Identification and characterization of regions of the rice genome associated with broad-spectrum, quantitative disease resistance

Abstract

Much research has been devoted to understanding the biology of plant-pathogen interactions. The extensive genetic analysis of disease resistance in rice, coupled with the sequenced genome and genomic resources, provides the opportunity to seek convergent evidence implicating specific chromosomal segments and genes in the control of resistance. Published data on quantitative and qualitative disease resistance in rice were synthesized to evaluate the distributions of and associations among resistance loci. Quantitative trait loci (QTL) for resistance to multiple diseases and qualitative resistance loci (R genes) were clustered in the rice genome. R genes and their analogs of the nucleotide binding site-leucine-rich repeat class and genes identified on the basis of differential representation in disease-related EST libraries were significantly associated with QTL. Chromosomal segments associated with broad-spectrum quantitative disease resistance (BS-QDR) were identified. These segments contained numerous positional candidate genes identified on the basis of a range of criteria, and groups of genes belonging to two defense-associated biochemical pathways were found to underlie one BS-QDR region. Genetic dissection of disease QTL confidence intervals is needed to reduce the number of positional candidate genes for further functional analysis. This study provides a framework for future investigations of disease resistance in rice and related crop species.

Figure 1.—

Integrated disease QTL map of rice. The scale at the top represents centimorgan values from the JRGP 2000 genetic map. Rice chromosomes are represented as gray bars and labeled 1–12, with the centromeres colored black within the chromosomes. Disease QTL and lesion mimic loci are shown as bars and R genes are represented by shapes above the chromosomes. Letter designations for QTL indicate the parental alleles associated with resistance at the corresponding loci (L, Lemont; T, Teqing; Mo, Moroberekan; Mi, Minghui 63; Z, Zhenshan 97B; S, SHZ-2; A, Azucena; J, Jasmine 85; O, Owarihatamochi). Resistance gene analogs (RGAs) of the NBS-LRR class are shown below the chromosomes.

Figure 2.—

Percentage of FL-cDNA representative gene families with a minimum of 20 members, residing in the disease QTL genomic fraction. Each gene family was defined as the set of FL-cDNAs harboring an individual InterPro domain. The solid line at 59% indicates the percentage of the entire catalog of nonredundant FL-cDNAs (n = 21,446) that were located in the QTL genomic fraction, or the mean value. The false discovery rate (FDR) for the most significant gene families is shown.

Figure 3.—

Cluster analysis of genes with significantly different EST abundances detected in a digital Northern analysis (Bonferroni correction, α = 0.05). The EST libraries described in Table 2 are indicated at the top horizontal axis. The following abbreviations were used to indicate the type of plant-pathogen interaction: (1) I, incompatible; (2) C, compatible; and (3) P, partial. Genes are represented by branches along the left vertical axis. The FL-cDNA clone name corresponds to the longest (in base pairs) cDNA for a given FL-cDNA cluster. The centimorgan values correspond to the rice JRGP 2000 high-density genetic map (unknown map locations are symbolized by a question mark). The symbols in the QTL column indicate whether the FL-cDNA colocalized with a QTL (+) or did not colocalize (−) or that the map position was unknown (?). Arabidopsis homologs were identified by BLASTp searches and match with an E-value ≤e⁻¹⁸. The relative abundance of ESTs across libraries is indicated by the intensity of the color red.