QTL Mapping For Rice Grain Shape And Grain Weight Based On Constructed Genetic Linkage Map

In this study, an F₂ population was established from a cross between rice lines BG1 with large grain and XLJ (Xiaolijing ) with small grain for constructing whole genome genetic linkage map . The F₂ population and F3 families derived were planted in Hangzhou in 2009 and Hainan in 2010, respectively. Using the linkage map and grain shape, grain weight data of F₂ populations in Hangzhou and Hainan for QTL mapping based on analysis with QTLNetwork-2.0 and WinQTLCart-2.5. Through the F_{2: 3} progeny test to verify the main effect QTL, and sequence alignment methods were used to analyze whether BG1 and XLJ containing the grain shape genes which have been cloned. Specific conclusions were as follows.(1) BG1 and XLJ with extreme difference in grain shape and grain weight. Grain length was 12.77±0.54 mm and 5.91±0.22 mm, 4.32±0.18 mm and 2.65±0.06 mm for grain width, 57.66 g and 12.57 g for grain weight. Bimodal distribution was shown on grain length, grain length/width and thousand-grain weight in the F₂ population, and normal distribution of grain width and grain thickness , and existence of transgressive segregation for grain weight,grain length and grain length /width. Only grain length/width had higher value than BG1 and XLJ in reciprocal cross F1 material, other grain shape and grain weight values of F1 were close to BG1, and there were no differences in reciprocal cross. For example, the grain weight of reciprocal cross F1was 44.82 g and 44.47 g, thus the grain shape and grain weight of BG1 and XLJ was not completely dominant traits, and mainly controlled by nuclear genes.(2) Correlation analysis based on grain shape and grain weight data in Hangzhou in 2009 and Hainan in 2010,Grain length and length / width with significant positive correlation, and correlation coefficient of 0.872 and 0.774, while small correlation coefficient between grain length and grain width , grain length and grain thickness. Grain width and length / width were significant negative correlation, grain width and grain thickness was significantly correlated at 0.654 and 0.380. Grain thickness was only significant positive correlation with grain weight. There were significant correlation between grain weight and grain length, grain weight and grain width, grain weight and grain thickness, grain weight and length / width , but the highest correlation were between grain weight and grain length, grain weight and grain thickness, grain weight and length / width , the correlation coefficient were 0.854 and 0.720, 0.668 and 0.697, 0.601 and 0.425 , respectively, indicating that grain length, grain thickness and length / width have a great impact on grain weight. Compare the differences between the paired data under the same grain trait in different environments of Hangzhou in 2009 and Hainan in 2010.The results for grain length |t|=0.09t0.01=2.63,length/width |t|=3.68>t0.01=2.63,grain thickness |t|=2.16t0.01=2.64, and at the 0.05 level grain thickness |t|= 2.16>t0.05 = 1.99. Thus the two environments have small impact on grain length, but other traits are environmentally sensitive.(3) Using Mapmaker / Exp 3.0 to construct genetic linkage map based on the genotype data of 150 SSR markers. Except chromosome7, 8, 9, part of SSR markers in other chromosomes were not linked, which result in only 137 of 150 SSR markers linked to the 12 chromosomes with 16 linkage groups, of which markers in chromosome 1, 2 , 3, 10 were divided into two linkage groups, named as Chr1a, Chr1b, Chr2a, Chr2b, Chr3a, Chr3b, Chr10a and Chr10b. RM8213 and RM5414 on Chromosome 4, RM1200, RM1248 and RM289, RM7449, RM249 and RM3381 on chromosome 5, RM5515, RM444 and RM3609 on chromosome 9, RM3311, RM216 and RM474 on chromosome 10a, the order of SSR markers on which were different from physical order of public SSR markers. In the 16 linkage groups, the average genetic distances were 6.4 cM 20.9 cM, the total genetic distance of 1595.9 cM, the largest genetic distance was 44.0 cM, the smallest genetic distance 1.2 cM, the average genetic distance of 13.2 cM.(4) Using grain shape and grain weight data in Hangzhou in 2009 and Hainan in 2010, QTLNetwork-2.0 software detected seven grain shape QTLs and one grain weight QTL, which were distributed on chromosome 2, 3 and 5. On chromosome 2, qSW-2 in RM5791-RM324 and qST-2 in RM492-RM5356 were detected as main effect QTL affecting grain width and grain thickness, respectively, meanwhile qSLWR-2 was a minor QTL for grain length/width. Only detected a minor QTL qSL-3 in RM3381-RM3638 on chromosome 3. qSL-5 and qSLWR-5 were detected as main effect QTL in RM3381-RM3638 on chromosome 5, which affected grain length and length / width, and qSWT-5 in RM249-RM3381 was detected as main effect QTL of thousand-grain weight . All QTLs detected had additive and dominance effect, and the effect of additive×environment was detected in grain thickness and thousand-grain weight traits. Meanwhile, epistasis effect was detected between RM3278-RM3280 and RM5414-RM3471 for grain thickness.(5) Genotype distribution with 48 big grain homozygotes, 49 small grain homozygotes and 105 big grain heterozygotes were detected from progeny test using F₂:3 population, which fits the 1:2:1 segregation ratio for a single dominance gene of grain length. The total individuals of three genotypes were used for gene mapping, and the results were consistent with QTL mapping. The gene qSL-5 was mapped between RM3381 and RM3638 with genetic distance of 10.7 cM from RM3381 and 21.9 cM from RM3638. Sequence alignment analysis showed that BG1 had a identity of 95.8% to the gene qSW5 (GW5) , whereas XLJ only had 24.6%～56.1% similarity.(6) F₂ homozygotes were selected based on the genotype at RM3381 and subjected to t test between the two genotype groups. Mean values of grain length were 11.2 mm and 7.15 mm (t = 10.65> t0.01 = 2.41), 3.34 and 2.23 (t = 6.79> t0.01 = 2.41) for length / width, 38.436 g and 18.395 g (t = 10.83> t0.01 = 2.41) for grain weight. The mean values between dominant and recessive homozygous of different traits were highly significant, indicating RM3381 not only closely related to grain shape and grain weight, and could be used as selective marker in field. In this study, requiring further work on molecular marker-assisted selection for grain shape and grain weight, as well as fine mapping, cloning of the gene and elucidating the mechanism of rice yield traits.