Identification of GWA-QTLs and germplasm useful for enhancing sheath blight resistance

Authors: Eizenga, Georgia; Pinson, Shannon; LI, DANTING - Guangxi Academy Of Agricultural Sciences; ZHANG, FANTAO - Jiangxi Normal University; Edwards, Jeremy; Jackson, Aaron

Submitted to: Rice Technical Working Group Meeting Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 12/4/2019
Publication Date: 1/6/2021
Citation: Eizenga, G.C., Pinson, S.R., Li, D., Zhang, F., Edwards, J., Jackson, A.K. 2021. Identification of GWA-QTLs and germplasm useful for enhancing sheath blight resistance. Proceedings of 38th Rice Technical Working Group Meeting, February 24-27, 2020, Orange Beach, Alabama. p. 114-115. Electronic Publication.

Interpretive Summary:

Technical Abstract: Sheath blight (ShB) disease is one of the most economically damaging rice (Oryza sativa L.) diseases worldwide, reducing grain yields up to 50%. The causal agent, Rhizoctonia solani Kühn, is a soil-borne fungus that spreads by runner hyphae. There are no known major resistance genes, only partial resistance, and many reported ShB-QTL are associated with plant architectural traits detrimental to grain yields such as tall height, late maturity, wide culm angle, or reduced tiller number (TN). To date, only three genes have been fine-mapped, one on chromosome (chr.) 9 at 21.4-21.5 Mb, and two on chr. 11 at 4.8-4.9 Mb and within 27.0–28.3 Mb. Also, a R. solani phototoxin gene, Rsn1, is on chr. 7 (18.1 Mb). To identify novel ShB QTL not confounded by plant architecture traits, genome-wide association (GWA) mapping studies were conducted in both Arkansas, USA and Nanning, China, and accessions carrying potentially novel resistance alleles were identified for use as breeding donor parents. This study included 417 Rice Diversity Panel 1 (RDP1) accessions from the five major subpopulations, indica (99 accessions), aus (62 accessions), tropical japonica (109 accessions), temperate japonica (116 accessions), aromatic (15 accessions) and 16 admixtures. RDP1 was evaluated for ShB disease with the greenhouse microchamber (MC) method (both locations, 3 replications), and field studies in Arkansas (two years, two replications) and Nanning (one year, three replications) using local R. solani isolates. The disease index was calculated from the lesion length for the MC studies. ShB disease was rated on a “0” (no disease) to “9” (nearly dead) scale and days to 50% heading (DH) and plant height (PH) measured for the field studies. RDP1 was evaluated for TN at early and late tillering (three replications), and panicle number (PN) in the greenhouse in Arkansas. Using the SAS 9.4 generalized linear mixed model procedure, both LSmeans and BLUPs were calculated for the ShB, PH and DH data, and BLUEs for the TN and PN data. Summary statistics, correlations and regression analyses were conducted in JMP 14. For RDP1 GWA mapping, the mixed linear model in Tassel 5 was used with the 700K SNP genotypes for 396 accessions. As validation, the GWA mapping of the previously reported ShB MC data on the Rice Minicore Collection (RMC) was analyzed with the 3.2 million SNPs for 173 accessions. Overall, disease severity was higher in Nanning than Arkansas. Based on MC disease index data, more resistant accessions were found in indica than other subpopulations. Arkansas field data identified 21 resistant indica accessions but only 11 in Nanning, six of which were resistant in both environments. Eight tropical japonica accessions were resistant in Nanning field tests but only two in Arkansas and none in common, suggesting differences between the local R. solani isolates. There was a strong correlation for ShB ratings between the Arkansas field trials but a low correlation between the field and MC ShB scores in both environments. Field ShB was highly negatively correlated with height in Arkansas, but positively correlated in Nanning, where RDP1 headed about 19 days earlier, and was shorter overall. Reduced ShB severity was associated with desirable increases in TN and PN. GWA ShB-QTL were selected from regions with only a few or no significant SNPs associated with PH or DH because many prior ShB-field QTL were confounded with undesirable PH and DH. Significant SNPs were then surveyed to find ShB-QTL regions where field and MC QTL overlapped, Arkansas and Nanning QTL overlapped, or where either the field or MC QTL were supported by multiple SNPs. Ten ShB QTL regions were chosen: qShB3 (17.7-20.8 Mb), qShB4 (24.4-24.9 Mb), qShB5 (7.6-8.5 Mb), qShB6-1 (6.5 Mb), qShB6-2 (23.0 Mb), qShB7 (19.8-24.2 Mb), qShB9 (21.2 Mb), qShB10 (4.9-5.3 Mb), qShB11 (16.4-20.9 Mb) and qShB12 (14.