Genome-wide analysis of yield in Europe: allelic effects as functions of drought and heat scenarios

Assessing the genetic variability of plant performance under heat and drought scenarios can contribute to reduce the negative effects of climate change. We propose here an approach that consisted of (1) clustering time courses of environmental variables simulated by a crop model in current (35 years × 55 sites) and future conditions into six scenarios of temperature and water deficit as experienced by maize (Zea mays L.) plants; (2) performing 29 field experiments in contrasting conditions across Europe with 244 maize hybrids; (3) assigning individual experiments to scenarios based on environmental conditions as measured in each field experiment; frequencies of temperature scenarios in our experiments corresponded to future heat scenarios (+5°C); (4) analyzing the genetic variation of plant performance for each environmental scenario. Forty-eight quantitative trait loci (QTLs) of yield were identified by association genetics using a multi-environment multi-locus model. Eight and twelve QTLs were associated to tolerances to heat and drought stresses because they were specific to hot and dry scenarios, respectively, with low or even negative allelic effects in favorable scenarios. Twenty-four QTLs improved yield in favorable conditions but showed nonsignificant effects under stress; they were therefore associated with higher sensitivity. Our approach showed a pattern of QTL effects expressed as functions of environmental variables and scenarios, allowing us to suggest hypotheses for mechanisms and candidate genes underlying each QTL. It can be used for assessing the performance of genotypes and the contribution of genomic regions under current and future stress situations and to accelerate breeding for drought-prone environments.