To what extent do wild apples in Kazakhstan retain their genetic integrity?

Abstract

Kazakhstan belongs to the center of origin of apple. Malus sieversii (Ledeb.) M. Roem., the ancestral progenitor of the cultivated apple is native to this region. Pressure on the natural habitats of this wild apple has been intensified due to agriculture, grazing, and urbanization in the last century. For decades, M. sieversii in Kazakhstan has been subjected to the “Red Book of the Kazakh SSR” and today, this species is threatened with extinction. Wild apple undergoes exceptional losses in habitats, and the risk for losing the genetic integrity becomes worse due to increasing cultivation of cultivated apples and frequently occurring crosspollination events. The present study was focused on the current state of M. sieversii in Kazakhstan, the level of its diversity, its genetic integrity, and the identification of regions where future activities for conservation will have a good chance of success. A total of 311 M. sieversii samples of 12 populations collected in the wild, 16 previously selected wild apple genotypes, and 50 grown cultivars were studied using 16 simple sequence repeat (SSR) markers for genetic analysis. The results suggest that the level of genetic diversity is high. The differentiation between the populations was low, although the within-population heterozygosity was relatively high. A significant number of hybrids (8–95%) between M. sieversii and cultivated apples were found suggesting frequent crop-to-wild gene flow. The percentage of pure wild apple genotypes was highest in Krutoe truct and Tauturgen. These sites should be taken into account for future in situ long-term preservation activities.

Full text

ORIGINAL ARTICLE
To what extent do wild apples in Kazakhstan retain their genetic
integrity?
Madina Y. Omasheva
1
&Henryk Flachowsky
2
&Natalya A. Ryabushkina
1
&
Alexandr S. Pozharskiy
1
&Nurbol N. Galiakparov
1
&Magda-Viola Hanke
2
Received: 7 June 2016 /Revised: 14 March 2017 /Accepted: 20 March 2017
#Springer-Verlag Berlin Heidelberg 2017
Abstract Kazakhstan belongs to the center of origin of apple.
Malus sieversii (Ledeb.) M. Roem., the ancestral progenitor of
the cultivated apple is native to this region. Pressure on the
natural habitats of this wild apple has been intensified due to
agriculture, grazing, and urbanization in the last century. For
decades, M. sieversii in Kazakhstan has been subjected to the
BRed Book of the Kazakh SSR^and today, this species is
threatened with extinction. Wild apple undergoes exceptional
losses in habitats, and the risk for losing the genetic integrity
becomes worse due to increasing cultivation of cultivated ap-
ples and frequently occurring crosspollination events. The
present study was focused on the current state of M. sieversii
in Kazakhstan, the level of its diversity, its genetic integrity,
and the identification of regions where future activities for
conservation will have a good chance of success. A total of
311 M. sieversii samples of 12 populations collected in the
wild, 16 previously selected wild apple genotypes, and 50
grown cultivars were studied using 16 simple sequence repeat
(SSR) markers for genetic analysis. The results suggest that the
level of genetic diversity is high. The differentiation between
the populations was low, although the within-population
heterozygosity was relatively high. A significant number of
hybrids (8–95%) between M. sieversii and cultivated apples
were found suggesting frequent crop-to-wild gene flow. The
percentage of pure wild apple genotypes was highest in Krutoe
truct and Tauturgen. These sites should be taken into account
for future in situ long-term preservation activities.
Keywords Malus sieversii .Microsatellite markers .Genetic
diversity .Population structure
Introduction
Kazakhstan covers the northern region of the South Western
Asian Centre of origin according to N. Vavilov where a number
of plants, like the cultivated apple (Malus domestica Borkh.),
were domesticated (Vavilov 1931). Malus sieversii (Ledeb.) M.
Roem., the ancestral progenitor of modern apple cultivars, is
native to this region. Natural habitats of this wild apple species
can be found in Kazakhstan, Kyrgyzstan, Tajikistan,
Uzbekistan, and the western part of China (Harris et al. 2002;
Vel a s c o et a l . 2010; Cornille et al. 2014). M. sieversii was first
discovered in Kazakhstan in 1796 by Johann August Carl
Sievers in the Tarbagatai Mountains and later described by
the German botanist Carl F. von Ledebour in 1830
(Dzhangaliev 1977). Trees of M. sieversii are 2 to 10 m (some-
times14m)tallwithyellowishgreen fruits tinged with red.
Fruits of M. sieversii have a globose or depressed globose shape
and a size of 3 to 4.5 cm (sometimes up to 7 cm) in diameter
(Cuizhi and Spongberg 2003). Its close genetic relationship to
the cultivated apple (Harris et al. 2002; Velasco et al. 2010),
together with some special traits like fruit size, taste, resistance
to apple scab, fire blight, drought, and numerous soil pathogens
(Ignatov and Bodishevskaya 2011;Pons2006) make this spe-
cies interesting for fruit breeders and botanists.
Communicated by E. Dirlewanger
Electronic supplementary material The online version of this article
(doi:10.1007/s11295-017-1134-z) contains supplementary material,
which is available to authorized users.
*Magda-Viola Hanke
viola.hanke@julius-kuehn.de
1
Institute of Plant Biology and Biotechnology, Timiriazev 45,
Almaty, Kazakhstan 05040
2
Julius Kühn-Institute, Federal Research Institute for Cultivated
Plants, Institute for Breeding Research on Fruit Crops, Pillnitzer Platz
3a, 01326 Dresden, Germany
Tree Genetics & Genomes (2017) 13:52
DOI 10.1007/s11295-017-1134-z

Although this species was neglected in the past, it has al-
ways contributed significantly to human’s nutrition. Towards
the end of the twentieth century and the beginning of the
twenty-first century, the international scientific community
started to pay more attention to M. sieversii. This attention
was a result of the activities of the Kazakh scientist, Aimak
D. Dzhangaliev, who carried out inventory of this species in
its natural habitats. He also described numerous apple trees
growing naturally in the mountains on northern, north-eastern,
and north-western slopes at 800 to 1500 m, and in some loca-
tions at up to 2000 m above sea level. He discovered trees of
M. sieversii mainly within the low forest belt, in forest
meadows, in forest steppes, and in steppe belts of deciduous
forest associations. Most of these trees were grown at forest
borders and in shrub communities along the mountainous
ridges extending from eastern through south-eastern to south-
ern Kazakhstan (Dzhangaliev 1977,2003).
Furthermore, Aimak D. Dzhangaliev described the high
level of intraspecific diversity of M. sieversii connected with
mountain regions differing in their geological history, their
geographical structure, and their natural conditions. Based
on observations in the twentieth century, a number of different
mountain floristic regions were identified in Kazakhstan
where wild apples occur naturally (Turehanova 2012).
However, the high level of variation between individual
trees of wild-grown apples in Kazakhstan has led to several
debates regarding the exact number of different species in the
past. Six different apple species were described by Bykov
(1961), whereas other authors suggested two or three species
with unclear species boundaries. Dzhangaliev defined the
number of species as three and assigned all individuals (with
exception of a few aliens) to Malus kirghisorum,Malus
niedzwetzkyana,andM. sieversii as the most prominent one
in Kazakhstan (Dzhangaliev 1977,2003). In 2009, Gayle M.
Volk assigned wild apples collected in Kyrgyzstan to a com-
mon genetic lineage with M. sieversii individuals from the
Karatau Mountain range of Kazakhstan (Volk et al. 2009).
Based on genetic fingerprint information, it became obvious
that M. sieversii is a highly diverse species with a range in
genetic and phenotypic trait variations (Volk et al. 2013).
Pressure on the wild apple forests in Kazakhstan has inten-
sified for many decades due to urbanization, agriculture, graz-
ing, and wood harvesting. In the late nineteenth century, many
wild fruit trees in the foothills of Zailiysky Alatau were
grubbed by local people and replaced by agricultural crops
(Severtsov 1873). Between 1932 and 1967, wild apple trees
were often used as rootstocks and cultivated apples were
grafted on their top in the forest. As a consequence, the natural
forest orchard systems lost species-specific dynamic features
and genetic integrity. Existing wood stands of apple were not
destroyed permanently, but their ability for natural renewal
was significantly reduced (Dzhangaliev 2003). In 1981,
M. sieversii and M. niedzwetzkyana were included in the
BRedBookofKazakhSSR.^Nevertheless, the progressive
reduction in the number of wild apple representatives contin-
ued and as a result, the natural habitat of M. sieversii has
declined by over 70% in Kazakhstan during the last 30 years
(Eastwood et al. 2009). This bad situation has been worsened
by the adoption of the new Land Code in 2003 which ap-
proves the private property on land. Nowadays, M. sieversii
is a species with continuously declining number of individuals
(Ivashenko 2005) thus threatened with extinction in its natural
environment (Eastwood et al. 2009).
The work presented here was initiated to study the current
status of Kazakhstan wild apple populations. Three hundred
eleven individual trees of 12 natural M. sieversii populations
belonging to three geographical areas were evaluated on their
genetic diversity and integrity using a set of 16 simple se-
quence repeat (SSR) markers. The populations were collected
in a geographical range covering the eastern, south-eastern,
and southern mountainous regions of Kazakhstan. A genetic
population structure analysis revealed the presence of genetic
structures across all populations regardless of existing dis-
tances between their geographical locations, but also a high
level of admixture with M. domestica in some populations.
The relatively high number of hybrids in some populations,
which results most probably from crosspollination events be-
tween M. sieversii and M. domestica in regions of spatial
proximity between both species, underscores the alarming sit-
uation of the ongoing fragmentation of wild species popula-
tions in their natural habitats.
Materials and methods
Plant material
Leaf materials of 311 M. sieversii trees were collected at 12
different sites assorted as three different geographical groups
in Kazakhstan (Fig. 1, Table 1). In large populations (≥30
individuals), 25–40 individuals were selected at random with
regards to the age of the trees. The approximate age of the
trees was determined using a conventional method used in
forest taxation (Forest encyclopedia 1986). In each region,
the mean value of annual increments of randomly selected
trees was estimated and then the age was determined by the
diameter of tree trunk at a height of 1.3 m above soil level. In
small populations, mostly all individuals were selected. In
addition, 16 apple clones were collected in Tarbagatay (five
clones), Dzhungarskiy Alatau (nine clones), and Zailiysky
Alatau (two clones) and included in this study. These clones
were selected in 1990 by A. D. Dzhangaliev based on their
superior fruit size and other parameters with interest for breed-
ing. Finally, 50 apple cultivars being grown in Kazakhstan
were included as references to estimate the level of admixture.
52 Page 2 of 12 Tree Genetics & Genomes (2017) 13:52

Genomic DNA extraction
Genomic DNAwas extracted from 100 mg of fresh leaf tissue
as described by Aubakirova et al. (2014). Quality was tested
by 1% agarose gel electrophoresis and the NanoDrop 2000
(Thermo Scientific, USA) was used for calculating the con-
centration. DNA was diluted to 30 ng/μL.
SSR marker analysis
Sixteen SSR markers namely GD12, GD96, GD142, GD147,
GD162, GD100, CH01h10, CH01h01, CH04c07, Hi02c07,
CH01f03b, CH02d08, CH02c11, CH04e05, CH01f02 and
CH02c09 (Hokanson et al. 1998;Hemmatetal.2003;
Liebhard et al. 2002;Richardsetal.2009a)wereused.
These markers, which are distributed across the apple genome,
were chosen due to their use in similar studies (e.g., Richards
et al. 2009a) and their belonging to the marker set suggested
for genotyping of apple genetic resources by the European
Cooperative Programme for Plant Genetic Resources
(ECPGR). For example, CH04e05, CH02c11, CH02c09,
CH02d08, CH04c07, CH01h01, Hi02c07, and CH01h10 be-
long to the priority group 1 of the ECPGR marker set whereas
CH01f02, CH01f03b, GD12, and GD147 belong to priority
group 2. Markers GD12, GD96, GD142, GD147, GD162, and
GD100 were used in combination with three universal tail
oligos D8S1132 (VIC), D12S1090 (NED), and DYS437
(FAM) as described by Missiaggia and Grattapaglia (2006).
PCR was performed using a Mastercycler Pro S thermocycler
(Eppendorf, Hamburg, Germany). A first-step amplification
was carried out in a volume of 20 μl, containing 20 ng geno-
mic DNA, 1× Taq buffer (750 мМ Tris HCl, pH 8.8, 200 мМ
(NH
4
)
2
SO
4
, 0.1% Tween 20), 2.5 мМ MgCl
2
,0.2мМ dNTPs,
0.2 мМ of each of the respective SSR primers, and 1 unit Tag
polymerase (Thermo Scientific, USA). After denaturation at
94 °Сfor 2 min, seven PCR cycles were performed with 1 min
at 94 °С,2minat60°С, and 2 min at 72 °С.Subsequently,20
cycles with 1 min at 94 °С,2minat54°С,2minat72°С,and
final extension of 10 min at 72 °Сwere performed. Five mi-
croliters of the PCR products were checked on a 1% agarose
gel. The remaining products were tenfold diluted with ddH
2
O.
One-microliter diluted PCR product was used in a second
PCR where the tail primers were used instead of the forward
primers. The same PCR program described previously was
used. After a final extension of 30 min, the products of all
six standard markers per sample were combined and mixed
with formamide and size standard 500 LIZ (Applied
Biosystems, Foster City, USA). Hence, in a total volume of
10 μl, fluorescent dye VIC was diluted 540-fold, NED 120-
fold, and FAM 360-fold. Samples were analyzed on a capil-
lary sequencer ABI Prizm 310 (Applied Biosystems, Foster
City, CA, USA). Data were processed by GeneMapper
Software 4.0 (Applied Biosystems, Foster City, CA, USA).
The Type-it® Microsatellite PCR Kit (Qiagen GmbH,
Hilden, Germany) was used for PCR reactions on markers
CH01h10, CH01h01, CH04c07, Hi02c07, CH01f03b,
CH02d08, CH02c11, CH04e05, CH01f02, and CH02c09.
Markers were divided into four multiplex groups and each
PCR reaction contained 10 ng genomic DNA, 1× PCR
Master Mix, 0.5 μM of each primer and 0.5× Q-Solution in
Fig. 1 Map of Kazakhstan with geographical distribution of the 12 sites
for collecting Malus sieversii samples examined in this study. Sites of
M. sieversii populations were marked by differently colored triangles.All
populations marked with blue triangles belong to the geographical region
I. Populations marked with green triangles belong to the geographical
region II, whereas populations with pink triangles belong to III. TB
Tar bag at a y, CR Chernoff River, KT Krutoe truct, LRB Lepsy right bank,
KM Ketmen, UK Uryukty, TT Tauturgen, AR Almaty reserve, BB
Belbulak, GA Great Almaty gorge, BG Bozturgay gorge, AD Aksu
Dzhabagly
Tree Genetics & Genomes (2017) 13:52 Page 3 of 12 52

Tab l e 1 Physical description of the 12 sites in Kazakhstan, where samples of M. sieversii were collected
Geographical
group
Region FR Sampling site (population) Latitude Longitude Elevation
(m)
Habitat Number Number of trees of different age
Up to
20 years
20–
50 years
Older than
50 years
I Tarbagatay 23 Tarbagatay (TB) 47° 16,830 81° 36,150 1015 Shrub meadow on a slope,
south-eastern exposure
34 23 9 2
Dzhungarskyi Alatau 24 Chernoff river (CR) 45° 31,275 80° 42,830 1240 Apple grove on the southern
slopes and terraces
30 6 0 24
Krutoe truct (KT) 45° 33,130 80° 43,840 1515 Forest-meadow border on the
slope of the watershed,
western exposure
28 8 15 5
Lepsy right bank (LRB) 45° 32,955 80° 41,830 1160 Apple grove on the western
slope alongside a pasture
on the hillside
22 4 17 1
Great Kirgizsay 25a Ketmen (KM) 43° 19,195 79° 30,900 1440 Single apple trees on the
bottom of the gorge
70 4 3
Chu-Ili Mountains 26 Uryukty (UK) 43° 02,318 75° 09,637 1240 Single apple trees among
small trees and shrubs
(mini-ecosystem tugai)
60 4 2
II Zailiysky Alatau 25 Tauturgen (Kuznetsov cleft) (TT) 43° 21,460 77° 40,355 1585 Apple forest-meadow on the
northern slopes
40 10 11 19
Almaty reserve (right Talgar) (AR) 43° 13,720 77° 16,880 1560 Apple grove on the terrace
slope, north-eastern
exposure
48 18 20 10
Belbulak (BB) 43° 16,275 77° 10,255 1255 Northern exposure of the
forest slope
30 5 6 19
Great Almaty gorge (GA) 43° 09,135 76° 54,690 1320 Apple grove on the northern
slope exposure
30 0 7 23
III Karatau 28 Bozturgay gorge (BG) 42° 40,570 70° 15,630 845 At the floodplain among
small trees and shrubs
(mini-ecosystem tugai)
17 10 7 0
Talas Alatau 29 Aksu Dzhabagly (AD) 42° 19,545 70° 22,270 1365 Apple grove on the northern
exposure slope
19 8 10 1
Collection sites of this study were in accordance to Dzhangaliev 1977)
FR floristic region, Nsample size
52 Page 4 of 12 Tree Genetics & Genomes (2017) 13:52

a total volume of 10 μl. After an initial denaturation at 95 °C
for 5 min, 28 cycles with 1 min at 95 °C, 1.5 min at 60 °C, and
0.5 min at 72 °C were performed using the same PCR machine
as mentioned before. After a final extension at 60 °C for
30 min, the PCR products were diluted in a 1:20 or 1:30 ratio
with ddH
2
O depending on the multiplex type. Samples for
separation consisted of 24.9 μl GenomeLab™Sample
Loading Solution, 0.1 μlGenomeLab™DNA size standard-
400, 1 μl diluted PCR product, and a drop of mineral oil. The
GenomeLab™GeXP software (Beckman Coulter GmbH,
Krefeld, Germany) was used to calculate size of PCR
fragments.
Statistical analysis
Conversion factor for each SSR marker was estimated with
the TANDEM v1.09 software program (Matschiner and
Salzburger 2009). The numbers of different alleles (N
a
)and
their frequency, the number of effective alleles (N
e
), number
of private alleles (N
o
), expected and observed heterozygosity
(H
e
;H
o
), the inbreeding coefficient within individuals relative
to the subpopulation (F
IS
), the inbreeding coefficient within
subpopulations relative to the total as a measure for genetic
differentiation (F
ST
), the Shannon-Weaver Index of ecology
(I), and Nei’s genetic distance were calculated with GenAlEx
6.5 (Peakall and Smouse 2012). Polymorphism information
content (PIC) was estimated according to Botstein et al.
(1980) using the Excel Microsatellite Toolkit. Allelic richness
(Rs) and private allele richness (PR) were calculated by the
rarefaction method with HP-RARE software (Kalinowski
2005).
Cluster analysis, genetic distance analysis, and principal
coordinates analysis (PCoA) were performed using R ver.
3.2.2 (R Development Core Team 2008) with additional pack-
ages Bgeosphere^,Becodist^(Goslee and Urban 2007),
Bpoppr^(Kamvar et al. 2015), Bggplot2^(Wickham 2009),
Badegenet^(Jombart 2008), Bade4^(Dray and Dufour 2007)
(dependencies for Bpoppr^), and Bape^(Paradis et al. 2004).
Mantel test was conducted in order to evaluate correlation of
genetic and geographic distances between populations. A geo-
graphic distance matrix was calculated using coordinate data
from Table 1by geosphere::distm function. Mantel test was
performed using ecodist::mantel function with 10,000 permu-
tations (the level of confidence limits was set as 0.95; confi-
dence limits were estimated with 10,000 iterations of
bootstrap).
Population structure and hybridization analysis
For genetic structure analysis, a Bayesian model-based ap-
proach was used as implemented in STRUCTURE 2.3.4.
Software (Pritchard et al. 2000). The admixture model was
used as recommended by Hubisz et al. (2009). The burning
period and MCMC were consisted of 100,000 iterations each.
Kvalues in a range from 2 to 10 were tested in 20 independent
runs, and the method by Evanno et al. implemented in
STRUCTURE HARVESTER was used to find the most ap-
propriate Kvalue (Evanno et al. 2005; Earl and von Holdt
2012). Subsequently, 100 runs for each Kfrom 2 to 6 were
performed and aligned by CLUMPP 1.2.2. (Jakobsson and
Noah 2007). DISTRUCT 1.1 was used for visualization of
the results (Rosenberg 2004). Individuals with a membership
coefficient to the wild gene pool larger than 0.9 were consid-
ered as pure wild enabling to estimate the proportion of pure
wild individuals in each population (Cornille et al. 2015).
Results
Overall frequencies of the alleles detected using SSR
markers
A set of 16 SSR markers was used to evaluate 311 wild apple
samples of 12 natural M. sieversii populations and 50 selected
cultivars of M. domestica. All markers were found to be poly-
morphic with 9 to 24 or 9 to 18 alleles per locus for wild
apples and cultivars, respectively (Table S1). The PIC values
of the markers were always >0.6, except for marker CH02d08.
This marker showed a lower PIC value (0.37) for the wild
apples. Thus, all markers were informative as also shown by
the Shannon’sindex(I) ranging from 0.86 for marker
CH02d08 to 2.45 for marker GD142. The 16 SSR markers
amplified a total of 259 alleles for 311 wild apples, and 203
alleles for 50 cultivars. CH01f02 was the most variable marker
with 24 and 18 alleles and CH02d08 was the least variable
one.
Wild apples represent a unique genetic diversity
The 311 wild apple samples represent a unique genetic diver-
sity. Each individual could be assigned to a unique genotype.
No duplicates were found, which could have originated from
clonal propagation via root suckers. Triploids were also not
detected.
Overall genetic diversity
The total number of alleles (N
a
) was noticeably higher in
comparison to the number of effective alleles (N
e
)leadingto
the conclusion that only a few alleles contributed to the diver-
sity. This was the case for wild apples, but also for cultivars.
The values for H
o
and H
e
were relatively high suggesting a
good level of diversity within the collected plant material.
Tree Genetics & Genomes (2017) 13:52 Page 5 of 12 52

Genetic diversity within the populations
Mean numbers of 16.2 and 12.7 alleles per locus and 6.3 and
9.1 alleles per population were found for the wild apples and
cultivars, respectively (Tables S1 and S2). The highest (9.12)
and lowest (5.0) mean number of alleles per locus was found
for cultivars followed by the Bozturgay gorge (BG) and Aksu
Dzhabagly (AD) wild apple populations, respectively
(Table S2). The mean number of effective alleles per locus
ranged between 5.7 for the BG population and 2.3 for the
AD population. The private allelic richness was maximal in
cultivars and minimal in the Tauturgen (TT) population
(Table S2). For most of the populations, H
o
was slightly less
than H
e
except for the cultivars and the AD apple population.
The inbreeding coefficient (F
IS
) was always low. The highest
F
IS
(0.1) was detected in the Almaty reserve (AR) population.
F
IS
was slightly negative in the cultivars and in the AD pop-
ulation. The Ketmen (KM) and Uryukty (UK) populations
were not taken into account because of their small size.
Allelic richness
Allelic richness was analyzed per locus and population for
standardized number of individuals in each population
(Table 2). The highest allelic richness (13.23) was found with
marker CH04e05 for the GA population. Among the wild
apple populations, the highest values were detected for the
BG population at the loci CH01f02, GD96, GD142, and
GD147. The lowest allelic richness was in the AD population
at 9 of the 16 loci with a minimal mean per site of 4.8.
Although the mean allelic richness was comparable for the
geographical groups I, II, and III (Table 2), there were differ-
ences at individual loci. In M. domestica compared to the
groups, allelic richness at locus CH01f02 was maximal with
16.18 and minimal at locus CH01h10 with 7.71.
Population structure analysis
STRUCTURE analysis revealed the presence of genetic struc-
tures across populations of the three geographical groups (I, II,
and III). The most likely number of clusters (K) was evaluated
considering the plateau criterion (Fig. 2a) and the ΔKmethod
with the highest ΔKvalue observed at K= 2 (Fig. 2b). The
estimated population structure inferred from this analysis is
shown in Fig. 2c. The highest ΔKvalue (K=2)reflectsthe
presence of two clusters in the inferred population structure
analysis. Each cluster clearly corresponds to M. sieversii
(green cluster) and M. domestica (red cluster) representatives,
respectively (while not taking clones selected by Dzhangaliev
in natural populations into account).
Populations located in the floristic regions 23 and 24
(eastern Kazakhstan, Table 1) were not separated from the
TT and AR populations of the south-eastern mountain system
of Zaylisky Alatau. However, there was high admixture with
M. domestica in Lepsy right bank (LRB) as the most eastern
population and Belbulak (BB), Great Almaty gorge (GA), and
BG as the south-eastern populations. The red cluster included
M. domestica, the AD population, and Dzhangaliev’sapple
clones with some admixture with M. sieversii.AtK=3,pop-
ulations located in the floristic regions 23 and 24 were clearly
separated (blue cluster) from the populations of the floristic
regions 25 (green cluster). AD was admixed. At K=4,theAD
populationformed its own genetic pool, whereas the UK, KM,
and BG populations were admixed. At K= 5, 6, and more,
there were no new clusters formed.
All samples of M. domestica were classified correctly and
did not show admixture at any Kchecked, except the cultivar
BNiedzwiecki^(MD33 Fig. S2). Niedzwiecki is known as a
hybrid of M. sieversii f. niedzwetzkyana and M. domestica.Its
hybrid nature is clearly supported by the STRUCTURE
analysis.
Genetic distance analysis among populations
To evaluate genetic distance, genetic differentiation and struc-
turing among populations the Nei’s minimum distance and
pairwise F
ST
were calculated (Table 3). Relationships among
populations were illustrated by a dendrogram and a PCoA plot
(Fig. 3). Based on Fig. 3a and Table 3, the populations of the
geographical group I collected in Krutoe truct (KT), Chernoff
River (CR), LRB, and Tarbagatay (TB) were very close to
each other. The same was found for the populations (GA,
BB, AR, and TT) of the geographical group II. According to
Nei’s algorithm, the genetic distance between groups I and II
was 0.24, between II and III 0.4, and between I and III 0.59.
The genetic distance was positively correlated with the geo-
graphic distance. It was also confirmed by Mantel test, that
showed a significant positive correlation of R=0.57
(p<0.01).
F
ST
ranged in group I from 0.02 (CR/LRB) to 0.06 (KT/
LRB). In group II, F
ST
values ranged between 0.01 (BB/GA)
and 0.04 (TT/GA), and in group III, it was 0.15. Among
groups I and II, F
ST
was 0.07; among II and III, 0.10; and
among I and III, 0.15.
The two-dimensional PCoA plot (Fig. 3b) separated the
311 samples into two distinct clusters with the eastern group
(geographical group I) and the south-eastern group (geograph-
ical group II). The third gene pool of the southern regions
(geographical group III) was less uniform. The possible cause
is the smaller number of samples.
Admixture with M. domestica
A unweighted pair group method with arithmetic mean
(UPGMA) dendrogram based on Nei’s genetic distances
was calculated from the dataset of 16 SSRs across the 311
52 Page 6 of 12 Tree Genetics & Genomes (2017) 13:52

wild apple genotypes, 50 cultivars, and 16 clones selected
by Dzhangaliev. The dendrogram is composed of two main
clusters (Fig. S1). Cluster 1 (samples MD35 to MD01) is
composed of 35 samples of M. domestica and the
Dzhangaliev clones DZh03, DZh05, DZh14, and DZh15.
Cluster 2 is divided into three sub-clusters. Sub-cluster 2.1
(samples MD32 to CR27) is composed of seven samples of
M. domestica, six Dzhangaliev clones, and 19 samples of
M. sieversii. Sub-cluster 2.2 (samples Dzh08 to KT26) con-
tains four Dzhangaliev clones and three samples of
M. sieversii. Sub-cluster 2.3 is further divided into two
sub-sub-clusters. Sub-sub-cluster 2.3.1 (samples GA20 to
AR45) are three samples of M. sieversii and the
M. domestica sample MD38. Sub-sub-cluster 2.3.2 is divid-
ed into two sub-sub-sub-clusters with sub-sub-sub-cluster
2.3.2.1 (samples MD18 to AD12) containing the remaining
six samples of M. domestica (except of MD33), two
Dzhangaliev clones and 58 samples of M. sieversii.Sub-
sub-sub-cluster 2.3.2.2 (sample KM04 to KT02) is the larg-
est cluster. It contains all the remaining samples of
M. sieversii. In this cluster, there is no Dzhangaliev clone
andnosampleofM. domestica grouped, except of the cul-
tivar Niedzwiecki (sample MD33).
Several individuals of the M. sieversii populations showed
a variable level of admixture with M. domestica (Fig. 2c,S2)
ranging from nearly no admixture to almost 100%. Nearly no
admixture (membership coefficient >0.9) was found in the TT
(92% pure wild apples) and KT (89% pure wild apples) pop-
ulations located at altitudes of 1585 and 1515 m above sea
level, respectively. The percentages of pure wild genotypes
was slightly lower (83 and 85%) in the south-eastern AR
and the eastern TB populations located at altitudes of 1560
and 1015 m, respectively. In other south-eastern populations
admixture was much higher. For example, the percentage of
pure M. sieversii individuals was about 40% in the GA (alti-
tude 1320 m) and 47% in the BB (altitude 1255 m) popula-
tions, but nearly all individuals of the BG and AD populations
(southern site) were admixed. No correlation was found be-
tween the altitude and the level of admixture, but there was a
tendency for admixture increasing from the eastern to the
southern sites. No correlation was found between the level
of admixture and the mean tree age per population. For exam-
ple, in TB (two third of the trees aged up to 20 years) 85%, in
CR (four fifths aged over 50 years) 67%, in LRB (four fifths
aged 20–50 years) 36%, and in TT (~50% aged over 50 years)
92% of the individuals were pure M. sieversii genotypes. Too
Tabl e 2 Allelic richness
Site
a
TB CR KT LRB TT AR BB GA BG AD Groups
b
M. ×domestica
Locus I II III I II III
CH01h10 3.75 7.26 4.49 6.32 5.09 7.05 5.94 5.05 8.00 5.57 7.03 7.74 9.86 7.71
CH01h01 7.60 7.63 5.38 6.63 6.58 7.85 7.73 8.46 6.00 4.77 8.90 10.1 6.99 8.31
CH04c07 5.25 6.83 2.61 2.95 4.10 6.79 7.22 5.45 8.00 7.46 6.96 8.041 8.94 9.61
Hi02c07 5.51 5.95 5.04 5.94 5.65 6.12 7.72 6.46 7.00 4.78 6.53 8.18 7.94 9.41
CH01f03b 5.63 7.91 8.88 8.44 5.65 7.09 8.90 7.16 9.00 2.89 8.89 9.34 8.94 11.4
CH02d08 2.25 4.87 1.61 5.27 3.87 4.04 4.38 4.13 6.00 3.68 5.06 4.91 5.97 8.67
CH02c11 8.39 7.30 4.55 7.32 5.78 6.59 8.44 8.44 10.0 4.68 9.33 9.11 9.99 12.4
CH04e05 10.6 7.64 5.82 7.76 8.94 10.5 9.19 13.2 11.0 3.89 11.7 13.9 10.9 13.9
CH01f02 6.01 9.42 8.32 10.3 7.95 8.60 10.9 9.35 12.0 5.68 12.0 12.1 11.9 16.2
CH02c09 6.42 8.5 1 6.59 6.95 5.42 8.45 7.35 8.35 7.00 4.79 8.27 9.54 7.00 10.9
GD12 6.00 6.30 6.08 7.61 7.66 8.23 6.89 7.33 9.00 3.68 8.45 9.44 10.8 12.1
GD96 6.58 7.22 4.58 6.75 8.49 9.93 8.30 8.51 12.0 2.94 8.23 12.2 13.0 14.4
GD142 9.16 8.12 6.40 9.26 8.50 11.12 10.3 10.1 12.0 6.47 10.3 13.2 12.9 12.9
GD147 8.97 7.69 5.58 8.44 6.06 7.43 7.30 8 .21 12.0 5.79 9.66 9.17 13.9 11.0
GD162 6.29 9.03 6.56 10.9 7.19 11.2 10.4 11.0 6.00 5.68 11.2 13.6 7.89 13.2
GD100 3.75 4.70 2.98 4.99 3.00 5 .06 3.59 5.16 4.00 3.98 4.82 5.49 4.99 10.6
mean/site 6.39 7.27 5.34 7.24 6.28 7.88 7.79 7.90 8.69 4.80 8.59 9.77 9.51 11.4
N 34302822404830301719114148 3650
Allelic richness was measured per locus and site (upper part of the table) and per locus and cluster (lower part of the table)
TB Tar bag at a y, CR Chernoff River, KT Krutoe truct, LRB Lepsy right bank, TT Tauturgen, AR Almaty reserve, BB Belbulak, GA Great Almaty gorge, BG
Bozturgay gorge, AD Aksu Dzhabagly
a
Allelic richness standardized to 17 diploid individuals (34 genes)
b
Allelic richness standardized to 35 diploid individuals (70 genes)
Tree Genetics & Genomes (2017) 13:52 Page 7 of 12 52

small populations (KM and UK) were not taken into account.
Sixteen clones selected in wild populations by Dzhangaliev
showed different levels of admixture with M. domestica
(Fig. 1S,2S).
Discussion
Malus sieversii in natural habitats is increasingly threatened
by forest destruction, with its populations being restricted to
areas that have been rapidly decreasing in size over the last
decades (Zhang et al. 2007). Conservation of wild plant pop-
ulations in their natural habitats is becoming a major concern.
It requires detailed information about the occurrence and dis-
tribution of individual species, their genetic diversity, differ-
entiation, and level of admixture with other species. The
knowledge about these parameters is required for establishing
sustainable management and conservation strategies (Cornille
et al. 2015). The crabapple - M. sieversii (Ledeb.) M. Roem.,
is of particular interest as it is considered the main progenitor
Fig. 2 Genetic structure analysis using 311 samples of M. sieversii,16
clones selected by Dzhangaliev, and 50 cultivars of M. domestica.a
Estimation of likelihood of STRUCTURE. The mean values of the
logarithmic likelihood function L(K) and standard deviation from 20
runs for each value of K(number of clusters) = 1–10. bThe distribution
of ΔKover K=1–9withΔKas a ratio of the mean of [L(K+1)−2
(L(K)+L(K−1)] to the standard deviation of L(K). cPopulation structure
inference by Bayesian assignment with K=2–6. Each individual is
represented by a vertical line. Populations are separated by a vertical
black line.Different colors in the same line indicate the individual’s
admixture proportion (Qvalue) in Kclusters
52 Page 8 of 12 Tree Genetics & Genomes (2017) 13:52

of the cultivated apple (Cornille et al. 2014;Harrisetal.2002;
Ve l a s c o e t a l . 2010). This species has contributed substantially
to human’s nutrition for thousands of years and contains a
number of economically important traits urgently desired in
many international apple breeding programs (Isutsa and
Merwin 2000; Bassett et al. 2011;Wangetal.2012;
Janisiewicz et al. 2008). M. sieversii is not only native to
Kazakhstan but also to Kyrgyzstan, Tajikistan, Uzbekistan,
and China. The main part of Kazakhstan belongs to the Iran-
Turan floristic region (Takhtadzhian 1986), which is charac-
terized by its specific taxonomic, ecological and climatic con-
ditions, geology, soil, and a multifaceted topography with
large plains and hillocks, but also with high mountains of up
to 7000 m covering about 7% of the total area. Each mountain
ecosystem represents a separate floristic sub province which
provides a suitable environment for many wild plants like
M. sieversii (Flora of Kazakhstan 1956). Apple forests are
distributed in the low mountain zones (steppes) at 800–
1100 m, in middle mountains (forest steppes) at 1100–
1500 m, and in forest-meadow-steppes at 1500–1800 m
(Dzhangaliev 2003). Based on some changes in land use,
M. sieversii is currently threatened with extinction in
Kazakhstan (Eastwood et al. 2009). On this account, several
plant explorations were made to collect plant propagation ma-
terial (seeds, graft sticks) of M. sieversii in the center of its
origin in Kazakhstan, Tajikistan, and Uzbekistan. Four of
these explorations were sponsored by the National Plant
Germplasm System of the USDA-Agricultural Research
Service (Forsline et al. 2003). Between 1989 and 1996,
American scientists collected M. sieversii at eight sites located
Tabl e 3 Genetic distance analysis using Nei’s(1972) minimum distance (top diagonal) and pairwise F
st
comparisons (bottom diagonal) among 12
populations of M. sieversii
Population FR TB CR KT LRB TT AR BB GA BG AD
I TB 23 *** 0.11 0.11 0.11 0.28 0.27 0.30 0.35 0.46 0.88
CR 240.03*** 0.090.110.340.290.320.380.460.80
KT 24 0.05 0.04 *** 0.14 0.37 0.38 0.40 0.47 0.57 1.03
LRB 24 0.02 0.02 0.06 *** 0.30 0.26 0.21 0.28 0.37 0.80
II TT 25 0.08 0.09 0.14 0.07 *** 0.10 0.14 0.18 0.36 0.70
AR 25 0.07 0.08 0.13 0.06 0.02 *** 0.10 0.12 0.32 0.71
BB 25 0.07 0.07 0.14 0.04 0.03 0.02 *** 0.09 0.29 0.64
GA 25 0.09 0.09 0.15 0.05 0.04 0.02 0.01 *** 0.27 0.66
IIIBG 280.100.090.170.060.070.060.040.04*** 0.41
AD 29 0.27 0.24 0.34 0.23 0.22 0.21 0.19 0.20 0.15 ***
Among geographical groups
IIIIII
I *** 0.24 0.58
II 0.07 *** 0.40
III 0.150.10***
All values (both matrices) are significant after permutation test. Pvalues were adjusted for a nominal 5% level for multiple comparisons
FR floristic region, TB Tarbagatay, CR Chernoff River, KT Krutoe truct, LRB Lepsy right bank, TT Tauturgen, AR Almaty reserve, BB Belbulak, GA
Great Almaty gorge, BG Bozturgay gorge, AD Aksu Dzhabagly
Fig. 3 UPGMA cluster analysis
and principal coordinate analysis
for the 311 M. sieversii
genotypes. aDendrogram
obtained by UPGMA cluster
analysis based on Nei’s(1972)
genetic distances, bwith group
membership defined accordingly
to mountain ranges using
principal coordinate analysis
Tree Genetics & Genomes (2017) 13:52 Page 9 of 12 52

in six different regions in Kazakhstan (Tarbagatay,
Dzhungarskiy Alatay, Ketmen, Zailisky Alatau, Talas
Alatau, and Karatau). Seeds of mother trees were brought to
different international apple research institutions, where seed-
lings were planted and evaluated on their genetic variability
among others (Richards et al. 2009b;Volketal.2005,2013;
Gross et al. 2013).
Since changes in land use are still proceeding, the present
study aimed at investigating the current status of M. sieversii
in Kazakhstan. Therefore, leaves of 311 individual plants of
12 M. sieversii wild apple populations natively growing in the
eastern, south-eastern, and southern mountain regions of
Kazakhstan were analyzed using a set of 16 SSR markers.
All markers were polymorphic and informative with a mean
PIC value of 0.75. Observed and expected heterozygosity
were high suggesting a high level of diversity within the col-
lected genotypes. According to Nei’s algorithm, the present
analysis revealed that the collection sites in the eastern, south-
eastern, and southern mountainous regions of Kazakhstan are
well differentiated with a genetic distance of 0.24 between
sites I and II, 0.4 between II and III, and 0.59 between I and
III. Despite the existing differences (e.g., half-sib families ver-
sus random seedlings in natural habitats, 8 versus 12 popula-
tions located at close but not the same coordinates (Table S4),
a time interval of 25 years between both studies as well as 8
versus 16 microsatellite markers) between the experimental
features published by Richards et al. (2009a) and here, both
studies revealed high levels of Ho and He, different levels of
within-site variation, as well as unique alleles, which is con-
sistent with the previous analysis of M. sieversii populations.
The mean number of alleles per locus was high, but only
slightly larger compared to Richards et al. (2009a), although a
threefold number of trees was characterized in the present
work versus 88 mother trees previously.
The samples of M. sieversii showed a high level of genetic
diversity, and consistently low inbreeding coefficient. Genetic
differentiation was measured using the F
ST
index considering
the genetic differentiation between the 12 populations. Low
levels (F
ST
<0.05) of differentiation were detected between
the populations from eastern Kazakhstan (Tarbagatay (TB),
Chernoff River (CR), Krutoe truct (KT), and Lepsy right bank
(LRB)). Only between Krutoe truct and Ketmen (KM) was a
moderate differentiation found. The observed within-
population heterozygosity was between 0.62 and 0.71. A sim-
ilar situation was observed for populations of the south-
eastern region. The F
ST
index was always below 0.05 for
populations originating from the subprovince Zailiysky
Alatau. This suggested a low differentiation between the pop-
ulations, but relatively high within-population heterozygosity
(0.67–0.75). A similar situation with a level of differentiation
of 0.05 between sites was described by Richards et al. (2009a)
while studying plant materials collected about 25 years ago.
The level of differentiation between sites in the present study
is similar to the findings of Richards et al. (2009a).
Furthermore, the genetic distance was positively correlated
with the geographic distance. Hence, both studies provide a
good insight into the situation of M. sieversii at the time point
of investigation. Richards et al. (2009a)arguedthatfragmen-
tation and population isolation inthe eastern and south-eastern
areas of Kazakhstan may have occurred only recently from a
large ancestral population as indicated by high levels of ge-
netic diversity and the degree of genetic differentiation.
Results of the present study are also in accordance with opin-
ion of Cornille et al. (2012), that M. sieversii and M. domestica
form distinctly separated groups with no significant differ-
ences in levels of genetic variation.
A significant number of hybrids between M. sieversii and
M. domestica were detected for most of the populations rang-
ing from 8 to 95%. This relatively high level of admixture
suggests the occurrence of a frequent crop-to-wild gene flow.
Significant levels of crop-to-wild gene flow between
M. domestica and M. sieversii have previously been reported
(Gross et al. 2012, Cornille et al. 2013). High rates of intro-
gression suggest that hybrids are often viable and contribute to
interspecific gene flow to a similar degree. The number of
apple orchards surrounding wild populations and the total or-
chard area are positively related to recent introgression rates in
wild apple populations (Cornille et al. 2015). The present
situation of wild apple populations in Kazakhstan is the result
of both current and past incidences. It is clear that the areas in
the low mountain zones primarily in the piedmont-mountain
border zone possessed the largest anthropogenic impact. The
loss of biodiversity in these regions is due to the intensifica-
tion of agriculture and urbanization. Moreover, in the 1960s
and 1970s, the use of plots for private gardens in the moun-
tainous areas was permitted (e.g., in Zailiysky Alatau).
Plantings of cultivated apples in such private gardens in-
creased the risk for crop-to-wild gene flow. For example, be-
tween 1932 and 1967, wild apples were often used as root-
stocks directly at their place of growth and grafted with
M. domestica. This additionally increased the risk of out cross-
ing. Since these trees are still present everywhere in Kazakhstan,
they continuously contribute to the crop-to-wild gene flow and
increase the level of admixture in wild apple populations.
Evidence for this is the 16 M. sieversii clones collected by
Dzhangaliev collected in the wild, but which did not belong to
M. sieversii. Whether these clones are hybrids or cultivated ap-
ple cultivars misclassified by Dzhangalievisanentirelydifferent
debate. Nevertheless, based on their very low level of admixture
with M. sieversii, it could be assumed they belong to
M. domestica.
The high risk for crop-to-wild gene flow needs to be taken
into account if future sample collection, evaluation, utiliza-
tion, or conservation activities are planned (Gross et al.
2012). Especially pollen movement over long distances can
detrimentally affect the genetic integrity of wild species.
52 Page 10 of 12 Tree Genetics & Genomes (2017) 13:52

Pollen movement in apple can lead to successful pollination
events over distances of up to 10.7 km (Reim et al. 2015).
These authors showed that a decrease in tree density resulted
in an increase in pollen dispersal distances and the number of
hybridization events. Conservation measures leading to an
enhancement of the density of pure individuals will help to
reduce the likelihood of hybridization. This can be achieved
by repatriating plants from seed orchards, which were pro-
duced by controlled crossings between pure genotypes of
the species of interest (Reim et al. 2015). On one hand,
crop-to-wild gene flow is disadvantageous for the genetic in-
tegrity of wild species, but on the other hand, it could be a
driving force for evolution. Hybridization between different
species creates numerous new variations across genes and
gene combinations. This increases the chance that individual
genotypes with superior trait combinations will emerge, which
will be better adaptable to the rapidly changing climatic
conditions.
In conclusion, our results showed that nearly no admixture
was found at Krutoe truct and Tauturgen. Both sites are locat-
ed at altitudes of 1515 and 1585 m above sea level, respec-
tively. We therefore recommend these sites for future in situ
long-term preservation activities of mostly pure M. sieversii
populations in Kazakhstan.
Acknowledgements The research was funded by the Grant on the
subpriority BFundamental Studies in the Area of Natural Sciences,^
Budget Program 101 of the Scientific Committee of the Ministry of
Education and Science of the Republic of Kazakhstan (grant no.
1105/GF4). We gratefully acknowledge Dr. S.V. Chekalin for collecting
Malus sieversii samples and providing information on the wild apple
germplasm. We also acknowledge R. Gläß for her technical assistance
and Dr. F.O. Emeriewen for improving the English language of the
manuscript.
Data archiving statement All data are provided in Table S3.
References
Aubakirova K, Omasheva M, Ryabushkina N et al (2014) Evaluation of
five protocols for DNA extraction from leaves of Malus sieversii,
Vitis vinifera and Armeniaca vulgaris. Genet Mol Res 13(1):1278–
1287
Bassett CL, Glenn DM, Forsline PL, Wisniewski ME, Farrell RE Jr
(2011) Characterizing water use efficiency and water deficit re-
sponses in apple (Malus × domestica Borkh. and Malus sieversii
Ledeb.) M. Roem. HortSci 46(8):1079–1084
Botstein D, White RL, Skolnick M, Davis RW (1980) Construction of a
genetic linkage map in man using restriction fragment length poly-
morphisms. Am J Hum Genet 32:314–331
Bykov BA (1961) In: Pavlov NV (ed) Flora of Kazakhstan [in Russian].
Kazakh Academy of Science, Alma-Ata, Volume 4, pp 402–405
Cornille A, Gladieux P, Smulders MJM, Roldan-Ruiz I, Laurens F, Le
Cam B, Nersesyan A, Clavel J, Olonova M, Feugey L, Gabrielyan I,
Zhang X-G, Tenaillon MI, Giraud T (2012) New insight into the
history of domesticated apple: secondary contribution of the
European wild apple to the genome of cultivated varieties. PLoS
Genet 8(5):e1002703
Cornille A, Gladieux P, Giraud T (2013) Crop-to-wild gene flow and
spatial genetic structure in the closest wild relatives of the cultivated
apple. Evol Appl 6:737–748
Cornille A, Giraud T, Smulders MJ, Roldan-Ruiz I, Gladieux P (2014)
The domestication and evolutionary ecology of apples. Trends
Genet 30(2):57–65
Cornille A, Feurtey A, Gelin U, Ropars J, Misvanderbrugge K, Gladieux
P, Giraud T (2015) Anthropogenic and natural drivers of gene flow
in a temperate wild fruit tree: a basis for conservation and breeding
programs in apples. Evol Appl 8:373–384
Cuizhi G, Spongberg SA (2003) Malus Miller. In: Wu Z, Raven PH (eds)
Flora of China, volume 9. Science Press, Beijing, and Missouri
Botanical Garden Press, St. Louis, pp 179–189
Dray S, Dufour AB (2007) The ade4 package: implementing the duality
diagram for ecologists. J Stat Software 22(4):1–20
Dzhangaliev AD (1977) The wild apple tree of Kazakhstan. Nauka
Publishing House of Kazakh SSR, Alma Ata
Dzhangaliev AD (2003) The wild apple tree of Kazakhstan. Hortic Rev
29:63–285
Earl DA, von Holdt BM (2012) STRUCTURE HARVESTER: a website
and program for visualizing STRUCTURE output and
implementing the Evanno method. Conserv Genet Res 4(2):359–
361
Eastwood A, Lazkov G, Newton A (2009) The red list of trees of Central
Asia. Fauna & Flora International, Cambridge, ISBN: 9781 903703
27 4
Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters
of individuals using the software STRUCTURE: a simulation study.
MolEcol14:2611–2620
Flora of Kazakhstan (1956) Edited by Pavlov NV. KazSSR Academy of
Science Publishing, Аlma-Аtа,V.1,pp30–32
Forest encyclopedia (1986) Edited by Anuchin NP, Vorobiev GI. Soviet
Encyclopedia, USSR, V.2., pp 180–182
Forsline PL, Aldwinckle HS, Dickson EE, Luby JJ, Hokanson SC (2003)
Collection, maintenance, characterization, and utilization of wild
apples of Central Asia. Hortic Rev 29:1–62
Goslee SC, Urban DL (2007) The ecodist package for dissimilarity-based
analysis of ecological data. J Stat Software 22(7):1–19
Gross BL, Henk AD, Forsline PL, Richards CM, Volk GM (2012)
Identification of interspecific hybrids among domesticated apple
and its wild relatives. Tree Genet Genom 8(6):1223–1235
Gross BL, Volk GM, Richards CM, Henk AD, Forsline PL, Szewc-
Mcfadden AK, Fazio G, Chao CT (2013) Diversity captured in the
USDA-ARS National Plant Germplasm System apple core collec-
tion. J Am Soc Hortic Sci 138(5):375–381
Harris SA, Robinson JP, Juniper BE (2002) Genetic clues to the origin of
the apple. Trends Genet 18(8):426–430
Hemmat M, Weeden NF, Brown SK (2003) Mapping and evaluation of
Malus × domestica microsatellites in apple and pear. J Am Soc
Hortic Sci 128(4):515–520
Hokanson SC, Szewc-McFadden AK, Lamboy WF, McFerson GR
(1998) Microsatellite (SSR) markers reveal genetic identities, genet-
ic diversity and relationships in a Malus domestica Borkh. Core
subset collection. Theor Appl Genet 94:671–683
Hubisz MJ, Falush D, Stephens M, Pritchard JK (2009) Inferring weak
population structure with the assistance of sample group informa-
tion. Mol Ecol Res 9(5):1322–1332
Ignatov A, Bodishevskaya A (2011) Malus. In: Kole C (ed) Wild crop
relatives: genomic and breeding resources: temperate fruits.
Springer-Verlag, Berlin Heidelberg, pp 45–64
Isutsa DK, Merwin IA (2000) Malus germplasm varies in resistance or
tolerance to apple replant disease in a mixture of New York orchard
soils. HortSci 35(2):262–268
Tree Genetics & Genomes (2017) 13:52 Page 11 of 12 52

Ivashenko AA (2005) The treasures of flora in Kazakhstan. Through the
pages of the red Book. Almaty Kitap, Almaty, p 119
Jakobsson M, Noah A (2007) Rosenberg CLUMPP: a cluster matching
and permutation program for dealing with label switching and
multimodality in analysis of population structure. Bioinformatics
23(14):1801–1806
Janisiewicz WJ, Saftner RA, Conway WS, Forsline PL (2008)
Preliminary evaluation of apple germplasm from Kazakhstan for
resistance to postharvest blue mold in fruit caused by Penicillium
expansum. HortSci 2(43):420–426
Jombart T (2008) Adegenet: an R package for the multivariate analysis of
genetic markers. Bioinformatics 24:1403–1405
Kalinowski ST (2005) HP-Rare: a computer program for performing
rarefaction on measures of allelic diversity. Mol Ecol Notes 5:
187–189
Kamvar ZN, Brooks JC, Grünwald NJ (2015) Novel R tools for analysis
of genome-wide population genetic data with emphasis on clonality.
Front Genet 6:208
Liebhard R, Gianfranceschi L, Koller B, Ryder CD, Tarchini R, Van de
Weg WE, Gessler C (2002) Development and characterization of
140 new microsatellites in apple (Malus xdomestica Borkh.) Mol
Breed 10:217–241
Matschiner M, Salzburger W (2009) TANDEM: integrating automated
allele binning into genetics and genomics workflows.
Bioinformatics 25(15):1982–1983
Missiaggia A, Grattapaglia D (2006) Plant microsatellite genotyping with
4-color fluorescent detection using multiple-tailed primers. Genet
Mol Res 5:72–78
Nei M (1972) Genetic distance between populations. Am Nat 106(949):
283–292
Paradis E, Claude J, Strimmer K (2004) APE: analyses of phylogenetic
and evolution in R language. Bioinformatics 20:289–290
Peakall R, Smouse PE (2012) GenAlEx 6.5: genetic analysis in excel:
population genetic software for teaching and research update.
Bioinformatics 19:2537–2539
Pons L (2006) Remarkable Kazak apples. AgResearch Magazine 54(1):
4–6
Pritchard JK, Stephens M, Donnelly P (2000) Inference of population
structure using multilocus genotype data. Genetics 155:945–959
R Development Core Team (2008) R: A language and environment for
statistical computing. R Foundation for Statistical Computing,
Vienna, Austria. URL http://www.R-project.org
Reim S, Proft A, Heinz S, Lochschmidt F, Höfer M, Tröber U, Wolf H
(2015) Pollen movement in a Malus sylvestris population and
conclusions for conservation measures. Plant Genet Res 1–9. doi:
10.1017/S1479262115000301
Richards CM, Volk GM,Reilley AA, Henk AD, Lockwood DR, Reevers
PA, Forsline PL (2009a) Genetic diversity and population structure
in Malus sieversii, a wild progenitor species of domesticated apple.
Tree Genet Genomes 5:339–347
Richards CM, Volk GM, Reeves PA, Reilley AA, Henk AD, Forsline PL,
Aldwinckle HS (2009b) Selection of stratified core sets representing
wild apple (Malus sieversii). J Am Soc Hortic Sci 134(2):228–235
Rosenberg NA (2004) Distruct: a program for the graphical display of
population structure. Mol Ecol Notes 4:137–138
Severtsov NA (1873) Travel on Turkestan region and research of moun-
tain country Tien Shan. Accomplished on a Commission of Russian
Geographic Society, St. Petersburg, pp 82–83
Takhtadzhian AL (1986) Floristic regions of the world. University of
California Press, Berkley, ISBN-10: 0520040279, ISBN-13: 978–
0520040274
Turehanova RM (2012) Sievers apple tree in Kazakhstan, research and
selection of forms. Terrain 12(1):44–61
Vavilov NI (1931) The wild relatives of fruit trees of the Asian part of the
USSR and Caucasus and the problem of the origin of fruit trees.
Trans Appl Bot Gene Breed 26(3):132–134
Velasco R, Zharkikh A, Affourtit J et al (2010) The genome of the do-
mesticated apple (Malus × domestica Borkh.) Nature Genet 42:833–
839
Volk GM, Reilley AD, Henk PL, Forsline HS, Aldwinckle, Richards CM
(2005) Ex situ conservation of vegetatively-propagated species: de-
velopment of a seed-based core collection for Malus sieversii.JAm
Soc Hortic Sci 130:203–210
Volk GM, Richards CM, Henk AD, Reilley AD (2009) Novel diversity
identified in a wild apple population from the Kyrgyz Republic.
HortSci 44(2):516–518
Volk GM, Henk AD, Richards CM, Forsline PL, Chao CT (2013) Malus
sieversii: a diverse Central Asian apple species in the USDA-ARS
national plant germplasm system. HortSci 48:1440–1444
Wang A, Aldwinckle H, Forsline P, Main D, Fazio G, Brown S, Xu K
(2012) EST contig-based SSR linkage maps for Malus ×domestica
cv Royal Gala and an apple scab resistant accession of M. sieversii,
the progenitor species of domestic apple. Mol Breed 29:379–397
Wickham H (2009) ggplot2: elegant graphics for data analysis. Springer-
Ve r la g , N ew Yo r k
Zhang C, Chen X, He T, Liu X, Feng T, Yuan Z (2007) Genetic structure
of Malus sieversii population from Xinjiang, China, revealed by
SSR markers. J Genet Genomics 34:947–955
52 Page 12 of 12 Tree Genetics & Genomes (2017) 13:52