DNA variation and polymorphism in Tunisianplum species (Prunus spp): contribution offlow cytometry and molecular markersH. Ben Tamarzizt1, D. Walker2, S. Ben Mustapha1, D. Abdallah1, G. Baraket1,A. Salhi Hannachi1 and S. Zehdi Azzouzi1Laboratory of Molecular Genetics, Immunology & Biotechnology,Faculty of Sciences of Tunis, University of Tunis El Manar, Campus University,El Manar, Tunis, Tunisia2Departamento de Recursos Naturales,Instituto Murciano de Investigación y Desarrrollo Agrario y Alimentario (IMIDA),Estación Sericícola, Calle Mayor s/n, La Alberca, Murcia, Spain1Corresponding author: A. Salhi HannachiE-mail: [email protected] Mol. Res. 14 (4): 18034-18046 (2015)Received August 18, 2015Accepted October 25, 2015Published December 22, 2015DOI ACT. Plums (Prunus spp) are among the most important stonefruit crops in the world. European (Prunus domestica) and Japanese(Prunus salicina) plums are characterized by different levels of ploidy.Because genetic variability is the prerequisite for any plant-breedingprogram, we aimed to establish the taxonomic status of Tunisian plumsand study their genetic variability. The nuclear DNA content of 45 wild andcultivated Tunisian plums was determined by flow cytometry. Two arbitraryprimers (AD10, AD17) were used to elaborate SCAR markers useful toidentify plum species. Three wild trees, Zenou 1, Zenou 6, and Zenou3, which had 2C nuclear DNA contents of 1.99, 2.05, and 2.13 pg, wereshown to be hexaploid (2n 6x 48), whereas the others were diploid(2n 2x 16). These results suggest that the three hexaploid wild plumsGenetics and Molecular Research 14 (4): 18034-18046 (2015) FUNPEC-RP

DNA variation in Tunisian plum species18035belong to Prunus insititia, and the others belong to Prunus salicina. NoSCAR markers were revealed using the AD10 and AD17 RAPD primersin relation to the ploidy of plums. We note also that AD17 primer appearsto be the most informative concerning the genetic diversity. Morphologicaland pomological traits revealed similarity between introduced and Tunisianplum cultivars. Despite the significant morphological differences found,all the cultivars studied belong to P. salicina. The information obtained inthis analysis provided on local plum genetic resources will be helpful toestablish a core collection, to evaluate genetic diversity, and to initiate animprovement and selection program.Key words: Flow cytometry; Plum; Prunus insititia; Prunus salicina;Molecular markers; RAPDINTRODUCTIONPrunus is a large and diverse genus of plants that belongs to the subfamily Prunoideaeof the family Rosaceae (Rehder, 1940).The genus Prunus comprises about 400 species of treesand shrubs that produce drupes as fruits, commonly called “stone fruits” It is mainly found intemperate regions in both the northern and southern hemispheres, and it constitutes the third mosteconomically important group of plants in the temperate regions of the world. The large numberof Prunus species and the frequent interspecific hybridization make the systematic classificationin Prunus controversial (Dosba et al., 1994). Plum species occur at three levels of ploidy: diploid,tetraploid, and hexaploid. Prunus domestica L. (6x), which is one of the European plums, is thoughtto be derived from a natural cross between Prunus spinosa L. (4x) and Prunus cerasifera Ehrh (2x).The term ‘Japanese plum’ was originally appliedto Prunus salicina Lindl. (2x). The wild speciesin the Prunus genus constitute an important genetic resource and include species that are usedmedicinally, as rootstocks, as ornamentals, or for food (Pandey et al., 2008). The introduction ofpromising cultivars of different species of Prunus and their subsequent selection to fit agro-climaticregions have allowed considerable diversity to develop in major cultivated species, but this alsoleads to the evolution of new species and varieties and the extinction of local ones. In addition,the introduction of genes from related species through inter-specific hybridization has been usedin several breeding programs throughout the world in order to develop better-adapted cultivarsand rootstocks. Rootstock breeding programs that use inter-specific hybridization have introduceduseful traits, including size control, adaptation to new environments, and pest resistance, thusproducing numerous new varieties (Martinez-Gomez et al., 2003). Nevertheless, breeding barriersexist among taxa that possess different ploidy levels, even within the same section like the sectionPrunophora of the Prunus genus, but hybrids are generally successful when both parents havethe same ploidy level (Okie and Weinberger, 1996). In fact, many cultivated genotypes result fromcross-pollination making the systematic classification of numerous cultivars extremely difficult.Hence, knowledge of the taxonomic level is important to identify and recognize the gene pool ofplum species. Hybridization can induce rapid genomic changes and subsequent changes in theDNA content (Baack et al., 2005). Hence, in recent years, many molecular studies have beenestablished with the aim of identifying and characterizing plum species. Moreover, since tree fruitcultivars are maintained by vegetative propagation, accurate identification of vegetative material isGenetics and Molecular Research 14 (4): 18034-18046 (2015) FUNPEC-RP

H. Ben Tamarzizt et al.18036crucial for nurserymen and growers, and is needed for plant breeder’s rights (Goulao et al., 2001).Therefore, molecular markers, such as restriction fragment length polymorphism (RFLP) (Quartaet al., 1996), random amplified polymorphic DNA (RAPD) (Gregor et al., 1994; Warburton andBliss, 1996; Ortiz et al., 1997; Bellini et al., 1998; Shimada et al., 1998; Casas et al., 1999; Lisek etal., 2007; Li et al., 2007; Ben Tamarzizt et al., 2009), inter simple sequence repeat (ISSR) (Yilmazet al., 2009), simple sequence reapet SSR (Mnejja et al., 2004; Baraének et al.,2006; Bouhadida etal., 2009), and amplified fragment length polymorphism (AFLP) (Ilgin et al., 2009), and sequencesof non-coding region of chloroplast DNA (Ben Mustapha et al., 2013) have been tested. Knowledgeof chromosome number and ploidy level is important, especially in plant families and genera wherehybridization between species with different chromosome number or ploidy level occurs frequently.To clarify the taxonomic status of plums in Tunisia, we investigated the Tunisian germplasmby means of DNA quantification using flow cytometry, RAPD markers generated by AD10 and AD17primers as suggested by Ortiz et al. (1997) to advance their collection, management, and rationalutilization. Flow cytometry constitutes a convenient technique that can be used to study ploidylevels, by estimating the nuclear DNA content (Dolezel et al., 2007), and RAPD markers haveproven to be a reliable and useful molecular marker for genetic fingerprinting. The specific aimsof this study were to 1) identify the taxonomical status of Tunisian plums, 2) to detect duplicatedor mislabeled accessions, 3) to evaluate diversity in order to facilitate its use in breeding andin developing a collection strategy, and 4) to analyze the genetic relationship of plum species,focusing on the origin of local resources.MATERIAL AND METHODSPlant materialsForty-five accessions were considered, which represented plum species and their wildrelatives. All samples were collected from several localities in northern Tunisia (Ras Jebel, Rafraf, ElAlia, Sounine, Ghar El Melh, Douar Hamouda, Bejou, Cap bon, Thibar, Djebba, and Kairouan) (Table1).Genomic DNA extractionTotal cellular DNA was purified from young frozen leaves according to two procedures:Bernatzky and Tanksley (1986) and a modified procedure as described by Ahrens and Seemüller(1992). The DNA quality was examined by electrophoresis on 0.8% agarose gels, as described bySambrook et al.(1989), and the DNA concentration was quantified spectrophtometrically.Flow cytometry procedureEstimation of nuclear DNA content was performed with a Partec PA II flow cytometer(Partec GMBH, Münster, Germany). The method was based on the protocol described by Bukhari(1997). Samples of growing leaf tissue of Prunus and soya were prepared together. Soya has a 2Cnuclear DNA content of 2.50 pg. Leaf material was chopped with a razor blade for 30-60 s, in a 60mm plastic Petri dish containing 0.4 mL extraction buffer (Cystain PI absolute P, Partec GMBH), towhich polyvinylpyrrolidone-10 (2.5% w/v), ascorbic acid (12 mM), dithiothreitol (9 mM), and TritonX-100 detergent (0.25% v/v) had been added. The resulting extract was passed through a 30-mLGenetics and Molecular Research 14 (4): 18034-18046 (2015) FUNPEC-RP

DNA variation in Tunisian plum species18037filter into a 15-mL centrifuge tube. The Petri dish was washed twice with 0.8 mL extraction bufferand the samples were thenfiltered into the 15-mL tube. After centrifugation at 1.100 g for 10 min, thesupernatant was removed and the pellet was re-suspended in 1.6 mL Cystain PI absolute P stainingbuffer (Partec GMBH) to which propidium iodide and RNase had been added (final concentrations of50 and 17.5 µg/mL, respectively). All stages of the extraction were performed at 4 C. The sampleswere kept in the dark for 15 min at 37 C, before being filtered through a 30-mL filter. The linearity ofthe cytometer fluorescence scale was checked regularly using propidium iodide-stained calibrationbeads (Partec GMBH). At least 5000 nuclei were analyzed in each sample. The equivalent numberof base pairs was calculated assuming that 1 pg DNA 978 Mbp (Dolezel et al., 2007; Greilhuber etal., 2007). One-way ANOVA was performed using SAS (1990), version 6.12. The mean nuclear DNAcontent was tested by the Student-Newman-Keler test (5%), using SPSS v.11.0.Table 1. Tunisian plums studied and their locality of origin.CultivarLocalityBedri1Japonia safraJanhaAin kounouliaCidre1Adham hmémNeb zaroukHamdaAouina hamra badriAin thaer nomanGolden JapanAin torkiaSanta Rosa1Aouina safra morraZaghwéniaAin tasstouriaAin ben moussaBaydha arbiMeski kahla1Meski safra1Meski kahla2Bedri2Bedri hamra1Black GoldBedri hamra2Bedri hamra3Bedri hamra4StanleyGolden Japan2Sauvage1ChaaraouiyaZenou5Zenou1Cidre 2Jelya1Golden Japan3Zenou6Zenou3Jelya2Zenou7Sauvage3Black diamondSandidSafra jridiSanta Rosa2Ras rafRafrafEl rafRafrafRafrafRas apbonThibarGhar el melhDouar HamoudaDouar HamoudaGhar el melhBejouGhar el melehDouar hamoudaDouar HamoudaBejouDouar HamoudaDjebbaCapbonRafrafRafrafCapbonGenetics and Molecular Research 14 (4): 18034-18046 (2015) FUNPEC-RP

H. Ben Tamarzizt et al.18038RAPD analysisTwo RAPD primers were tested: AD10 (AAGAGGCCAG) and AD17 (GGCAAACCCT)according to Ortiz et al. (1997), these primers produce specific patterns for diploid and hexaploidspecies. The reactions were carried out in a 25 µL volume reaction mixture containing 20 ngtotal cellular DNA, 50 pM primer, 2.5 µLTaq DNA polymerase buffer, 1.5 U Taq DNA polymerase(QBIOgéne, France), and 200 mM dNTP (DNA polymerization mix, Pharmacia). The PCRs werethen performed in a DNA thermocycler (TC 512, TECHNE) programmed to execute the followingcycles: reaction mixtures were heated at 94 C for 5 min as an initial denaturation step before entering35 cycles, each composed of 30 s at 94 C, 1 min at 35 C, 1 min at 72 C, and a final step of 5 minat 72 C. To reduce the possibility of cross contamination in the amplifications, reaction mixtureswithout DNA were used as negative controls. Only reproducible products were taken into accountfor further data analysis. The products of amplification were separated on 1.5% agarose gel at 100mV for 2 h in 0.5X TBE buffer and detected after ethidium bromide staining according to the methoddescribed by Sambrook et al. (1989). To confirm the results, acrylamide gels (10%) were also used.Data analysisPolymorphic RAPD bands were scored as present (1) or absent (0) across the 45genotypes for two RAPDs primers as a binary data matrix. The percentage of polymorphic bands(PPB) was estimated and the ability of RAPD tested primers to differentiate between plums wasappreciated by the estimation of their resolving power (Rp) (Prevost and Wilkinson, 1999) The Rphas been described by Gilbert et al. (1999) such that: Rp Ib where: Ib 1 - (2 x 0.5 -p ) where pis the accessions proportion containing the I band. The generated binary matrix was computed withthe Gendist program (version 3.572c), using the computer program PHYLIP (phylogeny inferencepackage, version 3.5c) (Felsenstein 1995), producing a genetic distance matrix according to theformula described by Nei and Li (1979). The neighbor program produces a tree-file using theunweighted pair group method with the arithmetic averaging (UPGMA) algorithm.RESULTSDNA quantificationQuantification of DNA by flow cytometry (Table 2, Figure 1) suggests that genome sizeof the Tunisian plum cultivars studies varies significantly. Of the 45 accessions studied, three wildtrees ‘Zenou 1’, ‘Zenou 6’, and ‘Zenou 3’, which had 2C nuclear DNA contents of 1.99, 2.05, and2.13 pg, respectively, were shown to be hexaploid (2n 6x 48), whilst the others (0.44-0.97 pgDNA) were diploid (2n 2x 16). The wild tree ‘Zenou1’ contained the highest number of bases(2083.1 Mbp) and cultivar ‘Hamda’ contained the lowest, 432 Mbp (Table 3). The results also showthat the ploidy of the introduced cultivars: ‘Black gold’; ‘Black diamond’; ‘Golden Japan’, and ‘SantaRosa’ were diploid and included in P. salicina.Molecular markersHere we used two RAPD primers that were suggested by Ortiz et al. (1997) to be ableGenetics and Molecular Research 14 (4): 18034-18046 (2015) FUNPEC-RP

DNA variation in Tunisian plum species18039to differentiate between diploid and hexaploid plum and generate specific SCAR markers that areuseful to define the taxonomy statute of Tunisian plum species. The results of the RAPD analysisof 54 plum genotypes are given in Table 3 and Figure 2. Molecular polymorphism was revealed,as demonstrated by RAPD patterns and no specific markers were detected in relation to the ploidylevel of plums. Nine and 10 RAPD markers were generated by the AD10 and AD17 primers. Percentpolymorphic bands were 90 and 100 for AD10 and AD17, respectively. The AD17 primer appears tobe the most informative with a resolving power of 6.4 (Rp of AD10 5.5).Table 2. 2C nuclear DNA contents, the equivalent number of base pairs, and the corresponding ploidy for thePrunus cultivars studied.CultivarAin kounouliaAin torkiaAin thaer nomanAin ben moussaAin tasstouriaAouina hamra badriAouina safra morraAdham hmémBaydha arbiBedri1Bedri2Bedri hamra1Bedri hamra2Bedri hamra3Bedri hamra4JanhaHamdaJaponia safraSafra jridiNeb zaroukSandidZaghwéniaCidre1Cidre2Meski kahla1Meski kahla2Meski 1Zenou3Zenou5Zenou6Zenou7Black GoldBlack diamondStanleyGolden Japan1Golden Japan2Golden Japan3Santa Rosa1Santa Rosa2DNA .6150.5870.7650.7150.5510.699DNA iploidDiploidDiploidFollowing ANOVA (P 0.001 for the effect of “cultivar”), the LSD values obtained (P 0.05) were 0.184 pg for thenuclear DNA content and 180 Mbp for the equivalent number of base pairs.Genetics and Molecular Research 14 (4): 18034-18046 (2015) FUNPEC-RP

H. Ben Tamarzizt et al.18040Figure 1. Histogram of the relative nuclear DNA content of Prunus plants 13 (Prunus salicina; cultivar Ain torkia) and45 (Prunus insititia; cultivar Zenou6), determined by flow cytometry analysis of propidium iodide-stained nuclei withsoya (Glycine max; 2C nuclear DNA content 2.50 pg) as an internal standard. Nuclei of Prunus and soya leaves wereisolated, stained, and analyzed simultaneously.The genetic distances (Nei and Li, 1979) ranged from 0.00 to 1.33 with a mean of 0.53,which suggest a high level of polymorphism at the genomic DNA level of the studied accessions.The lowest distance value (0.00) was observed between [‘Ain kounouliya’ and ‘Adham hmem’];[‘Ain kounouliya’ and ‘Golden Japan1’]; [‘Cidre1’ and ‘Golden Japan1’]; [‘Meski kahla1’ and ‘Bedrihamra3’]; [‘Meski safra1’ and ‘Bedri hamra2’] cultivars, which seem to be closely related. While thehighest distance of 1.33 was calculated between [‘Neb zarrouk’ and ‘Meski safra1’]; [‘Meski kahla1’and ‘Sauvage1’]; [‘Bedri1’ and ‘Sauvage1’]; [‘Ain torkia’ and ’Safra jridi’]; [‘Golden Japan2’ and ‘Safrajridi’] accessions, suggesting their divergence. Topology of the UPGMA dendrogram (Figure 3) showsthat varieties can be classified into two main clusters, the first one labeled (I) is subdivided into twosubgroups (Ia and Ib), which contain wild and introduced cultivars, respectively. The second groupis divided into two major subgroups (IIa and IIb), which contain the remaining accessions analyzed.Some cultivars presented an important similarity [‘Meski safra1’ and ‘Meski kahla2’]; [‘Meki kahla1’and ‘Bedri2]; [‘Ain taher noman’ and ‘Ain kounouliya’ ‘Cidre1’]; [‘Sandid’ and ‘Safra jridi’] despite theirappellation; these may be explained by misidentification or homonymy problems.Genetics and Molecular Research 14 (4): 18034-18046 (2015) FUNPEC-RP

DNA variation in Tunisian plum species18041Table 3. Random amplified polymorphic DNA (RAPD) markers generated by AD10 and AD17 primers.CultivarAin kounouliaAin torkiaAin thaer nomanAin ben moussaAin tasstouriaAouina hamra badriAouina safra morraAdham hmémBaydha arbiBedri1Bedri2Bedri hamra1Bedri hamra2Bedri hamra3Bedri hamra4JanhaHamdaJaponia safraSafra jridiNeb zaroukSandidZaghwéniaCidre1Cidre2Meski kahla1Meski kahla2Meski 1Zenou3Zenou5Zenou6Zenou7Black GoldBlack diamondStanleyGolden Japan1Golden Japan2Golden Japan3Santa Rosa1Santa Rosa2AD10AD171140 bp890 bp520 bp - - - : presence; -: absence of bands.DISCUSSIONLittle is known about the taxonomic status of the Tunisian plum germplasm.Results of DNAquantification suggest that the three hexaploid wild plums belong to the species Prunus insititia andall the other cultivars to Prunus salicina, taking into account their diploidy. The wild tree ‘Zenou1’,which belongs to P. insititia, had the highest number of bases (2083.1 Mbp) and cultivar ‘Hamda’(P. salicina) had the lowest number of bases (432 Mbp). The results also confirm the ploidy ofthe introduced cultivars: ‘Black gold’; ‘Black diamond’; ‘Golden Japan’, and ‘Santa Rosa’ werediploid and included in P. salicina,as suggested by Goulao et al. (2001). The cultivar ‘Stanley’ wasthought to belong to the hexaploid species P. domestica; however, flow cytometry showed thatGenetics and Molecular Research 14 (4): 18034-18046 (2015) FUNPEC-RP

H. Ben Tamarzizt et al.18042this cultivar is diploid and must be included in P. salicina. In this case, flow cytometry permittedthe detection and resolution of mislabeling problems. All the remaining cultivars were found to bediploid. In fact, the accessions ‘Sauvage1’, ‘Sauvage3’, ‘Chaaraouiya’, ‘Jelya1’, and ‘Jelya2’, whichare considered wild cultivars, do not belong to the spontaneous species P. insititia or P. spinosa. Allthe other cultivars are diploid and could belong to P. salicina. These findings corroborate the workof Mzali et al. (2002) on the basis of morphological characteristics.Figure 2. Random amplified polymorphic DNA (RAPD) patterns generated by the AD17 primer. Lanes: L Ladder (100bp; Invitrogen); 41 Zenou1; 42 Cidre2; 43 Jelya1; 44 Golden Japan3; 45 Zenou6; 46 Zenou3; 47 Jelya2;48 Zenou7; 50 Sauvage3; 51 Black diamond; 52 Sandid.Polymorphisms at the DNA level have been used in several studies to examine geneticdiversity in plums. Previous molecular studies on plums using RAPD markers revealed wide geneticpolymorphism among accessions (Gregor et al.,1994; Ortiz et al., 1997; Bellini et al., 1998; Shimadaet al., 1999; Lisek et al., 2007; Liu et al., 2007; Ben Tamarzizt et al., 2009), which was explained bythe floral biology and different ploidy levels. As demonstrated by Ortiz et al. (1997), use of the arbitraryprimers, AD10 and AD17, yields polymorphic amplification products that are specific to the diploidor hexaploid species. In fact, primer AD10 produces a fragment of approximately 1140 bp, which ispresent only in the Japanese plum. The primer AD17 produces only two patterns, one specifically forthe European and the other for the Japanese plum cultivars. These patterns were distinguished byone fragment of approximately 890 bp that is specific to Japanese cultivars and an other amplificationproduct of approximately 520 bp that is characteristic of the European cultivars (Ortiz et al., 1997).Here, we used the AD10 and AD17 primers to reveal the specific markers.Genetics and Molecular Research 14 (4): 18034-18046 (2015) FUNPEC-RP

DNA variation in Tunisian plum species18043Figure 3. UPGMA dendrogram based on RAPD markers showing the relationships among plum cultivars.The results reported here are not consistent with those of previous studies investigatingRAPD markers as descriptors of the diploid and hexaploid plum species, since the amplifiedfragments (1140 bp produced by AD10 and 890 bp produced by AD17), which should identify diploidtrees (Ortiz et al., 1997), are also amplified in the hexaploid ‘Zenou1’; ‘Zenou3’, and ‘Zenou6’.Similarly, the 1140-bp band produced by the AD10 primeris absent in some diploid samples:‘Japounia safra’; ‘Janha’; ‘Ain kounoulia’; ‘Cidre1’; ‘Neb zarouk’; ‘Ain Taher noman’; ‘Ain torkia’;‘Bedri hamra1’; ‘Bedri hamra2’; ‘Black gold’; ‘Stanley’; ‘Cidre2’, and ‘Santa Rosa 2’. Additionally,the AD17 primer that is used to amplify a specific band present in diploids (890 bp) was also tested(Figure 2). This 890-bp fragment was obtained in all cultivars except ‘Bedri1’; ‘Ain kounoulia’,‘Ain taher noman’, ‘Santa Rosa1’, ‘Aouina safra morra’, ‘Ain zaghwénia’, ‘Aouina arbi baydha’,‘Bedri2’, ‘Stanley’, ‘Zenou5’, ‘Zenou7’, ‘Cidre2’, ‘Jelya2’, ‘Sandid’, and ‘Santa Rosa 2’. TheseGenetics and Molecular Research 14 (4): 18034-18046 (2015) FUNPEC-RP

H. Ben Tamarzizt et al.18044results contradic those obtained by flow cytometry. Indeed, according to the flow cytometry results,‘Bedri1’, ‘Ain kounoulia’, ‘Ain Taher noman’, ‘Santa Rosa1’, ‘Aouina safra morra’, ‘Ain zaghwénia’,‘Aouina arbi baydha’, ‘Bedri2’, ‘Stanley’, ‘Zenou5’, ‘Zenou7’, ‘Cidre2’, ‘Jelya2’, ‘Sandid’, and ‘SantaRosa 2’ cultivars are diploids. Similarly, the characteristic 520bp fragment, which should be presentonly in hexaploid plums and is amplified using the AD17 primer (Ortiz et al., 1997), was observedin the diploids ‘Japounia safra’, ‘Ain torkia’, ‘Ain zaghwénia’, ‘Ain tasstouria’, and ‘Jelya1’. Theseresults do not confirm the findings of Ortiz et al. (1997) and the specific RAPD markers do not allowthe generation of SCAR markers to recognize plum species. Flow cytometry data reveals thatthere is intra- and inter-specific DNA variation (Table 3). Trees of the same variety did not presentequivalent DNA contents, namely [‘Cidre1’; ‘Cidre2’], [‘Golden Japan1’; ‘Golden Japan2’; ‘GoldenJapan3’], and [‘Santa Rosa1’; ‘Santa Rosa2’]. This variation is also observed among the wild trees[‘Zenou1’; ‘Zenou3’; ‘Zenou5’; ‘Zenou6’; ‘Zenou7’].Cluster analysis of plum cultivars revealed a strong distinctness of the genotypes fromdifferent geographical regions. As shown by Figure 3, plums are grouped independently of theirploidy level. Cultivar distribution occurs separately from their geographic origins, so typicallycontinuous genetic diversity characterizes local plum germplasm. Additionally, we note thathexaploid cultivars [‘Zenou1’, ‘Zenou3’, and ‘Zenou6’] do not diverge from diploid cultivars, whichconfirms the previous results.Comparison between phenotypic analysis and flow cytometryMorphological analysis of 20 accessions was performed and important inter-cultivarphenotypic variability was observed by Ben Tamarzizt et al. (2009). Principal component analysis(PCA ) was performed using 25 morphological and pomological parameters and showed that therewas similarity between the introduced variety ‘Santa Rosa’ and the local variety ‘Cidre1 according totheir pomological traits related to fruit and seed characteristics: ‘Fruit form’, ‘Skin color’, ‘Firmness’,‘Juiciness’, and ‘Acidity’. The introduced cultivars do not differ from the Tunisian ones, indicatingthe good performance of local cultivars (Ben Tamarzizt et al., 2009). It is important to note that flowcytometry clustered these varieties in the same group of diploid trees, belonging to the Japanesespecies P. salicina. Thus, this cultivar is more important than the introduced one ‘Santa Rosa’,which have resulted from selection by American breeders such as Luther Burbank since 1883(Shimada et al., 1999) and Wellington (USA) (Anonymus, 2002; Ben Tamarzizt et al., 2009).Furthermore, significant divergence between local cultivars was observed, especially between “AinTasstouria” and “Meski safra 1”. In fact, these two Tunisian plums are morphologically distantaccording to their leaf, branch, fruit, and seed parameters: ‘Leaf length’, ‘Branch length 2006’,‘Branch length 2007’, ‘Fruit length’, ‘Fruit width’, ‘Fruit weight’, ‘Fruit form’, ‘Skin color’, ‘Pulp color’,‘Firmness’, ‘Acidity’, ‘Aroma’, ‘Sweetness’, and ‘Seed form. Furthermore, similar varieties, withsimilar morphological traits, were observed, especially for trees of the same varieties [‘Meski kahla1’; ‘Meski kahla 2’] according to their branch, leaf, fruit, and seed parameters. Additionally, theclusters were independent to the geographic origin of the plum cultivars. In this analysis, dispersionof cultivars in the PCA plot appears without any clear aggregation correlated to geographical origin.Despite major differences in morphological appearance, all tested trees have the same ploidy leveland belong to the Japanese species P. salicina. In fact, many pomological traits were discriminatingand characterized each variety. This result can be explained by the environmental adaptation ofdifferent varieties.Genetics and Molecular Research 14 (4): 18034-18046 (2015) FUNPEC-RP

DNA variation in Tunisian plum species18045Morphological study has revealed wide genetic polymorphism among plum accessions,because plums are a very complex group, which includes diploid, tetraploid, and hexaploidspecies, and because floral biology differs among plum groups (Erturk et al., 2009). The distributionof cultivars occurs independently from their geographic origin. We also note that introduced andTunisian plums are clustered together in the PCA plot, confirming the efficiency of the localgermplasm. These results underline the importance of preserving the genetic resources of plumspecies since this may enable breeders to select the most diverse genotypes, with interesting fruitcharacteristics, for crossing and selection programs.Conflicts of interestThe authors declare no conflict of interest.ACKNOWLEDGMENTSResearch supported by grants from the Tunisian ‘‘Ministère de l’Enseignement Supérieuret de la Recherche Scientifique’’.REFERENCESAhrens U and Seemüller E (1992). Detection of DNA of plant pathogenic mycoplasma-like organisms by a polymerase chainreaction that amplifies the sequence of the 16S rRNA gene. Phytopathology 82: 828-832.Anonymus (2002). Arboricultures fruitières. Variétés fruitières reccomandées en Tunisie. Documents techniques INRAT n 114.Baack EJ, Whitney KD and Rieseberg LH (2005). Hybridization and genome size evolution: timing and magnitude of nuclearDNA content increases in Helianthus homoploid hybrid species. New Phytol. 167: 623-630.Baraének M, Raddová J and Pidra M (2006). Comparative analysis of genetic diversity in Prunus L. as revealed by RAPD andSSR markers. Sci. Hortic. 108: 253-259.Bellini E, Giordani E, Nencetti V and Paffetti V (1998). Genetic releationships in Japanese plum cultivars by molecular markers.Acta Hortic. 478: 53-69.Ben Mustapha S, Ben Tamarzizt H, Baraket G, Abdallah D, et al. (2013). A signature of balancing selection in the plastid trnLUAA intron and investigation of the phylogenetic relationships among Tunisian plum species (Prunusspp).Sci. Hortic.151:51-56.Ben Tamarzizt H, Baraket G, Ben Mustapha S, Marrakchi M, et al. (2009). Genetic relatedness among Tunisian plum cultivarsby random amplified polymorphic DNA analysis and evaluation of phenotypic characters. Sci Hortic. 121: 440-446.Bernatzky R and Tanskley SD (1986). Genetics of actin-related saquences in tomato. Theor. Appl. Genet. 72: 314-321.Bukhari YM (1997). Cytoevolution of taxa in Acacia and Prosopis (Mimosaceae). Australian J. Bot. 45: 879-891.Casas AM, Igartua E, Balaguar G and Moreno MA (1999). Genetic diversity of Prunus rootstocks analyzed by RAPD markers.Euphytica 110: 139-149.Dolezel J, Greilhuber J and Suda J (2007). Estimation of nuclear DNA content in plant using flow cytometry. Nat. Protoc. 2:2233-2244.Dosba F, Bernhard

Flow cytometry constitutes a convenient technique that can be used to study ploidy levels, by estimating the nuclear DNA content (Dolezel et al., 2007), and RAPD markers have . Estimation of nuclear DNA content was performed with a Partec PA II flow cytometer (Partec GMBH, Münster, Germany). The method was based on the protocol described by .