Morphological Variation and Relationships of Azerbaijan Cultivated and Wild Grape Populations

Transcription

1 POSTER PRESENTATION Morphological Variation and Relationships of Azerbaijan Cultivated and Wild Grape Populations Samira Salayeva 1, Javid Ojaghi 1, Roham Eshghi 1, and Zeynal Akparov 2 1 Science worker of Azerbaijan Genetic Resources Institute of ANAS, Baku, Azerbaijan; 2 Prof. and director of Azerbaijan Genetic Resources Institute of ANAS, Baku, Azerbaijan. salayeva_genetic@mail.ru Abstract: The aim of this study was the assessment of the genetic diversity of 60 accessions, comprising 30 local table grape varieties (Vitis vinifera ssp. sativa) and 30 wild types (Vitis vinifera ssp. sylvestris) originating from regions near the Caspian Sea of Azerbaijan Republic by the multivariate statistics analysis of morphological traits. The phenotypic coefficients of variation and variance analysis were highly significant, indicating high genetic diversity among investigated grape accessions. The results of path analysis showed that the traits, such as the number of bunches per vine, the bunch length in cultivated grape population and the leave width, the bud yield in wild grape population have the greatest effect on yield. By principle component analysis the cultivars Haji Abbas, Gavangir, Sarigile gejyetishan, Ala shani and the wild grape samples 32, 34, 38, 43 and 44 were selected as high productive accessions. The cluster analysis leading according the Ward method on the basis Euclidean distance classified cultivars into three groups and wild grape accessions into 4 groups, while the genotypes with high yield were placed in a separated group on the both of the dendrograms. Cluster analysis of all genotypes showed that the observed genetic diversity among cultivated and wild populations is not geographically structured. Key words: Vitis vinifera ssp. sativa and sylvestris, morphological traits, multivariate statistics, genetic diversity Introduction: Grapevine (Vitis vinifera L.) is one of the oldest and most important perennial crops in the world. More than 7000 grapevine cultivars are believed to exist around the world (Riccardi et al. 2006). In the year 2009, the global grapes production was estimated over 66.9 million tones harvested from million hectares (FAO 2010). Azerbaijan is one of the richest regions of Globe for the wealth of grapevines. The existence of a large assortment of local grape varieties, as well as the occurrence of aborigine varieties well adapted to local conditions and wild plant material, which is represented by typical fluffy-leafed (typica Negr.) and naked-leafed (aberrans Negr.) forms, points out Azerbaijan as one of the centre of origin of grapevine (Vitis vinifera L.) (Negrul 1973). Historical information and recent archaeological excavation confirmed this postulate (Negrul 1973). Occurrence of wild vine in Azerbaijan disappears into very far times of ancient history. The imprint of the leaf of wild vine Vitis silvestris Gmel., which was found by archaeologists on the stone on top pliosam of the Arax river of Zangelan area (near Minjivan railway station, located on border with Iran), testifies to it (Amanov 1998). The wild vine Vitis silvestris Gmel has spread on the vast territory of Azerbaijan Republic from 18 meters below sea-level (Kura riverside in Salyan area) to 2000 meters above sea-level (Gusar area) (Amanov 1997). Besides existence of resistant, morphologically different wild grape forms, this region also known with high-quality transportable table, rise and wine grape varieties (Nagiyev 1974). The evolution process of local grape grades in the country is very complicated and interesting. Local grape assortment originates from national breeding programs issued from the hybridization of wild forms with domesticated cultivars and from long-term selection. As a result of natural and artificial selection was created the rich fund of local grape grades with various economic values (Amanov 1998). The large grape diversity represented by landraces, improved varieties, hybrids and wild relatives has to be preserved as a potential source of valuable genetic variability for different breeding purposes. To preserve and use efficiently in the future the current genetic pool, it is necessary to evaluate the extent of this diversity by identification and distinction of Azerbaijan 1055 I n t e r n a t i o n a l C a u c a s i a n F o r e s t r y S y m p o s i u m

2 Morphological Variation and Relationships of Azerbaijan Cultivated and Wild Grape Populations grape accessions, as well as the determination of genetic relationship between local cultivars and wild relatives (Negrul 1973). Modern objectives in plant breeding may be achieved by evaluation of traits among genetic resources and combining those of interest in one cultivar. New methods for cultivar fingerprinting using molecular markers such as isozymes (Jahnkea et al. 2009; Cevera et al. 2001), RAPD (Salayeva et al. 2010; Maia et al. 2009), microsatellite (Santana et al. 2008) and AFLP (Stajner et al. 2009), have proved useful in distinguishing varieties, however, these methods are expensive and require wellequipped laboratories. On the other hand, morphological characterization can be a first and basic step in the description and classification of germplasm. Materials and Methods: A total of 60 accessions, comprising 30 local table grape varieties (Vitis vinifera ssp. sativa) and 30 wild types (Vitis vinifera ssp. sylvestris) belonged to Near-Caspian zone of Azerbaijan Republic, were analyzed. All plant materials were sampled from the grape collection of the Genetic Resources Institute of Azerbaijan National Academy of Sciences and from the Azerbaijan Scientific Research Institute of Viticulture and Wine-making. Local and wild accessions used in this study and their origins are listed in Table 1. Table1. Names and origins of grape accessions selected in the study Cultivated samples Wild samples Names Origin Origin 1 Davachi Halimasi Absheron 31 Nabran 2 Fikrati Absheron 32 Nabran 3 Malahati Absheron 33 Nabran 4 Farashi Absheron 34 Nabran 5 Sharabi Absheron 35 Davachi 6 Nardaran Absheron 36 Davachi Darbandisi 7 Ag uzum Absheron 37 Davachi 8 Sarigila tezyetishan Absheron 38 Davachi 9 Ag gulabi Absheron 39 Davachi 10 Danaguzu Absheron 40 Davachi 11 Giziluzum Absheron 41 Davachi 12 Haji Abbas Absheron 42 Guba 13 Sikhgara Absheron 43 Davachi 14 Absheron kishmishi Absheron 44 Davachi 15 Uzunsov gilali ag shani Absheron 45 Guba 16 Gavangir Absheron 46 Guba 17 Iri gilali ag shani Absheron 47 Davachi 18 Ag shani Absheron 48 Guba 19 Absheron Absheron 49 Davachi galinbarmagi 20 Fatmayi Absheron 50 Guba 21 Sarigila Absheron 51 Davachi ortayetishan 22 Sarigila gejyetishan Absheron 52 Davachi 23 Absheron Absheron 53 Davachi kechimamasi 24 Shireyi Absheron 54 Davachi 25 Gara shani Absheron 55 Davachi 26 Shabrani Absheron 56 Guba 27 Ala shani Absheron 57 Davachi 28 Ag shani (clone) Absheron 58 Guba 29 Absheron khatinisi Absheron 59 Guba 30 Iri salkhim Absheron 60 Guba Vines were planted in 3 2 meters (row vine), trained to a bilateral cordon trellis system and spur pruned. The experiment was arranged in a completely randomized block design with three two-vine replications. For each accession, 23 morphological traits (length of annual bud, length and width leave, length of stalk leave, length of central vessel, number of bunches per bush, length and width bunch, length of stalk bunch, number of berry per bunch, length and width berry, weight of bunch, number of seeds, length and width seed, 100 weight of berries, 100 weight of seeds, number of fertile buds, number of green buds (this trait has not measured in wild grape) yield of bud, yield of per bush and time of vegetation) were measured according to the international descriptors for grape (I.B.P.G.R 1983). Data were analyzed using MSTATC (1991) and SPSS v.12.0 (1991) software packages. Results and Discussion: Genetic diversity of cultivated grapevine: Significant differences (p < 0.01) were detected among the aborigine grapevines for all characters by analyses of variance (data not shown). On average, the range of traits were: bunch weight, 96.7 to g; bud yield, 99.6 to g; number of berries per bunch, 47.9 to 150.6; time of vegetation, 130 to 182 days; 100 weight of berries, to g; number of bunches per bush, 16.6 to 42.3; bush yield, 1.4 to 8.8 kg; berry width, 10 to 21.6 mm and bunch length, 9.54 to 26.8 mm; showing substantial genetic variability among 1056 I n t e r n a t i o n a l C a u c a s i a n F o r e s t r y S y m p o s i u m

3 Samira Salayeva, Javid Ojaghi, Roham Eshghi, and Zeynal Akparov accessions that would be valuable for breeding programs. It is well known that yield is a polygenic trait and that it is difficult to increase directly, although benefits can be acquired due to those traits strongly correlated with the yield and that indirectly increase the yield. Simple correlation coefficients (Table 2) indicated that yield per vine correlated positively with number of fertile buds (r=0.883), bunch weight (r=0.852), bud yield (r=0.928), number of green buds (r=0.533), number of bunches per vine (r= 0.96), bunch length (r=0.513), bunch width (r=0.369), number of berries per bunch (r=0.796), berry width (r=0.36) and time of vegetation (r=0.366). But no correlations were observed between the yield and the other traits. Correlation values can indicate the degree of genetic relationship between two or more properties. Three significant factors may influence the correlation of agronomical traits, namely linkage of genes controlling those properties, peliotropic effects of genes for two or more traits, and environmental effects (Steel and Torrie 1982). It should not however be forgotten that correlation coefficient has only a mathematical interpretation and does not always imply cause and effect relationships. Therefore, in most cases, path analysis is employed in order to identify and understand the cause and effect relationships among different traits. Table 2. Correlation between yield and other traits in the cultivated grapes Traits Yield per vine Number of fertile buds Bunch weight 100 weight of seeds 100 weight of berries Bud yield Length of annual bud Number of green buds Stalk bunch length Number of bunches per vine Leave length Leave width Stalk leave length Central vessel length Bunch length Bunch width Number of berries per bunch Berry length Berry width Number of seeds Yield per vine ** 0.852** -0.17n.s 0.288n.s 0.928** 0.07n.s 0.533** -0.11n.s 0.96** -0.09n.s 0.01n.s -0.14n.s -0.02n.s 0.513** 0.369* 0.796** 0.09n.s 0.36* 0.12n.s Seed length Seed width Time of vegetation 0.125n.s -0.01n.s 0.366* **, *-significant at P=0.01 and P=0.05, respectively On the basis of path analysis it was found that number of bunches per vine had the most direct effects on yield per vine (0.689). It means that selection on the basis of this trait in populations under study lead us to achieve high yield grapevine. Also, number of bunches per vine in spite of big direct effects showed high indirect effects. Direct effects of bud yield and bunch length on yield per vine were 0.3 and 0.131, respectively. In addition, the 23 morphological traits were analyzed using principle component analysis. Six principle components accounted for 72.64% of the total variance. In the first component, traits such as, number of bunch per vine, yield per vine, number of fertile buds, weight of bunches and bud yield were of greater importance. This component alone accounted for 30% of the total variance; therefore, these traits are more important in assessing the quantitative traits than the others. Gavangir, Sarigile gejyetishen, Haji Abbas, Ala shani and Absheron galinbarmagi were selected by the first component. Cluster analysis, based on all morphological characters, grouped the local grapevine accessions into three clusters with respective distances at a similarity level of 8 (Figure 1). Fatahi et al. (2004) evaluated genetic diversity of 90 Vitis vinifera cultivars for 23 characters. The results of phenotypic correlation, multiple regression and path analysis showed that bunch number, bunch weight, berry number and berry weight had the greatest effect on yield. Principle component analysis showed that seven factors with eigenvalues >1 contributed 81% of total variation. Cluster analysis was also run on the basis of full characters using the Ward method and cultivars 1057 I n t e r n a t i o n a l C a u c a s i a n F o r e s t r y S y m p o s i u m

4 Morphological Variation and Relationships of Azerbaijan Cultivated and Wild Grape Populations grouped in four groups. Al.Saif et al. (2001) studied some grape cultivars by phonological and morphological traits in Riyadh region. They found that significant differences were noticed among grape cultivars in percentages of viable pollen grains, length, diameter and shape of pollen. Significant differences were found among grape cultivars in average number of clusters per vine. Halwany cultivar gave the higher average number of cluster per vine as compared with other cultivars. Significant differences were found in length and width of fruit clusters among grape cultivars under study. Figure1. Dendrogram showing the genetic relationships among 30 cultivated grape accessions based on morphological traits. Genetic diversity of wild type: The statistics analyses of traits are showed in table 3. The coefficient of variance showed high genetic variability in the most of traits. The traits such as length of annuals bud, yield per vine, bunch width, number of bunch in vine, number of berry in bunch, berry length and leave width had high genetic diversity between studied traits. But seed length and time of vegetation had low genetic variability. Correlation analysis was used to determine the linear relationship between morphologic traits (Table 4). There is a significant correlation between bud yield (0.957**), bunch weight (0.951**), 100 weight of berry (0.94**), number of fertile bud (0.879**), berry width (0.421*), berry length (0.421*), leave width (0.378*) and bunch width (0.377*) with yield per vine. The correlation was negative between yield per vine and number of berry in bunch. But no correlation was observed between the yield and the other traits. On the basis of path analysis it was found that leave width (0.769) and bud yield (0.487) had the most direct effects on yield per vine. It means that the selection on the basis of these traits lead us for has high yield with compare of the other morphological traits. Traits such as number of fertile buds and 100 weight of berry showed middle indirect effect on yield. But leave length had 1058 I n t e r n a t i o n a l C a u c a s i a n F o r e s t r y S y m p o s i u m

5 Samira Salayeva, Javid Ojaghi, Roham Eshghi, and Zeynal Akparov high negative indirect effect on yield (-0.865). Whereas by simple correlation appeared positive correlation between yield and leave length. There were not studied big direct effects by other traits on yield. By the path analysis were found that the leave width, bud yield, number of fertile buds and 100 weight of berry are main components of yield. According to principle component analysis, 6 components could justify 80.67% of the total variance. In the first component 100 weight of berry, bud yield, bunch weight and number of bunch per vine were of greater importance. This component alone accounted for 31.39% of the total variance; therefore, these traits are more important in assessing the quantitative traits than the others. The most important traits in the second component were the bunch width, berry length and berry width, and length of annual bud, seed length and seed width was the most important trait in the third component. This component could justify 10.69% of the variance. Also, bunch length, number of seed, length of central vessel and length of stalk bunch in fourth component, length and width of leave in fifth component were important. Finally, in sixth component length of stalk leave was important trait. The existing diversity in wild samples is very significant as the domestication pressure on cultivated samples has caused considerable losses of many useful genes, such as those responsible for resistance to different types of diseases and environmental stresses (drought, salinity, cold, heat, etc). Therefore, wild samples are very valuable reservoirs of useful genes which can resolve many of our present and future problems. Table3. Estimation of maximum, minimum, mean, standard error, standard deviation and coefficient of variance for 22 morphological traits in wild grape samples. Traits Min. Max. Mean S.E. S.D. C.V. Yield of per vine Number of fertile buds Weight of bunch weight of seeds weight of berries Bud yield Length of annual bud Length of stalk bunch Number of bunches per vine Leave length Leave width Length of stalk leave Length of central vessel Bunch length Bunch width Number of berries per bunch Berry length Berry width Number of seeds Seed length Seed width Time of vegetation I n t e r n a t i o n a l C a u c a s i a n F o r e s t r y S y m p o s i u m

6 Morphological Variation and Relationships of Azerbaijan Cultivated and Wild Grape Populations Table 4. Correlation between yield and other traits in wild grape accessions Traits Yield of per vine Number of fertile buds Weight of bunch 100 weight of seeds 100 weight of berries Bud Yield Length of annual bud Length of stalk bunch Number of bunches per bush Leave length Leave width Length of stalk leave Length of central vessel Bunch length Width of bunch Number of berries per bunch Berry length Berry width Number of seeds Seed length Seed width Time of vegetation Yield ** 0.95 ** ** 0.96 ** ** 0.4 * 0.4 * * -0.4 * 0.41 * 0.4 * **, * significant at P=0.01 and P=0.05, respectively Cluster analysis according to the morphological traits using the Ward method and the basis of similarity index classified all the wild type grapevine accessions into four groups (Figure 2). According to the results obtained from the dendrogram, wild samples No. 53, 54, 55 and 58 from Davachi, samples No. 57, 59, 60 and 61 from Quba region which grouped in first cluster had lowest yield. In the second cluster 5 genotypes from Davachi region, also, one genotype from Nabran region were resided. The third group consisted of 11 genotypes, which made up 36.6% of all examined genotypes. In this cluster, most of the accessions were from Davachi region. At last, the fourth cluster includes five accessions (samples No. 34 and 32 (from Nabran region), 43 (from Quba region), 38 and 44 from Davachi region), with some characters related to yield such as number of fertile buds, bunch weight, yield per vine, 100 weight of berry, bud yield, number of bunches per vine, length and width bunch had were high in comparison with the other clusters. Figure2. Dendrogram showing the genetic relationships among 30 wild grape accessions based on morphological traits 1060 I n t e r n a t i o n a l C a u c a s i a n F o r e s t r y S y m p o s i u m

7 Samira Salayeva, Javid Ojaghi, Roham Eshghi, and Zeynal Akparov Cluster analysis for all samples (local and wild accessions) was performed according to Euclidean distance following the Ward method. It is allowed to classify all the genotypes into three main groups (Figure 3). Although on the basis of this distance, cultivated and wild individuals clustered in separate groups (except Nardaran darbandisi), the results of this dendrogram showed that the observed genetic diversity among these populations is not geographically structured. However, this information is very useful for further management of Azerbaijan grape germplasm. Figure3. Dendrogram showing the genetic relationships among 60 cultivated and wild grape accessions based on morphological traits Mallikarjuna et al. (2003) were analyzed 222 cultivated and 22 wild grapes (V. vinifera ssp. Sylvestris) for genetic diversity by morphological traits and eight microsatellite markers. French cultivars appeared to be distinct and showed close affinity to the wild progenitor, ssp. sylvestris from south-western France and Tunisia, probably reflecting the origin and domestication history of many of the old wine cultivars from France. Irina and Zora (2003) were studied genetic diversity of 13 wild grapes in Slovenia by morphological characters. Kartash and Sabit (2008) evaluated 45 wild grape accessions in Turkey by morphological traits. Aros et al. (2005), Melek et al. (2005), Russo et al. (2004) studied genetic diversity of wild species and cultivated varieties of grapevine using morphological traits and RAPD markers. References Al-Saif, A.M. M.A. Aly and M.A. Bacha Phenological, Morphological and Yield Characteristics Studies on Some Grape Cultivars Grown in Riyadh Region. Journal of king Saud University. Agricultural Sciences, 13: Amanov, M.V., Local and Introduced Grape Grades of Azerbaijan Republic, Azerbaijan Ministry of Agriculture, Baku I n t e r n a t i o n a l C a u c a s i a n F o r e s t r y S y m p o s i u m

8 Morphological Variation and Relationships of Azerbaijan Cultivated and Wild Grape Populations Amanov, M.V A new variety of wild vine in Azerbaijan vine and vine of Russia. Aros, D., C. Meneses and R. Infante Genetic diversity of wild species and cultivated varieties of grapevine estimated through morphological descriptors and RAPD markers. Hereditas, 129: Cevera, M.T., I. Rodriguez, J.A. Cabezas, J. Chavez, J.M. Martinez-Zapater and F. Cabello Morphological and molecular characterization of grapevine accession known as Albillo. Amer. J. Enol. Vitic. 52: FAO (Food and Agriculture Organization of the United Nations) FAOSTAT. Fatahi, R., A. Ebadi, A. Vezvaei, Z. Zamani and M.R. Ghanadha Relationship among quantitative and qualitative characters in 90 grapevine (Vitis vinifera L.) cultivars. Acta Hort., 640: I.B.P.G.R: descriptors for grapes Intern. Broad Plant Genetic Resources, Rome. Irina, T and K. Zora Validity of phyllometric parameters used to differentiate wild grape. Genetic Resources and Crop Evoluation, 50: Jahnkea, G., J. Májera, A. Lakatosa, J. Györffyné Molnára, E. Deákb, É., Stefanovits-Bányaib and P. Varga Isoenzyme and microsatellite analysis of Vitis vinifera L. varieties from the Hungarian grape germplasm. 120: Kartash, H and Y. Sabit Genetic diversity among Turkish wild grape accessions using morphological traits. Hereditas, 145: Maia, S.H.Z., C.A. Mangolin, S.A.O. Collet and M.F.P.S., Machado Genetic diversity in somatic mutants of grape (Vitis vinifera) cultivar Italia based on random amplified polymorphic DNA. Genet. Mol. Res., 8: Mallikarjuna K, S. Gerald and H. Bernard Genetic structure and differentiation in cultivated grape Vitis vinifera L. Genet. Res. Camb., 81: Melek, N and H. Celik The ampelographic characters of some Isobella grape (Vitis labrusca L.) types grown in Sinop. Six Viticulture Symposium September, Tekirdag. MSTATC User s guide. East Lansing, Michigan, USA: Michigan State University. Nagiyev, M.P Short course about Agrometeorology. Ganga. Negrul, A.M Ampelography of Azerbaijan USSR. Azerbaijan State Publishing, Baku. Riccardi, P., G. Lauria, R. Grillo, Mc. Fiore, R. Cifarelli and F. Sunseri Assessing genetic diversity of Nero d Avola grapevine cultivar by using SSR markers. Proceedings of the 50th Italian Society of Agricultural Genetics Annual Congress, Ischia, Italy-10/14 September. Russo, G., L.D. Andra, V. Liuzzi and G. Alviti Study genetic diversity of some wild grape using morphological traits and RAPD markers. ISHS 26th Intl. Hortic. Cong., 640: Salayeva, S., E. Akhundova and A. Mammadov Evaluation of DNA Polymorphism among Cultivated and Wild Grapevine Accessions from Azerbaijan. Czech J. Genet. Plant Breed., 46: Santana, J.C., E. Hidalgo, A.I. de Lucas, P. Recio, J.M. Ortiz, J.P. Martı n, J. Yuste, C. Arranz and J.A. Rubio Identification and relationships of accessions grown in the grapevine (Vitis vinifera L.) Germplasm Bank of Castilla y Le on (Spain) and the varieties 1062 I n t e r n a t i o n a l C a u c a s i a n F o r e s t r y S y m p o s i u m

9 Samira Salayeva, Javid Ojaghi, Roham Eshghi, and Zeynal Akparov authorized in the VQPRD areas of the region by SSR-marker analysis. Genet Resour Crop Evol, 55: SPSS, SPSS Base 12.0 for Windows User s Guide SPSS Inc., Chicago, IL, Stajner, N., J. Jakse, B. Javornik, R.W. Masuelli and L.E. Martínez Highly variable AFLP and S-SAP markers for the identification of Malbec and Syrah clones. Vitis, 48: Steel, R.G.D. and J.H. Torrie Principles and Procedures of Statistics: A Biometrical Approach. 2nd edition. McGraw-Hill. Tokyo I n t e r n a t i o n a l C a u c a s i a n F o r e s t r y S y m p o s i u m