nonetheless, few robust and accurate methods are available for estimating the rate of occurrence of genotyping errors and for identifying individual erroneous genotypes at a locus. Methods based on duplicate genotyping are expensive, and estimate genotype inconsistency rather than error rate at a locus. Methods based on Hardy-Weinberg equilibrium tests have low robustness and low power, and apply only to those particular errors that cause excessive homozygosity. Methods based on pedigrees are powerful, robust and accurate. However, they rely on known and complete pedigrees that are unfortunately rarely available from natural populations in the wild. 3.I proposed a maximum likelihood method to reconstruct pedigrees from genotype data with errors occurring at a roughly estimated (presumed) rate. In this paper, I describe how to use the method and inferred pedigree in estimating allelic dropout (or null allele) rate and false allele rate jointly at each marker locus, in identifying the erroneous genotypes, and in inferring the most likely genotypes at each locus of each individual. I examine the power, accuracy and robustness of the method by extensive simulations, and demonstrate the usefulness of the method by analysing three empirical datasets. 4.It is concluded that, both pedigrees and the rates of genotyping errors at each locus can be reliably estimated from the same genotype data by the same likelihood method, when marker information is sufficient and some sampled individuals are first-degree relatives. The erroneous genotypes are however inferred conservatively, and are reliably detected only when they occur in large families and/or at highly polymorphic loci. Estimation of genotyping error rates per locus and identification of erroneous genotypes of each individual at each locus should be routinely conducted to assess and improve data quality, to highlight markers for optimization of genotyping protocols or for replacement, and to enable the integration of genotyping errors in a robust statistical analysis. This article is protected by copyright. All rights reserved.