Analysis of proteomic changes in roots of soybean seedlings during recovery after flooding

J O U RN A L OF P ROTE O M IC S 7 5 ( 2 01 2 ) 8 7 8 –89 3

Available online at www.sciencedirect.com

www.elsevier.com/locate/jprot

Analysis of proteomic changes in roots of soybean seedlings during recovery after flooding Afshin Salavatia, b, 1 , Amana Khatoona, c, 1 , Yohei Nanjoa, 1 , Setsuko Komatsua,⁎ a

National Institute of Crop Science, National Agriculture and Food Research Organization, Tsukuba 305-8518, Japan Department of Agronomy and Plant Breeding, College of Agriculture and Natural Resources, University of Tehran, Karaj 31587-77871, Iran c Department of Plant Sciences, Kohat University of Science and Technology, Kohat 26000, Pakistan b

AR TIC LE I N FO

ABS TR ACT

Article history:

A proteomic approach was used to identify proteins involved in post-flooding recovery in soy-

Received 3 August 2011

bean roots. Two-day-old soybean seedlings were flooded with water for up to 3 days. After the

Accepted 3 October 2011

flooding treatment, seedlings were grown until 7 days after sowing and root proteins were

Available online 13 October 2011

then extracted and separated using two-dimensional polyacrylamide gel electrophoresis (2DE). Comparative analysis of 2-D gels of control and 3 day flooding-experienced soybean

Keywords:

root samples revealed 70 differentially expressed protein spots, from which 80 proteins were

Soybean

identified. Many of the differentially expressed proteins are involved in protein destination/

Recovery from flooding

storage and metabolic processes. Clustering analysis based on the expression profiles of the

Proteomics

70 differentially expressed protein spots revealed that 3 days of flooding causes significant

Root

changes in protein expression, even during post-flooding recovery. Three days of flooding resulted in downregulation of ion transport-related proteins and upregulation of proteins involved in cytoskeletal reorganization, cell expansion, and programmed cell death. Furthermore, 7 proteins involved in cell wall modification and S-adenosylmethionine synthesis were identified in roots from seedlings recovering from 1 day of flooding. These results suggest that alteration of cell structure through changes in cell wall metabolism and cytoskeletal organization may be involved in post-flooding recovery processes in soybean seedlings. © 2011 Elsevier B.V. All rights reserved.

1.

Introduction

Legumes are autonomous plants that are able to fix nitrogen and carbon [1]. Soybean is the most important legume crop, and its production is increasing and is expected to continue to increase as a result of global demand for soybean oil for human consumption and biodiesel fuel production, and the demand for high-protein meal for animal feed [2]. The availability

of the soybean genome sequence [3] provides unprecedented opportunities for investigations of the genes responsible for valuable traits in this important crop [4]. Although the molecular and cellular responses of soybeans to environmental stimuli have been extensively analyzed, the genetic basis of soybean stress responses is not well understood [5]. Soybean is sensitive to poor soil aerification caused by flooding or waterlogging [6], which significantly reduces growth and yield [7].

Abbreviations: 2-DE, two-dimensional polyacrylamide gel electrophoresis; CBB, Coomassie brilliant blue; ROS, reactive oxygen species; MS, mass spectrometry; pI, isoelectric point; IEF, isoelectric focusing; MALDI-TOF, matrix-assisted laser desorption ionization time-offlight; 1DFE, 1 day flooding experienced; 2DFE, 2 days flooding experienced; 3DFE, 3 days flooding experienced; FWR, flooded without recovery; CWR, 3-day-old seedling as control without recovery; Control, 7-day-old seedlings without flooding stress. ⁎ Corresponding author at: National Institute of Crop Science, National Agriculture and Food Research Organization, 2-1-18 Kannondai, Tsukuba 305-8518, Japan. Tel.: +81 29 838 8693; fax: +81 29 838 8694. E-mail address: [email protected] (S. Komatsu). 1 These authors equally contributed to this work. 1874-3919/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jprot.2011.10.002

J O U RN A L OF P ROT EO M IC S 7 5 ( 2 01 2 ) 8 7 8 –8 93

Flooding has a major negative impact on worldwide agricultural productivity. The water superfluity in root surroundings associated with flooding induces a decrease in the concentrations of cellular oxygen and of carbon dioxide, which leads to injurious effects on plants [8]. The yield of major crops is reduced by 50% annually as a result of damage caused by flooding stress [1]. Flooding stress is a complex phenomenon involving multiple stressors, such as hypoxic and light stress [9]. Flooding-induced increases in the production of reactive oxygen species (ROS) and a reduced capacity to detoxify ROS in susceptible plants lead to oxidative damage [10]. Flooding stress also decreases photosynthesis [11], nitrogen fixation [12], and carbon assimilation [13]. The evaluation of model systems can provide insights into low-oxygen sensing mechanisms and metabolic adjustments associated with controlling the use of carbohydrates and ATP [9]. It has been reported that the effects of flooding stress on soybean seedlings are manifested in every aspect of growth under stress lasting longer than 1 day [14,15]. Transcriptional and proteomic studies of flooded soybean seedlings have revealed that genes involved in alcohol fermentation, ethylene biosynthesis, and cell wall loosening are part of the anaerobic response of soybean to flooding stress [16]. Previous studies also demonstrated that flooding stress affects the expression of proteins involved in glycolysis [17], ROS scavenging [18], protein storage, disease resistance/defense [14], antioxidative systems, and signaling systems [19]. Furthermore, experiments involving nitrogen substitution have demonstrated that there are overlaps between the proteomic changes induced by flooding and those induced by low-oxygen stress [20]. While the responses of plants to flooding stress are generally well-known [21], relatively little work has focused on understanding the mechanisms associated with plant recovery after flooding. The processes involved in the recovery of plants from flooding stress have not been fully elucidated. In the present study, a proteomic approach was used to ascertain the changes that occur in the soybean seedling root proteome during recovery from flooding stress.

2.

Materials and methods

2.1.

Plant growth condition

Soybean (Glycine max L. cultivar Enrei) seeds were sterilized with sodium hypochlorite solution and rinsed in water. The sterilized seedlings were sown in a plastic case (180 ×140 × 45 mm) containing 400 mL of quartz sand wetted with 100 mL of water and grown at 25 °C and 70% humidity in a growth chamber (Sanyo, Tokyo, Japan) under white fluorescent light (600 μmol m− 2 s− 1, 12 h light period/day). For flooding treatments, two-day-old seedlings were flooded with 700 mL of water. The flooding condition was maintained at 2 cm of water above the quartz sand surface. The roots of soybean seedlings were collected and were used for analyses. In order to study the expression trend of proteins in recovery from flooding, seven-day-old soybean seedlings with or without experience of flooding stress were used. Seven-day-old seedlings grown for 7 days without flooding treatment were used as control (Control, Fig. 1). Seven-day-old soybean seedlings with

879

experience of flooding for 1, 2 and 3 days were used as 1, 2 and 3 days flooding experienced (1, 2 and 3DFE), respectively. Furthermore, three-day-old soybean seedlings with or without flooding treatment were also used for analysis. Three-day-old soybean seedlings with flooding were used as flooded without recovery (FWR, Fig. 6). Three-day-old soybean seedlings without flooding were used as control without recovery (CWR).

2.2.

Protein extraction

A portion (500 mg) of fresh roots was ground to powder in liquid nitrogen with a mortar and pestle. The powder was transferred to 10% trichloro acetic acid and 0.07% 2-mercaptoethanol in acetone and the mixture was vortexed. The suspension was sonicated for 5 min and then incubated for 45 min at −20 °C. After incubation, the suspension was centrifuged at 9000×g for 20 min at 4 °C. The supernatant was discarded and resulting pellet was washed twice with 0.07% 2-mercaptoethanol in acetone. The resulting pellet was dried using a Speed-Vac concentrator (Savant Instruments, Hicksville, NY, USA) and resuspended with 8 M urea, 2 M thiourea, 5% CHAPS and 2 mM tributylphosphine by vortexing for 1 h at 25 °C. The suspension was centrifuged at 20,000×g for 20 min at 25 °C. Supernatant was collected as protein extract. Protein contents were determined using the Bradford [22] method with bovine serum albumin as the standard.

2.3.

Two-dimensional polyacrylamide gel electrophoresis

Protein samples in a final volume of 180 μL of lysis buffer containing 0.4% Bio-Lyte pH 3/10 (Bio-Rad, Hercules, CA, USA) were directly loaded into a focusing tray. The immobilized pH gradient strips (3-10NL, 11 cm, Bio-Rad) were rehydrated for 14 h at 50 V. Isoelectric focusing (IEF) was carried out with the Protean IEF Cell (Bio-Rad) using following conditions: 250 V for 15 min with a linear ramp, 8000 V for 1 h with a linear ramp, and finally 8000 V at 35,000 V/h with a rapid ramp at 20 °C. After IEF, the strips were equilibrated with 6 M urea, 2% SDS, 0.375 M Tris– HCl (pH 8.8), 20% glycerol and 130 mM dithiothreitol for 30 min. The last equilibration step was done with 6 M urea, 2% SDS, 0.375 M Tris–HCl (pH 8.8), 20% glycerol, and 135 mM iodoacetamide for 30 min. The equilibrated strips were placed onto 15% SDS-polyacrylamide gels with 5% stacking gels and sealed with 1% agarose. Electrophoresis in the second dimension was performed at a constant current of 35 mA. After the electrophoresis, gels were stained with Coomassie brilliant blue (CBB).

2.4.

Gel image analysis

2-DE images were obtained using a GS-800 calibrated densitometer scanner (Bio-Rad) and the position of individual proteins on gels was evaluated with PDQuest software (version 8; Bio-Rad). The isoelectric point (pI) and molecular mass of each protein were determined using 2-DE standard marker (Bio-Rad). The amount of protein in a spot was estimated using the PDQuest software with local regression model normalization.

2.5.

Peptide preparation for mass spectrometry analysis

To identify proteins in protein spots using MS, protein spots were excised from 2-DE gels and washed with water. Proteins

880

J O U RN A L OF P ROTE O M IC S 7 5 ( 2 01 2 ) 8 7 8 –89 3

A

Sowing

Sampling

Control

Flooding

1DFE

Flooding

2DFE

Flooding

3DFE 0

1

2

3

4

5

6

7

days

B

Control

1DFE

2DFE

3DFE

Fig. 1 – Effect of flooding duration on the growth of soybean seedlings. (A) Experimental design. Two-day-old soybean seedlings were flooded for 1, 2, or 3 days. Roots were collected 7 days after sowing. Control, 1, 2, and 3DFE indicate control, 1, 2, and 3 days flooding stress, respectively. The duration of the flooding period is denoted by the rectangular box above the line. The times of sowing, initiation of flooding, and sampling are specified by downward triangles, upward triangles, and circles, respectively. (B) The effect of flooding duration on the growth of soybean seedlings. Representative 7-day-old soybean seedlings with or without flooding treatment are shown. Bar indicates 20 mm.

in the excised gel pieces were reduced with 10 mM dithiothreitol 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 (Wako, Osaka, Japan) at 37 °C overnight. The tryptic peptides were extracted from the gel grains with 0.1% trifluioroacetic acid in 50% acetonitrile 3 times. The procedure described above was performed with DigestPro (Intavis Bioanalytical Instruments AG, Cologne, Germany). The peptide solution obtained was dried and reconcentrated with 30 μL of 0.1% trifluioroacetic acid in 5% acetonitrile. The resulting peptides solutions were desalted with NuTip C-18 pipet tips (Glygen, Columbia, MD, USA). The desalted peptide solution was analyzed by matrix-assisted laser desorption ionization time-offlight (MALTI-TOF) MS or nano-LC–MS/MS.

2.6.

Protein identification by MALDI-TOF MS

Desalted peptide solution was mixed with α-cyano-4hydroxycinnamic acid. For analysis using a Voyager-DE-RP MALDI-TOF mass spectrometer (Perseptive Biosystems Inc., Framingham, MA, USA), calibration was external, and data were collected in the reflector mode. The resulting peptide

mass data were used to search this database using the MASCOT search engine (Matrix Science, London, UK). Soybean genome sequences were downloaded from the Department of Energy (DOE) database [3] (Phytozome, version 6.0, http:// www.phytozome.net/soybean) and converted into FASTA format. Search parameters used fixed cysteine carbamidomethylation and variable methionine oxidation as modifications, peptide mass tolerance ± 0.5 Da, fragments ions 1 Da, one missed cleavage, and trypsin was specified as the proteolytic enzyme. Peptides were selected in the 500–4000 Da mass range. For positive identification, the score result of (− 10 × Log (P)) had to be over the significance (> 60) threshold level (p < 0.05). Four criteria were used to assign a positive match with a known protein. These are as follows: (i) The deviation between the experimental and the theoretical peptide masses should be less than 50 ppm. (ii) At least seven different predicted peptide masses needed to match the observed masses for an identification to be considered valid. (iii) The coverage of protein sequences by the matching peptides must reach a minimum of 12%. (iv) The score that was obtained from the analysis with MASCOT search engine indicates the probability of a true positive identification and must be at least 60. The positives matches were BLASTP searched

881

J O U RN A L OF P ROT EO M IC S 7 5 ( 2 01 2 ) 8 7 8 –8 93

against the NCBI protein database (http://www.ncbi.nlm.nih. gov) for updated annotation and identification of homologous proteins.

2.7. Protein identification by nano-liquid chromatography–tandem MS A nanospray LTQ XL Orbitrap MS (Thermo Fisher Science, San Jose, CA, USA) was operated in data-dependent acquisition mode with the installed XCalibur software. Using an Ultimate 3000 nanoLC (Dionex, Germering, Germany), peptides in 0.1% formic acid were loaded onto a 300 μm ID × 5 mm C18 PepMap trap column. The peptides were eluted from the trap column and their separation and spraying were done using 0.1% formic acid in acetonitrile at a flow rate of 200 nL/min on a nano-capillary column (NTTC-360/75-3, Nikkyo Technos, Tokyo, Japan) with a spray voltage of 1.8 kV. Fullscan mass spectra were acquired in the Orbitrap over 150–2000 m/z with a resolution of 15,000. The three most intense ions above the 1000 threshold were selected for collision-induced fragmentation in the linear ion trap at 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 peptides. Acquired MS/MS spectra were converted to individual DTA files using BioWorks software (version 3.3.1) (Thermo Fisher Science). The following parameters were set to create a list of peaks: parent ions in the mass range with no limitation, one grouping of MS/MS scans and threshold at 100. The resulting peptide sequence data were used to search the soybean peptide database obtained from the soybean genome database using the MASCOT search engine. 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. The search parameters were peptide mass tolerance (10 ppm), fragment mass tolerance (0.2 Da), maximum missed cleavages 3, peptide and charges +1, +2, and +3. The minimal requirement for accepting a protein as identified was at least 5 peptide sequence matches above the identity threshold in coincidence with at least 14% sequence coverage. The Mowse score of more than 21 peptides from the MS data was significant with P < 0.05. The positives matches were BLASTP searched against the NCBI protein database for updated annotation and identification of homologous proteins.

2.8. Functional assignment of proteins and cellular localization Identified proteins were annotated to their biological function according to Bevan et al. [23]. Information of the identified proteins obtained from Universal Protein Resource (http://www. uniprot.org/), Phytozome (http://www.phytozome.net/) were used for classification. Also all identified proteins were analyzed with WoLF PSORT prediction [24] (http://wolfpsort.org/), ESLpred [25] (http://www.imtech.res.in/raghava/eslpred/), and Subloc [26] (http://www.bioinfo.tsinghua.edu.cn/SubLoc/ eu_predict.htm) to predict their subcellular localization.

2.9.

Clustering and statistical analyses

For the clustering analysis, the log2-transformed expression ratios of protein spots were used. Hierarchical clustering of proteins based on their expression profiles was performed using Gene Cluster 3.0 software [27] with Euclidean distance similarity metric and complete linkage method. The resulting clusters were visualized using JAVA TREEVIEW software [28]. The statistical significance of the results was evaluated with the Student's t-test when only two means were compared or with two-way ANOVA followed by Duncan's multiple comparisons test otherwise. Data analyses and graphical representations were performed using Microsoft Office Excel 2007 or R, a language and environment for statistical computing and graphics [29].

A pI kDa 76.0

4.5

5.0

6.5

8.5

2 1 5 6 3 9 8 7 15 17 11 13 10 12 20 16 14 18 19 21 22 23 24 25 29 26 27 33 30 28 31 32 36 38 42 35 40 34 37 43 41 45 44 47 48 39 50 53 46 49 55 54 51 52 57 56 58 61 59 60 63 62 65 64 69 70 66 68 67 4

43.0 37.0 32.0

21.5 20.0

B

kDa 76.0

43.0 37.0 32.0

21.5 20.0

Fig. 2 – 2-DE pattern of proteins from soybean seedling roots after 3DFE. Soybean seeds were germinated for 2 days and then flooded for 3 days. Proteins were extracted from the roots 2 days after removal of flood water, separated by 2-DE, and stained with CBB (B). Roots of 7-day-old soybean seedlings were used as a control (A). Open circles indicate differentially expressed spots. Upward arrows indicate upregulation of expression, and downward arrows indicate downregulation.

882

3.

J O U RN A L OF P ROTE O M IC S 7 5 ( 2 01 2 ) 8 7 8 –89 3

monitored in soybean seedlings. Two-day-old soybean seedlings were flooded for 1, 2, or 3 days, and then the seedlings were grown for an additional 4, 3, or 2 days, respectively, under non-flooded conditions (1, 2, and 3DFE, Fig. 1). The 7day-old soybean seedlings with and without flooding treatment were compared in terms of morphology. Root elongation, hypocotyl elongation, and the development of first leaves in the flooding-treated soybean seedlings were significantly delayed compared to control seedlings (Fig. 1B). The length of the delay depended on the duration of the flooding and recovery periods.

Results

3.1. Effect of flooding duration on the growth of soybean seedlings Although understanding of plant responses to flooding stress has increased in recent years, knowledge regarding the recovery process that occurs following a period of flooding is limited. To analyze the recovery process, changes in morphological factors brought about by flooding and recovery were

Relative intensity

A

120 100

*

80

*

**

* **

60 40

**

*

20

*

** *

0

1

4

5

7

8

9

10

11

*

*

**

** * **

*

12

13

14

15

16

17

20

21

22

Relative intensity

120 100

**

80

**

*

60

**

40 20 0

* **

*

** **

*

*

*

*

** *

*

** *

23 24 25 28 29 30 31 32 33 34 35 36 37 38 39 40 42 44

Relative intensity

120

**

100 80

40

Relative intensity

**

** ** ** *

* **

*

**

** * *

*

*

20

*

0

B

*

*

60

45 46 49 50 51 52 53 55 58 59 60 63 64 65 66 67 69 70

1000

**

800

*

600 400 200 0

** 2

* 3

** 6

18

19

*

*

* **

** 26

27

41

43

* ** * * 47

48

54

56

* 57

** * 61

62

68

Spot number Fig. 3 – Relative staining intensity of differentially expressed protein spots from 3DFE soybean seedling roots. Staining intensities were quantified using PDQuest software. Results are presented as the mean ± SE of relative protein intensity for gels from 3 biological replicates. White and black bars represent control and flooding treatment, respectively. Asterisks indicate significant differences between control and flooding treatment (*P < 0.05, **P < 0.01). Spot numbers are same as those shown in Fig. 2.

Table 1 – Differentially expressed proteins of soybean roots during recovery stage after flooding stress. Spot no. a 1 2

3 4

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

Copper amine oxidase Copper amine oxidase Glycyl-tRNA synthetase HSC70-cognate binding protein precursor Unknown 2,3-Bisphosphoglycerate-independent phosphoglycerate mutase RNA helicase-like protein V-H(+)-ATPase subunit A Coatomer subunit delta Selenium binding protein S-adenosyl-L-homocystein hydrolase Alpha tubulin TCP-1/cpn60 chaperonin family protein UDP-glucose dehydrogenase Beta-tubulin Pyruvate decarboxylase 1 ATP citrate lyase a-subunit ND UDP-glucose pyrophosphorylase GroEL-like chaperone, ATPase Elongation factor 1-gamma 6-Phosphogluconate dehydrogenase 26S proteasome subunit 7 6-Phosphogluconate dehydrogenase ND Actin isoform PEAc14-1 Actin isoform B Glutamate decarboxylase 26S proteasome AAA-ATPase subunit RPT4a Cytosolic glutamine synthetase GS beta 1 S-adenosylmethionine synthase phosphoglycerate kinase Leucin zipper-ef-hand containing transmembrane protein Alcohol dehydrogenase 1F Aspartate aminotransferase Phospholipase D delta TGF-beta receptor-interacting protein Ran binding protein 1 Adenosine kinase Unknown Short-chain dehydrogenase/reductase SDR

Accession no. b

Score c

Cov. (%) d

M. P. e

Blast score f

Mr (kDa)/pI Theo. g

Exp. h

Ratio i

p value j

MS k

Function l

Cluster m

CAE47488.1 CAE47488.1 NP_564337.1 AAB86942.1 ACU23913 XP_002519975

76 2495 876 341 60 1422

36 66 38 22 57 72

15 35 23 9 12 28

1284 1284 1092 1122 212 947

76.1/6.21 76.1/6.21 81.7/7.03 73.6/5.11 29.2/9.39 61.1/5.51

67.1/5.8 66.5/5.9 66.5/5.9 66.5/5.9 66.2/5.9 59.3/5.3

1.86 2.60 2.60 2.60 1.66 0.83

0.044 0.005 0.005 0.005 0.013 0.028

MS MS/MS MS/MS MS/MS MS MS/MS

SecMet SecMet ProtSyn ProtDes UnCl Ene

ADG27844.1 ABU87506.1 NP_568147.1 CAC67501.1 ACJ11250.1 ACJ60905.1 AAT77033.1 Q96558.1 ABS50666.1 AAO72533.1 CAC86995.1

281 216 215 1837 1711 263 340 2709 599 2416 1655

31 27 31 55 44 29 28 65 44 50 48

9 10 10 18 18 8 11 27 12 26 25

640 1217 744 924 882 831 959 931 842 1079 1113

55.4/5.44 69.0/5.48 58.5/5.42 53.8/5.67 53.85.60 50.3/4.92 57.4/5.55 53.5/5.74 51.3/4.77 66.3/5.85 66.4/7.95

58 62 58 22 21 41

23 40 23 11 6 16

828 1040 622 871 812 899

51.6/5.20 61.6/5.75 47.7/5.77 53.9/5.68 48.2/6.14 53.8/6.11

MS MS/MS MS/MS MS/MS MS/MS MS/MS

Met ProtDes ProtSyn Met ProtDes Met

ADP09679.1 BAA89214.1 BAF80896.1 BAC23035.1 AAG24873.1 A4ULF8.1 AAF85975.1 XP_002514773.1

462 363 712 79 27 920 173 63

45 55 31 18 14 72 50 44

52 14 12 6 6 20 15 15

754 756 1011 718 718 744 733 416

42.0/5.23 41.9/5.31 57.6/5.59 44.7/8.24 39.1/5.48 43.4/5.50 42.4/5.96 41.1/4.60

0.028 0.028 0.028 0.004 0.004 0.004 0.001 0.001 0.001 0.005 0.007 0.040 0.008 0.024 0.049 0.014 0.043 0.005 0.024 0.010 0.000 0.001 0.013 0.004 0.007 0.010 0.006

TranCri TranPor TranPor UnCl Met CellStr ProtDes Met CellStr Ene Ene

68 2821 1394 189 121 337

0.83 0.83 0.83 2.92 2.92 2.92 2.78 3.54 3.54 3.47 0.15 0.19 3.34 0.36 1.71 2.01 0.37 0.15 2.25 0.45 2.34 0.39 0.46 0.20 2.82 0.42 0.26

MS/MS MS/MS MS/MS MS/MS MS/MS MS/MS MS/MS MS/MS MS/MS MS/MS MS/MS

AAL33919 ABE86688.1 AAL82617.1 AAB41553.1 Q41365.1 BAA22812.1

59.3/5.3 59.3/5.3 59.3/5.3 59.2/5.4 59.2/5.4 59.2/5.4 53.6/5.5 53.8/5.6 53.8/5.6 54.0/5.6 52.8/8.5 50.1/4.7 46.9/5.0 47.6/5.1 47.7/5.7 46.3/5.7 48.0/5.8 47.3/5.9 50.0/8.0 43.3/4.9 42.2/5.0 44.1/5.5 40.2/8.6 40.5/4.9 39.4/4.9 38.9/5.2 39.0/5.3

MS/MS MS/MS MS/MS MS/MS MS/MS MS/MS MS/MS MS

CellStr CellStr Met ProtDes Met Met Ene UnCl

5 1 2 5 2 4 4 2 1 5 2 5 2 2 1 5 2 2

CAA80691.1 AAA33942.1 XP_002510602.1 AAK49947.1 ACF72671.1 XP_002531678.1 ACU19182.1 ABD32398.1

62 225 61 269 306 72 63 128

48 21 12 58 48 41 45 29

10 7 6 14 9 13 18 6

340 893

28.8/5.93 51.1/8.49 96.3/6.96 36.0/6.26 25.0/4.78 38.1/5.29 35.2/6.13 32.0/7.01

39.3/5.7 40.2/5.8 38.1/7.1 38.1/7.3 37.5/4.7 38.0/4.9 37.5/5.2 37.0/6.3

2.70 2.95 1.86 0.37 2.21 0.27 0.61 3.22

0.015 0.022 0.000 0.016 0.010 0.005 0.002 0.001

MS MS/MS MS MS/MS MS/MS MS MS MS/MS

Ene Met SigTra SigTra TranPor UnCl UnCl UnCl

5 4 4 2 5 2 3 4

635 286 584 587 509

4 5

4 3

5

4 5

883

(continued on next page)

J O U RN A L OF P ROT EO M IC S 7 5 ( 2 01 2 ) 8 7 8 –8 93

5

Homologous protein

884

Table 1 (continued)

34 35 36 37 38 39 40 41 42 43 44

45 46 47 48 49 50 51 52 53 54 55

Homologous protein

UDP-rhamnose:soyasaponin III-rhamnosyltransferase Sorting nexin 3 Splicing factor Chalcone reductase NAD(P)H-dependent 6′-deoxychalcone synthase Membrane attack complex component/perforin/ complement C9 Annexin Guanine nucleotide-binding protein Cyclophilin ND Globulin-like protein Ubiquinone biosynthesis protein COQ9, mitochondrial precursor Eukaryotic translation initiation factor 3f, eif3f inorganic pyrophosphatase Beta-1,3-endoglucanase ND GroES chaperonin Proteasome subunit alpha type Proteasome subunit alpha type-6 1-Cys peroxiredoxin ND Glutathione S-transferase GST 24 Flavodoxin-like quinone reductase 1 Chalcone-flavanone isomerase 1A Quinone oxidoreductase Chalcone isomerase A 1-Cys peroxiredoxin

Accession no. b

Score c

BAI99585.1 XP_002530347.1 XP_002530164.1 ACH42079.1 P26690.1

Cov. (%) d

M. P. e

Blast score f

231 64 62 66 124

27 58 34 33 68

7 23 6 9 14

64

20

XP 002513910.1 Q39836.1 AAX94775.1

85 71 323

AAM65577.1 XP_002516247.1

Mr (kDa)/pI

Ratio i

p value j

MS k

Function l

Cluster m

Theo. g

Exp. h

943 661 102 448 424

53.9/5.87 46.6/7.74 19.2/9.58 35.3/6.13 29.2/8.66

33.9/5.2 32.8/5.7 33.5/5.8 32.0/5.8 33.4/5.9

2.44 0.18 0.41 2.58 0.35

0.036 0.00 0.033 0.013 0.025

MS/MS MS MS MS MS

UnCl CellTra TranCri SecMet SecMet

5 1 2 5 2

11

604

64.2/8.85

31.9/5.9

2.22

0.030

MS

UnCl

4

49 48 73

14 15 8

523 659 310

36.1/6.48 36.1/7.62 18.3/7.68

72 52

20 13

289 343

39.0/5.12 32.3/8.39

0.59 0.17 0.65 0.40 3.94 3.94

0.001 0.001 0.014 0.010 0.013 0.013

MS MS MS/MS MS/MS MS/MS MS/MS

CellStr SigTra ProtDes

1581 248

32.8/6.0 31.9/8.0 33.8/8.0 30.8/4.6 32.6/4.8 32.6/4.8

3 1 3 2 5

XP_002526864.1 ADN34257.1 Q03773.1

234 202 166

57 29 33

15 10 6

421 461 694

31.8/5.13 32.5/6.37 38.1/8.72

XP_002516232.1 XP_002513374.1 O48551.2 Q6E2Z6.1

304 396 158 149

64 44 25 62

14 10 6 14

389 453 489 398

26.6/6.77 25.6/5.51 27.5/5.58 24.6/6.44

AAG34814.1 NP_200261.1 Q93XE6.1 CAD31838.1 ABI54176.1 ACF06438.1

488 118 70 895 126 513

45 45 49 62 46 85

12 16 11 14 9 16

399 334 394 384 390 348

24.9/5.74 21.7/6.09 23.3/6.23 21.7/6.43 23.2/6.23 24.5/6.44

32.6/4.8 32.6/4.8 31.1/8.7 29.8/4.6 30.5/5.3 30.3/5.5 28.6/5.3 28.0/5.6 26.4/5.6 26.9/5.6 26.9/5.6 27.0/5.8 26.1/5.8 26.1/5.8 26.9/6.9

3.94 3.94 2.29 0.32 0.57 0.55 0.49 2.59 1.85 0.44 0.44 0.65 0.40 0.40 3.34

0.013 0.013 0.012 0.022 0.021 0.004 0.001 0.010 0.004 0.040 0.040 0.008 0.018 0.018 0.004

MS/MS MS/MS MS/MS MS/MS MS/MS MS/MS MS/M MS/MS MS/MS MS/MS MS/MS MS MS/MS MS/MS MS/MS

ABE79564.1

ProtDes Met ProtSyn Met Met ProtDes ProtDes ProtDes Dis/Def Dis/Def Ene SecMet Met SecMet Dis/Def

4 2 3 3 2 5 5 2 3 2 5

J O U RN A L OF P ROTE O M IC S 7 5 ( 2 01 2 ) 8 7 8 –89 3

Spot no. a

56 57 58 59 60 61 62 63 64 65 66 67 68

a

AAT45389.1 BAA02117.1 ADD51186.1

129 119 106

42 29 49

7 5 6

194 398 195

22.0/5.44 22.7/5.09 22.0/5.12

AAW83328.1 AAQ03092

169 274

21 42

6 8

337 305

30.5/7.96 18.7/6.59

ACA23207.1 NP_001119147.1 ADN34205.1 XP_002530493.1 XP_002530493.1 AAR11455.1 P26987.1 AAD50376.1 CAQ56034.1 ACG24471.1 AAB68027 Q8W171.1

677 218 179 200 279 179 247 210 163 102 145 156

37 35 32 30 38 66 24 73 31 61 62 77

6 7 6 5 6 6 5 10 5 10 9 9

397 160 254 245 246 221 286 321 305 209 183 321

22.9/5.24 18.5/5.22 19.9/5.11 18.5/5.22 18.6/4.92 16.5/4.68 16.8/4.73 17.7/5.38 17.5/5.41 15.3/5.50 11.5/5.53 18.4/8.70

24.1/4.6 24.1/4.6 24.1/4.8 23.7/5.0 23.7/5.7 24.6/5.9 20.1/8.5 21.5/4.5 22.3/4.8 22.3/4.8 21.3/4.8 22.3/4.9 19.4/4.5 19.5/5.0 20.6/5.1 20.6/5.1 20.6/5.1 20.0/5.6 18.9/9.0

13.1 13.12 11.43 1.61 3.87 2.13 1.84 189.32 3.33 3.33 0.33 9.43 3.73 1.58 0.46 0.46 0.46 0.84 4.19

0.027 0.027 0.012 0.013 0.001 0.028 0.006 0.006 0.014 0.014 0.010 0.032 0.009 0.037 0.018 0.018 0.018 0.019 0.003

MS/MS MS/MS MS/MS MS/MS MS/MS MS/MS

Dis/Def TranPor UnCl

MS/MS MS/MS MS/MS MS/MS MS/MS MS/MS MS MS/MS MS/MS MS/MS MS/MS MS/MS

ProtDes UnCl Ene UnCl ProtDes DisDef DisDef DisDef DisDef ProtSyn ProtDes ProtDes

UnCl Dis/Def

6 6 4 4 5 4 7 5 2 6 5 4 2

3 5

Spot no., spot number as given in Figs. 2 and 3. Accession no., accession number according to the NCBI database. c Score, ions score of identified protein using soybean genome sequence databases. d Cov., sequence coverage, the proteins with less than 14% sequence coverage was excluded from the result. e M.P., number of query matched peptides, the proteins with ≥ 5 matched peptides were considered. f Blast score, The score of the high-scoring segment pair (HSP) from that database sequence. g Theo., theoretical; Mr, molecular weight; pI, isoelectric point. h Exp., experimental. i Ratio, the ratio of change in abundance of protein spots compared to the control analyzed by LSD test. j p value, indicates the significance of up- or down-regulation of spots according to the t-test through analysis of variance. k MS, the type of mass spectrometry used in this study. MS means MALDI-TOF-MS and MS/MS means nano-LC MS/MS. l Function, category using functional classification: ProtDes, protein destination/storage; Met, metabolism; Dis/Def, disease/defense; ProtSyn, protein synthesis; SecMet, secondary metabolism; CellStr, cell structure; Ene, Energy; SigTr, signal transduction; CellTra, cellular traffic; TranCri, transcription; TranPor, transporter; UnCl, unclear- and un-classified. m Cluster, belonging cluster which were divided in clustering analysis (Fig. 5). b

J O U RN A L OF P ROT EO M IC S 7 5 ( 2 01 2 ) 8 7 8 –8 93

69 70

Pathogen-inducible trypsin-inhibitor-like protein GTP-binding protein Tumor-related protein ND Mitochondrial prohibitin 1 Glutathione peroxidase ND Kunitz trypsin protease inhibitor Phosphatidyl ethanolamine-binding protein ATP synthase d chain Phosphatidylethanolamine binding protein Phosphatidylethanolamine binding protein Pathogenesis-related class 10 protein SPE-16 Stress-induced protein SAM22 Ripening related protein Peroxiredoxin 40S ribosomal protein S12 Late embryogenesis-abundant protein Peptidyl-prolyl cis-trans isomerase 1

885

886

J O U RN A L OF P ROTE O M IC S 7 5 ( 2 01 2 ) 8 7 8 –89 3

3.2. Identification of differentially expressed proteins induced in soybean roots during recovery from flooding To study the changes that occur during recovery from flooding, the root proteomes of 3DFE and control plants were compared. Extracted root proteins were separated using 2-DE and the spot patterns were compared. A total of 664 protein spots were reproducibly detected. It was determined that the intensity of 70 of these protein spots differed significantly between flooding-treated and control plants (Fig. 2A). Of these 70 protein spots, 38 (54.3%) were upregulated and 32 (45.7%) were downregulated in 3DFE plants compared to control plants (Fig. 2B). The intensity of 54 of the 70 differentially expressed protein spots changed by more than 2-fold (30 spots (55.6%) were upregulated by more than 2-fold, and 24 (44.4%) were downregulated by more than 2-fold). Moreover, the intensity of 4 of the upregulated spots (Nos. 56, 57, 62, and 65) and 5 of the downregulated spots (Nos. 9, 10, 16, 35, and 41) changed by more than 5-fold (Fig. 3). To identify the proteins that are differentially expressed during recovery from flooding, 2-DE protein spots were analyzed using MALDI-TOF-MS and nano-LC–MS/MS. A total of 80 proteins were identified from 63 of the differentially expressed protein spots (Table 1). No identifications were obtained for 7 spots, however. The identified proteins were grouped into classes based on their presumed biological function, as described by Bevan et al. [23]. A majority of the differentially expressed proteins identified were classified into the functional categories of protein destination/storage, metabolism, and disease resistance/defense processes (Fig. 4A). The expression of GroEL-like chaperone ATPase (spot 12), 26 S proteasome regulatory subunit 7 (spot 15), 26S proteasome regulatory subunit S10B (spot 21), cyclophilin (spot 42), and GroES chaperonin (spot 47) was downregulated in 3DFE seedlings. In contrast, globulin-like protein (spot 44), Kunitz trypsin protease inhibitor (spot 62), and peptidyl-prolyl cis-trans isomerase 1 (spot 70) were upregulated in 3DFE seedlings. All identified proteins were analyzed with WoLF PSORT, ESLpred, and Subloc to determine their subcellular localization. More than 50% of the differentially expressed proteins identified were predicted to be cytoplasmic, while most of the remaining proteins were predicted to localize in the nucleus and plastids (Fig. 4B).

3.3. Analysis of the expression profiles of proteins differentially expressed during recovery from flooding Differences in the expression data for the 70 differentially expressed protein spots were analyzed using the two-way ANOVA Duncan's multiple comparisons test. This analysis showed that there were significant differences in the expression of many protein spots. The analysis also indicated that the duration of the flooding stress as well as the recovery stage had a significant impact on the level of protein expression. For many of the identified proteins, the expression level varied greatly depending upon whether the flooding stress was mild or severe, and there was no linear pattern to the trend in protein expression (Supplementary Fig. 1). A hierarchical clustering analysis was performed with the expression data for the 70 differentially expressed

A

Protein destination/storage Metabolism Diseas/Defence Protein synthesis Secondary metabolism Cell structure Energy Signal transduction Cellular traffic Transcription Transporter Unclear-and Un-classified 0

4

8

12

16

20

Number of proteins

B Cytoplasm Nucleus Plastid Mitochondria Extracelular Vacuole Endoplasmic reticulum Plasma Membrane 0

10

20

30

40

50

60

70

Number of proteins Fig. 4 – Classification of proteins that are differentially expressed during post-flooding recovery in soybean seedling roots. (A) Differentially expressed proteins were identified and their functions were assigned using the classification scheme described by Bevan et al. [23]. Proteins were classified into 12 groups, including unclear and unclassified. (B) The subcellular localization of the identified proteins was predicted using WoLF PSORT (http://wolfpsort. org/), ESLpred (http://www.imtech.res.in/raghava/eslpred/), and Subloc (http://www.bioinfo.tsinghua.edu.cn/SubLoc/ eu_predict.htm). The proteins were classified based on their localization into 8 subcellular compartments. White and black bars indicate upregulation and downregulation, respectively.

protein spots from the control, 1DFE, 2DFE, and 3DFE samples to determine the nature of any trends. The clustering analysis led to the division of the 70 protein spots into 3 prominent categories, which were then subdivided into 7 clusters (Fig. 5, Table 1). The intensity of protein spots belonging to clusters 1 and 2 gradually decreased as the duration of flooding rose from 1 to 3 days, indicating that the expression of these proteins is dependent upon the duration of flooding or may return to a normal level depending on the duration of recovery. Moreover, the degree of the decrease in the expression of protein spots in cluster 1 was greater than that of the protein spots in cluster 2. The expression of protein spots belonging to cluster

887

J O U RN A L OF P ROT EO M IC S 7 5 ( 2 01 2 ) 8 7 8 –8 93

Log2 ratio

1DFE/C

2DFE/C

3DFE/C

-3 0 3

Cluster 9 16 22 35 41 10 25 31 21 49 20 46 24 36 54 18 43 68 29 38 15 52 12 64 4 69 32 40 48 42 53 47 6 1 3 67 45 33 14 28 61 13 58 39 27 59 2 8 17 19 11 23 7 44 55 70 63 66 5 37 34 26 50 30 60 51 57 56 65 62

1

2

1

3

Cluster 1

0

2

-1

1

-2

0

-3 1

-1 5

Cluster 2

4

0

Cluster 6

3

-1

2 1

-2

3

Cluster 5

0 -1 9

-3 1

Cluster 3

7

Cluster 7

5 0 3

4

1 -1 3

-1

1

2

3

DFE/C

Cluster 4

2 1 0

5 -1

1

2

3

DFE/C

6 7

Fig. 5 – Hierarchical clustering of proteins associated with post-flooding recovery in plants exposed to different durations of flooding stress. Hierarchical clustering based on the log2-transformed expression ratios of protein spots was performed using Gene Cluster 3.0 software with the Euclidean distance similarity metric and complete linkage method. The resulting clusters were visualized using JAVA TREEVIEW software.

3 was significantly lower in 3DFE plants, indicating that the decrease in the expression of these proteins is triggered by more than 3 days of flooding treatment. The intensity of protein spots belonging to cluster 4 was higher in 3DFE plants, indicating that more than 3 days of flooding treatment triggers an increase in the expression of these proteins. The intensity of protein spots belonging to clusters 5, 6, and 7 was higher in all flooding-treated samples, indicating that expression of these proteins increases upon flooding for more than 1 day. The increase in the expression of these proteins was sustained even during the recovery stage. Differences among the proteins in clusters 5 through 7 represented the degree of increases in expression.

3.4. Identification of root proteins involved in recovery from flooding To identify root proteins that may play a role in recovery from flooding, proteins from samples with and without a postflooding recovery period were analyzed using 2-DE (Fig. 6). A total of 365 protein spots were reproducibly detected. A comparison of the expression of root proteins from untreated 3day-old seedlings (CWR, Fig. 6B) and seedlings flooded for 1 day without a recovery period (FWR, Fig. 6C) showed that the expression of 21 protein spots was significantly different in the roots of the flooded seedlings (Table 2). A total of 16 of these protein spots (76%) were upregulated and 5 (24%) were downregulated in the FWR plants (Fig. 7). Analysis of these

888

J O U RN A L OF P ROTE O M IC S 7 5 ( 2 01 2 ) 8 7 8 –89 3

A

CWR Flooding

FWR Control

Flooding

1DFE 0 kDa

1

pI 4.5

5.0

2

3

6.5

days

7

D

(Control)

(CWR) 72

72

23 76

19 31 77

73 74

73 74 33

82 65 84

33

31 77

76

78 79

75

19

23

75

80

21.5

6

71

B

71

32.0

5

8.5

76.0

43.0

4

78

79 80

81

82

81 55

55 83 70

65

84

C

83 70

E

(FWR)

(1DFE)

Fig. 6 – 2-DE pattern of differentially expressed proteins in the roots of soybean seedlings after FWR treatment. (A) Experimental design. CWR indicates 3-day-old control seedlings without flooding treatment. FWR indicates 3-day-old seedlings with 1 day of flooding treatment and no recovery period. The duration of the flooding period is denoted by the rectangular box above the line. The times of sowing, initiation of flooding, and sampling are specified by downward triangles, upward triangles, and circles, respectively. (B to E) 2-DE patterns of proteins in the roots of soybean seedlings after CWR (B), FWR (C), Control (D) and 1DFE (E) treatments. Roots of control (D) and CWR (B) plants were used as controls. Open circles indicate spots with altered expression relative to controls. Upward arrows indicate upregulation of expression, and downward arrows indicate downregulation. Of the 21 differentially expressed protein spots, 7 protein spots were also detected in experiments involving 3DFE plants (see Fig. 2), and their numbers are the same as in Fig. 2 (spots 19, 23, 31, 33, 55, 65, and 70). For the 14 newly identified protein spots shown here, the numbering continued following Fig. 2 (spots 71–83).

21 protein spots using MS/MS resulted in the identification of 20 proteins, which included proteins involved in cell structure maintenance, metabolism, energy production, disease resistance/defense, and protein destination/storage processes (Table 2). Subsequently, the intensity of these 21 protein spots was compared between the 1DFE and control samples (Fig. 6D and E), and differences were found in the intensities of only 7 protein spots. This result suggests that expression of the proteins within the other 14 spots in 1DFE plants may have returned to normal levels due to the recovery period. The 7 protein spots that did not return to control levels after recovery included 6 spots for which expression increased (Nos. 19, 23, 55, 65, 70,

and 78) and 1 spot (spot 31) for which expression decreased relative to the control. Five of the 7 protein spots (those numbered 70 and below) were also identified as being differentially expressed relative to controls in 3DFE plants (Table 1). Five of these 7 proteins spots were grouped in either cluster 5 (spots 19, 23, 55, and 70) or 6 (spot 19) (Fig. 5, Table 2).

4.

Discussion

Flooding has a major negative impact on soybean seedling growth. It has been reported that the effects of flooding stress on soybean seedlings are manifested in every aspect of

Table 2 – Recovered proteins of soybean roots during recovery stage after flooding stress. Spot no. a

a

Actin isoform B S-adenosylmethionine synthase Adenosine kinase Short-chain dehydrogenase/reductase SDR 1-Cys peroxiredoxin Phosphatidylethanolamine binding protein Peptidyl-prolyl cis-trans isomerase 1 Heat shock protein 70 ND Serine hydroxymethyltransferase Sucrose binding protein Alcohol dehydrogenase-1F Phosphoglycerate kinase Fructikinase2 dTDP-glucose 4-6 dehydratase Ni-binding urease accessory protein Seed maturation protein PM 34 Annexin Cysteine proteinase inhibitor Glycinin G1 Nucleoside diphosphate kinase

Accession no b

Score c

BAA89214.1 A4ULF8.1 XP_002531678.1 ABD32398.1 ACF06438.1 XP_002530493.1 Q8W171.1 ACJ11741.1 AAZ67146.1 Q04672.1 CAA80691.1 AAF85975.1 AAQ10000.1 XP_002531297.1 AAD44338.1 AAF89645.1 XP_002513910.1 BAA19608.1 P04776.2 ssss

Cov. (%) d

M. P. e

Blast score f

363 920 72 128 513 279 156 3371

55 72 41 29 85 38 77 36

14 20 13 6 16 6 9 25

853 182 413 380 2076 103 354 843 718 235 1558 691

41 16 13 19 50 16 25 57 42 31 22 46

21 6 5 6 19 5 7 14 12 8 15 9

Ratio i

Mr (kDa)/pI

p value j

Theo. g

Exp. h

1DFE/Control

FWR/CWR

1DFE

FWR

756 744 584 509 348 246 321 1134

41.9/5.31 43.4/5.50 38.1/5.29 32.0/7.01 24.5/6.44 18.6 /4.92 18.4/8.70 71.8/5.05

859 723 734 735 558 734 489 407 523 417 880 290

56.3/8.35 58.4/6.08 41.6/5.97 42.3/6.28 35.5/5.29 42.9/6.11 31.6/5.99 32.1/6.38 36.07/6.48 27.7/6.57 56.2/5.89 16.3/6.30

42.2/5.0 39.4/4.9 38.0/4.9 37.0/6.3 26.9/6.9 22.3/4.9 18.9/9.0 76.3/4.8 50.1/5.0 47.7/7.3 44.1/8.0 35.2/5.7 33.9/4.7 33.9/4.7 32.8/5.7 32.8/5.7 28.0/5.8 28.3/6.3 27.5/5.8 22.3/9.2 20.2/5.7

2.73 3.63 0.77 0.76 2.04 2.93 3.46 0.63 0.94 1.36 0.71 0.80 0.59 2.49 3.72 0.61 1.03 1.61 0.76 1.03 0.66

1.70 2.57 0.08 1.679 3.27 3.20 1.44 78.18 0.63 2.084 3.53 9.97 2.59 4.49 1.85 12.68 2.68 0.58 1.39 6.31 1.36

0.036* 0.034* 0.041* 0.157 0.048* 0.033* 0.029* 0.096 0.236 0.561 0.631 0.786 0.295 0.089 0.008* 0.215 0.983 0.44 0.800 0.702 0.171

0.021 0.024 0.001 0.006 0.001 0.029 0.001 0.002 0.017 0.013 0.002 0.006 0.007 0.03 0.012 0.011 0.018 0.015 0.010 0.010 0.021

Func k

Cluster l

CellStr Met Ene Met Dis/Def Met Met ProtDes

5 5 2 4 5 6 5

ProtDes ProtDes Met Ene Met Met ProtDes UnCl ProtDes ProtDes ProtDes Met

Spot no., spot number as given in Figs. 5 and 6. Accession no., accession number according to the NCBI database. c Score, ions score of identified protein using soybean genome sequence databases. d Cov., sequence coverage, the proteins with less than 14% sequence coverage was excluded from the result. e M.P., number of query matched peptides, the proteins with ≥ 5 matched peptides were considered. f Blast score, The score of the high-scoring segment pair (HSP) from that database sequence. g Theo., theoretical; Mr, molecular weight; pI, isoelectric point. h Exp., experimental. i Ratio, the ratio of change in abundance of protein spots compared to the control analyzed by LSD test. j p value, indicates the significance of up- or down-regulation of spots according to the t-test through analysis of variance. Asterisks indicate p value was less than 0.05. k Func, Function, category using functional classification: ProtDes, protein destination/storage; Met, metabolism; Dis/Def, disease/defense; CellStr, cell structure; Ene, Energy; UnCl, unclear- and unclassified. l Cluster, belonging cluster which were divided in clustering analysis (Fig. 5). b

J O U RN A L OF P ROT EO M IC S 7 5 ( 2 01 2 ) 8 7 8 –8 93

19 23 31 33 55 65 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84

Homologous protein

889

890

J O U RN A L OF P ROTE O M IC S 7 5 ( 2 01 2 ) 8 7 8 –89 3

500

** **

Relative intensity

400

** *

300

200

* **

**

*

** *

19

33

*

*

*

*

* 0

*

*

*

*

100

55

65

70

72

73

74

100

2500

** 2000

Relative intensity

*

**

80

* 1500

60

1000

40

75

76

77

78

80

81

82

84

CWR FWR Control 1DFE

* *

* *

500 0

20

** 71

79

83

0

23

31

Spot number Fig. 7 – Relative expression of protein spots that are differentially expressed in soybean roots following FWR treatment. The relative intensities of protein spots that were differentially expressed in FWR plants compared to CWR plants were quantified using PDQuest software. Results are presented as the mean ± SE of the relative protein intensity for gels from 3 biological replicates. Asterisks indicate significant differences between control and flooding-treated samples (*P < 0.05, **P < 0.01). White, black, diagonal brick, and dotted bars indicate CWR, FWR, control, and 1DFE, respectively. Spot numbers are the same as in Fig. 6.

growth when plants are flooded for longer than 1 day [14,15]. Consistent with these previous reports, identical negative effects such as suppression of hypocotyls and roots elongation, of hypocotyl pigmentation, of fresh weight increase were observed in the present study (Fig. 1B). Although proteomic changes in soybean seedlings in response to flooding have been demonstrated, proteomic changes occurring during the recovery processes after flooding have not been examined. This study was therefore conducted to characterize the proteomic changes that occur in the roots of soybean seedlings during the recovery period after flooding. Comparative analyses of FWR/CWR and 1DFE/control identified 7 proteins that the protein amounts did not return to control levels after the recovery period (Fig. 6, Table 2). It suggests that the 7 proteins differentially express and involve in the recovery process. The expression of 6 of these proteins (actin isoform B, S-adenosylmethionine synthase, 1-Cys peroxiredoxin, phosphatidylethanolamine binding protein, peptidyl-prolyl cis-trans isomerase 1, and dTDP-glucose 4-6 dehydratase) increased during recovery, while expression of the other decreased. The upregulated proteins may thus be

required for recovery from flooding-induced effects. Actin is a component of the actin cytoskeleton. Modification of the actin cytoskeleton is necessary for building, modifying, or expanding cell walls [30]. The increased expression of actin during the post-flooding recovery period might be involved in cell wall expansion for root elongation. However, the levels of other isoforms of actin decreased (Table 1), suggesting that these actins may play different roles in the roots of floodingtreated soybean seedlings. S-adenosylmethionine synthase catalyzes S-adenosylmethionine synthesis from methionine and ATP. S-adenosylmethionine functions as a primary methyl-group donor and as a precursor for metabolites such as ethylene, polyamines, and vitamin B1 [31]. Previous proteomic studies demonstrated that the expression of Sadenosylmethionine synthases in soybean seedlings decreases significantly during flooding [14,17]. The increase in S-adenosylmethionine synthase expression during postflooding recovery may be necessary to produce metabolites required to overcome the effects of flooding. The 1-Cys peroxiredoxin enzyme has antioxidant activity and functions in the removal of hydrogen peroxide. This protein has been shown

J O U RN A L OF P ROT EO M IC S 7 5 ( 2 01 2 ) 8 7 8 –8 93

to be expressed at greater levels during flooding [14]. It has also been reported that the increased level of 1-Cys peroxiredoxin in flooded seedlings is the result of a delay in protein degradation [32]. The increase observed here during the post-flooding recovery period thus may also be the result of a delay in protein degradation. Phosphatidylethanolamine binding proteins are known as PEBPs, and their genes are part of a family that includes flowering-related genes such as FLOWERING LOCUS T, MOTHER OF FT AND TFL1, and TERMINAL FLOWER 1. In general, PEBP genes seem to regulate various signaling pathways to control growth and differentiation [33]. The expression of phosphatidylethanolamine binding protein was found to be decreased in 3DFE samples (Table 1). The molecular weight of the downregulated protein was lower than that of the upregulated form, indicating that the downregulated protein may have been degraded or processed. Therefore, the proteins were upregulated during the recovery stage and might function as root growth signaling regulators in flooding-treated soybean seedlings. Proteins possessing peptidyl-prolyl cis-trans isomerase activity are known as immunophilins, and these proteins are involved in directing protein folding [34]. During post-flooding recovery, de novo protein synthesis may be activated, thereby increasing the demand for synthesis of immunophilins. The conversion of dTDP-D-glucose into dTDP-4-keto-6-deoxy-Dglucose, a key intermediate for most deoxysugars, is catalyzed by dTDP-glucose 4-6 dehydratase [35]. In plants, deoxysugars such as deoxyribose, fucose, and rhamnose are involved in a number of critical cellular processes. For example, nucleotide sugars such as UDP-rhamnose and GDP-fucose are intermediates in cell wall biosynthesis [36]. The production of deoxysugars through the dTDP-glucose 4-6 dehydratase reaction might be required for cell wall biosynthesis that takes place during recovery from flooding. Taken together with previous reports, the results of this study suggest that regulating root growth through cell wall expansion, including cell wall biosynthesis and the synthesis of S-adenosylmethioninerelated metabolites, may be necessary for the recovery process in flooding-stressed soybean roots. Comparative analyses of 2-DE gels of root proteome samples from control and 3DFE seedlings resulted in the detection of 70 differentially expressed protein spots, from which 80 proteins were identified (Figs. 2 and 3; Table 1). Among the identified proteins, those involved in protein destination/storage and metabolism were the most abundant (Fig. 4A). Proteins categorized as being involved in protein destination/storage included proteins involved in proteolysis and protein folding, such as proteasome subunits and chaperoins. Metabolism-related proteins included enzymes involved in the metabolism of molecules derived from amino acids and sugars. Furthermore, 50% of the differentially expressed proteins were predicted to be cytoplasmic (Fig. 4B). These results suggest that flooding has a significant impact on protein turnover and that post-flooding changes in metabolism occurring in the cytoplasm may be required for recovery. Subsequent expression profile-based clustering analysis of proteins from samples of 1, 2, and 3DFE plants led to grouping of the protein spots into 7 clusters (Fig. 5). Clusters 5, 6, and 7 included those proteins that were expressed at significantly higher levels during the post-flooding recovery period after more than

891

1 day of flooding. However, those proteins in the clusters 5, 6 and 7 include at least two kinds of proteins. These clusters included well-known flooding-inducible proteins, such as alcohol dehydrogenase and pyruvate decarboxylase. Meanwhile, storage proteins such as Kunitz trypsin protease inhibitor, globulin-like protein and 1-Cys peroxiredoxins were also included in those clusters. Although these proteins were higher level during post-flooding recovery period, these proteins should be distinguished based on the protein information. Described above, in flooding condition, increased level of some kinds of storage proteins is the result of a delay in protein degradation [32] (schematic diagram shown in Supplementary Fig. 2). The increase of storage proteins is probably not due to the recovery. The increase in the expression of proteins except for the storage proteins during flooding may thus be sustained during the post-flooding recovery period, suggesting that these proteins might function in both flooding-response and recovery processes. Cluster 3 included proteins for which expression was significantly decreased during the post-flooding recovery period after 3 days of flooding treatment. The expression of many proteins included in cluster 4 was significantly increased during the post-flooding recovery period after 3 days of flooding treatment. The expression profiles of clusters 3 and 4 imply that more than 3 days of flooding has a more severe impact on soybean root cells. It has been demonstrated that 3 days of flooding causes significant growth suppression as measured by shoot length and the fresh weight of soybean seedlings [15]. The suppression of growth may represent negative effects resulting from the severe conditions. Even in the post-flooding recovery stage, soybean seedling root cells may require further metabolic changes to cope with the deleterious conditions resulting from more than 3 days of flooding. V-H(+)-ATPase subunit A and annexin are among the proteins in cluster 3. V-H(+)-ATPase subunit A is a subunit of the vacuolar membrane proton-pump and plays a role in pH-homeostasis [37]. Annexin is a multifunctional lipidbinding protein that regulates membrane dynamics, redox reactions, reactive oxygen species scavenging, ion transport, and calcium signaling [38]. Furthermore, annexins have been detected in the root elongation zone of Arabidopsis[39] and maize [40]. These findings imply that decreases in the expression of V-H(+)-ATPase and annexin downregulate ion transport and influence root elongation of soybean in the postflooding recovery process. Cluster 4 includes TCP-1/cpn60 chaperonin family protein, copper amine oxidase, and membrane attack complex component/perforin/complement C9. TCP-1/cpn60 chaperonin family protein is a chaperonin that contains a TCP-1 domain that assists in the folding of cytoskeletal proteins such as tublin and actin, and consequently maintains proper organization of the cytoskeleton [41]. Copper amine oxidase catalyzes the oxidative de-amination of polyamines, which are ubiquitous compounds essential for cell growth and proliferation [42]. Additionally, it has been suggested that H2O2 derived from the enzyme reaction is involved in signaling related to defense responses, hypersensitive response cell death, developmental programmed cell death, and root hair cell expansion. The isoform of copper amine oxidase identified in this

892

J O U RN A L OF P ROTE O M IC S 7 5 ( 2 01 2 ) 8 7 8 –89 3

Flooding duration

Responses at recovery Upregulated

Downregulated

fermentation protein folding cell wall biosynthesis cell expansion cytoskeletal organization

amino acid metabolism proteolysis glycolysis

programmed cell death

ion transport

1 2

3

Fig. 8 – Schematic diagram of hypothetical model of recovery responses after flooding period. Upregulated and downregulated metabolisms were listed in the boxes. The upper sides in the boxes indicate common responses among 1, 2 and 3DFE. The lower responses indicate specific responses in 3DFE.

study corresponded to the GmCuAO1 form identified by Delis et al. [43]. They reported that the GmCuAO1 gene and protein are expressed in roots and hypocotyls, and GmCuAO1 expression correlated with cell expansion in the fast-growing root and hypocotyl tissues. Membrane attack complex component/perforin/complement C9 (MACPF C9) is a member of the MACPF protein family, which is composed of pore-forming proteins involved in innate immunity in animals [44,45]. The C9 component is required for membrane insertion and pore formation with lytic activity [46]. Reports suggest that the MACPF domain-containing proteins CAD1 and NSL1 are negative regulators of the salicylic acid-mediated pathway of programmed cell death important in plant immunity [47,48]. The identification of these proteins in this study suggests that cytoskeletal organization processes, cell expansion, and programmed cell death are activated during the recovery period in the root cells of soybean seedlings that have undergone prolonged flooding. In concluding the comparative analysis of proteomic responses among 1, 2 and 3DFE, hypothetical schematic diagram was drawn (Fig. 8). Upregulation of fermentation, sugar metabolism, protein folding, cell wall biosynthesis, cell expansion related proteins were common responses among 1, 2 and 3 DFE. Meanwhile, downregulation of amino acid metabolism, proteolysis and glycolysis were common responses among 1, 2 and 3 DFE. Upregulation of programmed cell death and downregulation of ion transport may be occurred by experience of more than 3 days of flooding, suggesting that these responses may be involved in the occurrence of injurious symptom after flooding stress such as delayed growth. This study identified a number of proteins that may be involved in flooding recovery processes in the roots of soybean seedlings. The observed changes in protein expression suggest that the regulation of root growth through cell wall modification and the synthesis of S-adenosylmethionine-related metabolites are important components of the recovery process. Furthermore, the results presented here also indicate that prolonged flooding affects protein expression even during the postflooding recovery stage. Changes in protein expression in root cells during post-flooding recovery induce a downregulation of ion transport and an upregulation of cytoskeletal reorganization processes, cell expansion, and programmed cell death. These results suggest that alteration of the root cell structure through modulation of cell wall metabolism and reorganization of the cytoskeleton likely play an important role in the recovery processes induced by flooding injury.

Supplementary materials related to this article can be found online at doi:10.1016/j.jprot.2011.10.002.

Acknowledgment The authors thank Dr. Mohammad Zaman Nouri, Dr. Keito Nishazawa and Dr. Yuki Yanagawa for their valuable discussion. This work was supported by the grants from National Agriculture and Food Research Organization, Japan.

REFERENCES [1] Bhushan D, Pandey A, Choudhary MK, Datta A, Chakraborty S. Comparative proteomic analysis of differentially expressed proteins in chickpea extracellular matrix during dehydration stress. Mol Cell Proteomics 2007;6:1868–84. [2] Graham PH, Vance CP. Legumes: importance and constraints to greater use. Plant Physiol 2003;131:872–7. [3] Schmutz J, Cannon SB, Schlueter J, Ma J, Mitros T, Nelson W, et al. Genome sequence of the palaeopolyploid soybean. Nature 2010;463:178–83. [4] Kim MY, Lee S, Van K, Kim TH, Jeong SC, Choi IY, et al. Whole-genome sequencing and intensive analysis of the undomesticated soybean (Glycine soja Sieb. and Zucc.) genome. Proc Natl Acad Sci USA 2010;107:22032–7. [5] Tran LS, Mochida K. Identification and prediction of abiotic stress responsive transcription factors involved in abiotic stress signaling in soybean. Plant Signal Behav 2010;5:255–7. [6] Linkemer G, Board JE, Musgrave ME. Waterlogging effect on growth and yield components of late-planted soybean. Crop Sci 1998;39:1576–84. [7] Sullivan M, VanToai TT, Fausey N, Beuerlein J, Parkinson R, Soboyejo A. Evaluating on-farm flooding impacts on soybean. Crop Sci 2001;41:93–100. [8] Jackson MB, Ishizawa K, Ito O. Evolution and mechanisms of plant tolerance to flooding stress. Ann Bot 2009;103:137–42. [9] Voesenek LACJ, Colmer TD, Pierik R, Millenaar FF, Peeters AJM. How plants cope with complete submergence. New Phytol 2006;170:213–26. [10] Colmer TD, Voesenek LACJ. Flooding tolerance: suites of plant traits in variable environments. Funct Plant Biol 2009;36:665–81. [11] Jung G, Matsunami T, Oki Y, Kokubun M. Effects of waterlogging on nitrogen fixation and photosynthesis in supernodulating soybean cultivar kanto 100. Plant Product Sci 2008;11:291–7. [12] Bacanamwo M, Purcell LC. Soybean dry matter and N accumulation responses to flooding stress, N sources and hypoxia. J Exp Bot 1999;50:689–96.

J O U RN A L OF P ROT EO M IC S 7 5 ( 2 01 2 ) 8 7 8 –8 93

[13] Oosterhuis DM, Scott HD, Hampton RE, Wullschleger SD. Physiological response of two soybean [Glycine max L. Merr] cultivars to short-term flooding. Environ Exp Bot 1990;30:85–92. [14] Hashiguchi A, Sakata K, Komatsu S. Proteome analysis of early-stage soybean seedlings under flooding stress. J Proteome Res 2009;8:2058–69. [15] Komatsu S, Kobayashi Y, Nishizawa K, Nanjo Y, Furukawa K. Comparative proteomics analysis of differentially expressed proteins in soybean cell wall during flooding stress. Amino Acids 2010;39:1435–49. [16] Komatsu S, Yamamoto R, Nanjo Y, Mikami Y, Yunokawa H, Sakata K. A comprehensive analysis of the soybean genes and proteins expressed under flooding stress using transcriptome and proteome techniques. J Proteome Res 2009;8:4766–78. [17] Nanjo Y, Skultety L, Ashraf Y, Komatsu S. Comparative proteomic analysis of early-stage soybean seedlings responses to flooding by using gel and gel-free techniques. J Proteome Res 2010;9:3989–4002. [18] Shi F, Yamamoto R, Shimamura S, Hiraga S, Nakayama N, Nakamura T, et al. Cytosolic ascorbate peroxidase 2 (cAPX 2) is involved in the soybean response to flooding. Phytochemistry 2008;69:1295–303. [19] Komatsu S, Wada T, Yann A, Nouri MZ, Nanjo Y, Nakayama N, et al. Analysis of plasma membrane proteome in soybean and application to flooding stress response. J Proteome Res 2009;8:4487–99. [20] Komatsu S, Sugimoto T, Hoshino T, Nanjo Y, Furukawa K. Identification of flooding stress responsible cascades in root and hypocotyl of soybean using proteome analysis. Amino Acids 2010;38:729–38. [21] Perata P, Armstrong W, Voesenek LA. Plants and flooding stress. New Phytol 2011;190:269–73. [22] Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Anal Biochem 1976;72:248–54. [23] Bevan M, Bancroft I, Bent E, Love K, Goodman H, Dean C, et al. Analysis of 1.9 Mb of contiguous sequence from chromosome 4 of Arabidopsis thaliana. Nature 1998;391:485–8. [24] Horton P, Park KJ, Obayashi T, Fujita N, Harada H, Adams-Collier CJ, et al. WoLF PSORT: protein localization predictor. Nucl Acids Res 2007;35:585–7. [25] Bhasin M, Raghava GP. ESLpred: SVM-based method for subcellular localization of eukaryotic proteins using dipeptide composition and PSI-BLAST. Nucl Acids Res 2004;32:414–9. [26] Chen H, Huang N, Sun Z. SubLoc: a server/client suite for protein subcellular location based on SOAP. Bioinformatics 2006;22:376–7. [27] de Hoon MJL, Imoto S, Nolan J, Miyano S. Open source clustering software. Bioinformatics 2004;20:1453–4. [28] Saldanha AJ. Java Treeview-extensible visualization of microarray data. Bioinformatics 2004;20:3246–8. [29] Ihaka R, Gentleman R. R: a language for data analysis and graphics. J Comput Graph Stat 1996;5:299–314. [30] Wasteneys GO, Galway ME. Remodeling the cytoskeleton for growth and form: an overview with some new views. Annu Rev Plant Biol 2003;54:691–722. [31] Hesse H, Kreft O, Maimann S, Zeh M, Hoefgen R. Current understanding of the regulation of methionine biosynthesis in plants. J Exp Bot 2004;55:1799–808.

893

[32] Nishizawa K, Komatsu S. Characteristics of soybean 1-Cys peroxiredoxin and its behavior in seedlings under flooding stress. Plant Biotechnol 2011;28:83–8. [33] Karlgren A, Gyllenstrand N, Källman T, Sundström JF, Moore D, Lascoux M, Lagercrantz U. Evolution of the PEBP gene family in plants: functional diversification in seed plant evolution. Plant Physiol 2011;156:1967–77. [34] Romano P, Gray J, Horton P, Luanm S. Plant immunophilins: functional versatility beyond protein maturation. New Phytol 2005;166:753–69. [35] Chen F, Lin L, Wang L, Tan Y, Zhou H, Wang Y, et al. Distribution of dTDP-glucose-4,6-dehydratase gene and diversity of potential glycosylated natural products in marine sediment-derived bacteria. Appl Microbiol Biotechnol 2011;90:1347–59. [36] Seifert GJ. Nucleotide sugar interconversions and cell wall biosynthesis: how to bring the inside to the outside. Curr Opin Plant Biol 2004;7:277–84. [37] Gaxiola RA, Palmgren MG, Schumacher K. Plant proton pumps. FEBS Lett 2007;581:2204–14. [38] Laohavisit A, Davies JM. Annexins. New Phytol 2011;189:40–53. [39] Clark GB, Lee DW, Dauwalder M, Roux SJ. Immunolocalization and histochemical evidence for the association of two different Arabidopsis annexins with secretion during early seedling growth and development. Planta 2005;220:621–31. [40] Bassani M, Neumann PM, Gepstein S. Differential expression profiles of growth-related genes in the elongation zone of maize primary roots. Plant Mol Biol 2004;56:367–80. [41] Himmelspach R, Nick P, Schäfer E, Ehmann B. Developmental and light-dependent changes of the cytosolic chaperonin containing TCP-1 (CCT) subunits in maize seedlings, and the localization in coleoptiles. Plant J 1997;12:1299–310. [42] Cona A, Rea G, Angelini R, Federico R, Tavladoraki P. Functions of amine oxidases in plant development and defence. Trends Plant Sci 2006;11:80–8. [43] Delis C, Dimou M, Flemetakis E, Aivalakis G, Katinakis P. A root- and hypocotyl-specific gene coding for copper-containing amine oxidase is related to cell expansion in soybean seedlings. J Exp Bot 2006;57:101–11. [44] Esser AF. The membrane attack complex of complement. Assembly, structure and cytotoxic activity. Toxicology 1994;87:229–47. [45] Trapani JA, Smyth MJ. Functional significance of the perforin/ granzyme cell death pathway. Nat Rev Immunol 2002;2: 735–47. [46] Rosado CJ, Kondos S, Bull TE, Kuiper MJ, Law RHP, Buckle AM, et al. The MACPF/CDC family of pore-forming toxins. Cell Microbiol 2008;10:1765–74. [47] Morita-Yamamuro C, Tsutsui T, Sato M, Yoshioka H, Tamaoki M, Ogawa D, et al. The Arabidopsis gene CAD1 controls programmed cell death in the plant immune system and encodes a protein containing a MACPF domain. Plant Cell Physiol 2005;46:902–12. [48] Noutoshi Y, Kuromori T, Wada T, Hirayama T, Kamiya A, Imura Y, et al. Loss of NECROTIC SPOTTED LESIONS 1 associates with cell death and defense responses in Arabidopsis thaliana. Plant Mol Biol 2006;62:29–42.