Genetic variation in cucumber (Cucumis sativus L.) as assessed by random amplified polymorphic DNA

Isozyme and restriction fragment length polymorphisms (RFLPs) have been app lied to studies of genetic relationships and germplasm management in cucumb er (Cucumis sativus L.). However, isozymes identify relatively few polymorp hisms, and RFLPs are technically complex, expensive, and not compatible for the high through-put required for rigorous assessment of this narrow-based germplasm. Since random amplified polymorphic DNA (RAPD) markers do not ma nifest such shortcomings, a study was conducted in cucumber to examine gene tic relationships in diverse germplasm, assess the usefulness of RAPD marke rs in distinguishing elite accessions, and compare the relative effectivene ss of RAPD markers to that of isozyme and RFLP markers. One hundred and eig hteen C. sativus accessions were analyzed using variation at 71 RAPD loci ( 44 mapped and 27 unmapped). Genetic distances among accessions were estimat ed using the simple matching coefficient complement, and analyzed using mul ti-dimensional scaling. Each accession had a unique marker profile, indicat ing that RAPD analysis was useful in genotypic differentiation. Germplasm g rouping patterns were consistent with individual accession origins, theoret ical dispersal routes and discriminating morphological characters (i.e., se x expression and fruit length to diameter ratio). Although elite accessions were discriminated by RAPD profiling, their genetic distances were relativ ely small (between 0.01 and 0.58), indicating limited genetic diversity in this germplasm array. Assessment of a subset of the germplasm array using R APDs resulted in genetic distance measurements more similar to published ge netic distance estimates by RFLP markers (Spearman rank correlation, r(s) = 0.7-0.8) than estimates by isozyme markers (r(s) = 0.4). Data indicate tha t RAPD markers have utility for analysis of genetic diversity and germplasm management in cucumber.