Submitted to: Rice Technical Working Group Meeting Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 12/1/2003
Publication Date: 6/1/2004
Citation: Pinson, S.R., Shahjahan, A.M., Rush, M.C. 2004. QTLs for panicle blight resistance and their association with resistance to other diseases. In: Rice Technical Working Group Meeting Proceedings, February 29-March 4, 2004, New Orleans, LA. 2004 CDROM.
Technical Abstract: After the discovery of the bacterial pathogen Burkholderia glumae as the cause of panicle blight disease in rice, an inoculated nursery screening procedure was established at the LSU Agricultural Center Rice Research Station in Crowley, LA for identifying genetically resistant germplasm. Major genes and QTLs conferring resistance and/or tolerance to other rice diseases were previously mapped by various researchers within a set of Lemont/Te-qing recombinant inbred lines. Genetic loci associated with resistance to sheath blight (SB, causal org. Rhizoctonia solani), bacterial leaf blight (BLB, causal org. Xanthomonas oryzae), and several races of blast (causal org. Pyricularia grisea) are now known within this gene-mapping population. The aim of the present study was to identify genetic loci conferring resistance to bacterial panicle blight (BPB) within this well-studied population of 300 RILs. The Lemont/Teqing RILs were evaluated for BPB severity in inoculated field plots in 2001 and 2002 with three replications each year. Susceptible checks within the study included Bengal (2-yr average disease severity rating = 7.1), Cocodrie (avg. 6.7) and Lemont (avg. = 6.1). Resistant lines were Nipponbare (avg. = 4.4) and Teqing (avg. = 4.3). Disease severity among the RIL plots ranged from 3 to 8. Rep-to-rep correlation coefficients for both 2001 and 2002 data were as low as 0.4 - 0.5. The 2001 and 2002 yearly averages were poorly correlated as well (r = 0.4). Disease resistance evaluations are often hampered by the appearance of resistance among genetically susceptible plots that escape disease. In spite of weekly application of B. glumae in order to inoculate each plot according to its individual optimum timing and growth stage, escapes were apparent within the present BPB evaluation. When we re-evaluated our ratings and removed low (resistant) scores when a particular RIL appeared much more susceptible in the other two replications in that year, the rep-to-rep correlation coefficients increased to r = 0.7-0.8, but the year-to-year correlation remained low (r = 0.4). QTLs are identified and mapped through statistical association between phenotypic data and molecular marker/allele data. This statistical association weakens with every error in the phenotypic or molecular data sets. The low year-to-year correlation in our phenotypic data was thus of some concern. We elected to conduct marker linkage analyses using several sets of phenotypic data - individual year plus 2-Yr averages of actual ratings, ratings adjusted for escapes, and maximum ratings per year. The raw, adjusted, and maximum data sets all identified similarly located loci, though they did differ for LOD size and percentage variance explained. Five BPB-QTLs (on chromosomes 1, 3a, 3b, 8 and 10) were solidly identified (LOD > 2.4) from analysis of 2-Yr-Averages. All five QTLs also acquired LOD greater than 2.0 in analysis of 2001 and 2002 individual year data, indicating that their resistance was expressed/effective in both years. An additional BPB-QTL on chromosome 11 was identified from 2002 and 2-Yr-Av analyses, but was not evident in the 2001 data. The BPB-QTL near RG348a on chromosome 3b had the largest effect in each of the data sets, explaining from 10 - 17% of the total variance. This locus, plus the BPB-QTLs located on chromosomes 8 and 11 are co-located with QTLs reportedly conferring resistance to both SB and BLB, with the resistance allele originating from Teqing in each case. Resistance at the BPB-QTLs on chromosomes 1 and 3b originates from Lemont. We report a total of six BPB-QTLs. Though they are considered putative in that they have been identified in a single population using data with relatively low reliability (year-to-year r = 0.4), confidence in three of the BPB-QTLs is increased by the fact that they are co-located with SB and BLB resis