Dissection Of The Genetic Variation And Candidate Genes Of Lint Percentage By A Genome-wide Association Study In Upland Cotton

Cotton is a major source of natural textile fiber and a significant cash crop worldwide.Upland cotton(Gossypium hirsutum),an allotetraploid species,occupies approximately 95% of global cotton production.Improving the yield of cotton has always been a primary goal of cotton breeding.Lint yield,an important measure of cotton yield,depends on boll number(BN),lint percentage(LP),boll weight(BW)and other factors.However,the improvement of the LP trait via phenotypic selection in the field by traditional breeding approaches is ineffective.Identifying genetic variation in LP and the genes underlying this trait is therefore essential.In this study,a population consisting of 276 upland cotton accessions were planted and observed in multi-environment and genotyped by CottonSNP63 K array.In all,10,660 high-quality SNPs were identified.We then performed a GWAS to determine the relationship between genetic loci and the LP trait.1.The construction,population structure and kinship analyses of upland cotton populationIn this study,a total of 276 upland cotton materials were collected,including representative cotton varieties that played an important role in different historical stages and obtained from the National Mid-term Gene Bank for Cotton,China,and the breeding basic materials accumulated in the long-term high-yield breeding practice of the research team,to construct natural population.Moreover,the population genetic structure was analyzed using those 10,660 SNPs.Those results showed that this population was divided into two subgroups.In addition,the kinship analysis revealed that most kinship coefficients were less than 0.2.According to all of the above results,the tested accession panel was not highly structured and exhibited weak relatedness.2.Phenotypic analysis of LPIn this study,LP were investigated and statistically analyzed in multiple environments.The results showed that LP was distributed between 10.49% and 49.62% in seven environments,and the phenotypic data presented rich variation and approximate normal distribution.The generalized heritability of LP was up to 90.7%.Analysis of variance showed that genotype,environment and interaction between genotype and environment reached extremely significant level.Correlation analysis showed that there was a positive correlation between the clothing phenotypes in different environments.3.Marker-trait GWASIn all,twenty-three SNP loci randomly were identified as significantly associated with the LP using the mixed linear model(MLM).Among these loci,eleven significant SNPs were consistently detected in at least two environments.Following the definition of QTL,15 QTLs were detected in total.Moreover,nine QTLs were co-localized with the previously reported QTLs,and six QTLs was first identified.4.Identification of candidate genesIn total,434 candidate genes were identified in the QTL regions.Analysis of the TM-1 RNA-seq data revealed that 263 of these genes were specifically highly expressed in different organs.Moreover,two non-synonymous SNP loci located on Dt05 chromosome were significantly correlated with the LP in multiple environments.qRT-PCR analysis for the expression pattern of Gh_D05G0313 and Gh_D05G1124 in 17 different upland cotton tissues revealed that they were highly expressed during ovule and fiber development.We speculate that these two genes may be candidate genes for LP.By investigating the genotypes of natural population materials at the above two SNP loci,it was found that the phenotypes of LP of materials containing different genotypes were also different,showing that the more favorable alleles were pyramided in varieties with the larger average LP values increasing.These results suggest that LP is positively correlated with the number of favorable alleles and these favorable alleles displayed pyramiding effects on LP.