Fine mapping of stable QTLs related to eating quality in rice (Oryza sativa L.) by CSSLs harboring small target chromosomal segments

Abstract

Amylose content (AC) and viscosity profile are primary indices for evaluating eating and cooking qualities of rice grain. Using chromosome segment substitution lines (CSSLs), previous studies identified a QTL cluster of genes for rice eating and cooking quality in the interval R727-G1149 on chromosome 8. In this study we report two QTLs for viscosity parameters, respectively controlling setback viscosity (SBV) and consistency viscosity (CSV), located in the same interval using rapid viscosity analyzer (RVA) profile as an indicator of eating quality. Previously reported QTL for AC was dissected into two components with opposite genetic effects. Of four QTLs, qCSV-8 and qAC-8-2 had stable genetic effects across three and four environments, respectively. qSBV-8, qCSV-8 and qAC-8-1 partly overlapped, but were separated from qAC-8-2. Based on data from an Affymetrix rice GeneChip, two genes related to starch biosynthesis at the qAC-8-2 locus were chosen for further quantitative expression analysis. Both genes showed enhanced expression in sub-CSSLs carrying the target qAC-8-2 allele, but not in sub-CSSLs without the target qAC-8-2 allele, indicating their possible role in rice quality determination. Molecular markers closely linked to the two stable QTL provide the opportunity for marker-assisted selection (MAS) in breeding high quality rice.

Keywords: Chromosome segment sub-situtution lines; Eating and cooking quality; QTL; Rapid viscosity analyzer profiles.

Fig. 1

Breeding scheme for the development of the CSSLs and progeny selection. To obtain the substituted segments on target chromosome, CSSL₅₀ was backcrossed to Asominori followed by self-pollination until BC₄F₁, and the genetic background was selected by SSR markers on target/non-target chromosome. The BC₄F₁ individuals, designated as CSSL50-1, harbored a small segment spanning R727–X397. Sub-CSSLs, the progenies of CSSL50-1, were employed for further study. AA type and II type, represent lines homozygous for Asominori (AA) and IR24 (II) alleles, respectively, were used to identify the phenotype. And IA type, represents lines heterozygous for Asominori and IR24 alleles, was used to further shorten the segments on target interval.

Fig. 2

Graphical genotypes of the long arm of chromosome 8 of sub-CSSLs and QTL regions for grain quality detected in previous studies. A: Relative locations of genes controlled AC on the chromosome 8. Putative gene regions for ISA previonsly detected by Fujita et al. (1999). Putative QTL regions for amy-8 and qAC-8 mapped by and Aluko et al. (2004) and Hao et al. (2009). Putative gene cluster for AC and RVA profile detected by Wan et al. (2004). The region labelled by dotted line denotes positions of the gene clusters. The relative positions of RFLP markers used in the analyses are shown on the left. The new SSR markers in this study are shown on the right. The genetic distances (cM) between adjacent SSR markers are shown on the left of SSR markers. B: Graphical genotypes of the long arm of chromosome 8 of sub-CSSLs and four identied QTLs in this study. Blocks represent chromosomes. Horizontal lines show the positions of SSR markers investigated. To graphically represent the genotypes of CSSLs, the recombination point was arbitrarily determined at the mid-point between adjacent markers in different genotype. Black and gray blocks denote homozygous IR24 and heterozygous Asominori alleles, respectively. The physical distance (Kb) of adjacent SSR markers are shown on the left of SSR markers.

Fig. 3

Substitution mapping of QTL for rice quality and phenotype (AC and RVA profile) for sub-CSSLs located in the interval G1149–R727 on chromosome 8 in four environments. Blocks represent chromosomes. Horizontal lines show the positions of SSR markers investigated. To graphically represent the genotypes of CSSLs, the recombination point was arbitrarily determined at the mid-point between markers. Black and white blocks denote homozygous IR24 alleles and Asominori alleles, respectively. NJ and HN, Naijing and Hainan, respectively. ** and * represent significant differences from Asominori at P = 0.001 and P = 0.01, respectively by t-tests. A: Substitution mapping of QTL for SBV, CSV and AC located in the interval RM5351–RM5485 on the left. Phenotype means for 4 sub-CSSLs (Lines 1, 2, 3 and 4) and Asominori are shown on the right table. B: Substitution mapping of QTL for AC located in the interval RM5485–RM23653 on the left. Phenotypic means for 5 sub-CSSLs (Lines 10, 11, 12, 13 and 14) and Asominori are shown on the right Table. The serial numbers for sub-CSSLs corresponded to the final sub-CSSLs. ND, no data.

Fig. 4

Ten bacterial artificial chromosome (BAC) clones and two up-regulated genes in the interval RM23510–RM23579. The two up-regulated genes with higher expression ratios are marked by the arrows.

Fig. 5

Expression of two up-regulated genes during seed development of CSSL50-1 and seven sub-CSSLs. Total RNA was extracted from seeds at 6, 9, 12, 15 or 18 DAF. For each gene, the expression in Asominori seeds at 6, 9, 12, 15 or 18 DAF was set as a control. The relative expression value was obtained by the ratio of sub-CSSLs (or CSSL50-1) and Asominori. All data are means ± SD from three replicates. A: Os08g0534900, B: Os08g0536000