AFLP analysis of genetic diversity within and between Arabidopsis thaliana ecotypes

Abstract
The degree of genetic diversity within and between 21 Arabidopsis thaliana (L.) Heynh ecotypes was estimated by AFLP analysis. Within seven of the 21 ecotypes, a low but significant level of polymorphism was detected, and for five of these ecotypes two or three distinct subgroups could be distinguished. As these ecotypes represent natural populations, this intra-ecotypic diversity reflects natural genetic variation and diversification within the ecotypes. The source of this diversity remains unclear but is intriguing in view of the predominantly self-fertilizing nature of Arabidopsis. Interrelationships between the different ecotypes were estimated after AFLP fingerprinting using two enzyme combinations (EcoRI/MseI and SacI/MseI) and a number of selective primer pairs. SacI recognition sites are less evenly distributed in the genome than EcoRI sites, and occur more frequently in coding sequences. In most cases, AFLP data from only one enzyme combination are used for genetic diversity analysis. Our results show that the use of two enzyme combinations can result in significantly different classifications of the ecotypes both in cluster and ordination analysis. This difference most probably reflects differences in the genomic distribution of the AFLP fragments generated, depending on the enzymes and selective primers used. For closely related varieties, as in the case of Arabidopsis ecotypes, this can preclude reliable classification.