ArticlePDF Available

Abstract and Figures

With 3 figures and 3 tables The genetic relationships between 40 wild hazelnuts collected in northern Spain and cultivated hazelnuts, including 62 local selections and 14 reference cultivars, were investigated using 13 microsatellite loci. Microsatellite analysis revealed considerable diversity; 91 different alleles were identified with a mean of 7 per locus, and polymorphic information content values ranged from 0.43 to 0.83 with a mean of 0.69. The plot obtained from principal coordinate analysis, the unrooted neighbour-joining tree constructed, and the population structure analysis revealed a high level of differentiation between the locally cultivated forms and the remaining materials. Introgressions within these groups were detected in the three analyses. The results indicate that hazelnuts in northern Spain contain (i) a group of accessions clearly differentiated within the Spanish–Italian gene pool, (ii) a group with intermediate forms probably derived from hybridization and (iii) accessions probably derived through exchange with other geographical areas, especially north-eastern Spain. The presence of these different groups within the local cultivated hazelnut germplasm has consequences for its preservation and use.
Content may be subject to copyright.
Genetic relationship between cultivated and wild hazelnuts (Corylus avellana L.)
collected in northern Spain
A. C
,N. T
,E. P
,M. R
and J. J. F
´rea de Cultivos Hortofrutı
´colas y Forestales, Servicio Regional de Investigacio
´n y Desarrollo Agroalimentario (SERIDA),
33300 Villaviciosa, Asturias, Spain;
IRTA – Mas de Bover, Departament Olivicultura, Elaiote
`cnia i Fruita seca, Crta. Reus–El
Morell. Km, 3,8, 43120 Constantı
´, Tarragona, Spain;
Corresponding author, E-mail:
With 3 figures and 3 tables
Received March 24, 2010/Accepted October 8, 2010
Communicated by J. Le
The genetic relationships between 40 wild hazelnuts collected in
northern Spain and cultivated hazelnuts, including 62 local selections
and 14 reference cultivars, were investigated using 13 microsatellite
loci. Microsatellite analysis revealed considerable diversity; 91 different
alleles were identified with a mean of 7 per locus, and polymorphic
information content values ranged from 0.43 to 0.83 with a mean of
0.69. The plot obtained from principal coordinate analysis, the
unrooted neighbour-joining tree constructed, and the population
structure analysis revealed a high level of differentiation between the
locally cultivated forms and the remaining materials. Introgressions
within these groups were detected in the three analyses. The results
indicate that hazelnuts in northern Spain contain (i) a group of
accessions clearly differentiated within the Spanish–Italian gene pool,
(ii) a group with intermediate forms probably derived from hybrid-
ization and (iii) accessions probably derived through exchange with
other geographical areas, especially north-eastern Spain. The presence
of these different groups within the local cultivated hazelnut germ-
plasm has consequences for its preservation and use.
Key words: Corylus avellana L. — microsatellite — genetic
diversity — principal coordinate analysis — domestication
European hazelnut (Corylus avellana L.) is one of the main nut
crops in the world. Its native distribution extends from Asia
Minor, the Caucasus Mountains and Ural Mountains to
western Europe (British Isles, Scandinavian islands, and
Central and Southern Europe) and the Mediterranean coast
of North Africa (Thompson et al. 1996). However, in recent
years, the main production areas have been in Turkey, Italy,
the United States and Spain (FAOstat 2008). In Spain,
hazelnut is cultivated mainly in the north-east (Catalonia),
although there are other minor hazelnut-growing areas in the
north, such as Asturias (Tous et al. 2001).
The traditional characterization of many plants such as
hazelnut was based on morpho-agronomic traits (UPOV 1979,
Bioversity International 2008), which can be influenced by
environmental factors. Microsatellite markers or simple
sequence repeats (SSR) are considered robust markers, highly
reproducible and generally inherited as co-dominant loci not
affected by environmental factors. A set of 51 SSR markers
was recently developed in this species (Bassil et al. 2005a,b,
Boccacci et al. 2005) of which 30 were placed on a genetic map
(Mehlenbacher et al. 2006). SSR markers were used to analyse
genetic relationships among cultivars, cultivar fingerprinting
and the identification of synonyms in collections (Boccacci
et al. 2006, 2008, Go
¨kirmak et al. 2009, Gu
¨rcan et al. 2010).
The variation detected by SSR markers has also been used to
investigate the structure and organization of the genetic
diversity in this species. The variation detected at 21 SSR loci
in 270 European hazelnut accessions showed four major
geographical groups or gene pools (Go
¨kirmak et al. 2009):
Central European, Black Sea, English and Spanish–Italian.
Chloroplast microsatellite analysis indicated that hazelnut
genetic diversity was organized in three main groups with
different chloroplast type frequencies, suggesting that hazelnut
was domesticated in three main areas: the Mediterranean,
Turkey and Iran (Boccacci and Botta 2009). In many plant
species, microsatellite markers have been used to investigate
the structure and genetic relationships between wild and
cultivated forms (Grassi et al. 2003, Kwak and Gepts 2009).
The study of the relationship between wild and cultivated
forms in a geographical area can supply information about the
putative domestication events, the evolutionary relationships
or the gene flow between them (Gepts et al. 1986, Papa and
Gepts 2003, Kwak and Gepts 2009). Information about wild
forms may also be useful to breeders wishing to enlarge the
genetic base of their breeding programmes. However, there
is limited information concerning the relationship between
cultivated and wild forms in hazelnut.
In the past, hazelnut was an important crop in Asturias
(Alvarez Requejo 1965); however, at the end of the twentieth
century, the exodus of people from rural areas, together with
problems with hazelnut commercialization, led to abandon-
ment of local crops. In Asturias, wild and cultivated forms of
hazelnut coexist. Wild forms can be found along the banks of
streams or forming small woods in isolated areas. The
cultivated forms can be found in small orchards, gardens
and hedgerows, where they delimit the fields and produce fruit
and wood. A hazelnut germplasm exploration was carried out
in northern Spain (Asturias) over three consecutive years
(2003–2005), and as a result of this exploration, 90 cultivated
trees were selected from the region. The morphological
characterization of this local germplasm revealed a wide
phenotypic variation in nut and husk traits, and many
selections had characteristics appreciated by the market
(Ferreira et al. 2010). The genetic relationships among 50
local selections derived from the survey and 17 well-known
cultivars, including seven cultivars from north-eastern Spain,
were investigated using inter-simple sequence repeat (ISSR)
markers. The results indicated that the local accessions were
Plant Breeding doi:10.1111/j.1439-0523.2010.01835.x
2011 Blackwell Verlag GmbH
closely related, but relatively distant from the standard
cultivars of north-eastern Spain, Italy, Turkey and the United
States (Ferreira et al. 2010). These results could indicate local
domestication. In many regions, cultivated forms coexist with
wild forms, and it has been suggested that some local cultivars
were selected from local wild populations (Lagerstedt 1975,
Tasias Valls 1975, Thompson et al. 1996, Go
¨kirmak et al.
2009). In this study, the variation detected by SSR markers is
used to (i) verify the differentiation of hazelnuts cultivated in
Asturias from other hazelnut gene pools and (ii) investigate the
genetic relationships between cultivated and wild local hazel-
nuts. This information will be useful to identify local materials
for preservation in collections and to improve knowledge
concerning domestication events of this species.
Materials and Methods
Plant material: A total of 116 trees were analysed in this study
(Table 1): (i) 62 local forms cultivated by farmers, of which 58 were
collected in Asturias (Ferreira et al. 2010) and the remaining four are
local cultivars believed to have originated in Asturias (ÔCasinaÕ,
ÔAmandiÕ,ÔEspinaredoÕand ÔQuiro
´sÕ); (ii) 40 wild trees collected in
autumn of 2008 in the main areas of Asturias in which hazelnut wild
populations are present; and (iii) 14 well-known reference cultivars
representing the worldÕs most important production areas.
The local cultivars (ÔCasinaÕ,ÔAmandiÕ,ÔEspinaredoÕand ÔQuiro
and the fourteen reference cultivars are maintained in the hazelnut
collection of the Servicio Regional de Investigacio
´n y Desarrollo
Agroalimentario (SERIDA, Villaviciosa, Spain); part of the local
cultivated materials are also being incorporated into this collection.
The corresponding gene pool of the 14 reference cultivars, according to
¨kirmak et al. (2009), is shown in Table 1.
DNA extraction and microsatellite analysis: Genomic DNA was
extracted from young leaves or from immature catkins (only in wild
materials) according to Boccacci et al. (2006). DNA was quantified
using a spectrophotometer and maintained at )20C.
A total of 13 SSR markers were analysed: CAC-B029b, CAC-B109,
CAC-B101, CAC-A014b, CAC-C115, CAC-B010, Cat-B507, Cat-
B107, Cat-B50, Cat-B502, Cat-C504, Cat-B106 and Cat-A114 (Bassil
et al. 2005a, Boccacci et al. 2005). Of these markers, 10 have been
placed on the genetic map of hazelnut (Mehlenbacher et al. 2006). This
SSR set was selected after considering the pattern of bands produced
by the SSRs and the level of polymorphism previously reported (Bassil
et al. 2005b, Boccacci et al. 2005, 2006, Go
¨kirmak et al. 2009). PCR
amplification was performed in a volume of 20 ll containing 50 ng
DNA; 0.5 U Taq-DNA polymerase (BIOTAQTM DNA Polymerase;
BIOLINE, London, UK); 1·buffer; 2.5 m
/l; 200 l
each of
dATP, dCTP, dGTP and dTTP; and 0.5 l
each of forward and
reverse primers. The amplifications were carried out using the
programs described by the respective authors. PCR products were
resolved on 8% polyacrylamide gels, stained with SYBR Safe DNA gel
stain (Invitrogen, Eugene, OR, USA) and visualized under UV light. A
100-bp ladder (G.E. Healthcare Life Science, Fairfield, CT, USA) and
the software G
V4.01 (Syngene, Cambridge, UK) were used
to measure the size of the amplification products.
Data analysis: Alleles with molecular sizes in the range previously
described for each microsatellite were considered in this study.
Accessions with the same fingerprint were considered synonyms, and
only one accession in the group was used in subsequent analysis. The
software P
3.25 (Liu and Muse 2005) was used to
calculate the following parameters of genetic diversity: the number of
alleles per locus, observed heterozygosity (Ho), gene diversity (synon-
ymous with expected heterozygosity, He), estimated frequency of null
alleles (r) and the polymorphic information content (PIC). Ho is
calculated as the number of heterozygous genotypes, divided by the
number of total genotypes observed at the locus. He estimates the
probability that two alleles at any locus are different from each other.
The PIC value measures the polymorphism observed in a group of
genotypes at a specified locus.
From the variation provided by the 13 SSR markers, three types of
analysis were performed to group the accessions and investigate the
relationships and structure of the genetic diversity. Using the data
matrix for the presence or absence of each band (allele), a principal
coordinate analysis (PCoA) was performed with the software
V.2.1 (Rohlf 2002), and the two principal coordinates were
used to visualize the dispersion of the accessions in a graph. From a
data matrix that listed the alleles at each SSR marker locus, an
unrooted neighbour-joining tree (NJ) was constructed with the
software PowerMarker using the C.S. Chord distance (Cavalli-Sforza
and Edwards 1967), and the tree was visualized with the software
1.6.6 (Page 1996). Data were also analysed with
V2.2 software (Pritchard et al. 2000) to determine the number of
Table 1: Hazelnut materials inclu-
ded in the SSR markers analysis
Type of material Geographical origin
Material (accession
or cultivar) Gene pool
Local cultivated selections (58) Spain (Northern) As1 to As69
Local cultivars (4) Spain (Northern) Casina
Local wild germplasm (40) Spain (Northern) Astw1 to Astw40
Reference cultivars (14) Spain (North-eastern) Gironell
Italy Camponica
Tonda di Giffoni
Tonda Romana
Santa Maria del Gesu
Turkey Tombul Black Sea
England Daviana English
USA Butler
SSR, simple sequence repeats.
According to Go
¨kirmak et al. (2009).
2A. Campa, N. Trabanco, E. Pe
´rez-Vega et al.
divergent groups (K). Runs were carried out ten times for each K
value ranging from 2 to 4. Each run was performed using the
admixture model and 20 000 MCMC (Markov Chain Monte Carlo)
repetitions following a burn-in of 20 000 iterations during the
analysis, and the run showing the lowest likelihood value was
selected. The statistic DK (Evanno et al. 2005) was used to choose the
optimal Kvalue. The results supplied by this program were visualized
in a graphical bar plot of membership coefficients for each Kvalue.
software was also used to calculate F
values in the
populations that were eventually selected. F
measures genetic
differences among groups.
Genetic diversity
PCR products were obtained at all SSR loci for all cultivars.
Fingerprint data revealed that 82 of the 116 cultivars had
unique genotypes. Thirty-six trees had two fingerprints: a
group of 30 accessions named the Casina group (Casina,
Amandi, Quiro
´s, Espinaredo, As1, As2, As3, As4, As6, As7,
As8, As9, As17, As19, As20, As28, As32, As38, As43, As46,
As47, As51, As52, As53, As61, As64, As65, As67, As68 and
As69) and another group of six accessions named the As16
group (As16, As31, As41, As48, As49, As50). These two
groups only differed at the microsatellite Cat B502. Table 2
shows five genetic diversity parameters (number of alleles, He,
Ho, r and PIC) obtained per SSR locus in the analyses of the
82 unique accessions. The number of alleles totalled 91 and
ranged from four (Cat-A114) to 10 (CAC-B101) alleles per
locus, with an average of seven. He ranged from 0.47 (for
CAC-B010) to 0.85 (for CAC-B101), with an average of 0.73.
PIC values ranged from 0.43 (CAC-B010) to 0.83 (for CAC-
B101 and CAC-A014b), with an average of 0.69. The most
polymorphic loci were CAC-B101, CAC-A014b, CAC-C115
and CAC-B109, while the least were CAC-B010 and Cat-B106.
Four parameters of genetic diversity for each SSR locus in
the three groups of materials considered in this work (local
cultivated germplasm, local wild materials and reference
cultivars) are shown in Table 3. The group of local cultivated
germplasm generally had lower values for parameters He and
PIC. The values for the group of reference cultivars were
within the range of the local wild materials.
PCoA and cluster analysis
In the two-dimensional PCoA plot (Fig. 1), the first coordinate
explained 14.85% of the variation and the second coordinate
an additional 7.9%. The figure shows two main groups. The
majority of the local cultivated materials appear at the left
(Casina group, As16 group and 12 trees), and the remaining
accessions appear at the right. The reference cultivars appear
at the top right, along with several local cultivated accessions,
and at the top right appear several local accessions (As22, As
35, As36, As57, As58, As60 and As66) and four local wild trees
(AstW8, AstW13 and AstW14). At the bottom right appear
the majority of local wild materials and the locally cultivated
accession As55. Finally, four local cultivated accessions (As12,
As15, As30, and As63) and three local wild accessions
(AstW12, AstW17 and AstW25) had an intermediate position
between the three main groups described previously.
Figure 2 shows the NJ tree obtained. Three distinct main
branches can be observed. One branch (A in Fig. 2) was
formed by 53 locally cultivated hazelnuts (17 accessions and
the groups Casina and As16), four local wild accessions
Table 2: Genetic diversity parameters obtained per SSR locus in the
analyses of 82 unique genotypes
Locus No. of alleles He Ho rPIC
CAC-B029b 7 0.75 0.80 )0.03 0.72
Cat-B507 6 0.70 0.96 )0.16 0.65
Cat-B107 9 0.74 0.55 0.11 0.70
Cat-B504 6 0.73 0.94 )0.12 0.69
Cat-B502 6 0.76 0.92 )0.09 0.73
CAC-B109 8 0.82 0.78 0.02 0.80
CAC-B101 10 0.85 0.58 0.15 0.83
Cat-C504 6 0.70 0.44 0.16 0.67
Cat-B106 7 0.60 0.63 )0.01 0.56
CAC-A014b 8 0.84 0.88 )0.02 0.83
CAC-C115 9 0.83 0.80 0.01 0.80
CAC-B010 5 0.47 0.02 0.30 0.43
Cat-A114 4 0.72 0.63 0.05 0.66
Mean 7 0.73 0.68 0.05 0.69
SSR, simple sequence repeats; He, expected heterozygosity; Ho,
observed heterozygosity; r, estimated frequency of null alleles; PIC,
polymorphic information content.
Table 3: Summary statistics of microsatellites diversity in the three populations analysed in this study considering unique genotypes: reference
cultivars (Ref.), locally cultivated material (Cult.) and wild materials (Wild)
N. of Alleles He Ho PIC
Ref. Cult. Wild Ref. Cult. Wild Ref. Cult. Wild Ref. Cult. Wild
CAC-B029b 6 6 6 0.68 0.60 0.79 0.64 0.78 0.87 0.64 0.55 0.76
Cat-B507 5 3 6 0.75 0.55 0.71 0.93 1.00 0.95 0.71 0.45 0.67
Cat-B107 7 5 5 0.81 0.72 0.59 0.43 0.81 0.43 0.79 0.67 0.54
Cat-B504 6 6 6 0.74 0.64 0.69 0.64 1.00 1.00 0.71 0.58 0.64
Cat-B502 5 4 6 0.69 0.56 0.80 0.92 0.90 0.94 0.64 0.47 0.77
CAC-B109 7 7 8 0.76 0.78 0.71 0.93 0.93 0.63 0.72 0.75 0.68
CAC-B101 9 9 8 0.82 0.85 0.83 0.50 0.65 0.58 0.80 0.83 0.82
Cat-C504 6 5 5 0.69 0.49 0.77 0.43 0.31 0.54 0.66 0.47 0.74
Cat-B106 4 6 5 0.62 0.58 0.56 0.62 0.63 0.62 0.56 0.52 0.50
CAC-A014b 7 7 8 0.79 0.78 0.84 0.86 0.89 0.87 0.75 0.74 0.82
CAC-C115 4 6 9 0.67 0.70 0.79 0.86 1.00 0.68 0.60 0.65 0.76
CAC-B010 4 5 3 0.54 0.38 0.19 0.00 0.07 0.00 0.50 0.36 0.18
Cat-A114 4 4 4 0.72 0.62 0.66 0.79 0.78 0.5 0.67 0.55 0.60
Mean 5.7 5.6 6.08 0.71 0.63 0.69 0.66 0.75 0.66 0.67 0.58 0.65
He, expected heterozygosity; Ho, observed heterozygosity; PIC, polymorphic information content.
Genetic relationship of C. avellana 3
(AstW12, AstW17, AstW19 and AstW25), and the reference
cultivar ÔRibetÕ; a second branch (B) was constituted of 35
accessions, all classified as local wild material; and a third
branch (C) included 13 reference cultivars, nine local cultivated
hazelnuts (As30, As33, As35, As36, As57, As58, As60, As62
and As66) and the local wild accession AstW13. The local
accessions (wild or cultivated) were not grouped by collection
site (passport data are not shown).
Analysis of the population structure
The organization of genetic diversity was further analysed with
program in a stepwise fashion from 2 to 4
populations. The estimation of the statistic DK revealed the
highest value for K=2 (DK = 126.5) indicating two main
clusters were present: locally cultivated forms and the remain-
ing materials (Fig. 3). The F
value was 0.068 for these two
groups. Shown in the bar graph for K= 2 (Fig. 3 top) are
several accessions that are not clearly placed in a single group
(e.g. As35, As36, As58 and As55 in the local cultivated group
and AstW12, AstW17, AstW24 and AstW34 in the other
group), but rather contain genes from both groups. At K=3
(DK = 13.8),
revealed less differentiation between
the three groups (local cultivated germplasm, local wild
germplasm and reference cultivars). The F
values were
0.094 for local cultivated vs. reference cultivars, 0.091 for local
cultivars vs. local wild accessions and 0.107 for reference
cultivars vs. local wild accessions. Again, some local accessions
showed introgression from other groups. In the local cultivated
forms, As12 and As55 showed introgression from the local
wild germplasm, while trees As15, As22, As30, As58, As33,
As35, As36, As60, As62 and As66 showed a high similarity to
the reference cultivars. In the local wild population, individ-
uals AstW8, AstW13, AstW14, AstW26, AstW31 and AstW40
showed a high similarity to the reference cultivars, while
AstW12, AstW17, AstW25 and AstW34 showed a high
similarity to the local cultivars.
Microsatellite analysis showed considerable genetic diversity in
116 analysed hazelnuts, of which 82 had unique genotypes
(different patterns of bands). A total of 91 alleles were
identified, with high values for PIC and number of alleles.
The 13 SSR loci analysed in this study were previously used in
genetic diversity analysis (Bassil et al. 2005a, Boccacci et al.
2005, 2006, Go
¨kirmak et al. 2009) supplying high levels of
polymorphism (PIC > 0.70; number of alleles per locus >7).
In general, the values obtained in this study for these two
parameters were slightly lower than previously reported.
Among the three groups of materials included in this study,
the highest genetic diversity was in the well-known reference
cultivars and the local wild germplasm. The lower values for
the estimated parameters were generally in the group of local
cultivated germplasm. These results are consistent with the
different geographical origin of each group of materials and
with sexual or asexual reproduction. The group of reference
hazelnut cultivars is a diverse set representing the worldÕs most
important production areas (north-eastern Spain, Italy,
Turkey and USA), while the local cultivated and wild materials
included in this analysis were collected in a limited area from
northern Spain (10 602 km
). Locally cultivated materials are
commonly propagated using vegetative methods (grafting or
rooted suckers), while local wild materials typically use sexual
reproduction (seeds derived from cross-pollination) that gen-
erates high levels of diversity.
Results revealed that a total of 36 locally cultivated
accessions had two profiles of bands; the Casina group
comprised 30 accessions and the As16 group comprised six
accessions. Go
¨kirmak et al. (2009) used 21 SSR markers, eight
Fig. 1: Two-dimensional plot obtain-
ed from principal coordinate anal-
ysis for 82 unique genotypes.
Locally cultivated materials, wild
hazelnuts and reference cultivars
are indicated using squares, stars
and circles, respectively
4A. Campa, N. Trabanco, E. Pe
´rez-Vega et al.
of them common to the present study, and they also found an
identical profile of alleles in the local cultivars ÔCasinaÕ,
ÔAmandiÕ,ÔEspinaredoÕand ÔQuiro
´sÕ. However, within the
locally cultivated germplasm, a high level of differentiation
among accessions was found using morphological traits and
ISSR markers, as described by Ferreira et al. (2010). The
majority of locally cultivated hazelnuts used in this work were
also analysed using 11 ISSR primers (48 trees), which
generated 66 polymorphic bands. Of the 50 local trees analysed
by Ferreira et al. (2010), only two (As42 and As43) showed an
identical pattern of bands. The difference in levels of genetic
diversity between these two studies is probably due to the
different number of loci examined and perhaps the nature of
the markers. If each ISSR fragment is considered a putative
locus, Ferreira et al. (2010) analysed variation at 66 ISSR loci,
while in the present study, we analysed 13 SSR loci. ISSRs can
identify polymorphisms at both ends, manifested as length
polymorphisms, as well as insertion/deletion events within the
ISSR region.
The PCoA, the NJ tree and the population structure analysis
revealed a high level of differentiation between the locally
cultivated forms and the remaining materials. In the PCoA
plot (Fig. 1), the reference and wild materials could easily be
considered one group rather than two, especially considering
the higher weight of the first axis. In the NJ tree (Fig. 2), the
very short-arm distances for B (local wild hazelnuts) and C
(mainly reference cultivars) are similar to bifurcations within
those groups. Most of the polymorphism is among individuals
rather than among groups. The very low F
values for the
contrast of reference and wild populations, and the structure
analyses also suggest that most of the variation is among
individuals and not subpopulations. The differentiation
between most of the locally cultivated accessions and the
group of reference cultivars is in agreement with the differen-
tiation between the major part of locally cultivated forms and
a set of 17 standard cultivars revealed by the analysis of 66
ISSR markers (Ferreira et al. 2010). Four major geographical
groups or gene pools were described in European hazelnut:
Fig. 2: Neighbour-joining tree of microsatellite diversity based on the C. S. Chord distance implemented in the P
software. Three
main branches are indicated: A, group of the locally cultivated hazelnuts; B, group of the local wild hazelnuts; C, group of the reference cultivars
Genetic relationship of C. avellana 5
Central European, Black Sea, English and Spanish–Italian
¨kirmak et al. 2009, Boccacci and Botta 2010). Cultivars
from the Spanish–Italian, English and Black Sea gene pools
were included in the present analysis (Table 1), and the locally
cultivated hazelnuts were clearly differentiated from them.
¨kirmak et al. (2009) included the local cultivars ÔCasinaÕ,
ÔEspinaredoÕ,ÔAmandiÕand ÔQuiro
´sÕin their analysis, and these
were clustered into a separate subgroup within the Spanish–
Italian group. In addition, chloroplast microsatellite variation
revealed that Casina had the same haplotype as the majority of
Spanish–Italian germplasm (Boccacci and Botta 2009). All
these results suggest that locally cultivated hazelnuts belong to
the Spanish–Italian gene pool but constitute a separate
These findings raise an interesting question about the origin
of locally cultivated forms of hazelnut. Differentiation between
wild and most locally cultivated hazelnuts, and the strong
relationship among the locally cultivated accessions suggest
that locally cultivated forms may have been derived from the
introduction of a few primitive cultivars, followed by a
relatively local evolution that could include crosses among
them and with local wild hazelnut. Evidence of cultivated trees
derived from crosses between natural and cultivated popula-
tions was found in the present study (Fig. 1). In addition, the
close relationship between most of the local cultivated germ-
plasm (Casina group) suggests that the introduction of
cultivated forms may have been recent and has not led to a
broad diversification.
The plot generated from PCoA indicated putative intermedi-
ate forms and introgressions, mainly of some locally cultivated
accessions in the group of reference cultivars. The analysis of
population structure also revealed some introgressions and
admixtures of genotypes, particularly in the locally cultivated
and wild populations. The NJ tree also showed some introgres-
sion in the reference cultivar and the local cultivar groups. The
intermediate forms identified may be derived from natural
crosses between trees belonging to different groups. Hazelnut is
monoecious, dichogamous, wind-pollinated and has sporo-
phytic incompatibility controlled by a single S-locus with
multiple alleles (Thompson 1979, Mehlenbacher 1997, Pomper
et al. 1998). Introgressions in the locally cultivated population of
genotypes close to reference cultivars (e.g. accessions As60, As58
or As57) could be because of the introduction of new cultivars in
northern Spain. Exchange of hazelnut materials between Cata-
lonia and Asturias was reported beginning in the twentieth
century, and several campaigns to promote the use of cultivars
from Catalonia were conducted in the mid-twentieth century
(Alvarez Requejo 1965). This could explain (i) the position in the
group of reference cultivars of some local cultivated accessions in
the plot generated from PCoA and (ii) the presence in the
graphical output of the
software of accessions with
similar genotypes to reference cultivars. Finally, the inclusion of
some materials classified as cultivated forms in the group of wild
materials (accessions As30 or As55) could be a mistake in the
classification or the use of wild trees. The exodus of people from
rural areas in the twentieth century led to abandonment of local
hazelnut cultivation in many areas. Many orchards or gardens
with hazelnuts are now forest in which it is difficult to distinguish
the trees originally cultivated from those which arose from
hybridization with local wild hazelnuts.
Fig. 3: Hierarchical organization of genetic relatedness of 82 unique genotypes based on 13 microsatellite markers and analysed by the
program considering two and three populations (K=2 and K=3)
6A. Campa, N. Trabanco, E. Pe
´rez-Vega et al.
In summary, this study differentiated between the locally
cultivated accessions and remaining materials analysed in this
study. The local cultivated forms contained (i) a group of
accessions clearly differentiated within the Spanish–Italian
gene pool, (ii) a group with some intermediate forms probably
derived from hybridization and (iii) a group of accessions
probably derived from exchange with other geographical
origins (mainly Catalonia). The differentiation of these groups
within the local material cultivated is of great interest for the
preservation and use of the local genetic diversity.
This work was supported by grants RF2008-0014-CO3-02 from the
Ministerio de Ciencia y Tecnologı
´a, Spain. Elena Pe
´rez-Vega was
recipient of a salary fellowship from the Instituto Nacional de
´n y Tecnologı
´a Agraria y Alimentaria (INIA, Spain).
Alvarez Requejo, S., 1965: El avellano. Manuales Te
´cnicos No 32.
Ministerio de Agricultura, Madrid, Spain.
Bassil, N. V., R. Botta, and S. A. Mehlenbacher, 2005a: Microsatellite
markers in hazelnut: isolation, characterization, and cross-species
amplification. J. Am. Soc. Hortic. Sci. 130, 543—549.
Bassil, N. V., R. Botta, and S. A. Mehlenbacher, 2005b: Additional
microsatellite markers of the European hazelnut. Acta Hortic. 686,
Bioversity, FAO, and CIHEAM, 2008: Descriptors for Hazelnut
(Corylus avellana L.). Bioversity International, FAO, Rome, Italy;
CIHEAM, Zaragoza, Spain.
Boccacci, P., and R. Botta, 2009: Investigating the origin of hazelnut
(Corylus avellana L.) cultivars using chloroplast microsatellite.
Genet. Resour. Crop Evol. 56, 851—859.
Boccacci, P., and R. Botta, 2010: Microsatellite variability and genetic
structure in hazelnut (Corylus avellana L.) cultivars from different
growing regions. Sci. Hortic. 124, 128—133.
Boccacci, P., A. Akkak, N. V. Bassil, S. A. Mehlenbacher, and
R. Botta, 2005: Characterization and evaluation of microsatellite
loci in European hazelnut (Corylus avellana L.) and their transfer-
ability to other Corylus species. Mol. Ecol. Notes 5, 934—937.
Boccacci, P., A. Akkak, and R. Botta, 2006: DNA typing and genetic
relations among European hazelnut (Corylus avellana L.) cultivars
using microsatellite markers. Genome 49, 598—611.
Boccacci, P., M. Rovira, and R. Botta, 2008: Genetic diversity of
hazelnut (Corylus avellana L.) germplasm in northeastern Spain.
HortScience 43, 667—672.
Cavalli-Sforza, L. L., and A. W. F. Edwards, 1967: Phylogeneticanalysis:
models and estimation procedures. Am. J. Hum. Genet. 19, 233—257.
Evanno, G., S. Regnaut, and J. Goudet, 2005: Detecting the number of
clusters of individuals using the software STRUCTURE: a simula-
tion study. Mol. Ecol. 14, 2611—2620.
FAOstat, 2008: Agriculture date. Available at:
(last accessed on February 8, 2010).
Ferreira, J. J., C. Garcia-Gonza
´lez, J. Tous, and M. Rovira, 2010:
Genetic diversity revealed by morphological traits and ISSR
markers in hazelnut germplasm from northern Spain. Plant Breed.
129, 435—441.
Gepts, P., T. C. Osborn, K. Rashka, and F. A. Bliss, 1986: Phaseolin-
protein variability in wild forms and landraces of the common bean
(Phaseolus vulgaris): evidence for multiple centers of domestication.
Econ. Bot. 40, 451—468.
¨kirmak, T., S. A. Mehlenbacher, and N. V. Bassil, 2009: Charac-
terization of European hazelnut (Corylus avellana L.) cultivars using
SSR markers. Genet. Resour. Crop Evol. 56, 147—172.
Grassi, F., M. Labra, S. Imazio, A. Spada, S. Sgorbati, A. Scienza, and
F. Sala, 2003: Evidence of a secondary grapevine domestication
centre detected by SSR analysis. Theor. Appl. Genet. 107,
¨rcan, K., S. A. Mehlenbacher, and V. Erdog
˘an, 2010: Genetic
diversity in hazelnut (Corylus avellana L.) cultivars from Black
Sea countries assessed using SSR markers. Plant Breed. 129,
Kwak, M., and P. Gepts, 2009: Structure of genetic diversity in the two
major gene pools of common bean (Phaseolus vulgaris L., Faba-
ceae). Theor. Appl. Genet. 118, 979—992.
Lagerstedt, H. B., 1975: Filberts. In: J. Janick, and J. N. Moore (eds),
Advances in Fruit Breeding, 456—489. Purdue University Press,
West Lafayette, Indiana.
Liu, K. J., and S. V. Muse, 2005: PowerMarker: an integrated analysis
environment for genetic marker analysis. Bioinformatics 21,
Mehlenbacher, S. A., 1997: Revised dominance hierarchy for S alleles
in Corylus avellana L. Theor. Appl. Genet. 94, 360—366.
Mehlenbacher, S. A., R. N. Brown, E. R. Nouhra, T. Go
N. V. Bassil, and T. L. Kubisiak, 2006: A genetic linkage map for
hazelnut (Corylus avellana L) based on RAPD and SSR markers.
Genome 49, 122—133.
Page, R. D. M., 1996: TREEVIEW: an application to display
phylogenetic trees on personal computers. Comput. Appl. Biosci.
12, 357—358.
Papa, R., and P. Gepts, 2003: Asymmetry of gene flow and differential
geographical structure of molecular diversity in wild and domesti-
cated common bean (Phaseolus vulgaris L.) from Mesoamerica.
Theor. Appl. Genet. 106, 239—250.
Pomper, K. W., A. N. Azarenko, N. V. Bassil, J. W. Davis, and
S. A. Mehlenbacher, 1998: Identification of random amplified
polymorphic DNA (RAPD) markers for self-incompatibility alleles
in Corylus avellana L. Theor. Appl. Genet. 97, 479—487.
Pritchard, J., M. Stephens, and P. Donnelly, 2000: Inference of
population structure using multilocus genotype data. Genetics 155,
Rohlf, F. J., 2002: NTSYSpc: Numerical Taxonomy System, ver 2.1.
Exeter Publishing, Ltd., Setauket, NY.
Tasias Valls, J., 1975: El avellano en la provincia de Tarragona. Excma
´n Provincial de Tarragona, Tarragona, Spain.
Thompson, M. M., 1979: Genetics of incompatibility in Corylus
avellana L. Theor. Appl. Genet. 54, 113—116.
Thompson, M. M., H. B. Lagerstedt, and S. A. Mehlenbacher, 1996:
Hazelnuts. In: J. Janick, and J. N. Moore (eds), Fruit Breeding:
Nuts, vol 3, 125—184. Wiley, New York.
Tous, J., M. Rovira, and A. Romero, 2001: El Avellano. In: F. Nuez,
and G. Llacer (eds), 275—278. SECH, Co
´rdoba, Spain
UPOV, 1979: Guidelines for the conduct of test for distinctness,
homogeneity and stability (hazelnut). TG/71/3 Available at: http:// (last accessed
on March 1, 2010).
Genetic relationship of C. avellana 7
... Most genetic diversity studies in hazelnut have focused on cultivars (Bacchetta et al. 2014) with only recent interest in wild individuals and landraces (Ferreira et al. 2010;Campa et al. 2011;Leinemann et al. 2013;Martins et al. 2013Martins et al. , 2014Martins et al. , 2015. In addition to its contribution to biodiversity, wild germplasm is widely recognized as a potential resource of interesting traits for improved cultivars (Tanskley and McCouch, 1997). ...
... As expected, SSR markers gave fewer alleles than AFLP but proved to be more informative with an average gene diversity of 0.30 compared to 0.26 for AFLP. The 49 SSR markers had an average of 10.3 alleles per SSR which was consistent with other work with the same or similar genomic SSR markers which resulted in 3 to 10.6 alleles per marker Boccacci et al. 2005Boccacci et al. , 2008Gokirmak et al. 2009;Boccacci and Botta 2010;Gurcan and Mehlenbacher 2010;Gurcan et al. 2010aGurcan et al. , 2010bCampa et al. 2011;Bassil et al. 2013). The distance matrices generated from the AFLP and SSR marker data had only low correlation to each other perhaps because they sampled different parts of the hazelnut genome. ...
... Gene diversity values for the AFLP and SSR markers and the genetic dissimilarity matrix calculated with the Dice coefficient indicated high genetic diversity of hazelnut in agreement with many other studies (Gokirmak et al. 2009;Gurcan et al. 2010aGurcan et al. , 2010bCampa et al. 2011;Boccacci and Botta 2010;Boccacci et al. 2006Boccacci et al. , 2008Boccacci et al. , 2013. Levels of diversity are expected to be high in hazelnut because it has a self-incompatible mating system which prevents selfpollination. ...
Full-text available
European hazelnut (Corylus avellana L.), cultivated in several areas of the world including Europe, Anatolia, and the USA, is an economically important nut crop due to its high mineral, oleic acid, amino acid, and phenolic compound content and pleasant flavor. This study examined molecular genetic diversity and population structure of 54 wild accessions and 48 cultivars from the Slovenian national hazelnut collection using amplified fragment length polymorphism (AFLP) and simple sequence repeat (SSR) markers. Eleven AFLP primer combinations and 49 SSR markers yielded 532 and 504 polymorphic fragments, respectively. As expected for a wind-pollinated, self-incompatible species, levels of genetic diversity were high with cultivars and wild accessions having mean dissimilarity values of 0.50 and 0.60, respectively. In general, cultivars and wild accessions clustered separately in dendrogram, principal coordinate, and population structure analyses with regional clustering of the wild material. The accessions were also characterized for ten nut and seven kernel traits and some wild accessions were shown to have breeding potential. Morphological principal component analysis showed distinct clustering of cultivars and wild accessions. An association mapping panel composed of 64 hazelnut cultivars and wild accessions had considerable variation for the nut and kernel quality traits. Morphological and molecular data were associated to identify markers controlling the traits. In all, 49 SSR markers were significantly associated with nut and kernel traits [P < 0.0001 and LD value (r2) = 0.15–0.50]. This work is the first use of association mapping in hazelnut and has identified molecular markers associated with important quality parameters in this important nut crop.
... The 30 SSR markers provided sufficient polymorphism in the 402 hazelnut accessions with 13.6 fragments per marker. This value is slightly higher than those obtained by others using SSR markers in hazelnut which varied from 3 to 10.6 fragments per primer pair Boccacci et al. 2005Boccacci et al. , 2008Gökirmak et al. 2009;Boccacci and Botta 2010;Gurcan and Mehlenbacher, 2010;Gurcan et al. 2010a, b;Campa et al. 2011;Bassil et al. 2013;Ozturk et al., 2017). In previous work, a higher annealing temperature was used for these SSR primers (Gurcan et al., 2010a) (60 rather than 55°C). ...
... Subsets were selected based on molecular data (A), morphological features (B), and identity as a named cultivar (C). Subpopulation assignment for each accession is given in parentheses (1) FAI318 (1) Palaz (1) FAI081 (1) FAI324 (2) Sivri (1) FAI093 (1) FAI333 (1) Tombul (1) FAI096 (1) FAI351 (2) Uzunmusa (1) FAI112 (1) FAI388 (2) Yassıbadem (1) FAI144 (1) FAI406 (1) Yuvarlakbadem (1) (1) genetic diversity were observed in cultivars and wild accessions from Spain (Campa et al. 2011), Portugal (Martins et al. 2015), and Slovenia (Ozturk et al. 2017). Most of the accessions were collected in Giresun which is an area with extensive hazelnut cultivation and production. ...
... Dendrogram and PCoA analyses of the hazelnut collection indicated that cultivars were loosely clustered but not genetically distinct from landraces and wild material. In contrast, a clear separation between wild and cultivated accessions was observed in materials from Spain, Portugal, and Slovenia (Campa et al. 2011;Martins et al. 2014Martins et al. , 2015Ozturk et al., 2017). The difference between these studies and ours may lie in the fact that reference cultivars were used in the other studies. ...
Full-text available
European hazelnut (Corylus avellana L.) is an economically and nutritionally important nut crop with wild and cultivated populations found throughout Europe and in parts of Asia. This study examined the molecular genetic diversity and population structure of 402 genotypes including 143 wild individuals, 239 landraces, and 20 cultivars from the Turkish national hazelnut collection using simple sequence repeat (SSR) markers. A total of 30 SSR markers yielded 407 polymorphic fragments. Diversity analysis of the Turkish hazelnut genotypes indicated that they fell into three subpopulations according to ad hoc statistics and neighbor-joining algorithm. Although all cultivars clustered together, they overlapped with the wild accessions and landraces. Thus, the dendrogram, principal coordinate, and population structure analyses suggest that they share the same gene pool. A total of 78 accessions were selected as a core set to encompass the molecular genetic and morphological diversity present in the national collection. This core set should have priority in preservation efforts and in trait characterization.
... The number of alleles ranged from one to 12 with an average of 3.2 alleles. This average allele number was in the range of the previous studies (from 3 to 13.6) which used SSR markers in hazelnut Boccacci et al. 2005Boccacci et al. , 2008Gökirmak et al. 2009;Boccacci and Botta 2010;Gurcan and Mehlenbacher 2010;Gurcan et al. 2010a, b;Campa et al. 2011;Bassil et al. 2013;Ozturk et al. 2017a, b). A dendrogram was constructed using SSR data and the accessions fell into three subclusters. ...
... In contrast, the current set of tested SSR markers was randomly selected from the thousands of primers that were designed. Low variability is not unexpected because hazelnuts have been selected for similar characters and certain allelic combinations which may reduce their diversity (Campa et al. 2011;Martins et al. 2015;Ozturk et al. 2017a, b). Moreover, cultivars are propagated as clones, a fact that limits their diversity as compared to wild accessions. ...
Full-text available
European hazelnut (Corylus avellana) is a diploid tree species and is widely used in confections. Hazelnuts are, to a large part, produced in Turkey with the cultivar “Tombul” widely grown in the Black Sea region. In this work, the “Tombul” genome was partially sequenced by next-generation sequencing technology yielding 29.2% (111.85 Mb) of the ~ 385 Mb (1C). This sequence information was used to develop genetic markers in order to enable differentiation of material before the long maturation process and to facilitate future breeding strategies. A total of 90,142 simple sequence repeats (SSRs) were identified in the contigs giving a frequency of 1 SSR per 1240 nt in the assembly. Mononucleotides were the most abundant SSR marker type (60.9%) followed by di- and trinucleotides. Primer pairs were designed for 75,139 (83.3%) of the SSRs. Fifty SSR primers were applied to 47 hazelnut accessions from nine countries to test their effectiveness and polymorphism. The markers amplified an average of 3.2 fragments. The highest polymorphism information content value was for cavSSR11062 (0.97) and the lowest (0.04) was for cavSSR13386. Two markers were monomorphic: cavSSR12855 and cavSSR13267. Single-copy SSR primers were also assessed for their ability to discriminate 19 Turkish cultivars, and it was found that seven primer pairs (Cav4217, Cav14875, Cav14418, Cav2704, Cav12862, Cav3909, Cav1361) were sufficient for this task. Thus, this study developed new SSR markers for use in hazelnut breeding and genetic studies and also provide a method to distinguish and identify true-type Turkish cultivars.
... A genetic diversity analysis using ISSR and SSR markers revealed that the collected local materials were strongly related, constituting a group different from the genotypes grown in northeast Spain and Italy. The local materials also included intermediate forms that were probably derived from crosses between cultivated and wild materials (Ferreira et al., 2010;Campa et al., 2011). ...
Full-text available
Hazelnut is a traditional crop in northern Spain, where it grows wild as well as being cultivated. A field collection of 41 local and 17 non-local accessions, including 15 well-known cultivars, was established at SERIDA in Villaviciosa, Spain. Here, phenotypic variation was documented for phenological and morphological traits and chemical composition. A large degree of variation for most morphological and phenological traits, except nut maturity date, was revealed. Estimates of broad-sense heritability were high (>0.75) for most of the assessed characters, except for the first male bloom date (0.65), male and female flowering periods (0.40, 0.31), kernel weight (0.69), and kernel percentage (0.33). Local accessions produced smaller nuts and kernels than well-known cultivars but with higher kernel percentage. Limited overlapping between the male and female flowering periods (dychogamy) was observed, except for ‘Forcinas 1’, ‘Forcinas 2’, and ‘Morell’. The local accessions generally exhibited significantly later male and female flowering compared with the reference cultivars. The local materials showed similar nutritional values to those reported previously for hazelnut. Moreover, the local accessions presented average values similar to the non-local accessions for total fat, ash and carbohydrate contents, as well as energy value, but their protein contents were lower. Their oils were rich in functional compounds, such as unsaturated fatty acids (average: 90.1%), tocopherols (514 mg/kg) and squalene (294.3 mg/kg). A hierarchical clustering on principal components analysis grouped the accessions and differentiated eight local accessions from the rest, including the landrace ‘Casina’. This finding provides potential new cultivars, as well as sources of desirable traits, for European hazelnut breeding programs.
... Nearly all efforts have been focused on cultivated forms largely to better understand their origin, to fingerprint germplasm accessions, and to evaluate genetic diversity Botta 2009, 2010;Boccacci et al. 2006Boccacci et al. , 2008Gökirmak et al. 2009;Gürcan et al. 2010). Recently, a first on-farm exploration was conducted on local ecotypes and on wild accessions in northern Spain (Asturias) by Ferreira et al. (2010) and Campa et al. (2011). Finally, the long history of utilization and production by humans, probably predating the Roman era (Rosengarten 1984;Bacchetta et al. 2011;Boccacci and Botta 2009), makes this crop interesting from a social point of view, with a precious role in sustainable traditional agricultural systems. ...
... More than 700 polymorphic SSR markers were developed in C. avellana (Bassil et al., 2005a(Bassil et al., , 2005b(Bassil et al., , 2013Boccacci et al., 2005Boccacci et al., , 2015Mehlenbacher, 2010a, 2010b;Peterschmidt, 2013;Bhattarai, 2015). They have been used for linkage map construction (Bhattarai, 2015;Mehlenbacher et al., 2006), quantitative trait locus (QTL) analysis (Beltramo et al., 2016), assessment of genetic relationships among cultivars (Bassil et al., 2013;Boccacci et al., 2006Boccacci et al., , 2008Boccacci et al., , 2013Boccacci and Botta, 2010;Campa et al., 2011;Gökirmak et al., 2009;, and for identification and parentage confirmation Botta et al., 2005;Gökirmak et al., 2009;Sathuvalli and Mehlenbacher, 2012). Dinucleotide-containing SSRs are the most abundant throughout the genome, but suffer from stuttering, split peaks and binning errors which lead to difficulties in unequivocal genotyping and genetic profile discrepancies among different laboratories (Baldoni et al., 2009;Testolin and Cipriani, 2010). ...
The objective of this study was to develop a robust and cost-effective fingerprinting set for hazelnuts using microsatellite markers, also known as simple sequence repeat (SSR) markers. Twenty SSRs containing repeat motifs of three or more nucleotides distributed throughout the hazelnut genome were screened on eight cultivars to assess polymorphism, allele size range, and ease of scoring. Six SSRs were discarded after genotyping 96 hazelnut samples either due to large allele bin widths and/or alleles that do not match the motifs, which complicates allele scoring. Fourteen polymorphic, easy-toscore SSRs were selected and amplified in a single multiplex. The 14-SSR multiplexed set generated the same alleles that were obtained when amplifying each SSR individually in the eight test accessions. SSR primer concentrations were then optimized to generate a clear signal for each locus. This 14-SSR fingerprinting set was used to genotype 102 hazelnut accessions from different origins. The fingerprinting set distinguished unique accessions mainly according to parentage and in some cases based on geographic origin. They identified each of the cultivars released from the Oregon State University breeding program and confirmed the parentage of six cultivars. Tools for DNA fingerprinting of clonally propagated horticultural crops like hazelnut are in demand and this multiplexed set constitutes a reliable, less-time consuming and cost-effective procedure for identity and parentage confirmation in hazelnut.
Full-text available
Understanding species-specific responses to climate change allows a better assessment of the possible impact of global warming on forest growth. We studied the responses of the shrub species Corylus avellana L., Corylus maxima Mill. and intermediate forms, together stated as the Corylus taxa, and Cornus sanguinea L. upon periodically elevated temperatures in spring and/or in summer–autumn. Experiments were performed in a common garden, with Belgian and Pyrenean provenances for Corylus avellana and Cornus sanguinea. In the Corylus taxa, a warmer spring resulted in a reduction in height and diameter growth. Remarkably, the reduced diameter increment was restored with full compensation in the following year. The height increment for Cornus sanguinea was larger upon a warmer summer–autumn, concurring with a later leaf senescence. Our results suggest that Corylus is more sensitive to spring warming, influencing growth negatively, whereas Cornus is more sensitive to summer–autumn warming, influencing height growth positively. These deviating responses can be explained, at least partly, by their diverging ecological niches, with the Corylus taxa being more shade-tolerant compared to Cornus sanguinea. The warm spring conditions advanced bud burst in all studied taxa, whereas the warm summer–autumn advanced leaf senescence but prolonged its duration in the Corylus taxa, as well as delayed this phenophase in Cornus sanguinea. Little to no after-effects of the temperature treatments were detected. Although Corylus avellana and Cornus sanguinea plants originated from similar origins, their growth and phenological responses in the common garden diverged, with Corylus being more stable and Cornus displaying more variation between the Belgian and Pyrenean provenances.
Hazelnut ( Corylus spp.), also known as filbert, is an important temperate nut that has been used as a food crop since ancient times. There is a tremendous amount of genetic diversity in Corylus and more than 500 Corylus avellana cultivars have been described in the literature. World production is based on fewer than 20 ancient C. avellana cultivars that originated with the selection and clonal propagation of superior plants primarily in the Mediterranean Basin and along the Black Sea coast. Large nuts are sold in‐shell, while the majority of the world's hazelnuts are sold as kernels for use by the confectionary industry. In Turkey, the most important producing country, nut clusters are picked from the tree by hand, often in orchards on very steep terrain. The husks of Turkish cultivars are long and clasping and retain the clusters on the tree for harvest. In other countries, nuts are mechanically harvested from the ground, and nuts that fall free of the husk at maturity are desired. As newer production systems are adopted, including high‐density orchards of compact trees, appropriate cultivars, management practices, and machinery will be needed. Because of high demand for kernels, new orchards are being planted in many countries, including areas where the climate is less than ideal for the species. Interspecific hybridization offers an opportunity to expand production to new areas. Hazelnut has many arthropod pests, of which big bud mite (primarily Phytoptus avellanae ) is a serious one. Many diseases affect hazelnut, of which eastern filbert blight (EFB) caused by the fungus Anisogramma anomala is the most severe threat in North America. Genetic resistance to big bud mite and EFB has been identified and incorporated into new cultivars. Early genetic improvement efforts were largely by nurseries and growers. Over the past 60 years, 11 breeding programs have been active in different countries, but today there are only two relatively large programs, both in the USA. Standard methods to breed hazelnuts have been very effective although it takes 17 years from the time a controlled cross is made until a new cultivar is released. Biotechnology has added tools and knowledge to genetic improvement efforts. Thousands of DNA markers have been developed and scored, linkage maps have been constructed, and our knowledge of genetic diversity and the control of traits in Corylus have improved. To date, however, the use of marker‐assisted selection has been limited to EFB resistance from ‘Gasaway’. Genome sequences have been assembled into chromosome‐length scaffolds and genes have been annotated, allowing efficient marker development for genetic studies and marker‐aided breeding. The use of genome editing is limited by an inability to regenerate whole plants from single cells. The huge amount of genetic variability in C. avellana and other species can be exploited to develop improved cultivars with wider adaptation. Worldwide demand for hazelnuts continues to increase. With the proven effectiveness of traditional plant breeding approaches, the diverse genetic resources that have been collected and preserved, and the considerable promise of modern genetic and genomic tools, the future looks very bright for further genetic improvement of hazelnut.
Hazelnut is an economically important tree nut whose production is mostly destined to the confectionery industry with a demand that currently exceeds supply. Its cultivation remains substantially based on named selections from local, wild vegetation. Public breeding programs were not initiated until the 1960s and only two, both in the USA, are in operation today that are relatively large. Oregon State University has produced new cultivars with Gasaway resistance to the fungus Anisogramma anomala, causal agent of eastern filbert blight (EFB), a major disease in North America; these cultivars are being widely planted. In China, cold-hardy hybrid cultivars from Corylus heterophylla and C. avellana were recently released and are planted in northeastern China. In the past 25 years, molecular markers have facilitated a much better understanding of genetic diversity in the genus Corylus, aided the construction of linkage maps and allowed for marker-assisted selection for disease resistance. The genome of C. avellana was sequenced and assembled, and DNA markers identified from the transcriptome, providing the basis for the isolation of important genes, including those related to nut quality and adaptive and phenological traits. Many new genotypes expressing eastern filbert blight (EFB) resistance have been identified in the germplasm, and subsequent linked DNA markers developed, allowing new approaches to breeding for durable resistance. Micropropagation is routinely used in the USA, Chile and Italy for multiplication, but work with other in vitro techniques is less advanced. Genetic engineering has not been developed in hazelnut due to regeneration difficulties from somatic tissues but recent advances have established a protocol for organogenesis. More research is being carried out to assemble a high-quality hazelnut genome and achieve somatic embryogenesis. The results from this research will provide knowledge and tools enabling the isolation of genes and molecular markers, and the application of genome editing techniques to hazelnut.
Full-text available
Three microsatellite-enriched libraries of the european hazelnut (Corylus avellana L.) were constructed: library A for CA repeats, library B for GA repeats, and library C for GAA repeats. Twenty-five primer pairs amplified easy-to-score single loci and were used to investigate polymorphism among 20 C. avellana genotypes and to evaluate cross-species amplification in seven Corylus L. species. Microsatellite alleles were estimated by fluorescent capillary electrophoresis fragment sizing. The number of alleles per locus ranged from 2 to 12 (average = 7.16) in C. avellana and from 5 to 22 overall (average = 13.32). With the exception of CAC-B110, di-nucleotide SSRs were characterized by a relatively large number of alleles per locus (≤5), high average observed and expected heterozygosity (Ho and He > 0.6), and a high mean polymorphic information content (PIC ≤ 0.6) in C. avellana. In contrast, tri-nucleotide microsatellites were more homozygous (Ho = 0.4 on average) and less informative than di-nucleotide simple sequence repeats (SSRs) as indicated by a lower mean number of alleles per locus (4.5), H e (0.59), and PIC (0.54). Cross-species amplification in Corylus was demonstrated. These microsatellite markers were highly heterozygous and polymorphic and differentiated among genotypes of C. avellana irrespective of geographical origin. They will aid in fingerprinting genotypes of the european hazelnut and other Corylus species, genome mapping, and genetic diversity assessments.
Full-text available
In Spain, hazelnut is mainly cultivated in Catalonia, a region in the northeast. The province of Tarragona accounts for 88% of the total Spanish area planted to hazelnut. Almost 80% of the production in Tarragona is of the local cultivar Negret, with others cultivated to a lesser extent. Minor cultivars are only sporadically present in older orchards, farm yards and gardens, and have been collected for preservation. In this work, 16 SSR markers were used to fingerprint 18 minor hazelnut cultivars from northeastern Spain. Their microsatellite profiles were combined with those of 15 Spanish cultivars characterized in a previous work, and used to study the genetic diversity in 33 genotypes including local Spanish germplasm. The SSR analysis allowed development of unique profiles of each of the 18 cultivars, and no new case of synonymy was detected. A high level of genetic diversity (mean He = 0.7) was observed in 33 genotypes, although a high number of them showed a close genetic relationship. The dendrogram generated by UPGMA cluster analysis placed the 33 accessions into nine main groups, related to their putative pedigrees or geographical area of cultivation. All investigated Negret-type cultivars were found to be distinct from Negret, and only a few cultivars within this germplasm appeared to be seedlings of Negret. The results will be useful in the conservation of hazelnut germplasm and in the selection of parents for use in breeding.
Full-text available
In this work, 18 microsatellite loci were developed in the European hazelnut (Corylus avellana L.) using three enriched genomic libraries. They were evaluated on a set of 20 accessions of this species on the basis of number of alleles (mean: 7.1), expected heterozygosity (mean: 0.67), power of discrimination (mean: 0.77) and polymorphism information content (mean: 0.64). Cross-species transferability was evaluated using seven other Corylus species. All primer pairs amplified in all species, except for CaT-C505 in Corylus ferox and CaT-A114 in Corylus californica.
Full-text available
World hazelnut production is based primarily on selections from the wild. In this study, we used 21 pairs of simple sequence repeat (SSR) primers to investigate genetic diversity in 270 clonal accessions of European hazelnut (Corylus avellana) representing a wide geographic range. Of the 270 accessions, 198 had unique fingerprints while 72 were duplicates. Based on the 198 unique accessions, the number of detected alleles per locus averaged 9.81 and observed heterozygosity (Ho) averaged 0.67. Of the 206 total alleles amplified, 20 were unique to a single accession. A genetic similarity matrix was constructed and the resulting dendrogram revealed four major geographical groups: Central European, Black Sea, English and Spanish-Italian. SSR alleles indicated the parentage of 31 accessions. The fingerprints are publicly available through the Germplasm Resources Information Network (GRIN) database. The identification of duplicate and mislabeled accessions will improve management of hazelnut genebanks, and information on genetic variation in hazelnut will assist the international research community.
We describe a model-based clustering method for using multilocus genotype data to infer population structure and assign individuals to populations. We assume a model in which there are K populations (where K may be unknown), each of which is characterized by a set of allele frequencies at each locus. Individuals in the sample are assigned (probabilistically) to populations, or jointly to two or more populations if their genotypes indicate that they are admixed. Our model does not assume a particular mutation process, and it can be applied to most of the commonly used genetic markers, provided that they are not closely linked. Applications of our method include demonstrating the presence of population structure, assigning individuals to populations, studying hybrid zones, and identifying migrants and admixed individuals. We show that the method can produce highly accurate assignments using modest numbers of loci—e.g., seven microsatellite loci in an example using genotype data from an endangered bird species. The software used for this article is available from
Conference Paper
Sixteen microsatellite-containing sequences were identified in enriched genomic libraries of the European hazelnut, Corylus avellana L. Of these, seven amplified multiple loci, one was difficult to score and eight amplified polymorphic single loci. Amplification and polymorphism of these eight primer pairs in 20 cultivars of C. avellana were initially determined by electrophoresis using 3% agarose gels. Forward primers were then fluorescently labelled and allele sizes determined by capillary electrophoresis. The addition of PIG-tails to five reverse primers was necessary to eliminate split peaks. Three microsatellite markers (CAC-AC115, CAC-B005 and CAC-B014) reliably distinguished among the cultivars. Microsatellite markers will be used to fingerprint the hazelnut genotypes in our collection and identify duplicate accessions.
Diallele crosses of sibs and parents of 3 parental combinations demonstrated sporophytic type of incompatibility inCorylus avellana L. One gene with multiple alleles was indicated. All 6 alleles present in the 4 parental cultivars and in their progeny exhibited dominance in pollen and independent action in the pistil. Individuals homozygous for S-alleles appeared in progeny of parents having one allele in common. Reciprocal differences occurred in some crosses. The stigmatic surface is the site of the incompatibility reaction. Incompatible pollen germinated abundantly but failed to penetrate into the stigma cells.
Pollen-stigma compatibility was studied in cultivars and more than 1800 seedlings of the European hazelnut (Corylus avellana L). Four new S-alleles were identified, bringing the total to 25 unique alleles within C. avellana. The new alleles are the recessive alleles in ‘Tonda di Giffoni’ and ‘Segorbe’ (S23), in ‘Neue Riesennuss’ (S25), in ‘Gasaway’ (S26), and a dominant allele in a seedling of Turkish origin (S24). Dominance relationships in 233 of the possible 300 pairs of alleles were determined in both pistil and pollen. All alleles exhibited independent action in the pistil, whereas in the pollen either dominance or codominance was exhibited. The dominance hierarchy of alleles in the pollen was revised in light of the new information obtained. All 25 alleles have been assigned to a level in the hierarchy that is linear and now has eight levels. S6 and S9 were reassigned to lower levels in the hierarchy. Thirteen of the alleles are on the level of S1, while S4, S6, S11, and S23 occupy unique positions in the hierarchy. Improved pollen tester clones were identified for several S-alleles. The alleles in 55 cultivars were determined. The alleles identified in ‘DuChilly’ (S10 S14) did not agree with previous reports. Four cultivars have the same alleles as ‘Rmische Nuss’ (S10 S18) and are morphologically indistinguishable from it: ‘Frutto-grosso’, ‘Istarski Okrogloplodna’, ‘Payrone’, and ‘Romai’. ‘Belle di Giubilino’ and ‘Tonda di Biglini’ are both S1 S10 and appear to be synonyms for the same cultivar.
With 3 figures and 4 tablesAbstractHazelnut (Corylus avellana L.) has been a traditional crop in northern Spain. As a result of germplasm exploration over 3 years (2003–05), 90 trees were selected in this region. This study describes phenotypic variation in nut and husk traits and investigates genetic relationships among selections and cultivars using inter simple sequences repeat (ISSR) markers. The local selections were phenotypically diverse and many had characteristics appreciated by the market. Eleven ISSR primers, which generated 66 polymorphic bands, were used in the analysis. The graph from principal coordinates analysis of the molecular marker data showed two main groups, one for the local selections and the other for the standard cultivars. The dendrogram generated from UPGMA cluster analysis showed the same two main groups. The results suggest that the local accessions are closely related to each other, but are relatively distant from the standard cultivars of eastern Spain, Italy and the USA. Selections from northern Spain may be directly useful as new cultivars or alternatively as parents in breeding programmes. The collection and preservation of this genetic diversity is important.
With 6 figures and 6 tablesAbstractEuropean hazelnut (Corylus avellana L.) is an important crop in Turkey, Georgia and Azerbaijan, where cultivars were selected from the native vegetation. Accessions from Turkey have been assigned to the Black Sea group, and cultivars from Georgia and Azerbaijan have a similar phenotype. Genetic diversity was investigated in 88 accessions from these three countries and compared with cultivars from Spain and Italy using 12 microsatellite loci. A high level of genetic diversity (He = 0.71, Ho = 0.70) was observed in the Black Sea accessions. Six Turkish accessions in the US hazelnut collections were found to be synonyms of cultivars in the Turkish collection in Giresun. An unweighted pair-group method using arithmetic average dendrogram and principal component analysis of 109 unique accessions showed a tendency to form subgroups by country of origin, and high diversity within each subgroup. A moderate shift in allelic frequencies (FST = 0.114–0.131) was seen between accessions from the Black Sea and the Spanish-Italian accessions. Simple sequence repeat analysis identified the putative parents of two Turkish cultivars.