Population structure and mating system of Ascochyta rabiei in Tunisia: evidence for the recent introduction of mating type 2

The population structure of Ascochyta rabiei (teleomorph: Didymella rabiei ) in Tunisia was estimated among five populations sampled from the main chickpea growing regions using simple sequence repeat markers (SSR) and a mating type ( MAT ) marker. Mating type 2 isolates ( MAT1-2 ) had reduced genetic and genotypic diversity relative to mating type 1 isolates ( MAT1-1 ). This result, coupled with previous observations of lower overall frequency and restricted geographical distribution of MAT1-2 in Tunisia, and recent (2001) observation of the sexual stage, support the hypothesis of a recent introduction of MAT1-2 . Despite the presence of both mating types in Nabeul, Kef and Jendouba, the hypothesis of random mating was rejected in these locations with multilocus gametic disequilibrium tests. Highly significant genetic differentiation ( θ = 0·32, G ST = 0·28, P < 0·001) was detected among populations and genetic distance and cluster analyses based on pooled allele frequencies revealed that populations from Nabeul and Kef were distinct from those in Beja, Bizerte and Jendouba. More than 70% of total gene diversity ( H T = 0·55) detected was attributable to variation within populations compared to 28% among populations. This result, coupled with the occurrence of private alleles in each population, suggests that gene flow is currently limited among populations, even those separated by short geographic distances. The presence of two main genetic clusters was confirmed using Bayesian model-based population structure analyses of multilocus genotypes (MLGs) without regard to geographic origin of samples. The presence of MAT1-2 isolates in both clusters suggests at least two independent introductions of MAT1-2 into Tunisia that are likely to be the result of importation and planting of infected chickpea seeds.