Variation for Resistance to Kernel Infection and Toxin Accumulation in Winter Wheat Infected with Fusarium graminearum

Host resistance is the main way to control Fusarium head blight (FHB) in wheat. Despite improved levels of resistance to infection and spread in vegetative tissue, the toxin deoxynivalenol (DON) can still accumulate to unacceptable concentration levels. In this study, our objectives were to assess the genetic variation for resistance to kernel infection (RKI) and resistance to toxin accumulation (RTA) and their role in controlling DON. We collected spikes with different levels of visual symptoms from each of 32 wheat genotypes and at four environments and determined DON and fungal biomass (FB) from each sample. We assessed RKI by regressing FB on the level of visual symptoms and RTA by regressing DON on FB for each genotype. Significant genetic effects were found for RKI and RTA. Some genotypes consistently had low FB in their grain despite increasing visual symptoms suggesting RKI. Additionally, some genotypes consistently had low DON in their grain despite increasing FB levels suggesting a higher RTA in these genotypes. The variation for RKI and RTA explained a significant fraction of the variation for DON among genotypes with moderate visual symptoms using independent grain samples. Although RKI and RTA were significantly correlated (r == 0.58, P == 0.05), RTA was more predictive of DON accumulation because it modeled 32 to 44% of the genotype sum of squares for DON, while only 9 to 10% were predicted using RKI. Thus, variation for RTA was important in explaining variation for DON among genotypes with acceptable levels of resistance to fungal infection and spread. This work indicates that there is a need to develop a better understanding of RTA and rapid screening methods for this trait.