Assessing molecular and morpho-agronomical diversity and identification of ISSR markers associated with fruit traits in quince (Cydonia oblonga)

• Published in: Genetics and molecular research 2011-11, Vol.10 (4), p.2729-2746
• Main Authors: Ganopoulos, I, Merkouropoulos, G, Pantazis, S, Tsipouridis, C, Tsaftaris, A
• Format: Article
• Language: English
• Subjects: Fruit - genetics; Genetic Markers; Greece; Phylogeny; Polymorphism, Genetic - physiology; Quantitative Trait Loci; Rosaceae - genetics
• ISSN: 1676-5680; 1676-5680
• DOI: 10.4238/2011.November.4.7

Quince is a deciduous tree known to the countries around the Mediterranean since antiquity. Nowadays, quince is used as an ornamental plant, and as a rootstock for pear trees, with its fruit being appreciated mainly for production of jam and sweets rather than for raw consumption. Quince leaves contain compounds with antioxidant, antimicrobial and anticancerous properties that have been the focus of recent research on pharmaceutical and medical uses as well as for food preservatives. An orchard has been established in Greece, composed of quince varieties (Cydonia oblonga, N = 49) collected from different sites of the country (mainly from home gardens), constituting a unique quince gene bank collection for southeast Europe. We made a phenotypic analysis using 26 morphological plus seven agronomical descriptors coupled with molecular techniques in order to examine the genetic diversity within the collection. Principal component analysis using the 33 descriptors identified 10 components explaining the existence of more than 70% of the total variation. Subsequent cluster analysis classified most of the previously identified productive varieties of the quince orchard in the same clade of a dendrogram. Molecular analysis generated by 13 inter-simple sequence repeat primers amplified 139 bands, including 109 polymorphic bands, indicating a level of polymorphism of 79%; mean gene diversity was calculated to be 0.309. Using stepwise multiple regression analysis, a number of markers significantly associated with fire blight susceptibility, yield, mean fruit weight, citric acid content, soluble solid content, and fruit drop were identified. Hence, data extracted by multiple regression analysis could be useful in marker-assisted breeding programs, especially when no previous genetic information is available.