JOURNAL OF PLANT PHYSIOLOGY 2011 proteome

Cm

BFa

b

c

d

e

f

g

h

i

j

a

ARRA

KLPPTCTM

I

twCbqrti

0d

Journal of Plant Physiology 168 (2011) 392–402

Contents lists available at ScienceDirect

Journal of Plant Physiology

journa l homepage: www.e lsev ier .de / jp lph

omparative analysis of proteome changes induced by the two spotted spiderite Tetranychus urticae and methyl jasmonate in citrus leaves

.E. Maserti a,∗, R. Del Carratoreb, C.M. Della Crocea,e, A. Poddaa, Q. Migheli c,d, Y. Froelicher f,. Lurog, R. Morillonh, P. Ollitraulth, M. Taloni, M. Rossignol j

CNR-IBF, Istituto di BioFisica, Area della Ricerca, Via Moruzzi 1, I-56124 Pisa, ItalyCNR-IFC, Istituto di Fisiologia Clinica, Area della Ricerca, Via Moruzzi 1, I-56124 Pisa, ItalyUNISS-DPP, Università di Sassari, Dipartimento di Protezione delle Piante, I-07100 Sassari, ItalyIstituto Nazionale di Biostrutture e Biosistemi, Via De Nicola 9, I-07100 Sassari, ItalyCNR-IBBA, Istituto di Biologia e Biotecnologie Agrarie, Area della Ricerca, Via Moruzzi 1, I-56124 Pisa, ItalyCIRAD-UPR, Amélioration génétique d’espèces à multiplication végétative, F-20230, San Nicolao, Corse, FranceINRA-UR GEQUA, San Giuliano, F-20230, San Nicolao, Corse, FranceCIRAD-UPR, Amélioration génétique d’espèces à multiplication vegetative, Avenue Agropolis, 34398 Montpellier, Cedex 5, FranceIVIA, Centro de Genómica Instituto Valenciano de Investigaciones Agrarias, Moncada, 46113 Valencia, SpainINRA-UR 1199 Protéomique Fonctionnelle, Place Pierre Viala, 34060 Montpellier, Cedex 2, France

r t i c l e i n f o

rticle history:eceived 9 April 2010eceived in revised form 29 July 2010ccepted 30 July 2010

eywords:eaf proteomelant defenselant–arthropod interactionwo-dimensional electrophoresis

a b s t r a c t

Citrus plants are currently facing biotic and abiotic stresses. Therefore, the characterization of molecu-lar traits involved in the response mechanisms to stress could facilitate selection of resistant varieties.Although large cDNA microarray profiling has been generated in citrus tissues, the available proteinexpression data are scarce. In this study, to identify differentially expressed proteins in Citrus clementinaleaves after infestation by the two-spotted spider mite Tetranychus urticae, a proteome comparison wasundertaken using two-dimensional gel electrophoresis. The citrus leaf proteome profile was also com-pared with that of leaves treated over 0–72 h with methyl jasmonate, a compound playing a key role inthe defense mechanisms of plants to insect/arthropod attack. Significant variations were observed for110 protein spots after spider mite infestation and 67 protein spots after MeJA treatments. Of these, 50

itrusetranychus urticaeethyl jasmonate

proteins were successfully identified by liquid chromatography–mass spectrometry–tandem mass spec-trometry. The majority constituted photosynthesis- and metabolism-related proteins. Five were oxidativestress associated enzymes, including phospholipid glutathione peroxidase, a salt stressed associated pro-tein, ascorbate peroxidase and Mn-superoxide dismutase. Seven were defense-related proteins, such asthe pathogenesis-related acidic chitinase, the protease inhibitor miraculin-like protein, and a lectin-likeprotein. This is the first report of differentially regulated proteins after T. urticae attack and exogenous

s leav
MeJA application in citru
ntroduction

Citrus fruits represent one of the most important fruit produc-ions worldwide, with 109 million tons produced annually in theorld. This fruit crop is currently facing biotic and abiotic stresses.onsequently, the generation of new varieties and rootstocks com-ining tolerance to major biotic and abiotic stresses and good
uality characteristics constitutes a major challenge of the cur-ent citrus industry. To this end, the molecular characterization ofhe physiological and genetic determinants of selected character-stics is necessary. Although large cDNA microarray profiles have
∗ Corresponding author. Tel.: +39 0503152748; fax: +39 0503152760.E-mail address: [email protected] (B.E. Maserti).

176-1617/$ – see front matter © 2010 Elsevier GmbH. All rights reserved.oi:10.1016/j.jplph.2010.07.026

es.© 2010 Elsevier GmbH. All rights reserved.

been generated in citrus tissues (Martinez-Godoy et al., 2008), aproteomic approach could help to complement the genomic andtranscriptomic data. To date, only a few investigations have con-ducted comparative proteomic studies in citrus plants affected byabiotic and biotic stress (Lliso et al., 2007; Cantú et al., 2008; Shi etal., 2008).

The two spotted spider mite, Tetranychus urticae Koch (Acari:Tetranychidae), is considered a pest of many economically impor-tant plant species. Mites are cell-content feeders; they laceratecells and consume cellular contents by means of their stylets. The
withdrawal of cell content leads to loss of chlorophyll, resultingin whitish or yellowish speckled areas on the upper surfaces ofleaves. Citrus plants are particularly sensitive to outbreaks of spi-der mite, especially when grown in dry climates (Aucejo-Romeroet al., 2004).
dx.doi.org/10.1016/j.jplph.2010.07.026

http://www.sciencedirect.com/science/journal/01761617

http://www.elsevier.de/jplph

mailto:[email protected]

dx.doi.org/10.1016/j.jplph.2010.07.026

lant Ph

piaJou

ctts

tFngmwtrmTi

M

P

Ilsrmcr5Muecf6FNicttaoarw

S

soina

B.E. Maserti et al. / Journal of P

In order to reduce attack by phytopathogenic pests andathogens, plants have developed several defense mechanisms,

ncluding the induction of defensive proteins that can negativelyffect the feeding, growth or reproduction of pests (Howe andander, 2008). Indeed, Kant et al. (2004) have found over-expressionf chitinase and proteinase inhibitors in tomato leaf infested by T.rticae.

In this work, a comparative proteomic analysis of Citruslementina (Hort. ex Tan) leaves infested by T. urticae was under-aken with the aim to evaluate the potential of proteomics as aool for identifying candidate proteins involved in the response topider mite attack in citrus plants.

C. clementina is the main small citrus variety, and it is par-icularly important for the Mediterranean Basin citrus industry.urthermore, C. clementina has recently been chosen by the Inter-ational Citrus Genome Consortium as a model system for wholeenome sequencing from a haploid line. The proteome profile ofite-infested leaves was also compared with that of leaves treatedith methyl jasmonate over a time course 0–72 h. The rationale of

his comparison is the dominant role played by jasmonic acid andelative compounds in promoting the plant defense response toany arthropod herbivores (Howe and Jander, 2008). For instance,

. urticae attack triggered a rapid and direct response of jasmonate-nduced genes in tomato (Li et al., 2002).

aterial and methods

lant material

C. clementina cv Tomatera (clone SRA 63; Station de RechercheNRA, S. Nicolao, Corse, France) plants were used as the source ofeaves. All plants used were reproduced by scions excised from aingle mother plant and grafted on Poncirus trifoliata (cv Kryder)ootstocks. Plants were grown in separate pots (7 L) in a com-ercially available substrate for citrus plants, under greenhouse

ondition for two years. Plants were watered every three days andeceived fertilizer for citrus plants (NPK 6.5.11—Gesal, Italy: 6% N;% P2O5; 6% K2O; 0.01% B; 0.002% Cu; 0.02% Fe; 0.01% Mn; 0.001%o; 0.002% Zn, and all the elements were chelated with water sol-

ble EDTA), once a week. Three months before the start of thexperiments, the plants were transferred into controlled growthhambers for acclimation. The experimental conditions were asollows: 28 ± 1 ◦C; light/dark cycle of 16/8 h; relative humidity0–80%. Fluorescent tubes (General Electric F36W/54 and Osramluora L36W/77) were used to produce a PAR of 100 �mol m−2 s−1.o other plants were present in the room at the time of the exper-

ment. Different experiments were performed in separate growthhambers in which the same environmental conditions were main-ained. Each experiment was performed in triplicate, and each ofhe three biological replicates consisted of a pool of four leaves ofbout the same age, collected from four independent plants. Leavesf approximately similar size (6 × 2.5 cm) were used. Treatmentsnd leaf harvesting were carried out in the middle of day photope-iod. Harvested leaves were carefully washed with double distilledater, frozen in liquid nitrogen and stored at −80 ◦C until analysis.

pider mite infestation

Infestation by T. urticae was carried out by transferring about 50
pider mites collected from naturally infested C. clementina plantsnto leaves of the experimental plants. Infested leaves were closednto nest bags to avoid mite escape. Control leaves were closed intoest bags without mites. Leaves were harvested at 0 h, 2 h, 6 h, 24 hfter infestation.
ysiology 168 (2011) 392–402 393

Chemical treatment

MeJA treatments were performed by spraying the upper andlower leaf sheet with an aqueous solution of 1% (v/v) glycerol sup-plemented with MeJA 0.1 mM. MeJA was diluted in water from a1000-fold stock solution in 96% ethanol. Control leaves were treatedwith 1% glycerol and an equal amount of ethanol. Leaf material washarvested at 0 h, 2 h, 6 h, 24 h, 48 h, 72 h after chemical treatment.

Protein extraction and solubilization

For each treatment, total protein extracts were prepared induplicate for each biological replicate according to previously pub-lished protocols (Maserti et al., 2007) with a slight modification asfollows: 0.01% protease inhibitor cocktail (P 9599 Sigma–Aldrich,Italy) was added to lysis buffer. The protein concentration of crudeextract was assayed using an RC/DC assay (BioRad, USA). Bovineserum albumin (BSA) was used to create a standard curve. Thefinal sample average leaf protein concentration was 3 ± 1.5 mg/gF.W. After protein assay, samples were stored at −80 ◦C until 2-DEanalysis.

2-DE analysis

A total of 500 �g of proteins were dissolved in 340 �L of rehydra-tion buffer (7 M urea, 2 M thiourea, 4% CHAPS, 10% isopropanol, 1.5%DeStreak Reagent, 0.5% IPG Buffer 3–10 NL) and loaded on 18 cm IPGstrips pH 3–10 NL (GE-Healthcare), belonging to the same batch. In-gel rehydration was performed in passive mode for 5 h and at 50 Vfor an additional 7 h in a Protean IEF system (Bio-Rad). Then, IPGswere focused under the following conditions: 5 h at 300 V, 6 h at500 V, 2 h gradient at 500–1000 V, 4 h gradient at 1000–5000 V, 2 hat 5000 V, 2 h gradient at 5000–8000 V, and 50,000 V h at 8000 V at25 ◦C with the current limited to 50 mA strip−1. Focused strips werefrozen to −80 ◦C, awaiting second-dimension, when they were sub-jected to reduction, alkylation and detergent exchange by two stepincubations in 15 mL equilibration buffer containing DTT 2% for thefirst incubation and IAA 4% for the second. The strips were thenapplied to the top of lab-cast polyacrylamide 13% w/v gel and fixedin place with 1 mL 0.5% agarose, 0.002% bromo-phenol blue. Seconddimension electrophoresis was carried out in a Protean XL cell (Bio-Rad) using a two-step program (60 min at 15 mA/gel followed by30 mA/gel until the dye front was within 0.5 cm of the bottom edgeof the gel). Molecular masses were determined by running stan-dard protein markers from Sigma–Aldrich (S8445). The resultinggels were stained with colloidal Coomassie. The stained gels weredigitized at an optical resolution of 300 dpi with a GS-800 scanner(BioRad, USA).

Image and statistical analysis

The 2D gel images were analyzed and the protein volume ofeach identified spot was quantified using REDFIN basic software(http://www.ludesi.com). Gel images were warped to each otherafter setting vectors points. Stacking gel images, a fusion gel imageis created, and spots are detected and transferred back to all theanalyzed gels. The intensity of each protein spot was normalized tothe total intensity of all valid spots detected on each gel. For eachtreatment, six 2-DE gels representing three biological replicatesand two technical replicates were used for data analyses.

Comparison of protein expression levels from the various
samples involved pairwise t-tests (p ≤ 0.01) between the threetreatment types (spider mite vs control, methyl jasmonate vs con-trol, and spider mite vs methyl jasmonate), with each time pointbeing treated separately. Protein spots that demonstrated a ratio atleast ≥1.5-fold between treatments were defined as up-regulated.
http://www.ludesi.com/

394 B.E. Maserti et al. / Journal of Plant Physiology 168 (2011) 392–402

F stationd protea ith red

Pr“wd

P

aaunt1at

ig. 1. Representative 2-DE patterns of proteins from C. clementina leaves after infeimension (IEF) and 13% polyacrylamide gels in the second dimension. Numbers ofnd S2. Circles indicate proteins with changes in accumulation in treated plants wetected in the gel.

rotein spots showing a ratio of at least ≤0.6 were defined as down-egulated. Proteins present only in treated leaves were defined asinduced” and their ratio set ∞. 50 spots fulfilling the above criteriaere manually excised from gels and subjected to LC–MS/MS andatabase search for identification.

rotein identification by LC–MS/MS

Spots were processed using a MultiProbe II liquid handlingutomate (PerkinElmer). Briefly, after extensive washing withmmonium bicarbonate and acetonitrile, proteins were digestedsing trypsin (0.1 �g) and, after peptide extraction with ammo-
ium carbonate and acetonitrile, the volume of digests was reducedo ca 10 �L. LC–MS experiments were performed using an Agilent200 LC chain fitted with the Chip Cube (Agilent) and coupled ton Esquire-HCT ion trap mass spectrometer (Bruker). Tryptic pep-ides (2 �L) were resolved on a Chip containing an enrichment
by T. urticae. Proteins were separated on a 3–10 non-linear pH gradient in the firstin spots correspond to those listed in Table 1 and supporting information Tables S1spect to controls after pairwise t-test (p ≤ 0.01). Dotted squares indicate spots not

precolumn and a 43 mm × 75 �m RP column, using a linear gra-dient (7 min) from 3% to 45% of a 80% acetonitrile mobile phasein 0.1% formic acid. The overall system performance was assessedby introducing a BSA sample (25 fmol) every 10 samples. Raw datawere processed using Data Analysis software (Bruker) to generatefiles for database searching with Mascot software (Matrix Sciences).The NCBI Viridiplantae taxonomy was selected to search the MSDBdatabase (20063108 release) using the following parameters: onemissed cleavage allowed, carboxymethylated cystein as fixed mod-ification, mass tolerance of ±0.6 Da for both precursor and productions. Ion scores above 38 were considered significant (p < 0.05).

Gene ontology (GO) annotation

The identified proteins were mapped to Uniprot (www.uniprot.org) to assess whether their function was known. These proteins

http://www.uniprot.org/

lant Ph

woppp

R

2p

smaS

sNipo

lwdstvdtuieods

(mr

istt(dr

tdt

irMw7(pg

(y

B.E. Maserti et al. / Journal of P

ere divided into different groups based on three independent setsf ontology: (1) the cellular component through which the generoducts can be found; (2) the biological process in which the generoduct participates; (3) molecular function, describing the generoduct’s activities.

esults

-DE analysis and identification of differentially expressedroteins by LC–MS/MS

Leaves of C. clementina plants were infested with the two spottedpider mite T. urticae for 0 h, 2 h, 6 h and 24 h or in parallel experi-ents treated over 0–72 h with MeJA. Total proteins were extracted

nd subjected to 2-DE (Fig. 1, and supporting information Figs. S1,2, S3, S4, S5, S6, S7).

Image analysis revealed about 650 highly reproducible proteinpots across pI and a molecular mass range of 4–8 and 14–70 kDa.o significant differences were observed between spots represent-

ng 2D separated proteins from control leaves at different timeoints, either between control leaves from spider mite experimentsr control leaves from MeJA experiments.

Relevant changes in protein abundance in spider mite-infestedeaves were occurring after only 24 h following treatment. There

ere also several proteins that appeared to be altered in abun-ance in response to the treatments at early time points, but theseamples showed a much higher degree of variation among thehree independent biological replicates. The increased biologicalariation may be due to a slower onset of feeding by starving spi-er mites at the beginning of the experiments, or differences inhe rate of feeding. Consequently, the statistical power achievedsing replicates for 2 h and 6 h time points with spider mite feed-

ng was not sufficient to allow conclusive selection of differentiallyxpressed proteins. Consistent differences in the proteome patternf MeJA-treated leaves compared to the pattern of control and spi-er mite-infested leaves were noted, especially in a group of proteinpots at 55 kDa and pI 6–7.

Statistical analysis revealed significant differences at 24 hp ≤ 0.01) for 110 protein spots in the leaves infested by spider

ites and for 67 protein spots in the leaves treated with MeJA,espectively.

50 differentially regulated protein spots were successfullydentified by LC–MS/MS analysis followed by Mascot databaseearches of Viridiplantae in the NCBI database. The results allowedhe identification of 32 different protein accessions, suggestinghe occurrence of post-translational modifications. Three proteinsspots 37, 72 and 81) were not matched with other proteins in theatabase (Table 1). Peptide sequences used for identification areeported in supporting information Table S1.

Out of these, the levels of 35 proteins were changed by bothreatments, 8 proteins were only differentially regulated by spi-er mite infestation and 7 proteins were only detected after MeJAreatment (Fig. 2A).

Globally, 18 proteins were up-regulated, four proteins werenduced (spots 13, 14, 17 and 43) and three proteins were down-egulated (spots 10, 79 and 82) in spider mite-infested andeJA-treated leaves compared to control leaves. Four proteinsere found exclusively in spider mite treated leaves (spots 53, 72,

7 and 81). Changes after MeJA treatment occurred since early stage2 h), and were generally stable up to 48 h. However, the levels of
rotein accumulation induced by the two-spotted spider mite wereenerally stronger than those induced by MeJA (Fig. 3).
MeJA strongly induced the expression of seven protein spotsspots 41, 76, 190, 194, 195, 197 and 199) that, after MS anal-sis, resulted as ribulose-1,5-biphosphate carboxylase/oxygenase

ysiology 168 (2011) 392–402 395

(RubisCO) large subunit fragments, probably due to protein degra-dation.

The identified proteins were classified according to electronicannotation (www.uniprot.org). Most were involved in photosyn-thesis and carbohydrate metabolism. Five proteins are involved inresponse to oxidative stress. Seven proteins, potentially involvedin defense against biotic and abiotic stress, were also identified(Fig. 4).

Discussion

Mites are major constraints to the growth of many economicallyimportant plant species such as Citrus sp. However, at present, lit-tle is known about the molecular response mechanisms to thesepests in citrus plants. In the present work, 50 proteins were dif-ferentially regulated by the two spotted spider mite T. urticaeattack in C. clementina leaves. Changes in the expression of mite-responsive proteins were also paralleled by application of MeJA,a compound recognized as playing a dominant role in promotingplant defense/response to many arthropod herbivores (Howe andJander, 2008).

Photosynthesis-related proteins were up-regulated in mite-infested leaves, although the expression of these proteins isgenerally reduced during plant–pest interaction. However, the dataare in accordance with those observed in rice challenged withthe fungal pathogen Rhizoctonia solani by Lee et al. (2006). In thiswork, maintenance of photosynthesis activity during mite attackmight be explained in terms of infestation length (24 h) or inten-sity. However, spider mites induced up-regulation of glutaminesynthetase. Oliveira et al. (2002) suggested that increasing levelsof gluthamine synthetase in transgenic tobacco plants may accom-pany an increase in photosynthetic capacity. Giordanengo et al.(2010) found that M. persicae infestation of potato plant stronglyincreased glutamine synthase at the site of feeding. Spider mitesalso induced up-regulation of several proteins involved in primarymetabolism, including several glycolytic enzymes and two dehy-drogenases (malate and dihydrolipoamide dehydrogenase). Takentogether, the results reported in this work seem to support theidea that a reconfiguration of leaf primary metabolism might itselfbe a defensive strategy against herbivore attack, as suggested bySchwachtje and Baldwin (2008).

MeJA induced down-regulation of large and small RubisCO sub-units and other photosynthesis related proteins. Lower levels ofproteins involved in photosynthetic carbon assimilation have beenobserved by Rakwal and Komatsu (2000) in rice exposed to exoge-nous jasmonic acid. A strong degradation of RubisCO occurred,as several fragments of large RubisCO subunit was found in gelfrom MeJA treated leaves (spots 41, 43, 76, 190, 194, 195 and199). This finding may be the cause of the noticeable differencesobserved in proteome pattern from MeJA-treated compared tountreated and spider mite-infested leaves. MeJA is known to affectvarious steps in the formation of chloroplast proteins, leading tocharacteristic senescence symptoms within plastid compartments,such as chlorophyll loss and Rubisco degradation (Reinbothe etal., 1994). Desimone et al. (1996) observed that oxidative stressinduced degradation of the large subunit of RubisCO in isolatedchloroplasts of barley. Martins dos Santos Soares et al. (2004)observed that MeJA application induced RubisCO degradation inRicinus communis leaves and suggested that such degradation couldbe promoted by H2O2 accumulation. Thus, RubisCO degradation
induced by MeJA in C. clementina leaves may be due to increasedproduction of ROS. Degradation of chloroplast proteins has beenproposed to be used by plants to provide amino acids necessaryto satisfy the high demand of residues to synthesize defense pro-teins (Reinbothe et al., 1994).A rapid increase in reactive oxygen

396B.E.M

asertietal./JournalofPlant

Physiology168 (2011) 392–402

Table 1List of identified C. clementina leaf proteins differentially regulated proteins by T. urticae infestation or MeJA application. Spot number; NCBI accession number; putative protein name; reference organism; theoretical/experimentalprotein molecular weight and pI; putative subcellular compartment, biological processes and molecular function as inferred by Uniprot database; protein MASCOT score; peptide number used to data search.

Spot no. Acc. no. Putative protein name Referenceorganism

Theor./Exper. MW kDa/pI Subcellularcompartment

Biological processes Molecular function M.ascotScore

P. n.

Chaperones6 P08927 Probable chaperonin 60

beta chainPisumsativum

63/64 5..9/5..2 Chloroplast Protein folding Unfolded protein binding 701 13

7 Q8GTB0 Putative heat shock 70 kDaprotein

Oryzasativa

71/71 5.4/5..2 Mitochondrion Protein folding Protein binding 237 7

8 P08824 Chaperonin groEL Ricinuscommunis

52/57 4.7/5.0 Chloroplast Protein folding Protein binding 481 8

Defence response47 Q8H985 Acidic class II chitinase Citrus

jambhiri32/33 5.1/4.8 N/A Chitin catabolic process Chitinase activity 81 2

48 Q9FQ12 Lectin-related protein Citrusparadisi

29/33 5.1/4.9 N/A N/A Sugar binding 177 5

56 Q8H985 Acidic class II chitinase Citrusjambhiri

32/23 5.1/5.1 N/A Chitin catabolic process Chitinase activity 121 3

75 Q2HXG9 Miraculin-like protein 1 Citrusjambhiri

26/31 8.1/5.7 N/A N/A Endopeptidase activity 223 5

182 B9T1D5 Heat shock protein 26 K Ricinuscommunis

24/23 9.3/4.8 Chloroplast Stress response N/A 335 6

183 B9T1D5 Heat shock protein 26 K Ricinuscommunis

24/22 9..3/4.8 Chloroplast Stress response N/A 128 3

514 Q8W418 Lipoxygenase EC 1.13.11.12 Citrusjambhiri

102/97 5..9/6.0 N/A Oxidation reductionOxylipin biosyntheticprocess

Lipoxygenase activity 187 5

Oxidative stress response57 Q06652 Glutathione peroxidase EC

1.11.1.12 PHGPXCitrussinensis

19/23 5.7/5.4 Cytoplasm Oxidation reductionSalt stress response

GSH peroxidase activity 232 7

74 Q6JRB8 Ascorbate peroxidase EC1.11.1.11

Ipomoeabatatas

28/31 5..3/5..5 Cytoplasm Oxidation reduction l-Ascorbate peroxidase activity 141 2

79 Q38PL3 Superoxide dismutase EC1.15.1.1

Citrus limon 16/24 6.0/5.8 Mitochondrion Oxidation reduction Metal ion binding superoxidedismutase activity

184 3

90 Q38PL3 Superoxide dismutase EC1.15.1.1

Citrus limon 16/24 6.0/5.7 Mitochondrion Oxidation reduction Metal ion binding Superoxidedismutase activity

176 2

95 Q8S568 Catalase EC 1.11.1.6 Vitisvinifera

57/57 6.7/6.7 Peroxisome Oxidation reduction Catalase activity 341 9

ATP synthesis10 Q9THW1 ATP synthase subunit beta

EC 3.6.3.14Poncirustrifoliata

52/57 5.0/5.3 ChloroplastThylakoidmembrane

ATP synthesis coupledproton transport

ATP binding 1004 22

15 Q9XPB2 ATP synthase subunit alphaEC 3.6.3.14

Phaseoulusaureus

56/58 6.2/5.7 Mitochondrion ATP synthesis coupledproton transport

ATP binding 370 9

Glycolysis13 P42896 Enolase EC 4.2.1.11 Ricinus

communis48/57 5.5/5.4 Cytoplasm Glycolysis Phosphopyruvate hydratase

activity463 9

14 P35494 Phosphoglyceromutase EC5.4.2.1

Nicotianatabacum

61/72 5.9/5.7 Cytoplasm Glycolysis Phosphoglycerate mutase activity 225 5

34 Q42961 Phosphoglycerate kinaseEC 2.7.2.3

Nicotianatobacum

50/49 8.4/5.5 Chloroplast Glycolysis Kinase activity ATP binding 269 5

39 P17783 Malate dehydrogenase EC1.1.1.37

Citrulluslanatus

36/42 8.8/5.9 Mitochondrion Glycolysis TCA Cycle l-Malate dehydrogenase activity 240 5

Photosynthesis related proteins16 Q95633 Ribulose-1,5-bisphosphate

carboxylase/oxygenaselarge subunit

Bruceamollis

52/58 6.5/5.9 Chloroplast Photosynthesis Ribulose-biphosphate carboxilaseactivity

648 18

20 Q95633 Ribulose-1,5-bisphosphatecarboxylase/oxygenaselarge subunit

Bruceamollis


648 18

B.E.Masertiet

al./JournalofPlantPhysiology

168 (2011) 392–402397

Table 1 (Continued)

31 Q8L5T3 Rubisco activase Chenopodiumquinoa

48/48 6.5/6.6 Chloroplast Photosynthesis ATP binding 345 10

32 P09559 Phosphoribulokinase EC2.7.1.19

Spinaciaoleracea

45/43 5.8/5.2 Chloroplast PhotosunthesisReductivepentose-phosphatecycle

ATP binding 313 8

33 P09559 Phosphoribulokinase EC2.7.1.19

Spinaciaoleracea

45/43 5.8/5.3 Chloroplast PhotosynthesisReductivepentose-phosphatecycle

ATP binding 244 4


Bruceamollis


97 3


Bruceamollis


60 2


Bruceamollis


60 2


Bruceamollis


56 2


Bruceamollis


56 2


Bruceamollis


56 2


Bruceamollis


56 2


Bruceamollis


56 2

50 P23322 Photosystem IIoxygen-evolving complexprotein 1 OEC 33 kDa

Lycopersiconesculentum

35/32 5.9/5.1 Chloroplast PhotosynthesisPhotosystemIIstabilization

Calcium ion binding 189 6


Lycopersiconesculentum

35/37 5.9/5.1 Chloroplast PhotosynthesisPhotosystemIIstabilization

Calcium ion binding 93 4


Pisumsativum

28/22 8.2/8.4 Chloroplast Photosynthesis Calcium ion binding 104 12


Pisumsativum

28/22 8.2/5.9 Chloroplast Photosynthesis Calcium ion binding 181 4

92 Q84LN2 Ribulose-1,5-bisphosphatecarboxylase/oxygenasesmall subunit

Citrus limon 14/22 8.6/6.0 Chloroplast Photosynthesis Ribulose-biphosphate carboxilaseactivity

183 5

93 Q02071 Type III Chlorophylla/b-binding protein

Pinussylvestris

31/24 8.6/5.2 Chloroplast Energy-photosynthesis Light harvesting 73 1

Aminoacid metabolism and transport35 Q9XFJI Glutamine synthetase EC

6.3.1.2M. crys-tallinum

48/47 6.7/5.2 N/A Glutamine biosyntheticprocess

ATP binding 175 4

96 P50433 GlycinehydroxymethyltransferaseEC 2.1.2.1

Solanumtuberosum

57/57 8.5/7.0 Mitochondrion One-carbon metabolicprocesses

GlycineLoadinghydroxymethyltransferase activity

434 9

398B.E.M

asertietal./JournalofPlant

Physiology168 (2011) 392–402

Table 1 (Continued)

Spot no. Acc. no. Putative protein name Referenceorganism

Theor./Exper. MW kDa/pI Subcellularcompartment

Biological processes Molecular function M.ascotScore

P. n.

Lactoglutathione lyase activity54 O04428 Hypothetical protein Citrus

paradisi33/37 5.4/5.3 N/A N/A Lactoglutathione lyase activity 436 10

70 Q75GB0 Putative glyoxalase Oryzasativa

29/33 4.9/5.3 N/A N/A Lactoglutathione lyase activity 120 4

Others17 Q93WQ1 Dihydrolipoamide

dehydrogenase EC 1.8.1.4Bruguieragymnorhiza

54/58 6.7/5.9 Cytoplasm Oxidation reduction Dihydrolipoyl dehydrogenaseactivity

227 5

77 Q65XW4 Putative 3-betahydroxysteroiddehydrogenase/isomerase

Oryzasativa

31/31 9.1/6.0 N/A Steroid biosyntheticprocess

Isomerase activity 89 1

82 Q53KB6 Retrotransposon protein Oryzasativa

132/23 8.7/8.4 Nucleus RNA-dependent DNAreplication

RNA binding 36 1

Unknown function37 Unknown Unknown /42 /5.3 N/A N/A N/A72 Unknown Unknown /32 /5..3 N/A N/A N/A81 Unknown Unknown /22 /7.2 N/A N/A N/A

B.E. Maserti et al. / Journal of Plant Physiology 168 (2011) 392–402 399

F T. urtu

sd(ioMocttiMo(afAaHlprdgoa2t

ig. 2. Venn diagrams showing numbers of overlapping and unique proteins inp-regulated and induced (in bold) (B); down-regulated (C).

pecies (ROS) appears to be a common event in induced processesirected against herbivores (both chewing and piercing/sucking)Maffei et al., 2006). MeJA was shown to induce antioxidant defensen sunflower seedlings (Parra-Lobato et al., 2009). The occurrencef an oxidative burst in C. clementina leaves after T. urticae oreJA challenge may be supported by the differential regulation

f five scavenging enzymes. Superoxide dismutase (SOD), whichatalyzes disproportionation of O2

·− to H2O2 and O2, is consideredo be one of the first lines of defense in plants. Interestingly, inhis work, Mn-SOD in spot 90 was up-regulated, while anothersoform (spot 79) was down-regulated by both spider mites and

eJA treatment. However, the significance of their regulation withpposite signs is unclear. Up-regulation of ascorbate peroxidaseAPX), which catalyzes the reduction of H2O2 using ascorbate asn electron donor, was found in rice leaves infested by the phloem-eeding brown plant hopper Nilaparvata lugens by Wei et al. (2009).PX was up-regulated in C. clementina leaves by spider mitesnd down-regulated by MeJA. Catalase, which converts H2O2 to2O and O2, was found to be down-regulated in C. clementina

eaves after both treatments. A putative phospholipid gluthationeeroxidase (PHGPX) (spot 57), a salt-associated protein, was up-egulated by both treatments. PHGPX, which is involved in theetoxification of lipid peroxides, showed no increased activity orene expression after insect feeding (Maffei et al., 2006). On the
ther hand, a GPX like protein conferred tolerance to oxidativend environmental stress in transgenic Arabidopsis (Gaber et al.,006). Taken together, the data reported in this work suggest thathe differences in the antioxidant responses may have an impor-
icae and/or MeJA treated C. clementina leaves: total differentially expressed (A);

tant role in mediating citrus responses to T. urticae and MeJAchallenge.

In accordance with the observations of Lippert et al. (2007) inSitka spruce bark infested by weevils, T. urticae induced changesin accumulation of small heat shock proteins (sHSP) and HSP70.The expression of one isoform of sHSP was also triggered by MeJA.sHSPs are thought to play a protective role against oxidative stress.Thus, the increased levels of the two isoforms of HSP20 may sup-port the hypothesis of an oxidative burst triggered by spider mitesand MeJA in C. clementina leaves. The abundance of an HSP70 wasstrongly induced in spider mite-infested leaves. The HSP70 fam-ily is known for its involvement in binding to polypeptide chainsof other proteins, catalyzing their proper folding and thus protect-ing them from denaturation and degradation. In Citrus reticulata,distinct HSP70 mRNAs were differentially expressed during Xylellainfection and Citrus tristeza viral infection (Gandia et al., 2007). HSP70 may play a role in citrus leaves by maintaining the integrity ofother proteins and by facilitating the intercellular transportation ofenzymes during stress.

High levels of lipoxygenase (LOX) were induced in C. clementinaleaves by both treatments. Accumulation of LOX transcripts hasbeen previously reported by Gomi et al. (2002a) in Citrus jambhirileaves within 30 min after wounding or inoculation with Alternariaalternata. LOX proteins, a family of enzymes involved in the syn-
thesis of the phytohormone jasmonic acid, play important roles inplant growth and development, in maturation and senescence, andin the metabolic responses to pathogen attack and wounding (Blée,2002).

400 B.E. Maserti et al. / Journal of Plant Physiology 168 (2011) 392–402

Fig. 3. Expression profiles of differentially expressed proteins in: control leaves; 24 h infested by T. urticae leaves; MeJA treated leaves at 2 h; 6 h; 24 h; 48 h; 72 h. Numbers ofprotein spots correspond to those listed in Fig. 1 and supporting information Tables S1 and S2. Results are representative of mean ± standard deviation from three biologicaland two technical replicates.

B.E. Maserti et al. / Journal of Plant Ph

Fcw

iahwif3ticslai(io

ftftpti2cUpdch

ccltc

Comptes Rendus Biologies 2010;333:516–23.

ig. 4. Gene ontology classification of differentially expressed proteins based onellular component (A) and biological/metabolic pathway (B) as inferred fromww.uniprot.org.

Chitinases are pathogenesis-related (PR) proteins putativelynvolved in resistance against chitin-containing plant pathogensnd pests (Wan et al., 2008). Induction of acidic chitinase mRNAas been observed in C. jambhiri leaves after pathogen infection,ounding and MeJA treatment (Gomi et al., 2002b). Interestingly,

n this study, the levels of two of acidic chitinase isoforms, dif-ering in molecular weight, were affected by both treatments. The3 kDa isoform, whose experimental molecular weight correspondo a theoretical one, was found to be up-regulated after spider mitenfestation and down-regulated beginning at 6 h after MeJA appli-ation. Conversely, the smaller isoform (23 kDa) was found to betrongly up-regulated in both treatments. A much smaller molecu-ar weight of a weevil-induced spot was also observed by Lippert etl. (2007). The role of a PR protein in response to herbivores adopt-ng a piercing/sucking mode of feeding is still unclear. Chen et al.2007) have shown that plant-derived enzymes can remain activen the insect gut, provoking detrimental effects on the developmentf insect herbivores through the degradation of essential nutrients.

Miraculines are highly glycosylated proteins that belong to aamily of protease inhibitors. In this work, higher levels of pro-ein were observed after both treatments. Although their specificunction in the stress response has not yet been elucidated, induc-ion of miraculin-like genes has been reported in many compatibleathogen–plant interactions, including some caused by fungi orreatment with MeJA (Tsukuda et al., 2006), and in C. sinensis leavesnfested by the leafhopper Homalodisca coagulata (Mozoruk et al.,006). Conversely, down-regulation of miraculines was found initrus plants affected by Citrus Sudden Disease (Cantú et al., 2008).pon arthropod attack, Zavala et al. (2004) suggested that potentialroteinase inhibitors such as miraculin-like proteins might impairigestive protease in the insect midgut, resulting in amino acid defi-iencies that negatively affect the growth and development of theerbivore.

The lectin-like protein belongs to a heterogeneous group ofarbohydrate-binding proteins suggested to play a role as insecti-
ides in plant tissue (Van Damme, 2008). In C. clementina, increasingevels of lectin-like proteins were induced by spider mites and MeJAreatment. Induction of lectin genes was found in the compatibleitrus–Citrus leprosis virus interaction (Freita-Astua et al., 2007).
ysiology 168 (2011) 392–402 401

Conclusion

In this study, the proteome-level changes that occur during T.urticae or MeJA treatment in C. clementina leaves were investigatedusing two-dimensional electrophoresis and mass spectrometry toidentify the proteins involved in these interactions. This approachappears to be a powerful way to elucidate proteomic informa-tion in citrus plant, although the citrus sequence genome is notyet available. The identities of some stress-responsive proteinspointed to biochemical processes that may be differentially alteredfollowing challenge by T. urticae infestation or exogenous MeJAapplication. These include enzymes involved in the detoxificationof ROS as well as energy and primary metabolism-related pro-tein and HSPs. The data will stimulate additional investigation toelucidate the role in stress response mechanisms of three defense-related proteins: the pathogenesis-related protein acidic chitinase,the protease inhibitor miraculin-like protein, and a lectin-like pro-tein, which was previously found to be widely overexpressed incitrus-stressed leaves. To the best of our knowledge, this paperestablishes a survey of previously undiscovered changes inducedin the proteome of a citrus plant by T. urticae feeding or MeJAchallenge.

Acknowledgements

Work at IBF, UNISS-INBB and SRA was carried out with finan-cial support from the Provincia di Livorno (Italy) with Mis 3.1 –PIC INTERREG IIIA-Italia – Francia - Isole 2000/2006, research pro-gram CITRUS: Citrus as a model system for the Mediterranean area:study on varieties resistant to biotic and abiotic stresses. 2-DE pro-tein separations were performed at CNR-IBF; Protein identificationby LC–MS/MS was performed at INRA-UR 1199 Protéomique Fonc-tionnelle. The PhD fellowship for A.P. was provided by the Master& Back program (Regione Sardegna, Italy).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.jplph.2010.07.026.

References

Aucejo-Romero S, Gomez-Cadenas A, Jacas-Miret JA. Effects of NaCl-stressed citrusplants on life-history of Tetranychus urticae (Acari: Tetranychidae). Exp ApplAcarol 2004;33:55–67.

Blée E. Impact of phyto-oxylipins in plant defense. Trends Plant Sci 2002;7:315–22.Cantú MD, Mariano AG, Palma MS, Carrilho E, Wulff NA. Proteomic analysis reveals

suppression of bark chitinases and proteinase inhibitors in Citrus plants affectedby the Citrus Sudden Death Disease. Phytopathology 2008;98:1084–92.

Chen H, Gonzales-Vigil E, Wilkerson C, Howe GA. Stability of plant defence proteinsin the gut of insect herbivores. Plant Physiol 2007;143:1954–7.

Desimone M, Henke A, Wagner E. Oxidative stress induces partial degradation ofthe large subunit of ribulose 1,5-biphosphate carboxylase/oxygenase in isolatedchloroplasts of barley. Plant Physiol 1996;111:789–96.

Freita-Astua J, Bastianel M, Locali-Fabris EC, Novelli VM, Silva-Pinhati AC, Basilio-Palmieri AC, et al. Differentially expressed stress-related genes in the compatiblecitrus-Citrus leprosis virus interaction. Genet Mol Biol 2007;30:980–90.

Gaber A, Yoshimura K, Yamamoto T, Yabuta Y, Takeda T, Miyasaka H, et al. Glu-tathione peroxidase-like protein of Synechocystis PCC 6803 confers tolerance tooxidative and environmental stresses in transgenic Arabidopsis. Physiol Plant2006;128:251–62.

Gandia M, Conesa A, Ancillo G, Gadea J, Forment J, Pallàs V, et al. Transcriptionalresponses of Citrus aurantifolia to infection by Citrus tristeza virus. Virology2007;367:298–306.

Giordanengo P, Brunissen L, Rusterucci C, Vincent C, van Bel A, Dinant S, et al.Compatible plant–aphid interactions: how aphids manipulate plant responses.

Gomi K, Yamamoto H, Akimitsu K. Characterization of a lipoxygenase gene in roughlemon induced by Alternaria alternata. J Gen Plant Pathol 2002a;68:21–30.

Gomi K, Itoh N, Yamamoto H, Akimitsu K. Characterization and functional analysisof Class I and II acidic chitinase cDNA from rough lemon. J Gen Plant Pathol2002b;68:191–9.


http://dx.doi.org/10.1016/j.jplph.2010.07.026

4 lant Ph

H

K

L

L

L

L

M

M

M

M

M

resistance to the brown planthopper (Nilaparvata lugens): a proteomic approach.Proteomics 2009;9:2798–808.

02 B.E. Maserti et al. / Journal of P

owe GA, Jander G. Plant immunity to insect herbivores. Annu Rev Plant Biol2008;59:41–66.

ant MR, Ament K, Sabelis MW, Haring MA, Schurink RC. Differentially timingof spider-mite direct and indirect defences in tomato plants. Plant Physiol2004;135:483–95.

ee J, Bricker TM, Lefevre M, Pinson SRM, Oard JH. Proteomic and genetic approachesto identify defence-related proteins in rice challenged with the fungal pathogenRhizoctonia solani. Mol Plant Pathol 2006;7:405–16.

i CY, Williams MM, Loh YT, Lee GI, Howe GA. Resistance of cultivated tomato tocell content-feeding herbivores is regulated by octadecanoid-signaling pathway.Plant Physiol 2002;130:494–503.

ippert D, Chowrira S, Ralph SG, Zhuang J, Aeschliman D, Ritland C, et al. Coniferdefense against insects: proteome analysis of Sitka spruce (Picea sitchensis) barkinduced by mechanical wounding or feeding by white pine weevils (Pissodesstrobi). Proteomics 2007;7:248–70.

liso I, Tadeo FR, Phinney BS, Wilkerson CG, Talon M. Protein changes in the albedoof citrus fruits on post-harvesting storage. J Agric Food Chem 2007;55:9047–53.

affei ME, Mithofer A, Arimura G, Uchtenhagen H, Bossi S, Bertea CM, et al. Effects offeeding Spodoptera littoralis on lima bean leaves. III. Membrane depolarizationand involvement of hydrogen peroxide. Plant Physiol 2006;140:1022–35.

artinez-Godoy MA, Mauri N, Juarez J, Marques MC, Santiago J, Forment J, et al. Agenome-wide 20 K citrus microarray for gene expression analysis. BMC Genom2008;9:318.

artins dos Santos Soares A, Freita de Souza T, de Souza Domingues SJ, Jacinto T,Lima Tavares Machado O. Methyl jasmonate promotes the transient reductionof the levels of 2-Cys peroxiredoxin in Ricinus communis plants. Plant PhysiolBiochem 2004;42:543–7.

aserti BE, Della Croce CM, Luro F, Morillon R, Cini M, Caltavuturo L. A general
method for the extraction of citrus leaf proteins and separation by 2D elec-trophoresis: a follow up. J Chromatogr B 2007;849:351–6.
ozoruk J, Hunnicutt LE, Cave RD, Hunter WB, Bausher MG. Profiling transcriptionalchanges in Citrus sinensis (L.) Osbeck challenged by herbivory from the xylem-feeding leaf hopper Homalodisca coagulate (Say) by cDNA macroarray analysis.Plant Sci 2006;170:1068–80.

ysiology 168 (2011) 392–402

Oliveira IC, Brears T, Knight TJ, Alexandra Clark C, Coruzzi GM. Overexpression ofcytosolic glutamine synthetase. relation to nitrogen. light, and photorespiration.Plant Physiol 2002;129:1170–80.

Parra-Lobato MC, Fernandez-Garcia N, Olmos E, Alvarez-Tinaut MC, Gómez-JiménezMC. MeJA induced antioxidant defence in root apoplast from sunflower seedling.Environ Exp Bot 2009;66:9–17.

Rakwal R, Komatsu S. Role of jasmonate in the rice (Oryza sativa L.) self-defensemechanism using proteome analysis. Electrophoresis 2000;21:2492–500.

Reinbothe S, Mollenhauer B, Reinbothe C. JIPs and RIPs: the regulation of plant geneexpression by jasmonate in response to environmental cues and pathogens.Plant Cell 1994;6:1197–209.

Schwachtje J, Baldwin IT. Update on reconfiguration of primary metabolism afterherbivore attack: why does herbivore attack reconfigure primary metabolism?Plant Physiol 2008;146:845–51.

Shi JX, Chen S, Gollop N, Goren R, Goldschmidt EE, Porat R. Effects of anaerobic stresson the proteome of citrus fruit. Plant Sci 2008;175:478–86.

Tsukuda S, Gomi K, Yamamoto H, Akimitsu K. Characterization of cDNAs encod-ing two distinct miraculin-like proteins and stress-related modulation ofthe corresponding mRNAs in Citrus jambhiri. Plant Mol Biol 2006;60:125–36.

Van Damme EJM. Plant lectins as part of the plant defence system against insects.In: Schaller A, editor. Induced plant resistance to herbivory. Springer ScienceBusiness Media; 2008, chapter 14.

Wan J, Zhang XC, Stacey G. Chitin signaling and plant disease resistance. Plant SignalBehav 2008;3:831–3.

Wei Z, Hu W, Lin Q, Cheng XY, Tong MJ, Zhu LL, et al. Understanding rice plant

Zavala JA, Patankar AG, Gase K, Hui DQ, Baldwin IT. Manipulation of endogenoustrypsin proteins a inhibitor production in Nicotiana attenuata demon-strates their function as anti-herbivore defences. Plant Physiol 2004;134:1181–90.

Documents

JOURNAL OF PLANT PHYSIOLOGY 2011 proteome