Combining Linkage Analysis And GWAS To Identify QTL Linked To Soybean Seed Protein And Oil Content

Soybean is the most important source of vegetable protein and edible oil with an average composition of approximately 40%protein and 20%oil content.Detecting quantitative trait loci（QTL）associated with protein/oil content is important for directing molecular breeding,and identifying genes whose functions affect the accumulation of protein/oil content is significant for understanding their mechanisms.In this study,linkage analysis and genome-wide association analysis（GWAS）were applied to detect QTL that are associated with protein/oil content in soybean.Genes related to protein/oil metabolism were excavated in significant and reproducible intervals,and preliminary functional verification was conducted for candidate genes.The main contents and results are as follows:1.Linkage analysis was applied using 308 F_2:7:7 recombinant inbred lines（RILs）that derived from the cross between high-oil soybean cultivar Heihe 27（HH27）and high-protein cultivar Zigongdongdou（ZGDD）,phenotypic data of protein and oil content were collected in six environments.In combination with the genetic map constructed by 3454 high-density single nucleotide polymorphism（SNP）markers,which were obtained by simplified whole genome resequencing（2b-RAD）before,we detected 24protein QTL and 24 oil QTL by two software.Among them,10 QTL were located by both software and10 QTL could be stably detected in two more environments.And q Oil5-1,qPro15-1,q Oil15-1,qPro20-1,qOil20-2 were reproducible and significant in multiple environments,making them suitable for further exploration of candidate genes.2.GWAS was conducted using 203 soybean accessions from China and America.Phenotypic data of protein and oil content collecting from four environments and 94462 SNP/InDel markers obtaining from whole-genome resequencing were combined to do GWAS.The association population could be divided into two subgroups,and the linkage disequilibrium（LD）decay distance of the four-region-population and the total population was among 132 kb～190 kb.The mixed linear model was used to discover 19,12 and 36 significant SNP loci in two subgroups and total population,respectively.Among them,the region of 34.99 Mb～41.46 Mb on chromosome（Chr.）20 was associated with ample SNP loci that linked to protein/oil content in multiple environments.3.Four significant SNP loci regions were co-detected by both linkage analysis and GWAS.A total of 25 candidate genes related to protein/oil metabolism were excavated in two regions:a co-detected SNP region of 34.60 Mb～35.70 on Chr.20 Mb that based on 200 kb upstream and downstream,and an extra QTL interval qPro15-1/qOil15-1 of 2.60 Mb～3.50 Mb on Chr.15.The relative expression levels of these candidate genes in two cultivars（ZGDD and HH27）for one photoperiod condition and in two photoperiod conditions[short-day（12 h light/12 h dark）and long-day（16 h light/8 h dark）]for one cultivar were examined.It was found that four genes（Glyma.15g034600,Glyma.15g033200,Glyma.20g106900 and Glyma.20g108800）were differently expressed during the late reproductive growth stages between different cultivars/photoperiods,suggesting that the four genes may be linked to the accumulation of soybean protein/oil content,but further verification should be conducted.In summary,this study obtained significant and reproducible QTL and excavated four genes differentially expressed among different cultivars/photoperiod,which provides help for soybean high-oil/high-protein molecular breeding and understanding the genetic mechanism of protein/oil metabolism.