Fiber strength is an important trait among cotton fiber qualities due to ongoing changes in spinning technology. Major quantitative trait loci (QTL) for fiber quality enable molecular marker-assisted selection (MAS) to effectively improve fiber quality of cotton cultivars. We previously identified a major QTL for fiber strength derived from 7235 in Upland cotton. In the present study, in order to fine-map fiber strength QTL, we chose three recombinant inbred lines (RIL), 7TR-133, 7TR-132, and 7TR-214, developed from a cross between 7235 and TM-1 for backcrossing to TM-1 to develop three large mapping populations. Phenotypic data for fiber strength traits were collected in Nanjing (JES/NAU) and Xinjiang (BES/XJ) in 2006 and 2007. Three simple sequence repeat (SSR) genetic linkage maps on Chro.24(D8) were constructed using these three backcrossed populations. The SSR genetic maps were constructed using 907 individuals in (7TR-133 x TM-1)F(2) (Pop A), 670 in (7TR-132 x TM-1)F(2) (Pop B), and 940 in (7TR-214 x TM-1)F(2) (Pop C). The average distance between SSR loci was 0.62, 1.7, and 0.56 cM for the three maps. MapQTL 5 software detected five-clustered QTL (2.5 < LOD < 29.8) on Chro.D8 for fiber strength following analysis of three RIL backcrossed F(2)/F(2:3) progenies at JES/NAU and BES/XJ over 2 years. Five QTL for fiber strength exhibited a total phenotypic variance (PV) of 28.8-59.6%.

Fiber quality is an important economic index and a major breeding goal in cotton, but direct phenotypic selection is often hindered due to environmental influences and linkage with yield traits. A genome-wide association study (GWAS) is a powerful tool to identify genes associated with phenotypic traits. In this study, we identified fiber quality genes in upland cotton (Gossypium hirsutum L.) using GWAS based on a high-density CottonSNP80K array and multiple environment tests. A total of 30 and 23 significant single nucleotide polymorphisms (SNPs) associated with five fiber quality traits were identified across the 408 cotton accessions in six environments and the best linear unbiased predictions, respectively. Among these SNPs, seven loci were the same, and 128 candidate genes were predicted in a 1-Mb region (±500 kb of the peak SNP). Furthermore, two major genome regions (GR1 and GR2) associated with multiple fiber qualities in multiple environments on chromosomes A07 and A13 were identified, and within them, 22 candidate genes were annotated. Of these, 11 genes were expressed [log2(1 + FPKM)>1] in the fiber development stages (5, 10, 20, and 25 dpa) using RNA-Seq. This study provides fundamental insight relevant to identification of genes associated with fiber quality and will accelerate future efforts toward improving fiber quality of upland cotton.

No abstract