January 13-17, 2007
Town & Country Convention Center
San Diego, CA
Sarah B. Ware1 , Alexander H.J. Wittenberg1 , Els C.P. Verstappen1 , Theo VanderLee1 , Henk J. Schouten1 , Steve Goodwin2 , Jane Grimwood3 , Jim Bristow4 , Igor Grigoriev4 , Gert H.J. Kema1
Specificity of the haploid, plant-pathogenic fungus Mycosphaerella graminicola towards durum (tetraploid, genomes AABB) versus bread (AABBDD) wheat is well documented. To estimate the number of genes involved in host species specificity, we crossed an isolate of M. graminicola collected from bread wheat (IPO323) with one from durum wheat (IPO95052) and analyzed the pathogenicity of 163 progeny to four cultivars of bread wheat and three of durum. Segregation data for pathogenicity to the seven cultivars of wheat were added to a high-density genetic linkage map containing 1,144 Diversity Array Technology (DArT) markers. Combined analyses of the phenotypic data and the DArT map identified nine quantitative trait loci (QTL) for cultivar specificity on seven linkage groups. These genetic loci were aligned to the whole-genome sequence of M. graminicola to identify the genes involved in specificity. To our knowledge, this is the first comprehensive QTL analysis of virulence performed on a plant-pathogenic fungus. The previously accepted hypothesis that host species specificity among isolates of M. graminicola is under controle of a single gene appears to be incorrect. Instead, specificity of M. graminicola to durum versus bread wheat appears to be under multigenic control. Therefore, pathogenicity to different species of cultivated wheat most likely is conditioned by a mix of genetically inherited factors, and the distinction between host species and cultivar specificity is not clearly defined.