Variation and covariation for Gibberella ear rot resistance and agronomic traits in testcrosses of doubled haploid maize lines

Gibberella ear rot (GER) of maize caused by Fusarium graminearum reduces grain yield and leads to contamination of the grains with deoxynivalenol (DON), a mycotoxin that adversely affects the health of humans and animals. The objectives of this study were to (1) analyze means and genotypic variances for line per se performance (LP) and testcross performance (TP) of doubled haploid (DH) lines for GER severity and DON concentration as well as for some agronomic traits, (2) examine correlations among these traits, (3) validate QTL for resistance detected in previous studies for LP and their effect on TP and (4) investigate the relative efficiency of indirect selection (RE) for LP to improve TP. Testcross progenies of 94 DH lines originating from four flint populations were developed using a susceptible dent tester as pollinator. Artificial inoculations with F. graminearum led to appreciable disease development. Average TP for GER severity and DON concentration were lower than the mean mid-parent values of the tester and DH lines, indicating mid-parent heterosis for resistance. Genotypic variation for resistance was significant for LP and TP. Genotypic correlations between LP and TP were low and resistance QTL for LP had no significant effects on TP. Accordingly, RE for resistance was low, suggesting to allocate resources mostly to the evaluation of testcrosses. Correlations of resistance to GER and DON contamination with grain yield (measured under non-inoculated conditions) were not significant, indicating that selection for resistance and higher grain yield can be carried out simultaneously.