Guozhong Zhu1, Sen Hou1, Xiaohui Song1, Xing Wang1, Wei Wang2, Quanjia Chen3, Wangzhen Guo4. 1. State Key Laboratory of Crop Genetics and Germplasm Enhancement, Cotton Germplasm Enhancement and Application Engineering Research Center (Ministry of Education), Nanjing Agricultural University, Nanjing, 210095, China. 2. Institute of Agricultural Sciences in Coastal Area of Jiangsu Province, Yancheng, 224002, China. 3. Engineering Research Center for Cotton (the Ministry of Education), Xinjiang Agricultural University, Urumqi, 830052, China. 4. State Key Laboratory of Crop Genetics and Germplasm Enhancement, Cotton Germplasm Enhancement and Application Engineering Research Center (Ministry of Education), Nanjing Agricultural University, Nanjing, 210095, China. moelab@njau.edu.cn.
Abstract
BACKGROUND: Numerous quantitative trait loci (QTLs) and candidate genes associated with yield-related traits have been identified in cotton by genome-wide association study (GWAS) analysis. However, most of the phenotypic data were from a single or few environments, and the stable loci remained to be validated under multiple field environments. RESULTS: Here, 242 upland cotton accessions collected from different origins were continuously investigated for phenotypic data of four main yield components, including boll weight (BW) and lint percentage (LP) under 13 field environments, and boll number per plant (BN) and seed index (SI) under 11 environments. Correlation analysis revealed a positive correlation between BN and LP, BW and SI, while SI had a negative correlation with LP and BN. Genetic analysis indicated that LP had the highest heritability estimates of 94.97%, followed by 92.08% for SI, 86.09% for BW, and 72.92% for BN, indicating LP and SI were more suitable traits for genetic improvement. Based on 56,010 high-quality single nucleotide polymorphisms (SNPs) and GWAS analysis, a total of 95 non-redundant QTLs were identified, including 12 of BN, 23 of BW, 45 of LP, and 33 of SI, respectively. Of them, 10 pairs of homologous QTLs were detected between A and D sub-genomes. We also found that 15 co-located QTLs with more than two traits and 12 high-confidence QTLs were detected under more than six environments, respectively. Further, two NET genes (GH_A08G0716 and GH_A08G0783), located in a novel QTL hotspot (qtl24, qtl25 and qlt26) were predominately expressed in early fiber development stages, exhibited significant correlation with LP and SI. The GH_A07G1389 in the stable qtl19 region encoded a tetratricopeptide repeat (TPR)-like superfamily protein and was a homologous gene involved in short fiber mutant ligon lintless-y (Liy), implying important roles in cotton yield. CONCLUSIONS: The present study provides a foundation for understanding the regulatory mechanisms of yield components and may enhance yield improvement through molecular breeding in cotton.
BACKGROUND: Numerous quantitative trait loci (QTLs) and candidate genes associated with yield-related traits have been identified in cotton by genome-wide association study (GWAS) analysis. However, most of the phenotypic data were from a single or few environments, and the stable loci remained to be validated under multiple field environments. RESULTS: Here, 242 upland cotton accessions collected from different origins were continuously investigated for phenotypic data of four main yield components, including boll weight (BW) and lint percentage (LP) under 13 field environments, and boll number per plant (BN) and seed index (SI) under 11 environments. Correlation analysis revealed a positive correlation between BN and LP, BW and SI, while SI had a negative correlation with LP and BN. Genetic analysis indicated that LP had the highest heritability estimates of 94.97%, followed by 92.08% for SI, 86.09% for BW, and 72.92% for BN, indicating LP and SI were more suitable traits for genetic improvement. Based on 56,010 high-quality single nucleotide polymorphisms (SNPs) and GWAS analysis, a total of 95 non-redundant QTLs were identified, including 12 of BN, 23 of BW, 45 of LP, and 33 of SI, respectively. Of them, 10 pairs of homologous QTLs were detected between A and D sub-genomes. We also found that 15 co-located QTLs with more than two traits and 12 high-confidence QTLs were detected under more than six environments, respectively. Further, two NET genes (GH_A08G0716 and GH_A08G0783), located in a novel QTL hotspot (qtl24, qtl25 and qlt26) were predominately expressed in early fiber development stages, exhibited significant correlation with LP and SI. The GH_A07G1389 in the stable qtl19 region encoded a tetratricopeptide repeat (TPR)-like superfamily protein and was a homologous gene involved in short fiber mutant ligon lintless-y (Liy), implying important roles in cotton yield. CONCLUSIONS: The present study provides a foundation for understanding the regulatory mechanisms of yield components and may enhance yield improvement through molecular breeding in cotton.
Entities:
Keywords:
Genome-wide association study; Gossypium hirsutum; Multiple field environments; Quantitative trait loci; Yield components
Authors: Z Jeffrey Chen; Brian E Scheffler; Elizabeth Dennis; Barbara A Triplett; Tianzhen Zhang; Wangzhen Guo; Xiaoya Chen; David M Stelly; Pablo D Rabinowicz; Christopher D Town; Tony Arioli; Curt Brubaker; Roy G Cantrell; Jean-Marc Lacape; Mauricio Ulloa; Peng Chee; Alan R Gingle; Candace H Haigler; Richard Percy; Sukumar Saha; Thea Wilkins; Robert J Wright; Allen Van Deynze; Yuxian Zhu; Shuxun Yu; Ibrokhim Abdurakhmonov; Ishwarappa Katageri; P Ananda Kumar; Yusuf Zafar; John Z Yu; Russell J Kohel; Jonathan F Wendel; Andrew H Paterson Journal: Plant Physiol Date: 2007-12 Impact factor: 8.340