Identification of novel candidate loci and genes for seed vigor-related traits in upland cotton (Gossypium hirsutum L.) via GWAS

Abstract

Seed vigor (SV) is a crucial trait determining the quality of crop seeds. Currently, over 80% of China's cotton-planting area is in Xinjiang Province, where a fully mechanized planting model is adopted, accounting for more than 90% of the total fiber production. Therefore, identifying SV-related loci and genes is crucial for improving cotton yield in Xinjiang. In this study, three seed vigor-related traits, including germination potential, germination rate, and germination index, were investigated across three environments in a panel of 355 diverse accessions based on 2,261,854 high-quality single-nucleotide polymorphisms (SNPs). A total of 26 significant SNPs were detected and divided into six quantitative trait locus regions, including 121 predicted candidate genes. By combining gene expression, gene annotation, and haplotype analysis, two novel candidate genes (Ghir_A09G002730 and Ghir_D03G009280) within qGR-A09-1 and qGI/GP/GR-D03-3 were associated with vigor-related traits, and Ghir_A09G002730 was found to be involved in artificial selection during cotton breeding by population genetic analysis. Thus, understanding the genetic mechanisms underlying seed vigor-related traits in cotton could help increase the efficiency of direct seeding by molecular marker-assisted selection breeding.

Keywords: GWAS; candidate genes; germination rate; seed vigor; upland cotton.

Figure 1

Single-nucleotide polymorphism distributions in the upland cotton genome. The number of SNPs within a 10-Mb window. A01–A13 and D01–D13 on the Y axis are the numbers of the 26 chromosomes. The X axis represents chromosome length (Mb).

Figure 2

Phenotypic variation analysis of seed vigor (SV)-related traits. (A–I) Distributions of the mean values for the germination index (GI), germination potential (GP), and germination rate (GR) in three environments, respectively. (J) Correlation analysis of SV-related traits (GI, GP, and GR) in three environments (**P < 0.01). E1, Huanggang-2021; E2, Sanya-2021; and E3, Sanya-2022.

Figure 3

Genome-wide association study results for seed vigor-related traits. (A–C) Manhattan plots of GI-BLUP, GP-BLUP, and GR-BLUP for single-nucleotide polymorphism (SNP) markers, respectively. Significant SNP markers are distinguished by black lines. (D–F) QQ plots for GI-BLUP, GP-BLUP, and GR-BLUP, respectively.

Figure 4

Variation analysis of the germination rate (GR)-associated gene Ghir_A09G002730 in the candidate region. (A) Local Manhattan plots for GR-related genes on chromosome A09 and linkage disequilibrium heat map for the candidate region within 21.9 kb. (B) Box plots for GR of the two haplotypes mentioned above (**P < 0.01). (C) Differentiation of the genetic diversity distribution of Ghir_A09G002730 in four geographic areas (NIR, Northwest Inland region; YZRR, Yangtze River region; YRR, Yellow River region; and NSER, Northern Specific Early-Maturity region). (D) Gene structure diversity of Ghir_A09G002730 across three breeding stages. (E) Heat map of candidate gene expression patterns in the seed germination stage (0, 5, and 10 h) on chromosome A09.

Figure 5

Variation analysis of seed vigor (SV)-related traits associated with qGI/GP/GR-D03-3 in the candidate region. (A) Local Manhattan plots for SV-related genes on chromosome D03 from 30 to 33 Mb. (B) Top two haplotypes of qGI/GP/GR-D03-3 in 355 upland cotton accessions. (C) Comparison of germination potential between accessions containing Hap1, Hap2, Hap3, Hap4, and Hap5. Letters on the violin plot indicate significant differences according to one-way ANOVA (LSD test; P < 0.05). (D) Comparison of seed germination status for 3 days between Hap1 and Hap2. (E) Heat map of candidate gene expression patterns in the seed germination stage (0, 5, and 10 h) on chromosome D03.