Journal of Biotechnology 111 (2004) 263–267

Short communication
Analysis of genetic relationships among Rosa damascena plants
grown in Turkey by using AFLP and microsatellite markers
Nilgün Göktürk Baydar a,∗ , Hasan Baydar b , Thomas Debener c
a Department of Horticulture, Faculty of Agriculture, University of Süleyman Demirel, 32260 Isparta, Turkey
b Department of Field Crops, Faculty of Agriculture, University of Süleyman Demirel, 32260 Isparta, Turkey
c Federal Centre for Breeding Research on Cultivated Plants, Institute for Ornamental Plant Breeding,

Bornkampweg 31, D-22926 Ahrensburg, Germany

Received 24 November 2003; received in revised form 26 April 2004; accepted 30 April 2004

Abstract

Rosa damascena Mill. is the most important rose species for rose oil production. The main rose oil producers in the world are
Turkey and Bulgaria and they obtain the rose oil almost exclusively from R. damascena. In spite of coming from the same original
populations, R. damascena plants grown in Turkey show some morphological differences. In this study, it was aimed to investigate
the genetic relationships among R. damascena plants grown in Turkey by using microsatellite and AFLP markers. Twenty three
AFLP and nine microsatellite primer pairs were used for this aim. No polymorphism could be detected among the plants, as
the marker patterns obtained from different plants are identical. The conclusion from these data is that all R. damascena plants
under study are derived from the same original genotype by vegetative propagation. Furthermore, the observed morphological
differences originate from point mutations not detectable by molecular markers. Therefore, they are equivalent to sport mutations
frequently observed in cut and garden rose varieties.
© 2004 Elsevier B.V. All rights reserved.

Keywords: Rosa damascena; AFLP; Microsatellite; Genetic relationships

1. Introduction flavorings. For example, rose oil known as rose otto

Rose is one of the most important crop for floricul-
ture industry. The genus Rosa includes 200 species
and more than 18,000 cultivars (Gudin, 2000). Apart
from the use as cut flowers, potted plants and gar-
den plants, roses also have an economic importance
for their petal as a source of natural fragrances and
∗ Corresponding author. Tel.: +90-246-211-14-18;
fax: +90-246-237-16-93.
E-mail address: nilgun@ziraat.sdu.edu.tr (N. Göktürk Baydar).
is the most important commercial product, espe-
cially in the flavor and fragrance industry (Kovats,
1987).
Despite the large number of cultivated rose varieties,
only a few of them exhibit the marked fragrance that
is sought by perfumeries in the world (Antonelli et al.,
1997). There are mainly four species of roses for oil
production. These are Rosa damascena Mill., Rosa
gallica L., Rosa moschata Herrm. and Rosa centifolia
L. (Tucker and Maciarello, 1988). Rose oil is produced
in Morocco and France from R. centifolia, in Egypt

0168-1656/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jbiotec.2004.04.014
264 N. Göktürk Baydar et al. / Journal of Biotechnology 111 (2004) 263–267
from R. gallica, in China from R. rugosa and other
Rosa spp. But the main rose oil producers in the world
are Turkey and Bulgaria and they obtain the rose oil
from R. damascena (a hybrid between R. gallica L.
and R. phoenicia Boiss.). In addition to rose oil, some
important base materials of the cosmetic industry such
as rose concrete, absolute and rose water from these
species are also obtained.
The first R. damascena plantations in Turkey were
established with plants brought from Bulgaria in
the 1880s. After that new gardens were established
especially in the province of Isparta and its sur-
rounding by vegetatively propagating the original
material.
In spite of coming from the same origin, R. dam-
ascena plants grown in Turkey show some morpho-
logical differences such as flower shape, leaf and
plant size. As it was not absolutely clear whether the
plant material is based on a single or a few original
genotypes, the observed differences might either be
the result of the original differences between these
genotypes or the result of newly arisen sport muta-
tions. Therefore, it is of great importance for future
breeding of R. damascena genotypes with improved
horticultural characteristics to elucidate the genetic
relationships between the different R. damascena
accessions in Turkey.
Molecular markers, which directly display dif-
ferences on the DNA level and which are inde-
pendent of the phenotype, represent a significant
resource for creating genetic and physical genome
maps, distinguishing individuals, investigating ge-
netic relatedness and studying genome organiza-
tion (Thomas et al., 1993; Debener and Mattiesch,
1998).
Molecular markers have been applied in a number
of research projects to investigate the polymorphism
in roses (Hubbard et al., 1992; Torres et al., 1993;
Reynders-Aloisi and Bollereau, 1996; Debener et al.,
1996). But the only reported study conducted on ana-
lyzing the genetic diversity of R. damascena by using
the Random Amplified Polymorphic DNA (RAPD), a
kind of molecular marker, was represented by Ağaoğlu
et al. (2000).
This study was aimed at the investigation of the
genetic relationships among R. damascena plants
grown in Turkey by using microsatellite and AFLP
markers.
2. Materials and methods
All experiments were conducted in the laborato-
ries of the Institute for Ornamental Plant Breeding in
Ahrensburg-Germany.
2.1. Plant materials
Leaves from 15 individuals were sampled from
15 different R.damascena Mill. plantations of Is-
parta province of Turkey (latitude 37◦ 45 N, longitude
30◦ 33 E, altitude 997 m) in May 2003. In each planta-
tion, plants were propagated vegetatively by cuttings.
2.2. DNA extraction
Genomic DNA from young rose leaves was ex-
tracted with DNeasy plant mini kits (Qiagen, Inc., CA,
USA).
2.3. AFLP analysis
AFLP analyses were performed as described in
Debener and Mattiesch (1999). Briefly genomic
DNA (250 ng) was digested overnight with the
HindIII and MseI restriction enzymes in a total vol-
ume of 25 ␮l and ligated with HindIII and MseI
site-specific double stranded adaptors. For pream-
plification, 5 ␮l of ligation mixture was amplified
in a Genius thermocycler with 20 cycles of 94 ◦ C
for 30 s, 60 ◦ C for 30 s and 72 ◦ C for 1 min using
HindIII (5 -GACTGCGTACCAGCTT-3 ) and MseI
(5 -GATGAGTCCTGAGTAA-3 ) primers carrying
one selective nucleotide at the 3 end. For selective
amplification, 2.5 ␮l of a 20-fold diluted preamplifica-
tion mixture was amplified in Genius E Thermocycler
consisting of one cycle of 94 ◦ C/30 s, 65 ◦ C/30 s and
72 ◦ C/1 min, then lowering the annealing temperature
to 56 ◦ C in the next 11 cycles; and then 23 cycles
of 94 ◦ C for 30 s, 56 ◦ C for 30 s and 72 ◦ C for 1 min
using HindIII and MseI primers with three selective
nucleotides. HindIII primers were also fluorescently
labelled with the IRD 800 or the IRD 700 dyes at
their 5 end (MWG Biotech, Ebersberg, Germany).
Primer combinations used in this study are listed in
Table 1. Samples were analyzed on a LICOR 400
Automatic Sequencer (MWG Biotech, Ebersberg,
Germany).
 N. Göktürk Baydar et al. / Journal of Biotechnology 111 (2004) 263–267 265

Table 1 and extention temperatures were 94 ◦ C for 1 min and

AFLP primer combinations and selective nucleotides 72 ◦ C for 2 min, respectively. However, the annealing
MseI HindIII primers HindIII primers temperature changed according to primer pairs used.
primers labelled with IRD 700 labelled with IRD 800 The repeat sequences and annealing temperatures
ATG – AGG of primer pairs are given in Table 2. After adding
AAA – AGG loading-dye–buffer at a ratio of 1:8, 1 ␮l of final sam-
AAC – AGG
ple was loaded on acrylamide gel and analyzed on a
AAT – AGG
AGG – AGG LICOR 400 Automatic Sequencer (MWG Biotech,
AGA – AAA Ebersberg, Germany).
ACT – ACC
ACA – ATG
ACT – ATC
3. Results and discussion
ATA AAT AAA
AGA AAC AGT
AGC AAT ACG AFLP and microsatellite analyses are useful and
AGC AAC ACG powerful techniques for detecting genetic varia-
AAT AAC ACG tion between individuals (Thomas and Scott, 1993;
ATA AAT AGG
Rongwen et al., 1995). In this study, 23 AFLP
ACG AAT AGG
primer combinations and nine microsatellite primer
pairs were used in order to obtain more marker in-
2.4. Microsatellite analysis formation for the inference of genetic relationships
among R. damascena plants originating from dif-
The rose microsatellite markers were obtained from ferent plantations. AFLP analysis of 15 individuals
Concipio GmbH, Sangerhausen, Germany. Forward using 23 primer combinations provided a total of
and IRD labelled reverse primers were synthesized by 966 amplified bands with an average of 42 bands
MWG Biotech, Ebersberg, Germany. per primer combination. The average over all pop-
Each PCR reaction was prepared as follows: 25 ng ulations of bands amplified per primer combination
DNA, 5 pmol of each labelled reverse primer and un- varied from 25 (MseI–AAC/HindIII–AGG) to 98
labelled forward primer, 1 U Tag DNA polymerase, (MseI–AAT/HindIII–AGG). In none of the marker
2 ␮l of 10× reaction buffer, 1 ␮l of 1× W1-detergent, analyses any polymorphism could be detected among
2 ␮l of dNTPs (1 mM), 2 ␮l of magnesium chloride the plants analyzed. Marker patterns were always
(15 mM) in a total volume of 20 ␮l. identical (Fig. 1).
The PCR reaction was carried out in a Techne This is in accordance to earlier investigations with
Genius thermocycler. An initial denaturation of RAPD markers where no variability could be detected
94 ◦ C/30 s was followed by 30 cycles. Denaturation among R. damascena plants (Ağaoğlu et al., 2000).
It can be concluded that R. damascena plants from
different plantations in Isparta are all derived from the
Table 2
The repeat sequences and annealing temperatures of primer pairs
same original clone by vegetative propagation. How-
used in microsatellite analyses ever, the plants can be distinguished by morphologi-
cal characters. According to our observations through
Name Repeat Annealing temperature (◦ C)
the survey, there were some morphological differences
RMS023 GT 52 in flower shape, leaf size and plant size among the
RMS027 AT&GT 55 plants grown in different plantations. These morpho-
RMS029 GA 68
RMS037 GA 62
logical differences are most probably caused by point
RMS057 GAA/GA 58 mutations that are not detected by molecular mark-
RMS070 GA 57 ers, which only scan a very small fraction of the rose
RMS088 GA 68 genome.
RMS089 AT&GT 55 Previous observations have shown that sport mu-
RMS146 GT 57
tants with clearly mutated phenotypes show identical
266 N. Göktürk Baydar et al. / Journal of Biotechnology 111 (2004) 263–267

Fig. 1. AFLP fingerprints produced by HindIII–AAC/HindIII–ACG/MseI–AAT primer combinations (on the left) and microsatellite PCR
products amplified with RMS 070 primer pairs (on the left) for 15 Rosa damascena plants.

DNA marker patterns as the single mutations leading
to the altered phenotypes are derived from only a neg-
ligible small part of the DNA of the mutant genotype
(Debener et al., 2000).
From the results of the present study it can be con-
cluded that there is very little genetic variation among
R. damascena plants grown in Turkey.
For breeding of R. damascena varieties with im-
proved agronomic characters especially with improved
yields of rose oil and novel scent compositions two
major strategies could be followed. One strategy
could be the search for novel genotypes of R. dam-
ascena from other growing areas and the use of
these genotypes in classical breeding programs. This
strategy could benefit from methods for marker as-
sisted selection which were recently proposed for
roses (Debener et al., 2003). The second strategy
could involve novel biotechnological methods like,
e.g. the introduction of scent related genes into roses
via genetic transformation. However, this strategy is
hampered by technical problems encountered during
rose transformation and by public concerns related
to the use of genetically modified plants (Dohm,
2003).
Therefore, it is necessary to create genetic
variability especially for improving novel scent
 N. Göktürk Baydar et al. / Journal of Biotechnology 111 (2004) 263–267 267

types of R. damascena thanks to the classical and (Herausgeber) (Eds.), Encyclopedia of Rose Sciences. Elsevier
biotechnological breeding methods. Science, Oxford, UK.
Gudin, S., 2000. Rose: genetics and breeding. Plant Breed. Rev.
17, 159–189.
Hubbard, M., Kell, J., Rajapake, S., Abbott, A., Ballard, R., 1992.
References Restriction fragment length polymorphism in rose and their
use for cultivar identification. HortScience 27, 172–173.
Kovats, E., 1987. Composition of essential oils. Part 7. Bulgarian
Ağaoğlu, Y.S., Ergül, A., Baydar, N., 2000. Moleculer analyses
oil of rose (Rosa damascena Mill.). J. Chromatogr. 406, 185–
of genetic diversity oil rose (Rosa damascena Mill.) grown
222.
Isparta (Turkey) region. Biotechnol. Biotechnol. Eq. 14, 16–18.
Reynders-Aloisi, S., Bollereau, P., 1996. Characterization of
Antonelli, A., Fabbri, C., Giorgioni, M.E., Bazzocchi, R., 1997.
genetic diversity in genus Rosa by random amplified
Characterisation of 24 old garden roses from their volatile
polymorphic DNA. Acta Hort. 424, 253–259.
compositions. J. Agric. Food Chem. 45, 4435–4439.
Rongwen, J., Akkaya, M.S., Bhagwat, A.A., Lavi, U., Cregan,
Debener, T., Mattiesch, L., 1998. Effective pairwise combination
P.B., 1995. The use of microsatellite DNA markers for soybean
of long primers for RAPD analyses in roses. Plant Breeding
genotype identification. Theor. Appl. Genet. 90, 43–48.
117, 147–151.
Thomas, M.R., Scott, N.S., 1993. Microsatellite repeats in
Debener, T., Mattiesch, L., 1999. Construction of a genetic linkage
grapevine reveal DNA polymorphisms when analysed as
map for roses using RAPD and AFLP markers. Theor. Appl.
sequence-tagged sites (STSs). Theor. Appl. Genet. 86, 985–
Genet. 99, 891–899.
990.
Debener, T., Bartels, C., Mattiesch, L., 1996. RAPD analysis of
Thomas, M.R., Satsumoto, S., Cain, P., Scott, N.S., 1993.
genetic variation between a group of rose cultivars and selected
Repetitive DNA of grapevine: classes present and sequences
wild rose species. Mol. Breed. 2, 321–327.
suitable for cultivar identification. Theor. Appl. Genet. 86,
Debener, T., Janakiram, T., Mattiesch, L., 2000. Sports and
173–180.
seedlings of rose varieties analysed with molecular markers.
Torres, A.M., Millan, T., Cubero, J.T., 1993. Identifying rose
Plant Breeding 119, 71–74.
cultivars using random amplified polymorphic DNA markers.
Debener, T., Malek, B., Schreiber, M., Drewes-Alvarez, R., 2003.
HortScience 28, 333–334.
Marker assisted background selection for the introgression of
Tucker, A.O., Maciarello, M., 1988. Nomenclature and chemistry
black spot resistance into cultivated roses. Eur. J. Hort. Sci.
of the Kazanlak Damask Rose and some potential alternatives
68, 245–252.
from the horticultural trade of North America and Europe.
Dohm, A., 2003. Biotechnologies for Breeding: Genetic
In: Flavors and Fragrances: A world Perspective. Elsevier,
Transformation. In: Roberts, A., Debener, T., Gudin, S.
Amsterdam, pp. 99–114.