| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Molecular & Cellular Proteomics 6:207-230, 2007.
© 2007 by The American Society for Biochemistry and Molecular Biology, Inc.
,
,**
From the Research Center for Molecular and Developmental Biology, Key Laboratory of Photosynthesis and Environmental Molecular Physiology, Institute of Botany, Chinese Academy of Sciences, Beijing 100093, China, Graduate School of Chinese Academy of Sciences, Beijing 100049, China, ¶ Key Laboratory of Molecular and Developmental Biology, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing 100101, China, and || Beijing Genomics Institute, Chinese Academy of Sciences, Beijing 101300, China
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2300 protein spots revealed 186 that were differentially expressed in mature and germinated pollen. Most showed a changed level of expression, and only 66 appeared to be specific to developmental stages. Furthermore 160 differentially expressed protein spots were identified on mass spectrometry to match 120 diverse protein species. These proteins involve different cellular and metabolic processes with obvious functional skew toward wall metabolism, protein synthesis and degradation, cytoskeleton dynamics, and carbohydrate/energy metabolism. Wall metabolism-related proteins are prominently featured in the differentially expressed proteins and the pollen proteome as compared with rice sporophytic proteomes. Our study also revealed multiple isoforms and differential expression patterns between isoforms of a protein. These results provide novel insights into pollen function specialization.
Recent analyses of mature pollen of Arabidopsis revealed the transcriptome to have reduced complexity and a higher proportion of selectively expressed transcripts than sporophytic tissues (8–10). As well, about one-third of the genes expressed in vegetative tissues are not expressed in the pollen (8). The observation suggests that the transcriptional characteristics involve pollen function specialization. Furthermore these studies determined that pollen transcriptome has a functional skew toward transcripts implicated in cell wall metabolism, signaling, and cytoskeletal dynamics. These results also support the early notion that mature pollen has stored presynthesized mRNAs (11) and present a novel insight into pollen function specialization at the whole genome level.
However, the transcriptomic data of Arabidopsis pollen showed that transcripts related to translation and glycolysis/energy metabolism were underrepresented in pollen (8–9). All transcripts encoding putative or known ribosomal proteins seem to be undetectable in the transcriptome (9). Given the ability of pollen to initiate protein synthesis rapidly upon germination and the carbon skeleton/energy needs for active tube growth, the transcriptome data raise questions, although these results might be explained by the proteins related to the functional categories being presynthesized during the early stage of male gametophytic development (8).
Recent proteomics studies from Arabidopsis (12, 13) and rice (14) have revealed that mature pollen presynthesizes a complement of proteins required for pollen function, including those implicated in protein synthesis and carbohydrate/energy metabolism. Dai et al. (14) showed that these functional categories related to carbohydrate/energy metabolism, wall metabolism, protein synthesis and degradation, and signaling are overrepresented in the proteome. This finding indicated that proteomic data are a necessary complement to transcriptomic data of pollen and are essential to our understanding of pollen function. Importantly proteomics analyses showed that mature pollen contains multiple isoforms; this suggests that posttranslational modifications seem to be crucial to pollen function.
In contrast to our increased understanding of molecular information in mature pollen, our knowledge of genome-scale events underlying germination and fast tube growth are still lacking, although these data are essential for understanding the molecular mechanisms involved in fast tube growth and invasion in pistils. Such knowledge will help dissect the molecular regulation of sexual reproduction and important cellular programs such as polarized cell growth and cell recognition. Recently Fernando (15) compared 2-D1 electrophoresis (2-DE) protein maps of mature and germinated pollen from the gymnosperm Pinus strobus, which produces unlimited quantities of pollen grains, the pollen having in vitro germination activity over an extended period. The author detected 57 protein spots that were specifically expressed or increased in expression in the germinated pollen. However, because of a lack of genomic sequence information about the species, 33% of these spots were not identified.
Arabidopsis and rice have been accepted as the model plants of dicots and monocots of angiosperms, respectively. They have small genome size, and the two genomes have been completely sequenced. Transcriptomic and proteomic data about pollen tubes will greatly increase our knowledge about tube development. Several efforts have established an in vitro germination system for Arabidopsis pollen (16), but collecting enough germinated pollen grains from the plant is difficult. In contrast, we can collect a relatively large amount of mature pollen grains from rice, but a main difficulty in using rice is that the mature pollen quickly loses the ability to germinate under in vitro conditions after being released from anthers (17). Therefore, some technological breakthroughs are required for proteomics and transcriptomics studies of germinated pollen from the two model plants.
Inhibitor experiments have shown that early tube growth strictly depends on protein synthesis that is relatively independent of transcription (11). So proteomics identification of proteins differentially expressed in mature and germinated pollen will generate important molecular information. In the present study, we wanted to address the difference between proteome maps of mature and germinated pollen, which proteins are synthesized during tube growth, and the functional characteristics of newly synthesized proteins. We first established an in vitro germination system of rice pollen, and then used 2-DE followed by MALDI-TOF MS and ESI-Q-TOF MS/MS to identify 160 differentially expressed proteins (representing 120 unique proteins) associated with germination and tube growth. These proteins were implicated in 12 groups of known function and one of unknown function with obvious functional skew toward carbohydrate/energy metabolism, wall metabolism, protein synthesis and degradation, cytoskeleton dynamics, and stress response. Of the differentially expressed unique proteins, 25% had isoforms, and isoforms of some protein species showed distinct changes in expression. Our results suggest that phosphorylation and glycosylation are involved in the generation of these isoforms.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
30 °C) to generate synchronously germinated rice pollen grains (GPGs). The amount of GPGs was examined under a microscope (also see below). After anther debris were removed by filtering the pollen culture through cheesecloth, GPGs were collected by centrifuging at 500 x g. MPGs and GPGs were used immediately for extracting proteins or stored at –80 °C until use.
Observation of Pollen Morphology—
Morphological characteristics of MPGs were examined under a microscope (Axioskop 40 fluorescence microscope, Zeiss) by staining with 1% I2-KI, 1% triphenyltetrazolium chloride in 50% sucrose or 0.2 µg/µl 4',6-diamidino-2-phenylindole (DAPI) (Molecular Probes). The GPGs stained with DAPI or FM 1–14 (0.2 µM; Molecular Probes) were observed by microscopy with a confocal laser scanning microscope (Zeiss) or Axioskop 40 fluorescence microscope.
Preparation of Proteins from MPGs and GPGs—
MPGs and GPGs were homogenized in a homogenate buffer (50 mM Tris-HCl, pH 7.5, 20 mM KCl, 2% Nonidet P-40, one tablet of protease inhibitor mixture/25 ml (Roche Applied Science), 13 mM DTT) with use of a chilled mortar and pestle. Supernatant was collected by centrifugation at 18,000 x g for 20 min at 4 °C and then supplemented with trichloroacetic acid to a final concentration of 12.5% to precipitate proteins on ice for 2 h. Proteins were pelleted by centrifugation at 15,000 x g for 20 min at 4 °C and then resuspended in 80% cold acetone containing 0.07% ß-mercaptoethanol. The mixture was placed at –20 °C for 30 min to allow proteins to precipitate, and proteins were collected by centrifugation at 15,000 x g for 20 min at 4 °C. After being rinsed with cold acetone with 0.07% ß-mercaptoethanol and dried by vacuum, the resulting proteins were dissolved in a lysis buffer (7 M urea, 2 M thiourea, 4% Nonidet P-40, 13 mM DTT, 2% Pharmalyte 3–10) and used for 2-DE immediately or stored in aliquots at –80 °C after debris were removed by centrifugation at 20,000 x g for 20 min at 4 °C. For protein preparation of MPGs or GPGs, triplicate biological samples were used. Protein concentrations were determined according to the Bradford method (18) by DU640 UV-visible spectrophotometry (Beckman). Bovine serum albumin was used as the standard.
2-DE Gel Staining and Image Analysis—
An aliquot (about 600 µg of proteins) of protein sample prepared from MPGs or GPGs was diluted with rehydration buffer (7 M urea, 2 M thiourea, 1% Nonidet P-40, 13 mM DTT, 0.5% IPG buffer 3–10, 0.002% bromphenol blue) and loaded onto an IPG strip holder of a 24-cm, pH 3–10 or pH 4–7 linear gradient IPG strip (Amersham Biosciences). Isoelectric focusing involved the Ettan IPGphor isoelectric focusing system following the protocol of the manufacturer (Amersham Biosciences). For SDS-PAGE, the equilibrated IPG gel strips were placed onto 12.5% ExcelGel SDS gels (Amersham Biosciences) by use of an Ettan DALT Six electrophoresis unit. Low molecular mass (relative molecular mass (MM)) protein markers (Fermentas) were co-electrophoresed as MM standards. The proteins in gels were visualized by Coomassie Brilliant Blue (CBB) staining. 2-DE experiments were repeated three times for each protein sample, and images were obtained by scanning each stained gel at 300 dots/inch with use of an ImageScanner (Amersham Biosciences). The apparent MM of each protein in gels was determined with the co-electrophoresed MM protein markers used as standards. An apparent pI of each protein was determined by its migration on IPG linear strips.
Protein Quantification and Expression Abundance Analysis—
All 2-DE gel images were analyzed by use of the ImageMaster 2D platinum 5.0 program (Amersham Biosciences). After spot detection, quantification, and background subtraction, the relative volume (RV) of each protein spot was obtained by dividing the volume of the spot by the total volume of all spots in a gel. Furthermore to compare protein abundance between two independent datasets of each sample directly (pH 4–7 and pH 3–10), we generated a normalized RV whereby the RV of each spot was multiplied by a correction constant (C). Correction constants for pH 4–7 (CpH4–7) and pH 3–10 (CpH3–10) were calculated according to the formulas described by Hajduch et al. (19).
The normalized RV values of protein spots in triplicate biological repeats for each sample underwent further statistical analysis to evaluate their relative standard deviation (S.D.). These reproducible protein spots were used further to identify proteins differentially expressed in MPGs and GPGs. The significance of differentially expressed proteins between MPGs and GPGs was examined by t test. Proteins considered differentially expressed were those that displayed at least a 2-fold significant change in RV values (p value <0.05) between the two samples.
MALDI-TOF MS—
The differentially expressed protein spots were excised from 2-DE gels by use of the Ettan Spot Cutter (Amersham Biosciences), destained in a destaining buffer (25 mM ammonium bicarbonate, 50% (v/v) acetonitrile), and then dehydrated by use of acetonitrile and spun dry. Enzyme digestions in gels were carried out in 25 mM ammonium bicarbonate buffer containing 10 ng/µl sequencing grade modified trypsin (Roche Applied Science) at 37 °C for 16 h. For MALDI-TOF MS, one aliquot of the enzyme digest solution was spotted onto a sample plate with matrix (
-cyano-4-hydroxcinnamic acid, 8 mg/ml in 50% (v/v) TFA) and allowed to air dry. MALDI-TOF MS acquisition involved use of an Autoflex MALDI mass spectrometer (Bruker Daltonics) equipped with a flight tube (reflex mode, 2.6 m long), laser (N2, 337 nm), and scout 384 target system. Accelerating voltage was 20 kV, and the microchannel plate detector was at 1.6 kV. Mass spectra were acquired in a positive mode. To ensure the accuracy of protein identification, MALDI-TOF MS was internally calibrated with the use of the peptide calibration standard (Bruker) to reach a typical mass measurement accuracy of 100 ppm. The known trypsin-autocleavable peptide masses of 9.07 and 22.73 kDa were used for a two-point internal calibration for each spectrum.
Peptide mass fingerprintings (PMFs) were searched in the National Center for Biotechnology Information non-redundant (NCBInr) protein databases with use of the search engine MASCOT (Matrix Science). O. sativa was chosen for the taxonomic category. All peptide masses were assumed to be monoisotopic and [M + H]+ (protonated molecular ions). Searches involved use of a mass accuracy of ±100 ppm, and one missed cleavage site was allowed for each search. The identified proteins had to be in the top hit with more than four peptides matched and a sequence coverage of more than 10%.
ESI Q-TOF MS/MS—
For nanospray ESI Q-TOF MS/MS (Micromass) analysis, 1–2 µl of each digested peptide sample underwent trapping column filtration for desalting treatment before being loaded in the nanoflow probe tip (Micromass). The instrument accuracy was calibrated by the external calibration of Glu-fibrinogen (3 ppm). The applied spray voltage was 800–1000 V with a sample cone of 25–40 V. The microchannel plate detector was at 2250 V, and the energy-adjustable collision cell was filled with pure argon gas. MS/MS data processing involved use of MassLynx 3.5, and data were searched in the NCBInr protein sequence databases by use of the MS/MS ion search program MASCOT (Matrix Science). The identified proteins had to be in the top hit with more than two peptide sequences matched. Under less-than-optimal circumstances, a matched protein was accepted if it ranked as the top hit with a single peptide match. In this case (only four identities), the MS/MS fragment ion pattern was verified by manual inspection. The manual inspection of MS/MS data processing involved use of the Peptide Sequencing Program with the following strict criteria: 1) MM tolerance was set at 0.3 Da, and mass type was assumed to be monoisotopic; 2) the peak threshold was 0.15%, and the fragment ion tolerance 0.15 Da; and 3) most importantly, nearly complete Y-ion series and partial complementary B-ion series needed to be present, and the Y-ions should correspond to peaks with high relatively intensity.
Western Blot Analysis—
After separation by 12.5% SDS-PAGE, proteins were electrophoretically transferred on a semidry blot apparatus to a PVDF membrane (Pierce) and then immunodetected according to the methods of Dai et al. (14). The primary antibodies used were against maize pollen coat-specific ß-1,4-xylanase (20), pumpkin catalase 1 (21), plasma membrane H+-ATPase PMA2 from Nicotiana plumbaginifolia (22), and rice OsRad21-3 (prepared in our laboratory).
Semiquantitative RT-PCR—
Semiquantitative RT-PCR was used to analyze the accumulation patterns of the following transcripts in anthers, MPGs, and GPGs. Primers Pr1 (5'-ACG CGC TGA CCA AGG AGA TC-3') and Pf1 (5'-CAG GTT GTA GTC GCC GTG TG-3') were used to detect the transcript for ß-1,4-xylanase (GenBank accession number AAP53220), primers Pr2 (5'-ACT TGG CCT GGA CGT TGG AC-3') and Pf2 (5'-ATG GCA TCC TCC TCC CTT CT-3') were used for ß-expansin (NP_91258), primers Pr3 (5'-GCA AGC ATG CAG CAA CAC AT-3') and Pf3 (5'-GCA GCA ATG GCA TCC TCC T-3') were for major pollen allergen Ory s 1 (Q40638), primers Pr4 (5'-TTT GCA TAT GGC TCC ACA GC-3') and Pf4 (5'-ATG GCC TCC ATG TCC TCC TTC-3') were for pollen allergen 1 (CAD40508), primers Pr5 (5'-TCA GAT CAT GTT GGA GAG G-3') and Pf5 (5'-ATG GCG AGA TCA CTG GCG C-3') were for pectin methylesterase inhibitor (NP_912762), and primers Pr6 (5'-ACA GAG CCT GCC ACG ACA GA-3') and Pf6 (5'-GAC CGT TGC CTT CAA TAG CG-3') were for ß-galactosidase (NP_920740). The amplified tubulin tubA cDNA (X91806) was used as a constitutive control (23). First strand cDNA was synthesized with SuperScript II RNase H– reverse transcriptase (200 units/µl, Invitrogen) according to the manufacturer’s protocol. PCR was performed in a mixture of 50 µl that contained 1 µl of first strand cDNA, 10 pmol each of the gene-specific primers, 0.4 mM dNTPs, 1x PCR buffer, and 2.5 units of LA Taq DNA polymerase (5 units/µl, TaKaRa) for 25 cycles.
Detection of Glyco-/Phosphoprotein Candidates—
After separation by 2-DE according to the protocol described under "2-DE Gel Staining and Image Analysis," proteins in the gel were stained with Pro-Q diamond phosphoprotein and Pro-Q Emerald 488 glycoprotein gel stain kits according to the manufacturer’s protocol (Molecular Probes). Images were acquired on a Typhoon 9400 variable mode imager (Amersham Biosciences) with a 532 nm laser excitation and 560-nm bandpass emission filter for Pro-Q diamond-stained proteins and by use of 488 nm laser excitation with a 530-nm bandpass emission filter for Pro-Q Emerald 488-stained proteins. Finally the gel was stained by CBB. Unequivocal identification of glyco- and/or phosphoproteins was by grayscale adjustment of images depending on the co-electrophoresed "PeppermintStick" phosphoprotein and "CandyCane" molecular mass standards (Molecular Probes).
Bioinformatics Analysis—
Protein functional domains were predicted by use of the PHI and PSI-BLAST programs (www.ncbi.nlm.nih.gov/BLAST/). Signal peptides and subcellular localization information were obtained by use of SignalP and PSORT. Chromosome loci of protein-coding genes were detected at mpss.udel.edu/rice/. Amino acid sequence identity/similarity analysis was conducted by use of DNAman software and the BLAST program (www.ncbi.nih.gov/blast/).
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
30 °C). The collected MPGs showed an active metabolism (Fig. 1, A and B). After germination, anther debris were removed by filtering the culture, and GPGs were collected by centrifugation. Purity examination by microscopy revealed no contamination by other tissues in the filtered culture (Fig. 1C). This procedure gave rise to a germination ratio of more than 80% (Fig. 1C). The germinating pollen can transport an entire germ unit into the pollen tube (Fig. 1, D and E).
|
|
|
Protein Expression Profiles of GPGs—
To identify differentially expressed proteins associated with germination and early tube growth, we analyzed protein expression profiles of MPGs and GPGs by 2-DE and CBB staining. Originally we separated the two distinct protein samples by 2-DE on pH 3–10 gel strips, performed in at least triplicate biological repeats to ensure the reproducibility of protein patterns. The experiments revealed 559 ± 10 spots (n = 3) for MPGs (Fig. 4A) and 544 ± 11 (n = 3) for GPGs (Fig. 4B). Most of the spots were distributed around pH 4–7 with 413 ± 7 for MPGs and 404 ± 11 for GPGs and only about 26% around pH 7–10 (146 ± 6 for MPGs and 140 ± 5 for GPGs). The protein spots around pH 7–10 were well separated with 
90% well matched spots in triplicate biological repeats of MPGs and GPGs. Therefore we identified differentially expressed proteins from these well matched spots in the pH 7–10 range of the two distinct gels by comparison and quantification. Statistical analysis revealed five RV-increased, 10 RV-reduced (p value <0.05), eight MPG-specific, and 17 GPG-specific spots (Fig. 4 and Tables I and II).
|
|
|
|
|
In this study, we tried to detect these released proteins from germination medium but were unsuccessful mainly because of the high concentration of sucrose and polyethylene glycol in the germination medium interfering with the separation by one-dimensional electrophoresis, 2-DE, or liquid chromatography (data not shown). However, 21 of the 66 identified RV-reduced/disappearing protein spots (Table II), such as ß-1,4-xylanase, pollen allergens, ß-expansin, polygalacturonase, profilin A, enolase, and ascorbate peroxidase, have been detected in coat/wall-associated and pollen-released protein fractions of mature rice (14) and maize pollen (20). Further analyses revealed that 1) most of the identified RV-reduced/disappearing proteins, such as members of the class III peroxidase family, polygalacturonase, and disulfide isomerase, have a potential extracellular targeting signal peptide in their N termini (Table II), and 2) 24 have been documented to be secreted/released extracellular proteins and/or cell surface/wall-associated proteins in other organisms, for example 2,3-bisphosphoglycerate-independent phosphoglycerate mutase (26), fructose-1,6-bisphosphate aldolase (27), and triose-phosphate isomerase (28). Thus, we propose that 51 of the 66 RV-reduced/disappearing identities (39 Unipros) were probably released, and the other 15 (14 Unipros) may really be down-regulated (Table II).
Functional Categories of These Differentially Expressed Identities—
Among the 160 differentially expressed identities, 128 (80%) have been deposited in the current database as putative functional proteins. We further examined their annotated function by domain searching and similarity comparison. The remaining 32 (20%) were deposited in the database as unknown or hypothetical proteins. Domain searches and similarity analyses revealed that 20 of the unknown proteins contained conserved entire domains associated with known activities and/or showed high sequence similarity with known functional proteins in the database. The other 12, with no known conserved domains or sequence similarity with known functional proteins, may represent a group of novel proteins.
From the above analyses, in combination with metabolic and functional features of pollen, we grouped all the identities into 13 major categories (Fig. 7A). An impressive 74% of these identities were implicated in five functional groups, including carbohydrate/energy metabolism (24%), wall remodeling and metabolism (24%), protein metabolism (11%), cytoskeleton dynamics (8%), and stress response (7%) (Fig. 7A). Analysis of their relative expression level in each category revealed an identical trend (Fig. 7B); these cellular/metabolic process-related proteins were overrepresented either in number or expression level of the identified differentially expressed proteins, suggesting the functional importance of these processes in pollen tube growth.
|
-1,4-glucan phosphorylase H and sucrose-6F-phosphate phosphohydrolase), glycolysis, the tricarboxylic acid cycle, and the pentose phosphate pathway (Table I). These up-regulated proteins also included ATP generation-related proteins such as ATP synthase F0 subunit 1, ATP synthase, NADPH-thioredoxin reductase, and cytochrome oxidase c subunit 6b (Table I), and basic energy supply-related soluble inorganic pyrophosphatase (five isoforms: spots -/G189, -/G42, -/G394, -/G39, and -/G103) (Table I). Therefore, these data demonstrate that active carbohydrate/energy metabolism is a basis of fast pollen tube growth. Unexpectedly 10 carbohydrate metabolism-related identities were detected as RV-reduced/disappearing protein spots in the germinated pollen (Table II). They represented eight unique proteins, including enolase, triose-phosphate isomerase, phosphoglycerate kinase, phosphoglycerate mutase, fructose-1,6-bisphosphate aldolase, fructokinase I, ß-phosphoglucomutase, and 6-phosphogluconolactonase. The first five are found in the cell wall of yeast or pathogens (26–30). Our proteomics analysis of MPGs also revealed that all but phosphoglycerate kinase were present in the pollen-released protein fraction (14). Therefore, their decrease in level or disappearance in the germinated pollen should result from their release. Why are these carbohydrate metabolism-related proteins releasable? One possibility is that these proteins may have a role in a related metabolism on the pollen/stigma surface or they may be deposited onto the pollen coat/wall from degraded tapetal cells (20).
Germinated Pollen Seems to Display Active Wall Dynamics Both in Wall Synthesis and Degradation—
The differentially expressed wall-related proteins (39 identities, 24% total) could be divided into two distinct function groups. The first group consists of 16 up-regulated identities (Table I) seeming to function in wall synthesis and hydrolyzing/loosening with seven (representing five enzymes), including UDP-glucose pyrophosphorylase, UDP-glucose dehydrogenase, dTDP-4-dehydrorhamnose reductase, reversibly glycosylated polypeptide with three isoforms (spots P902/G874, P95/G35, and -/G119), and myo-inositol-1-phosphate synthase, involved in the interconversion and transportation of sugars required for cell wall biosynthesis. Another nine, including multifunctional class III peroxidase (POD) 79, POD 78 with two isoforms (spots P444/G245 and P702/G699), POD 31 with two isoforms (spots P605/G326 and P194/G45), polygalacturonase Os02g10300 with two isoforms (spots -/g242 and -/g281), and expansin with two sequence-related isoforms (Os10g40090, spot -/g93; Os03g01640, spot -/g398), may be responsible for wall loosening and hydrolyzing. Class III PODs have been documented to function in promoting elongation of Arabidopsis roots (31, 32) or physically inhibiting the elongation process by creating cell wall cross-linking (33). Polygalacturonase is a key enzyme-degrading pectin, and expansins are involved in cell elongation growth. The up-regulation of these proteins in germinated pollen suggests that they may be necessary for fast polar tube growth through regulating synthesis and wall loosening.
The second group involves 23 possibly pollen-released identities (16 Unipros) (Table II) such as polygalacturonase Os06g35320 with three isoforms (spots p952/g433, p565/g148, and p621/-), ß-1,4-xylanase, 12-kDa ß-expansin-like (spots P18/G86, P3/G17, P11/G34, and P17/G116), pollen allergen Phl p 11, pollen-specific protein C13 with two isoforms (spots P150/- and P47/G269), 35-kDa ß-expansin, and ß-expansin OsEXPB13 with two isoforms (spots P29/- and P9/G22). Secretory activity of polygalacturonase has been detected in pollen of many species (1, 34). ß-1,4-Xylanase is one of the most abundant pollen coat proteins in maize and functions in hydrolyzing the stigma and track cell wall (20). Pollen allergen Phl p 11 contains the pollen Ole e I domain (pfam01190), whereas Ole e 1 has been localized in the pollen wall and can be released into the culture medium during in vitro germination (35). Several members of the ß-expansin subfamily in grass pollen are proposed to facilitate pollen tube growth by loosening the cell wall in pistils (36). This group also contained one isoform of POD 31 (spot P15/G273) and two isoforms of POD 36 (spots P1/G133 and p594/g126). Most proteins of the group have been documented to have roles in hydrolyzing walls. Given that they are possibly released in the germinated pollen, these proteins may involve hydrolysis and/or loosening of walls of the stigma and tract cells in pistils.
Most of the Differentially Expressed Proteins in the Protein Metabolism Category Involve Protein Degradation—
Among 17 identities implicated in protein translation, assembly, and degradation, besides disulfide isomerase and translationally controlled tumor protein (TCTP), which were decreased in level, others were increased in level in the germinated pollen (Tables I and II). Proteins eukaryotic initiation factors 4A (eIF4A) and 4G (eIF4G), involved in protein translation regulation at the level of ribosome recruitment, showed a 2.7- and 12.9-fold increase, respectively. eIF4A, a member of the eIF4F complex, has ATP-dependent RNA helicase activity. eIF4G, another member of the complex, is a large modular protein that serves as a docking site of translation-related proteins (37). TCTP is the key inhibitor of translation elongation; it inhibits guanine nucleotide dissociation from protein elongation factor 1A (EF1A) (38). Up-regulation of eIF4A and eIF4G and down-regulation of TCTP, together with up-regulation of molecular chaperones and amino acid metabolism-associated proteins (Table I), suggest greatly increased translational efficiency in growing pollen tubes; this is consistent with the notion that active protein synthesis is required for fast pollen tube growth (9).
Interestingly 10 of the 17 protein metabolism-related identities were implicated in protein degradation with an increased level in the germinated pollen. These included six subunit components of the 14-subunit 20 S core complex of the 26 S proteasome as well as peptidase M3, peptidase S8, subtilisin-like proteinase, and leucine aminopeptidase (Table I). This finding suggests the importance of protein turn-over in pollen tube growth.
Change in Level of Actins and Tubulins during Germination Was Distinct—
Our analysis revealed 13 differentially expressed identities involved in cytoskeletal dynamics. All seven identities representing actins, including Os03g50890 with four isoforms, Os11g06390 with two isoforms, and Os03g61970, were up-regulated or newly appearing (Table I). Previous studies have revealed that subapical axial actin cables in pollen tubes act to target the vesicle to the tip and/or maintain proper cytoplasmic organization of the tube; thus they are essential for polar pollen tube growth (3). These results suggest that actin cables and dynamics in the growing tube may depend on new synthesis of actins.
In contrast, except for the up-regulated -tubulin (Os03g51600) (Table I), the other identified tubulin proteins, including three ß-tubulins (Os06g46000, Os05g34170, and Os03g45920) and one -tubulin (Os07g38730), were decreased in level (Table II). The functions of microtubulins (MTs) in pollen tubes are not fully understood. Several studies showed that MTs had roles in transporting the male germ unit and maintaining generative cell shape (39). But depolymerization of MTs seems not to affect the initial rate of the pollen tube growth in vitro (1). Although evidence is lacking to explain the importance of down-regulated tubulins in polarly growing pollen tubes, their reduced level as compared with that of actin proteins suggests that the two types of cytoskeletal proteins may have different functions in tube growth.
The Presence of Multiple Isoforms—
It is generally accepted that multiple isoforms result from sequence-related proteins encoded by distinct genes and/or polypeptide variants encoded by the same gene (splice variants and/or posttranslational modifications (PTMs)). Our results show that 25% (n = 30) of the differentially expressed Unipros have isoforms: six are sequence-related (actin, Os03g61970, Os11g06390, and Os03g50890; ß-tubulin, Os06g46000, Os05g34170 and Os03g45920; -tubulin, Os07g38730 and Os03g51600; voltage-dependent anion channel protein (VDAC), Os06g45210 and Os01g51770; polygalacturonase, Os02g10300 and Os06g35320; and ß-expansin, Os10g40090 and Os03g01640) (Tables I and II), and the other 24 (Fig. 8 and Table III) appear to have PTM/splice variant-generated isoforms.
|
|
10%) (spots -/G601, -/G89, P240/G47, -/g344, -/G121, -/G184, -/G122, P902/G874, P194/G45, P605/G326, and P702/G699, representing seven Unipros) (Table III). However, a data search in The Institute for Genomic Research (TIGR) rice database (www.tigr.org/tdb/e2k/osal/locusNameSearch.shtml) demonstrated that each of the seven genes seemed to give rise to multiple alternative splice variants, some appearing to be close to these identities in MM. Therefore, these isoforms could have resulted from splice variants rather than from proteolytic degradation. This conclusion was also evidenced by the reproducibility of all the differentially expressed identities in triplicate biological repeats and the Western blot experiments with four marker proteins, which have different subcellular localization (20–22, 25), that did not detect protein fragments possibly resulting from proteolytic degradation in the MPG and GPG protein samples (Fig. 2). Taken together, these data demonstrate the presence of these isoforms in MPGs and GPGs. To evaluate the kinds of modifications related to the occurrence of these isoforms during pollen germination, we applied two newly developed specific staining methods (see "Detection of Glyco-/Phosphoprotein Candidates"), which are efficient to predict phosphorylated and glycosylated proteins with 2-DE gels (41), to visualize 2-DE gels of MPG and GPG proteins. A total of 14 protein spots showed positive reactions to glycoprotein detection reagent Pro-QTM Emerald 488, four reacted to phosphoprotein detection reagent Pro-Q diamond, and one reacted to both reagents (Supplemental Fig. S2, A, B, and C, and Table III). The 19 spots represent 13 of 24 Unipros with possible PTM/splice variant-derived isoforms (Table III). In fact, 17 of the 24 protein species are glycosylated and/or phosphorylated in other organisms (Table III), and seven of the 17 have modifications identical to our results (Table III). Furthermore of the identified possible glycoproteins (Supplemental Fig. S2A and Table III), class III peroxidase has been known as a heme-containing glycoprotein (42). Polygalacturonase, pollen-specific protein C13 (Os09g39950), ß-expansin OsEXPB13, and ß-expansin have been well documented as pollen allergens (43–45). Previous investigations showed that most pollen allergens were glycoproteins, whereas glycan probably contributes to their allergenic activity (46). And Os09g39950 contains the function domain pfam01190 of the Ole e I family, and its homolog Ole e10 from olive pollen tubes has been detected to bind 1,3-ß-glucans preferentially (45). In addition, among these detected phosphoproteins (Supplemental Fig. S2B and Table III), soluble inorganic pyrophosphatase has been found to be phosphorylated in the pollen tubes of Papaver rhoeas (47). These data suggest that at least glycosylation and/or phosphorylation is involved in the generation of isoforms.
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Recent proteomics analyses from Arabidopsis (12, 13) and rice (14) suggest that mature pollen has presynthesized a complement of proteins required for germination and early tube growth. Importantly mature pollen has stored translation-related proteins, including ribosome proteins (12, 14). Together transcriptomic and proteomic data showed that mature pollen has presynthesized these mRNA and proteins to prepare for quick pollen germination and polar pollen tube growth.
In the present study, we screened proteins expressed differentially in MPGs and GPGs by use of 2-DE and MS. 2-DE revealed 1004 ± 22 spots for MPGs and 1030 ± 45 spots for GPGs (pH 4–7 range) with high reproducibility. Of about 2300 protein spots (pH 4–7 range and pH 4–10 part of the pH 3–10 range) in the two samples, 120 spots showed changes in level, whereas only 66 appeared to be sample-specific, suggesting that the changes in development and metabolism were not associated with, at least for the high and medium abundance proteins, appearance or disappearance of proteins. Relatively identical protein patterns were also observed in mature pollen and pollen tubes of Gymnospermae pine (15). Together with the proteomic data of mature pollen of Arabidopsis and rice (12–14), our data further indicate that pollen germination and early tube growth depend mainly on these presynthesized proteins in mature pollen.
Preferential Representation of Wall-related Proteins in Pollen Proteome Is an Important Property of Pollen Function Specialization—
Although recent transcriptome studies of Arabidopsis pollen have not resulted in similar conclusions about the distribution of pollen-expressed transcripts in functional categories (8–10, 49, 50), overrepresented transcripts in the transcriptome are generally considered to be implicated in wall metabolism, cytoskeleton, and signaling (8, 9, 49). Our proteomics analysis revealed that these proteins are involved in not only these three functional categories but also carbohydrate/energy metabolism, protein synthesis and degradation, and stress response (14).
To evaluate the proteomic difference between pollen and sporophytic tissues, we compared proteomic data from mature rice pollen (14) and rice sporophytic tissues, including leaves (811 proteins), roots (1051 proteins), and seeds (702 proteins) identified by Koller et al. (51). Comparison of 10 function categories (Supplemental Table S2) revealed that the most important difference between pollen and sporophytic proteomes is the high representation of wall-related proteins in the pollen proteome (11%) (14) but the lowest representation in the sporophytic proteomes (0.25% in leaves, 0.67% in roots, and 0.28% in seeds) (51). Thus, highly active wall dynamics is a prominent property of pollen function specialization.
Just like the mature pollen proteome, the differentially expressed proteins showed a functional skew toward wall and carbohydrate/energy metabolism (Fig. 7A). The wall metabolism-related proteins displayed two distinct changed patterns during pollen germination. These up-regulated proteins mainly involve wall synthesis (Table I) and wall hydrolysis/loosening enzymes, such as UDP-glucose pyrophosphorylase, reversibly glycosylated polypeptide, polygalacturonase Os02g10300, class III peroxidases, and ß-expansins Os10g40090 and Os03g01640 (Table I). Our previous study revealed that mature rice pollen contains all the enzymes for pectin degradation (14), which suggests that pectin may be important for rice pollen function. Recent evidence has suggested that alternative activity changes in polygalacturonase, as well as pectin methylesterase, mediated by apical proton gradient are required for polar growth of the tube (7). Polygalacturonase in Aspergillus niger has a role in stimulating pollen germination and tube growth (52). Therefore, these proteins, up-regulated on germination, might be mainly responsible for fast pollen tube growth. In contrast, down-regulated wall-related proteins such as ß-1,4-xylanase, polygalacturonase Os06g35320, ß-expansin OsEXPB13, and one isoform of ß-expansin Os10g40090 were mainly involved in wall degradation and loosening (Table II). The down-regulation of most of these proteins should result from their release from the germinating pollen into the medium (Table II), which is possibly similar to their situation in vivo. In the monocots maize and rice, ß-1,4-xylanase is the major protein component in the pollen coat (14, 53) and possibly facilitates pollen tube invasion by hydrolyzing the xylan on the stigma surface (53). Besides being responsible for degradation of demethylesterified pectin (7), polygalacturonase-mediated release of oligogalacturonides can act as a signal for different development processes (54). The in vivo release of these proteins might be necessary for pollen tube invasion growth in pistils.
Selective Degradation of Proteins May Be Vital for Fast Pollen Tube Growth—
Mature rice pollen has presynthesized a complement of translation-related proteins with multiple molecular chaperones (14). After germination, TCTP, an effective inhibitor of translation elongation (38), was down-regulated (Table II), whereas eIF4A and eIF4G, involved in protein translation regulation at the level of ribosome recruitment, were greatly up-regulated (Table I). This finding suggests that active protein synthesis is initiated in growing tubes.
Besides the regulation of protein synthesis, the temporal regulation of protein function is vital to cellular and developmental processes as much evidence has suggested. 26 S proteasome-based selective protein degradation is an important mechanism for temporal regulation and involves numerous cellular and development processes, including pollen-stigma interaction (4) and polarized cell morphogenesis (55). The disruption of pollen germination by inhibitors of 26 S proteasome activity (56) and presynthesis of multiple subunit components of the complex in mature rice pollen (14) suggest that the pathway is implicated in regulation of pollen function. Furthermore our differential proteomics analysis identified six of 14 subunits of the 20 S proteolytic particle of the proteasome, all of which were up-regulated in growing tubes (Table I) as were some peptidases. Therefore, 26 S proteasome-mediated protein degradation might be an important regulator of pollen function.
The Presence of Multiple Isoforms Underlines the Importance of Their Function in Pollen Germination and Tube Growth—
Besides closely related sequences from different genes and splice variants from a gene, more than 200 PTMs contribute to the generation of isoforms (57). Increasing data from the studies of animal cells show that isoforms have different subcellular localization (58, 59) and/or functions (57). Therefore, isoforms are generally considered to diversify the function of a protein. Recent proteomics studies have revealed that isoforms are common in analyzed seeds (60), anthers (61), and mature pollen (12–14). In our identified Unipros from mature rice pollen, 23% have isoforms (14) with wide distribution in different functional categories (Fig. 9B). A similar proportion was found in the identified proteins from mature Arabidopsis pollen (13), implying the functional importance of isoforms.
|
-tubulin, ß-tubulin, polygalacturonase, ß-expansin, and VDAC, have sequence-related isoforms (Fig. 8 and Tables I, II, and III). In animal sperm cells, isoforms of tubulin participate in different microtubulin structures, thus possibly having different functions (62). Arabidopsis pollen-expressed pectin methylesterase isoforms appear to have different functions for pollen germination and pollen tube growth (7). Our data show that the sequence-related isoforms had distinct expression patterns in rice pollen (Tables I, II, and III) during the developmental switch from mature pollen to germination and tube growth. For example, polygalacturonase Os02g10300 newly appeared, whereas Os06g35320 was probably released in the growing tubes (Table III). Furthermore the 24 protein species giving rise to isoforms possibly by PTMs/splice variants appeared as 59 identities (Fig. 8 and Table III); 37 of these were up-regulated, and 22 were down-regulated or possibly released during pollen germination and tube growth. Interestingly for proteins such as ß-expansin and class III peroxidase 31, some isoforms were up-regulated, and others were down-regulated (Table III), suggesting their different function. In addition, our results preliminarily demonstrate that at least glycosylation and/or phosphorylation is involved in the generation of some isoforms. The importance of these predicted modified proteins in pollen tube growth remains largely unknown, but some evidence indicates that glycosylation is required for subcellular localization and transportation of proteins (63). This process might be important for wall loosening/hydrolysis-related proteins such as polygalacturonase and ß-expansin, which have roles in facilitating pollen tube invasion growth in pistils by localizing around the tube wall and/or being released into tract tissues of the pistil. Soluble inorganic pyrophosphatase is involved in the generation of ATP and synthesis of biopolymers; the latter are material for cell membranes and walls. In the self-incompatibility response of Papaver rhoeas, Ca2+ gradient-mediated hyperphosphorylation of this protein in the apical pollen tube was proposed to be associated with the inactivity or down-regulated activity of the enzyme, consequently resulting in inhibition of pollen tube growth (47). Our results showed that this protein has multiple up-regulated isoforms and that one of them is phosphorylated in the germinated rice pollen, suggesting a strict functional regulation of this protein. All this evidence demonstrates the importance of isoforms in pollen germination and tube growth, although direct information about function is still lacking.
In summary, we have established an in vitro germination method for mature rice pollen and screened 186 protein spots differentially expressed in mature and germinated pollen by 2-DE-based differential proteomics approaches. Most of the proteins showed differences in level between the two proteomes, and only 66 were specific to development stage. Furthermore 160 differentially expressed protein spots were identified on MS to match with 120 diverse protein species. These differentially expressed proteins showed an obvious functional skew, like those in the mature pollen proteome (14). Wall metabolism-related proteins are predominant in the pollen proteome as compared with rice sporophytic proteomes (51) (Supplemental Table S2). The study has also revealed multiple isoforms in the differentially expressed proteins and different expression patterns between isoforms of a protein, a finding not achieved with transcriptomics approaches. These results have provided novel insights into pollen function specialization important for understanding the molecular regulation of polar tube growth.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Published, October 10, 2006
Published, MCP Papers in Press, November 27, 2006, DOI 10.1074/mcp.M600146-MCP200
1 The abbreviations used are: 2-D, two-dimensional; 2-DE, 2-D electrophoresis; MPG, mature rice pollen grain; GPG, germinated rice pollen grain; DAPI, 4',6-diamidino-2-phenylindole; MM, molecular mass; CBB, Coomassie Brilliant Blue; RV, relative volume; PMF, peptide mass fingerprinting; Unipros, unique proteins; POD, peroxidase; TCTP, translationally controlled tumor protein; MT, microtubulin; PTM, posttranslational modification; VDAC, voltage-dependent anion channel protein; eIF, eukaryotic initiation factor. 
* This work was supported by National Science Foundation of China Grants 30570932 and 30370138, Chinese Ministry of Sciences and Technology Grants 2005CB120802 and 2005CB120804, Chinese Academy of Sciences Grant KSCX-SW-307, and an innovation grant from the Institute of Botany, Chinese Academy of Sciences. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
S The on-line version of this article (available at http://www.mcponline.org) contains supplemental material.
** To whom correspondence should be addressed: Key Laboratory of Photosynthesis and Environmental Molecular Physiology, Institute of Botany, Chinese Academy of Sciences, 20 Nanxincun, Xiangshan, Haidianqu, Beijing 100093, China. Tel.: 86-10-62836210; Fax: 86-10-62594170; E-mail: twang{at}ibcas.ac.cn
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
