Department of Genetics, Federal University of Viçosa, Viçosa, Minas Gerais, 36570-900, BRAZIL.
Drought is the most important abiotic stress resulting in significant yield losses in maize (Zea mays L.). Maize now recognized as one of the major and eminent food security crops due high yield potential compared with another important crops as rice and wheat. Development of more drought-tolerant genotypes can contribute to ensure food security mainly in developing areas in Africa, Asia and Latin America where this crop is a staple food. However, selection for drought tolerance is difficult because it is a complex trait with strong interactions between genotypes and environments and limited knowledge about the role and regulation of tolerance mechanisms. Classical breeding have identified morphophysiological traits related to grain yield under drought conditions. Most of these traits are polygenic, but grain yield probably remains the most polygenic and complex trait. The availability of molecular markers allowed mapping Quantitative Trait Loci (QTLs). It is a promising tool for detection constitutive and adaptive genomic regions controlling drought tolerance as well as for studying changes in the expression of these loci across varying environmental conditions. These genomic regions may be considered target for markers-assisted selection (MAS) program to develop drought tolerant genotypes. We evaluated three tropical maize populations from CIMMYT’s Global Maize Program under water stress (WS) and well-watered (WW) regimes in Mexico, Kenya and Zimbabwe to provide an understanding of the genetic basis of yield and secondary traits involved in response to water-limited conditions at flowering time. To achieve this goal we conducted QTL mapping studies in single and multiple environments as well as a meta-QTL analysis to identify the most prone genomic regions across populations to be useful in MAS program. Grain yield (GY) and anthesis-silking interval (ASI) were measured in Mexican and African environments, while secondary traits, as ears per plant (EPP), stay-green (SG) and plant and ears height (PEH) were realized only in Mexican environments. Drought stress reduced GY around 50% and increased ASI above 80%. Interestingly another morphophysiological traits like EPP, SG and PEH were not markedly affected by water shortage. In general drought stress tends to reduce genetic variance of GY, while secondary traits remain to be stable or even higher under water scarcity. Also, high correlation between morphophysiological traits and GY were observed under drought condition. Grain yield QTLs showed strong interactions with the environment (QEI) and changed their positions on the genome across environments. Whereas QTLs for secondary traits tend to be more stable across water regimes. Meta-QTL analysis reveals clusters of QTLs for grain yield and secondary traits, such as anthesis-silking interval, ears per plant, stay-green and plant and ears heights on chromosomes 1 (bin 1.06 at 161.07-183.83 Mb) and 10 (bin 10.04-06 at 111.26-141.82 Mb) while one interesting cluster of all secondary traits were detected on chromosome 3 (bin 3.06 at 169.75-178.23 Mb) under both water regimes. The confidence interval of metaQTLs regions harbored several genes, available in maize database (http://www.maizegdb.org), that were involved in diverse networks controlling development, metabolism and responses to biotic and abiotic stresses. The target regions identified by QTL mapping can contribute to complementing the evaluation and selection of improved germplasm, especially in poor areas with high risk of drought as in sub-Saharan Africa.