Proteomic and Metabolic Profiling of Rice Suspension Culture Cells as a Model to Study Abscisic Acid Signaling Response Pathways in Plants Sushma R. Rao,† Kristina L. Ford, Andrew M. Cassin, Ute Roessner, John H. Patterson,‡ and Antony Bacic* Australian Centre for Plant Functional Genomics, School of Botany, University of Melbourne, VIC 3010, Australia Received July 28, 2010
Rice (Oryza sativa cv Taipei 309) suspension culture cells (SCCs) were used as a simple, single cell model system to gain insights into the complex abscisic acid (ABA) signaling response pathways in plants. Following system establishment involving morphological observations and transcript profiling of genes known to be ABA responsive in planta, a comprehensive proteomic and metabolomic study was performed. A total of 759 buffer-soluble proteins that included 3284 peptides categorized into 656 protein families are reported. Using iTRAQ, only 36 of these proteins showed statistically significant changes in abundance in response to ABA. In addition, a GC-MS based metabolite profiling study allowed the identification of 148 metabolites that included 25 amino acids (AAs), 45 organic acids (OAs), 35 sugars, 19 fatty acids, 2 polyamines, 4 sterols, 5 sugar acids, 4 sugar alcohols, and 9 miscellaneous compounds. Of these, only 11 (8.8%) changed in a statistically significant manner in response to ABA treatment. These studies provide important insights into plant responses to ABA at the protein and metabolite level. Keywords: Oryza sativa (rice) suspension culture cells • ABA • proteomics • metabolomics
Introduction The plant hormone abscisic acid (ABA) is a mediator of cellular responses during growth and development as well as to environmental stresses, including drought. ABA is produced in the roots in response to a water deficit and this signal moves throughout the plant where it is perceived by ABA receptors. This interaction activates a complex web of signaling pathways that include protein phosphorylation, activation of G-proteins and modulation of RNA metabolism.1,2 The pH and redox status of the cell are crucial factors in mediating ABA signal transduction.3 These signals lead to appropriate physiological, metabolic, transcriptional and translational modifications. Studying ABA signal transduction at the molecular level is helping to identify new targets and pathways involved in these responses. Part of the rationale for the present study was to evaluate the utility of SCCs, given they are a simple single cell type that can be generated in high yield, as a model system to study plant ABA responses using high throughput ’omics technologies. Advances in proteomic and metabolomic techniques have allowed the large-scale profiling of samples from many species. * To whom correspondence should be addressed. Antony Bacic. E-mail:
[email protected]. Phone.: +613-83445041. Fax: +913-93471071. Website: www.plantcell.unimelb.edu.au. † Current address: Sushma Rao, Children’s Medical Research Institute, 214 Hawkesbury Road, Westmead, NSW 2145. ‡ Current address: John Patterson, Viewbank College, Warren Rd, Viewbank VIC 3084. 10.1021/pr100788m
2010 American Chemical Society
The ability to analyze and compare dynamic changes quantitatively in a plant proteome in response to a stimulus, using the iTRAQ approach is a significant added advantage of modern proteomics.4 Proteome profiling has been carried out in various plant species such as Arabidopsis mitochondrion,5 Pinus radiata,6 Holm oak leaf,7 wheat lemma8 and leaf.9 Comprehensive comparative studies of protein profiling during either development or stress responses, and functional analysis and characterization of regulatory processes are now needed to understand plant complex signaling and response mechanisms. Likewise, with the detection and relative quantification of many of the major metabolites, it should be possible to link them to biochemical pathways and understand their role. It is predicted that there are around 200 000 metabolites in the plant kingdom.10 Metabolite profiling has been used to study temperature, water, salinity, sulfur, phosphorus, oxidative and heavy metal stress responses in a number of plant species11-13 and have resulted in the identification of previously unknown metabolic responses and networks. In this study, GC-MS based metabolite profiling was used, as its utility has been demonstrated in the analysis of metabolites from Solanum tuberosum,14 Lycopersicon esculentum,15 Hordeum vulgare,12 Lotus japonicus,16 Oryza sativa17,18 and Arabidopsis thaliana.19 The integrated metabolomic and quantitative proteomic approach adopted in this study reveals some of the known components of the ABA signaling process and also suggests the possible involvement of other factors. The time-course design enabled Journal of Proteome Research 2010, 9, 6623–6634 6623 Published on Web 10/20/2010
research articles the temporal tracking of changes together with identification of relevant proteins and metabolites in response to ABA.
Materials and Methods Plant Material and ABA Treatment. All chemicals were from Sigma Aldrich unless otherwise specified. Rice (Oryza sativa cv Taipei 309) suspension cell cultures (SCCs), a gift from the late Prof. Hans Kende, Michigan State University, East Lansing, MI, were grown in Murashige and Skoog medium lacking IAA, kinetin or sucrose (MP Biomedicals). This was supplemented with 6% (w/v) sucrose, 60 mM 2,4-dichlorophenoxyacetic acid (2,4-D) and 2.5 mM 2-(N-morpholino)ethanesulfonic acid (MES). The pH of the medium was adjusted to 5.6 with 1 M KOH. The cultures were maintained in 250 mL flasks on shakers running at 100 rpm at a temperature of 23 °C in the dark. The cells were subcultured every 7 days by transferring 50 mL of cells and media into fresh MS medium. Rice SCCs (1 L) were grown to their midlog phase (7 d after subculture). Cell growth curves were obtained by monitoring the fresh weight of cells for 2 weeks in the presence or absence of 100 µM ABA (added from a 100 mM stock, dissolved in ethanol). At 48 h intervals, the fresh weight of cells was recorded for control cells and cells from cultures supplemented with either 100 µL ethanol or 100 µL MS media (3 replicates in separate flasks used for each treatment). In each experiment, the cells were harvested and weighed in triplicate (Supporting Information I). Cells in the exponential phase of their growth (7 days) were used for all further experimentation. For proteomic and metabolomic experiments, the parental culture (1 L) was divided into 4 separate culture flasks (250 mL each). One flask from each set was used as the untreated control. The 3 remaining flasks were treated with 100 µM ABA for 0.5, 2, and 6 h, respectively. Three individual biological replicate parental flasks were grown and divided as described for each time point. The triplicates were harvested by filtering the SCCs through a vacuum filter setup with a cellulose filter disc (Whatman) to remove the media. The cells were then snap frozen in liquid nitrogen. Biological replicates, three for proteomics and five for metabolomics were used. Real Time-Q-PCR. Cells treated with 100 µM ABA for 0 (without ABA), 2, 4, 6, and 8 h were used for the Q-PCR experiments. Separate flasks were used for each time sampling point. The primer pairs indicated in Supplementary Table 4 (Supporting Information) were used for RT-QPCR experiments carried out at the University of Adelaide with the assistance of Dr. Neil Shirley (ACPFG, http://www.acpfg.com.au/). The size and the identity of the products were confirmed by sequencing. Q-PCR analysis was carried out as per the method described by Burton et al.20 Three replicate PCRs for each of the cDNAs were included in every run. The data obtained were normalized with respect to the control genes, actin (OsActin), ribosomal (10S) RNA (Os10SRNA), tubulin (OsTubulin) and glyceraldehyde-3-phosphate dehydrogenase (OsGAPDH) and analyzed. Protein Extraction. The frozen samples were ground in liquid nitrogen to a fine powder and suspended in an extraction buffer containing Tris.HCl (50 mM pH 8.5), MgCl2 (15 mM), NaCl (75 mM), β-glycerophosphate (15 mM), para-nitrophenylphosphate (pNPP) (15 mM), NaF (1 mM), EDTA (0.25 M), PMSF (1 mM), Na3VO4 (0.5 mM), DTT (1 mM), aprotinin (40 g/L), leupeptin (20 g/L) and pepstatin A (20 g/L) at 4 °C. Samples were vortexed and centrifuged at 3000× g for 30 min. The supernatant was collected and further centrifuged at 100 000× g for 30 min. Proteins in the membrane-depleted 6624
Journal of Proteome Research • Vol. 9, No. 12, 2010
Rao et al. supernatant were collected and concentrated by mixing with six volumes of -20 °C equilibrated acetone, left at -20 °C for 2.5 h and centrifuged at 3000× g for 10 min. The protein pellet was resuspended in triethylammonium bicarbonate (0.5 M, pH 8.5) containing 0.1% SDS. Soluble protein concentration was determined by the Bradford assay using bovine serum albumin (BSA) as a standard.21 The acetone precipitated proteins (40 µg) were reduced with tris-(2-carboxylethyl) phosphine (5 mM), at 60 °C for 1 h. Cysteine residues were blocked by incubating proteins in methyl methane thiosulphate (MMTS) (90 mM) in isopropanol for 10 min at RT. Proteins were then digested with trypsin (4 µg; 1:10 w/w, sequencing grade, Promega) for 16 h at 37 °C. iTRAQ Labeling. iTRAQ tags (Applied Biosystems) were resuspended in ethanol (70 µL) and added to the peptide mixtures. One set of iTRAQ tags were used per biological replicate (replicate 1: control 114; 0.5 h 115; 2 h 116; 6 h 117, replicate 2: control 115; 0.5 h 116; 2 h 117; 6 h 114, replicate 3: control 116; 0.5 h 117; 2 h 114; 6 h 115). iTRAQ tagging was carried out by incubation at RT for 1 h and resulting labeled peptide mixtures were pooled and lyophilized for 2D-LC-MS/ MS. Offline Strong Cation Exchange of the iTRAQ-Labeled Peptides. The iTRAQ tagged peptide sample was resuspended in 0.1% formic acid and passed over a C18 SepPak column (Waters, Milford, MA) and concentrated under vacuum to a final volume of 100 µL. The concentrated tryptic peptides were separated on a PolySULFOETHYL Aspartamide SCX column (4.6 mm × 200 mm, 5 µm, 300 Å, PolyLC Inc., Columbia, MD) attached to an Agilent 1100 series HPLC system (Agilent Technologies, Palo Alto, CA) with the following separation gradient: buffer A (25% (v/v) acetonitrile in 5 mM phosphate buffer, pH 3) for 10 min, then up to 100% buffer B (300 mM potassium chloride, 25% (v/v) acetonitrile in 5 mM phosphate buffer, pH 5) over 30 min at a flow rate of 0.7 mL/min with 0.5 min fractions being collected in a 96-well plate. Inline Reversed-Phase LC-ESI-MS/MS. Fractions obtained from SCX-HPLC were reduced under vacuum and resuspended in 0.1% formic acid (60 µL), filtered through a minisart membrane (0.2 µm; Sartorius Stadim Biotech, Aubagne, France) and one-quarter of each fraction was loaded onto a reversedphase precolumn (300 µm × 5 mm Zorbax 300SB-C18; Agilent Technologies, Palo Alto, CA) attached to a Shimadzu Prominence nano LC system (Shimadzu Corporation, Kyoto, Japan). The precolumn was washed with 0.1% formic acid in 5% acetonitrile for 15 min before placing in-line with a 75 µm i.d. × 150 mm Zorbax 300SB-C18 (Agilent Technologies, Palo Alto, CA) reversed-phase column. Peptides were eluted using a gradient of 5-65% (v/v) acetonitrile in 0.1% formic acid over 60 min, at a flow rate of 0.25 µL min-1. Peptides were analyzed via electrospray ionization (ESI) on a QSTAR Elite hybrid quadrupole time-of-flight mass spectrometer (Applied Biosystems/MDS Sciex, Foster City, CA). Each SCX-HPLC fraction was chromatographed and analyzed three times. The MS was operated in the positive ion mode, ion source voltage of 2200 V, using 10 µm uncoated SilicaTips (New Objective, Woburn, MA). Analyst QS 2.0 software (Applied Biosystems/MDS Sciex, Foster City, CA) was used to collect data in a data-dependent acquisition mode for the three most intense ions fulfilling the following criteria: m/z between 450 and 2000; ion intensity 40 counts; and charge state between +2 and +5. After MS/MS analysis, these ions were dynamically excluded for 18 s, using a mass tolerance of 50 mDa. MS scans
research articles
Oryza sativa Cell Suspension Culture Proteomics were accumulated for 0.5 s, and MS/MS scans were collected in automatic accumulation mode for a maximum of 2 s. Mass and charge state-dependent rolling collision energy was used and the MS instrument was calibrated daily with [Glu]fibrinopeptide B (Sigma-Aldrich, St. Louis, MO). Proteomic Data Analysis. Peak lists from the MS/MS spectra were made using ProteinPilot software version 2.0.1 (Applied Biosystems/MDS Sciex Foster City, CA). The peak list was searched against Oryza sativa proteins downloaded from NCBI in July 2009 using MASCOT22 and the Paragon Algorithm (Applied Biosystems/MDS Sciex).23 The rationale for using two separate algorithms was to reduce the false positive rates of the peptides identified. The false positive rate for this study was calculated using randomized version of the rice protein database from NCBI and was found to be 0.2%. The MASCOT parameters were: enzyme: trypsin, fixed modifications: iTRAQ4plex (N-term); iTRAQ4plex (K); Methylthiol (C), variable modifications: iTRAQ4plex (Y), MS peptide tolerance: 0.25 Da, MS/MS tolerance: 0.15 Da, number of missed cleavages: up to 1. The Paragon Algorithm parameters were: sample type: iTRAQ 4plex (peptide labeled); Cys Alkylation: MMTS; Digestion: Trypsin; Search effort: Thorough ID. The outputs from both search algorithms were combined and only proteins with 2 or more peptides with a P < 0.05 in both search algorithms were reported. The reporter ion peak areas generated in ProteinPilot were used for quantification. Any peptide with a reporter ion peak area of less than 20 was removed from quantification. When a peptide was detected more than once, the peak area for each reporter ion was summed, each peptide was then normalized by the sum of its channel intensities (114, 115, 116, and 117). Peptides were ignored when the normalized peptide value was more than 2 standard deviations from the calculated mean of the protein the peptide matched to. The mean was then calculated for proteins with 3 or more peptides that fulfilled the above criteria. The time points were normalized to the control and again by the average protein ratio for each time point. The results for each replicate were combined and only those proteins seen in all 3 replicates are reported. This analysis therefore focused primarily on proteins that showed changes in abundance in response to ABA, rather than proteins that may have become differentially post-translationally modified in response to ABA. Proteins were classified into their respective functional categories as specified either in NCBI or Uniprot (Universal protein resource: http://www.uniprot.org/). Extraction, Derivatization, Profiling and Statistical Analysis of Metabolites. The harvested rice SCCs were frozen in liquid nitrogen, homogenized to a fine powder which was used for metabolite extraction using a modified method of Jacobs et al.18 To account for biological and technical variability, 5 biological replicates (SCCs grown in separate flasks) were analyzed for each particular time point.15 Approximately 70-100 mg of the sample (accurate weights were recorded) was extracted in 100% methanol (500 µL) and a polar internal standard (20 µL of 0.2 mg/mL ribitol/norleucine in water) was added. The mixture was extracted for 15 min at 70 °C and water (500 µL) was added and vortexed. The mixture was centrifuged at 14 000 rpm for 10 min. The dried polar residue was derivatized using both the trimethylsilyl (TMS) and tributyldimethylsilyl (TBS) methods.14,15 as described by Roessner et al.12 and the samples (1 µL) were injected onto a GC column in the splitless mode and run on a GC-MS system comprising of an AS3000 auto sampler, an ultra trace GC and a DSQ quadrupole MS (Thermo Electron Cor-
poration, Madison, WI) using the conditions described in Roessner et al.12 The analysis of TBS samples was performed as described in Jacobs et al.18 and the mass spectra were recorded at two scans per second over a mass range of massto-charge (m/z) ratio of 70 to 600 atomic mass units (amu). Both total ion chromatograms (TIC) and mass spectra were evaluated using the Xcalibur program (ThermoFinnigan) and the resulting data prepared and normalized as described in Roessner et al.14 Mass spectra of eluting compounds were identified using a combination of an in-house constructed mass spectra library of authentic standards, the public domain mass spectra library of the Max-Planck-Institute for Molecular Plant Physiology, and the commercial mass spectra library of the National Institute of Standards and Technology.14,18,24,25 Where possible, all matching mass spectra were additionally confirmed by determination of the retention time and mass spectra by analysis of authentic standards. Resultant relative response ratios were normalized per gram extracted fresh weight as previously described.12 Metabolite profile data is presented as fold changes with respect to the control which is set at 1. Differences between the treated samples were considered as significant when the p-value using student’s t test was
