LAUSR

Population genetic structure of the worldwide-distributed wheat pathogen Zymoseptoria tritici has been extensively studied at large geographical scales, but to a much less extent at small or local spatial scales. A total of 627 single-conidial fungal isolates were sampled from several locations in northern France (Hauts-de-France Region) to assess fungal genetic structure at region, field, plant, and leaf layer scales, using highly polymorphic microsatellite markers and mating type idiomorphs. Important and overall similar levels of both gene and genotype diversities (gene diversity values of ≥0.44 and haplotype frequencies of ≥94%) were found at all the examined scales. Such rates of diversity are likely due to an active sexual recombination in the investigated areas, as revealed by equal proportions of the two mating types scored in all sampled populations. Interestingly, a rare occurrence of clones among lesions from a same leaf, as well as among leaves from different plant leaf layers (e.g., upper versus lower leaves), was highlighted, indicating that ascospores contribute much more than expected to Z. tritici epidemics, compared with pycnidiospores. Population structure and analyses of molecular variance revealed significant genetic differentiation at the regional scale (GST = 0.23) and, as expected, not at the other more local scales (GST ≤ 0.01). Further analyses using Bayesian and unweighted neighbor-joining statistical methods detected six genetic clusters within the regional population, overall distributed according to the locations from which the isolates were sampled. Neither clear directional relative migration linked to the geographical distribution of the locations, nor isolation by distance, were observed. Separate evolutionary trajectories caused by selection and adaptations to habitat heterogeneity could be the main forces shaping such structuration. This study provides new insights into the epidemiology and the genetic structure of Z. tritici at small local and, for the first time, at single plant and leaf layer scales. Such findings would be helpful in implementing effective control strategies.