Genetic Study of Malvasia and Torrontes Groups through Molecular Markers

Transcription

1 Genetic Study of Malvasia and Torrontes 125 Genetic Study of Malvasia and Torrontes Groups through Molecular Markers J. Borrego, M.T. de Andrés, J.L. Gómez, and J. Ibáñez* Two important grapevine groups, Malvasia and Torrontes, have been studied by random amplified polymorphic DNA (RAPD) markers and sequence-tagged microsatellite site (STMS) analyses. Study of 18 accessions of Malvasia group has allowed the establishment of 13 different varieties, 11 of which are represented by only one accession each. It could also be determined that only the two accessions named Turruntes (both from Haro, La Rioja) showed identical RAPD profiles and STMS alleles, while the other four were different varieties. The discrimination power of the techniques is high: any of the 10 RAPD primers or two of the eight STMS loci are enough to distinguish all of the different varieties. A dendrogram obtained from the entire data shows clearly that Malvasia and Torrontes are artificial groups. Key words: Vitis vinifera, microsatellite, RAPD, STMS, polyclonal variety One of the major problems for varietal identification in grapevines (Vitis vinifera L.) is the existence of homonyms (different varieties with the same name) and synonyms (different names for the same variety), particularly with varieties that have been cultivated for centuries and are widely distributed. There are many grapevine varieties, but only a few are included as authorized or recommended by the Spanish Denominación de Origen (DO). Accurate identification of any grapevine plant has thus become an important issue, as legal implications are present. Different methods are available to perform such characterization in grapevines, but DNA molecular markers have become the marker of choice. Because they are not limited in number, their expression is environmentally independent and results are obtained faster than with other markers. Three molecular techniques are widely used for grapevine characterization: random amplified polymorphic DNA (RAPD) [10, 21], sequence-tagged microsatellite sites (STMS) [2, 19, 23], and amplified fragment length polymorphism (AFLP) [3, 20]. All reveal sufficient genetic variation among grape varieties, although STMS is preferred to construct a database and compare results within and between laboratories [22]. In general, a few STMS loci are enough to distinguish any grapevine variety [8,17,18,19]. Nevertheless, some grapevine varieties could have a polyclonal origin, normally from related plants. In these cases, particularly if close relatives (sibs) are integrated in the same variety, it can Instituto Madrileño de Investigación Agraria y Alimentaria (IMIA), Finca El Encín, Apdo. 127, Alcalá de Henares, Madrid, Spain. *Corresponding author [Fax: ; javier.ibanez@imia.madrid.org] Acknowledgments: The authors thank K.M. Sefc and H. Steinkellner for providing primer sequences of microsatellite loci before their publication and L. Hidalgo and F. Cabello for their useful suggestions. This work was developed under the project RF funded by INIA. MTA wishes to thank INIA for her Ph.D. grant. Manuscript submitted January 2001; revised October 2001 Copyright 2002 by the American Society for Enology and Viticulture. All rights reserved. be difficult to discover the polyclonal origin using a few microsatellite loci, as the probability of having the same genotypes is higher for sibs. It would be convenient to increase the number of loci studied or to combine this technique with others: Filippetti et al. used ampelography to distinguish between two sibs that showed identical alleles at the 10 microsatellite loci tested [5]. In this work we have used the RAPD technique, which easily distinguishes sib grapevines. A number of accessions included in 13 different Spanish DOs, initially belonging either to Malvasia or Torrontes groups, were studied. To our knowledge, there is no relationship between these two groups. Initially, Malvasia wines were produced only in Greece. Due to their reputation, other regions began to produce this type of wine, using either authentic Malvasia or other similar varieties. Count Odart (1863, cited by [12]) already described the existence of 16 white and four colored Malvasias. Torrontes are thought to be native to Spain, and several different varieties named Torrontes have long been cultivated in Spain, although they are similar in the quality of their fruits for vinification [24]. We have applied two different PCR-based techniques: RAPD and STMS to solve problems of synonymies and homonymies and to study the genetic relatedness existing among the different varieties. Materials and Methods Plant material. Grapevine young leaves were obtained from plants conserved at the Vitis Germplasm Bank (BGV), at Finca El Encin, Alcala de Henares, which contains more than 2400 accessions. Table 1 lists 31 accessions studied, their geographical origin, and, when appropriate, the DO where they are included. Initially, 18 accessions were considered as belonging to Malvasia group and six to Torrontes group [1]. The other seven accessions were included in Table 1 because of their identity with several accessions of the studied groups. Geographical 125

2 126 Borrego et al. Table 1 Name and geographic origin of accessions studied; the group (Malv: Malvasia; Torr: Torrontes) where each accession was initially included; abbreviation (Abbr) of the accessions; and the corresponding Denominación de Origen (DO), when appropriate. Group Accession name Geographic origin Abbr DO Alarije Cáceres Al Ribera del Guadiana Albillo Valladolid ARD Ribera de Duero Torr Aris Guadalajara Ar Blanco del País Burgos BP Ribera de Duero Doña Blanca Ourense DB Valdeorras Macabeo Figueras (Girona) Mo Ampurdán-Costa Brava Malv Malvasía Vitoria (Álava) MV Rioja Malv Malvasía Navarra MN Navarra Malv Malvasía Logroño MLo Rioja Malv Malvasía Tenerife MT Tenerife a Malv Malvasía Alcaraz (Salamanca) MS Malv Malvasía Morales de Toro (Zamora) MZ Toro Malv Malvasía León ML Bierzo Malv Malvasía Blanca Arrecife (Lanzarote, Gran Canaria) MA Lanzarote Malv Malvasía Blanca Lanzarote (Gran Canaria) MB Lanzarote Malv Malvasía Común Jerez (Cádiz) MC Malv Malvasía de Sitges Barcelona MSi Malv Malvasía Dorada Portugal MD Pardilla Mondejar (Guadalajara) P Mondejar Malv Rojal Logroño RLo Malv Rojal Albacete RA Malv Rojal Cuenca RC Malv Rojal Blanco Tarazona de la Mancha (Albacete) Rb Malv Royal Gordo Cadalso de los Vidrios (Madrid) RG Malv Subirat Parent Villafranca del Penedés (Barcelona) SP Penedés Torr Torrontés Ourense TO Ribeiro Torr Torrontés Barbante (Ourense) TB Ribeiro Torr Torrontés Córdoba TC Montilla-Moriles Torr Turruntés Haro (Logroño) TH Torr Turruntés Logroño TL Valenciana León V Bierzo a Five different DO in Tenerife Island include Malvasia among their authorized varieties. locations of the different accessions studied are shown in Figure 1. DNA extraction. Genomic DNA was extracted from young leaves according to [11] and diluted to 5 ng/µl. STMS analysis. Amplifications were performed in a reaction volume of 20 µl containing 10mM Tris-HCl ph 8.3; 50mM KCl; 200µM of each dntp; 1.5 mm MgCl 2 ; 0.2µM upper primer, with an attached fluorescent dye; 0.2µM lower primer; 0.5 unit Ampli Taq Gold (Applied Biosystems, Foster City, CA) and 50 ng of genomic DNA. The amplification program was 95 C for 5 min, 40 cycles of (95 C for 45 sec; 50 C for 30 sec; 72 C for 1 min 30 sec), and 72 C for 7 min. Eight different STMS were analysed: VVS2 (labeled with the fluorochrome HEX), VVS5 (6-FAM), VVS29 (HEX) [23], VVMD5 (6-FAM) and VVMD7 (TET) [2] in an ABI PRISM 377 DNA Sequencer (Applied Biosystems), and ssrvrzag 47 (6- FAM), ssrvrzag 62 (TET), and ssrvrzag 79 (HEX) [19] in an ABI PRISM 310 Genetic Analyzer (Applied Biosystems). Amplified, electrophoresed fragments were sized with the GeneScan software (Applied Biosystems). Two different plants per accession were analyzed, and the average of the two GeneScan analyses calculated. A genotypic data table was constructed from these average values. These data were translated to a 1/0 code considering twice the alleles from homozygotes, except in the case of VVS5, where null alleles were detected [6,17,22]. In cases where two different varieties share at least one allele per locus (compatibility as parent/progeny), nine additional STMS loci were studied: ssrvrzag 15 (HEX), ssrvrzag 21 (TET), ssrvrzag 25 (TET), ssrvrzag 29 (HEX), ssrvrzag 30 (6-FAM), ssrvrzag 64 (TET), ssrvrzag 67 (HEX), ssrvrzag 83 (6-FAM), ssrvrzag 112 (6-FAM) [19]. RAPD analysis. Only the 24 accessions initially considered as belonging to the Malvasia or the Torrontes group were analyzed (Table 1). Amplifications Figure 1 Map of Spain showing the geographical origin of the accessions studied. White lines delimit the different provinces. Denominación de Origen (DO) appear in light grey.

3 Genetic Study of Malvasia and Torrontes 127 were performed in a reaction volume of 25µL containing 10mM Tris-HCl ph 8.3; 50mM KCl; 200µM of each dntp; 5mM MgCl 2 ; 0.4µM primer; 2 units Ampli Taq DNA Polymerase, Stoffel Fragment (Applied Biosystems); and 25 ng of genomic DNA. The amplification program was 94 C for 6 min and 40 cycles of (94 C for 1 min; 35 C for 1 min; 35 C to 72 C for 1 min 30 sec; 72 C for 6 min 30 sec). Ten different oligonucleotide primers (10-mers) were used, purchased from Operon Technologies (Alameda, CA) (Table 2). Amplification products were separated on 2% agarose gels and images were captured with an image analyzer. PCR reactions were performed twice. Amplified fragments (bands) were recorded as present (1) or absent (0), and results were assembled in a binary data matrix. Data analysis. Polymorphic RAPD and STMS data were pooled in a single binary matrix. Dice similarity coefficients were calculated [4] and subjected to cluster analyses using unweighted pair-group method analysis (UPGMA) to construct phenograms. The option FIND of the program allowed us to obtain the possible alternative dendrograms. Cophenetic values were obtained from these dendrograms and compared with those of the similarity matrix to obtain cophenetic correlations. NTSYS-pc software version 2.0 [15] was used for these calculations. From microsatellite data, a search for genotype compatibility as parent/progeny was performed with a selfwritten BASIC program. Results and Discussion Microsatellite analysis. STMS genotypes were established for eight microsatellite loci (Table 3). When a given accession presented only one peak, it was considered as a homozygote, except in the case of VVS5 where the presence of null alleles has been described [6,17,22]. The results obtained for the two different plants of each accession were essentially identical: GeneScan values observed in each repetition for a given allele differed in less than one nucleotide. Analysis of genotypes obtained for 18 accessions of Malvasia group showed the existence of many homonyms and allowed establishing 13 different varieties. Of these, 11 varieties are represented by only one accession each. Other variety is formed by six accessions: Rojal from Logroño (RLo) is a synonym for Malvasia from Navarra (MN), Malvasia from Vitoria (MV), and Malvasia from Logroño (MLo). This is not surprising, as Logroño, Vitoria, and Navarra are geographically close (Figure 1). Subirat Parent (SP) from Barcelona also belongs to this variety, as does Aris from Guadalajara (Ar), which was initially thought to be of the Torrontes group. These six accessions are cultivated in the northern half of the Iberian Peninsula, geographically relatively close. The remaining two Malvasia accessions are Table 2 Primer sequences used for RAPD study. Code Sequence 5' to 3' OPA-02 TGCCGAGCTG OPA-07 GAAACGGGTG OPA-08 GTGACGTAGG OPA-11 CAATCGCCGT OPD-07 TTGGCACGGG OPP-02 TCGGCACGCA OPP-08 ACATCGCCCA OPP-09 GTGGTCCGCA OPP-18 GGCTTGGCCT OPP-19 GGGAAGGACA Table 3 Allelic sizes (in bp) obtained for the different accessions studied at eight microsatellite loci. Microsatellite loci Accession VVS2 VVS5 VVS29 VVMD5 VVMD7 ZAG 47 ZAG 62 ZAG 79 Al, Ar, MV, MN, MLo, RLo, SP ARD, BP, TH, TL DB, MZ, V Mo, MS, P MT, MSi ML MA MB MC MD RA RC Rb RG TB TC TO

4 128 Borrego et al. identical despite cultivation in distant places: Tenerife (MT) in the Canary Islands, and Barcelona (MSi) in northeast Spain. Interestingly, another variety cultivated in the Canary Islands, Malvasia Blanca from Lanzarote (MB), has shown the same genotypes as Malvasia Chianti, cultivated in Italy (data not shown). With respect to the remainder of the accessions of Torrontes group, only the two accessions named Turruntes (both from Logroño) (TL and TH) showed identical STMS genotypes, while Torrontes from Ourense (TO), Torrontes from Barbantes (TB), and Torrontes from Córdoba (TC) are homonyms. Interestingly, Barbantes and Ourense are in the same province, but their Torrontes accessions are different. The power of STMS analysis is high: only two microsatellite (vg. VVS5 and VVMD5) are needed to distinguish the different varieties. One advantage of this type of analysis is that it permits the comparison with results obtained previously or in other laboratories. Comparison of Malvasia and Torrontes genotypes with our database has revealed interesting results. Surprisingly, Malvasia de Salamanca (MS) shows the same STMS genotypes as Macabeo (Mo), a widely cultivated variety in Spain and included in many DOs. Mo is one of the varieties used to produce Spanish cava. Another synonym for both varieties is Pardilla from Mondejar (P; which is different from Pardilla from DO La Mancha, data not shown). The identity between Doña Blanca (DB) and Valenciana (V) was expected, as they are known synonymies, cultivated in northwest Spain, but it was not known that Malvasia from Zamora (MZ) is also a synonym for them, as has been demonstrated here by STMS analysis. Again, the provinces where these accessions are cultivated are close. Alarije (Al), a well-regarded variety cultivated in Caceres, has the same alleles for all the microsatellites as the major group of accessions found within Malvasia. Finally, Turruntes from Haro and from Logroño (TH and TL) are identical to Albillo from Valladolid (ARD) and Blanco del Pais from Burgos (BP). RAPD analysis. RAPD banding patterns were highly polymorphic among the 24 accessions tested (Figure 2). A total of 174 bands were considered, 136 of which were polymorphic (78%). The polymorphism found is higher than in previous studies in grape, between 50 and 73% [13,16,25,26]. The protocol for obtaining RAPDs involves a long polymerization time (8 min including the ramp). This was originally performed for overcoming the particular characteristics of the enzyme used, the Stoffel fragment, including its low processivity [9]. As a consequence of such protocol, the profiles obtained contain numerous bands and are very reproducible. RAPD analysis absolutely confirmed STMS results in both Malvasia and Torrontes groups. These results have been obtained not only with the whole of the RAPD primers but also with each primer. This is due to two factors: the previous selection of the most discriminant primers and the high polymorphism level existing in grapevines. Cluster analysis. Individual dendrograms obtained for STMS and for RAPD data show the 17 different genotypes found Figure 2 RAPD profiles generated with primer OP-P02 with the 24 accessions studied of the Malvasia and Torrontes groups. Abbreviations correspond to Table 1. Lane M: molecular weight marker: Ladder 100pb (Gibco). Lane B: blank reaction (without a template). among Malvasia and Torrontes groups (13 and 4, respectively) among the 24 accessions analyzed. Dice coefficients of similarity between different varieties range from 0.06 to 0.69 for microsatellite data and from 0.47 to 0.81 for RAPD data. Varieties belonging to each group appear mixed in the individual trees (data not shown): there is no grouping according to Malvasia and Torrontes characteristics. The groupings obtained with each technique are quite different, as a consequence of the poor correlation between the two similarity matrices (r=0.771, normalized Mantel statistic Z). The dendrogram obtained from the whole data shows a similar result: there is a high heterogeneity within these groups, accessions from both groups appearing mixed in the tree (Figure 3). This dendrogram reflects well the similarity matrix, as can be deduced from the high cophenetic correlation coefficient obtained: The highest similarity (0.75) was found among the varieties that include Malvasia from Logroño (Al/Ar/MLo/ MN/MV/RLo/SP) and Malvasia from Salamanca (Mo/MS/P). The analysis of their genotypes shows that these varieties could be closely related, as they share one allele per locus; that is, one could be the progeny of the other. Two other similar varieties are Torrontes from Barbante (TB) and Torrontes from Ourense (TO). Both are from the same province in Galice and share eight alleles in five loci. Malvasia from Tenerife (MT) and Malvasia from Arrecife (MA), cultivated in the Canary Islands, also share A MLo MN MV Rlo SP MS Rb TH TL RA RC MZ TC MC MD TB TO MB RG MA MSi MT ML Figure 3 Dendrogram obtained from microsatellite and RAPD data with the 24 accessions studied of the Malvasia and Torrontes groups. Abbreviations correspond to Table 1.

5 Genetic Study of Malvasia and Torrontes 129 one allele in each of the eight loci studied. Malvasia from Leon (ML) is the most dissimilar variety. The lowest similarity (0.38) has been found between ML and Malvasia Dorada (MD). The heterogeneity within these groups has been long known, particularly in the case of Malvasia (Count Odart, 1863, cited by [12]). The denomination Malvasia referred initially to the seaport in Greece from where the wine was exported. The name was later generalized to those varieties producing certain types of wines: liquored, perfumed, and sweet. These types of varieties extended through the Mediterranean countries, as can be deduced from its culture in southern France, Spain, and central and southern Italy. A remarkable point is that the accession from Portugal (MD) did not differ from the others, which supports this origin. Malvasia Dorada (MD) and Torrontes from Ourense (TO) even share one allele per locus in seven out of eight loci. Their relationship could be due to the cultural flow between Portugal and Galice. The origin of the denominations Rojal or Royal for some of the red Malvasia varieties may be related to berry color: the Spanish rojo translates to red in English. Of the Malvasia varieties cultivated in Spain, MSi/MT is most probably the original from Greece (L. Hidalgo, personal communication). The wine produced in Tenerife from this variety was exported to England with the name Canary during the sixteenth century. Madeira Island later produced a similar wine, using a variety called Malvasia fina from Madeira. The morphological comparison between the description of this variety found in [12] and MSi/MT shows that they are probably the same variety (data not shown). Torrontes group was also known to include different varieties. Viala and Vermorel described Turruntes from Rioja and indicated that it was different than other Torrontes cultivated in the east and south of Spain [24]. Polyclonal origin of the varieties. Polyclonal origin in grapevine varieties is not frequent, but it is suspected to occur [14]. Two issues regarding such origin should be considered. First, a limited number of STMS can be insufficient for discovering differences between very close relatives [5]. Second, accessions with the same name and different DNA profiles could be considered as belonging to a polyclonal variety, if they are relatives. Filippetti et al. [5] found that two out of 24 grapevine sibs, deriving from a single self-pollinated vine, share the same microsatellite alleles at the 10 tested loci, although they were morphologically distinguishable. In our work, RAPD markers have found no differences within the varieties defined by microsatellite analysis. As RAPD technique easily distinguishes full sibs in grapevine [6], it can be concluded that none of the groups established by STMS is a clustering of different genotypes. The second issue is whether any of the different groups established are close relatives and could be considered as belonging to the same variety. Microsatellite analysis allows the study of whether one plant can be the progeny of the other. For that, the two plants should share at least one allele per locus, being then compatible as parent/progeny. We have found two cases where this occurs: the group of accessions Al/Ar/MLo/MN/MV/ RLo/SP with the group formed by Mo, MS and P, and the group MSi/MT with MA. One accession of each of these groups has been further studied with additional microsatellite loci. Two other varieties, MD and TO, which belong to the Malvasia and the Torrontes group, respectively, share one allele per locus except in the locus VVS5. The genotype of MD at VVS5 is 92:- while the genotype of TO is 94:120. In this locus, the presence of null alleles has been described. The difference between the allele of MD and the short allele of TO is two nucleotides, simply the repetition unit of this microsatellite. The presence of mutations in the microsatellite sequence that allows distinguishing between two synonym varieties has also been described in grapevine [7]. For both reasons these two varieties were also studied with more microsatellite loci; results are shown in Table 4. The group of MLo and of Mo are incompatible at the loci ZAG 64 and ZAG 67. These varieties can then be excluded as a parent or progeny of the other. The same can be said of TO and MD, which are incompatible at ZAG 112, besides VVS5. On the contrary, MSi and MA are compatible at all the 17 loci studied. MSi is identical to MT, which is cultivated in Tenerife, while MA is cultivated in Lanzarote, and Tenerife and Lanzarote are two islands of the Canary archipelago. A speculative hypothesis could be drawn about the origin of the varieties MA and MT: one could be the progeny of the other, and their identical denomination (Malvasia) could correspond to a polyclonal variety. Supporting this hypothesis is the fact that MA and MT (plus MSi) constitute a well-defined group in the dendrograms obtained from RAPD and STMS data, individually (data not shown) and Table 4 Genotypes obtained for nine additional STMS loci to test the compatibility progenitor/progeny in three pairs of accessions. Alleles shared within each pair in comparison are highlighted in bold. STMS loci Accession ZAG 15 ZAG 21 ZAG 25 ZAG 29 ZAG 30 ZAG 64 ZAG 67 ZAG 83 ZAG 112 Malvasía (MLo) Macabeo (Mo) Malvasía de Sitges (MSi) Malvasía Blanca (MA) Malvasía Dorada (MD) Torrontés (TO)

6 130 Borrego et al. together (Figure 3). For the remainder of the varieties, data do not allow claiming for a polyclonal origin, although the possibility of the presence of sibs among them has not been investigated. Summary RAPD and STMS analyses have independently shown the same results: 13 out of 18 accessions of the Malvasia group and five out of six of the Torrontes group are different varieties. Cluster analysis of the different accessions show varieties belonging to both groups mixed in the dendrogram. The names Malvasia and Torrontes do not reflect any general genetic characteristic. Only two Malvasia accessions of the Canary Islands could be genetically related and then be considered as belonging to a polyclonal variety. Literature Cited 1. Borrego, J., J.F. Gallego, L. Serrano, J.L. Gómez, and I. Martínez. Descripciones ampelográficas nacionales.1st ed. 250 pp. Comunidad de Madrid, Madrid (1990). 2. Bowers, J.E., G.S. Dangl, R. Vignani, and C.P. Meredith. Isolation and characterization of new polymorphic simple sequence repeat loci in grape (Vitis vinifera L.). Genome 39: (1996). 3. Cervera, M.T., J.A. Cabezas, J.C. Sancha, F. Martinez de Toda, and J.M. Martinez Zapater. Application of AFLPs to the characterization of grapevine Vitis vinifera L. genetic resources. A case study with accessions from Rioja (Spain). Theor. Appl. Genet. 97(1-2):51-59 (1998). 4. Dice, L.R. Measures of the amount of ecologic association between species. Ecology 26: (1945). 5. Filippetti, I., O. Silvestroni, M.R. Thomas, and C. Intrieri. Diversity assessment of seedlings from self-pollinated Sangiovese grapevines by ampelography and microsatellite DNA analysis. Vitis 38:67-71 (1999). 6. Ibáñez, J. Estudio genético de variedades de uva de mesa sin semillas (Vitis vinifera L.) mediante marcadores moleculares y su aplicación a la protección legal. Thesis, Universidad Complutense (2000). 7. Ibáñez, J., M.T. de Andrés, and J. Borrego. Allelic variation observed at one microsatellite locus between the two synonym grape cultivars Black Currant and Mavri Corinthiaki. Vitis 39(4): (2000). 8. Lamboy, W.F., and C.G. Alpha. Using simple sequence repeats (SSRs) for DNA fingerprinting germplasm accessions of grape (Vitis L.) species. J. Am. Soc. Hort. Sci. 123: (1998). 9. Lawyer, F.C., S. Stoffel, R.K. Saiki, S.-Y. Chang, P.A. Landre, R.D. Abramson, and D.H. Gelfand. High-level expression, purification, and enzymatic characterization of full-length Thermus aquaticus DNA polymerase and a truncated form deficient in 5' to 3' exonuclease activity. PCR Meth. Appl. 2: (1993). 10. Lodhi, M.A., N.F. Weeden, and B.I. Reisch. Characterization of RAPD markers in Vitis. Vitis 36: (1997). 11. Lodhi, M.A., G.N. Ye, N.F. Weeden, and B.I. Reisch. A simple and efficient method for DNA extraction from grapevine cultivars and Vitis species. Plant Mol. Biol. Rep.12: 6-13 (1994). 12. Marés, H. Description des Cépages Principaux de la Région Méditerranéenne de la France. 120 pp. Coulet, Montpellier and Masson, Paris (1890). 13. Qu, X., J. Lu, and O. Lamikanra. Genetic diversity in muscadine and American bunch grapes based on randomly amplified polymorphic DNA (RAPD) analysis. J. Am. Soc. Hort. Sci. 121(6): (1996). 14. Rives, M. Bases génétiques de la sélection clonale chez la vigne. Ann. Amélior. Plant. 11: (1961). 15. Rohlf, F.J. NTSYS-pc. Numerical taxonomy and multivariate analysis SYStem. Exeter Software, Setauket, New York (1990). 16. Sanchez-Escribano, E.M. Identificación de variedades de uva de mesa (Vitis vinifera L.) mediante marcadores genéticos (isoenzimas, microsatélites, RAPDs y AFLPs). Thesis, Universidad de Murcia (1998). 17. Sanchez-Escribano, E.M., J.P. Martin, J. Carreño, and J.L. Cenis. Use of sequence-tagged microsatellite site markers for characterizing table grape cultivars. Genome 42:87-93 (1999). 18. Sefc, K.M., M.S. Lopes, F. Lefort, R. Botta, K.A. Roubelakis-Angelakis, J. Ibáñez, I. Pejic, H.W. Wagner, J. Glössl, and H. Steinkellner. Microsatellite variability in grapevine cultivars from different European regions and evaluation of assignment testing to assess the geographic origin of cultivars. Theor. Appl. Genet. 100: (2000). 19. Sefc, K.M., F. Regner, E. Turetschek, J. Glössl, and H. Steinkellner. Identification of microsatellite sequences in Vitis riparia and their applicability for genotyping of different Vitis species. Genome 42: (1999). 20. Sensi, E., R. Vignani, W. Rohde, and S. Biricolti. Characterization of genetic biodiversity with Vitis vinifera L. Sangiovese and Colorino genotypes by AFLP and ISTR DNA marker technology. Vitis 35: (1996). 21. Tessier, C., J. David, P. This, J.M. Boursiquot, and A. Charrier. Optimization of the choice of molecular markers for varietal identification in Vitis vinifera L. Theor. Appl. Genet. 98: (1999). 22. Thomas, M.R., P. Cain, and N.S. Scott. DNA typing of grapevines: A universal methodology and database for describing cultivars and evaluating genetic relatedness. Plant Mol. Biol. 25: (1994). 23. Thomas, M.R., and N.S. Scott. Microsatellite repeats in grapevine reveal DNA polymorphisms when analysed as sequence-tagged sites (STSs). Theor. Appl. Genet. 86: (1993). 24. Viala, P., and V. Vermorel. Traité Général de Viticulture Ampelographie. Vols. 4, 6. Masson et Cie, Paris ( ). 25. Vidal, J.R., S. Moreno, A. Masa, and J.M. Ortiz. Study of the genetic homogeneity of Albarino (Vitis vinifera L.) growing in Galicia (Spain) using isozyme and RAPD markers. Vitis 37: (1998). 26. Ye, G.N., M. Hemmat, M.A. Lodhi, N.F. Weeden, and B.I. Reisch. Long primers for RAPD mapping and finger-printing of grape and pear. BioTechniques 20(3): (1996).