Estimation of deoxynivalenol (DON) content by symptom rating and exoantigen content for resistance selection in wheat and triticale

Fusarium head blight (FHB) in wheat and triticale leads to contamination of the grain with the mycotoxin deoxynivalenol (DON) that is harmful to animal and man. A fast, low-cost, and reliable method for quantification of the DON content in the grain is essential for selection. We analysed 113 wheat and 55 triticale genotypes for their symptom development on spikes, Fusarium exoantigen (ExAg) and DON content in the grain after artificial inoculation with a highly aggressive isolate of E culmorum in three (wheat) and six (triticale) location-by-year combinations. Additionally, in triticale the amount of Fusarium damaged kernels (FDK) was assessed. ExAg content was analysed by a newly developed Fusarium-specific plate-trapped antigen enzyme-linked immunosorbent assay (PTA-ELISA) and DON content by an immunoassay. A moderate disease severity resulted in an ExAg content of 0.87 optical density (OD) units in wheat and 1.02 OD in triticale. DON content ranged from 12.0 to 105.2 mg kg(-1) in wheat and from 24.2 to 74.0 mg kg(-1) in triticale. Genotypic and genotype-by-environment interaction variances were significant (P < 0.01). Coefficient of phenotypic correlation between DON content analysed by the immunoassay and ExAg content was r = 0.86 for wheat and r = 0.60 for triticale. The highest correlation between DON content and symptom rating was found by FHB rating in wheat (r = 0.77) and by FDK rating in triticale (r = 0.71). In conclusion, selection for reduced FHB symptoms should lead to a correlated selection response in low fungal biomass and low DON content in the grain.