Analysis of Plasma Membrane Proteome in Soybean and Application to Flooding Stress Response Setsuko Komatsu,†,* Takuya Wada,†,‡ Yann Abale´a,†,§ Mohammad-Zaman Nouri,† Yohei Nanjo,† Norikazu Nakayama,† Satoshi Shimamura,† Ryo Yamamoto,† Takuji Nakamura,† and Kiyoshi Furukawa‡ National Institute of Crop Science, Tsukuba 305-8518, Japan, Nagaoka University of Technology, Nagaoka 940-2188, Japan, and Pierre et Marie Curie University, Paris 75005, France Received March 28, 2009
The plasma membrane acts as the primary interface between the cellular cytoplasm and the extracellular environment. To investigate the function of the plasma membrane in response to flooding stress, plasma membrane was purified from root and hypocotyl of soybean seedlings using an aqueous two-phase partitioning method. Purified plasma membrane proteins with 81% purity were analyzed using either two-dimensional polyacrylamide gel electrophoresis followed by mass spectrometry and protein sequencing (2-DE MS/sequencer)-based proteomics or nanoliquid chromatography followed by mass spectrometry (nanoLC-MS/MS)-based proteomics. The number of hydrophobic proteins identified by nanoLC-MS/MS-based proteomics was compared with those identified by 2-DE MS/sequencer-based proteomics. These techniques were applied to identify the proteins in soybean that are responsive to flooding stress. Results indicate insights of plasma membrane into the response of soybean to flooding stress: (i) the proteins located in the cell wall are up-regulated in plasma membrane; (ii) the proteins related to antioxidative system play a crucial role in protecting cells from oxidative damage; (iii) the heat shock cognate protein plays a role in protecting proteins from denaturation and degradation during flooding stress; and (iv) the signaling related proteins might regulate ion homeostasis. Keywords: flooding stress • soybean • plasma membrane • proteome
Introduction The plasma membrane is an organized system that plays a structural role and functions as a communication interface with the extracellular environment for the exchange of information and substances. Both biotic and abiotic stresses cause significant intracellular restructuring in plants.1 In plant cells, the processing of signals involved in responses to biotic and abiotic stresses occurs in the plasma membrane. Therefore, a better understanding of the plasma membrane proteome would help in developing strategies to increase the natural defenses or tolerance of plants. The plasma membrane controls many primary cellular functions such as metabolite and ion transport, endocytosis, and cell differentiation and proliferation. In addition, the degree of association of proteins with the membrane varies, with some proteins embedded in the membrane lipid core, whereas other proteins are peripheral proteins that occasionally associate with the membrane through reversible interactions.2 Plant membrane proteomics can provide valuable information on plant-specific processes; however, the objective * To whom correspondence should be addressed: Setsuko Komatsu, National Institute of Crop Science, 2-1-18 Kannondai, Tsukuba 305-8518, Japan. Fax: +81-29-838-8694; E-mail:
[email protected]. † National Institute of Crop Science. ‡ Nagaoka University of Technology. § Pierre et Marie Curie University. 10.1021/pr9002883 CCC: $40.75
2009 American Chemical Society
of much research on the plasma membrane is to find methods of extracting and identifying the entire set of mainly hydrophobic plasma membrane proteins.3,4 Complete genome sequences, which are available for rice5 and Arabidopsis,6 provide insights into many fundamental aspects of plant biology; they do not, however, address some important aspects of legume biology. Legumes are important for maintenance of human health and as crops for sustainable agriculture. Two model species of legume, Lotus japonicus and Medicago truncatula,7 have been the focus of projects on genome sequencing and functional genomics. Using genome sequence information, plasma membrane results of proteomics studies have been reported for rice,8,9 Arabidopsis2,10 and M. truncatula.11 Part of the genome sequence of the agricultural legume soybean was released in 2008, but functional genomics studies of this crop are in their infancy, and therefore proteomics approaches could be a powerful tool for functional analysis.12 Soybeans are susceptible to flooding stress at germination,13 early vegetative and early reproductive growth stages,14 and their seed yields are substantially reduced in response to this stress. Higher plants are aerobic organisms that die when oxygen availability is limited due to soil flooding.15 Gene expression studies of plants exposed to low oxygen revealed the up-regulation of genes coding for transcription factors,16 Journal of Proteome Research 2009, 8, 4487–4499 4487 Published on Web 08/06/2009
research articles
Komatsu et al. 17
18
signal transduction components, nonsymbiotic hemoglobin, ethylene biosynthesis,19 nitrogen metabolism,20 and cell wall loosening.21 However, research at the protein level is insufficient. It has been reported that flooding injury of soybean seeds before radicle protrusion is caused by physical disruption of the normally rapid uptake of water and can be alleviated by using seeds with high moisture content.22 The causes of flooding injury in stages after radicle protrusion, however, have not been elucidated. Shi et al.23 reported that cytosolic ascorbate peroxidase 2 is involved in flooding stress responses in young soybean seedlings using a proteomic technique. Hashiguchi et al.24 suggested using proteome analysis that flooding stress in soybean seedlings included not only hypoxic stress, but also other stresses such as those due to weak light, disease, and water stress. Furthermore, Komatsu et al.25 reported that the expression of many proteins that changed due to flooding showed the same tendencies in protein expression changes observed for nitrogen substitution; however, the expression of proteins classified as involved in protein destination or storage did not. The plasma membrane was considered as an important player in anoxia stress during flooding. Cytosolic pH regulated the root water transport during anoxic stress through gaiting of aquaporins, which are water channel proteins of plasma membrane intrinsic protein.26 This achieved via a gating mechanism common to different plant species.27 Channel closure resulted from the protonation of a conserved histidine residue of aquaporin following a drop in cytoplasmic pH due to anoxia during flooding.28 In addition, participation of plasma membrane in flooding stress was suppression of solute uptake via breakdown of the H+ gradient across it in azuki bean epicotyls.29 However, there are not many researches on the importance of plasma membrane under flooding. In the present study, to investigate the function of the plasma membrane in response to flooding stress plasma membrane proteins were purified with an aqueous two-phase partitioning method from root and hypocotyl of soybean seedlings. Purified plasma membrane proteins were analyzed by two-dimensional polyacrylamide gel electrophoresis followed with mass spectrometry and protein sequencing (2-DE MS/sequencer)-based proteomics, and by nanoliquid chromatography followed with mass spectrometry (nanoLC-MS/MS)-based proteomics. 2-DE MS/sequencer and nanoLC-MS/MS-based proteomics techniques were also applied to identify the plasma membrane proteins in soybean that are responsive to flooding stress.
Experimental Procedures Plant Growth and Treatment. Seeds of soybean (Glycine max L.) cultivar Enrei were sterilized by sodium hypochlorite solution then germinated on sand for 3 days under white fluorescent light (600 µmol m-2 s-1, 12 h light period/day) at 25 °C and 70% relative humidity in a growth chamber. For stress experiments, soybean seeds that had germinated on sand for 2 days were treated with or without flooding for 1 day. The treatment group was submerged in additional tap water. After treatment, physiological parameters including fresh weight and length of root and hypocotyl were measured, and protein analysis was carried out. The experiments were repeated more than three times. Plasma Membrane Purification. All protein extraction and purification steps were carried out keeping samples on ice, in a centrifuge held at 4 °C, or in a 4 °C cold room. A portion (20 g) of root and hypocotyl from 3-day-old soybean seedlings was 4488
Journal of Proteome Research • Vol. 8, No. 10, 2009
collected. Roots and hypocotyls were ground using a mortar and pestle with 140 mL of grinding buffer containing 400 mM sucrose, 75 mM MOPS, 5 mM EDTA, 5 mM EGTA, 10 mM potassium fluoride, 1 mM dithiothreitol and 2% polyvinylpyrrolidone 40 on ice. The homogenate was filtered with four layers of Miracloth (Calbiochem, San Diego, CA) and centrifuged at 10 000× g for 15 min to pellet insoluble material. The supernatant was adjusted to a volume of 180 mL with grinding buffer, then transferred to ultracentrifuge tubes and centrifuged at 200 000× g for 30 min. After discarding the supernatant, the precipitate, which included the proteins, was collected by dissolving it into 3 mL of grinding buffer and diluting with 6 mL of buffer containing 330 mM sucrose, 5 mM potassium phosphate buffer (pH 7.8) and 3 mM potassium chloride. The solution was mixed with 27 mL of buffer containing 330 mM sucrose, 5 mM K3PO4 (pH 7.8), 3 mM KCl, 6.2% polyethylene glycol (MW: 3350, Sigma-Aldrich, St. Louis, MO) and 6.2% dextran (Sigma-Aldrich). The plasma membrane fraction of the mixture was purified by two-phase partitioning. The mixture was centrifuged at 2450× g for 4 min and the upper phase of the solution was collected. Partitioning was carried out three more times and the final plasma membrane-enriched solution was diluted four times with 20 mM Tris-HCl (pH 7.5). This solution was then centrifuged at 200 000× g for 30 min and the precipitated plasma membrane fraction was dissolved in sample buffer containing 250 mM sucrose and 20 mM TrisHCl (pH 7.5). The solution was again centrifuged at 50 000× g for 1 h and the pellet dissolved in 100 µL of sample buffer containing 250 mM sucrose and 20 mM Tris-HCl (pH 7.5) according to Kawamura and Uemura.30 Plasma Membrane Purity Measurement. Purity of the plasma membrane fraction in solution was determined by measurement of plasma membrane-specific H+-ATPase activity compared to total ATPase activity. The assay was based on quantification of liberated phosphate from ATP using the molybdenum blue method.31 Vanadate (Na3VO4), nitrate (KNO3) and azide (NaN3) were used as inhibitors of ATPases that are associated with a specific site, respectively plasma membrane, vacuolar membrane and mitochondrial membrane. The reaction solution was composed of 30 mM MES-Tris (pH 6.5), 50 mM KCl, 3 mM MgSO4, 3 mM ATP and the respective inhibitors, 0.1 mM Na3VO4, 50 mM KNO3 or 10 mM NaN3. Water was added instead of inhibitor as a control. Purified protein (1 µg) was aliquoted to the reaction solutions and incubated for 15 min at 30 °C. The reaction was terminated by adding two times the volume of a stop solution containing 0.8 N H2SO4, 0.5% ammonium molybdate and 1% SDS. Ascorbate was then added at a final concentration of 0.3% and the mixture was placed at room temperature for 30 min. The concentration of the reaction product, ammonium phosphomolybdate chelate, was measured by absorbance at 750 nm using an Ultramark microplate imaging system (Bio-Rad, Hercules, CA). To generate a standard curve, K2HPO4 was used at a range of 0.1-0.5 mM.32 Two-Dimensional Polyacrylamide Gel Electrophoresis. The plasma membrane fraction was concentrated by addition of 50% trichloroacetic acid to a final concentration of 10%. The solution was kept for 30 min at 4 °C and centrifuged at 15 000× g for 10 min. The resultant precipitate was suspended in 200 µL of lysis buffer33 containing 8 M urea, 2% Nonidet P-40, 0.8% Ampholine (pH 3.5-10, GE Healthcare, Piscataway, NJ, USA), 5% 2-mercaptoethanol and 5% polyvinylpyrrolidone 40 using a glass mortar and pestle. The homogenate obtained above was
Plasma Membrane Proteome of Soybean centrifuged at 15 000× g twice for 10 min. The supernatant was used as a plasma membrane protein extract. Proteins (50 µL, 100 µg) were separated by 2-DE in the first dimension by an isoelectric focusing (IEF) tube gel33 and in the second dimension by SDS-PAGE. The IEF gel was composed of 8 M urea, 3.5% acrylamide, 2% Nonidet P-40, 2% Ampholine (pH 3.5-10 and pH 5-8; 1:1 mixture). Electrophoresis was carried out at 200 V for 30 min, followed by 400 V for 16 h and 600 V for 1 h. After IEF, SDS-PAGE in the second dimension was performed using 15% polyacrylamide for the separation gel with 5% polyacrylamide for the stacking gel. The protein spots were detected by silver staining (Sil-Best stain, Nacalai, Kyoto, Japan) or Coomassie brilliant blue (CBB) staining (Phast Gel Blue R, GE Healthcare). All procedures were performed according to the manufacturer’s protocol. Gel Image Analysis. 2-DE images were obtained using a GS800 calibrated densitometer (Bio-Rad). The position of individual proteins on gels was evaluated automatically with ImageMaster 2D Elite software (version 2.0) (GE Healthcare). The isoelectric point and molecular mass of each protein were determined using a 2-DE marker (Bio-Rad). The amount of a particular protein spot was estimated using the ImageMaster 2D Elite software, and expressed as the volume of that spot, which was defined as the sum of the intensities of all the pixels that make up the spot. For differential analysis, the position of individual proteins on gels was evaluated with PDQuest software (version 7.1) (BioRad). To correct the variability due to silver staining and to reflect the quantitative variations in intensity of protein spots, the spot volumes were normalized as a percentage of the total volume in all of the spots present in the gel. Five biological replications were used for the analysis, and spots that were upor down-regulated in at least three biological replications were considered reproducibly regulated. The expression ratio was assigned a cutoff value of 1.5, and spots which exceeded that threshold were used. To determine which protein spots displayed a statistically significant quantitative difference in expression, Fisher’s analysis of variance (ANOVA) method was used in Excel software. N-Terminal Amino Acid Sequence Analysis. To analyze N-terminal amino acid sequences following separation using 2-DE, proteins were electroblotted onto a polyvinylidene difluoride membrane (Pall, Port Washington, NY) using a semidry transfer blotter (Nippon-Eido) and detected by CBB staining. The stained protein spots were excised from the membrane and directly subjected to Edman degradation on a gas-phase protein sequencer (Procise cLC, Applied Biosystems, Foster City, CA). A FASTA search service provided by the National Institute of Agrobiological Sciences of Japan DNA bank (http:// www.dna.affrc.go.jp) was used for searching the Swiss-Prot or Uniprot-Sprot databases, which cover various sequences from plants in other taxons, to identify novel proteins in soybean. Protein Preparation for Mass Spectrometry. Protein spots were excised from CBB stained gels and destained with 50 mM NH4HCO3 for 1 h at 40 °C. Proteins were reduced with 10 mM DTT in 100 mM NH4HCO3 for 1 h at 60 °C and incubated with 40 mM iodoacetamide in 100 mM NH4HCO3 for 30 min. The gel pieces were minced and allowed to dry, then rehydrated in 100 mM NH4HCO3 with 1 pM trypsin (Sigma-Aldrich, St. Louis, MO) at 37 °C overnight. The tryptic peptides were extracted from the gel grains with 0.1% trifluoroacetic acid in 50% acetonitrile three times. The procedure described above was performed with robot (DigestPro, Intavis Bioanalytical Instru-
research articles ments AG, Cologne, Germany). The peptide solution obtained was dried and reconcentrated with 30 µL of 0.1% trifluoroacetic acid in 50% acetonitrile and desalted with NuTip C-18 pipet tips (Glygen, Columbia, MD). The desalted peptide solution was analyzed by matrix-assisted laser desorption ionization timeof-flight (MALDI TOF) MS (Voyager-DE RP mass spectrometer, Applied Biosystems) or nanoliquid chromatography-tandem MS (Ultimate3000, Dionex, Sunnyvale, CA; LTQ Orbitrap, Thermo Fisher Scientific, Waltham, MA). Analysis using Matrix-Assisted Laser Desorption Ionization Time-of-Flight Mass Spectrometry and Data Analysis for MS Spectra from Protein Spots. Calibration was external, and data were collected in the reflector mode. Data were searched on the Internet using an in-house licensed MASCOT search engine (version 2.2.18) software platform (Matrix Science, London, UK) against all entries in the GenBank (NCBI) database (release data 7 May 2008; 6,495,087 sequences) or the soybean genome database (version 4; 62,199 sequences), which was especially constructed for this research based on preliminary soybean genome sequences from the Department of Energy (DOE) Joint Genome Institute and Soybean Genome Sequencing Consortium. Soybean genome sequences were downloaded from the DOE database (http://www.phytozome. net, release data 24 January 2008) and then converted into FASTA format. Carbamidomethylation of cysteines was set as a fixed modification and oxidation of methionines was set as a variable modification. Trypsin was specified as the proteolytic enzyme and one missed cleavage was allowed. In the case of peptides matching among multiple members of a protein family, the proteins were selected based on the highest score and the highest number of matching peptides. For analysis, four criteria were used to assign a positive match with a known protein: (i) the deviation between the experimental and theoretical peptide masses needed to be less than 50 ppm; (ii) at least six different predicted peptide masses needed to match the observed masses for an identification to be considered valid; (iii) the matching peptides needed to cover at least 30% of the known protein sequence; and (iv) individual ions had to score more than 72 identity or extensive homology (P < 0.05). Analysis using Nano-Liquid Chromatography-Tandem Mass Spectrometry and Data Analysis for MS/MS Spectra from Protein Spots. A nanospray LTQ XL Orbitrap MS (Thermo Fisher, San Jose, CA) was operated in data-dependent acquisition mode with Xcalibur software installed on instrument. Using an Ultimate3000 nanoLC (Dionex, Germering, Germany), peptides were loaded in 0.1% formic acid onto a 300 µm ID × 5 mm C18 PepMap trap column. Elution of the peptides from the trap column and their separation on a 75 µm ID × 15 cm C18 PepMap100, 3 µm nanocolumn were done using 0.1% formic acid in acetonitrile at a flow rate of 200 nL/min. To spray sample into the MS, a PicoTip emitter (20 µm ID, 10 µm Tip ID, Woburn, MA) was used with a spray voltage of 1.8 kV. Full scan mass spectra were acquired in the Orbitrap over 150-2000 m/z with a resolution of 15 000. The three most intense ions at a threshold above 1,000 were selected for collision-induced fragmentation in the linear ion trap at a normalized collision energy of 35% after accumulation to a target value of 1000. Dynamic exclusion was employed within 30 s to prevent repetitive selection of the peptides. Acquired MS/MS spectra were converted to single DTA files using BioWorks software (version 3.3.1, Thermo Fisher Scientific). The following parameters were set for creation of the peak Journal of Proteome Research • Vol. 8, No. 10, 2009 4489
research articles lists: parent ions in the mass range with no limitation, one grouping of MS/MS scans and threshold at 100. Precursor ion tolerance was 10.00 ppm. Data were searched using the MASCOT search engine against all entries in the NCBI or DOE database. Carbamidomethylation of cysteines was set as a fixed modification and oxidation of methionine was set as a variable modification. Trypsin was specified as the proteolytic enzyme and one missed cleavage was allowed. Parameters for search were peptide mass tolerance 10 ppm, fragment mass tolerance 1 Da, maximum missed cleavages 3, peptide charges +1, +2, +3, variable modification methylation and fixed modification carbamidomethylation. The instrument setting was specified as “ESI-Trap”. Protein hits were validated if the identification was with at least four top ranking peptides with a P-value
