Expression profiling of two stress-inducible genes encoding for miraculin-like proteins in citrus plants under insect infestation or salinity stress

G

J

M

Ems

Aa

b

c

d

I

a

ARRAA

KBCCMS

I

abtcwcumwe

t

0h

ARTICLE IN PRESS Model

PLPH-51781; No. of Pages 10

Journal of Plant Physiology xxx (2013) xxx– xxx

Contents lists available at ScienceDirect

Journal of Plant Physiology

j o ur na l ho me page: www.elsev ier .com/ locate / jp lph

olecular Biology

xpression profiling of two stress-inducible genes encoding foriraculin-like proteins in citrus plants under insect infestation or

alinity stress

. Poddaa, M. Simili a, R. Del Carratoreb, W. Mouhayac, R. Morillonc, B.E. Maserti a,d,∗

CNR-IBF, Istituto di Biofisica, Dipartimento di Scienze fisiche e tecnologie della materia, ItalyCNR-IFC, Istituto di Fisiologia Clinica, Dipartimento di Scienze Biomediche, Area della Ricerca CNR, Via Moruzzi 1, I-56124 Pisa, ItalyCIRAD-UPR, Amélioration génétique d’espèces à multiplication vegetative, Avenue Agropolis, 34398 Montpellier Cedex 5, FranceCNR-IPP, Istituto di Protezione delle Piante, Dipartimento di Scienze BioAgroAlimentari, Area della Ricerca CNR, Via Madonna del Piano 10, 55100 Firenze,

taly

r t i c l e i n f o

rticle history:eceived 5 June 2013eceived in revised form 1 August 2013ccepted 1 August 2013vailable online xxx

eywords:iotic stressitrusysteine proteaseiraculin-like protein

a b s t r a c t

The expression of two genes, namely Mir1 and Mir3 and the abundance of their encoded proteins, theputative miraculin-like proteins, MLP1 and MLP3, showing similarity to the Kunitz family of proteaseinhibitors, were monitored in the leaves of the citrus variety, ‘Clementine’ after Tetranychus urticae infes-tation and elicitor treatments, or in the leaves of three other diploid citrus: ‘Willow leaf’, ‘Cleopatra’mandarins and ‘Trifoliate’ orange, as well as their respective doubled diploids and the allotetraploidsomatic hybrid ‘FLHORAG1’ under salt stress. RT-PCR and 2-DE indicated that Mir1 and Mir3 and theirproducts were present at low-basal expression in all citrus genotypes. Both genes and products wereinduced in the ‘Clementine’ leaves infested by T. urticae, but a contrasting profile was observed underelicitor treatments. Under salt stress, the two genes showed an expression pattern contrasting eachother and depending on the genotypes. ‘Cleopatra’ mandarin, ‘Trifoliate’ orange and ‘FLHORAG1’ pre-

alinity sented overexpression of Mir3 and MLP3 and decreased levels of Mir1 and MPL1. The opposite behaviourwas found in ‘Willow leaf’ mandarin. The positive correlation of the expression profile of the two geneswith that of a gene encoding a putative apoplastic cysteine protease (CysP) might suggest a possibleinteraction of the respective encoded proteins during the response to biotic stress. Under salt stress, CysPand Mir 1 showed a similar expression pattern but only at transcript level. The possible occurrence ofpost-translational CysP regulation is discussed.

ntroduction

Citrus is one of the major fruit crops in the Mediterraneanrea and its productivity and fruit quality is frequently affectedy biotic and abiotic stress. Among biotic stress, infestation by thewo-spotted spider mite, Tetranychus urticae Koch (Acari: Tetrany-hidae) is often observed in citrus orchards. T. urticae producesebbing and dense colonies on the lower side of leaves, causing

hlorotic spots on the upper side of leaves (Urbaneja et al., 2008). T.rticae may also cause fruit scarring and the decrease in the com-ercial value of fruit (Ansaloni et al., 2008). The impact of T. urticae

Please cite this article in press as: Podda A, et al. Expression profiling of twoplants under insect infestation or salinity stress. J Plant Physiol (2013), htt

as enhanced when associated with salt stress (Aucejo-Romerot al., 2004).

∗ Corresponding author at: CNR-IPP, Istituto per la Protezione delle Piante, Dipar-imento di Scienze Agroalimentari. Tel.: +39 055 5225585; fax: +39 055 5225560.

E-mail address: [email protected] (B.E. Maserti).

176-1617/$ – see front matter © 2013 Elsevier GmbH. All rights reserved.ttp://dx.doi.org/10.1016/j.jplph.2013.08.001

© 2013 Elsevier GmbH. All rights reserved.

Most citrus varieties are salt-sensitive (Arbona et al., 2003), andthe adaptation to soil salinity is intimately related to the use of root-stocks that promote chloride exclusion at the root level (Hussainet al., 2012). Citrus doubled diploid (4x) seedlings were shown tobe more tolerant to mild salt stress than their respective diploid(2x) parents (Saleh et al., 2008). Recently, 4x ‘Rangpur’ lime root-stock was shown to be much more tolerant to water deficit than itsrespective diploid (Allario et al., 2011).

Miraculin-like proteins are a family of glycoproteins havingsequence similarity to native miraculin, a family of proteins withthe particular feature to switch the sour taste to sweet taste(Masuda et al., 1995). Induction of miraculin-like genes is one of themost prominent responses to stress observed in microarray exper-iments (Talon and Gmitter, 2008). The abundance of miraculin-likeproteins have been reported to change in ‘Clementine’ (Citrus

stress-inducible genes encoding for miraculin-like proteins in citrusp://dx.doi.org/10.1016/j.jplph.2013.08.001

clementina Hort. ex Tan.) leaves exposed to infestation of the twospotted-spider mite T. urticae (Maserti et al., 2011), in Sour orange(Citrus aurantium L.) (Tanou et al., 2009) and ‘Cleopatra’ mandarin(Citrus reshni, Hort. Ex Tan.) (Podda et al., 2013) exposed to salt

dx.doi.org/10.1016/j.jplph.2013.08.001

dx.doi.org/10.1016/j.jplph.2013.08.001

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

http://www.elsevier.com/locate/jplph

mailto:[email protected]

dx.doi.org/10.1016/j.jplph.2013.08.001

ING Model

J

2 ant Ph

ssplbhftpuhpetpne

i(iaie

wprmtgs‘gampJtno

ooeatldrscctbelopt(Mso

o

ARTICLEPLPH-51781; No. of Pages 10

A. Podda et al. / Journal of Pl

tress, as well as in ‘Satsuma’ mandarin (Citrus unshiu) after coldtress (Lang et al., 2005). However, the functions of miraculin-likeroteins in the citrus stress response remain unclear. Miraculin-

ike proteins have been suggested to play a defensive role againstiotic stress by limiting cell damage (Tsukuda et al., 2006) due to theigh homology between their amino acid sequences and the Kunitz

amily of proteinase inhibitors. In accordance with this observa-ion, miraculin-like proteins found in the leaves of rough lemonossessed trypsin inhibitor activity (Tsukuda et al., 2006). Mirac-lins play an important role in the regulation of different types ofypersensitive cell death (Xu et al., 2012) which will lead to theroduction of reactive oxygen species (ROS), one of the earliestvents following pathogen infection or elicitor treatment in cul-ured plant cells (Grant and Loake, 2000). Other than hydroperoxideroduction, many other parallel processes, such as the activation ofucleases and proteinases, are involved in hypersensitive cell deathxecution (Montillet et al., 2005).

Proteinase inhibitors can modulate cysteine protease, which isnvolved in the ubiquitin-mediated protein degradation pathwayHabib and Fazili, 2007). Also, cysteine proteases were found to benvolved in the response to biotic and abiotic stresses (Atkinsonnd Urwin, 2012). The abundance of a cysteine protease was foundncreased in salt stressed leaves of ‘Cleopatra’ mandarin (Poddat al., 2013).

The aim of this work was to identify useful genes associatedith defence response to abiotic and/or biotic stress in citruslants, and provide information on their possible function in stressesponse. To achieve this objective, the expression profile of twoiraculin-like protein genes, namely Mir1 and Mir3 was charac-

erized using a semi quantitative PCR analysis, in different citrusenotypes exposed to T. urticae infestation, elicitor treatments oralinity stress. We investigated the expression of both genes inClementine’ infested by T. urticae. The expression profile of the twoenes was also characterized after methyl jasmonate and salicyliccid (SA) treatments. The rationale of this comparison is that jas-onic acid and relative compounds play a major role in promoting

lant defence response to many arthropod herbivores (Howe andander, 2008). On the other hand, SA operates along with other phy-ohormones, such as jasmonic acid or ethylene as part of a signallingetwork playing a role in coordinating an optimal response to singler combination of pest invaders (Bostock, 2005).

In another experiment, three different seedling rootstocks andne variety, were exposed to salt stress: ‘Cleopatra’ mandarin isne of the most salt-tolerant citrus rootstocks and it is a good Cl−

xcluder, explaining the salt-tolerant nature of this rootstock (Zekrind Parsons, 1992). According to Moya et al. (2003), if Cl− absorp-ion is not limited at the root level, the ions are translocated to theeaves via the transpiration stream and cause necrosis and evenefoliation. ‘Trifoliate’ orange (Poncirus trifoliata L., Raf) is a closeelative of Citrus sp. in the Rutaceae family and is used as root-tock to confer tolerance to Citrus Tristeza Virus (CTV) and betterold-tolerance to the citrus scion. However, ‘Trifoliate’ orange isonsidered to be salt-sensitive because it accumulates Cl− fromhe soil. ‘Willow leaf’ mandarin is a variety known to accumulateoth Na+ and Cl− (Mouhaya et al., 2010). It has been included in thexperiment because it is one of the parents of ‘FLHORAG’1 (‘Willoweaf’ + ‘Trifoliate’ orange) which is an allotetraploid somatic hybridbtained by fusion of ‘Willow leaf’ and ‘Trifoliate’ orange proto-lasts (Ollitrault et al., 2000). This genotype inherited the tendencyo accumulate Cl− in roots and leaf from its ‘Trifoliate’ orange parentMouhaya et al., 2010). Finally, to try to determine what functionir1 and Mir3 may have in the response to mite infestation or salt


tress, the expression pattern of the two genes was compared thatf a CysP gene which encodes a cysteine protease.

The transcript profile of the three genes was compared to thatf their encoding proteins, the miraculin-like protein 1 (MLP1) and

PRESSysiology xxx (2013) xxx– xxx

miraculin-like proteins 3 (MLP3) as well as the cysteine protease(CysP) detected using two-dimensional gel electrophoresis (2-DE).The correlation between mRNA and protein level is discussed.

Materials and methods

Plant materials used to investigate Tetranychus urticae infestation

Clementine (Citrus clementina cv. ‘Tomatera’; SRA535) plantswere kindly provided by Station de Recherche SRA, San Giuliano,Corsica, France). Plants were produced by scions excised from asingle mother plant and grafted on ‘Trifoliate’ orange cv. Pomeroy).

Plant materials for salt stress

Diploid (2x) and doubled diploid (4x) seedlings of ‘Willow leaf’mandarin (Citrus deliciosa Ten.; SRA133), ‘Trifoliate’ orange (Pon-cirus trifoliata L., Raf. cv. ‘Pomeroy’; SRA1074), ‘Cleopatra’ mandarin(Citrus reshni Hort. Ex Tan.; SRA948) and the allotetraploid somatichybrid ‘FLHORAG1’ (‘Willow leaf’ + ‘Trifoliate’ orange, SRA951)were kindly provided by Station de Recherche Agronomique INRA-CIRAD (SRA, San Giuliano, Corsica, France). Originally, spontaneous4x were selected at SRA among a population of seedlings from thediploid plants. The ploidy status of 2x and 4x seedlings was checkedand confirmed by flow cytometry according to Aleza et al. (2011).The genetic constitution of genotypes was checked using 5 simplesequence repeat (SSR) according to Froelicher et al. (2008). All 2xand 4x seedlings presented the same SSR profiles on polyacrylam-ide gels as their parental tree attesting their nucellar origin.

Experimental conditions

Plants were grown in separate pots (3L) in commercially avail-able substrate for citrus plants (1/3 sand, 1/3 soil and 1/3 compost),under greenhouse condition for two years. Plants were wateredevery three days and received fertilizer for citrus plants (NPK6.5.11 – Gesal, Italy: 6% N; 5% P2O5; 6% K2O; 0.01% B; 0.002% Cu;0.02% Fe; 0.01% Mn; 0.001% Mo; 0.002% Zn, and all the trace ele-ments were chelated with water soluble EDTA) once a week. Threemonths before the start of the experiments, plants were transferredinto controlled growth chambers for acclimation (temperature:28 ± 1 ◦C; light/dark period: 16/8; relative humidity: 60–80% andPAR: 200 mmol photons m−2 S−1). No other plants were present inthe room at the time of the experiment. T. urticae infestation or elic-itor application experiments were performed in separate growthchambers in which the same environmental conditions were main-tained.

Each experiment was performed in triplicate for control andtreated plants, each of the three biological samples consisting ofa pool of four leaves of the same age, same stage of developmentand similar leaf area. Treatments and leaf harvest were carried outaround 14 p.m. The leaves were carefully washed with double dis-tilled water and immediately frozen in liquid nitrogen and storedat −80 ◦C until analysis.

Spider mite infestation

Infestation by T. urticae was carried out by transferring about50 spider mites collected from naturally infested Clementine plants


onto leaves of the experimental plants. Infested leaves were closedinto nest bags to avoid mite escape. Control leaves were closedinto nest bags without mites. Analysis was performed on leavesharvested at 0 h and 24 h after infestation.

dx.doi.org/10.1016/j.jplph.2013.08.001

ING Model

J

nt Ph

L

lmM1wwt

L

av

S

weosa(

R

ipoPs(

s(wantwBv(PauwacottpeTldus


A. Podda et al. / Journal of Pla

eaf hormonal treatments

Treatments were performed by spraying the upper and lowereaf faces with an aqueous solution of 1% (v/v) glycerol supple-

ented with 100 �M methyl jasmonate or 1 mM salicylic acid.ethyl jasmonate or salicylic acid was diluted in water from a

000-fold stock solution in 96% ethanol. Control leaves were treatedith aqueous solution containing 1% (v/v) glycerol. Leaf materialas harvested at 2 h, 6 h, 24 h, 48 h and 72 h after methyl jasmonate

reatment or at 0 h, 24 h after salicylic acid treatment.

eaf wounding

Wounding was performed by scratching the citrus leaves withutoclaved carborundum (Geonatura, Spain). Leaves were har-ested at 0 h and 24 h after wounding.

alt stress experiments and mineral analysis

After one month of acclimation in growth chambers, plantsere watered by soaking the pots for one hour, twice weekly,

ither in water supplemented with half-diluted nutrient solution,r in 50 mm NaCl solution supplemented with half-diluted nutrientolution. The experiments lasted eight weeks. Sodium and chloridenalyses were carried out on dried leaves according to Saleh et al.2008).

NA extraction and RT-PCR analysis

The cDNA sequences used for Mir1 and Mir3 and CysP were foundn genebank starting from the peptides matching the miraculin-likeroteins 1 and 3 and the cysteine protease identified in a previ-us proteomic analysis (Supplementary Table S1). The web basedrimer3 software provided by the Whitehead Institute at Mas-achusetts Institute of Technology (was used to design the primersSupplementary Table S2).

Total RNA extraction was performed modifying the methoduggested for the Taqman Gene Expression Cells-to-CT TM KitApplied Biosystems). Ten mg of liquid-nitrogen frozen leavesere extracted with PBS; 5 �L of the extracted suspension was

dded to lysis buffer plus DNAse to exclude genomic contami-ation as described by the manufacturer. Two �L of the reverseranscribed samples was used for PCR amplification performedith GoTaq Green Mastermix (Promega, USA). As the Appliediosystems extraction method was performed for cell culture wealidated the method using a standard RNA extraction procedureQuiagen). The following standard thermal profile was used for allCRs: 94 ◦C for 3 min; 35 cycles of 90 ◦C for 30 s, 59 ◦C for 40 s,nd 72 ◦C for 40 s; 72 ◦C for 7 min as final extension. PCR prod-cts were separated by 1% agarose gel electrophoresis and stainedith ethidium bromide. cDNA fragments were purified from gels

nd sequenced by BMR-Genomics (Italy). For each PCR, the con-entration of cDNA and the number of cycles were optimized tobserve a quantifiable signal within the linear range of amplifica-ion, according to the putative level of each mRNA amplified andhe size of the corresponding PCR product. Control reactions wereerformed in the absence of reverse transcriptase. Transcript lev-ls were compared with citrus elongation factor 1-alpha gene (EF1).hree independent biological replicates were used. The transcript


evels were expressed as mean ± SD. The statistical significance ofifferential findings between samples was determined by ANOVAsing NIA software (Sharov et al., 2005). P < 0.05 was consideredtatistically significant.

PRESSysiology xxx (2013) xxx– xxx 3

Proteomic analysis

Proteomic analysis on ‘Clementine’, ‘Cleopatra’ mandarin, ‘Wil-low leaf’ mandarin, ‘Trifoliate orange’ and ‘FHLORAG1’, 2-DE imagedetection and statistical analysis were performed as described byMaserti et al. (2011), Podda et al. (2013). Detailed information areavailable in the ‘Proteomic method paragraph’ reported in Supple-mentary material.

Results

Two-dimensional gel electrophoresis (2-DE) of protein leaves

Total proteins were extracted and separated by 2-DE from allsamples (Supplementary Fig. S1). The normalized % volume of thespots matching with MLP1, MLP3 and CysP were detected and theselected spots were subjected to mass spectrometry to identify theaminoacid sequences.

Sequence analysis of the Mir, Mir3 and CysP genes

RT-PCR analyses were used to monitor the expression of thetwo Mir and CysP genes in citrus leaves. Primers were designedusing the peptide sequences of MLP1, MLP3 and CysP found by 2-DEanalysis (Supplementary Table S1 and S2). Gene sequence analysisshowed a cDNA of 159 bp with 100% homology with Mir1 gene fromC. jambhiri (AB213395), a 109 bp cDNA with 100% homology withMir3 gene (AF283534) from C. paradisi and a cDNA of 633 bp with100% homology with a CysP gene (EF690284) from C. sinensis.

The miraculin-like proteins encoded by Mir1 and Mir3 con-tained highly conserved motifs typical of the Soybean trypsininhibitor Kunitz family (Supplementary Fig. S2). The alignment ofthe encoded proteins of the two citrus genes with the miraculinfrom Richadella dulcifica revealed that the proteins have secondarystructural idiosyncratic traits similar to that of miraculin (Supple-mentary Fig. S3).

Analysis in the peptidase database MEROPS(http://merops.sanger.ac.uk) showed that the cysteine pro-tease encoded by CysP belongs to the clan CA, family C1, subfamilyC01. This group of peptidases are known to have structures similarto that of papain.

A putative signal peptide for secretion was found in MLP1, MLP3and CysP (Supplementary Table S3a–c) using SecreTary the soft-ware tools available at solgenomics.net.

Analysis performed in the Plant Small RNA Target Analy-sis Server http://www.plantgrn.org/psRNATarget showed that thethree genes might be putatively target by miRNAs (SupplementaryTable S4).

Mir1 and Mir3 expression after T. urticae infestation

Semi-quantitative RT-PCR analysis showed constitutive expres-sion of both Mir1 and Mir3 in ‘Clementine’ leaves prior T. urticaeinfestation (Fig. 1). Twenty-four hours after spider mite infesta-tion both Mir1 and Mir3 transcripts were 2-fold higher than in thecontrol leaves.

Mir1 and Mir3 expression after stress elicitor treatments and leafwounding

The effect of a class of elicitors known to have a role in plantresponse to insect/arthropod attacks was assessed via the expres-


sion profile of the two genes. Twenty-four hours after methyljasmonate treatment, Mir1 expression was increased to a levelsimilar to that induced by T. urticae. Compared to control, thewounding treatment promoted the expression of Mir1 and the

dx.doi.org/10.1016/j.jplph.2013.08.001

http://merops.sanger.ac.uk/

http://www.plantgrn.org/psRNATarget

ARTICLE IN PRESSG Model

JPLPH-51781; No. of Pages 10

4 A. Podda et al. / Journal of Plant Physiology xxx (2013) xxx– xxx

F , �) anutivel

i .05; *

dMi(

Tt

wRe7sa

S

sbtcdS‘(dpa(s2aaF

F(

ig. 1. RT-PCR products of Mir1, Mir3 and CysP in control leaves of C. clementina (C), wounding (W, ), or 100 �M methyl jasmonate (MeJA, ) treatments. Constit

ndicate the mean value (n = 3) ± SD. Asterisks indicate significant differences (*P < 0

own regulation of Mir3, while the salicylic acid had no effect onir1 expression. Salicylic acid and methyl jasmonate treatments

nduced over expression and down regulation of Mir3, respectivelyFig. 1).

ime-course of Mir1 and Mir3 expression after methyl jasmonatereatment

A time-course experiment in which citrus leaves were sprayedith a solution of 100 �M methyl jasmonate was performed.esults showed that the level of Mir1 transcripts increased lin-arly during the first 48 h and then decreased at constitutive level2 h after methyl jasmonate treatment. Conversely, Mir3 expres-ion decreased from 6 h to 48 h after methyl jasmonate treatmentnd increased again at 72 h (Fig. 2).

odium and chloride leaf contents in 2x and 4x citrus genotypes

One year old 2x and doubled diploid seedlings (4x) wereubjected to a 50 mM salt stress for 8 weeks in a growth cham-er. At the end of the experiment 2x ‘Trifoliate’ orange washe only genotype that presented leaf fall and symptoms ofhlorosis. 4x “Trifoliate” orange, ‘Cleopatra’ and ‘Willow leaf’ man-arin genotypes and ‘FLHORAG1’ did not present any symptoms.odium accumulation was very low in salt treated leaves of

Cleopatra’ mandarin, ‘Trifoliate’ orange genotypes and ‘FLHORAG1’Supplementary Fig. S4A, E and G). Conversely ‘Willow leaf’ man-arin genotypes accumulated Na+ (Supplementary Fig. S4C), 4xlants significantly accumulating more sodium than 2x. No Cl−

ccumulation was observed in ‘Cleopatra’ mandarin genotypesSupplementary Fig. S4B). ‘Trifoliate’ orange and ‘FLHORAG1’ pre-ented a significant increase of Cl− content that was similar in


x and 4x genotypes (Supplementary Fig. S4F and H). Chlorideccumulation was limited in ‘Willow leaf’ mandarin genotypes,ccumulation being similar in 2x and 4x plants (Supplementaryig. S4D).

ig. 2. RT-PCR products Mir1 and Mir3 monitored at 0, 6, 24, 48 and 72 h after 100 �m MeEF1) was amplified as a reference gene. Vertical bars indicate the mean value (n = 3) ± SD

d after 24 h of infestation by Tetranychus urticae (T.ur, ), 1 mM salicylic acid (SA,y expressed elongation factor (EF1) was amplified as a reference gene. Vertical bars*P < 0.01).

Mir1 and Mir3 expression in 2x and 4x citrus genotypes subjectedto salt stress

Constitutive expression of Mir1 and Mir3 was found indepen-dent of the genotypes (Fig. 3). Higher transcript levels of Mir1 andMir3 were found in control leaves of 4x mandarin genotypes com-pared to the respective 2x, but not in ‘Trifoliate’ orange. Whenplants were exposed to mild salt stress, Mir1 transcript levelsmarkedly decreased in both 2x and 4x ‘Cleopatra’ mandarin, in2x and 4x ‘Trifoliate’ orange and in the allotetraploid ‘FLHORAG1’,whereas Mir1 was strongly induced in ‘Willow leaf’ mandarin,especially in the 4x genotype. A different expression profile wasobserved for Mir3 which was induced by salt stress in 2x and 4x‘Cleopatra’ mandarin and ‘Trifoliate’ orange as well as in the allote-traploid ‘FLHORAG1’, whereas Mir3 decreased in 2x and 4x ‘Willowleaf’ mandarin genotypes.

Expression of CysP

CysP was present at constitutive level in ‘Clementine’ leavesand it was strongly induced by T. urticae, salicylic acid, methyljasmonate and in a more limited level by artificial wounding (Fig. 1).

Salt stress treatment induced down regulation of CysP in‘Cleopatra’ mandarin and ‘Trifoliate’ orange, while it induced overexpression of the gene in ‘Willow leaf’ mandarin. In control con-dition, CysP was expressed in all genotypes, independently of theploidy (Fig. 3).

Abundance values of the encoded proteins MLP1, MLP3 and CysPafter T. urticae infestation and elicitor treatments

The three proteins were present at basal levels in control leavesof ‘Clementine’, however, in infested leaves the abundance of MLP1,


MLP3 and CysP increased. Salicylic acid did not affect the levels ofMLP1and cysteine protease. Higher levels of MLP3 were found inSA-treated leaves respect to control ones. Wounding and did notaffect MLP3 levels (Fig. 4).

JA application in leaves of C. clementina. Constitutively expressed elongation factor. Asterisks indicate significant differences (*P < 0.05; **P < 0.01).

dx.doi.org/10.1016/j.jplph.2013.08.001

Please cite this article in press as: Podda A, et al. Expression profiling of two stress-inducible genes encoding for miraculin-like proteins in citrusplants under insect infestation or salinity stress. J Plant Physiol (2013), http://dx.doi.org/10.1016/j.jplph.2013.08.001



A. Podda et al. / Journal of Plant Physiology xxx (2013) xxx– xxx 5

Fig. 3. RT-PCR products of Mir1, Mir3 and CysP in leaves of control and salt-treated diploid (2x), and doubled diploid (4x) ‘Cleopatra’ mandarin, ‘Willow leaf’ mandarin, ‘Trifo-liate’ orange and the allodoubled diploid (4x) FLHORAG1 (‘Willow leaf’ × ‘Trifoliate’ orange). Moderate salinity stress (50 mm NaCl) was performed for 8 weeks. Constitutivelyexpressed elongation factor (EF1) was amplified as a reference gene. Vertical bars indicate the mean value (n = 3) ± SD. Asterisks indicate significant differences (*P < 0.05;**P < 0.01).

Fig. 4. Close-up regions from the Coomassie stained 2D gels and abundance levels of MLP 1, MLP 3and CysP in Clementina leaves after control (C), T. urticae (T.ur) infestation,salicylic acid (SA), wounding (W) and methyl jasmonate (MeJA) treatments. Vertical bars indicate the mean log10 value (n = 3) ± SE. Asterisks indicate significant differences(*P < 0.05; **P < 0.01).

dx.doi.org/10.1016/j.jplph.2013.08.001



6 A. Podda et al. / Journal of Plant Physiology xxx (2013) xxx– xxx

F MLP

b ces (*

At

jed(

Aa

wpstalc

t2

D

sfitomvGgtnueh(

mpcbp

Ep

a

ig. 5. Close-up regions from the Coomassie stained 2D gels and abundance levelsars indicate the mean log10 value (n = 3) ± SE. Asterisks indicate significant differen

bundance values of the MLP1, MLP3 during methyl jasmonateime-course experiment

The abundance profiles of MLP1 and MLP3 during methylasmonate treatment are similar to the transcript profiles. How-ver the abundance values of MLP3 did not showed a significantifference when compared to those from the control leavesFig. 5).

bundance values of the encoded proteins MLP1, MLP3 and CysPfter salt stress

MLP1, MLP3 and CysP proteins were found in the control leavesith no significant differences in abundance between 2x and 4xlants. Under salt stress the abundance values of MLP1 and MLP3howed a behaviour similar to the transcript. It is worth nothinghat the ratio between the levels of the proteins in salt-treatednd control leaves not always reached the set threshold (1-foldog10 value) and P < 0.05 to considering the protein as differentiallyhanged.

On the contrary to transcript profile, CysP levels increased in allhe genotypes, but not significantly in 2x Willow leaf mandarin andx Trifoliate orange (Fig. 6).

iscussion

Citrus crops may undergo exposure to several biotic and abiotictresses, which may strongly affect the production and quality ofruit. Hence, information on citrus defence mechanisms is of greatmportance for plant breeders and for promoting the citrus indus-ry. Actually, induction of miraculin-like gene expression is onef the most prominent responses observed in microarray experi-ents performed in open field experiments, including fungal and

iral infection as well as cold, drought and salt stress (Talon andmitter, 2008; Agustí et al., 2009). However, the function of thoseenes in citrus tissues is still unclear. This study is the first attempto investigate the expression profiles of two miraculin-like genes,amely Mir1 and Mir3 and the abundance of their respective prod-cts MLP1 and MLP3 in citrus leaves under arthropod infestation,licitor treatments or salt stress. The selected genes showed 100%omology with the cDNA of miraculin-like proteins of rough lemonC. jambhiri L.) and grapefruit (C. paradisi Macf).

Additionally, to provide information on the possible function ofiraculin-like proteins in citrus response to stress, the expression

attern of Mir1 and Mir3 and the abundance of their products wasompared with that of CysP, a cysteine protease encoding gene,ecause MLP1 and MLP3 showed homology with a Kunitz-typerotease inhibitors.

xpression of Mir1 and Mir3 and abundance of their encoded


roteins in control leaves

The constitutive levels of Mir genes found in all genotypes are inccordance with the basal expression of Mir genes found in mature

1, MLP 3and CysP in Clementina leaves after time-course MeJA treatment. VerticalP < 0.05; **P < 0.01).

‘Satsuma’ mandarin fruits (Fujii et al., 2008), but not in the leavesof rough lemon (C. jambhiri L.). The basal constitutive level of CysP,a protease belonging to CA1 family, found in the citrus leaves inde-pendently to the genotype might be useful for maintenance of plantmetabolism, remodelling of cell protein components, degradationof damaged or unnecessary proteins and nutrient remobilization(Grudkowska and Zagdanska, 2004) and could be modulated byprotease inhibitors (Habib and Fazili, 2007). In accordance withthis hypothesis the encoded proteins MLP1 and MLP3 showedaminoacid motifs similar to that of Kunitz protease inhibitor andidiosyncratic traits similar to other miraculins (Costa Mondegoet al., 2011). The highest levels of stress-related Mir gene transcriptsin 4x genotypes might be involved in the higher tolerance to saltstress or water shortage of these genotypes respect to the cognate2x (Saleh et al., 2008; Mouhaya et al., 2010). Although changes inthe genome expression in double diploid citrus plants were verylimited compared to the cognate 2x genotype (Allario et al., 2011),in Arabidopsis thaliana, it was found that after gene duplication,ancestral stress responses tend to be retained (Zou et al., 2009).The lack of correlation between mRNA and protein profiles in 2xand 4x citrus plants might be due to post-transcriptional, transla-tional and/or post-translational regulations. It is known that mRNAis not always translated into protein, and the amount of proteinproduced for a given amount of mRNA depends on the gene it istranscribed from and on the current physiological state of the cell(Vogel and Marcotte, 2012).

Expression of Mir1 and Mir3 and abundance of their encodedproteins after T. urticae infestation and elicitor treatments

‘Clementine’ is sensitive to outbreaks of spider mite, espe-cially when grown in dry climates (Aucejo-Romero et al., 2004).Both Mir1 and Mir3 genes and their relative proteins, MLP1 andMPL3 were strongly induced by T. urticae. The good correlationbetween Mir 1 and Mir 3 transcript levels and their encoded pro-tein abundance indicates that a substantial proportion of proteinchanges is a consequence of changed mRNA levels, rather thanpost-transcriptional effects. Induction of miraculin-like transcriptshas been reported in many pathogen–plant interactions includingsome caused by fungi (Tsukuda et al., 2006). Additionally, induc-tion of transcripts and increased abundance of MLP1 was foundin sweet orange in response to Candidatus Liberibacter asiaticus(Fan et al., 2011). On the other hand, Cantú et al. (2008) reportedsuppression of miraculin-like protein in the bark of Citrus suddendisease-susceptible plants, but not in the tolerant ones. To defendthemselves from pest attacks, plants synthesize protease inhibitors,which may be involved in a defence strategy against insect andmany phytopathogenic microorganisms (Habib and Fazili, 2007).It seems that the defensive capabilities of protease inhibitors rely


upon the inhibition of proteases present in insect guts or secretedby microorganisms, causing a reduction in the availability of aminoacids necessary for their growth and development (Lawrence andKoundal, 2002).

dx.doi.org/10.1016/j.jplph.2013.08.001



A. Podda et al. / Journal of Plant Physiology xxx (2013) xxx– xxx 7

F LP 1, Mi (*P < 0

prisadob

ig. 6. Close-up regions of the Coomassie stained 2D gels and abundance levels Mndicate the mean log10 value (n = 3) ± SE. Asterisks indicate significant differences

To survive under adverse conditions, plants have evolved com-lex mechanisms to perceive external signals and start optimalesponse to environmental conditions. Phytohormones such as sal-cylic acid and compounds of jasmonic acid regulate the protectivetress response of plants to biotic and abiotic stresses. Thus, Mir1


nd Mir3 expression was also analyzed for their response to threeifferent stress elicitors in ‘Clementine’ leaves. The strong inductionf Mir1 in the leaves treated with methyl jasmonate and wounding,ut not in those treated with salicylic acid is in accordance with the

LP 3 and CysP in Clementina leaves after eight weeks of salt stress. Vertical bars.05; **P < 0.01).

observations of Tsukuda et al. (2006) and Huang et al. (2007). On thecontrary, Mir3 expression was promoted SA, while it was stronglydown regulated by wounding and MeJA. This MeJA inhibitor effectseems to be confirmed by the time-course experiment, duringwhich the Mir3 expression was restored only after 72 h from methyl


jasmonate treatment. Intriguingly the MLP3 abundance was notaffected by MeJA treatment at protein level and that could beexplained after further investigation on its regulation at proteinlevel. Recent studies in plant immunity have provided new insights

dx.doi.org/10.1016/j.jplph.2013.08.001

ING Model

J

8 ant Ph

idjtirwsmuunu

E2

tmo(tit

taaoeaiim‘tosstcCi2Foab

dwti

d

wPwMAoc‘i


A. Podda et al. / Journal of Pl

nto the underlying defence signalling networks suggesting thatifferent signalling pathways, such as those of salicylic acid and

asmonic acid cross-communicate in an antagonistic or synergis-ic manner, providing the plants with a powerful ability to tunets immune response (Pieterse et al., 2009). On the other hand, in aecent work on Coffea miraculin expression it was shown that CoMiras up-regulated after abscisic acid application and wounding

tress. However, the same gene was down-regulated by coffee leafiner herbivore (Costa Mondego et al., 2011). The differential mod-

lation of Mir1 and Mir3 genes, as well as of their encoded proteinsnder hormonal treatments indicates that a complex regulatoryetwork might modulate miraculin expression patterns in citrusnder biotic stress (Taheri et al., 2011).

xpression of Mir1 and Mir3 and abundance of MLP1 and MLP3 inx and 4x genotypes consecutively to a salt stress

Salt stress experiments were performed on four citrus geno-ypes presenting different tolerance to salt stress. Using SSRs

arkers, rootstocks seedlings (‘Cleopatra’ mandarin, ‘Trifoliate’range and ‘FLHORAG1’) as well as seedlings of a mandarin variety‘Willow leaf’) were shown to be identical to their parental motherree. In this salt stress experiment, we did not use ‘Clementine’ sincet is self-incompatible and plants from seeds will be different fromhe parental mother tree.

In salt stress conditions, 2x ‘Trifoliate’ orange was the only geno-ype showing salt stress symptoms such as leaf fall and chlorosis,s already reported by Saleh et al. (2008). Measurements of ionsccumulation showed that ‘Cleopatra’ mandarin and ‘Trifoliate’range rootstocks were Na+ excluders in accordance with Saleht al. (2008). The variety ‘Willow leaf’ accumulated mainly Na+ butlso Cl−. ‘FLHORAG1’ inherited the capacity to exclude Na+ fromts ‘Trifoliate’ orange parent +. In citrus trees, Cl− is the most toxicon and salt tolerance is correlated to the ability of rootstocks to

inimize Cl− accumulation (Hussain et al., 2012). Interestingly, 2xTrifoliate’ orange showing leaf chlorosis, also have a high Cl− con-ent. However, even though Cl− contents for 2x and 4x ‘Trifoliate’range genotypes were similar, no leaf fall and symptom of chloro-is were observed in 4x. In ‘FLHORAG1’, that did not display a salttress phenotype, Cl− accumulation did not differ significantly fromhat of 2x and 4x ‘Trifoliate’ orange genotypes. Investigation of ROSontents and detoxification enzymes in 4x Dioscorea zingiberensis.H. Wright indicated that 4x plants possessed a stronger antiox-

dant defense system and increased heat tolerance (Zhang et al.,010). Stronger antioxidant defense system in 4x genotypes andLHORAG1 may explain why no symptoms of leaf chlorosis werebserved compared to 2x ‘Trifoliate’ orange, even though chloridend sodium leaf accumulation were similar as previously reportedy Mouhaya et al. (2010).

Salt stress affected both Mir genes expression and MLPs abun-ance in citrus leaves, in accordance with Dombrowski (2003)hich discovered that salt stress causes the accumulation of pro-

einase inhibitors and the activation of other wound-related genesn tomato plants.

Interestingly, Mir1 and Mir3 displayed different regulationepending on genotype.

Cleopatra’ mandarin rootstock is the most salt tolerant genotypeith very limited accumulation of Cl− and Na+ (Hussain et al., 2012;

odda et al., 2013). In our hands, ‘Cleopatra’ mandarin seedlings,hich are Cl− or Na+ excluders, showed slight down regulation ofir1 and an over expression of Mir3 at transcript and protein level.lthough the MLP1 decrease did not reach the set threshold. In Sour


range (C. aurantium L.) subjected to salt stress, three forms of mira-ulin were shown to decrease under salt stress (Tanou et al., 2009).Sour’ orange is known as a salt tolerant rootstock, even thought is worth noting that the tolerance to salt stress of seedlings of


this genotype is lower than that of ‘Cleopatra’ mandarin seedlings(Hussain et al., 2012). Conversely, another genotype belonging tothe mandarin group, ‘Willow leaf’ mandarin, presented an oppo-site profile for Mir1 and Mir3, as well as their respective products.Interestingly, this genotype was shown to be salt sensitive andaccumulated Cl− and Na+ in leaves. On the other hand, ‘Trifoliate’orange belongs to the Poncirus genus and is one of the most sen-sitive genotype to salt stress due to the accumulation of large Cl−

contents. However, in this genotype the Mir1 and Mir3 expressionas well as the abundance profiles of the encoded proteins behave asin the ion excluder ‘Cleopatra’ mandarin. Cl− accumulation inducesoxidative stress, which may lead to a change of miraculin expres-sion and perturbation of metabolism. In stress condition, miraculinswere shown to be involved in the regulation of different types ofhypersensitive cell death (Xu et al., 2012) that will lead to ROS pro-duction (Grant and Loake, 2000). ‘FLHORAG1’ which inherited fromits parents, ‘Trifoliate’ orange and ‘Willow leaf’ mandarin, the prop-erty to accumulate low Na+ and high Cl− did not present symptomsof chlorosis like its 4x ‘Trifoliate’ orange parent. The lack of chlorosissymptoms in this genotype suggests that sensitivity traits are notonly due to ion accumulation, but by other mechanisms such asa stronger antioxidant defense system which may help to preventleaf chlorotic symptoms. In stress conditions, Mir1 and Mir3 expres-sions and MLP1 and MLP3 in ‘FLHORAG1’ were very similar to itsPoncirus parent and to the ‘ion excluder’ ‘Cleopatra’ mandarin. Thisfinding might suggest that the switch on/off of the two Mir trans-cripts may be dependent on the toxic ion content of cells and/orother effects consecutive to salt stress. To support this hypoth-esis, it was found in several plants that salt stress did not alterthe resistance to the herbivore Trichoplusia ni, but did alter thenegative signal interaction between the jasmonate and salicylateresponses (Thaler and Bostock, 2004). Dombrowski (2003) indi-cated that salt stress-induced accumulation of proteinase inhibitorsin tomato plants was jasmonic acid dependent.

As hypothesized for MLP3 after MeJA treatment, the similaritybetween the Mir 1 and Mir 3 transcript and protein profiles, but thelack of significant changes in protein abundances under salt stressmight due to translational mechanisms (Vogel and Marcotte, 2012).

Comparison of Mir1 and Mir3 profiles with those of the cysteineprotease encoding gene (CysP)

Protease inhibitors, such as miraculin-like proteins are involvedin the endogenous defence system as they help to regulate andbalance protease activities. Proteases are enzymes that catalyzethe hydrolytic cleavage of specific peptide bonds in their targetproteins and play key roles in many biological processes and abioticand biotic stress response (Atkinson and Urwin, 2012). Salicylicacid was shown to potentiate cell death and protease activityinduced by oxidative stress or by pathogen attack (Shirasu et al.,1997). Thus the increased CysP expression and the abundance of theencoded protein induced by T. urticae, wounding, salicylic acid andmethyl jasmonate in n ‘Clementine’ were expected. The correlationbetween Mir1 transcripts, their products and those of CysP claimedfor a complementary function of the two genes under biotic stress.A role of protease inhibitors and proteases in response to ROSproduction and in the regulation of programmed cell dead in plantsunder stress have been suggested (Solomon et al., 1999). Usinga dedicated software, a signal peptide for secretion was found inMLP1 and MLP3, in accordance with Takai et al., 2013, and in CysP.van der Linde et al. (2012) reported that apoplastic Cys proteasesplay a pivotal role in maize defence signalling and attributed a


pivotal role to the fine-tuning of protease activity in plant cell deathregulation associated with the pathogen response by the interplayof activating triggers and inhibitors. A possible interaction betweenMLPs and CysP in the apoplastic space to constitute a barrier

dx.doi.org/10.1016/j.jplph.2013.08.001

ING Model

J

nt Ph

aIsapUoiTetsm(ioapcl

eitnuii3aCtnbm

A

fj

R

A

A

A

A

A

A

A

B

C


A. Podda et al. / Journal of Pla

gainst pest attack should be assessed by further investigations.n particular, an important challenge will be to identify theignal(s) that are released or removed by extracellular proteasectivity into apoplast and to understand how such signals areerceived and transduced into downstream defence cascades.nder salt stress condition, the CysP expression correlates to thatf Mir1, however, the abundance of its encoding protein showedncreased profiles also when the transcript is down-regulated.he poor correlation between transcript and protein could bexplained with the occurrence of post-transcriptional and transla-ional processes (Vogel and Marcotte, 2012). The analysis of CysPequence revealed the presence of a motif complementary to theiRNA 1515, found in C. sinensis putatively targeting CysP mRNA

Xu et al., 2010). A number of work focussed to identify factorsnvolved in stress responses, has shown that different elementsf RNA metabolism and in particular small RNA molecules knowns microRNAs (miRNAs), are active players in these responses,articipating in the regulation of different pathways selected toope with environmental stress conditions, including but notimited to, drought and salinity (Covarrubias and Reyes, 2010).

In summary, the results described in this work show that thexpression of Mir 1, Mir 3 and CysP and their encoding proteinsn citrus leaves is affected in different way by arthropod infes-ation, elicitor treatments and salt stress. Also the study raises aumber of new questions on the Mir and CysP genes and their reg-lation in citrus plants under abiotic and biotic stress: the issues

nclude (i) whether salicylic and jasmonate pathways are involvedn the modulation of Mir 1 and Mir 3 genes; (ii) whether Mir 1, Mir

and CysP interact each other to modulate the effects of abioticnd biotic stress; (iii) whether the negative correlation betweenysP expression profile and its encoding protein is due to miRNA/sargeting the CysP transcripts. Data present in this work stimulateew investigations to understand the molecular function playedy miraculin-like genes in the response of citrus plants to environ-ental stress.

ppendix A. Supplementary data

Supplementary data associated with this article can beound, in the online version, at http://dx.doi.org/10.1016/.jplph.2013.08.001.

eferences

gustí J, Merelo P, Cercós M, Francisco R, Tadeo FR, Talón M. Comparative trans-criptional survey between laser-microdissected cells from laminar abscissionzone and petiolar cortical tissue during ethylene-promoted abscission in citrusleaves. BMC Plant Biol 2009;9:127.

leza P, Froelicher Y, Schwarz S, Agusti M, Hernandez M, Juarez J, et al. Doubleddiploidization events by chromosome doubling of nucellar cells are frequentin apomictic citrus and are dependent on genotype and environment. Ann Bot2011;108:37–50.

llario T, Brumos J, Colmenero-Flores JM, Tadeo F, Froelicher Y, Talon M, et al. Largechanges in anatomy and physiology between diploid Rangpur lime (Citrus limo-nia) and its autodoubled diploid are not associated with large changes in leafgene expression. J Exp Bot 2011;62:2507–19.

nsaloni T, Pascual-Ruiz S, Hurtado MA, Jacas JA. Can summer and fall vegetativegrowth regulate the incidence of Tetranychus urticae Koch on clementine fruit?Crop Prot 2008;27:459–64.

rbona V, Flors V, Jacas J, Garcıa-Agustın P, Gomez-Cadenas A. Enzymatic andnon-enzymatic antioxidant responses of Carrizo citrange, a salt-sensitive citrusrootstock, to different levels of salinity. Plant Cell Physiol 2003;44:388–94.

tkinson NJ, Urwin PE. The interaction of plant biotic and abiotic stresses: fromgenes to the field. J Exp Bot 2012;30:1–21.

ucejo-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.


ostock RM. Signal crosstalk and induced resistance. Straddling the line betweencost and benefit. Annu Rev Phytopathol 2005;43:545–80.

antú MD, Mariano AG, Palma MS, Carrilho E, Wulff NA. Proteomic analysis revealssuppression of bark chitinases and proteinase inhibitors in citrus plants affectedby the citrus sudden death disease. Phytopathology 2008;98(10):1084–92.

PRESSysiology xxx (2013) xxx– xxx 9

Costa Mondego JM, Pasini Duarte M, Kiyota E, Martínez L, Rodrigues de Camargo S, DeCaroli FP, et al. Molecular characterization of a miraculin-like gene differentiallyexpressed during coffee development and coffee leaf miner infestation. Planta2011;233:123–37.

Covarrubias AA, Reyes JL. Post-transcriptional gene regulation of salinity anddrought responses by plant microRNA. Plant Cell Environ 2010;33:481–9.

Dombrowski JE. Salt stress activation of wound-related genes in tomato plants. PlantPhysiol 2003;132(4):2098–107.

Fan J, Chen C, Yub Q, Brlanskyb RH, Lia Z-G, Gmitter FG Jr. Comparative iTRAQproteome and transcriptome analyses of sweet orange infected by CandidatusLiberibacter asiaticus. Physiol Plant 2011;143:235–45.

Froelicher Y, Dambier D, Costantino G, Lotfy S, Didout C, Beaumont V, et al. Character-ization of microsatellite markers in mandarin orange (Citrus reticulata Blanco).Mol Ecol Resour 2008;8:119–22.

Fujii H, Shimada T, Sugiyama A, Endo T, Nishikawa F, Nakano M, et al. Profilinggibberellin (GA3)-responsive genes in mature mandarin fruit using a citrus 22Koligoarray. Sci Hortic 2008;11:291–8.

Grant JJ, Loake GJ. Role of reactive oxygen intermediates and cognate redox signalingin disease resistance. Plant Physiol 2000;124:21–9.

Grudkowska M, Zagdanska B. Multifunctional role of plant cysteine proteinases. ActaBiochim Pol 2004;51:609–24.

Habib H, Fazili KM. Plant protease inhibitors: a defence strategy in plants. BiotechnolMol Biol Rev 2007;2:068–85.

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

Huang Y, Xiao B, Xiong L. Characterization of a stress responsive proteinaseinhibitor gene with positive effect in improving drought resistance in rice. Planta2007;226:73–85.

Hussain S, Luro F, Costantino G, Ollitrault P, Morillon R. Physiologicalanalysis of salt stress behavior of citrus species and genera: low chlo-ride accumulation is an indicator of salt tolerance. S Afr J Bot 2012.,http://dx.doi.org/10.1016/j.sajb.2012.06.004.

Lang P, Zhang CK, Ebel RC, Dane F, Dozier WA. Identification of cold acclimated genesin leaves of Citrus unshiu by mRNA differential display. Gene 2005;359:111–8.

Lawrence PK, Koundal KR. Plant protease inhibitors in control of phytophagousinsects. Electron J Biotechnol 2002;5:93–109.

Maserti BE, Del Carratore R, Della Croce CM, Podda A, Migheli Q, Froelicher Y, et al.Comparative analysis of proteome changes induced by the two spotted spidermite Tetranychus urticae and methyl jasmonate in citrus leaves. J Plant Physiol2011;168:392–402.

Masuda Y, Nirasawa S, Nakaya K, Kurihara Y. Cloning and sequencing ofa cDNA encoding a taste-modifying protein miraculin. Gene 1995;161:175–7.

Montillet JL, Chamnongpol S, Rustérucci C, Dat J, van de Cotte B, Agnel JP, et al. Fattyacid hydroperoxides and H2O2 in the execution of hypersensitive cell death intobacco leaves. Plant Physiol 2005;138:1516–26.

Mouhaya W, Allario T, Brumos J, Andres F, Froelicher Y, Luro L, et al. Sensitivity tohigh salinity in doubled diploid citrus seedlings increases with water availabilityand correlates with expression of candidate genes. Funct Plant Biol 2010;37:674–85.

Moya J, Gomez-Cadenas A, Primo-Millo E, Talon M. Chloride absorption in salt-sensitive Carrizo citrange and salt-tolerant Cleopatra mandarin citrus rootstocksis linked to water use. J Exp Bot 2003;54:825.

Pieterse CMJ, Leon-Reyes A, Van der Ent S, Van Wees SCM. Networking by small-molecule hormones in plant immunity. Nat Chem Biol 2009;5:308–16 [Review].

Podda A, Checcucci G, Mouhaya W, Centeno D, Rofidal V, Del Carratore R, et al.Salt-stress induced changes in the leaf proteome of diploid and tetraploid man-darins with contrasting Na(+) and Cl(–) accumulation behaviour. J Plant Physiol2013;170(12):1101–12.

Ollitrault P, Dambier D, Seker M, Froelicher Y. Rootstock breeding bysomatic hybridisation for the Mediterranean citrus industry. Acta Hortic2000;535:157–62.

Saleh B, Allario T, Dambier D, Ollitrault P, Morillon R. Doubled diploid citrusrootstocks are more tolerant to salt stress than diploid. C R Biol 2008;331:703–10.

Sharov AA, Dudeka DB, Ko MS. A web-based tool for PC and significance of mircroar-ray data. Bioinformatics 2005;21:2548–9.

Shirasu K, Nakajima H, Rajasekhar VK, Dixon RA, Lamb CJ. Salicylic acid potenti-ates an agonist-dependent gain control that amplifies pathogen signals in theactivation of defence mechanisms. Plant Cell 1997;9:261–70.

Solomon M, Belenghi B, Delledonne M, Menachem E, Levinea A. The involvement ofcysteine proteases and protease inhibitor genes in the regulation of programmedcell death in plants. Plant Cell 1999;11:431–43.

Taheri F, Nematzadeh G, Zamharir M, Nekouei MK, Naghavi M, Mardi M, et al. Pro-teomic analysis of the Mexican lime tree response to ‘Candidatus Phytoplasmaaurantifolia’ infection. Mol BioSyst 2011;7:3028–35.

Takai A, Satoh M, Matsuyama T, Ito A, Nakata R, Aoyama T, et al. Secretion of miraculinthrough the function of a signal peptide conserved in the Kunitz-type soybeantrypsin inhibitor family. FEBS Lett 2013;587:1767–72.

Talon M, Gmitter FG Jr. Citrus Genomics. Int J Plant Genom 2008.,http://dx.doi.org/10.1155/2008/528361.

Tanou G, Job C, Rajjou L, Arc E, Belghazi M, Diamantidis G, et al. Proteomics reveals


the overlapping roles of hydrogen peroxide and nitric oxide in the acclimationof citrus plants to salinity. Plant J 2009;60:795–804.

Tsukuda S, Gomi K, Yamamoto H, Akimitsu K. Characterization of cDNAs encod-ing two distinct miraculin-like proteins and stress-related modulation of

dx.doi.org/10.1016/j.jplph.2013.08.001

http://dx.doi.org/10.1016/j.jplph.2013.08.001

http://dx.doi.org/10.1016/j.jplph.2013.08.001

http://refhub.elsevier.com/S0176-1617(13)00317-9/sbref0005








































































































































































































































































































































































































dx.doi.org/10.1016/j.sajb.2012.06.004







































































































































































































































































































































dx.doi.org/10.1155/2008/528361






































ING Model

J

1 ant Ph

T

U

v

V


0 A. Podda et al. / Journal of Pl

the corresponding mRNAs in Citrus jambhiri Lush. Plant Mol Biol 2006;60:125–36.

haler JS, Bostock RM. Interactions between abscisic-acid-mediated responses andplant resistance to pathogens and insects. Ecology 2004;85:48–58.

rbaneja A, Pascual-Ruiz S, Pina T, Abad-Moyano R, Vanaclocha P, Montón H, et al.Efficacy of five selected acaricides against Tetranychus urticae (Acari: Tetrany-chidae) and their side effects on relevant natural enemies occurring in citrusorchards. Pest Manag Sci 2008;64:834–42.

an der Linde K, Hemetsberger C, Kastner C, Kaschani F, van der Hoorn RA, Kum-


lehn J, et al. A maize cystatin suppresses host immunity by inhibiting apoplasticcysteine proteases. Plant Cell 2012;24:1285–3000.

ogel C, Marcotte EM. Insights into the regulation of protein abundancefrom proteomic and transcriptomic analyses. Nat Rev Genet 2012;13(4):227–32.


Xu Q, Liu Y, Zhu A, Wu X, Ye J, Yu K, et al. Discovery and comparative profilingof microRNAs in a sweet orange red-flesh mutant and its wild type. Genomics2010;11:246.

Xu Q-F, Cheng WS, Li S-S, Zhang Z-X, Zhou X-P, Cai X-Z. Identification of genesrequired for Cf-dependent hypersensitive cell death by combined proteomicand RNA interfering analyses. J Exp Bot 2012;63:2421–35.

Zekri M, Parsons L. Salinity tolerance of citrus rootstocks: effects of salt on root andleaf mineral concentrations. Plant Soil 1992;147:171–81.

Zhang X, Hu C, Yao J. Tetraploidization of diploid Dioscorea results in activation of


the antioxidant defense system and increased heat tolerance. J Plant Physiol2010;167:88–94.

Zou C, Lehti-Shiu MD, Thomashow M, Shiu S-H. Evolution of stress-regulatedgene expression in duplicate genes of Arabidopsis thaliana. PLoS Genet2009;5(7):e1000581, http://dx.doi.org/10.1371/journal.pgen.1000581.

dx.doi.org/10.1016/j.jplph.2013.08.001
















































































































































































dx.doi.org/10.1371/journal.pgen.1000581

Documents

Expression profiling of two stress-inducible genes encoding for miraculin-like proteins in citrus plants under insect infestation or salinity stress