| With the progress made in cotton molecular biology and functional genomics,especially thanks to the large scale EST (expressed sequence tag) sequencing projects,more and more fiber-related genes have been cloned and characterized and an increasingnumber of EST sequences are available in public database. These data pave the way forprofiling gene expression and discovering regulation mechanisms involved in cotton fiberdevelopment, meanwhile establish the basis for proteomics research on this nonsequenced organism. The target of function genomics is to make up the network ofgenome-gene expression-phenotype. In the complicated network ofGenome-Transcriptome-Proteome-metabome-Phenome, protein analysis is a notoriouschallenge. However, proteins are final products of most genes and execute variousfunctions in a cell, therefore, Reliance on Genomics as the sole tool for profiling geneexpression has a number of limitations to reflect genome function. Regulatorymechanisms in protein translation, degradation, compartmentation, and post translationmodifications, such as glycosylation, phosphorylation, acetylation, lipidation etc. moredirectly relate to cell functions and phenotypes. It is of great importance to carry outcotton proteomic research to unravel the molecular mechanisms underlying yield, qualityand stress resistance.In this study, we compared the staining results and sensitivities of 16 methods basedon Coomassie bright blue G-250 or R-250 in visualization of proteins in poly acylamidegel, the results demonstrate that colloidal Coomassie bright blue G-250 and modifiedMeyer's method with Coomassie bright blue R-250 gave the best staining result and ahigh sensitivity of 20 nanogram per band, the sensitivity of silver staining method is10nanogram per band, eosin did not work in our experiments. These results clarifyvariegated descriptions of the results and sensitivities of protein staining methods based on Coomassie bright blue. Colloidal Coomassie bright blue G-250 method, which producegel with high staining sensitivity, low background, without the necessity of destaining andcompatible to mass spectrometry, is preferred to apply in regular research. In this method,the duration of gel fixation with trichloroacetic acid influences gel background, whereasthe quality of water used in the whole electrophoresis affects staining sensitivity. Fixinggels in trichloroacetic acid for 4 hours and using super pure water exclusively in allsolution preparation can give best staining result. The sensitivity in silver staining methodis only as twice as that acquired with Colloidal Coomassie bright blue G-250 method, butthe former can stain protein in higher intensity. Therefore, when we select ColloidalCoomassie bright blue G-250 method to stain protein in gels, we should increase loadingamount in gel electrophoresis. In practice, we may use Colloidal Coomassie bright blueG-250 method and silver staining method to visualize two gels of the same sample inparallel or one gel of a sample sequentially to get more information.Protein extraction method is another technological factor which directly influence theprotein spot number and map quality in 2-Dimension electrophoresis gel. Our researchresults indicate that phenol extraction method can be applied to isolate proteins from alltissues in cotton, such as ovule, fiber, root etc., but trichloroacetic acid method does notwork in extraction protein from cotton ovule and/or fiber. Phenol extraction method ischaracterized by high harvest rate, increased diversity in pâ… and molecular weight, as wellas high solubility allowing to prepare high concentration sample. In early procedure ofprotein isolation, high concentration of polyvinylpolypyrrolidone, reducing agent of2-mercaptoethanol and appropriate amount of NBDH were added to eliminate polyphenoliccompounds in order to avoid their interference in 2-dimension electrophoresis and obtainoptimal isolation. Adding thiourea in protein lysis buffer can dissolve protein sufficientl.Using cup-loading sample application can get better and replicable 2-DE profile with highresolution and more spots, especially in high molecular weight.In aspect of protein identification, we compare the identification results throughMALDI-TOF and MALDI-TOF/TOF analysis followed by peptide mass fingerprint (PMF)and fragment fingerprint (FFP) alignment by MASCOT, ProFound, MS-Fit and Aldenteengine respectively. We believe that MALDI-TOF is useful for preliminary protein identification in large scale. Identification can not be convincing unless proper sampletreatment, control sample setting, and data processing taken. When PMF data were used tosearch against protein databases with different software programs, the searching results arepossibly widely divergent depending on the different algorithm and score value systemssued in different programs, and database used. PMF data alignment should carry out byMASCOT and ProFound, their matching results verify and complement each other. Resultsby MS-Fit and Aldente matching can be only used as reference. Better results can beacquired when use an in-home version of MASCOT or other commercial software tosearch PMF against a local dataset, eg., Gossypium EST dataset. This is a very importantapproach for protein identification in non-sequenced organisms. Nevertheless, tandemMS/MS data-dependent identification is more reliable, which is currently, if not only onepeer-accepted approach. Tandem mass spectra can be searched in ion search, sequencequery, and novo sequencing interpretation modes to identify protein. Their results can beverified each other.Cotton fibers differentiate from a part of ovule epidermal cells, and undergo distinctbut overlapping stages of expansion, secondary wall deposition, and maturationdevelopmental phases in a period of 60 days. The duration, gene and protein expressionpattern of each stage will eventually influence fiber yield, and fiber qualities, such aslength, strength, and fineness. Using the depilated ovules from upland cotton line of TM-1as a control, we compared 2-DE profiles of fiber proteins of 6 DPA, 10 DPA, 14 DPA, 18DPA, and 22 DPA, which represent rapid polar elongation, peak rate of expansion, early,middle and late period of the transition from PCW to SCW synthesis respectively todetect differentially regulated spots among them, and detected 77 differentially-expressedproteins. All of these proteins were digested in gel with trypsin followed by MALDI-TOFmass spectrometry analysis, the resulting PMFs were searched against viridiplantaeprotein database in NCBInr, and Gossypium EST database downloaded from NCBInr inDecember of 2005, respectively, only 20 (26ï¼…) spots were identified with confidence, thislow percentage is likely attributed to relatively meager sequence data for cotton, and highfrequency of post translation modifications occurred during fiber development. Five of the20 identified proteins are unknown protein sequences matched to translated ESTs in silica, suggesting that there are some species-specific genes involved in fiber development. Therest 15 proteins are as follows: a transcription factor containing START domain, amaturase, a translation elongation factor, an Hsp83, a cell wall structural protein, anascorbate peroxidase, two retrotransposons, 6 enzymes as well as a seed storage protein,indicating that transcription regulation (probably for MYB gene), RNA alternativesplicing, new protein synthesis, protein assembly-disassembly induced structural changes,cell wall structural protein synthesis, reactive oxygen scavenge, gene expression regulatedat DNA level induced by transposable element activities, and changes in metabolismsoccur during fiber elongation and cellulose synthesis.The epidermal cells ovule of wild type cotton can develop lint or fuzz fibers, allexcept guard cells have the potential to differentiate into pre-fiber cells, although onlyabout one third of them become fibers in nature. It is difficult to study the lint fiberdifferentiation at -3 or -2 DPA by conventional methods. On the other hand, lint and fuzzfiber initiate successively and distribute miscellaneously, making it hard to separate twotypes of fiber to investigate their development independently. We compared the 2-DE mapsof the proteins isolated from the ovules of Xuzhou 142 wild-type and a fuzzless-lintlessmutant derived from Xuzhou 142 at -3 and 0 DPA respectively to explore key proteinsfunction in fiber differentiation and initiation and underlying molecular mechanism. Wealso compared the 2-DE profiles of the proteins isolated from the ovules of the wild-typeindividuals and lintless mutants segregated of a inbred line of the Ligon lintless mutant (Li1)at -3 and 0 DPA, and proteins from fibers of them at 4 and 8 DPA respectively, with an aimto detect and identify critical factors affecting early fiber elongation. Total 74 differentiallyregulated spots were found, the spots were dig and digested trypically in-gel followed byMALDI-TOF/TOF coupled with bioinformatics identification, 48 spots were identifiedunambiguously, 12 of those were identified exclusively by searching Gossypium ESTdatabase. They can be classified into as follow categories according to their putativefunctions: regulatory protein, transcription factor, protein biosynthesis, protein folding,assembly, proteolysis, defense reaction, energy metabolism, cytoskeleton, retrotransposon,metabolism, and unknown protein. Our results did not verify the hypothesis of themechanism proposed on transcriptomic level, such as MYB genes'roles in fiber differentiation and expansion's roles in fiber initiation, besides the fact that we onlyidentify 48 of 74 differentially expressed proteins, some intrinsic limits residing in thecurrent proteomic technology, such as difficulties in isolation of some proteins andregulatory factors usually expressed in low abundance making them resistant to bevisualized on 2-DE gels, may be responsible for this phenomena. However, we discoveredmany novel mechanisms, such as changes in RNA processing, protein synthesis, processing,isomerization, folding, degradation, localization, and consequent changes in signaltransduction, hormone transportation and responses, enzyme activity, and cytoskeletoncrystallization and substance and energy metabolism, may influence fiber development.These results offer us new clues in finding molecular mechanisms underlying fiberdevelopment. Seven identified proteins were predicted to localize in mitochondrion, two ofwhich are speed-limited enzymes in citrate cycle, one is an electron transporter inrespiration chain, and one peptidase involved in protein processing. Suggesting thatmitochondrial dysfunction must be the important reason for aberrant fiber development inthe two mutants. 6 enzymes are involved in carbohydrate metabolism, they catalyze criticalreactions in pentose phosphate pathway, glycolysis, citrate cycle, and polysaccharidesynthesis. Altered carbohydrate metabolism network is directly link abnormal fiberdevelopment in the mutants. The possible functions of all identified proteins are discussed.
|
PDF Full Text Request