Stem rust, caused by the fungus Puccinia graminis Pers. f. sp. tritici Ericks, is one of the most damaging diseases of wheat (Triticum aestivum L.). The recent emergence of the stem rust Ug99 race group poses a serious threat to world wheat production. Utilization of genetic resistance in cultivar development is the optimal way to control stem rust. Here, we report association mapping of stem rust resistance in a global spring wheat germplasm collection (2152 accessions) genotyped with the wheat iSelect 9K single-nucleotide polymorphism array. Using a unified mixed model method (or QK method), we identified a total of 47 loci that were significantly associated with various stem rust resistance traits including field disease resistance and seedling resistance against multiple stem rust pathogen races including BCCBC, TRTTF, TTKSK (Ug99), and TTTTF. The 47 loci could be further condensed into 11 quantitative trait locus (QTL) regions according to linkage disequilibrium information among adjacent markers. We postulate that these QTLs represent known stem rust resistance genes including Sr2, Sr6, Sr7a, Sr8a, Sr9h, Sr13, Sr28, and Sr36. We further employed a multilocus mixed model to explore marker-trait associations and identified two additional QTLs (one potentially represents Sr31) that were significantly associated with stem rust resistance against various races. Combinations of the most significant loci for each trait explained up to 38.6% of the phenotypic variance. Markers identified through this study could be used to track the genes or QTLs. Accessions with high numbers of resis-tance-associated alleles may serve as important breeding materials for stem rust resistance.
Bibliographical noteFunding Information:
This study is part of the Triticeae Coordinated Agriculture Project (www.triticeaecap.org), funded by USDA National Institute of Food and Agriculture Grant no. 2011-68002-30029. We thank Dr. Shiaoman Chao (USDA-ARS, Fargo, ND) for performing 9K SNP chip genotyping of the GWAS panel. We thank Sheri Rynearson (Washington State University) and Qi Wang (University of Minnesota) for technical assistance in DNA extractions and genotyping. We thank Dr. Katherine Frels (University of Minnesota) for reviewing an earlier draft of this manuscript. We acknowledge Amy Fox for facilitating the stem rust field and seedling experiments. Computing resources from the Minnesota Supercomputing Institute at the University of Minnesota are greatly appreciated.