Combined population analyses for mapping loci conditioning resistance to Southern corn leaf blight

Abstract

Southern leaf blight (SLB) is a fungal disease that attacks the leaves of maize plants, forming tan elliptical lesions. The causal agent is Cochliobolus heterostrophus, a necrotrophic ascomycete, which is endemic to hot, humid maize-growing regions. Most maize hybrids have at least a moderate level of quantitative resistance to SLB, which is the primary method of control for the disease. Quantitative resistance is conditioned by small effects of many genes; thus, to make most efficient use of the genetic variation present for SLB resistance, the number of genes involved, the effects they have singly and in combinations, and their potential interactions with the environment must be elucidated. Because the maize-SLB pathosystem is an excellent model for host-necrotroph genetics, the identification of genes underlying the disease response will not only accelerate breeding progress for SLB resistance, but also for necrotrophic resistance in other crop species. Several attempts have been made to map quantitative trait loci (QTL) responsible for conditioning resistance to SLB in biparental segregating populations; however, the precision of the positional estimates of the resultant loci does not allow for identification of genes underlying the response. In addition, the limited germplasm studied only offers the potential of examining two alleles per locus. Many more alleles exist in breeding germplasm, and for this reason, usefulness of results are limited. The objective of the two studies presented here is to address these limitations by combining data across distinct populations. The first study jointly analyzed data from four independently derived B73 x Mo17 populations to validate the existence of and fine map an SLB resistance QTL in bin 3.04, the most significant QTL detected in a study of the IBM population (a set of advanced intercross lines derived from a B73 x Mo17 cross). The four populations used consisted of the IBM population, a set of recombinant inbred lines (RILs), and two sets of F2:3 lines derived from crosses between near isogenic lines (NILs, lines genetically identical except for the 3.04 region). These F2:3 lines allowed for validation of the 3.04 QTL in a uniform genetic background. Combined analysis of data from the four populations yielded a smaller positional confidence interval than the IBM study interval, in which 3 canidate leucine repeat kinase genes lie. The second study utilized the Nested Association Mapping (NAM) population to provide a global view of the genetic architecture of SLB resistance across a broad range of maize germplasm. The NAM population consists of 5000 RIL derived from crosses between B73 and 25 diverse founder inbred lines. The full set of lines was phenotyped for resistance to SLB, and linkage analysis was used to identify 32 QTL segregating for SLB resistance. Additive effects explained 79.6% of the phenotypic variation; no significant epistatic interactions were detected between pairs of additive loci. To compare efficacy of the NAM approach with previously published studies, QTL identified using NAM and previously identified QTL were positioned on the IBM2008 map. 60% of NAM QTL had been detected previously, however, 13 novel QTL were detected in this study. The advantage to the NAM approach lies in not only the increased population size and statistical power, but also the wider sampling of alleles.