J O U R N A L O F P R O T E O M I C S 7 3 ( 2 0 0 9 ) 3 0 – 4 0

ava i l ab l e a t www.sc i enced i rec t . com

www.e l sev i e r. com/ loca te / j p ro t

Proteomics of Arabidopsis redox proteins in response tomethyl jasmonate

Sophie Alvareza,1, Mengmeng Zhua, Sixue Chena,b,⁎aDepartment of Botany and Zoology, Genetics Institute, The Plant Molecular and Cellular Biology Program, University of Florida, Gainesville,FL 32611, USAbCollege of Life Sciences, Heilongjiang University, Harbin, China

A R T I C L E I N F O

1874-3919/$ – see front matter © 2009 Elsevidoi:10.1016/j.jprot.2009.07.005

⁎ Corresponding author. Genetics Institute, Un8330; fax: +1 352 273 8284.

E-mail address: schen@ufl.edu (S. Chen).1 Current address: Donald Danforth Plant S

A B S T R A C T

Article history:Received 23 April 2009Accepted 15 July 2009

Protein redox regulation is increasingly recognized as an important switch of proteinactivity in yeast, bacteria, mammals and plants. In this study, we identified proteins withpotential thiol switches involved in jasmonate signaling, which is essential for plantdefense. Methyl jasmonate (MeJA) treatment led to enhanced production of hydrogenperoxide in Arabidopsis leaves and roots, indicating in vivo oxidative stress. Withmonobromobimane (mBBr) labeling to capture oxidized sulfhydryl groups and 2D gelseparation, a total of 35 protein spots that displayed significant redox and/or total proteinexpression changes were isolated. Using LC–MS/MS, the proteins in 33 spots were identifiedin both control and MeJA-treated samples. By comparative analysis of mBBr and SyproRubygel images, we were able to determine many proteins that were redox responsive andproteins that displayed abundance changes in response to MeJA. Interestingly, stress anddefense proteins constitute a large group that responded to MeJA. In addition, manycysteine residues involved in the disulfide dynamics were mapped based on tandem MSdata. Identification of redox proteins and their cysteine residues involved in the redoxregulation allows for a deeper understanding of the jasmonate signaling networks.

© 2009 Elsevier B.V. All rights reserved.

Keywords:Arabidopsis2-DEMass SpectrometryMethyl jasmonateRedox proteins

1. Introduction

Plants produce jasmonic acid andmethyl jasmonate (MeJA) inresponse to many abiotic and biotic stresses, particularlypathogen and insect herbivores [1,2]. Jasmonates are planthormones biosynthesized from linolenic acid through theoctadecanoid pathway [3]. They function as signaling mole-cules to activate genes involved in plant defense responses[4,5]. Over the past decades, intensive research has beenfocused on the jasmonate signaling pathway in Arabidopsisand tomato [6–8]. The perception of stress signal, the induction

er B.V. All rights reserved

iversity of Florida, 1376 M

cience Center, St Louis, M

and regulation of jasmonate biosynthesis, and the genesdifferentially expressed by jasmonates have beenwell-studied[3,6,7,9–11]. However, the molecular details of downstreamregulatory proteins and pathways remain to be discovered. Itwas suggested in 1994 that jasmonates could induce oxidativestress in parsley suspension cells [12]. Later, jasmonateinduced hydrogen peroxide (H2O2) accumulationwas observedin the cellwall of tomato plants [13]. H2O2, themost stable formof reactive oxygen species (ROS), is well known to function as asignalingmolecule to activate cellular antioxidantmechanisms,and can be used as an indicator of cellular oxidative stress [14].

.

owry Road, Room 438, Gainesville, FL 32610, USA. Tel.: +1 352 273

O 63132, USA.

31J O U R N A L O F P R O T E O M I C S 7 3 ( 2 0 0 9 ) 3 0 – 4 0

For instance, the production of H2O2 in the cell wall was shownto induce the activity of polygalacturonases [15]. The resultindicates that H2O2 can function as a secondary messengerand play a role in MeJA mediated metabolic processes [16].

Cellular protein redox cycling is regulated by several well-known systems such as the NADPH/ferredoxin–thioredoxinand glutathione–glutaredoxin systems [17,18]. Whereas thethioredoxin and glutaredoxin systems seem to function mainlyin the reduction of disulfide bonds, other thiol/disulfide contain-ing proteins (e.g., oxidoreductases) and ROS act to oxidize thiolgroups.Although it isknownthatH2O2can regulate theactivitiesof certain plant proteins, little is known about the molecularmechanisms underlying the regulation. It was suggested thatH2O2 mediated redox state change of protein thiols may play arole in oxidative stress signaling [19]. Thiol based redoxproteome is relatively complex. Under oxidative stress, cysteinefree thiols can be reversibly oxidized to form disulfides, sulfenicacids, S-nitrosylated andS-glutathiolatedadducts or irreversiblyoxidized to form sulfinic acids and sulfonic acids [17,20].

To date, many redox proteomics studies have been focusedon the identification of direct protein targets of thioredoxinandglutaredoxin function [18,21,22].Major approaches includeaffinity purification with mutant thioredoxin affinity column,2D gel separation of proteins with thiol groups fluorescentlylabeledwith a fluorescent dyemonobromobimane (mBBr), anddiagonal gel electrophoresis [22]. Mass spectrometry (MS) isused for protein identification in all the approaches. Progresshas been made in identifying disulfide proteins [23] andthioredoxin regulated proteins [21]. However, the cysteinesinvolved were largely unknown and whether the redoxregulation is a direct effect of sensing cellular redox state wasnot clear. Characterization of redox proteins in plant jasmo-nate signal transduction has not been reported. Here wepresent the identification of changes in protein redox regula-tion in response to oxidative stress induced by MeJA in Arabi-dopsis shoots and roots using a 2DE-based proteomicsapproach [24]. The fluorescent mBBr was used to label thethiol groups of proteins obtained after alkylation of free thiolgroups and reduction of reversible oxidized thiol groups. Thelabeled proteins were separated on 2D gels, followed byvisualization of mBBr-labeled proteins. Total proteins werestained using SyproRuby to determine protein expressionchanges and to compare with the mBBr signal indicative ofoxidation of protein thiol groups. A comparative proteomicmap of potential redox proteins regulated by MeJA wasestablished and the cysteine residues involved in the oxidativeregulation were localized.

2. Experimental procedures

2.1. Plant growth and treatment

Seeds from Arabidopsis thaliana ecotype Col-0 were obtainedfrom the Arabidopsis Biological Resource Center (Stock num-ber: CS3879). The seeds were sterilized in 50% bleach for10 min, and washed four times with sterilized water. Theywere then germinated on a half strength Murashige-Skoogagar medium containing 1% sucrose, and transferred to agrowth chamber under a photosynthetic flux of 140 µmol

photonsm−2 s−1 with a photoperiod of 16 h at 24 °C and 20 °C atnight for nine days. MeJA was applied evenly to the agarmedium at a final concentration of 500 μM. After 24 h ofincubation, the seedlingswere dissected into shoots and roots,weighed and immediately frozen in liquid nitrogen. Forcontrol samples, all the steps were the same except thatMeJA was replaced with 0.004% (v/v) ethanol in water. For allthe experiments, at least three independent replicates wereconducted.

2.2. Hydrogen peroxide detection

Hydrogen peroxide was visualized by staining with 3,3-diaminobenzidine (DAB). Once DAB encounters H2O2, itundergoes oxidative polymerization to produce a dark-brown precipitate [25]. After MeJA treatment, some seedlingswere collected and immediately incubated with 1 mg/ml DABfor 2 h in the dark. Then shoots were separated from the roots.The leaf stainwas fixed in 95% ethanol for 30min, and the rootstain was fixed by rinsing with distilled water. The leaves androots were imaged using a Leica DMRE microscope coupled toa digital camera and a computer with Leica Qwin Imagingsoftware (Leica Ltd., USA).

2.3. Protein extraction and mBBr labeling of protein thiols

Shoot and root samples were ground in liquid nitrogen to finepowders. Onemilliliter 10% trichloroacetic acid in acetonewasadded, followed by incubation on ice for 30 min. The sampleswere centrifuged at 14,000 rpm for 10 min, the supernatantwas removed and 1 ml cold acetone was added to the pellet.After centrifugation, the pellet was washed twice in 80%acetone, and resuspended in a 2D gel sample solubilizationsolution containing 5 M urea, 2 M thiourea, 2% CHAPS, 2% SB3–10, 40 mM Tris, 0.2% Bio-Lyte 3/10 ampholyte (Bio-Rad, CA,USA). Protein amounts were quantified using a CB-X ProteinAssay kit (GenoTech, MO, USA).

Two hundredmicroliters of alkylation buffer (100 mM Tris–HClpH7.5, 200mM iodoacetamide)were added to 200 μL (about250 μg) protein sample. Alkylation was performed at 75 °C for5 min, then at 37 °C for 1 h in the dark. Proteins were preci-pitated by adding 1 ml cold 80% acetone at −20 °C for 1 h. Thetubes were centrifuged at 14,000 rpm at 4 °C for 10 min. Afterremoving the supernatant, the pellet was dried briefly, andresuspended in 200 μL of reduction buffer (100mMTris−HCl pH7.5, 10 mM DTT). Reduction was performed at room tempera-ture for 1 h. Labeling of the proteins was performed by adding20 μl of mBBr solution (1 μg/100 μL) to each sample and thesampleswere incubated at room temperature for 30min in thedark. The labeling reactionwas terminated by adding 10 μl 10%SDS. Proteins were then precipitated by incubating with 1 mlcold acetone at −20 °C overnight. After centrifugation, theprotein pellet was resuspended in a destreak rehydrationbuffer (GE Healthcare, NJ, USA).

2.4. Two dimensional gel electrophoresis

For shoot samples, 1 mg protein was loaded onto a 24 cm Bio-Rad ReadyStrip (pH range 3–10 NL). The focusing conditionswere: 200 V for 30 min, then ramping to 500 V for 30 min, and

32 J O U R N A L O F P R O T E O M I C S 7 3 ( 2 0 0 9 ) 3 0 – 4 0

finally to 10,000 V for 1 h. The voltagewas held at 10,000 V untilthe total voltage hours reached 85,000. After the first dimen-sion, the strips were equilibrated and transferred onto 24 cm11% SDS gels. Electrophoresis was run at 15 W for 5 h. For rootsamples, 250 µg protein was loaded onto an 11 cm Bio-RadReadyStrip (pH range 3–10 NL). The isoelectric focusing andSDS gel electrophoresis were performed as previouslydescribed [26]. For each sample, at least three replicate 2-DEexperiments were conducted.

2.5. Gel image analysis

Fluorescence of the thiol groups of mBBr-labeled proteins wascaptured using an ImageQuant Gel Imager with a UV 365 nmlight-box (GE Healthcare, NJ, USA). The gels were then stainedwith SyproRuby following manufacturer's instructions (Invitro-gen, CA, USA). Protein spots stained with SyproRuby werevisualized using a Typhoon 9410 laser scanner (excitation450 nm, emission 650 nm) (GE Healthcare, NJ, USA). Replicategels from control and MeJA samples were analyzed withProgenesis Software (Nonlinear Dynamics, CA, USA) enablingspot detection, quantification and spotmatching across differentgels. Theautomatic spotdetectionandmatchingwas followedbya manual correction. Experimental molecular weights werecalibrated using commercial molecular mass standards run in aseparate marker lane on the 2D gels, and the experimentalisoelectric points were calibrated according to Bio-Rad IPG stripspecifications. After subtracting background with the non-spotmode, spot volumes were normalized by dividing each spotvolume by the total volume of all spots present in all gels. Thenormalized spot volumes were used to determine the quantita-tive information of protein redox and expression changes.ANOVA test was used to determine the statistical significance.

2.6. Protein identification using liquid chromatography–tandem MS (LC–MS/MS)

Spots from 2D gels were excised and digested with trypsin aspreviously described [27]. Ten microliters of protein digestswere injected onto a capillary trap column (PepMap, Dionex,USA) and desalted for 5 min before being loaded onto a C18PepMap nanoliter-flow column [25]. The elution gradient ofthe column started at 97% solvent A (0.1% v/v acetic acid, 3%v/v acetonitrile), 3% solvent B (0.1% v/v acetic acid, 96.9% v/vacetonitrile) and finished at 40% solvent A, 60% solvent B for20 min. Tandem MS analysis was carried out on a hybridquadrupole-time of flight mass spectrometer (QSTAR XL,Applied Biosystems, MA, USA). The focusing potential andion spray voltages were set to 275 V and 2600 V, respectively.The information-dependent data acquisition (IDA) wasemployed in which a survey scan from m/z 400–1500 wasacquired followed by collision induced dissociation of threemost intense ions. Survey scan and each MS/MS spectrum inan IDA cycle were accumulated for 1 s and 3 s, respectively.The acquired mass spectra were searched against IPI Arabi-dopsis database (version June 22, 2007) using Mascot searchengine (http://www.matrixscience.com). The following para-meters were selected: tryptic peptides with ≤1 missedcleavage site, mass tolerance of precursor ion and MS/MSion of 0.3 Da, fixed carbamidomethylation of cysteine, variable

methionine oxidation, and variable cysteine modification bymBBr. Unambiguous identification was judged by the numberof peptide sequences, sequence coverage, Mascot score, andthe quality of tandem MS spectra [26–28].

3. Results

3.1. H2O2production in leavesand roots afterMeJA treatment

We have tested different MeJA concentrations (50, 250 and500 µM) and treatment conditions. Treatment of Arabidopsisseedlings with 500 µM MeJA for 24 h led to marked productionof H2O2, which is indicative of oxidative stress and displace-ment of cellular redox equilibrium (Fig. 1). No obvious toxiceffect was observed on the treated plants with 500 µM MeJA, aconcentration that has been used in several previous studies[29–32]. H2O2 production in leaves and roots was visualizedusing DAB staining [25]. Restricted spots of staining wereobserved in MeJA-treated leaves, indicating H2O2 productionand accumulation in specific compartments (Fig. 1A, B). Inroots, root tip was constitutively stained in both control andMeJA-treated roots, and the intensity of the stain decreasedgradually towards the elongation zone. Across the entire roots,MeJA-treated roots displayed amuch darker stain than controlroots (Fig. 1C, D). A closer inspection of the roots revealed thatthe accumulation of the stain seemed to be mainly in thevascular cells and close to the plasma membrane/cell wallregion (Fig. 1E, F).

3.2. Thiol redox and protein abundance changes in responseto MeJA

Proteomic maps of thiol redox proteins and the total solubleproteins from control and MeJA-treated plants were comparedusing 2D gel image analysis. Altogether 27 spots from shootsamples and eight spots from root samples showed significantchanges in spot volumes (ANOVA test p<0.05) in response toMeJA. Images in Fig. 2 were included to show the positions ofthese spots on the 2D gels. Please refer to Supplemental Figs. 1and 2 for representative SyproRuby and mBBr gel images. TheVenn diagrams in Fig. 3 summarized the number of the spotsthat increased or decreased in intensity on replicate mBBrimages and/or SyproRuby images. In shoots, a total of 15 spotsshowed significant increases in spot intensity with mBBrlabeling (e.g., spot 437 in Fig. 4) and 16 (with ten increased andsix decreased) with SyproRuby staining. Four spots showedsignificant differences from both stains, two of which (spots888 and 1255) with similar intensity change in both stains,suggesting that the change associated with mBBr labeling wascaused by the change in abundance of the protein rather thana redox status modification. Two spots (spots 1278 and 1420)exhibited opposite intensity changes when the mBBr signalwas compared to the SyproRuby signal (Figs. 3, 4; Supple-mental Table 1). In roots, the intensities of three spots weresignificantly altered (spots 703 and 814 increased, and spot 323decreased) with mBBr labeling while six spots decreased inabundance with SyproRuby staining (Figs. 3, 4; SupplementalTable 1). Only spot 323 showed a decrease in abundance withboth mBBr and SyproRuby staining (Supplemental Table 1).

Fig. 2 – 2Dgel imagesofproteins stainedbySyproRuby (A, shootand B, root) showing the spots (27 from shoot and eight fromroot) with significant changes in spot intensity in response toMeJA treatment.

Fig. 1 – Hydrogen peroxide production in leaves (A–B) and roots (C–F) of 10 day old Arabidopsis seedlings. H2O2 production wasvisualized by incubation in DAB for 2 h. (A, C, E) control, (B, D, F) 24 h MeJA treatment.

33J O U R N A L O F P R O T E O M I C S 7 3 ( 2 0 0 9 ) 3 0 – 4 0

3.3. Identification of MeJA responsive proteins with redoxand/or abundance changes

Redox-regulated proteins as well as proteins with expressionchanges after MeJA treatment were identified using LC–MS/MS.In this study,protein spots frombothcontrol andMeJAgelswereexcised for identification to contrast and confirm the proteinsthat actually showed significant changes in response to MeJA(Fig. 4). For shoot samples, 25 out of 27 spots were confidentlyidentified. In root samples, a total of eight spots were pickedfrom the control and seven from theMeJA-treated (one spotwasnot detectable in the MeJA samples). All eight spots yielded

Fig. 3 – Venn diagrams showing the number of spots thatincreased or decreased in abundance after MeJA treatment ofshoot and root samples as visualized by mBBr labeling andSyproRuby staining. The numbers in the middle indicatecommon spots found on mBBr images and SyproRubyimages. In shoots, 4 spots were common, all with increasedabundance on mBBr images, but with 2 increased and 2decreased on SyproRuby images. In roots, 1 spot wascommon, it showed decreased abundance on both mBBrimages and SyproRuby images.

Fig. 4 – Examples of protein spots from leaves (# 437, 1255 and 1278) and roots (# 703 and 814) showing quantitative differencesbetween control (a and c) andMeJA-treated (b and d) plants aftermBBr labeling (a and b) and SyproRuby staining (c and d). Spotsidentified by mass spectrometry were labeled with the same numbers as in Tables 1, 2 and Supplemental Table 1.

34 J O U R N A L O F P R O T E O M I C S 7 3 ( 2 0 0 9 ) 3 0 – 4 0

confidentprotein identifications. Several spotsshowedmultipleprotein identities (Supplemental Table 1). Although highresolution 2D gels were used, it is not uncommon that multipleproteins were identified in a single gel spot [28]. When a spotcontains multiple proteins, the change of the spot intensitycaused by experimental treatment usually cannot be assignedto a particular protein. This weakness is often regarded as adisadvantage of the 2-DE technology [28]. In this study, based onthe presence of a spot on mBBr image and/or on SyproRubyimage, and the quantitative responses of the spot to MeJA, wewere able to relate the redox and/or the abundance changesback to the protein(s) affected for approximately half of thespots identified. Table 1 summarizes the identifiedproteins thatshowed redox or protein expression changes in response toMeJA treatment. For example, beta-glucosidase in spot 437 didnot show significant abundance changes in response to MeJA,but showed more than two-fold increase in mBBr signal afterMeJA treatment. This indicates that beta-glucosidase wasoxidized in response to MeJA (Fig. 4; Supplemental Table 1).When a spot contained more than one protein, we utilized twocriteria to determine proteins undergoing redox changes and/ornet synthesis, i.e., the cysteine content in each of the identifiedproteins and the peptide number (spectra counting). Forexample, spot 1278 showed significant decrease in abundanceand over two-fold increase inmBBr signal afterMeJA treatment.Three proteins were identified in this spot, i.e., photosystem Ilight harvesting complex protein, superoxide dismutase, andribulose bisphosphate carboxylase (Rubisco) small chain 1B.Because the first twoproteinsdonothaveany cysteine residues,the increase in mBBr signal (i.e., most oxidized protein) mustcome from Rubisco small chain 1B. Since the first protein wasthe only one not identified in the MeJA-treated sample, it musthave contributed to the decrease of the spot abundance. Usingthis strategy, we were able to identify potential redox proteinsandproteinswith significant abundance changes in response toMeJA (Table 1).

3.4. Identification of cysteine residues involved in redoxregulation

Based on the MS/MS spectra acquired from the gel spots withsignificant changes in response to MeJA, many mBBr-labeledcysteinescouldbe identified (Fig. 5; Table2). These cysteineswereeither oxidized in response to MeJA treatment or originallyinvolved in the formation of structural disulfide bonds and

reversible cysteine adducts. According to the changes of mBBrand SyproRuby intensities and the presence of mBBr-labeledcysteines in the MeJA-treated and/or the control samples, wecould determine whether the cysteines were involved in redoxregulation in response to MeJA. For example, the mBBr-labeledcysteines in the peptides from an isocitrate dehydrogenase wereonly found from the spot in theMeJA-treated samples, but not inthe control samples (Supplemental Table 1). The protein did notshow significant abundance changes after MeJA treatment.Therefore, the cysteines were determined to be redox-regulated(spot 672 inTable1). In contrast, thecysteines identified inmalatedehydrogenase (spot 946 in Supplemental Table 1) were mostlikely not to be involved in redox regulation by MeJA becausethere was no redox change, but significant abundant changecausedby theMeJA treatment (Table1). BasedonMS/MSdataandpredictions from the disulfide bond prediction software DiANNA(http://clavius.bc.edu/~clotelab/DiANNA/), the role of the labeledcysteines in disulfide bond formation can be proposed (Fig. 5).Given input of a protein sequence, the program outputs aprediction of the disulfide connectivity using a neural network-based approach [33]. In addition to intra-disulfide bonds, thecysteine residues might be involved in the formation of mixeddisulfides or intermolecular disulfides [34]. These possibilitiesneed to be discerned with further experiments.

3.5. Functional classification of the proteins showing redoxand/or expression changes

The proteins identified in shoots and roots were classifiedaccording to their biological functions (Table 1). As expected,proteins from the primary metabolism such as photosynthesis,respiration, and amino acid metabolism were identified. Moreinterestingly, proteins involved in sulfur metabolism (e.g.,glutathione S-transferase (GST)), stress and defense (e.g., mosaicdeath 1 (MOD1) and vegetative storage protein), and jasmonatebiosynthesis (allene oxide cyclase 2 (AOC2))were identified in theshoot samples.The largest groupofproteinsbelongs to stressanddefense (Table 1). In roots, proteins found to be differentiallyexpressed were involved in post-transcription process (RNA-binding proteins), signal transduction (nucleoside diphosphatekinase (NDPK1)), and response to stress (protease inhibitor andmajor latex protein). These data point to a clear differencebetween the redox proteomic responses of the two tissues toMeJA. Several proteins were identified to be affected by MeJA attheir redox levels: glutamate-1-semialdehyde 2,1-aminomutase

Table 1 – List of proteins identified as significantly affected in redox status or abundance in response to MeJA treatment.

Spot Organ Protein name Redox Abundance

Amino acid metabolism714 S Glutamate-1-semialdehyde 2,1-aminomutase

ATP sulfurylase 3U

1425 S Aconitase C-terminal domain-containing protein −1.8⁎

Respiration872 S Malate dehydrogenase, cytoplasmic 1 1.8⁎672 S Isocitrate dehydrogenase U814 S Fructose-bisphosphate aldolase −1.5⁎946 S Malate dehydrogenase, chloroplastic precursor 1.8⁎

Photosynthesis1278 S Photosystem I light harvesting complex gene 3 −1.4⁎1278 S Ribulose bisphosphate carboxylase small chain 1B 2.3⁎

Post-transcriptional factors793 R Isoform 1 of Glycine-rich RNA-binding protein 7 −2.6⁎807 R Isoform 1 of Glycine-rich RNA-binding protein 8 −1.7⁎

Stress and defense437 S Beta-glucosidase, PYK10 2.3⁎⁎⁎666 S Coronatine induced 1, jasmonic acid responsive 2 U946 S Mosaic death 1 (MOD1) 1.8⁎1188 S Vegetative storage protein 1 (VSP1) precursor U703 R Trypsin and protease inhibitor family protein 2.1⁎793 R Regulator of ribonuclease-like protein 1 −2.6⁎814 R Major latex protein-related/RNA-binding protein 1 1.8⁎

Secondary metabolism1277 S Allene oxide cyclase 2, chloroplast precursor 5.8⁎

Signaling pathway856 R nucleoside diphosphate kinase 1 −2.6⁎

Sulfur metabolism1258 S Glutathione S-transferase 6, chloroplast precursor U1517 S Glutathione S-transferase (class tau) 20 U

The proteins from shoots (S) and roots (R) were classified according to their biological functions. The asterisks indicate significant changes inredox and/or protein abundance in MeJA-treated samples in comparison with the controls (⁎p<0.05; ⁎⁎⁎p<0.001). Positive numbers stand forreversible oxidation in redox (mBBr) or increase in abundance (SyproRuby), and negative numbers denote reduction or irreversible oxidation(mBBr) or decrease in abundance (SyproRuby). The letter U indicates that the spot is present in only one of the conditions and thus the foldchange could not be calculated.

35J O U R N A L O F P R O T E O M I C S 7 3 ( 2 0 0 9 ) 3 0 – 4 0

1/ATP sulfurylase 3, malate dehydrogenase, isocitrate dehydro-genase, coronatine induced (CORI3), small chain of Rubisco, twoGSTs (6 and 20), and beta-glucosidase (PYK10) in shoots, and aprotease inhibitor protein and amajor latex protein/RNA-bindingprotein in roots. An increase in mBBr labeling indicated anincrease of oxidation in response toMeJA treatment,which led toH2O2 accumulation in shoots and roots (Fig. 1). Eleven proteinswere affected in their total abundance by MeJA with sevendownregulated (aconitase, fructose-bisphosphate aldolase andphotosystem I in shoots, and twoRNA-bindingproteins, inhibitorof ribonuclease protein and nucleoside diphosphate kinase 1(NDPK1) in roots) and four upregulated (chloroplastic malatedehydrogenase, AOC2, MOD1 and vegetative storage protein 1(VSP1) (Table 1). Three proteins, two myrosinase-associatedproteins (spot 323) and a cysteine synthase (spot 888) showedsimilar extents of changes in mBBr labeling and in SyproRubystaining (Supplemental Table 1). It is likely that the change inredox state is due to the change in protein levels. Since other

proteins with cysteines were also identified in the spots, thepossibility of redox regulation of these proteins could not beexcluded.

4. Discussion

4.1. MeJA induced significant protein expression changes

The protein that displayed nearly six-fold increases in abun-dance after MeJA treatment was identified as one of the keyenzymes involved in jasmonate biosynthesis, AOC2. Theinduction of endogenous levels of jasmonic acid (JA) byexogenous application of JA has been observed previously inArabidopsis leaves [35]. The increase in the abundance of AOC2confirmed the effectiveness of the MeJA treatment and thesignificance of the data obtained in this present study. Otherproteins that displayed significant increases in protein levels

Table 2 – Cysteines identified to be redox-regulated by MeJA.

Spot Protein name Redox-regulatedcysteines

Redox-insensitivecysteines

323 Similar to myrosinase-associatedprotein 1

K.FVIQTLAPLGCLPIVR.Q

323 Similar to myrosinase-associatedprotein 2

K.FVIQTLAPLGCLPIVR.Q

323 Peroxidase 32 precursor K.AAVETAC⁎PR.TK.NQC⁎QFIMDR.LR.MKAAVETAC⁎PR.T

437 Beta-glucosidase, PYK10 R.CNNDNGDVAVDFFHR.Y508 Ketol-acid reductoisomerase,

chloroplastK.NISVVAVCPK.G

508 Ribulose bisphosphate carboxylaselarge chain precursor

R.AVYECLR.G

666 Coronatine induced, jasmonicacid responsive 2

K.SFCDR.HK.KLTADDVFMTLGCK.Q

672 Isocitrate dehydrogenase K.CATITPDEGR.VK.SEGGYVWACK.NK.LVPGWTKPICIGR.HR.NILNGTVFREPIICK.N

703 Trypsin and proteaseinhibitor family protein

K.FVFCPR.TR.GGQPCPYDIVQESSEVDEGIPVK.F

714 Glutamate-1-semialdehyde2,1-aminomutase 1, chloroplastprecursor

R.FVNSGTEACMGVLR.L

751 Formate dehydrogenase,mitochondrial

R.LKPFGCNLLYHDR.L

814 Fructose-bisphosphate aldolase R.VLAACYK.A856 nucleoside diphosphate kinase 1 R.GLIGEVICR.F872 Malate dehydrogenase, cytoplasmic K.EFAPSIPEKNISCLTR.L

K.NISCLTR.L888 Porphobilinogen deaminase,

chloroplastK.DEEGNCIFR.G

888 Cysteine synthase, chloroplast/chromoplast precursor

K.GCVASVAAK.LK.LEIMEPCCSVK.D

926 Glutaredoxin-C2 K.TYCPYCVR.V946 Malate dehydrogenase, chloroplast

precursorR.DFTGPSELADCLK.DK.GVAADLSHCNTPSQVR.D

1064 Universal stress protein (USP) familyprotein

K.TDIACLDMLDTGSR.Q

1254 Dehydroascorbate reductase K.AAVGAPDHLGDC⁎PFSQR.A1516 Isoform 3 of Carbonic anhydrase,

chloroplast precursorK.ICLPAK.SK.YMVFACSDSR.VK.VISELGDSAFEDQCGR.CR.VCPSHVLDFQPGDAFVVR.N

The cysteines in the peptides listed in the middle between the two columns were ambiguous cysteines that could not be assigned as redox-regulated or redox-insensitive. The cysteines supported by the crystal structure of the proteins (http://www.rcsb.org/pdb/home/home.do) toform disulfide bonds were highlighted with stars. The MS/MS spectra supporting the identification of the cysteines were included asSupplemental Fig. 3.

36 J O U R N A L O F P R O T E O M I C S 7 3 ( 2 0 0 9 ) 3 0 – 4 0

include MOD1 and VSP1. MOD1 involved in program cell deathclearly contained disulfide bonds [36] and here we showed anincrease in protein synthesis in response to MeJA. ArabidopsisVSP1 gene has been found to contain a jasmonate-responsiveelement in its promoter region [37] and the gene expressionwasupregulated in response to herbivores [38], JA [39] andMeJA [40].VSP proteins may play important nutritional roles during plantdevelopment and stress adaptation.

Twoglycine-richRNA-bindingproteins (GR-RBP),GR-RBP7andGR-RBP8 were identified in the roots with decreased abundancein response to MeJA. This family of proteins is involved in the

regulation of the post-transcriptional gene expression processesincluding pre-mRNA splicing, mRNA transport, mRNA stabilityand translation [40–42]. It has been reported that GR-RBP7expression ismodulated via a circadian clock [43] and by a varietyof abiotic and biotic stress conditions [44,45]. Recently, GR-RBP7was shown to affect stomatal movement in response to abioticstress, and thus play a role in freezing tolerance and in theresponse to dehydration and high salinity stress [46]. GR-RBP7was localized in both the nucleus and cytoplasm of guard cells,but was also present in root cells [46]. In roots, the function ofthese post-transcriptional factors has never been studied. The

Fig. 5 – Tentative localization of the disulfide bonds in three representative proteins. (A) isocitrate dehydrogenase, (B) isoform 3carbonic anhydrase, and (C) cysteine synthase.

37J O U R N A L O F P R O T E O M I C S 7 3 ( 2 0 0 9 ) 3 0 – 4 0

decrease of GR-RBP protein expression in response toMeJAmightaffect the expression of other proteins essential inMeJA signalinginArabidopsis. Other proteins that exhibited significant decreasesin abundance include regulator of ribonuclease-like protein,NDPK1, aconitase small subunit, fructose-bisphosphate aldolaseand photosystem I light harvesting complex protein. Except forlight harvesting complex protein, microarray studies did notidentify significant expression changes in genes of the otherproteins, highlighting the importance of proteomic studies [39].

4.2. Definition of redox-regulated thiol switches

Redox proteomic technologies have been well-developed in themammalian field to study the effect of oxidative stress onregulatory mechanisms and dysfunctions such as immunode-ficiency and Alzheimer's disease [47,48]. In the plant field, theapplication of these technologies is still in its infancy. Experi-mental approaches including the mBBr labeling have beendeveloped and a number of potential thioredoxin and glutar-edoxin target proteins have been identified [18,22]. A recentstudy using Medicago truncatula identified in vivo thioredoxintarget proteins in the course of seed germination [21]. However,investigation of in vivo redox proteome and its response toenvironmental signals has been rare.

Our interesthere is todefinepotential protein candidateswiththiol switches in the jasmonate signaling networks usingcomparative proteomics. The principle underlying the approachemployed is that free thiols are susceptible to irreversible

modification by iodoacetamide, leading to thiol carbamido-methylation. When exposed to MeJA and oxidative stress, thesensitive cysteine thiols are oxidized. After reduction, proteinswith these thiols can be specifically tagged by labelingwithmBBrand separated by 2D gels. By identification of spots from bothcontrol and treatment gels, we were able to determine thecysteines thatwere redox-regulated byMeJA and those thatwerenot. If a protein from control and treatment displayed similarlevels of mBBr labeling relative to SyproRuby staining, thecysteines in the protein were not involved in redox regulation.If the labeled cysteines showed different levels in treatment or incontrol samples orwere identifiedonly in one condition, then thecysteinesmaybe targetsofoxidation in response toMeJA. IfmBBrsignal increases after MeJA treatment, it indicates that theoxidized cysteines can be reduced and labeled by mBBr. Thesereversible cysteines can be considered as redox-regulatedcysteines and the proteins as thiol redox proteins. If there is aloss ofmBBr signal, it implies that irreversible cysteine oxidationoccurs after MeJA treatment. With this comparative approach,thiol switches were defined for many proteins.

4.3. Redox-regulated proteins in MeJA signal transduction

In the present study, most of the proteins that changed theirredox states in response to MeJA were proteins involved indetoxification and defense mechanisms. Two glutathione-S-transferases (GSTs) contained sulfhydryl groups that wereoxidized in response to the oxidative stress induced by MeJA.

38 J O U R N A L O F P R O T E O M I C S 7 3 ( 2 0 0 9 ) 3 0 – 4 0

GSTs have multiple functions in plants, but they are mostlyknown as detoxifiers of endobiotic and xenobiotic com-pounds by covalently linking glutathione to hydrophobicsubstrates [49] for sequestration or removal. It was shownpreviously that MeJA induced the expression of GST genes intobacco, Arabidopsis or Bupleurum kaoi [50–52]. What is stillunknown is how MeJA regulates the GSTs during pathogenattack and how the GSTs act in the defense response. Herewe provide some insight into a possible redox regulation ofGSTs by the accumulation of H2O2 when plants are understress.

Proteins related to the myrosinase-glucosinolate systemwere identified with redox state differences after MeJA treat-ment. Myrosinases catalyze the hydrolysis of glucosinolates, agroup of sulfur and nitrogen containing plant specializedmetabolites present in species of the Brassicaceae family.Glucosinolate degradation products are involved in plantdefense against insects and pathogens [53]. A beta-glucosidase,PKY10 was identified in shoots with a greater than two-foldincrease in oxidized state. PKY10 was reported as a root andhypocotyl specific myrosinase in A. thaliana [54], although it isnot clear whether it has myrosinase activity. PKY10 occupiedmost of the endoplasmic reticulum (ER) bodies [55], which wereinduced in rosette leaves by jasmonate [56]. A recent studyshowed that the PYK10 active form was polymerized throughforming a homodimer linked by disulfide bonds [57]. Thissuggests that the increase of oxidation of the PYK10 in responseto MeJA is likely to lead to an increase in activity. Since myro-sinase activities in Arabidopsis hypocotyls were found to bereduced in response to MeJA [24], PYK10 may not be the majormyrosinase in Arabidopsis shoots, or its function may bedifferent from glucosinolate degradation. A recent studyshowed that PYK10 was essential in the interaction betweenArabidopsis and an endophytic fungus [58].

Trypsin and protease inhibitor family protein was alsoidentified in the root samples with a two-fold increase inoxidation in response to MeJA (Table 1). In plants, theseproteins function as anti-metabolic proteins, which interferewith the digestive process of insects. Previous reports havedemonstrated that protease inhibitors from the Kunitz familywere induced by pathogen and insect attacks [59] and waterstress [60]. The expression of the protease inhibitors in thisfamily increased in response to MeJA treatment [61]. In thisstudy, the increase in oxidation in response toMeJAmay be anindication of an activity change of the protein from inactive toactive. This redox regulation adds a new mechanism to MeJAsignal transduction.

In summary, a 2D gel proteomics approach was used toidentify the proteome changes in response to MeJA. BecauseMeJA induced oxidative stress through ROS production in rootsand leaves, the in vivo redox proteome was studied using mBBrlabeling, comparative 2D gel analysis and tandem MS basedprotein identification. Several redox proteins were identified inresponse to the oxidative stress induced by MeJA. For some ofthe proteins, the cysteines involved in the redox regulationwere mapped by MS/MS. The oxidation and reduction of thesesulfhydryl groups may serve as important thiol switches inMeJA signal transduction. This report not only introduced a wayto capture thiol redox proteins and their quantitative dyna-mics, but also unraveled interesting MeJA and/or oxidative

stress responsive proteins whose functions await furtherinvestigations.

Acknowledgements

This work was supported by University of Florida and theNational Science Foundation Grant (MCB 0818051) to S. Chen.The manuscript was significantly improved as a result of theexcellent suggestions made by three anonymous reviewers.

Appendix A. Supplementary data

Supplementary data associated with this article can be found,in the online version, at doi:10.1016/j.jprot.2009.07.005.

R E F E R E N C E S

[1] Creelman RA, Tierney ML, Mullet JE. Jasmonic acid/methyljasmonate accumulate in wounded soybean hypocotyls andmodulate wound gene expression. Proc Natl Acad Sci USA1992;89:4938–41.

[2] Seo HS, Song JT, Cheong JJ, Lee YH, Lee YW, Hwang I, et al.Jasmonic acid carboxyl methyltransferase: a key enzyme forjasmonate-regulated plant responses. Proc Natl Acad Sci USA2001;98:4788–93.

[3] Creelman RA, Mullet JE. Biosynthesis and action ofjasmonates in plants. Annu Rev Plant Physiol Plant Mol Biol1997;48:355–81.

[4] Gundlach H, Muller MJ, Kutchan TM, Zenk MH. Jasmonic acidis a signal transducer in elicitor-induced plant cell cultures.Proc Natl Acad Sci USA 1992;89:2389–93.

[5] Vijayan P, Shockey J, Levesque CA, Cook RJ, Browse J. A role forjasmonate in pathogen defense of Arabidopsis. Proc Natl AcadSci USA 1998;95:7209–14.

[6] Chini A, Fonseca S, Fernandez G, Adie B, Chico JM, Lorenzo O,et al. The JAZ family of repressors is the missing link injasmonate signalling. Nature 2007;448:666–71.

[7] Thines B, Katsir L, Melotto M, Niu Y, Mandaokar A, Liu G, et al.JAZ repressor proteins are targets of the SCF(COI1) complexduring jasmonate signaling. Nature 2007;448:661–5.

[8] Wasternack S. Jasmonates: an update on biosynthesis, signaltransduction and action in plant stress response, growth anddevelopment. Ann Bot 2007;100:681–97.

[9] Chen H, Jones AD, Howe GA. Constitutive activation of thejasmonate signaling pathway enhances the production ofsecondary metabolites in tomato. FEBS Lett 2006;580:2540–6.

[10] Howe GA, Lightner J, Browse J, Ryan CA. An octadecanoidpathway mutant (JL5) of tomato is compromised in signalingfor defense against insect attack. Plant Cell 1996;8:2067–77.

[11] Miersch O, Wasternack C. Octadecanoid and jasmonatesignaling in tomato (Lycopersicon esculentum Mill.) leaves:endogenous jasmonates do not induce jasmonatebiosynthesis. Biol Chem 2000;381:715–22.

[12] Kauss H, Jeblick W, Ziegler J, Krabler W. Pretreatment ofParsley (Petroselinum crispum L.) Suspension cultures withmethyl jasmonate enhances elicitation of activated oxygenspecies. Plant Physiol 1994;105:89–94.

[13] Orozco-Cardenas M, Ryan CA. Hydrogen peroxide isgenerated systemically in plant leaves by wounding andsystemin via the octadecanoid pathway. Proc Natl Acad SciUSA 1999;96:6553–7.

39J O U R N A L O F P R O T E O M I C S 7 3 ( 2 0 0 9 ) 3 0 – 4 0

[14] Neill SJ, Desikan R, Clarke A, Hurst RD, Hancock JT. Hydrogenperoxide and nitric oxide as signaling molecules in plants. JExp Bot 2002;53:1237–47.

[15] Orozco-Cardenas ML, Narvaez-Vasquez J, Ryan CA. Hydrogenperoxide acts as a second messenger for the induction ofdefense genes in tomato plants in response to wounding,systemin, and methyl jasmonate. Plant Cell 2001;13:179–91.

[16] dos Santos Soares AM, de Souza TF, de Souza Domingues SJ,Jacinto T, Tavares Machado OL. Methyl jasmonate promotesthe transient reduction of the levels of 2-Cys peroxiredoxin inRicinus communis plants. Plant Physiol Biochem 2004;42:543–7.

[17] Buchanan BB, Balmer Y. Redox regulation: a broadeninghorizon. Annu Rev Plant Biol 2005;56:187–220.

[18] Rouhier N, Villarejo A, Srivastava M, Gelhaye E, Keech O,Droux M, et al. Identification of plant glutaredoxin targets.Antioxid Redox Signal 2005;7:919–29.

[19] Hancock J, Desikan R, Harrison J, Bright J, Hooley R, Neill S.Doing the unexpected: proteins involved in hydrogenperoxide perception. J Exp Bot 2006;57:1711–8.

[20] Paget MSB, Buttner MJ. Thiol-based regulatory switches. AnnuRev Genet 2003;37:91–121.

[21] Alkhalfioui F, Renard M, Vensel WH, Wong J, Tanaka CK,Hurkman WJ, et al. Thioredoxin-linked proteins are reducedduring germination ofMedicago truncatula seeds. Plant Physiol2007;144:1559–79.

[22] Yano H, Kuroda S, Buchanan BB. Disulfide proteome in theanalysis of protein function and structure. Proteomics2002;2:1090–6.

[23] Lee K, Lee J, Kim Y, Bae D, Kang KY, Yoon SC, et al. Definingthe plant disulfide proteome. Electrophoresis 2004;25:532–41.

[24] Alvarez S, He Y, Chen S. Comparative investigations of theglucosinolate-myrosinase system in Arabidopsis suspensioncells and hypocotyls. Plant Cell Physiol 2008;49:324–33.

[25] Thordal-Christensen H, Zhang Z, Wei Y, Collinge DB.Subcellular localization of H2O2 in plants, H2O2 accumulationin papillae and hypersensitive response duringbarley-powdery mildew interaction. Plant J 1997;11:1187–94.

[26] Zhu J, Chen S, Alvarez S, Asirvatham VS, Schachtman DP, WuY, et al. Cell wall proteome in the maize primary rootelongation zone. I. Extraction and identification ofwater-soluble and lightly ionically bound proteins. PlantPhysiol 2006;140:311–25.

[27] Sheffield J, Taylor N, Fauquet C, Chen S. The cassava(Manihot esculenta Crantz) root proteome: proteinidentification and differential expression. Proteomics2006;6:1588–98.

[28] Chen S, Harmon AC. Advances in plant proteomics.Proteomics 2006;6:5504–16.

[29] Schneider K, Kienow L, Schmelzer E, Colby T, Bartsch M,Miersch O, et al. A new type of peroxisomal acyl-coenzyme Asynthetase from Arabidopsis thaliana has the catalyticcapacity to activate biosynthetic precursors of jasmonic acid.J Biol Chem 2005;280:13962–72.

[30] Brader G, Tas E, Tapio Palva E. Jasmonate-dependentinduction of indole glucosinolates in Arabidopsis by culturefiltrates of the nonspecific pathogen Erwinia carotovora. PlantPhysiol 2001;126:849–60.

[31] Liu C. Characterization and functional analysis ofArabinogalactan protein 31 in Arabidopsis. Ph.D. Dissertation,The University of Texas at Austin 2007 Dec.

[32] On-line resources: http://130.14.29.110/projects/geo/query/acc.cgi?acc=GSM339777 and http://www.ncbi.nlm.nih.gov/sites/entrez?db=nucest&term=500+um+Methyl+Jasmonate&report=DocSum.

[33] Ferre F, Clote P. DiANNA: a web server for disulfideconnectivity prediction. Nucleic Acids Res 2005;33:w230–2.

[34] Sheehan D. Detection of redox-based modification intwo-dimensional electrophoresis proteomic separations.Biochem Biophys Res Commun 2006;349:455–62.

[35] He Y, Fukushige H, Hildebrand DF, Gan S. Evidence supportinga role of jasmonic acid in Arabidopsis leaf senescence. PlantPhysiol 2002;128:876–84.

[36] Alvarez S, Wilson GH, Chen S. Determination of in vivodisulfide-bonded proteins inArabidopsis. J Chromatogr B2009;877:101–4.

[37] Guerineau F, Benjdia M, Zhou DX. A jasmonate-responsiveelement within theArabidopsis thaliana vsp1 promoter. J Exp Bot2003;54:1153–62.

[38] Berger S, Mitchell-Olds T, Stotz HU. Local and differentialcontrol of vegetative storage protein expression in responseto herbivore damage in Arabidopsis thaliana. Physiol Plant2002;114:85–91.

[39] Jung C, Lyou SH, Yeu S, Kim MA, Rhee S, Kim M, et al.Microarray-based screening of jasmonate-responsive genesin Arabidopsis thaliana. Plant Cell Rep 2007;26:1053–63.

[40] Dreyfuss G, Matunis MJ, Pinol-Roma S, Burd CG. hnRNP proteinsand thebiogenesisofmRNA.AnnuRevBiochem1993;62:289–321.

[41] Higgins CF. Stability and degradation of mRNA. Curr Opin CellBiol 1991;3:1013–8.

[42] Simpson GG, Filipowicz W. Splicing of precursors to mRNA inhigher plants: mechanism, regulation and sub-nuclearorganization of the spliceosomal machinery. Plant Mol Biol1996;32:1–41.

[43] Staiger D, Zecca L, Wieczorek Kirk DA, Apel K, Eckstein L. Thecircadian clock regulated RNA-binding protein AtGRP7autoregulates its expression by influencing alternativesplicing of its own pre-mRNA. Plant J 2003;33:361–71.

[44] Kim YO, Pan S, Jung CH, Kang H. A zinc finger-containingglycine-rich RNA-binding protein, atRZ-1a, has a negativeimpact on seed germination and seedling growth ofArabidopsis thaliana under salt or drought stress conditions.Plant Cell Physiol 2007;48:1170–81.

[45] Kwak KJ, Kim YO, Kang H. Characterization of transgenicArabidopsis plants overexpressing GR-RBP4 under highsalinity, dehydration, or cold stress. J Exp Bot 2005;56:3007–16.

[46] Kim JS, Jung HJ, Lee HJ, Kim KA, Goh CH, Woo Y, et al.Glycine-rich RNA-binding protein 7 affects abiotic stressresponses by regulating stomata opening and closing inArabidopsis thaliana. Plant J 2008;55:455–66.

[47] Fratelli M, Demol H, Puype M, Casagrande S, Eberini I,Salmona M, et al. Identification by redox proteomics ofglutathionylated proteins in oxidatively stressed humanT lymphocytes. Proc Natl Acad Sci USA 2002;99:3505–10.

[48] Sultana R, Butterfield DA. Redox proteomics studies of in vivoamyloid beta-peptide animal models of Alzheimer's disease:insight into the role of oxidative stress. Proteomics-ClinicAppl 2008;2:685–96.

[49] Edwards R, Dixon DP, Walbot V. Plant glutathioneS-transferases: enzymes with multiple functions insickness and in health. Trends Plant Sci 2000;5:193–8.

[50] Chen LR, Chen YJ, Lee CY, Lin TY. MeJA-inducedtranscriptional changes in adventitious roots of Bupleurumkaoi. Plant Sci 2007;173:12–24.

[51] Jost R, Altschmied L, Bloem E, Bogs J, Gershenzon J, Hahnel U,et al. Expression profiling of metabolic genes in response tomethyl jasmonate reveals regulation of genes of primary andsecondary sulfur-related pathways in Arabidopsis thaliana.Photosynth Res 2005;86:491–508.

[52] Xiang C, Miao ZH, Lam E. Coordinated activation of as-1-typeelements and a tobacco glutathione S-transferase gene byauxins, salicylic acid, methyl-jasmonate and hydrogenperoxide. Plant Mol Biol 1996;32:415–26.

[53] Bones AM, Rossiter J. The myrosinase–glucosinolatesystem, its organization and biochemistry. Physiol Plant1996;97:194–208.

[54] Nitz I, Berkefeld H, Puzio PS, Grundler FM. Pyk10, a seedlingand root specific gene and promoter from Arabidopsis thaliana.Plant Sci 2001;161:337–46.

40 J O U R N A L O F P R O T E O M I C S 7 3 ( 2 0 0 9 ) 3 0 – 4 0

[55] Matsushima R, Hayashi Y, Yamada K, Shimada T, NishimuraM, Hara-Nishimura I. The ER body, a novel endoplasmicreticulum-derived structure in Arabidopsis. Plant Cell Physiol2003;44:661–6.

[56] Matsushima R, Kondo M, Nishimura M, Hara-Nishimura I. Anovel ER-derived compartment, the ER body, selectivelyaccumulates a beta-glucosidase with an ER-retention signal inArabidopsis. Plant J 2003;33:493–502.

[57] Nagano AJ, Matsushima R, Hara-Nishimura I. Activation of anER-body-localized beta-glucosidase via a cytosolic bindingpartner in damaged tissues of Arabidopsis thaliana. Plant CellPhysiol 2005;46:1140–8.

[58] Sherameti I, Venus Y, Drzewiecki C, Tripathi S, Dan VM, Nitz I,et al. PYK10, a beta-glucosidase located in the endoplasmaticreticulum, is crucial for the beneficial interaction between

Arabidopsis thaliana and the endophytic fungus Piriformosporaindica. Plant J 2008;54:428–39.

[59] Ryan CA. Protease inhibitors in plants: genes for improvingdefenses against insects and pathogens. Annu RevPhytopathol 1990;28:425–49.

[60] Downing WL, Mauxion F, Fauvarque MO, Reviron MP, deVienne D, Vartanian N, et al. A Brassica napus transcriptencoding a protein related to the Kunitz protease inhibitorfamily accumulates upon water stress in leaves, not in seeds.Plant J 1992;2:685–93.

[61] Badri DV, Loyola-Vargas VM, Du J, Stermitz FR, Broeckling CD,Iglesias-Andreu L, et al. Transcriptome analysis of Arabidopsisroots treated with signaling compounds: a focus on signaltransduction, metabolic regulation and secretion. New Phytol2008;179:209–23.