Prospects for celeriac (Apium graveolensvar.rapaceum) improvement by using genetic resources ofApium, as determined by AFLP markers and morphological characterization

Abstract

Genetic relationships among elite celeriac varieties and celeriac accessions conserved in genebanks are generally unknown. The objective of this study was to use amplified fragment length polymorphism (AFLP) markers and morphological characterization to identify material that could be of use in celeriac breeding. Genetic relationships were estimated in 34 elite celeriac varieties bred in Europe and 28 celeriac accessions conserved at the German genebank. Two varieties of celery, two varieties of leaf celery and three genebank accessions of wildApiumspecies were additionally analysed. FifteenEcoRI/MseI-based AFLP primer combinations were used. Polymorphic AFLP fragments were scored for calculation of Jaccard's coefficient of genetic similarity (GS). Morphological distances (MD) were determined based on 11 morphological traits. Average GS estimate in elite germplasm (GS=0.90) was higher than in exotic germplasm (GS=0.80). An AMOVA (analysis of molecular variance) revealed that a high proportion of variation was due to variation within elite celeriac varieties and genebank accessions. Although GS and MD matrices were poorly correlated (r=0.22), UPGMA (unweighted pair group method using arithmetic averages) cluster analyses revealed clear genetic groupings of celeriac germplasm, which was supported by morphological traits. Elite, moderately bred and exotic varieties formed distinct clusters, indicating that only a part of the available genetic diversity in celeriac germplasm has been exploited in breeding. DistinctApiumspecies might be useful for the introgression of new genes into cultivated celeriac material. Broadening of celeriac collections in genebanks and detection of new genetic resources are vital for improvements in celeriac breeding.