| With the increasing of population and the continuous decrease of cultivated land, by 2030, China’s population will reach 1.6 billion, the demand for food will increase, consequently, the crop yield should increase more than 50% to meet the safety of the food supply. In order to fully satisfy the future demand for rice, the breakthrough breeding of new varieties become the urgent matter. And rice yield increase is essentially the process, which alternative improved of source-sink and continuously building new balance relationships between source and sink. The previous study of rice source-sink relationship mostly focused on the research of external morphological index, while the molecular level research on source-sink correlation properties are relatively few. This study takes 31 countries and regions of 261 indica rice core germplasm resources as the experimental materials to conduct the experiment at two sites (Hainan, Guangdong) in two years. The phenotypic data of source sink-related traits of the tested material were investigated. Using 6704 SNP markers, genome-wide association analysis of source-sink related traits was conducted to mining excellent gene loci. The purpose was to improve rice source-sink characteristics by molecular marker assisted breeding, provide germplasm resource, genes and markers information for cultivating source-sink relationship coordination, good comprehensive traits breakthrough rice varieties, and improve the efficiency of breeding. The main results were as follows:(1) The difference analysis of source sink-related traits in two sites indicated thatIn Hainan, the maximum coefficient of variation of related source traits is area of top second leaf, variation coefficient was 26.21%, while in Guangdong sites for sword leaf weight, the variation coefficient was 28.34%, the smallest variation coefficient of source properties of the two sites was sword leaf specific leaf weight,10.05% and 14.81%, respectively. The maximum coefficient variation of sink traits of two sites were flat grains number, up to 62.06% and 73.71%, the minimum coefficient of variation was seed setting rate,7.82% and 13.38%, respectively. Correlation analysis showed that there was the same related level between source traits and between sink traits, although there was partly the same related level between source and sink traits, the relationship almost the same. Those results provide important basis for further exploring the relationship between rice source sink-related traits.(2) The principal component analysis showed that in the two sites, the 19 source sink-related traits of indica rice core germplasm resources both can be simplified as independent 6 principal components, the six main component information account for total information 87.16%and 84.68%, respectively in Hainan and Guangdong, covering all 19 indicators. Related source sink traits accordance with each principal component in the two sites. They both classified into two source factors, three sink factors and one source-sink coordination factor. The germplasm of TD25,685, CT 6510, NARGUNI ANGIFOTSY, ELWEE comprehensive ranking in the top 20 both in the two sites, which indicating the five varieties not only have relative good source-sink comprehensive characteristics, but also have a better ability to adapt to different ecological environment.(3) Cluster analysis showed that the materials in the two sites were divided into three groups, which was source limited type, sink limited type and source-sink coordinate type. The relationship of source and sink of 81 materials in different sites has not changed, including source limiting type 17,18 libraries limit type,46 source-sink coordinate type, illustrating that those rice materials could be less affected by environment in the process of growth and development, and had strong ability to adapt to different environment. In the two sites, the effective panicles, grain number per panicle of source limited type were the largest in three types, while seed setting rate,1000-grain weight and yield per plant were the smallest in three types; The effective panicles, grains number per panicle of sink limited type varieties were the smallest in three types, while seed setting rate,1000-grain weight were the maximum in three types; The yield per plant of source-sink coordinate type was the highest in three types, the rest yield component were modest between the other two types(4) In the two sites, the dry matter accumulation regularity of the three types was almost the same. The dry matter accumulation in leaves, stem and sheath, the three types all decreased along with the growth and the development, the largest was the source-sink coordinate type; The panicle weight of source limited type was the largest, followed by source-sink coordinate type and sink limited type. In plant total dry matter accumulation, they were all the largest in mature stage, moderate in booting stage, and the smallest in full panicle stage. In the three types, the leaf dry matter output rate and conversion rate of source-sink coordinate type was the largest, the stem and sheatli dry matter output rate and conversion rate of sink limited type was the largest. In total, the organ dry matter output rate of the three types in the two sites both were stem and sheath larger than leaf, the contribution to yield formation was the stem and sheath larger than leaf, too.(5) 6704 SNP marker distributed on 1-12 chromosomes, the average physical distance between all the SNP marker was 58 kb, there are 3 756 minimum allele frequency less than 0.20 marked by SNP, account for 56.03%, 384 minimum allele frequency closed to 0.5 marked by SNP, account for 5.73%. Polymorphism information content (PIC) with a mean of 0.24, the range between 0.05 and 0.38, the average genetic diversity was 0.29, the range between 0.05 and 0.50,261 indica rice core germplasm resources have a total 1 749 744 SNP markers, the base changes related to AA, CC, GG, TT, AC, AQ AT, CG, CT, GT, the GG type was the largest, followed by AA, CC and TT.(6) 261 rice germplasm population structure analysis showed that the optimal subsets of group number was three, including 88,82 and 91 materials respectively. Linkage disequilibrium analysis indicated that 124 301 pairs combinations in 319 900 linear pairs combinations in the station of LD state (P< 0.01), accounting for 38.86% of the total linear combinations, when D’= 0.5 and r2= 0.1, LD attenuation distance was 590 and 509 kb, respectively, existing strong LD levels. Genetic relationship (Kinship) analysis showed that there were unrelated or weaker relationship of most of the 261 indica rice core germplasm resources, reflecting the universality and diversity of the selected materials, only a minority had strong genetic relationship.(7) Correlation analysis results showed that under the P≤0.001 significant level or less, it detected significantly associated sites of source-sink related traits 103 and 86 respectively in Hainan and Guangdong. The loci detected in Hainan mainly distributed in chromosome 2,3,4, respectively, including 14,16,22 related sites; The loci detected in Guangdong mainly distributed in chromosome 1,2, ll,including 32,13,13 related sites. It detected 96 QTLs associated with 10 source related traits,62 loci in Hainan, and 34 loci in Guangdong; It detected 93 QTLs associated with 9 sink related traits,41 loci in Hainan, and 52 loci in Guangdong. In Hainan, it detected 18 loci associated with multiple source related traits, and 9 loci in Guangdong; it detected 8 loci associated with multiple sink related traits in Hainan, and 13 loci in Guangdong; it detected 11 loci simultaneous associated with different source and sink related traits in Hainan, and 7 loci in Guangdong.25 loci associated with specific traits were detected in the two sites, account for 13.23% of the total loci, and the physical distance all in 590 kb.4 loci significantly associated with the same traits were detected in the two sites, including weight of top second leaf, flat grain number and yield per plant. |
PDF Full Text Request