ARS | Publication request: Genome-wide Association Mapping of Fusarium Head Blight Resistance in Contemporary Barley Breeding Germplasm

Submitted to: Molecular Breeding
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: April 6, 2010
Publication Date: April 1, 2011
Citation: Massman, J., Cooper, B., Horsley, R., Neate, S., Dill-Macky, R., Chao, S., Dong, Y., Muehlbauer, G., Smith, K. 2011. Genome-wide Association Mapping of Fusarium Head Blight Resistance in Contemporary Barley Breeding Germplasm. Molecular Breeding. 27:439-454.

Interpretive Summary: The disease Fusarium head blight (FHB) caused by a fungal pathogen has devastated the barley growing regions in the U.S., particularly in the Northern Plains. The mycotoxin deoxynivalenol (DON) is found in infected seed and renders seed unacceptable for the brewing industry. Molecular markers if found closely linked to the genes controlling disease resistance will allow breeders to apply marker-assisted selection (MAS) in their breeding programs to effectively screen and select lines resistant to the disease. The inheritance of FHB resistance is complex and strongly influenced by environment. Linkage mapping has been the common practice to map resistance genes on chromosome regions and locate closely linked markers. This process can be achieved through the use of populations derived from crossing two parents with genetic differences followed by an analysis of co-segregation of the markers with the trait. Previous results based on nine such bi-parental mapping studies have identified and located many quantitative trait loci (QTL) underlying the FHB resistance on all seven barley chromosomes. DNA markers for only a few of these QTL have been validated by subsequent studies and used in breeding programs. Recently, the association mapping approach has become an alternative tool to find DNA markers closely associated with a trait using a mechanism known as linkage disequilibrium (LD). LD is the association of a gene or DNA marker with its neighboring DNA markers, which form a block that can be passed on from parents to progeny without breaking up. It facilitates mapping of a complex trait with finer resolution. Because individuals used in association mapping need not be closely related, creation of a mapping population is not required. Breeding germplasm can thus be readily used for association mapping studies. In this report, we present results from mapping barley FHB resistance using 768 breeding lines. The DNA marker used is single nucleotide polymorphism (SNP), which is a marker system that can differentiate individuals based on variations detected at the level of a single nucleotide base in the genome. A set of 1,536 gene-derived SNP markers widely distributed on barley chromosomes was developed and used to genotype all the breeding lines. QTL were mapped using a statistical method that accounts for relatedness among lines to avoid detecting false marker-trait associations. Four QTL were identified for FHB and eight QTL were identified for DON, some of which agreed with previous bi-parental results. Our studies show that using breeding germplasm to map QTL can complement bi-parental mapping studies by providing independent validation, and potentially more precise QTL locations. Association mapping within breeding germplasm can bridge the gap between basic genetic studies and applied breeding.

Technical Abstract: Utilization of quantitative trait loci (QTL) identified in bi-parental mapping populations has had limited success for complex quantitative traits. The use of association mapping in contemporary breeding germplasm may lead to more effective marker strategies for crop improvement. To map Fusarium head blight (FHB) resistance in barley, 768 breeding lines were evaluated in 2006 and 2007, in four locations. All lines were genotyped with 1,536 SNP markers and QTL were mapped using a mixed-model that accounts for relatedness among lines. Average linkage disequilibrium within the breeding germplasm extended beyond four cM. Four QTL were identified for FHB and eight QTL were identified for the deoxynivalenol (DON), the toxin resulting from FHB, in two independent sets of breeding lines. The effects of all QTL were small explaining 1-5% of the phenotypic variation. We show that using breeding germplasm to map QTL can complement bi-parental mapping studies by providing independent validation, potentially more precise QTL locations, and the ability to resolve questions of linkage and pleiotropy. Association mapping within breeding germplasm can bridge the gap between basic genetic studies and applied breeding.