Mapping Of Immature Fiber Mutant Gene And Expression Profile During Fiber Secondary Cell Wall Development Using Immature Fiber Mutant (Gossypium Hirsutum L.)

Cotton(Gossypium spp.) is an important economic crop and cotton fiber is leading natural textile fiber in the world. Cotton fiber is a unicellular hair originating from the seed epidermis. Fiber development consists of four overlapping stages:fiber initiation, fiber elongation, secondary cell wall deposition and maturation. Because of some unique characters of cotton fiber, cotton fiber is regarded as an ideal model for studies of plant cell elongation and cell wall biogenesis. Mutants are powerful tools to help us understand genetic and molecular mechanism of important genes. The immature fiber mutant (im) is a specific cotton fiber mutant, which regarded to result in defective secondry cell wall. The research work about im mutant will help us better understand the complex process of fiber secondary cell wall biogenesis and explore the molecular mechanism of gene regulation. It could also help us find some potential excellent genes for cotton gene engineering in improving fiber quality traits. In the present study, the inheritance analysis and mapping of the im mutant and QTL analysis for multiple fiber quality traits and lint percentage are performed. Also, the differential genes expression profiling of im mutant and TM-1during fiber secondary cell wall development is analyzed. The majory results are listed as following:1. The im mutant and two cotton lines, G hirsutum acc. TM-1and14005were compared for differences in fiber characteristics. The im mutant has no fluffy fibers like as in normal cotton lines and fibers in im mutant are still tightly matted around seeds, although the cotton bolls have opened. Two hybrid combination,(im×TM-1) and im×I4005) are developed by using im mutant corssing with TM-1and14005. All F1plants had nomal fiber characteristics. However, among all of the F2segregation populations, only two phenotypes, normal and matted fibers, were found. The data in each segregation population fit a Mendelian pattern of inheritance with a ratio of3:1(Table1), which reconfirmed that im was ahomozygous recessive gene.2. A total of6,713SSR primer pairs were screened to detect polymorphisms among im,14005and TM-1. Two hundred and fifty-seven polymorphic SSR markers were detected between im and14005. One hundred and ninty-three polymorphic loci from186SSR markers were mapped on31linkage groups, and the im locus and four SSR loci (NAU1190, DPL0170, NAU3995and NAU1197) were identified in a linkage group. According to previous mapping information, the linkage group was assigned on chromosome3(A3), which covered55.5cM, with an average distance of about13.9cM. The DPL0170and NAU3995loci were linked to the flanking the im locus, with a distance of21.8cM and32.5cM, respectively. However, only3polymorphic SSR markers were detected between im and TM-1and anchored on chromosome A3. They were NAU3639, NAU1190and DPL0170, and all were linkaged with im locus, with a distance of10.2,24.3and23.3cM, respectively.3. In order to lay a good foundation for future map based cloning of the im gene, fine mapping of the im gene was performed. Three F2mapping populations, CSIL028×im F2, CSIL030×im F2and CSIL031×im F2, were constructed by using im mutant respectively corssing with three different chromosome segment introgression lines (CSILs), which were characterized by carrying one homozygous chromosome3(A3) segment from G. barbadense cv Hai7124in the background of a genetic standard line of the upland cotton, TM-1. Molecular mapping showed that47,7and12polymorphic SSR loci were individually detected and anchored on the three selected overlapping CSILs on chromosome3(A3). Of these, the linkage map constructed based on (CSIL028×im) F2contained the most markers and covered80.9cM. with an average distance of1.8cM. The im gene was anchored between BNL2443and cgr6528loci with a distance4.6cM and1.3cM, respectively. The SSR markers linked to the im locus on the linkage map were detected based on (CSIL030×im) F2and (CSIL031×im) F2. The nearest marker to the im locus was NAU5444with a distance of30.3cM and NAU3479with a distance of40.0cM respectively. In addition, the linkage analysis showed that the introgression chromosome segment transferred from G. barbadense cv. Hai7124in the CSIL030and CSIL031lines might carry an im allelic locus not from Hai7124, but from TM-1. 4. A comparison analysis was performed between im mutant and other five cotton lines,14005, TM-1, CSIL028, CSIL030and CSIL031for fiber quality traits and lint percentage. Except for short fiber index, TM-1had higher fiber quality index than im mutant on the tarits including fiber length, fiber strength, micronaire, fiber maturity, fiber uniformity ratio and lint percentage (P<0.001). The results of QTL analysis based on five populations showed that QTLs related to multi-traits in each population were clustered near the im locus with a high percentage of phenotypic variance, ranging from18.2%to82.03%. The meta-analysis for QTLs from five populations revealed some consensus QTLs by using BioMercator v2.1program. Except for fiber strength, QTLs for other six traits including fiber length, micronaire, fiber uniformity ratio, short fiber index, fiber maturity and lint percentage which clustered near by im locus were all regarded as on consensus QTL. They were located on a small interval, with95%confidence interval1.22,0.72,0.56,0.89,0.76and0.67cM, respectively. They shared same flanking markers, BNL2443and cgr6528and had mean of phenotypic variation that ranged from27.16%to69.61%.5. The cellulose content in fibers was measured at different fiber developmental stage including19,22,25,30,35and40DPA, and maturation both in im mutant and TM-1. The im mutant had significantly lower cellulose content in fibers than TM-1at each fiber developmental stage. The accumulation of cellulose in im might be propobably suppressed partially. The statistical analysis for thickness of fiber cell wall was also performed at13,19and25DPA and maturation developmental stage. There was no difference between two lines at13and19DPA, but im had thinner fiber cell wall (P<0.001) than TM-1at25DPA and maturity, suggesting the defect of the development of fiber secondary cell wall (SCW) in im mutant and thinner cell wall in im was most likely due to the reason for reduction of cellulose. The fructose, glucose and sucrose content in fibers at10,13,16,19,22,25,30,35, and40DPA were also measured. TM-1and im mutant shared similar changing trend. But the fructose and glucose content were higher in fibers of im mutant than that in TM-1during the later SCW development. During the developmental interval25～35DPA, higher sucrose content in fibers of TM-1was found than that in im mutant, which might be relative to different cellulose synthesis content between two accessions. A presumption was that the normal carbohydrate metabolism was affected in im mutant resulted from the im mutant gene, and sucrose distributed into the pathway of cellulose synthesis in im mutant was less than that in TM-1.6. In order to explore the mechanism of fiber development and reveal potential mutant candidate genes, we employed cotton cDNA microarray to evaluate the variation of transcription profiling within and between im and TM-1at13,16,19,22and25DPA. The comparison analysis for the variation of transcription profiling between adjacent time points during fiber development within two lines revealed that TM-1had more differentially expressed genes (DEGs) than im mutant (fold change>2, and FDR<0.05), during the fiber developmental interval13～16,16～19, and19～22DPA. In particular, during the fiber developmental interval13～16DPA, only98DEGs were detected in im mutant, whereas494DEGs were detected in TM-1at same fiber developmental interval. Conversely, more genes were differentially expressed in im mutant than that in TM-1between22and25DPA. There were236genes differentially expressed in im mutant but only60genes differentially expressed in TM-1. The results revealed different variation levels of transcription profiling within the two lines. And there was a putative delayed fiber developmental progress in im mutant. The maximum transcriptional variation between TM-1and im mutant occurred at16DPA and a total of1308genes altered their expression levels, which suggested that16DPA was a very important fiber developmental stage. The GO annotation for DEGs and enrichment analysis indicated that onset of secondary cell wall biosynthesis for fibers in TM-1approximately occurred at16DPA, whereas the same fiber developmental program in im was delayed until19DPA, suggesting an asynchronous fiber developmental program between TM-1and im mutant. At the some fiber developmental atage of SCW,16and19DPA, most enriched GO terms for DEGs between TM-1and im mutant were associated with carbohydrate metabolism and cell wall biogenesis progress. Many genes which have been confirmed to be required for normal development of SCW in Arabidopsis were included in these GO terms. The expression levels of these genes in TM-1were higher than that in im mutant at this fiber developmental stage. However, these GOs associated with carbohydrate metabolism and cell wall biogenesis progress were not over-represented at subsequent developmental stages, suggesting that im gene probably had a major effect on the early stage of secondary cell wall biosynthesis. The im gene might be an important regulatory gene that participated in regulating carbon flux into cellulose synthesis. 7. Using the recently released cotton D genome sequence information as reference and combined with the im gene mapping results, one hundary and fifty-five genes were predicted to be as im candidate genes. However, only8genes were differentially expressed between TM-1and im mutant. But the difference of gene expression level was not large. The result of GO analysis indicated that73genes involved in different biological processes. Out of them, twenty genes involved in signaling, biological regulation, growth, developmental process, reproduction and carbohydrate metabolic process, respectively. But these genes did not show significant expression difference between TM-1and im mutant. This might also be related to the representative genes in the chip. The KEGG analysis for155genes showed that18genes participated in different pathways, and the carbohydrate metabolism was one of the major pathways. That the roles of these genes played during fiber development and the functional analysis of im candidate gene from these predicted genes required further experimental validation.