Jasmonates (JAs) are a new class of plant hormone that regulate expression of diverse genes to mediate various plant responses. The Arabidopsis F-box protein COI1 is required for plant defense and male fertility in JA signal pathway. To further investigate the regulatory role of COI1 in male fertility, we compared the proteomics profiles of Arabidopsis wild type (WT) flowers with coi1-1 mutant male-sterile flowers using two-dimensional difference gel electrophoresis coupled with matrix-assisted laser desoption/ionization-time-of-flight mass spectrometry. Sixteen proteins with potential function in specific biological processes such as metabolism processes and defense/stress responses were differentially expressed in WT and coi1-1 mutant flowers. Verification on a phi class glutathione transferase AtGSTF9, one out of these 16 identified proteins, revealed that the expression of AtGSTF9 was severely downregulated in flowers of coi1-1 mutant compared with that of WT. Further function analyses of these genes would provide new insights into the molecular basis of COI1-regulated male fertility.
Recently, two-dimensional difference gel electrophoresis (2-D DIGE), a new emerging technology that was build on two-dimensional gel electrophoresis (2-DE) by adding a highly accurate quantitative dimension, was developed for proteomics analysis. It has been successfully carried out to identify proteins regulated by UV-B (Casati et al. 2005), salt and osmotic stresses (Ndimba et al. 2005), cold stress (Amme et al. 2006), fungal elicitors (Chivasa et al. 2006), gibberellins (Komatsu et al. 2006), and brassinosteroid (Deng et al. 2007; Tang et al. 2008) in many plant species. In the present study, we used 2-D DIGE and matrix-assisted laser desoption/ionization-time-of-flight (MALDI-TOF) mass spectrometry to compare the proteomics profiles of Arabidopsis WT flowers with coi1-1 mutant male-sterile flowers. A total of 16 differentially expressed proteins have been identified. Of these, 12 proteins were upregulated while four were downregulated in the flowers of WT compared with that of coi1-1 mutant. These proteins were predicted to play potential roles in specific biological processes, such as metabolism processes and defense/stress responses.
Results and Discussion
Analysis of proteomics changes between WT and coi1-1 mutant flowers
Similar to previous observations (Feys et al. 1994; Xie et al. 1998; Xu et al. 2002), we found that all of the flowers in the coi1-1 mutant could not develop normally to set seeds (Figure 1A, B). Closer examination of WT and coi1-1 mutant flowers showed that the filaments of coi1-1 mutant could not elongate sufficiently to position the anther onto the stigmatic papilla (Figure 1C). Meanwhile, the anther of the coi1-1 mutant could not dehisce at the time of flower opening (Figure 1C).
To examine the regulatory role of COI1 in male fertility, we harvested flowers (Figure 1C) from both 5-week-old WT and coi1-1 mutant plants for protein extraction and 2-D DIGE analysis. Comparing the protein profiles between WT and coi1-1 mutant showed that the abundance changes of 33 protein spots were significant (at least 50% increase or decrease, P < 0.05). Peptide mass fingerprinting via MALDI-TOF mass spectrometry on these 33 protein spots successfully generated protein assignments for 23 spots, which represented 16 unique proteins (Figure 2 and Table 1). Among these, 11 spots represented 11 unique proteins; the remaining 12 spots represented five proteins. More than two spots represented a single protein in some cases (Figure 2 and Table 1), which probably resulted from the posttranslational modifications that shifted the mobility in 2-DE. The discrepancy between the experimental and theoretical molecular weight (MW) and isoelectric point (pI) of some proteins (Table 1) might be caused by changes in posttranslational modifications, proteolytic degradation of polypeptides, or variability arising from alternate splicing of mRNAs. In WT flowers, 12 out of these 16 proteins were upregulated, whereas four were downregulated compared with coi1-1 mutant male sterile flowers.
Table 1. COI1-regulated proteins in Arabidopsis flowers
No. matching peptidesc
Observed MW (kDa)
Theoretical MW (kDa)
16 COI1-regulated proteins in Arabidopsis flowers were classified based on the biological process in which the protein participated and their corresponding spots were shown on Figure 2.
aCOI1-regulated genes identified previously in Arabidopsis flowers by transcriptomics analysis. bAverage volume ratio quantified by DeCyder 6.5 software. cNumber of peptide masses matching the top hit from MS-Fit peptide mass fingerprinting (PMF). dAmino acid sequence coverage for the identified proteins. MW, molecular weight; pI, isoelectric point.
These 16 COI1-regulated proteins were grouped following the gene ontology (GO) categories based on: (i) the biological process (BP) in which the protein participates; (ii) the cellular component (CC) where the protein can be found; and (iii) the molecular function (MF) that describes the protein activities at molecular level. Protein distributions according to the GO rules are shown in Figure 3.
Proteins involved in metabolism processes
In the present study, we identified seven COI1-regulated proteins that were involved in a variety of metabolism processes. Of these, five proteins (At5g26000, At5g25980, At4g08870, At5g24420 and At2g39990) were more strongly expressed in WT flowers, whereas two proteins (At3g13470 and At2g05990) were more strongly expressed in coi1-1 mutant male sterile flowers. At5g26000 and At5g25980 encode TGG1 and TGG2, respectively, two known functional myrosinase enzymes involved in the glucosinolate catabolic process. TGG1 and TGG2 both catalyze the hydrolysis of glucosinolates into compounds that are toxic to various microbes and herbivores (Xue et al. 1995). At4g08870 encodes an arginase that functions in the polyamine metabolic process. At5g24420 encodes a glucosamine/galactosamine-6-phosphate isomerase, which is known to catalyze the reaction in pentose-phosphate shunt of carbohydrate metabolic process. At2g39990 encodes a translation initiation factor eIF2 p47 subunit homolog involved in the protein metabolism process. At3g13470 encodes a chaperonin, which also functions in one branch of the protein metabolism process: protein folding. At2g05990 encodes a enoyl-ACP reductase that acts as a component of the fatty acid synthase complex (Mou et al. 2000). Anther development has been shown to be involved in an active stage of protein and lipid metabolism for pollen wall formation (Wang et al. 2005b). These proteins with predicted roles in carbohydrate, protein and lipid metabolism may function in pollen wall formation or other aspects in stamen development.
Proteins involved in defense/stress responses
In the present study, we found nine COI1-regulated proteins that functioned in different kinds of defense/stress responses. Of these, the expression of seven proteins (At5g24780, At5g24770, At3g16470, At1g52400, At1g52040, At1g19570 and At2g30860) was induced, whereas the expression of two proteins (At1g04410 and At5g01410) was repressed in WT flowers compared with coi1-1 mutant male sterile flowers. At5g24780 and At5g24770 encode two vegetative storage proteins VSP1 and VSP2, respectively, and their expression is induced by wounding and JA (Berger et al. 1995). At3g16470 also encodes a JA inducible protein isolog, which functions in plant responses to abiotic stimulus including cold, wounding and salt (Leon et al. 1998; Bae et al. 2003; Jiang et al. 2007). At1g52400, which encodes a member of the glycosyl hydrolase family 1, is required in wound-inducible endoplasmic reticulum (ER) body formation and defense response to fungus (Xu et al. 2004). At1g52040 encodes a myrosinase-binding protein MBP1 that functions in plant defense response. At1g19570 encodes a dehydroascorbate reductase that plays an important role in plant ozone tolerance and response to symbiotic fungus (Yoshida et al. 2006; Vadasserya et al. 2009). At2g30860 encodes AtGSTF9, a member of glutathione transferases belonging to the phi class, which is important in plant defense responses (Wagner et al. 2002). At1g04410 encodes a malate dehydrogenase involved in salt stress and cadmium exposure response (Sarry et al. 2006; Jiang et al. 2007). At5g01410 encodes a pyridoxine biosynthesis protein (PDX) 1, which is essential for the biosynthesis of vitamin B6, development and stress tolerance in Arabidopsis (Titiz et al. 2006). It has been reported that anther development was an environment-sensitive process (Dix et al. 1996). The corresponding stress responses of these proteins might correlate with the process of anther development.
Based on the above analyses, most of these genes encode enzymes that potentially mediate specific cellular and physiological processes and therefore function as the downstream targets of COI1. Primary COI1-regulated genes such as transcription factors were not detected in this study, which might be caused by less sensitivity of 2-D DIGE in low-abundance protein identification. To detect primary COI1-regulated proteins, protein enrichment should be essential.
Comparative analysis of proteomics and transcriptomics data
Changes at the RNA level might not necessarily lead to similar changes at the protein level (Huber et al. 2004). Gene expression is also regulated by translational or posttranslational processes (Griffin et al. 2002; Tian et al. 2004). By comparison of our proteomics data with previous transcriptomics data (Capella et al. 2001; Devoto et al. 2005), we found that seven of these 16 COI1-regulated proteins were identified as COI1-regulated RNAs in Arabidopsis flowers (Table 1). Six genes, including At5g26000, At4g08870, At5g24770, At3g16470, At1g52040 and At2g30860, were upregulated in WT flowers compared with coi1-1 mutant male sterile flowers at the transcriptional level (Capella et al. 2001; Devoto et al. 2005), which correlated well with our proteomics data (Figure 2 and Table 1). However, the transcriptional pattern of the remaining gene, At2g05990, was not consistent with its protein expression pattern. The transcripts of At2g05990 were upregulated (Devoto et al. 2005), whereas its protein was downregulated in WT flowers compared with coi1-1 mutant male sterile flowers (Figure 2 and Table 1), which might result from posttranscriptional processes.
Confirmation of 2-D DIGE data
To verify the 2-D DIGE data, we used a western blot approach to further examine the expression pattern of phi class glutathione transferase AtGSTF9, one out of the 16 COI1-regulated proteins. In our proteomics study, we found that the expression level of AtGSTF9 protein was severely reduced in the flowers of coi1-1 mutant compared with that in WT (Table 1). Consistent with our 2-D DIGE results, western blot analysis also showed that the expression of AtGSTF9 protein was greatly affected in coi1-1 male sterile flowers (Figure 4), which demonstrates that 2-D DIGE analysis yields relatively accurate measurement of protein expression. Molecular and genetic approaches will be used to examine whether AtGSTF9 has a possible role in COI1-regulated male fertility.
In the present study, we identified 16 COI1-regulated proteins by comparing the proteomics profiles of WT with coi1-1 mutant flowers. This study is the first application of proteomics to the study of COI1 function in JA-regulated male fertility. These proteins are predicted to function in specific biological processes such as metabolism processes and defense/stress responses. Further characterization of these proteins might contribute to our understanding of the molecular mechanism of COI1-regulated male fertility.
Materials and Methods
The Arabidopsis mutant coi1-1 was described previously (Feys et al. 1994). Seeds were surface-sterilized, plated on Murashige and Skoog medium (Sigma, St. Louis, MO, USA) supplemented with 1% sucrose, chilled at 4 °C for 3 d, and then grown under a 16 h light (23–25 °C):8 h dark (17–20 °C) photoperiod. The same photoperiod was followed for soil-grown plants.
Extraction of total proteins for 2-D DIGE
For 2-D DIGE analysis (Figure 2), flowers from both 5-week-old WT and coi1-1 mutant plants were harvested for protein extraction. Flowers were homogenized in ice-cold extraction buffer (0.7 M sucrose, 0.1 M KCl, 0.5 M Tris-HCl pH 7.4, 50 mM ethylenediaminetetraacetic acid [EDTA], 1% w/v polyvinylpolypyrrolidone [PVPP], 1 mM phenylmethanesulfonyl fluoride [PMSF], 0.2%ß-mercaptoethanol) supplemented with the protease inhibitor cocktail (Roche, Mannheim, Germany) and placed on ice for 30 min. The protein extracts were then mixed with an equal volume of ice-cold phenol (Tris-HCl pH8.0 buffered) and incubated for 30 min at 4 °C shaking. After centrifugation at 5 000g for 20 min, the proteins in the upper phenol phase were precipitated with five volumes of cold 0.1 M ammonium acetate in methanol at −20 °C overnight and centrifuged at 5 000g for 20 min. The pellet was then washed twice with ice-cold 0.1 M ammonium acetate in methanol, twice with 80% v/v acetone, once with 70% v/v ethanol and then dissolved in the lysis buffer (7 M urea, 2 M thiourea, 4% w/v CHAPS, 30 mM Tris-HCl pH 8.0, pH adjusted to 9.0). The proteins were cleaned up using a 2D clean up kit (GE Healthcare Bio-Sciences AB, Uppsala, Sweden) and resuspended in the lysis buffer after the pH was adjusted to about 8.5. Fifty microgram proteins were labeled with 400 pmol of DIGE specific Cy3 or Cy5 (GE Healthcare, Piscataway, NJ, USA) and incubated on ice in the dark for at least 30 min. The reaction was stopped by adding 1 μL of 10 mM lysine under the same conditions. The pooled sample internal standard was always Cy2-labeled.
Two-dimensional gel electrophoresis
Labeled protein samples to be separated in the same gel were mixed together with an equal volume of rehydration solution (8 M urea, 2% w/v CHAPS, 0.002 g/mL dithiothreitol [DTT] and 1% immobiline drystrip gel [IPG] buffer). The first dimension isoelectric focusing (IEF) was carried out on an Ettan IPGphor IEF system (GE Healthcare Bio-Sciences AB) with a 24 cm, pH gradient from 4.0 to 7.0 strip. After a total of 84 250 vhs IEF, IPG strips were treated with the equilibration buffer (6 M urea, 30% w/v glycerol, 2% w/v SDS, 50 mM Tris-HCl pH 8.0 and 1% bromophenol blue) first with 0.01 g/mL DTT, and then with 0.025 g/mL iodoacetamide. The strips were then overlayered onto 12.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels using the Ettan Dalt six apparatus (GE Healthcare Bio-Sciences AB) and electrophoresed at 1 W/gel for an hour followed by 17 W/gel at 15 °C until the bromophenol blue dye front had run off the bottom of the gels.
Image scanning and spot analysis
Cy2-, Cy3-, and Cy5-labeled images of each gel were acquired on a Typhoon laser scanner (GE Healthcare Bio-Sciences AB) at excitation/emission values of 488/520 nm, 532/580 nm, and 633/670 nm, respectively. Gel analysis was carried out with Decyder v6.5 software (GE Healthcare Bio-Sciences AB). First, First, the Decyder differential in-gel analysis (DIA) module was used to perform the spot detection, spot volume quantification and volume ratio normalization of different samples in the same gel. Then, the Decyder biological variation analysis (BVA) software module was used to perform the gel to gel matching and statistical analysis. The statistical significance of quantitative data was determined using the Student's t-test (n= 3). A spot with abundance ratio >1.5 or <−1.5 and P < 0.05 was set as a threshold to identify differently expressed proteins. Differently expressed protein spots were then subjected to in-gel trypsin digestion and peptide mass fingerprinting (PMF) analysis using a Voyager DE STR MALDI-TOF mass spectrometer (Applied Biosystems) with Proteomics Solution 1 (PS1) software (Applied Biosystems). The obtained spectrum was analyzed with the Data Explorer software. Self-degraded fragments of trypsin were used for internal calibration, while the standard peptide mixtures (angiotensin II, angiotensin I, substance P, bombesin, ACTH clip 1–17, ACTH clip 18–39, and somatostatin 28) were used for external mass calibration.
Protein identification and database search
The mass spectrum data were used to search for protein candidates using MS-Fit in the National Center for Biotechnology Information non-redundant (NCBInr) database. MS-Fit searching parameters were as follows: MW searched from 1 000 to 100 000 Da and pI searched from 0 to 14; allowing one trypsin missed cleavage site, four minimum numbers of peptides to match and a mass accuracy of 50 ppm; N terminus-Hydrogen (H) and C terminus-free acid (OH); fixed modifications-carbamidomethylation of Cys (C) and considered modifications-phosphorylation of Ser (S), Thr (T) and Tyr (Y).
The western blot analysis was carried out as per previously described procedures (Xu et al. 2002). The AtGSTF9 antibody was made by Alpha Diagnostic International Inc. The ribulose bisphosphate carboxylase oxygenase activase (RCA) antibody was generated by Jiang et al. (Jiang et al. 2001).
(Co-Editor: Chun-Ming Liu)
We thank Dr De-An Jiang for providing RCA antibody. This work was financially supported by the National Key Program for Transgenic Breeding (2008ZX08011-006), the State Key Basic Research and Development Plan of China (2006CB101606), the National Natural Science Foundation of China (30630044) and Singapore A-star grant to D.X.