Association Analysis And Candidate Genes Mining Of Plant Architecture-related Traits And 100-seed Weight In Soybean

Soybean[Glycine max(L.)Merr.]is an important economic and oil crop,because its seeds are rich in plant protein and oil and are widely cultivated all over the world.In recent years,with the rapid development of Chinese economy and the adjustment of the dietary structure of residents,the demand for soybeans has risen sharply.Increasing soybean yield has become the main task of soybean breeders in China,and improving soybean plant architecture and increasing 100-seed weight(100-SW)are effective ways to increase soybean yield.Soybean plant architecture traits such as plant height(PH),number of nodes on main stem(NN),branch number(BN)and stem diameter(DI)are closely related to plant biological yield,lodging and photosynthetic efficiency.100-SW is an important component of soybean yield and an important appearance quality trait.Soybean plant architecture traits and 100-SW are all complex quantitative traits controlled by multiple genes,and their genetic mechanism and breeding potential remain to be determined.At present,the development of high-throughput sequencing technology and statistical analysis methods provides an effective tool for the genetic analysis of important agronomic traits and the mining of related genes in soybean.Mining functional markers/genes related to soybean plant architecture and 100-SW from soybean germplasm resources can greatly accelerate the process of molecular breeding in soybean.In this study,soybean local germplasm population was used to association mapping based on genome-wide SNP markers for four plant architecture-traits of PH,NN,BN and DI.The aim is to identify markers that are significantly associated with target traits,to screen for candidate genes and abundand allelic variations.The secondary segregation populations were constructed derived from CSSL3228(a wild soybean chromosome fragment substitution line material)and Nannong(NN1138-2)hybridization.Fine mapping of PH QTLs based on QTL-seq and linkage analysis,and candidate genes were analyzed by qRT-PCR and bioinformatics.The restricted two-stage multi-locus genome-wide association analysis(RTM-GWAS)was used to detect the SNPLDBs associated with 100-SW,and the corresponding QTL-allele matrix was constructed to predict the breeding potential of 100-SW.The main results of this research are as follows:1.Genome-wide association studies of soybean plant architecture-related traitsPhenotypic identification of PH,NN,BN and DI of 133 soybean local germplasm under three environments(Jiangpu in 2016,Jiangpu in 2017 and Dangtu in 2017),and 82,187 high quality SNPs were used to GWAS for four plant architecture-related traits.The results showed that there were extensive phenotypic variation in PH,NN,BN and DI,and 4 traits were affected by environment.The heritability was 88.85%,93.53%,65.17%and 67.39,respectively.The heritability of PH and NN were higher,and BN and DI were lower.GWAS found that a total of 27 SNPs were significantly associated with target traits in at least two environments,with PH,NN,BN and DI of 10,12,2 and 3,respectively.Among these SNPs,7 SNPs identified were located within the reported target trait QTL,and 20 SNPs were new markers in the study.Single-domain analysis of candidate intervals(LD ±119.07 kb)in which four SNP markers(AX-90498802,AX-90317160,AX-90420194 and AX-90460927)with high phenotypic variation explanation,AX-90420194 and AX-90460927 were found to be located in the haplotypes of 124.69 kb and 134.51 kb,respectively.A total of 8 candidate genes related to plant architecture were predicted in candidate interval or haplotype interval of the above four markers.The significant association of SNP markers and predicted candidate genes identified in this study can help to analyze the genetic mechanism of soybean plant architecture-related traits,and lay a theoretical foundation for the mining of plant architecture-related trait genes and molecular marker-assisted selection breeding in soybean.2.Fine mapping and candidate genes analysis of the PH QTL in soybeanCSSL3228 is a chromosomesegment substitution line(CSSL)material constructed by using wild soybean N24852(G.soja)as the donor parent and NN1138-2(G.max)as the recipient parent.Its PH is significantly higher than the recipient parent NN1138-2.In this study,the 349 F2:3 and F2:4 populations derived from a cross of CSSL3228 and NN1138-2.Phenotypic identification of F2:3 families and F2:4 families in two environments was carried out.Fine mapping for PH QTLs based on QTL-seq and linkage mapping,and the candidate genes were analyzed by qRT-PCR and bioinformatics.The results showed that there were extensive phenotypic variation in PH in CSSL3228 and NN1138-2 and offspring populations.The heritability of F2:4 families populations were 88.97%and 85.24%in two environments,respectively.A PH QTL was detected in the 1.73 Mb genomic region of soybean chromosome 13 by QTL-seq.The SSR markers were encrypted in this candidate interval,and the PH QTL was mapped on the chromosome 13 between BARCSOYSSR₁3 1417 and BARCSOYSSR₁3 1421.The physical distance was about 69.3 kb,and the QTL was named qPH13.1.The LOD values were 8.11,14.35 and 11.86 in the three environments,and phenotype variation explanation were 10.31%,14.94%and 14.67%,respectively.qRT-PCR analysis of 12 genes in the candidate interval,it was found that there were 4 genes differentially expressed between the two parents.Bioinformatics analysis indicated that the CDS sequence of Glyma.13g249400 is 1122 bp in CSSL3228,encoding 373 amino acids,a protein molecular weight of 40.57 KD,an isoelectric point of 8.73,and a conserved U-box domain.Glyma.13g249400 gene has six differential amino acids in two parents CSSL3228 and NN1138-2,and there are differences in protein secondary structure and tertiary structure,which may be a candidate gene for PH in soybean.This study of soybean PH QTL fine mapping and candidate genes identification,for gene cloning of PH and marker-assisted selection provides theoretical basis in soybean.3.Genome-wide association studies of soybean 100-SW and breeding potential predictionIn this study,the RTM-GWAS method was used to analyze the 100-SW of 133 soybean local germplasm in the two environments of Jiangpu in 2017 and Dangtu in 2017.The QTL-allele matrix was constructed based on the effect value of the significantly associated SNPLDBs to predict the breeding potential of 100-SW and the parent materials with excellent allelic variation of 100-SW were selected.The results showed that there were 29 SNPLDBs significantly associated with 100-SW,which were located on 16 chromosomes except for chromosome 1,4,12 and 17,containing 73 allelic variations,a total of 98.32%of phenotypic variations were explained.The distribution of high-generation pure lines with 100-SW in 8,778 combinations of 133 soybean local germplasm was predicted,and 33 expected excellent combinations were obtained.The 95%quantile of the predicted population of 33 excellent combinations ranged from 35.74 to 39.53,which was 0-10.6%higher than the maximum 100-SW of the parental population.The best combination was P5F01(Lvrouheipidou)and P1E07(Cudou),and the parent P5F01 appeared multiple times in 33 excellent combinations.It may be a better complementary material with other materials of 100-SW.Among 33 excellent combinations,34 parent materials appeared at least once,34 materials were widely distributed in 16 provinces of China,among which Jilin Province(6),Jiangsu Province(5),Shandong Province(4),Shanxi Province(3),Guangdong Province(3),Zhejiang Province(2)and Fujian Province(2)had 100-SW materials with more excellent allelic variation.In this study,the 100-SW breeding potential prediction and the screening of excellent materials will provide a theoretical basis for the selection of parents in field breeding,and provide a material basis for high-yield breeding in soybean.