Molecular identification of Rosa x damascena growing in Taif region ( Saudi Arabia )

A fragment of 772 bp of the chloroplast maturase K gene was amplified and sequenced for Rosa x damascena trigintipetala variety growing in Taif region of Saudi Arabia. The data were aligned with their counterparts of other varieties already found in the Genbank database and were analyzed by maximum-parsimony, neighbor-joining and maximum-likelihood methods and a single rooted tree was executed. R. x damascena trigintipetala was paraphyletic where one sample [A] clustered with all varieties while the second [B] was basal. R. x damascena was sister to R. x chinensis semperflorens with the later being basal. R. x damascena gori was basal for all taxa studied. R. moschata was inside the clade of R. x damascena. Hybridization could be possible among R. damascena, R. chinensis and R. moschata. The genetic distance and tree topology indicated that [A] variety could be originated from R. moshata while [B] could be originated from gori or R. chinensis semperflorens. We, therefore, may consider that R. x damascena gori or R. chinensis could be the origin of all nowadays R. x damascena varieties.


Introduction
Rosa is a genus of an ornamental plant containing hundreds of species and is one of the most cultivated plants in the world. 1 Roses are economically valuable as they contain essential oils which are a source of perfumes and scents. 2 The researchers have identified significant pharmacological properties of rose extracts as it was used in radioprotective 3 and anti-inflammatory 4 purposes.Compounds like polysaccharides, polyphenols and other secondary metabolites are found with high level in rose plants. 5They tightly bind with DNA and inhibit its isolation. 6[9] Molecular study on the genetic framework of these plants requires large yield of DNA and therefore, method described by Ibrahim 10 was used for this purpose with few modifications.
The inter-specific relationships within Rosa remain controversial.Different names were given to Rosa species based on the morphological variants and hybrids. 11The genetic studies on the genus were primarily based on isozymes 12 and random amplification of polymorphic DNA (RAPD)-PCR. 13,14Microsatellite analyses for both wild and cultivated roses were also conducted by Baydar and colleagues. 157][18] The chloroplast matK gene 16,19 and the atpB-rbcL intergenic spacer 18 were conducted for phylogenetic resolution of the genus and its subgenera, however, the genetic controversial is still standing.These investigations explained partly the extremely low levels of sequence divergence observed across the genus. 18,19On the other hand DNAbased molecular marker systems are efficient and informative for genetic analysis of roses because DNA polymorphism indicated by these markers is not affected by environmental conditions.Several molecular marker systems, i.e., Restriction Fragment Length Polymorphism (RFLP), 20 RAPD, 21,22 Simple Sequence Repeat (SSR) 23 and Amplified Fragment Length Polymorphism (AFLP) 15 have been used for genetic analysis of rose species, cultivars and rootstocks.
R. x damascena is a famous species within this genus that possesses different varieties of which R. d. trigintipetala and R. d. semperflorens.R. x damascena trigintipetala (the socalled Ward taifi) is highly cultivated in Taif governorate and there are different debates about its origin.This Taif-rose is the famous rose cultivated in different areas of Taif city at the western region of Saudi Arabia and is also known as Damask rose.There are two varieties of Damask Rose, such as Autumn Damask rose (R. x damascena semperflorens) and Damask rose Kazanlik (Rosa x damascena trigintipetala).It is known that Kazanlik in Bulgaria and Isparta in Turkey are the main areas of origin of R. x damascena (references cited in El-Assal and colleagues). 24Moreover, Taif-rose has been suggested to be brought to Taif from Balkans by Turks, who occupied this area in the 14th century.However, El-Assal and colleagues 24 indicated that Taif-rose has closed genetic relations to the gory rose-Syrian cultivated in Syria.It was therefore necessary to conduct PCR amplification and sequencing for matK gene in order to identify genetically Taifrose and construct its relationship to other varieties.

Samples collection and DNA extraction
Fresh young leafs of 8 samples of R. x damascena were collected from the Botanical Garden at Faculty of Science, Taif University, Taif, Saudi Arabia (5 samples from Taif variety Rosa x damascena trigintipetala, 2 samples from Rosa x damascena gori and one sample from Rosa x damascena French variety).Approximately, 300 mg of the leaf tissues were ground separately to a fine powder under liquid nitrogen with the use of pre-chilled mortar and pestle.The method detailed by Ibrahim 10 (CTAB-PVP-βME) was used to extract DNA form the plant tissues with some modifications.These modifications were that the DNA precipitate was dissolved in 100 µL bi-distilled water.Finally, spin column (BioFlux, Tokyo, Japan) was used to purify the extracted DNA according to the manufacturer instructions.PCR products were visualized under UV light and photographed (Figure 1).The products were then purified using spin column (BioFlux, Tokyo) according to the manufacturer instructions.

Phylogenetic analysis
The sequenced matK gene fragment of R. x damascena trigintipetala and R. damascena gori was aligned manually with its counterparts of other related taxa collected from the database (see their accession numbers on the tree) and the aligned data were used for phylogenetic analysis with a total length of 772 bp.The same sequence for R. californica was used for rooting the tree.The phylogenetic analyses were done primarily by maximum-likelihood (ML) method with PAUP* 4.0b10 25 by heuristic searches with the nearest-neighbor interchange (NNI) branch swapping and 10 random taxon additions.The general reversible model (GTR+I) and parameters optimized by Modeltest 3.0 were used.Bootstrapping replicates were set to 300.Maximum-parsimony (MP) and neighbor joining (NJ) were also conducted with PAUP* 4.0b10 under bootstrapping replications of 10000 to confirm the robustness of the ML tree topology.The distance option of Tamura-Nei was adjusted for executing the neighbor-joining analysis.

Results
No morphological characteristics were observed to differentiate between any of R. x damascena samples and all specimens from each variety were identical.The extracted DNA of the studied samples was tested by electrophoreses using 1% agarose and all samples gave positive results (data not shown).The amplified fragments of the matK gene were also examined on the agarose gel electrophoreses and the PCR profile was shown in Figure 2.
Unambiguous nucleotides of 772 bp of matK gene were sequenced for the collected samples.The base compositions of these data were A=31.04%,C=17.50%,G=14.50% and T=36.96%.Among the 772 sites included for tree analysis, 766 were constant and 6 were variables.Five variable sites were parsimony uninformative and only 1 was informative Article Table 1.Pairwise genetic distances among the different varieties of R. damascena populations.The distances were calculated from matK data sequenced in this study.under parsimony criterion.The constructed parsimony tree showed consistency index (CI=1.00),homology index (HI=0.00),retention index (RI=1.00)and rescaled consistency index (RC=1.00).

R. x damascena trigintipetala
The ML analysis exhibited a negative log likelihood score -lnL=1049.6.Both MP and NJ methods executed a similar tree topology to that given by ML method.The three methods gave the same bootstrap probabilities at the node clustering all varieties (Figure 1).The best-fit model that explained the datasets was GTR+I exhibiting model parameters as follows: substitution rate matrix R (a)=3.97;R (b)=2.75;R (c)=0.00;R (d)=5.62,R (e)=2.76 and R(f)= 1.00.The proportion of invariable sites (I) was 0.947.
In the NJ tree (Figure 1), a clustering of all varieties in one clade was shown with R. x damascena trigintipetala being paraphyletic.One sample of R. x damascena trigintipetala [A] clustered with all available varieties while the second sample [B] was basal to this clus-ter.All varieties of R. x damascena showed a sister relationship to R. x chinensis semperflorens with the later being basal.R. x damascena gori appeared to the basal for all R. x damascena studied.R. moschata was found to be inside the clade of R. x damascena.It seemed, therefore, that there was a possible hybridization among the three taxa of R. damascena, R. chinensis and R. moschata.
The pairwise genetic distances among the studied varieties are recorded in Table 1.The distance showed an identity among R. moschata, R. x damascena trigintipetala [A] and the other varieties of R. x damascena (D=0.000).On the other hand, the second variety of R. x damascena trigintipetala [B] was similarly distant from the [A] variety, the other R. x damascena varieties and R. moshata with a distance of 0.003.R. x damascena gori and R. chinensis semperflorens were distant from all studied varieties with genetic a distance 0.0013.
The genetic distance together with the tree topology indicated, therefore, that one of both R. x damascena trigintipetala varieties [A] could be originated from R. moshata while the other [B] could be originated from the gori or R. chinensis semperflorens.We, therefore, may consider that R. x damascena trigintipetala the so-called Ward taifi was not the origin of all nowadays R. x damascena varieties but the gori or R. chinensis could be.
The sequenced matK fragment was aligned with the complete gene sequence (1512 bp) obtained from the Genbank database.This fragment was located between base 75 and base 846 with a length of 772 bp.The sequence showed only 2 base substitutions within R. x damascena trigintipetala [B].One of the two substitutions was a transition within pyrimidines in the third position of the codon where C 114 was substituted with T 114 .The second substitution was a transversion, in the second position, of C 220 to G 220 .Only the later substitution was non-synonymous recording a change of the amino acid number 74 from leucine into valine at C 220 G 220 (Figure 3).This mutation makes R. x damascena trigintipetala variety [B] close to both the gori and R. chinensis semperflorens while the [A] variety is identical to the other varieties of R. x damascena.The number below the letter referred to the corresponding position of the nucleotide inside the complete gene sequence.

Discussion and Conclusions
Using the protocol detailed by Ibrahim 10 with some modifications described herein, DNA necessary for molecular genetics and biodiversity applications were successfully extracted from eight different Taif-rose samples containing high levels of ployphenolics and secondary metabolites.According to the protocol of CITAB mixed with PVP and ME followed by DNA purification kit, a high pure yield of DNA from fresh leaf tissue grinned with liquid nitrogen was obtained.The DNA quantity and purity obtained herein was comparable to that obtained by other methods. 6,7,9ood DNA quality and quantity was clearly shown.We did not measure the A/A ratio (absorbance at 260/ absorbance at 280) and we went directly to the PCR amplification to ensure the little contamination of proteins, polysaccharides or polyphenolics.This extraction procedure was found suitable for direct PCR amplifications of nuclear or chloroplast DNAs.
In the present procedure, β-mercaptoethanol and high concentration of EDTA in the extraction buffer made some temporary pores in the cytoplasm membrane, eliminating some of soluble polysaccharides, polyphenols and pigments improving the DNA yield of the rose plants and similar ornamental plants. 5,6ticle The use of high levels of PVP and β-mercaptoethanol had a considerable effect in preventing oxidization of the secondary metabolites.Considering extraction time, DNA yield and purity and PCR amplifications by this procedure emerged to be the best for molecular diversity analysis of roses.Because different plants can vary considerably in the number and types of produced secondary compounds, it is unlikely that any one technique for isolating contaminant-free nuclear DNA will ever be developed.It is likely that our rose DNA isolation protocol can be used to isolate nuclear DNA from a variety of other plant species.R. x damascena was believed to be originated in the Middle East as a wild plant and was introduced to Western Europe twice approximately 500 years ago, 27 however, the present results could not support this suggestion.Rusanov and colleagues 28 stated that R. x damascena trigintipetala was the common ancestor genotype which has been propagated in Europe and Asia, however, the paraphyly of this variety in the present study disagreed with the authors statement.We may agree with Huxely, 29 who revealed that R. x damascena originated partly as a hybrid from R. moschata.The author mentioned that semperflorens is one variety of R. x damascena flowering during autumn.Based on random amplified polymorphic DNA, inter simple sequence repeat and simple sequence repeat markers Taif-rose was shown to have closed genetic relations to the gory rose. 24These results are in agreement with the present finding where R. x damascena gori was shown to be basal in a clade including both varieties of R. x damascena trigintipetala.
In conclusion, Taif-rose (Ward taifi) or R. x damascena trigintipetala is not considered as one homogenous variety but seemed to be two varieties.One variety is originated from R. moschata while the other is originated from R.
x. chinensis semperflorens or from R. x damascena gori.The present study is considered a preliminary trial for investigating the genetic variability within R. x damascena trigintipetala.However, further investigation is needed to clearly infer the origin of Taif varieties of R. x damascena trigintipetala.Such a study must be conducted on different collections of Taif-rose and on more molecular data.

(
Promega Corporation, Madison, WI) and 10.3 µL autoclaved deionized water.PCR was carried out using a PeX 0.5 thermal Cycler with the cycle sequence at 94°C for 5 min one cycle, followed by 35 cycles each of which consisted of denaturation at 94°C for one min, annealing at 50°C for one min and extension at 72°C for one min.A final strand elongation for one cycle at 72°C was done for additional 5 min.The PCR products were analyzed in 1% agarose gel electrophoresis in TAE buffer (40 mm Tris, 40mM acetic acid and 1 mM EDTA) with ethidium bromide staining.A 100-bp DNA ladder (Biolabs) was used as a molecular marker.

Figure 1 .
Figure 1.A neighbor-joining tree constructed from 722 bp matK gene fragment sequenced in this study.The values at some nodes refer to the bootstrapping of maximum-parsimony, neighborjoining and maximum-likelihood analyses.The numbers between brackets are the accession numbers of the sequences collected from the Genbank database.N o n c o m m e r c i a l u s e o n l y

Figure 3 .
Figure 3.The aligned amino acid sequences for the two varieties of R. x damascena trigintipetala [A] and [B] along with that of R. x damascena [C] found in Genbank database.The underlined letter showed the non-synonymous change in the variety [B].Stars indicate sequence identity.The missed data from both varieties A and B are denoted by X.