RESEARCH ARTICLE
Genetic diversity of Vitis vinifera in Georgia: relationshipsbetween local cultivars and wild grapevine, V. vinifera L.subsp. sylvestris
Jana Ekhvaia • Maia Gurushidze •
Frank R. Blattner • Maia Akhalkatsi
Received: 29 October 2013 / Accepted: 14 April 2014
� Springer Science+Business Media Dordrecht 2014
Abstract The Caucasus and Middle East regions are
considered to be theprimary centre of origin ofcultivated
grapevine, and, as confirmed by archaeobotanical,
archaeological, and cultural evidence, Georgia belongs
to this earliest centre of winemaking. This study aims to
investigate the genetic diversity, population structure
and relationships of local autochthonous wine cultivars
and wild grapevine, Vitis vinifera subsp. sylvestris.
Multiple accessions of 15 Georgian aboriginal cultivars
and 42 individuals of wild grapevine from different
regions of Georgia and adjacent Turkey were genotyped
at 17 nuclear microsatellite loci. A total of 160 alleles
were detected with a mean number of 9.41 alleles and the
effective number of 4.6 alleles (r) per locus, indicating
that the SSRs were highly informative. Despite high
genetic diversity, the level of genetic differentiation
among defined wild and cultivated populations is low
(Fst = 0,05; P***), which together with the outcome of
model-based cluster analyses and genetic assignment
methods point to gene flow among wild populations, as
well as among cultivated and wild accessions. Besides,
the data presented here suggest that local cultivars
‘Saperavi’ and ‘Tavkveri’ are independently derived
from different local wild populations, while the majority
of Georgian cultivars seem to have a single origin.
Overall, the present study takes important steps for better
characterization of Georgian cultivated and wild grape-
vines, and supports Georgia as one of the important
centres of grapevine domestication still harbouring
valuable genetic resources for grapevine breeding.
Keywords Domestication � Genetic diversity �Genetic similarity � Georgian grape cultivars �Microsatellites � Vitis vinifera subsp. sylvestris �V. vinifera subsp. vinifera
Introduction
Grapevine (Vitis vinifera L.) is one of the most valuable
crops worldwide and consists of two forms, domesti-
cated V. vinifera L. subsp. vinifera and wild V. vinifera
L. subsp. sylvestris (C.C.Gmel.) Hegi. The Caucasus
region is considered to be within the primary centre of
origin of domesticated grapevine, with high relevance
Electronic supplementary material The online version ofthis article (doi:10.1007/s10722-014-0125-2) contains supple-mentary material, which is available to authorized users.
J. Ekhvaia
Institute of Ecology, Ilia State University, Tbilisi, Georgia
M. Gurushidze (&) � F. R. Blattner
Leibniz Institute of Plant Genetics and Crop Research
(IPK), 06466 Gatersleben, Germany
e-mail: [email protected]
F. R. Blattner
German Centre for Integrative Biodiversity Research
(iDiv), 04103 Leipzig, Germany
M. Akhalkatsi
Department of Plant Genetic Resources, Institute of
Botany, Ilia State University, Tbilisi, Georgia
123
Genet Resour Crop Evol
DOI 10.1007/s10722-014-0125-2
for the further distribution of the crop throughout the
Mediterranean basin and for the development of
modern European cultivars (De Candolle 1885; Poteb-
nia 1911; Negrul 1946; Mullins et al. 1992; Jackson
1994; Damania et al. 1997; Sefc et al. 2003; Constan-
tini 2004; Forni 2006; This et al. 2006; Vouillamoz
et al. 2006). Grapevine was among the first fruits to be
cultivated in Georgia (Javakhishvili 1930). Confirma-
tions for long-lasting cultivation of grapevine in
Georgia stem from archaeological remains of berries
and seeds of domesticated grapes dated *8,000 years
before present (yBP) in southeastern Georgia (Ra-
mishvili 1988; McGovern et al. 1997). Other archae-
ological evidences of prehistoric winemaking are
found in near proximity of the Caucasian region in
northern Iran at the Hajji Firuz Tepe site in the northern
Zagros Mountains dated to about 7,400–7,000 yBP
(McGovern 2003) and in the Levant where archaeo-
logical findings are dated to *6,000 yBP (Zohary and
Spiegel-Roy 1975; Zohary and Hopf 2000).
Another indicator of a possible origin of cultivated
grapevine in the Caucasus region is high genetic and
morphological diversity of both wild and cultivated
grapes in this area (Vavilov 1931; Grassi et al. 2006;
Ekhvaia and Akhalkatsi 2010). About 500 names of
autochthonous grapevine varieties, including the cen-
turies-old cultivars ‘Rkatsiteli’, ‘Ojaleshi’, and ‘Sa-
peravi,’ are known from Georgia (Javakhishvili 1930;
Ketskhoveli et al. 1960). They show great ampelo-
metric variability and broad adaptability to different
climates and soils (Vinogradov-Nikitin 1929; Negrul
1946; Ketskhoveli et al. 1960; Ramishvili 1970;
Tsertsvadze 1989; Ekhvaia and Akhalkatsi 2006).
However, only half of these cultivars have been
conserved in some national collections, and today only
a small number of local varieties are still cultivated
(Chkhartishvili and Tsertsvadze 2004). This causes
genetic erosion on this rich ampelographic heritage,
involving loss of a valuable gene pool before it could
be evaluated.
Wild grapevine occurs in the Caucasus region
mainly in riparian forests and reaches upper vegetation
zones such as oak-hornbeam, beech and spruce forests
at up to 900 m a.s.l. (Ramishvili 1988). Like in other
parts of the world, the distribution area of the
subspecies was dramatically reduced in Georgia due
to human activities. Therefore, investigation and
preservation of genetic variability of wild grapevine
populations has become a priority to avoid genetic
erosion and maintain invaluable genetic resources for
cultivated grapevines (Arnold et al. 1998).
In recent years, particular attention has been paid to
elucidate the domestication history of cultivated grape
(Grassi et al. 2003; Sefc et al. 2003; Arroyo-Garcia et al.
2006; Myles et al. 2011). Myles et al. (2011) supposed a
Near East origin of subsp. vinifera, while others argue
that at least a second independent domestication centre
exists in the Mediterranean (Grassi et al. 2003; Sefc et al.
2003; Arroyo-Garcia et al. 2006; Lopes et al. 2009).
Special emphasis has been given to evaluation and
genetic characterization of different cultivars and/or
determination of the main events that enabled the
morphological transformation from wild subsp. sylves-
tris to cultivated grapevine (Aradhya et al. 2003;
Vouillamoz et al. 2006; Imazio et al. 2006; This et al.
2006; D’Onofrio et al. 2010; Bacilieri et al. 2013).
Unfortunately, the above-mentioned studies included
none or only a few autochthonous varieties from the
Caucasus region. A recent work of Imazio et al. (2013)
evaluated the genetic diversity of Georgian cultivars
from European germplasm repositories and investigated
their relationships with some worldwide distributed,
mainly European varieties. However, there is very
limited knowledge about genetic diversity and popula-
tion structure of Georgian wild grapevine. Conse-
quently, it is of high importance to study aboriginal
grape varieties from the supposed domestication area
and determine their genetic relationships to the local
wild populations. In addition, knowledge about genetic
variation in natural populations may provide relevant
information for biodiversity conservation strategies of
the region. In the present study, we have performed a
comprehensive comparative study of the native popula-
tions of wild subsp. sylvestris and the local autochtho-
nous cultivars. We employed 17 nuclear microsatellite
(SSR) markers to (1) study genetic diversity of Georgian
wild and domesticated grape germplasm and (2) assess
the relationships among autochthonous cultivars and
wild grapevine populations in order to shed light on the
origin of the locally cultivated varieties.
Materials and methods
Plant material
A total of 99 samples representing 57 accessions were
studied: 15 Georgian autochthonous cultivars (V.
Genet Resour Crop Evol
123
vinifera subsp. vinifera) and 42 individuals of wild V.
vinifera subsp. sylvestris from different regions of
Georgia and adjacent territory of Turkey. The wild
grapevine was represented by six populations located in
river basins: Ajaristskali population (n = 4), Alasan-Iori
population (n = 10), Aragvi population (n = 5), Chor-
okhi population (n = 8), Mtkvari population (n = 13),
and Tskhenistskali population (n = 2) (Table 1; Fig. 1).
All wild individuals were sampled from locations
characterized by environmental conditions typical for
wild grapevine habitats: riparian and floodplain forests
with a high degree of humidity and abundant tree species
on which grapevines grow as lianas. All individuals
within the studied populations of wild grape were
identified as dioecious plants, which clearly separates
them from the monoecious cultivated form.
Table 1 List of cultivated and wild accessions, colour of
berries (R—red, W—white); sex of individuals (F—female,
M—male), location and source institution name (MSEM—
Martvili State Ethnographical Museum collection, Martvili,
West Georgia; VOI—Viticulture and Oenology Institute col-
lection, Tbilisi, East Georgia)
N Variety and synonymy Sample number Colour Region of origin Source
Cultivated group
1 Aleksandrouli 4 R Racha-Lechkhumi reg., Georgia VOI
2 Avshiluri 4 R Samegrelo reg., Georgia VOI
3 Chodi 4 R Guria reg., Georgia MSEM
4 Chvitiluri 4 R Samegrelo reg., Georgia VOI
5 Kachichi 4 R Abkhazeti reg., Georgia VOI
6 Kamuri Tetri 4 W Guria re., Georgia MSEM
7 Khojishtoli, Kharistvala Kolkhuri 4 R Ajara, Guria, Samegrelo reg., Georgia VOI
8 Mujretuli 4 R Racha-Lechkhumi reg., Georgia VOI
9 Ojaleshi 4 R Samegrelo reg., Georgia MSEM
10 Rkatsiteli 4 W Kakheti reg., Georgia VOI
11 Saperavi 4 R Kakheti rek., Georgia VOI
12 Shkhucheshi 4 W Samegrelo reg., Georgia MSEM
13 Shonuri, Svanuri 4 R Svaneti reg., Georgia MSEM
14 Tavkveri 4 R Shida Kartli reg., Georgia VOI
15 Uchakhardani 4 R Samegrelo reg., Georgia VOI
N Population Individuals number Sex Location
Wild group
1 Ajaristskali 4 1M/3F Near vil. Godgadzeebi, Khulo distr., Ajara reg., Georgia
2 Alasan-Iori 10 7M/3F Near vil. Khalatsani, Akhmeta distr., Kakheti reg., Georgia
Lagodekhi Nature preserve, Lagodekhi distr., Kakheti reg., Georgia
Jumas Kure Nature preserve, Dedoplis Tskaro distr., Kakheti reg.,
Georgia
Iori Nature preserve, Sagarejo distr., Kakheti reg., Georgia
3 Aragvi 5 1M/4F Near vil. Zhinvali, Dusheti distr., Mtskheta-Mtianeti reg., Georgia
4 Chorokhi 8 5M/3F Near Artvin HES, Artvin, Turkey
5 Mtkvari 13 7M/6F Betw. vv. Atskuri and Likani, Borjomi distr., Samtskhe-Javakheti
reg., Georgia
Gardabani forest-park,Gardabani distr., Kvemo Kartli reg., Georgia
Near vil. Sakorintlo, Kaspi distr., Shida Kartli reg., Georgia; Suburb
of Tbilisi, Georgia
6 Tskhenistskali 2 2F Jonoula gorge, Tsageri distr., Racha-Lechkhumi reg., Georgia
Tskhaltubo-Tsageri road, Tsageri distr., Racha-Lechkhumi reg.,
Georgia
Genet Resour Crop Evol
123
Molecular methods
Genomic DNA was extracted with Qiagen DNeasy
Plant Mini Kit according to the manual provided by
the manufacturer or according to Lodhi et al. (1994)
from silica-dried leaves. 17 nuclear microsatellite loci,
well characterized in previous studies, were used:
VVS2, VVS4 (Thomas and Scott 1993); VVMD7,
VVMD24, VVMD25, VVMD27, VVMD28,
VVMD32, VVMD34 (Bowers et al. 1996, 1999);
scu04vv, scu14vv (Scott et al. 2000) used in studies on
V. vinifera, and VrZAG21, VrZAG47, VrZAG62,
VrZAG64, VrZAG79, VrZAG83 (Sefc et al. 1999),
originally identified in V. riparia Michx. Eight of these
markers (VrZAG62, VrZAG79, VVMD7, VVMD25,
VVMD27, VVMD28, VVMD32, VVS2) had been
previously selected by the European GENRES con-
sortium as the core set for genotyping grapevine
collections (Costantini et al. 2005). PCR amplification
was performed in 10 lL final volume containing about
10 ng template DNA, 0.2 lM of each dNTP, 0.2 lM
of each primer (one primer from each pair was
fluorescently labelled), 1.5 mM MgCl2 and 0.2 U Taq
DNA polymerase in 109 reaction buffer (QIAGEN).
The thermocycler (PE Biosystems, Gene Amp 9700)
was programmed for an initial step of 2 min at 94 �C
followed by 40 cycles at 92 �C for 30 s, 50–56 �C for
1 min, and 70 �C for 2 min, and a final extension step
at 72 �C for 10 min.
Length polymorphisms of the amplified products
were determined on a MegaBACE 1000 DNA
sequencer (Amersham Biosciences). Fragment lengths
were estimated in relation to an internal size standard
(Amersham Biosciences). In each run, we have
included 14 reference cultivars (Table 2), approved
by the two European projects GENRES-081 (Maul
and This 2008) and GrapeGen06 (Bacilieri 2007).
They served as standards in order to have consistent
allele sizes over all runs and they allowed allele size
comparison of our study with other published data.
Data analysis
Genotypes showing one or two alleles for each locus
were scored as homozygous and heterozygous, respec-
tively. Genetic polymorphism for each population
was assessed by calculating the mean number of
alleles per locus (MNA), the observed heterozygosity
Fig. 1 Geographical distribution of the studied wild V. vinifera subsp. sylvestris populations in Georgia and Turkey
Genet Resour Crop Evol
123
(Ho) (Brookfield 1996), expected heterozygosity (He)
under Hardy–Weinberg equilibrium (Nei et al. 1983),
probability of identity (PI) (Paetkau et al. 1995), and
estimation of null allele frequency from the heterozy-
gote deficiency (No) (Brookfield 1996) were calcu-
lated using genetic analysis package IDENTITY 4.0
(Wagner and Sefc 1999). The effective number of
alleles (r) was obtained according to the formula
Ne =P
(pi2)-1 (Morgante et al. 1994). The discrim-
ination power (D), which is an estimation of the
probability that two randomly sampled accessions
could be distinguished by their SSR profiles (Tessier
et al. 1999), was calculated as D = 1 - C, where C is
probability of coincidence, or the probability that two
varieties match by chance at one locus (C =P
pi2,
where pi is the frequency of different genotypes for a
given locus). The polymorphism information content
(PIC), which provides an estimate of the discrimina-
tory power of each SSR locus, was calculated as 1 -P
pi2 -
PP2pi
2pj2, where pi equals the frequency of
the ith allele and pj the frequency of the (i ? 1)th
allele (Botstein et al. 1980).
Population comparisons were made using measures
based on allele or genotype identity (exact tests, Fst)
because Gaggiotti and Excoffier (2000) have shown
that Fst is always the better estimator of gene flow/
population divergence when the sample sizes are small
to moderate. Allele frequencies were compared using
the probability test for genic differentiation described
by Raymond and Rousset (1995). Excess and defi-
ciency of heterozygotes, deviations from Hardy–
Weinberg equilibrium, genic and genotypic differen-
tiation were compared using GENEPOP, version 3.4
(Raymond and Rousset 1995) or GENEPOP on the Web
(http://genepop.curtin.edu.au/) with 1 million itera-
tions (1,000 batches with 1,000 iterations). Analysis of
molecular variance (AMOVA; Excoffier et al. 1992),
population pairwise genetic distances (Fst) (Weir and
Cockerham 1984), as well as probability (P) values for
Fst estimates generated based on 1,000 random per-
mutations (Excoffier et al. 1992) were computed in
ARLEQUIN, version 3.5. (Schneider et al. 2000) For the
Tskhenistskali population, the calculation of statistical
parameters was not considered executable because of
small number of genotypes.
In order to assign samples to each population and to
determine individuals that were first-generation
migrants by computing the probability of 1,000
simulated genotypes the program GENECLASS2 (Piry
et al. 2004) was used based on a Bayesian model and
the criterion described by Rannala and Mountain
(1997) and a simulation algorithm by Cornuet et al.
(1999).
Furthermore, individuals were clustered into a
given number of populations and assigned probabilis-
tically to clusters based on their multilocus genotypes
by using STRUCTURE version 2.3 (software available
from http://pritch.bsd.uchicago.edu/structure.html).
This Bayesian clustering approach estimates shared
ancestry of individuals based on their genotypes and
concludes individual proportions of ancestry from
‘‘K’’ clusters, where K is specified in advance and
corresponds to the hypothetical number of ancestral
populations. The best K was evaluated using STRUC-
TURE HARVESTER, version 0.9.93 (Earl and von Holdt
2012) available on http://taylor0.biology.ucla.edu/
structureHarvester/ and the algorithm described by
Evano et al. (2005). Consequently all sets of markers
tested were run with K ranging from K = 1 to
K = 15, and 15 replicates were performed at each
value of K. All STRUCTURE runs were performed
Table 2 List of reference cultivars, approved by the two
European projects GENRES-081 and GrapeGen06, including
colour of berries (R—red, W—white), location and source
institution name (INRA—French National Institute for Agri-
cultural Research, Montpellier UMR DIAPC, France)
Variety and synonymy Colour Origin Source
Alvarelhao N R Spain INRA
Cabernet Franc N R France INRA
Cabernet-Sauvignon N R France INRA
Chardonnay B W France INRA
Couderc 3309 (Interspecific
Cross: riparia; rupestris)
R France INRA
Fercal (Interspecific Cross:
includes berlandieri; vinifera)
R France INRA
Malegue 44–53 (interspecific
cross: cordifolia; riparia;
rupestris)
R France INRA
Mancin N R France INRA
Mouvedre N R Spain INRA
Muscat a petits grains blancs B W Greece INRA
Romorantin B W France INRA
Teleki 5C (interspecific cross:
berlandieri; riparia)
R Hungary INRA
Traminer rot RG R Italy INRA
Veltiner rot RG R Italy INRA
Genet Resour Crop Evol
123
without any prior population assignment, and used the
admixture model with a 20,000 step burn-in and
100,000 Markov Chain Monte Carlo (MCMC) itera-
tions, as recommended by Falush et al. (2007).
Genetic distances between wild grapevine acces-
sions and cultivars based on allele frequency data were
performed in the computer program POPTREE2 (Take-
zaki et al. 2010) using the neighbor-joining (NJ)
method (Saitou and Nei 1987) with Da distance
measure (Nei et al. 1983).
Results
Microsatellite variability
Seventeen nuclear SSR markers used for diversity
analysis in Georgian cultivated and wild grapevines
detected 160 alleles in the genotyped 99 individual
plants. The number of alleles per locus ranged from 3
(scu04vv, scu14vv) to 13 (VrZAG21, VrZAG62,
VrZAG64, VrZAG79 and VVMD28), which corre-
sponds to a mean number of 9.41 alleles (MNA) and
the effective number of alleles (r) was 4.6 per locus
(Table 3). The expected heterozygosity ranged
between 0.41 (scu14vv) and 0.85 (VVS2) with an
average 0.74, while observed heterozygosity varied
from 0.46 (VVS4) to 0.97 (VrZAG62) with an average
0.77. Seven loci (VrZAG21, VrZAG79, VrZAG83,
VVMD7, VVMD24, VVMD32, and VVS4) showed
an observed heterozygosity lower than Hardy–Wein-
berg conditions. The positive ratio of null alleles for all
seventeen loci is so close to zero (Table 3) that we
considered the samples with only one allele at a locus
homozygous. Polymorphism Information Content
(PIC) ranged from 0.12 (VVMD34) to 0.84 (VVMD7,
VVS2) with a mean value of 0.64 (Table 3), indicating
that the SSRs were highly informative. The least
informative markers were VVMD34, VVS4 and
scu14vv with PIC values of 0.12, 0.23, and 0.26,
respectively. Overall, the 17 SSR loci used in the study
allowed discrimination among the Georgian wild and
Table 3 Characteristics of the 17 SSR markers used for germplasm analysis
Marker name n Size range (bp) He Ho r No PI PIC
VrZAG21 13 188–218 0.822 0.792 5.488 0.017 0.097 0.796
VrZAG47 9 153–172 0.813 0.875 6.02 -0.034 0.107 0.82
VrZAG62 13 143–210 0.844 0.979 5.405 -0.073 0.075 0.792
VrZAG64 13 137–238 0.817 0.938 5.683 -0.066 0.099 0.805
VrZAG79 13 188–263 0.834 0.813 6.25 0.012 0.071 0.816
VrZAG83 7 186–235 0.718 0.689 3.571 0.018 0.212 0.651
VVMD7 12 210–261 0.829 0.771 6.704 0.032 0.088 0.84
VVMD24 10 208–241 0.684 0.604 3.283 0.047 0.191 0.568
VVMD25 8 176–270 0.776 0.792 3.791 -0.009 0.139 0.657
VVMD27 9 172–247 0.820 0.896 5.995 -0.042 0.103 0.82
VVMD28 13 227–277 0.773 0.787 5.195 -0.008 0.107 0.763
VVMD32 12 224–271 0.825 0.771 5.806 0.03 0.086 0.803
VVMD34 6 131–248 0.576 0.583 2.065 -0.004 0.287 0.119
VVS2 10 131–167 0.854 0.938 6.522 -0.045 0.073 0.837
VVS4 6 165–182 0.538 0.458 2.261 0.052 0.329 0.235
scu04vv 3 167–177 0.574 0.896 2.298 -0.205 0.459 0.377
scu14vv 3 167–181 0.416 0.521 1.836 -0.074 0.601 0.266
Total 160
average 9.41 0.736156118 0.770573 4.598 -0.021 0.184 0.645
Number of alleles (n), size range of alleles in basepairs (bp), expected heterozygosity (He), observed heterozygosity (Ho), number of
effective alleles (r), number of null alleles (No), probability of identity (PI) and polymorphism information content (PIC) of 17 SSR
loci used in cultivated and wild grapevines
Genet Resour Crop Evol
123
cultivated grapes. This is also supported by the
extremely low value obtained for the Total Probability
Identity (10-6 for all accessions), which indicates the
probability of falsely identifying these samples when
using all 17 SSR markers.
Genetic diversity of Georgian V. vinifera
The 17 nuclear SSR markers (Table 3) chosen to study
the diversity among wild populations of V. vinifera
subsp. sylvestris and Georgian aboriginal grapevine
cultivars identified 56 different genotypes in the 57
accessions: 8 for Chorokhi population (n = 8), 13 for
Mtkvari population (n = 13), 5 for Aragvi population
(n = 5), 10 for Alasan-Iori population (n = 10), 4 for
Ajaristskali population (n = 4), 2 for Tskhenistskali
population (n = 2) and 15 for cultivated grapevines
(n = 15). The number of alleles detected in the
samples of wild grapevine populations ranged from
34 for Tskhenistskali to 106 for Mtkvari with a total
number of 158 alleles, a mean number of alleles of
9.05 (MNA) and the effective number of alleles (r) of
4.43 for entire wild group. In cultivated samples, 111
alleles were identified with a MNA of 6.52 and the
effective number of 3.77 alleles (r) per locus
(Table 4). The five wild grapevine populations
showed moderate levels of gene diversity ranging
from 0.64 for Alasan-Iori to 0.77 for Mtkvari, in
contrast to the cultivated group, which showed high
level of expected heterozygosity (0.8). Values of
observed heterozygosity were higher than the
expected heterozygosity for all wild and cultivated
grapevines. The highest discrimination power at all
loci was found for Mtkvari (0.71) and the lowest for
Aragvi (0.63) and Ajaristskali (0.63) (Table 4).
Private alleles were found in all wild populations
with the exception of Tskhenistskali, and also within
the cultivated group. Numbers of private alleles were:
three in Ajaristskali, five in Alasan-Iori, six in Aragvi,
four in Chorokhi, six in Mtkvari and eight in the
cultivars (Table 5). Overall, the genetic diversity is
higher within the wild germplasm than within culti-
vated material.
Testing for Hardy–Weinberg equilibrium across all
loci and all populations resulted in a slight
Table 4 Genetic parameters obtained with 17 SSR loci for the 6 wild grapevine (V. vinifera subsp. sylvestris) populations and for the
Georgian aboriginal grapevine cultivars
Population n He Ho TNA MNA r No PI C D PIC
Cultivated 15 0.688 0.796 111 6.529 3.769 0.065 0.237 0.311 0.689 0.543
Ajaristskali 4 0.618 0.729 64 3.765 2.963 0.015 0.333 0.373 0.627 0.427
Alasan-Iori 10 0.64 0.641 88 5.176 3.241 0.001 0.259 0.36 0.64 0.424
Aragvi 5 0.677 0.741 80 4.706 3.075 0.039 0.294 0.374 0.626 0.425
Chorokhi 8 0.68 0.743 80 4.706 3.529 0.037 0.257 0.32 0.68 0.541
Mtkvari 13 0.711 0.774 106 6.235 4.001 0.037 0.216 0.289 0.711 0.589
Tskhenistskali 2 – – 34 – – – – – – –
Number of genotypes (n), expected heterozygosity (He), observed heterozygosity (Ho), total number of alleles (TNA), mean number
of alleles (MNA), number of effective alleles (r), number of null alleles (No), probability of identity (PI), probability of coincidence
(C), discrimination power (D) and polymorphism information content (PIC) of 17 SSR loci used in cultivated and wild grapevines
Table 5 Private alleles and their frequencies (in parentheses)
observed in wild grapevine populations and cultivated group at
12 out of 17 SSR loci
N Locus Wild Cultivated
1 VrZAG21 198 (0.125), 200 (0.063),
216 (0.039), 218 (0.077)
214 (0.09)
2 VrZAG47 169 (0.05)
3 VrZAG62 188 (0.188), 192 (0.033),
210 (0.125)
198 (0.033),
208 (0.033)
4 VrZAG64 157 (0.1), 161 (0.039)
5 VrZAG83 196 (0.077)
6 VVMD7 261 (0.063), 263 (0.1) 255 (0.033)
7 VVMD25 264 (0.033)
8 VVMD27 188 (0.1)
9 VVMD28 251 (0.05), 271 (0.039), 275
(0.2)
231 (0.033),
241 (0.033)
10 VVMD32 239 (0.039)
11 VVMD34 218 (0.25)
12 VVS2 127(0.05), 137 (0.1)
Genet Resour Crop Evol
123
disequilibrium (P = 0.05) with slight heterozygote
excess. However, a test for Hardy–Weinberg equilib-
rium in cultivated grapevines revealed a deviation
from equilibrium and significant heterozygote excess
(P = 0.0085), as expected for outbreeding species.
All pairwise tests of linkage disequilibrium between
loci were non-significant after corrections for multiple
tests.
Pairwise genetic distance (Fst) values were found to
be significantly different from zero (P \ 0.001) in all
pairwise comparisons between populations (Table 6),
except for the comparison between the Ajaristskali
and Chorokhi populations (P = 0.013). Fst values at a
99 % confidence level revealed a low level a genetic
differentiation (0.05***) between Georgian wild and
cultivated grapevines when all samples, including the
Tskhenistskali population of subsp. sylvestris, were
included in the analyses.
Genetic relationships between cultivated and wild
grapevines
Population assignment and exclusion simulation test,
which was performed in GENCLASS2 for the five main
wild grapevine populations (Tskhenistskali excluded
due to small sample size) and Georgian aboriginal
cultivars, correctly assigned all samples to their
population of origin (Table 7). Interestingly, several
wild grapevine accessions were assigned not only to
the predefined, but also to another population. In total,
11 potential first-generation migrants were recognized
within five wild populations of subsp. sylvestris with
high probability (P \ 0.01). These are: one sample of
Ajaristskali defined as a potential first-generation
migrant from Mtkvari; one sample of Ajaristskali as
a first generation migrant from Chorokhi; two plants of
Alasan-Iori are potential first generation migrants
from Ajaristkali and Mtkvari; one sample of Aragvi is
a potential first generation migrant from Mtkvari while
another sample of Aragvi is a potential first generation
migrant from Ajaristskali; two plants from Chorokhi
could be potential first generation migrants from
Ajaristskali; and three plants from Mtkvari are
potential first generation migrants from Ajaristskali,
Chorokhi and Aragvi populations, respectively. When
the samples were tested according to subspecies, four
samples belonging to Chorokhi (one sample) and
Mtkvari (three samples) were found to be potential
first generation migrants from the cultivated subsp.
vinifera grapevine samples, while four Georgian
cultivated samples (‘Chvitiluri’, ‘Ojaleshi’, ‘Saperavi’,
and ‘Tavkveri’) were identified as potential first
generation migrants from subsp. sylvestris.
To further evaluate intraspecific population differ-
entiation between subsp. sylvestris populations, and
their genetic similarity to the Georgian aboriginal
cultivars, we used STRUCTURE to group populations into
clusters. Number of genetic clusters (K) ranging from
1 to 15 was tested using the admixture model,
supposing that each individual comes simply from
one of the K population. STRUCTURE supported the
presence of differentiation among the populations, and
simulation summary showed that the K value showed
the highest peak at K = 3 (Fig. 2), meaning three
populations could account for all individuals with the
highest probability. The graphical representation of
the estimated membership coefficients to the three
clusters for the wild populations and cultivated group
is given in Fig. 3, where each individual is represented
by a single vertical bar composed of K coloured
segments. Here, coloured segments represent the
individual’s estimated membership fractions in K
Table 6 Genetic differentiation between wild grapevine populations and cultivated grapevines
Cultivated Ajaristskali Alasan-Iori Aragvi Chorokhi Mtkvari
Cultivated – 0.077 (0.009) 0.09 (0.004) Inf. (Inf.) 0.078 (0.008) 0.02 (0.004)
Ajaristskali 0.036 (19.524) – 0.001 (0.001) Inf. (Inf.) 0.311 (0.013) 0.003 (0.002)
Alasan-Iori 0.116 (24.003) 0.04 (15.581) – 0.007 (0.003) 0.021 (0.005) Inf. (Inf.)
Aragvi 0.08 (28.212) 0.051 (12.945) 0.059 (24.64) – 0.032 (0.005) Inf. (Inf.)
Chorokhi 0.043 (14.315) 0.07 (23.856) 0.09 (25.0) 0.075 (26.376) – 0.015 (0.004)
Mtkvari 0.016 (5.962) 0.02 (9.677) 0.062 (10.549) 0.04 (15.082) 0.014 (9.935) –
The upper triangle shows Pairwise values for Fst with P values in parentheses. The lower triangle shows k2 values for the
homogeneity of allele frequencies and for genotypes (in parentheses) in pairwise comparisons
Genet Resour Crop Evol
123
clusters. STRUCTURE analysis shows, that wild and
cultivated accessions are not clearly separated from
each other (Fig. 3a). Assignment of the individuals to
the clusters when K = 3 is shown in Fig. 3b. Cluster I
mainly comprised individuals from Alasan-Iori and
Aragvi populations and two representatives
Fig. 2 Bayesian assignment analysis performed in STRUCTURE showing A the posterior probability of the data [L(K)] after 15 runs
assuming K = 1–15. B DK as a function of K following Evano et al. (2005) using the correlated allele frequency model
Fig. 3 A Estimated population structure of the Georgian
aboriginal cultivars and wild V. vinifera subsp. sylvestris
populations (K = 3); each individual is represented by a single
vertical bar split into K segments, with lengths proportional to
the ratio of each of the K inferred clusters. B Summary plot of
estimated membership of 99 individuals in three clusters
Table 7 Assignment of 44 V. vinifera subsp. sylvestris individuals to 5 predefined populations using the algorithm of GeneClass2
Ajaristskali Alasan-Iori Aragvi Chorokhi Mtkvari
Number of individuals 5 10 5 8 13
% assigned to the predefined population 100 100 100 100 100
% assigned to the predefined population and another population 60 80 60 75 76.9
% assigned not to the predefined population but to another
population
% assigned not to neither the predefined population nor another
population
Genet Resour Crop Evol
123
(‘Kachichi’ and ‘Saperavi’) from the cultivated group;
Cluster II contained individuals from the rest of the
wild populations and five cultivars; and Cluster III
consisted of individuals primarily from the cultivated
group and a few individuals from Chorokhi and
Mtkvari populations.
The unrooted phylogram generated using Da dis-
tances is moderately resolved, and consists of four
clusters (A, B, C and D) and the unresolved group
comprising wild accessions (Fig. 4). Basically, all
clusters except cluster A (uniting most cultivars) are
represented nearly exclusively by wild grapevine
samples: Cluster A consists of predominantly Geor-
gian cultivars and three wild accessions (two of
Chorokhi and one of Mtkvari populations), similar to
the STRUCTURE results. Cluster B is the largest and
comprises almost entirely Georgian wild accessions,
with the exception of three aboriginal Georgian
cultivars (‘Aleksandrouli’, ‘Mujretuli’, and ‘Ojale-
shi’). Cluster C unites mostly wild individuals and the
autochthonous Georgian cultivar ‘Saperavi’ similar to
STRUCTURE Cluster I. However, another cultivar
‘‘Kachichi’’, which according to the STRUCTURE ana-
lysis is also assigned to the same cluster, based on NJ
analysis is placed in cluster A. Cluster D unites wild
grapevine accessions mainly from Mtkvari and Aragvi
and the cultivar ‘Tavkveri’, which according to the NJ
tree is derived from Mtkvari population (Fig. 4).
Discussion
Genetic diversity and relationships
within Georgian V. vinifera
Genotype analysis of Georgian aboriginal cultivars
and wild V. vinifera subsp. sylvestris populations at 17
nuclear microsatellite loci showed that both cultivated
and wild grapevine samples possess high levels of
genetic variability similar to those reported in different
Fig. 4 Neighbor-joining
dendrogram constructed
from allele-sharing
distances among 42 samples
of V. vinifera subsp.
sylvestris and 14 genotypes
of V. vinifera subsp.
vinifera: Cultivated group
(red), Ajaraistskali (green),
Alasan-Iori (black), Aragvi
(pale blue); Chorokhi
(cobalt blue), Mtkvari
(brown) and Tskhenistskali
(orange).
Genet Resour Crop Evol
123
grapevine datasets (Bowers et al. 1996; Sefc et al.
2003; Ibanez et al. 2003; Snoussi et al. 2004; Grassi
et al. 2006; Martin et al. 2006; Lopes et al. 2009;
Schaal et al. 2010; Beridze et al. 2011; Pipia et al.
2012; Zecca and Grassi 2013; Imazio et al. 2013).
High genetic diversity in wild grapevine found in this
study is consistent with the study of Grassi et al.
(2006), which despite very limited sampling from
Caucasus, reported highest chloroplast diversity in the
Caucasus region, harbouring all detected chloroplast
haplotypes of this subspecies. This result has
prompted the hypothesis about considering the Cau-
casian region as a possible centre of origin of subsp.
sylvestris. Slightly higher genetic diversity in the wild
subsp. sylvestris compared to the cultivated subspecies
was rather expected and is in agreement with the
previous study of Myles et al. (2011) suggesting that
grape domestication involved only a weak genetic
bottleneck.
The observed low level of genetic differentiation
among wild and cultivated populations (Fst = 0.05;
P***) indicate gene flow between wild and Georgian
domesticated grapevines and/or that in situ repeated
domestication of wild germplasm took place within
local populations (see below).
Population genetic structure analyses of all acces-
sions identified three groups of individuals clustered
irrespective of their collection region, and showed
admixture among cultivated and wild samples that is
similar to the earlier study (Ekhvaia et al. 2010)
performed with less microsatellite markers and
smaller sample size. The GENECLASS2 clustering and
the NJ phylogram provided comparable outcomes
indicating introgression among different wild popula-
tions, as well as among cultivated and wild accessions.
Gene flow from cultivated to wild grapevines has been
frequently addressed (Di Vecchi-Staraz et al. 2009;
Lopes et al. 2009; Myles et al. 2011), while introgres-
sion from wild to cultivated grapes has been relatively
rarely reported. Although the extent of introgression
from wild to the crop gene pool has been long debated
(Slatkin 1987; Arnold and Hodges 1995), there have
been few documented cases concerning natural or
farmer-assisted introgression into crop cultivars (for a
review see Jarvis and Hodgkin 1999). Our data from
assignment tests indicate that gene flow might indeed
occur in both directions. This is consistent with a
previous study suggesting that Western European
grape cultivars experienced introgression from local
subsp. sylvestris (Myles et al. 2011). Main mechanism
of introgression in V. vinifera populations is probably
pollen mediated gene flow, as V. vinifera is outcross-
ing and predominantly wind pollinated (Huglin 1988).
However, gene flow via seed transfer, actively med-
iated by birds, or exchange of cuttings by viticulturists
over the centuries could be also alternative scenarios
for introgression. The latter mechanism used to
enhance diversity of local germplasm might have
resulted in admixture of alleles between wild and
cultivated subspecies of V. vinifera irrespective of
their geographical distance.
Origin of local cultivars
The fact that most Georgian aboriginal cultivars group
together in all analyses (cluster III in STRUCTURE
output and cluster A in NJ phylogram) supports the
idea that these cultivars originated from a single
domestication event followed by diversification
within the domesticated genepool (Figs. 3, 4).
Another hypothesis, that domestication of these
cultivated varieties took place several times from
genetically similar material, is a less parsimonious
explanation. Interestingly, the ancient Georgian cul-
tivars ‘Ojaleshi’, ‘Saperavi’ and ‘Tavkveri’ occupy
derived positions in NJ phylogram (Fig. 4). In addi-
tion, these cultivars were identified as potential first-
generation migrants from subsp. sylvestris by GENE-
CLASS2, supporting the ‘‘in situ’’ domestication
hypothesis for these cultivars. Autochthonous cultivar
‘Saperavi’ falls within clusters consisting of exclu-
sively wild individuals according to STRUCTURE ana-
lysis and the NJ phylogram (Figs. 3b, 4). Moreover,
NJ groups this cultivar with accessions of the Alasan-
Iori populations that allows us to suppose that this
cultivar was derived from a local domestication event.
It is known that ‘Saperavi’ is a very old Georgian
cultivar, which currently is highly commercialized in
Kakheti (Eastern Georgia) for production of the
famous red wines ‘‘Saperavi’’ and ‘‘Kindzmarauli’’.
A similar case has already been mentioned for the
German-Austrian cultivars ‘Ortlieber’ and ‘Orange-
traube’ (Ambrosi et al. 1994), for Portuguese cultivar
‘Alvarinho’ (Lopes et al. 2009), and Azerbaijan
cultivars ‘Sargila ortayetishan’ and ‘Sargila gejyeti-
shan’ (Salayeva et al. 2010), which were probably
selected from local German, Portuguese and Azerbai-
jan wild populations, respectively.
Genet Resour Crop Evol
123
Another example confirming the genetic linkage
between Georgian cultivars and local wild grapevine
is the famous Georgian red cultivar ‘Tavkveri’, which
has functionally female flowers. This cultivar was
identified as potential first-generation migrant from
subsp. sylvestris and also according to NJ analysis
groups in cluster D together with four wild accessions
(three male and one female plant) from the suburbs of
Tbilisi (Fig. 4). ‘Tavkveri’ has a sister-group relation-
ship to the female wild individual Tbilisi5, the GS
value among these two accessions is 0.97, due to 32
out of 34 alleles being identical. According to
Ketskhoveli et al. (1960) ‘Tavkveri’ originated in
Georgia by natural reproduction, and later the best
form was selected by humans. Nowadays it is widely
spread in Kartli and in Kakheti (Eastern Georgia) as a
famous red wine cultivar. As mentioned above, our
genetic results have shown that ‘Tavkveri’ clustered
with wild populations from the suburbs of Tbilisi,
meaning that in addition to their genetic relationships
they are also closely situated geographically, which is
consistent with data obtained by Ketskhoveli et al.
(1960) that ‘Tavkveri’ originated in the Kartli region
(Eastern Georgia). We compared flower morphology
of ‘Tavkveri’ with that of the wild female individual
Tbilisi5 in order to exclude that this wild accession can
be an escaped (feral) ‘Tavkveri’ cultivar. The com-
parison has shown that the flowers of these two
samples are quite different from each other. In
particular, the wild individual has longer styles and
narrower pistils, while ‘Tavkveri’ has very short styles
and wide, peculiar pitcher-like formed pistils, which
determine the name of this cultivar meaning ‘‘pitcher’’
in Georgian language (Ekhvaia 2012). Thus, the
hypothesis that this cultivar could have been derived
and selected from local wild grape population seems
more plausible.
In conclusion, it should be mentioned that the
Georgian cultivated and wild grapevines represent a
unique and interesting genetic resource possessing
high genetic diversity. Low genetic differentiation and
strong admixture found between wild and cultivated
material indicate considerable gene flow in both
directions. The data presented here identifies a group
of genetically closely related autochthonous cultivars
as derived from local wild grapevine (cluster C), while
the cultivars ‘Saperavi’ and ‘Tavkveri’ are indepen-
dently derived from different local wild populations.
Using the molecular markers we have identified the
populations from which these cultivars were most
probably selected. Thus, the genetic data obtained in
our study support archaeobotanical, archaeological,
and historical evidence that Georgia belongs to one of
the important centres of winemaking and grapevine
domestication. In addition, our study shows, that local
wild grapevines still harbour high genetic diversity, a
valuable resource for grape breeding, and suggests that
wild grape populations in Georgia should be given a
high conservation priority.
Acknowledgments The study was conducted in the
framework of the EU 6th FP (2007–2011), ‘‘Management and
Conservation of Grapevine Genetic Resources (GrapeGen06)’’
grant under Council Regulation (EC) No 870/2004, Proposal
GEN RES 2005 008. This work was partially funded by the IPK
Gatersleben, Germany. We thank V. Laucoui (INRA,
Montpellier UMR DIAPC, France) for providing us grape
DNAs, Petra Oswald (IPK) for technical support, Dr.
V. Gotsiridze for supporting us during collecting of wild
grapevine materials in Georgia and supplying leaf material of
cultivars from the collection of the Institute of Viticulture and
Oenology, Tbilisi, Georgia, and G. Eliava for providing leaf
material of West Georgian autochthonous grapevine cultivars
from the collection of the Martvili Ethnographic Museum.
References
Ambrosi H, Dettweiler-Munch E, Ruhl EH, Schmid J, Schu-
mann F (1994) Farbatlas Rebsorten: 300 Sorten und ihre
Weine. Eugen Ulmer, Stuttgart
Aradhya MK, Dangl GS, Prins BH, Boursiquot JM, Walker MA,
Meredith CP, Simon CJ (2003) Genetic structure and dif-
ferentiation in cultivated grape (Vitis vinifera L.). Genet
Resour Crop Evol 81:179–182
Arnold ML, Hodges S (1995) Are natural hybrids fit or unfit
relative to their parents? Trends Ecol Evol 10:67–71
Arnold C, Gillet F, Cobat JM (1998) Situation de la vigne
sauvage (Vitis vinifera subsp. sylvestris) en Europe. Vitis
37:159–170
Arroyo-Garcia R, Ruiz-Garcia L, Bolling L, Ocete R, Lopez
MA, Arnold C, Ergul A, Soylemezoglu G, Uzun HI, Ca-
bello F, Ibanez J, Aradhya MK, Atanassov A, Atanassov I,
Balint S, Cenis JL, Constantini L, Goris-Lavets J, Gramdo
MS, Klein BY, McGovern PE, Merdinoglu D, Pejuc I,
Pelsy F, Primikirios N, Rissovannaya V, Roubelakis-An-
gelakis KA, Snoussi H, Satiti P, Tamankar S, This P,
Troshin L, Malpica JM, Lefort F, Martinez-Zapater JM
(2006) Multiple origins of cultivated grapevine (Vitis
vinifera L. subsp. sativa) based on chloroplast DNA
polymorphisms. Mol Ecol 15:3707–3714
Bacilieri R (2007) GrapeGen06—management and conserva-
tion of grapevine GR. Bioversity Newsl Europe 34:16
Bacilieri R, Lacombe T, Le Cunf L, Di Vecch-Staraz M, Laucou
V, Senna B, Peros JP, This P, Boursiquet JM (2013)
Genet Resour Crop Evol
123
Genetic structure in cultivated grapevines is linked to
geography and human selection. BMC Plant Biol 13:25
Beridze T, Pipia I, Beck J, Hsu SC, Gamkrelidze M, Gognashvili
M, Tabidze V, This P, Bacilieri R, Gotsiridze V, Glonti M,
Schaal B (2011) Plastid DNA sequence diversity in a
worldwide set of grapevine cultivars (Vitis vinifera L.
subsp. sativa). Bull Georgian Natl Acad Sci 5:98–103
Botstein D, White RL, Skolnick M, Davis RW (1980) Con-
struction of genetic linkage map in man using restriction
fragment length polymorphisms. Am J Hum Genet
32:314–331
Bowers JE, Dangl GS, Vignani R, Meredith CP (1996) Isolation
and characterization of new polymorphic simple sequence
repeat loci in grape (Vitis vinifera L.). Genome 39:628–633
Bowers JE, Dangl GS, Meredith CP (1999) Development and
characterization of additional microsatellite DNA markers
for grape. Am J Enol Viticult 50:243–248
Brookfield JFY (1996) A simple new method for estimating null
allele frequency from heterozygote deficiency. Mol Ecol
5:453–455
Chkhartishvili N, Tsertsvadze N (2004) Status of grapevine
genetic resources (Vitis vinifera), in Georgia. In: Pro-
ceedings of the international meeting ‘Conservation and
use of genetic resources of grapevine in the Caucasus and
Northern Sea region’, Tbilisi, Georgia
Constantini L (2004) The archaeobotany of grapevine in the
Caucasus and Northern Black Sea region. In: Proceedings
of the international meeting on development of national
programs in plant genetic resources in South-Eastern
Europe-Conservation of Grapevine in the Caucasus and
Northern Black Sea Region, Yalta, Ukraine, pp 73–75
Cornuet JM, Piry S, Luikart G, Estope G, Solignac M (1999)
New methods employing multilocus genotypes to select or
exclude populations as origin of individuals. Genetics
153:1989–2000
Costantini L, Monaco A, Vouillamoz JF, Forlani M, Grando MS
(2005) Genetic relationships among local Vitis vinifera
cultivars from Campania (Italy). Vitis 44:25–34
D’Onofrio C, De Lorenzis G, Giordani T, Natali L, Cavallini A,
Scalabrelli G (2010) Retrotransposons-based molecular
markers for grapevine species and cultivars identification.
Tree Genet Genomes 6:451–466
Damania AB, Valkoun J, Willcox G, Qualset CO (1997) The
origins of agriculture and crop domestication. In: Pro-
ceedings of the Harlan symposium, Aleppo, Syria
De Candolle A (1885) Origine des plantes cultivees. Germer
Bailliere, Paris
Di Vecchi-Staraz M, Laucou V, Bruno G, Lacombe T, Gerber S,
Bourse T, Boselli M, This P (2009) Low level of pollen-
mediated gene flow from cultivated to wild grapevine:
consequences for the evolution of the endangered subspe-
cies Vitis vinifera subsp. sylvestris. J Hered 100:66–75
Earl DA, von Holdt BM (2012) STRUCTURE HARVESTER: a
website and program for visualising STRUCTURE output
and implementing the Evanno method. Conserv Genet Res
4:359–361
Ekhvaia J (2012) Comparative morphometric and molecular-
systematic study of Georgian autochthonous grapevine
cultivars and wild grape populations of Vitis vinifera L.
subsp. sylvestris (C. C. Gmel.) Hegi. PhD thesis, Ilia State
University, Tbilisi, Georgia
Ekhvaia J, Akhalkatsi M (2006) Comparative study of the
quantitative parameters of berries and seeds in the
autochthonous red grape varieties in the Kolkhic (Western
Georgia). Proc Georgian Natl Acad Sci Biol B 4:38–46
Ekhvaia J, Akhalkatsi M (2010) Morphological variation and
relationships of Georgian populations of Vitis vinifera L.
subsp. sylvestris (C.C. Gmel.) Hegi. Flora 205:608–617
Ekhvaia J, Blattner FR, Akhalkatsi M (2010) Genetic diversity
and relationships between wild grapevine (Vitis vinifera
subsp. sylvestris) populations and aboriginal cultivars in
Georgia. In: Proceedings of 33rd world congress of vine
and wine, 8th general assembly of the OIV, Tbilisi, Georgia
Evano G, Kegnaut S, Goudes J (2005) Detecting the number of
clusters of individuals using the software STRUCTURE: a
simulation study. Mol Ecol 14:2611–2620
Excoffier L, Smouse PE, Quaratto JM (1992) Analysis of
molecular variance inferred from metric distances among
DNA haplotypes: application to human mitochondrial
DNA restriction data. Genetics 131:479–491
Falush D, Spepheng MW, Pritchard JK (2007) Inference of
population structure using multilocus genotype data: domi-
nant markers and null alleles. Mol Ecol Notes 7:574–578
Forni F (2006) Chloroplast microsatellites to investigate the ori-
gin of grapevine. Genet Resour Crop Evol 53:1003–1011
Gaggiotti OE, Excoffier L (2000) A simple method of removing
the effect of a bottleneck and unequal population sizes on
pairwise genetic distances. Proc R Soc B Biol 267:81–87
Grassi F, Labra M, Imazio S, Spada A, Sgorbati S, Scienza A,
Sala F (2003) Evidence of a secondary grapevine domes-
tication centre detected by SSR analysis. Theor Appl Genet
107:1315–1320
Grassi F, Labra M, Imazio S, Ocete R, Failla O, Scienza A, Sala
F (2006) Phylogeographical structure and conservation
genetics of wild grapevine. Conserv Genet 7:837–845
Huglin P (1988) Biologie et ecologie de la vigne, 2nd edn. Tec
and Doc, Bordeaux
Ibanez J, de Andres MT, Molino A, Borrego J (2003) Genetic
study of key Spanish grapevine varieties using microsat-
ellite analysis. Am J Enol Viticul 54:22–30
Imazio S, Labra M, Grassi F, Scienza A, Failla O (2006)
Chloroplast microsatellites to investigate the origin of
grapevine. Genet Resour Crop Evol 53:1003–1011
Imazio S, Maghradze D, De Lorenzis G, Bacilieri R, Laucou V,
This P, Scienza A, Kaill O (2013) From the cradle of
grapevine domestication: molecular overview and
description of Georgian grapevine (Vitis vinifera L.)
germplasm. Tree Genet Genomes 9:641–658
Jackson RS (1994) Wine science. Principles and applications.
Academic Press, New York
Jarvis DI, Hodgkin T (1999) Wild relatives and crop cultivars:
detecting natural introgression and farmer selection of new
genetic combinations in agroecosystems. Mol Ecol
8:S159–S173
Javakhishvili I (1930) Sakartvelos Ekonomiuri istoria (Eco-
nomic history of Georgia). Tbilisi, Georgia (in Georgian)
Ketskhoveli N, Ramishvili M, Tabidze D (1960) Sakartvelos
ampelografia (Ampelography of Georgia). Tbilisi, Georgia
(in Georgian)
Lodhi MA, Ye CN, Weeden NF, Reisch BI (1994) A simple and
efficient method for DNA extraction from grapevine cul-
tivars and Vitis species. Plant Mol Biol Rep 12:6–13
Genet Resour Crop Evol
123
Lopes MS, Mendonca D, Rodrigues dos Santos M, Eiras-Dias
JE, da Camara Machado A (2009) New insights on the
genetic basis of Portuguese grapevine and on grapevine
domestication. Genome 52:790–800
Martin JP, Santiago PC, Leel F, Martinez MC, Ortiz JM (2006)
Determination of relationships among autochthonous
grapevine varieties (Vitis vinifera L.) in the Northwest of
the Iberian Peninsula by using microsatellite markers.
Genet Resour Crop Evol 53:1255–1261
Maul E, This P (2008) GENRES 081—a basis for the preser-
vation and utilization of Vitis genetic resources. In: Maul
E, Dias JE, Kaserer E, Lacombe HT, Ortiz JM, Schneider
A, Maggioni L, Lipman E (eds) Report of a working group
on Vitis. pp 13–22
McGovern PE (2003) Ancient wine. Princeton University Press,
Princeton
McGovern PE, Hartung U, Badler VR, Glusker DL, Exner LJ
(1997) The beginnings of winemaking and viniculture in
the ancient Near East and Egypt. Expedition 39:3–21
Morgante M, Rafalski A, Biddle P, Tingey AM, Olivieri AM
(1994) Genetic mapping and variability of seven soybean
simple repeat loci. Genome 37:763–769
Mullins MG, Bouquet A, Williams LE (1992) Biology of the
grapevine. Cambridge University Press, Cambridge
Myles S, Boyko AR, Owens CL, Brown PJ, Grassi F, Aradhya
MK, Prins B, Reynolds A, Chia J, Ware D, Bustamante CD,
Buckler ES (2011) Genetic structure and domestication his-
tory of the grape. Proc Natl Acad Sci USA 108:3530–3535
Negrul AM (1946) Semeistvo Vitaceae Linde (Familiy Vitaceae
Linde). Ampelography of the USSR. Moscow, Russia (in
Russian)
Nei M, Tajima F, Tateno Y (1983) Accuracy of estimated
phylogenetic trees from molecular data. J Mol Evol
19:153–170
Paetkau D, Calvet W, Stirling I, Strobeck C (1995) Microsat-
ellite analysis of population structure in Canadian polar
bears. Mol Ecol 4:347–358
Pipia I, Gogniashvili M, Tabidze V, Beridze T, Gamkrelidze M,
Gotsiridze V, Melyan G, Musayev M, Salimov V, Beck J,
Schaal B (2012) Plastid DNA sequence diversity in wild
grapevine samples (Vitis vinifera subsp. sylvestris) from
the Caucasus region. Vitis 51:119–124
Piry S, Alapetite A, Cornuet JM, Paetkau D, Baudouin L, Estoup
A (2004) GENECLASS2: software for genetic assignment
and first-generation migrant detection. J Hered 95:536–539
Potebnia A (1911) Semena vinograda evropeiskikh sortov I ikh
znachenie dlia klassificatsii (Grape seeds of European
grapevine varieties and their importance for classification).
Trudi Biuro po Prikladnoi Botanike. pp 159–165 (in
Russian)
Ramishvili M (1970) Ampelography. Ganatleba, Tbilisi (in
Georgian)
Ramishvili R (1988) Dikorastushii vinograd Zakavkazia (Wild
grape of the South Caucasus). Ganatleba, Tbilisi (in
Russian)
Rannala B, Mountain JL (1997) Detecting immigration by using
multilocus genotypes. Proc Natl Acad Sci USA 94:
9197–9201
Raymond M, Rousset F (1995) GENEPOP (version 1.2): pop-
ulation genetic software for exact tests and ecumenicism.
J Hered 86:145–155
Saitou N, Nei M (1987) The neighbor-joining method: a new
method for reconstructing phylogenetic trees. Mol Biol
Evol 4:406–425
Salayeva S, Akhundova E, Mammadov A (2010) Evaluation of
DNA polymorphism among cultivated and wild grapevine
accessions from Azerbaijan. Czech J Genet Plant Breed
46:75–84
Schaal B, Beck J, Hsu SC, Beridze T, Gamkrelidze M, Gogni-
ashvili M, Pipia I, Tabidze V, This P, Bacilieri R, Gots-
iridze V, Ghlonti M (2010) Plastid DNA sequence diversity
in a worldwide set of grapevine cultivars (Vitis vinifera L.
subsp. sativa). In: Proceedings of 33rd world congress of
vine and wine, 8th general assembly of the OIV. Tbilisi,
Georgia
Schneider S, Roessli D, Excoffier L (2000) Arlequin ver. 2.000.
Software for population genetic data analysis. Genetics and
Biometry Laboratory, University of Geneva, Switzerland
Scott KP, Eggler P, Seaton G, Rosseto M, Ableft EM, Lec LS,
Henry RJ (2000) Analysis of SSR derived from grape
ESTs. Theor Appl Genet 100:723–726
Sefc KM, Regner F, Turetschek E, Glossi J, Steinkellner H
(1999) Identification of microsatellite sequences in Vitis
riparia and their applicability for genotypes of different
Vitis species. Genome 42:367–373
Sefc KM, Steinkellner H, Lefort F, Botta R, Da Camara
Machado A, Borrego J, Maletic E, Glossl H (2003) Eval-
uation of the genetic contribution of local wild vines to
European grapevine cultivars. Am J Enol Viticul 54:15–21
Slatkin M (1987) Gene flow and the geographic structure of
natural populations. Science 236:787–792Snoussi H, Harbi Ben Slimane M, Ruiz-Garcia L, Zapataer JM,
Arroyo-Garcia R (2004) Genetic relationships among
cultivated and wild grapevine accessions from Tunisia.
Genome 47:1211–1219
Takezaki N, Nei M, Tamura K (2010) POPTREE2: software for
constructing population trees from allele frequency data
and computing other population statistics with Windows
interface. Mol Biol Evol 27:747–752
Tessier C, David J, This P, Boursiquot JM, Charrier A (1999)
Optimization of molecular markers for varietal identifica-
tion in Vitis vinifera L. Theor Appl Genet 89:171–177
This P, Lacombe T, Thomas MR (2006) Historical origins and
genetic diversity of wine grapes. Trends Genet 22:511–519
Thomas MR, Scott NS (1993) Microsatellite repeats in grapevine
reveal DNA polymorphisms when analyzed as sequence-
tagged sites (STSs). Theor Appl Genet 86:985–990
Tsertsvadze N (1989) Sakartvelos kulturuli vazis klasifikatsia
(Classification of Georgian cultivated grapevine). Sab-
chota Sakartvelo, Tbilisi (in Georgian)
Vavilov NI (1931) Dikie rodichi plodovikh dereviev Aziatskoi
chasti SSSR I Kvakaza I problema proiskhozhdenia
plodovikh dereveiev (Wild progenitors of the fruit trees of
Turkestan and the Caucasus and the problem of the origin
of the fruit trees). Bull App Bot Genet Plant Breed
26:85–134 (in Russian)
Vinogradov-Nikitin P (1929) Plodovie I pishevie derevia lesov
Zakavkazia (Fruit and nourishing trees in the forests of
Transcaucasia). Bull App Bot Genet Plant Breed 22:3–211
(in Russian)
Vouillamoz JF, McGovern PE, Ergul A, Stella-Grando M
(2006) Genetic characterization and relationships of
Genet Resour Crop Evol
123
traditional grape cultivars from Transcaucasia and Anato-
lia. Plant Genet Res 4:144–158
Wagner HW, Sefc KM (1999) IDENTITY 4.0. Centre for
Applied Genetics, University of Agricultural Sciences,
Vienna
Weir B, Cockerham CC (1984) Estimating F-statistics for the
analysis of population structure. Evolution 38:1358–1370
Zecca G, Grassi F (2013) RPB2 gene reveals phylodemographic
signal in wild and domesticated grapevine (Vitis vinifera).
J Syst Evol 51:205–211
Zohary D, Hopf M (2000) Domestication of plants in the Old
World. University Press, Oxford
Zohary D, Spiegel-Roy P (1975) Beginning of fruit growing in
the world. Science 187:319–327
Genet Resour Crop Evol
123