Identification Of Fiber Quality Genes By Using Cotton Oligonucleotide Microarrays With Gene Expression Analysis

The two cultivated tetraploid cottons, Gossypium hirsutum with wide adaptation and high yield and Gossypium barbadense with high fiber quality produce important natural fiber for the textile industry in the world. Significant interspecific differences exist in morphology, physiology, and pest resistance between the two species. However, little is known on differential gene expressions on a genome-wide basis. Gossypium barbadense is known by its high fiber quality, but the two cultivated tetraploid species originated from a common ancestor 1-2 million years ago, significant interspecific differences exist and because of hybrid breakdown between the two species, it is hard to transfer desirable traits such as fine fiber quality from G..barbadense to Upland cotton.To solve the problem, backcross inbred lines(BIL) has become a very important introgressed lines. Although a number of QTLs for fiber quality traits were reported, gene expression is still unknown.A 3-year field trial on Giza75 and SG747 in 4 environments was conducted. Select the extremity ones for fiber quality, and then group them. The Affymetrix Cotton GeneChip of 17 BILs and Giza75,SG747 were then used to perform a genome-wide transcriptome analysis in 10 days post-anthesis (DPA) fibers. The fiber quality genes were screened from the large number genes based on the bioinformatics analysis and microarray. It provides important reference to the cloning and functional verification of genes related to fiber development and paves the way for the mining of candidate genes responsible for fiber quality and its molecular marker-assisted breeding. The main results were showed as follows:1. The Affymetrix Cotton GeneChip was used to perform a genome-wide transcriptome analysis in 10 days post-anthesis (DPA) fibers between Giza75 and SG747. Of the 24,029 transcripts, 5,757 (23.96%) showed significant differential expression (DE) and 3,095 transcripts (12.88%) showed 2-fold or higher levels of expression changes. COG software was taken to do function prognosis and group.2. Through COG database, we can see that genes expressed in Giza75(12837)and SG747(13387) had been differently grouping, according to the metabolic process in which they participated, though the largest two part was the same. According to Chi-square test, most of the definations have significance at the 0.05 and 0.01 level between Giza75 and SG747 except three ones 3. Through quantitative RT-PCR analyses on different plant tissues and developing fibers of five stages from 5 to 25 DPA, of 12 selected DE genes including SAHH, CESA4,α-expansin 2, 2α- tubulin genes, 3β- tubulin genes, and 4 unknown genes, 11 had similar results to the microarray analysis.The results indicated that the comparative microarray results were biologically reproducible, although none of the DE genes were fiber specific.This study represents the first investigation using microarray analysis to compare differential gene expressions between Upland and Pima cotton.4. Select the extremity ones for fiber quality, and then group them, by variance analysis. The Affymetrix Cotton GeneChip was then used to perform a genome-wide transcriptome analysis in 10 days post-anthesis (DPA) fibers to identify quality genes.By the significance of difference standard , selected 1490,1038,259 transcripts at fiber length,fiber strength and fiber micronaire, respectively, with P=0.05.5. Then use GO, COG, KOBAS software to do data analysis.There are 34,26,7 items of GO in fiber length, fiber strength, fiber micronaire group,respectively. The three largest categories of COG were General function prediction only, Translation, ribosomal structure and biogenesis and Posttranslational modification, protein turnover, chaperones.6. According to P-value and Fold-Change, we select 10 genes: GhiAffx.24025.1.S1_at, Ghi.10655.1.S1_s_at, Ghi.3578.1.S1_s_at, ACO1, ARF1, SAHH,β-tubulin1,β-tubulin10,TUA6 , TUA7.And then quantitative RT-PCR analyses on different plant tissues and developing fibers of five stages from 5 to 25 DPA, of 12 selected DE genes was detect.All had similar results to the microarray analysis. It provides important reference to the cloning and functional verification of genes related to fiber development and paves the way for the mining of candidate genes responsible for fiber quality and its molecular marker-assisted breeding.