Genome-wide association study reveals novel SNP loci and candidate genes linked to flowering time in upland cotton.

Key message: A total of 456 SNPs associated significantly with FT via GWAS and three candidate genes related to flowering were identified via RNA‒seq, qRT‒PCR and VIGS. Flowering time (FT) is one of the main traits associated with early maturity in upland cotton; however, genetic basis and candidate genes underlying FT remain inadequately understood. In this study, 1,574,032 high-quality single nucleotide polymorphisms (SNPs) were identified on the basis of resequencing data from 619 upland cotton lines, and among them, 418 core germplasms were selected and genome-wide association studies (GWASs) were conducted to identify 456 SNPs that were significantly associated with FT. Variant annotation of significant SNPs revealed that 25 of these SNPs resulted in nonsynonymous mutations in eight genes. Three early-flowering-favouring haplotypes (A02_Hap3, D10_Hap3 and D11_Hap3) and two early-flowering-favouring alleles (D09_6523710_GG and D09_50028094_AA) were identified by haplotype/allele analysis. By RNA‒seq and qRT‒PCR, three candidate genes (GhFRO7, GhCML1 and GhPCMP-E88) were also shown to be differentially expressed between early-flowering and late-flowering varieties. Virus-induced gene silencing (VIGS) experiments further verified the critical roles of the three genes, which were important regulators underlying the late-flowering phenotype in upland cotton, in the regulation of FT. Moreover, selection pressure analysis revealed that these three candidate genes might have experienced artificial or natural selection. In addition, we developed two cleaved amplified polymorphic sequence (CAPS) markers, which have potential application value. These findings will provide a new theoretical basis for the genetic improvement of early maturity-related traits in upland cotton and lay the foundation for the breeding of excellent early maturing varieties.