Gene expression in grapevine cultivars in response to Bois Noir phytoplasma infection

Plant Science 176 (2009) 792–804

Gene expression in grapevine cultivars in response to Bois Noir phytoplasmainfection

Giorgia Albertazzi a,*, Justyna Milc a, Alessandra Caffagni a, Enrico Francia a, Enrica Roncaglia b,Francesco Ferrari c, Enrico Tagliafico b, Emilio Stefani a, Nicola Pecchioni a

a Department of Agricultural and Food Sciences, University of Modena and Reggio Emilia, Via Amendola 2 (Padiglione Besta), 42100 Reggio Emilia, Italyb Department of Biomedical Sciences, University of Modena and Reggio Emilia, Via Campi 287, 41100 Modena, Italyc Department of Biology, University of Padova, via G. Colombo 3, 35131 Padova, Italy

A R T I C L E I N F O

Article history:

Received 22 December 2008

Received in revised form 26 February 2009

Accepted 4 March 2009

Available online 18 March 2009

Keywords:

Vitis vinifera

Bois Noir phytoplasma

Affymetrix

Gene profiling

Microarray

A B S T R A C T

Bois Noir phytoplasma is an emerging disease of Vitis vinifera in several regions of the world. No

completely resistant grapevine cultivars are known and the physiology of disease remains still poorly

understood so far.

Affymetrix GeneChip1 oligonucleotide arrays have been used to identify differentially expressed

genes between infected and recovered samples from Chardonnay and between infected and healthy

samples from Manzoni Bianco. In the field, Manzoni showed reduced symptoms, while Chardonnay was

highly susceptible to the disease. Results showed that expression levels of few hundreds genes were

altered in infected plants, both common and specific for each cultivar, with effects on various metabolic

pathways. In Chardonnay a serious inhibition of whole photosynthetic chain and photosystem I activity,

Calvin-cycle enzymes transcription, lipid metabolism and phenylpropanoid biosynthesis was observed.

Increasing physical barriers to limit phytoplsma spread in the plant was observed in infected

Chardonnay and Manzoni plants, with the repression of genes responsible for cell wall degradation and

the induction of genes involved in cell wall reinforcement. Interestingly, specifically in Manzoni the

expression of a Myb transcription factor, belonging to a gene family that has a role in defense response,

was induced.

This is the first analysis of gene expression profiling in a grapevine–phytoplasma interaction using

Affymetrix GeneChip1 array. Presented data provide an interesting picture of the transcriptional

response of grapevine to Bois Noir and allowed the selection of several candidate genes for future

functional analysis.

� 2009 Elsevier Ireland Ltd. All rights reserved.

Contents lists available at ScienceDirect

Plant Science

journal homepage: www.e lsev ier .com/ locate /p lantsc i

1. Introduction

In many countries, Vitis vinifera is severely affected by Bois Noir,a grapevine yellows disease [1]. It is associated with phytoplasma,an unculturable phloem-limited plant pathogen, which belongs tothe Stolbur group (16SrXII-A) [2] and is transmitted by thepolyphagous sap-sucking insect vector Hyalesthes obsoletus Sign-oret (Homoptera, Cixiidae). As with many plant viruses, a systemicinfection involves the movement of phytoplasmas in the networkof phloem sieve tubes, in directions determined by the relativestrengths of the major sinks.

Various symptoms and metabolic changes in the host havebeen associated with phytoplasma infection, including leafyellowing and discoloration, particularly along the vascular

* Corresponding author. Tel.: +39 0522522064; fax: +39 0522522027.

E-mail address: [email protected] (G. Albertazzi).

0168-9452/$ – see front matter � 2009 Elsevier Ireland Ltd. All rights reserved.

doi:10.1016/j.plantsci.2009.03.001

tissue. Such discoloration is due to inhibition of chlorophyllbiosynthesis in the young grapevine leaves, resulting in a net lossof chlorophyll and leaf chlorosis [3]. Reduced stomatal con-ductance has also been observed along with impaired photo-synthesis [4], together with the accumulation of solublecarbohydrates and starch in source leaves and decreased sugarcontent in sink-leaves [5,6]. It is likely that reduction ofphotosynthesis together with imparing of phloem transport areone of the main causes of the other symptoms observed: growthaberrations (proliferations, internode shortening, stunting),flower malformations (size reduction, virescence, phyllody),swollen veins [7], and a drastic drop in yield [8]. In fact,histological studies have shown that some phytoplasmas induceanatomical changes of phloem such as callose deposit on the sieveplates, followed by the collapse of sieve elements, while otherphytoplasmas affect phloem differentiation by increasing ordecreasing the number of sieve elements. Very sensitive hostsmay develop serious phloem necrosis to limit pathogen survival,

mailto:[email protected]

http://www.sciencedirect.com/science/journal/01689452

http://dx.doi.org/10.1016/j.plantsci.2009.03.001

G. Albertazzi et al. / Plant Science 176 (2009) 792–804 793

while others, with a less severe phloem reaction, may sustainhigher pathogen titres in the leaf tissues [2].

Sometimes, plants infected with phytoplasmas may show aspontaneous remission of symptoms, for which the term ‘recovery’has been proposed [9]. Recent studies have begun to explain thephysiological basis of recovery, as well as the possible factors thatmay lead to this phenomenon. A significant increase in the reactiveoxygen species (ROS) has been observed, as well as in the synthesisof signal molecules, such as pathogenesis related (PR) proteins, andthe concentration of H2O2 on the plasmalemma of the sieve tubesincrease through a selective and presumably stable down-regulation of enzymatic H2O2 scavengers [10].

Grapevine plants are sensitive to phytoplasma infection, andsymptoms range from very severe, as in high susceptible cultivars,to barely visible, as in tolerant ones [11,12]. No completelyresistant grapevine cultivars to phytoplasmas are known so far.However, it has been shown that several susceptible host cultivarsare not passive against the pathogen and can set up a defenceresponse which is however not sufficiently effective to stoppathogen replication and dissemination [13,14]. Few studies haveinvestigated the molecular mechanisms involved in the phyto-plasma–plant interaction and the effects of phytoplasmas infectionon gene expression in its host plants. Recent results emerging fromsequencing of two aster yellows phytoplasma strains haveprovided an initial insight into the physiology and requirementsof these organisms [15,16]. Molecular studies of the plant–pathogen interaction have been mainly based on the differentialdisplay of mRNAs, and they all have been reported in theexperimental host plant periwinkle (Catharanthus roseus L.) [17],as well as in Prunus armeniaca [18], tomato [19] and poinsettia(Euphorbia pulcherrima) [20]. Very recently, the expression level ofthree genes (sucrose synthase, alcohol dehydrogenase I and heatshock protein 70) in grapevine plants infected by phytoplasmas,have been investigated by real-time PCR [21]. However, no studieshave been reported so far that investigate the effects ofphytoplasmas infection in grapevine on a genomic scale, even ifthe GeneChip microarrays (Affymetrix) have been used by severalauthors to study other agronomic traits [22–28].

The aim of our study was to investigate the responses elicitedby phytoplasmas in grapevines, specifically in the transcriptprofiles of Chardonnay and Manzoni Bianco. Vines that werenaturally infected with Bois Noir were compared with healthy andrecovered grapevine plants by Affymetrix expression arrays (V.

vinifera genome array).

2. Materials and methods

2.1. Plant material

Two grapevine cultivars were examined: V. vinifera L. Char-donnay (clone R8) and Manzoni Bianco (formerly Incrocio Manzoni6.0.13, a product of a cross between Riesling Renano and PinotBlanc), from here also named as Manzoni. Phenotypical symptomsobservations, aimed to study the susceptibility/tolerance ofChardonnay and Manzoni to phytoplasma infection, were carriedout in summer 2006 on a total of 100 plants of both Chardonnayand Manzoni, in two experimental fields located respectively inCoviolo (Reggio Emilia) and Brescia, Italy.

The leaf samples used for microarray analysis were collectedfrom symptomless (control: recovered Chardonnay and healthyManzoni) and from naturally Bois Noir infected field-grown plantsin one of the two locations, Coviolo-Reggio Emilia, Italy, betweenJune 2006 and August 2006.

Biological replicates corresponded to samples, each constitutedon average of 8 leaves, collected once a week from differentbranches of the same plant.

Each sample was kept separated and screened for presence ofphytoplasmas, to select plant material that was infected by BoisNoir, but negative for Flavescence Doree.

To ensure sampling leaves of about the same age, medial leaves(approximately 6 cm diameter) were harvested. Petioles andcentral leaf midribs were immediately cut out from individualleaves, and then frozen in liquid nitrogen directly in the field forDNA and RNA extraction.

2.2. Phytoplasma detection and identification

Total DNA was extracted from leaf samples (vascular tissues ofleaf veins and petioles) using the method derived from [29] withmodifications described by [30]. Briefly, 1 g of tissue washomogenized at room temperature in disposable plastic bagswith a ball-bearing device in 7 ml of CTAB buffer (3% CTAB, 1 MTris–HCl pH 8, 20 mM EDTA, 1.4 M NaCl) with the extemporaneousaddition of 0.2% 2-Mercaptoethanol; 1 ml was transferred to anEppendorf tube and incubated in a water bath at 65 8C for 20 min.After extraction with 1 ml of Chloroform, nucleic acids wereprecipitated from the aqueous phase with an equal volume ofIsopropanol, collected by centrifugation, washed with 70% Ethanol,dried, dissolved in 150 ml of TE buffer (10 mM Tris, 1 mM EDTA, pH7.6) and stored at �20 8C until use.

DNA amplification of a region of the 16S rRNA phytoplasmagene, was performed in 25 ml total reaction volume in an AppliedBiosystem thermal cycler ‘‘GeneAmp PCR system 2700’’. The firstset of PCR primers was P1 [31] and P7 [32]. P1–P7 amplicons werethen used as target DNA in nested-PCR amplification with theuniversal primer pair for phytoplasmas 16r758f/M23Sr [33,34]. Inorder to identify the phytoplasma group or subgroup, theamplicons were then digested with the restriction enzyme TaqI.The restriction patterns were compared with those of selectedphytoplasma control strains [35].

In all the sampled plants, presence or absence of seven viralinfections were additionally screened by ELISA tests (Agritest andBioerba kits) following manufacturer conditions: Arabis mosaicvirus (ArMV), Grapevine Virus A (GVA), Grapevine Fleck Virus(GFkV), Grapevine Leaf-Roll-Associated Viruses 1 (GLRaV-1), Grape-vine Leaf-Roll-Associated Viruses 2 (GLRaV-2), Grapevine Leaf-Roll-Associated Viruses 3 (GLRaV-3) and Grapevine Fanleaf Virus (GFLV).

2.3. RNA isolation

Total RNA was extracted from 1.5 g of central leaf midribs andpetioles from Bois Noir infected and healthy V. vinifera Chardonnayand Manzoni using a modified Tris–LiCl method [22,36]. Briefly,samples were extracted with homogenization buffer (200 mM Tris–HCl, pH 8.5, 1.5% (w/v) lithium dodecylsulfate, 300 mM LiCl, 10 mMsodium EDTA, 1% (w/v) Sodium deoxycholate, 1% (v/v) Nonidet P-40, 2 mM Aurintricarboxylic acid, 20 mM Dithiothreitol, 10 mMThiourea and 2% (w/v) Polyvinylpolypyrrolidone). After RNAprecipitation with Sodium acetate and Isopropanol, the sampleswere extracted with 25:24:1 Phenol:Chloroform:Isoamyl alcoholand 24:1 Chloroform:Isoamyl alcohol before performing LiClprecipitations. The samples were then treated with DNase (Invitro-gen). RNA was further purified using Qiagen RNeasy columns(Qiagen Inc., Valencia, CA). Disposable RNA chips (Agilent RNA 6000Nano LabChip kit) were used to determine the concentration andpurity/integrity of RNA samples using the Agilent 2100 bioanalyzer.

2.4. Microarray hybridization

RNA samples were processed following the GeneChip1

Expression 30-Amplification Reagents One-cycle cDNA synthesiskit instructions (Affymetrix Inc., Santa Clara, CA, USA). Briefly,

G. Albertazzi et al. / Plant Science 176 (2009) 792–804794

single stranded cDNA was synthesized with 3 mg of total RNA andoligo-dT-T7Promoter Primers using the Superscript II (Invitrogen,USA). After the second strand synthesis, biotin labelled cRNA wasgenerated from the purified cDNA sample (Gene Chip SampleCleanup Module, Affymetrix) by an in vitro transcription reaction,following the manufacturer’s specifications. The labelled cRNA waspurified using the Affymetrix spin columns and the concentrationof biotin-labelled cRNA was determined by Agilent 2100 bioana-lyzer. Ten micrograms of the biotin labelled cRNA were fragmentedand added to the hybridization cocktail containing 4 biotinylatedhybridization controls (BioB, BioC, BioD, and Cre), as recommendedby the Affymetrix protocol (GeneChip1 Expression 30-Amplifica-tion Reagents for IVT Labelling kit instructions). The samples werehybridized in the GeneChip1 V. vinifera (Grape) Genome Array 1.0cartridge (Affymetrix1, Santa Clara, CA) using standard procedures(45 8C for 16 h). The arrays were washed and then stained in aFluidics Station 450. Scanning was carried out with the GeneChip1

scanner 3000 and image analysis was performed using GeneChip1

Operating Software. Three independent biological replicates wereperformed for Chardonnay (infected vs. recovered) and two forManzoni (infected vs. healthy), for a total of ten microarrayhybridizations.

2.5. Microarray data quality control, processing and analysis

GeneChip1 V. vinifera (Grape) Genome Arrays ver. 1.0 (Affyme-trix1 Inc., Santa Clara, CA) were first inspected using a series ofquality control steps. As recommended by the GeneChip1 OperatingSoftware Users Guide, the levels of average background and noise(RawQ) were examined for consistency across all ten arrays.

Arrays data were processed and normalized by RMA (RobustMulti-Array Average) [37] using the R statistical environment(http://www.r-project.org/) and the Bioconductor Affy package(http://www.bioconductor.org). Default options for backgroundadjustment, normalization and summarization were used.

Differentially expressed genes between each infected cultivarand its corresponding control sample were identified as describedby [38]. In details, The R package ‘‘limma’’ was used for fitting alinear model to RMA expression values and to perform ANOVAanalysis (http://www.bioconductor.org/packages/release/bioc/vignettes/limma/inst/doc/limma.pdf).

The following model was adopted for the analysis: yijk = -Ci + Tj + (CT)ij + eijk, where yijk represent the RMA signal measuredfor cultivar i, treatment j, and biological replicate k. The terms Ci

and Tj measure the effect of the cultivar and treatment,respectively, and the interaction term (CT)ij accounts for theinteraction between cultivar and treatment. ANOVA was per-formed using the linear model above and contrasts based ondifferences between cultivars, and each infected cultivar andcontrol state. Moderated statistics and corresponding p-valueswere obtained using the limma functions for empirical bayesstatistics for differential expression. In addition false discovery rate(FDR) [39] was computed on the p-values of the F statistics in orderto control multiple tests error rate. Genes with a false discoveryrate �0.01 were selected for further analysis.

All microarray expression data produced by this work areavailable at Plant Expression Database (http://www.plexdb.org/)under the accession number VV14.

2.6. Gene annotation and functional analysis

The differentially expressed probesets were annotated usingPLEXdb (Plant Expression Database) (http://www.plexdb.org/), aunified public resource for gene expression for plants and plantpathogens. The gene annotation was updated by a nucleotidesequence query of the probe consensus sequence against the

UniProt/TrEMBL and TAIR protein databases (The Arabidopsis

Information Resource) using BLASTX, and against TIGR (The Institutefor Genomic Research) and DFCI (Dana Faber Cancer Institute, GrapeGene Index Database) using BLASTN. The homology between thequery and database sequences was only considered significant if theE value was below 1E�10.

The Gene Ontology (GO) categories for biological processes,molecular function, and cellular component were assignedmanually, picking preferentially those probesets that matchedUniProt/TrEMBL and TAIR protein databases. When homologieswith protein data were not available, Tentative Consensus (TCs)from TIGR Grape Gene Index matching probesets of Arabidopsis

genome were used. Functional classification analysis was per-formed on selected lists of differentially expressed genes, using ahypergeometric test to evaluate GO categories overrepresentation.

2.7. Array validation

To validate expression profiles obtained using the AffymetrixGeneChip1 V. vinifera genome array, quantitative real-time RT-PCRwas performed. In order to assess the performance of the array in abiological context, we examined the transcript abundance of sixcandidate genes exhibiting changing expression patterns. Amplifiedgrapevine genes were zeaxanthin epoxidase (1618171_s_at,TC55939, exclusively repressed in Chardonnay); sucrose synthase(1609402_at, TC62599, exclusively induced in Chardonnay); auxin-regulated protein (1613468_at, TC59343, exclusively repressed inManzoni); flavonol synthase (1608791_at, TC59043, exclusivelyinduced in Manzoni); transketolase, chloroplast precursor(1620654_at, TC52548, repressed in both Chardonnay and Manzoni)and WRKY-type DNA binding protein 1 (1622778_at, TC54050,induced in both cultivars). Relative expression levels of each samplewere normalized to the expression level of actin (AF369524), whichwas expressed at a constant level in the present experimentalconditions. For each of the samples used for microarray experiments,cDNA was synthesized using the SuperScriptTM II Reverse Tran-scriptase kit (Invitrogen) according to the manufacturer instruc-tions. Then, the three cDNAs referring to each experimentalcondition were pooled in a balanced way, and real-time PCRs wereperformed starting from 1 ng/ml of each cDNA sample. Primers forgenes assayed by quantitative real-time reverse transcription wereselected using Primer3 software (http://primer3.sourceforge.net/)[40]. Quantitative real-time PCR reactions were prepared using SYBRGreen PCR Master Mix with ROX (Applied Biosystems) andperformed using the ABI PRISM1 7300 Sequence Detection System(Applied Biosystems, Foster City, CA, USA) under conditions of 95 8Cfor 10 min, then 40 cycles of 95 8C for 15 s and 60 8C for 1 min. PCRdata were analyzed with the ABI Prism 7300 SDS software. Meltingcurve analysis was performed on all samples to ensure amplificationof a single product with the expected melting temperature and theabsence of primer–dimers. The Ct value of each gene was normalizedwith actin to obtain the value of DCt. Values of DDCt relative tocontrol was used to calculate relative gene expression following theprocedure described in the Relative Quantification Using theComparative Ct Method in the User Bulletin #2, ABI Prism 7700Sequence Detection System (Applied Biosystems). The fold changeswere estimated in natural log values of the Relative Quantification(RQ) between Bois Noir infected and control leaf sample, which wascalculated using the equation: 2 � (DDCT).

3. Results

3.1. Phytoplasma detection and identification

Grapevine yellows incidence in the inspected vineyards wascalculated as percentage of positive symptomatic plants in 100

http://www.r-project.org/

http://www.bioconductor.org/

http://www.bioconductor.org/packages/release/bioc/vignettes/limma/inst/doc/limma.pdf

http://www.bioconductor.org/packages/release/bioc/vignettes/limma/inst/doc/limma.pdf

http://www.plexdb.org/

http://www.plexdb.org/

http://primer3.sourceforge.net/

Fig. 1. Heat map of the most significantly differentially expressed genes (false

discovery rate � 0.01) from Bois Noir infected versus healthy leaf samples. Genes

with similar expression patterns are grouped together. The colour scale ranges from

saturated green for log ratios �2.0 and below (repression), to saturated red for log

ratios 2.0 and above (induction), whereas black means no change in transcript level.

Each gene is presented by a single row of coloured boxes and each experiment is

represented by a single column. (a) Heat map visualization of the 579 most

significantly differentially expressed genes in Bois Noir infected Chardonnay plants

compared to recovered samples. (b) Heat map visualization of the 343 most

significantly differentially expressed genes in Bois Noir infected Manzoni Bianco

plants compared to healthy samples. (For interpretation of the references to colour

in this figure legend, the reader is referred to the web version of the article.)


plants of each variety. The results showed that the incidence ofphytoplasma infection largely varied between the two cultivars,indicating as tolerant the Manzoni. The yellow disease incidencewas of 73.6% and 62.7% for Chardonnay, and of 3.3% and 3.7% forManzoni, in the two experimental fields of Coviolo (Reggio Emilia,Italy) and Brescia (Italy), respectively (Elisa Angelini and MicheleBorgo, personal communication).

For microarray analysis, phytoplasma-free and naturally BoisNoir infected grapevine plants were selected on the basis of visiblesymptoms observed on field samples of Coviolo, Reggio Emilia.Bois Noir infection was later confirmed by nested-PCR, whichshowed also the absence of Flavescence Doree phytoplasma, andthe final samples for expression analyses were selected accord-ingly. Phytoplasma-free plant material was negative for both BoisNoir and Flavescence Doree phytoplasmas.

The absence of all tested virus (Fleck, ArMV, GFLV, GLRaV-1,GLRaV-2, GLRaV-3, GVA) was confirmed by ELISA test in all theManzoni Bianco plants, while all the Chardonnay material waspositive only for grapevine Fleck virus (Elisa Angelini and MicheleBorgo, personal communication).

3.2. Bois Noir affects grapevine transcriptome

The GeneChip1 V. vinifera genome array (Affymetrix) repre-sents a large and significant part of the 30,344 genes predicted in V.

vinifera [41]. It consists of 16,436 probesets: 14,496 derived from V.

vinifera transcripts and 1940 derived from other Vitis speciesor hybrids transcripts. Sequences used in the design of theVitis GeneChip1 were selected from GenBank, dbEST, and NCBIReference Sequences (RefSeq).

The sequence clusters were created from the UniGene database(Build 7, October 2003). V. vinifera sequences represented on thechip correspond to 10,042 TIGR Tentative Consensus and 1940Singletons (Release 4, September 2004), while 102 GenBankaccessions are not present in the TIGR database. Overall chipredundancy is estimated to be 16.6%.

Changes in gene expression in phytoplasma-infected grapevineplants were found in both cultivars.

Across the ten arrays, the present transcripts call rates rangedfrom 69% to 74%, indicating that about 11,300 genes were expressedat detectable levels in the leaf tissues. ANOVA with Benjamini–Hochberg (BH) adjustment of the false discovery rate (FDR)identified 765 probesets with FDR � 0.01 based on differencesbetween Bois Noir infected and healthy plants. As illustrated inFig. 1, of this 765 probeset, in Chardonnay 168 genes were up-regulated and 254 were down-regulated in phytoplasma infectedleaves (Fig. 1a), while in Manzoni 77 genes were induced and 109repressed in response to Bois Noir infection (Fig. 1b). Additionally, inresponse to phytoplasma infection, the two cultivars shared 54 up-regulated and 103 down-regulated probesets.

3.3. Functional classification of differentially expressed genes

The TAIR database was used to retrieve Gene Ontology (GO)annotations of Arabidopsis genes. Differentially expressed genelists were examined to identify significant enrichment of specificGene Ontology categories, using a hypergeometric test. Theresults of GO classes enrichment analysis for biological process,molecular function and cellular component are reported inSupplementary Tables 4 and 5. Statistically significant differ-ences in the distribution of genes within biological processcategories were observed. Since many over-represented cate-gories from Supplementary Tables 4 and 5 can actually beassociated to related biological processes, a custom GeneOntology slim (http://www.geneontology.org/GO.slims.shtml)was applied to group similar over-represented classes into

wider groups. Then the main groups of over-represented GO slimclasses were represented as pie charts in Fig. 2. As highlighted byFig. 2, some Gene Ontology slim categories were only repre-sented by repressed genes. In particular, only in infectedChardonnay and not in Manzoni, GO:0015979 (photosynthesis)and GO:0006629 (lipid metabolic processes) were down-regulated. In infected Chardonnay and Manzoni plants, genesof categories involved in cellular component organization andbiogenesis, protein metabolic processes and nucleic acid meta-bolic processes were inhibited. Carbohydrate metabolic pro-cesses were strongly affected, with about 30% and 33% of GO slimclasses associated to repressed genes in Chardonnay andManzoni, respectively (Fig. 2c and d). In Manzoni about 64% ofthe GO slim classes associated to induced transcripts wererelated to amino acid and derivate metabolic processes.

To gain insight into the cellular processes that are affected byBois Noir infection as manifested by the relative difference at atranscript level, the 765 selected genes were classified according totheir annotated putative function (Supplementary Tables 1, 2 and3). Representative list of the 168 and 77 specifically induced genesof phytoplasma-infected Chardonnay and Manzoni are shown inTables 1 and 2, respectively. The remaining common genes thatwere differentially induced during the infection in both cultivarswere functionally classified as well (Table 3).

Affymetrix probesets were annotated using the PLEXdb (PlantExpression Database) and the TAIR database. In particular, inorder to retrieve additional functional annotations, the probesetswere associated with V. vinifera genes and with theirhomologous Arabidopsis thaliana genes (see SupplementaryTable 1, 2 and 3).

3.3.1. Transcription factors, signalling pathways and regulatory genes

Different transcription regulators were affected by phyto-plasma infection (Tables 1–3; Supplementary Tables 1, 2 and 3).Interestingly, a Myb transcription factor was induced in Manzoni(Table 2) and repressed in Chardonnay (Supplementary Table 1).

http://www.geneontology.org/GO.slims.shtml

Fig. 2. Main groups of over-represented GO slim classes concerning the biological

processes affected in grapevine response to Bois Noir. The figure shows the

percentages of over-represented (p-value � 0.05) Gene Ontology biological process

terms of genes induced by phytoplasma infection for (a) Chardonnay (28 genes) and

(b) Manzoni Bianco (17 genes), and repressed for (c) Chardonnay (77 genes) and (d)

Manzoni Bianco (21 genes).


3.3.2. Carbohydrate metabolism and glycolysis

As shown in Tables 1–3 and Supplementary Tables 1, 2 and 3,carbohydrate metabolic processes are strongly affected, mainlyinhibited, by Bois Noir in both cultivars. Two genes implicated inthe glycolysis were repressed in infected Chardonnay plants: non-phosphorylating glyceraldehyde-3-phosphate dehydrogenase andfructose-bisphosphate aldolase (Supplementary Table 1). InManzoni, phosphomannomutase, UDP-glucose glucosyltransfer-ase and UDP-glucose dehydrogenase were repressed (Supplemen-tary Table 2). Chloroplast transketolase precursor and trehalose-phosphate phosphatase were repressed in infected Chardonnayand Manzoni plants (Supplementary Table 3).

3.3.3. Photosynthesis and carbon assimilation

In Chardonnay, transcripts encoding proteins with photosynth-esis-related functions were strongly repressed after Bois Noirinfection (Fig. 2c and d, Supplementary Table 1), while in Manzonionly one gene was down-regulated (Supplementary Table 2). In theChardonnay infected plants, the Calvin cycle was also stronglyinhibited. The genes that were suppressed in this pathwayincluded ribulose-5-phosphate-3-epimerase, a cytosolic fruc-tose-1,6-bisphosphatase, a chloroplast transketolase precursor

and NADP-dependent glyceraldehyde-3-phosphate dehydrogen-ase (Supplementary Table 1). In both cultivars, carbonic anhydrasewas repressed (Supplementary Table 3). In plants, carbonicanhydrase helps raise the concentration of CO2 within thechloroplast in order to increase the carboxylation rate of theenzyme RuBisCO; this reaction integrates CO2 into organic carbonsugars during photosynthesis.

3.3.4. Pathogen and disease resistance related proteins

As shown in Fig. 2, the GO slim class ‘‘response to stress’’ isrepresented in both the induced and repressed Chardonnay andManzoni pie charts. PR1 precursors, the marker for the salicylicacid (SA) biosynthesis, were strongly induced in Chardonnay inresponse to Bois Noir (Table 1), while the expression of PR5 wasinduced in both cultivars (Table 3).

3.3.5. Cell wall metabolism

Plant cell wall contains glycan (cellulose and callose), hemi-cellulose, pectin, mannan and various glycoproteins. In infectedChardonnay and Manzoni plants, genes responsible for cell walldegradation were repressed (Supplementary Table 1, 2 and 3), whilegenes involved in cell wall reinforcement were induced (Tables 1–3).

In Chardonnay, the genes that were induced in response to BoisNoir (Table 1) included those encoding a hydroxyproline-richglycoprotein, a proline-rich protein 1 and a proline-rich protein 2.All three proteins are known to be one of the structuralcomponents of cell walls and are involved in cell wall reinforce-ment [42]. A putative uncharacterized protein WAKL1 (WAK-likekinase), functioning as protein kinase that physically links theextracellular matrix and the cytoplasm, involved in the response topathogens [43], was also included in this category. Several genesassociated with wall polysaccharide hydrolysis were repressed ininfected Chardonnay and Manzoni plants. These genes include b-galactosidase (whose products are involved in the hydrolysis ofcomplex cell wall sugars), xyloglucan endotransglycosylase (whichhydrolyzes and transglycosylates xiloglucans), expansins, pectatelyase, a polygalacturonase inhibitor-like protein and an endoglu-canase 2 precursor (Supplementary Table 3).

3.3.6. Protein, amino acid and nitrogen metabolism

Fig. 2c and d highlights that GO slim class concerning proteinmetabolic processes are repressed in infected Chardonnay andManzoni plants. The majority of suppressed genes in Chardonnaywere involved in chloroplast protein synthesis and maturation viaproteolysis and protein folding. These genes included chloroplasticribosomal protein 30S and 50S, alanyl-tRNA synthetase, isoleucyl-tRNA synthetase, elongation factor G and peptidyl-prolyl cis–transisomerase (Supplementary Table 1). In Manzoni the genes involvedin protein ubiquitination were repressed (Supplementary Table 2).Several different proteases were down-regulated by Bois Noir ininfected Chardonnay and Manzoni plants, including a putativeserine protease, a cysteine protease, a papain-like cysteineproteinase and a serine carboxypeptidase (Supplementary Table 3).

Amino acids and derivate metabolic processes were mostlyinduced in infected Chardonnay and Manzoni plants (Tables 1–3).Few genes were down-regulated in Chardonnay, and among themthere were those encoding a serine hydroxymethyltransferase(involved in glycine biosynthesis), a glutamine synthetaseprecursor, a glutamate synthase (an important enzyme involvedin ammonia assimilation) and a NADH glutamate dehydrogenase(Supplementary Table 1).

3.3.7. Secondary metabolism

As shown in Tables 1–3 several genes were induced in infectedChardonnay and Manzoni plants. Among the genes that weresuppressed only in infected Chardonnay plants, there was a

Table 1Induced genes in Chardonnay plants during Bois Noir infection. The list includes genes that are induced more than twofold in response to Bois Noir

infection in Chardonnay plants (FDR � 0.01). Functional categories were assigned considering sequences of genes represented on the array and their

annotation.

Probeset ID Putative function Fold change

Transcription factors, signalling pathway and regulatory genes

1617012_at Ethylene-responsive transcription factor ERF003

(Arabidopsis thaliana)

9.4

1621255_at Nam-like protein 11 (Petunia hybrida) 6.6

1611361_at Putative MADS-box family transcription factor (Pinus radiata) 6.3

1622333_at WRKY transcription factor (Populus tremula � Populus alba) 4.8

1613776_at Hypothetical transcription factor (Prunus persica) 4.1

1620324_at Leucine-rich repeat receptor-like protein kinase 1 (Populus nigra) 4.1

1610775_s_at WRKY transcription factor-b (Capsicum annuum) 3.6

1612448_at No apical meristem protein, expressed (Oryza sativa) 3.3

1620797_at Putative CLAVATA1 receptor kinase (Oryza sativa) 3.3

1612310_at Heat shock factor protein HSF24 (Solanum peruvianum) 3.1

1618638_at Putative serine/threonine protein kinase (Nicotiana tabacum) 3.1

1615113_at Calmodulin-like protein (Arabidopsis thaliana) 2.4

1621641_at Receptor protein kinase-like protein (Capsicum annuum) 2.3

1619830_at Squamosa promoter binding protein 3 (Physcomitrella patens) 2.3

Hormones metabolism and signalling

1612224_s_at Phytosulfokine peptide precursor (Zea mays) 4.6

1609893_at GASA1 (GAST1 PROTEIN HOMOLOG 1) (Arabidopsis thaliana) 3.9

1620071_at Gibberellic acid receptor (Gossypium hirsutum) 2.6

1618213_at Ethylene-responsive protein, putative (Arabidopsis thaliana) 2.4

Carbohydrate metabolism and glycolysis

1620283_s_at Alpha amylase, catalytic region (Medicago truncatula) 4.8

1620865_at Enolase (Oryza sativa) 4.0

1609402_at Sucrose synthase (Citrus unshiu) 2.8

1612836_at Vacuolar invertase 2, GIN2 (Vitis vinifera) 2.4

1612546_at Malate dehydrogenase precursor (Medicago sativa) 2.3

Pathogen and disease resistance related proteins

1608692_s_at Putative pathogenesis related protein 1 precursor (Vitis vinifera) 23.2

1613816_x_at Putative pathogenesis related protein 1 precursor (Vitis vinifera) 11.2

1611490_at Putative pathogenesis related protein 1 precursor (Vitis vinifera) 10.9

1620390_s_at Thaumatin-like protein (Vitis vinifera) 8.6

1606794_at Thaumatin-like protein (Vitis vinifera) 7.4

1615595_at Beta-1,3-glucanase (Vitis vinifera) 6.9

1613180_at Thaumatin (Vitis riparia) 6.2

1606513_s_at Stress-related protein (Vitis riparia) 4.8

1616127_at Hypoxia-responsive family protein (Arabidopsis thaliana) 4.6

1611876_s_at Acidic endochitinase precursor (Nicotiana tabacum) 4.6

1621066_s_at glycosyl hydrolase family 1 protein (Arabidopsis thaliana) 4.3

1612108_at Avr9/Cf-9 rapidly elicited protein 146 (Nicotiana tabacum) 4.0

1618533_at Pathogenesis-related protein PR-1 precursor (Medicago truncatula) 4.0

1620505_at Basic endochitinase precursor (Vitis vinifera) 3.8

1617430_s_at Basic endochitinase precursor (Vitis vinifera) 3.5

1617527_at Glycosyl hydrolase family 1 protein (Arabidopsis thaliana) 3.4

1614514_at Beta-1,3 glucanase precursor (Pisum sativum) 3.1

1608864_s_at Acidic endochitinase precursor (Vitis vinifera) 3.1

1621414_at Phytochelatin synthetase-like protein (Phaseolus vulgaris) 2.3

1610722_at Beta-1,3-glucanase (Prunus persica) 2.1

1614975_at Wound induced protein [fragment] (Solanum lycopersicum) 2.0

1613464_at Resistance protein candidate [fragment] (Vitis amurensis) 2.0

Cell wall metabolism

1615495_at Putative uncharacterized protein WAKL1 (Nicotiana tabacum) 4.4

1607449_s_at Proline-rich protein 1 (Vitis vinifera) 4.2

1616822_at Proline-rich protein 1 (Vitis vinifera) 4.0

1614803_at Proline-rich protein 1 (Vitis vinifera) 3.9

1621384_at Proline rich protein 2 (Vitis vinifera) 3.8

1615469_at Hydroxyproline-rich glycoprotein precursor (Phaseolus vulgaris) 3.1

1612253_at Expansin (Sambucus nigra) 3.1

1608624_at Putative cellulose synthase-like protein OsCslE1 (Oryza sativa) 3.0

1617740_at Putative ripening-related P-450 enzyme (Vitis vinifera) 2.9

Protein metabolism

1611820_at Protein disulfide isomerase (Quercus suber) 3.8

1607709_at MATH domain-containing protein (Arabidopsis thaliana) 2.6

1615360_at Cystatin-like protein (Citrus paradisi) 2.5

1609252_at Ubiquitin conjugating enzyme (Arabidopsis thaliana) 2.5

1608422_at Ankyrin repeat family protein (Arabidopsis thaliana) 2.3

1618208_s_at Leucine-rich repeat protein (Triticum aestivum) 2.2

1608948_at Putative RNA-binding protein RBP37 (Oryza sativa) 2.1

1613079_at Probable tyrosine-protein phosphatase At1g05000


2.1

1616579_at Ankyrin repeat family protein (Arabidopsis thaliana) 2.0


Table 1 (Continued )


Amino acid and nitrogen metabolism

1614574_at Lysine decarboxylase-like (Oryza sativa) 4.4

1622358_s_at Lysine decarboxylase-like (Oryza sativa) 4.0

Secondary metabolism

1608761_at Flavanone 3-hydroxylase-like protein (Arabidopsis thaliana) 14.4


1612552_at S-adenosyl-L-methionine:carboxyl methyltransferase family

protein (Arabidopsis thaliana)

4.3

1622331_at Berberine and berberine like, putative (Medicago truncatula) 3.0


1608603_at Phloroglucinol O-methyltransferase (Rosa chinensis var. spontanea) 2.4

1613121_at Anthocyanidin 5,3-O-glucosyltransferase (Rosa hybrid cultivar) 2.2

1613159_at Short-chain dehydrogenase/reductase (SDR) family protein


2.0

Fatty acid, lipid and conjugates

1614923_at 3-hydroxy-3-methylglutaryl-coenzyme A reductase 3

(Solanum tuberosum)

2.2

Transport facilitation

1611350_at Putative lipid transfer protein (Solanum tuberosum) 6.6

1611326_at Putative sugar transporter (Solanum lycopersicum) 4.7

1610422_at Patellin-6 (Arabidopsis thaliana) 4.6

1610800_at Amino acid/polyamine transporter II (Medicago truncatula) 3.8

1620384_s_at Heavy-metal-associated domain-containing protein


3.4

1616115_at Metal ion binding (Arabidopsis thaliana) 3.3

1620972_at Putative permease (Gossypium hirsutum) 3.3

1621683_x_at Metal ion binding (Arabidopsis thaliana) 3.1

1622455_at Peptide transporter PTR3-A (Arabidopsis thaliana) 2.6

ROS metabolism

1615967_at Peroxidase 73 precursor (Arabidopsis thaliana) 2.7

1608802_s_at Glutathione S-transferase (Medicago truncatula) 2.3

1617100_at Putative glutathione S-transferase T3 (Solanum lycopersicum) 2.0


zeaxanthin epoxidase (encoding a carotenoid biosynthesisenzyme) and the phenylpropanoid synthesis pathway genesphenylalanine ammonia lyase, cinnamate-4-hydroxylase and 4-coumarate:CoA ligase (Supplementary Table 1).

GDP-mannose 3,5-epimerase 1 (a key enzyme in the biosynth-esis of vitamin C), was repressed in Chardonnay and Manzoniplants in response to Bois Noir (Supplementary Table 3).

3.3.8. Fatty acids, lipids and conjugates

Transcripts annotation by Gene Ontology vocabulary revealedthat genes involved in lipid metabolic processes were repressed inresponse to Bois Noir mainly in Chardonnay (Fig. 2). In infectedChardonnay plants the expression of the following gene productswas repressed: squalene monooxygenase (which is implicated incholesterol biosynthesis), b-ketoacyl-ACP synthase precursor(involved in fatty acid elongation), a putative phosphoethanola-mine N-methyltransferase 3 (involved in choline biosynthesis), aputative lipase (involved in triacylglycerol degradation) andsoluble diacylglycerol acyltransferase (involved in triacylglycerolbiosynthesis) (Supplementary Table 1). In Manzoni, a cholinepho-sphate cytidylyltransferase (which is involved in phospholipidbiosynthesis and choline biosynthesis) was repressed, togetherwith a lipoxygenase (which is important in the biosynthesis of theplant hormone jasmonic acid) and a diacylglycerol kinase 1 (whichphosphorylates diacylglycerol to produce phosphatidic acid)(Supplementary Table 2). In both cultivars, the glycerol-3-phosphate acyltransferase 4 (involved in triacylglycerol andphospholipid biosynthesis) and an omega-6 fatty acid desaturasewere repressed (Supplementary Table 3).

3.3.9. ROS metabolism

In infected Chardonnay plants, six genes encoding a putativeL-ascorbate proxidase, thioredoxins and an oxidoreductase were

repressed (Supplementary Table 1). In infected Manzoni plantsthe genes that were down-regulated included two genesencoding a peroxidase precursor, a gene for class III peroxidase(GvPx2b) and three genes encoding for a glutathione S-transferase (Supplementary Table 2), while no induced genesinvolved in ROS metabolism were induced exclusively in thiscultivar (Tables 1–3).

3.4. Array validation

Quantitative real-time PCR (qRT-PCR) assays was performed ona subset of six genes to verify differential expression measured inthe microarray analysis. The quantitative data were normalized asa ratio to actin expression and then calculated as a ratio ofexpression from Bois Noir infected to recovered Chardonnayplants, and from Bois Noir infected to healthy Manzoni Biancoplants. The data were correlated to expression ratios from theGeneChip1 microarray. Linear regression analysis indicated thatthere was a good correlation between qRT-PCR and the microarraydata with a goodness of fit (r2) of 0.93 for Chardonnay (Fig. 3a) and0.95 for Manzoni Bianco (Fig. 3b).

4. Discussion

Grapevine phytoplasmas are restricted to the phloem sievetubes in which the photosynthetically enriched sap circulates, thuscausing several symptoms such as stunting, yellowing andphyllody and stress responses in plants [44]. Despite theireconomic importance and unique biological features, phytoplas-mas remain the most poorly characterized plant pathogens, novirulence genes have been found in their genome [15], and onlyrecently few studies on single mechanisms involved in theirpathogenicity have been reported [4,5,6,10]. Therefore the present

Table 2Induced genes in Manzoni Bianco plants during Bois Noir infection. The list includes genes that are induced more than twofold in response to Bois Noir

infection in Manzoni Bianco plants (FDR � 0.01). Functional categories were assigned considering sequences of genes represented on the array and their

annotation.

Probeset ID Putative function (organism) Fold change

Transcription factors, signalling pathway and regulatory genes

1612620_at Protein binding (Arabidopsis thaliana) 9.5

1621723_at NAM-like protein (Prunus persica) 9.4

1607082_at Protein binding (Arabidopsis thaliana) 7.9

1621854_at Receptor-like protein kinase (Arabidopsis thaliana) 5.9

1607620_at NAM-like protein (Arabidopsis thaliana) 3.9

1619410_at HDZip I protein (Glycine max) 3.7

1611033_at DNA-directed RNA polymerase subunit beta (Vitis vinifera) 3.2

1613366_at Mutant cincinnata (Antirrhinum majus) 2.4

1607523_at TCP1 protein (Lupinus albus) 2.4

1616184_at Myb (Nicotiana tabacum) 2.3


1610880_s_at Indole-3-acetic acid-amido synthetase GH3.3 (Arabidopsis thaliana) 7.6

1612001_s_at Auxin and ethylene responsive GH3-like protein (Capsicum chinense) 4.6


1608393_at Glucose-1-phosphate adenylyltransferase (Citrullus lanatus) 15.5

1616116_at Seed imbitition protein-like (Arabidopsis thaliana) 9.6

1611723_at Seed imbitition protein-like (Arabidopsis thaliana) 6.9

1613795_at UDP-sugar pyrophospharylase (Pisum sativum) 2.4

1615374_at Sucrose-phosphate synthase 1 (Citrus unshiu) 2.1


1611666_s_at Protease inhibitor (PR6) (Vitis vinifera) 25.5

1616317_at Putative trypsin inhibitor (Arabidopsis thaliana) 9.2

1611611_at Germin-like protein precursor (Pisum sativum) 7.0

1608026_at Germin-like protein subfamily 1 member 7 precursor (Arabidopsis thaliana) 6.1

1622203_at Tumor-related protein (Nicotiana tabacum) 5.7

1616413_at Thaumatin-like protein VVTL1 (Vitis vinifera) 3.2

1607247_at Universal stress protein (USP) family protein (Arabidopsis thaliana) 2.2


1614008_at Polygalacturonase PG1 (Vitis vinifera) 5.2

1607069_at Putative cellulose synthase-like protein OsCslE1 (Oryza sativa) 2.7

Protein metabolism

1606575_at RING zinc finger protein-like (Arabidopsis thaliana) 2.9

1621142_s_at F-box/LRR-repeat protein 5 (Arabidopsis thaliana) 2.8

1613168_at Glycyl-tRNA synthetase 1, mitochondrial precursor (Arabidopsis thaliana) 2.7

1609170_at 40S ribosomal S4 protein (Medicago truncatula) 2.4

1613996_at RNA binding (Arabidopsis thaliana) 2.1


1614207_at Asparagine synthetase (Elaeagnus umbellata) 4.2

1611496_at Glycine cleavage T protein-like (Oryza sativa) 2.3


1608791_at Flavonol synthase (Vitis vinifera) 9.7

1619371_at Beta-carotene hydroxylase (Vitis vinifera) 6.9

1610410_at Putative glucosyltransferase (Arabidopsis thaliana) 6.3

1607029_at Carotenoid cleavage dioxygenase 1 (Lactuca sativa) 4.5

1618389_at Salutaridinol 7-O-acetyltransferase (Papaver somniferum) 2.6

Fatty acid, lipid and conjugates

1616273_at Phospholipid/glycerol acyltransferase (Medicago truncatula) 2.4


1618695_at Plasma membrane H+-ATPase (Solanum lycopersicum) 6.9

1612886_at Plasma membrane H+-ATPase (Sesbania rostrata) 4.8


work is the firstly reported transcriptional profiling study of suchinteractions.

Phytoplasmas lack a cell wall, therefore, their membraneproteins are in direct contact with the cytoplasm of plant or insecthost cells, and might have important roles in host–pathogeninteractions. Their genome lacks homologues of the type IIIsecretion system encoded in the hrp gene cluster, which is essentialfor the virulence of many Gram-negative phytopathogenic bacteria[45,46]. However, phytoplasmas may secrete virulence-relatedproteins via the Sec-dependent pathway [16]. Because phytoplas-mas lack most of the common metabolic pathways, it has beenspeculated that they must assimilate a wide range of materials

from the host cells, probably with detrimental effects on the hosts.Sugars seemed to play an important role in the pathogenicity ofcertain mollicutes [47]. It has been reported that the spiroplasmagenome includes a fructose metabolism operon [48] that isresponsible for their pathogenicity [49]. However, no genesinvolved in fructose metabolism were identified in the phyto-plasma genome. Therefore, although phytoplasmas are closelyrelated to spiroplasmas, they are metabolically different and mayuse different sugar molecules [50].

Differences between Chardonnay and Manzoni in symptomexpression in response to Bois Noir infections were observedduring field surveys. The Chardonnay showed high susceptibility,

Table 3Overlapping responses of grapevine Chardonnay and Manzoni Bianco. The table shows genes that exhibited a similar behaviour in both cultivars during infection

with Bois Noir. Genes with an expression level over twofold in response to phytoplasma infection were considered for this analysis (FDR � 0.01).


Chardonnay

Fold change

Manzoni

Transcription factors, signaling pathway and regulatory genes

1621876_at NAC1 protein (Glycine max) 8.9 13.6

1621448_at NAC domain protein NAC5 (Glycine max) 2.9 7.1

1609117_at NAM (No apical meristem)-like protein (Arabidopsis thaliana) 3.2 5.1

1622778_at WRKY-type DNA binding protein 1 (Vitis vinifera) 8.2 4.5


1609559_at Ethylene response factor 1 (Solanum lycopersicum) 13.4 5.9

1615079_x_at Auxin-induced protein PCNT115, putative, expressed (Oryza sativa) 2.6 2.2


1615552_at Galactinol synthase (Capsicum annuum) 9.9 20.8

1610427_at Inositol oxygenase 1 (Arabidopsis thaliana) 3.0 5.2

1615814_at Glyceraldehyde-3-phosphate dehydrogenase, cytosolic (Ranunculus acris) 3.4 3.9


1616695_s_at Thaumatin (Vitis riparia) 24.3 12.4

1608262_at Chitinase precursor (Vitis vinifera) 23.9 8.9

1622104_at Disease resistance-responsive family protein (Arachis hypogaea) 2.1 2.6

1622723_at Hairpin-induced family protein (Ipomoea nil) 2.2 2.4


1619147_at Merlot proline-rich protein 2 (Vitis vinifera) 43.5 44.8

1620342_at Caffeic acid O-methyltransferase (Acacia mangium � Acacia auriculiformis) 3.3 7.1

1607475_s_at Caffeic acid O-methyltransferase (Acacia mangium � Acacia auriculiformis) 2.5 3.1

1608799_at Pectin methylesterase 2 (Pyrus communis) 2.3 2.2

Protein metabolism

1608656_at Putative C3HC4-type RING zinc finger protein (Hevea brasiliensis) 2.2 5.5


1614680_at Putative 3-isopropylmalate dehydratase large subunit (Oryza sativa) 8.7 10.4

1618916_at Cysteine synthase (Vitis vinifera) 2.5 9.4

1614985_at Aspartate aminotransferase, cytoplasmic (Daucus carota) 2.1 2.4


1614862_at Short chain alcohol dehydrogenase-like (Arabidopsis thaliana) 12.1 13.5


1612795_at Heavy-metal-associated domain-containing protein/copper

chaperone (CCH)-related (Arabidopsis thaliana)

7.0 5.9

1607996_at Glucose-6-phosphate/phosphate translocator (Oryza sativa) 3.4 5.2

1614764_at Hexose transporter HT2 (Vitis vinifera) 2.3 2.1

ROS metabolism

1615827_at Alternative oxidase (Nicotiana attenuata) 3.5 3.8

1610989_at Glutathione S-transferase GST 13 (Glycine max) 6.6 3.6


1606607_at Glutathione S-transferase (Vitis vinifera) 2.6 3.1


1609781_s_at Lactoylglutathione lyase-like protein (Arabidopsis thaliana) 2.1 2.5

1611268_x_at Glutathione S-transferase GST 24 (Glycine max) 4.5 2.4


as already reported by other authors [51]. Notably, despiteManzoni is considered a medium-susceptible cultivar, our cloneshowed tolerance to phytoplasma infection in both the experi-mental fields, with an average of disease incidence of 3.5% only.Symptoms varied according to the cultivar; Chardonnay exhibitedtypical symptoms on the leaves, which were discoloured andwhose edges rolled downward, giving them an angular shape. InManzoni symptoms on the leaves were delayed, and yellowing wasonly found in the proximity of the veins. In both cultivars, leafblade became brittle, the flowers dried out, the bunches werebrown and shrivelled and the shoots were irregularly lignificated.

In Manzoni, all the viruses tested with ELISA were absent fromboth the Bois Noir infected and healthy plants. In Chardonnay, inboth the Bois Noir infected and recovered plant material, wedetected the presence of the Grapevine fleck virus (GFkV). In V.

vinifera this virus is latent and asymptomatic [52] and changes ingene expression in response to GFkV have not been reported inliterature. Thus, GFkV infection should not influence the detectionof differentially expressed genes addressed in this study. More-

over, since both infected and recovered Chardonnay materialswere positive for GFkV and the selection of differentially expressedgenes was done in the comparison between infected vs. recovered,no probesets induced by the virus could have been selected.

As illustrated in Fig. 4, the dendrogram of unsupervisedhierarchical clustering analysis for infected Chardonnay andManzoni plants, and healthy Chardonnay and Manzoni plantsamples data obtained from the array, indicated high similaritiesamong sample replications. The dendrogram reflects the relation-ships among replicates by separating four groups of samples,corresponding to biological replicates clustered together thatsegregate into discrete groups, according to the manifestation ofsymptoms in the field.

To explore host cellular processes affected by phytoplasmainfection, differentially expressed genes were classified into twelveclasses based on analysis of the GO slim (Supplementary Tables 4and 5).

As expected, microarray analysis revealed that Bois Noirinfection heavily affected several metabolic pathways remodelling

Fig. 3. Comparison between the gene expression ratios reported by the Affymetrix

GeneChip1 genome array and by real-time RT-PCR. The microarray log2

(expression ratio) values (x-axis) are plotted against the log2 (expression ratio)

obtained by quantitative real-time RT-PCR (y-axis). Six plots mean the results of six

genes. (a) Comparisons between Bois Noir infected and recovered Chardonnay. (b)

Comparisons between Bois Noir infected and healthy Manzoni Bianco.

Fig. 4. Unsupervised hierarchical clustering performed on the whole dataset using

Pearson correlation as similarity measure and average linkage. The dendrogram

shows that the different classes of samples are well separated by clustering

analysis.


the transcriptome of both Chardonnay and Manzoni. An overalldecrease in transcript abundance was found in response to BoisNoir infection in grapevine and a greater number of repressedgenes were obtained in comparison with the up-regulated group(Supplementary Table 1, 2 and 3). A shift from housekeeping todefense metabolism, which is associated with down-regulation ofgenes related to several cellular processes is a typical response tovarious pathogens, which is qualitatively similar in compatible andincompatible interactions [53,54].

A general decay of the main plant metabolisms, i.e. of proteins,carbohydrates and lipid metabolisms, was observed in bothcultivars, although stronger in Chardonnay (Supplementary Tables1, 2 and 3). A large quantity of genes repressed in Chardonnay inresponse to phytoplasma infection code for proteins associatedwith the chloroplast. In this cultivar there was a deal evidence indecreasing of transcripts for photosystem I reaction center subunitII, III, IV, N, psaK and O (Supplementary Table 1). This leads to aserious inhibition of the whole photosynthetic chain and ofphotosystem I activity, causing rapid leaf senescence, as observedin the field.

Biochemical analyses on phytoplasma-infected tobacco plantsindicated that soluble carbohydrates and starch were accumulatedin source leaves, while sink organs showed a marked decrease insugar levels [5,6]. In our study, infected Chardonnay showed adecrease in RNA level of ribulose-1,5-bisphosphate carboxylase/oxygenase small subunit N-methyltransferase I, ribulose-5-phos-phate-3-epimerase, fructose-1,6-bisphosphatase, transketolase,fructose-bisphosphate aldolase, NADP-dependent glyceralde-hyde-3-phosphate dehydrogenase (Supplementary Table 1).

These results show that an accumulation of carbohydratescould decrease the photosynthesis rate via a mechanism involvingan inhibition of several Calvin-cycle enzymes. The redistribution ofcarbohydrate reserves might be linked to the up-regulation ofgenes encoding vacuolar invertase, sucrose synthase and alphaamylase, in infected Chardonnay plants (Table 1), and sucrose-phosphate synthase 1, in infected Manzoni (Table 2); theseenzymes convert sucrose and starch into fructose and glucose.

The accumulation of pathogenesis-related protein transcripts,observed in both the infected cultivars, requires a certain thresholdconcentration of hexoses, suggesting that hexose sensing isessential for mediating the activation of some defence-relatedgenes [55]. However, the increase in sucrose-cleaving enzymescould also be viewed as being part of phytoplasma pathogenesis. Infact, as observed in Spiroplasma citri infected Catharanthus roseus

plants, spiroplasma competes with the companion cells forfructose utilization and impairs sucrose loading by the companioncells into the phloem sieve tubes [47,49]. In Manzoni, theexpression of sugar transporters was induced, as it has also beenreported in Arabidopsis and tomato in response to fungal infections[56,57], and in grapevine in response to viral infection [58],hypothesizing a re-direction of sugars to regions colonized by thepathogen.

As expected, a significant number of genes modulated by BoisNoir interaction were related to defense, cell wall and response tostress. The expression of a group of proteins induced duringdefense response to pathogen and collectively referred to as‘‘pathogenesis-related proteins’’ (PR) was modulated in our study.

Induction of PRs has been found in many plant speciesbelonging to various families, suggestive of a general role forthese proteins in adaptation to biotic stress conditions. Consis-tently with the report that phytoplasma infection induces theaccumulation of PR-5 (thaumatin-like protein) in Garland chry-santhemum plants [59], we found that the expression level ofgenes coding PR-5 increased in infected Chardonnay and Manzoniplants. PR-1 and PR-2 (beta-1,3-glucanase) were induced inChardonnay. In Manzoni, PR-6 (protease inhibitor) and two


Germin-like proteins (GLP), that have various proposed roles inplant development and defense, were induced. A number of GLPhave been demonstrated to have oxalate oxidase (OXO) [60] orsuperoxide dismutase (SOD) [61] activity resulting in theproduction of H2O2. Interestinly genes coding for PR-1 proteinswere strongly induced only in Chardonnay; PR-1 is a marker ofsalicylic acid signalling pathway [62], and this may suggest thatsalicylic acid is involved in response to phytoplasma infection. Thisconfirms previous reports that PR proteins are associated withgrapevine defence against pathogens [63,64]. Differential expres-sion of these defence-related genes indicates the activation ofdefence pathways even in compatible interactions betweengrapevines and phytoplasmas. In fact, as it has been shown inother pathosystems, several susceptible hosts are not passiveagainst the pathogen and can set up a defence response, which ishowever not sufficiently effective to stop pathogen spread ininfected plant tissues [13,14].

In agreement with the notion that the plant resistance responseis mediated by a complex interplay of activating and repressingtranscriptional regulators [65], a large number of genes codingtranscription factors were modulated in the present study.Different families of transcription factors, such as NAM, NACand WRKY were induced during Bois Noir infection in bothcultivars. NAC and WRKY have been reported previously to play arole in senescence [66] and defence responses [65], reflecting theoverlap between these processes. Although zinc finger transcrip-tion factors of several classes were shown to be differentiallyexpressed in plant–pathogen interactions [67], in Manzoni theirexpression was repressed.

Interestingly, a Myb transcription factor was induced in Manzoniand repressed in Chardonnay. This class of transcription factors canbind to promoters of defense-associated genes [68], as alreadyreported for Arabidopsis, where several genes encoding Mybtranscription factors were up-regulated by P. syringae infections[69]. In addition, Myb classes appear to play a key role in theregulation of phenylpropanoid biosynthesis in a wide range of plantspecies [70]; the biosynthesis of phenylpropanoid compounds is infact developmentally activated in specific tissues and cell types, butit can also be activated in response to environmental stresses such aswounding and pathogen infection. The enzyme phenylalanineammonia-lyase (PAL) catalyzes the first metabolic step fromprimary metabolism into phenylpropanoid metabolism, which isthe deamination of phenylalanine to produce cinnamic acid. Thisstep is the flux control point into the phenylpropanoid pathway [70].Cinnamic acid is further modified by the actions of cinnamate 4-hydroxylase (C4H) and 4-coumarate:coenzyme A ligase (4CL),which catalyzes the formation of CoA esters of hydroxycinnamicacids, an intermediates used in the lignin biosynthesis.

Consistently with the observation that one of the mostdevastating symptoms in grapevine plants infected by phytoplas-mas is improper lignification of the young shoots, in Chardonnay theexpression of Myb transcription factors and phenylpropanoidsynthesis pathway genes PAL, C4H and 4CL was repressed, indicatinga blockage in the lignin biosynthesis and an higher susceptibility toBois Noir infection in this cultivar, compared to Manzoni.

Consistent with the predicted importance of cell wall modifica-tions in plant defense we found in both cultivars, in response toBois Noir infection, the repression of genes responsible for cell walldegradation (Supplementary Table 1, 2 and 3), while genesinvolved in cell wall reinforcement were induced (Tables 1–3).Products of the latter can act as a physical barrier to pathogenpenetration and restrict the diffusion of pathogen-synthesizedtoxins [71] and could limit new infection and the spread ofphytoplasma cells in the plant.

Although ROS accumulation alone is insufficient to triggerdisease resistance [72], they contribute to plant defense by

reinforcing the plant wall, by their toxicity to invading pathogensand by signalling further defense responses. The differentialexpression in infected Chardonnay and Manzoni plants of genesrelated to the production and neutralization of reactive oxigenspecies, such as glutathione S-transferase, alternative oxidase andlactoglutathione lyase-like protein, indicate that the oxidativeburst plays an important role in the grapevine–phytoplasmainteraction. On the other hand, ROS can cause severe cellulardamage, so are tightly regulated and detoxified by complexenzymatic and nonenzimatic mechanism. In the present studygenes encoding antioxidant protein, glutathione S-transferase(GST), were up- or down-regulated by phytoplasma, suggestingthat accumulation and detoxification of ROS simultaneously occurin the infected plants.

It has been reported that hydrogen peroxide accumulates in thephloem plasmalemma of recovered grapevine leaves, but not ineither healthy or diseased material [10]. Recovered grapevineplants might be able to achieve such H2O2 accumulation through aselective and presumably stable down-regulation of enzymaticH2O2 scavengers, without altering the levels of other antioxidantsystems and without incurring an increased oxidative risk. In thepresent study, in Manzoni a strong down-regulation of transcriptscoding for peroxidises and glutathione S-transferases wasobserved during Bois Noir infection, supporting the notion thatdown-regulation of ROS scavengers may influence the plantresponse to pathogen.

Grapevine phytoplasmas are detrimental to infected plants,affecting vitality, reducing yields and decreasing the quality ofvines. L-tartaric acid (TA) is the primary nonfermentable solubleacid in grapes and the principal acid in wine, contributingimportant aspects to the taste. Uniquely for a fruit acid, TAbiosynthesis begins with L-ascorbic acid (vitamin C), and a novelenzyme, having L-idonate dehydrogenase activity, was proposed tobe the key component in the tartaric acid biosynthetic pathway inVitaceous plants, converting vitamin C to TA [73]. In Bois Noirinfected Chardonnay plants, three transcripts coding for L-idonatedehydrogenase were repressed, suggesting the working hypoth-esis that this enzyme might cause the reduced wine quality in thephytoplasma infected grapevine plants.

In conclusion, we found that many genes involved in severalmetabolic pathways were modulated in both cultivars in responseto phytoplasma. The results obtained by microarray analysis werethen validated by quantitative RT-PCR (Fig. 3). In particular, moregenes, representing more GO categories, were differentiallyexpressed in Chardonnay than in Manzoni (Fig. 2, SupplementaryTables 4 and 5). The symptoms developed by two cultivars weredifferent as well, and Manzoni presented always milder symptomsthan Chardonnay. The future research will focus on most likelycandidate genes involved in tolerance response to Bois Noir ingrapevine, such as the members of Myb transcription factor family.

Acknowledgments

We would like to thank Elisa Angelini and Michele Borgo,Istituto Sperimentale per la Viticoltura di Conegliano (TV), Italy, foruseful discussions and for providing disease data, and AnselmoMontermini and Roberto Bondavalli, Consorzio FitosanitarioProvinciale, Reggio Emilia, Italy, for access to the experimentalfield and for helpful initial suggestions. We would also like to thankClaudio Moser, IASMAA, San Michele all’Adige, Italy.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.plantsci.2009.03.001.

http://dx.doi.org/10.1016/j.plantsci.2009.03.001


References

[1] G.P. Martelli, E. Boudon-Padieu, Directory of infectious diseases of grapevines,Opt. Medit. B 55 (2006) 157–194.

[2] E. Seemuller, H. Stolz, H. Kison, Persistence of the European stone fruit yellowsphytoplasma in aerial parts of Prunus taxa during the dormant season, J. Phyto-pathol. 146 (1998) 407–410.

[3] P.Y. Tan, T. Whitlow, Physiological responses of Catharanthus roseus (periwinkle)to ash yellows phytoplasmal infection, New Physiol. 150 (2001) 757–769.

[4] J.A. Matteoni, W.A. Sinclair, Stomatal closure in plants infected with mycoplasma-like organisms, Phytopathology 73 (1983) 398–402.

[5] P. Lepka, M. Stitt, E. Moll, E. Seemuller, Effect of phytoplasmal infection onconcentration and translocation of carbohydrates and amino acids in periwinkleand tobacco, Physiol. Mol. Plant Pathol. 55 (1999) 59–68.

[6] B.E. Maust, F. Espadas, C. Talavera, M. Aguilar, J.M. Santamarıa, C. Oropeza,Changes in carbohydrate metabolism in coconut palms infected with the lethalyellowing phytoplasma, Phytopathology 93 (2003) 976–981.

[7] C.J. Chang, I.M. Lee, Pathogenesis of diseases associated with mycoplasma-likeorganisms, in: U.S. Singh, R.P. Singh, K. Kohmoto (Eds.), Pathogenesis and HostSpecificity in Plant Diseases, vol. 1, Elsevier Science Publishing Co., New York,1995, pp. 237–246.

[8] R. Garau, A. Sechi, V.A. Prota, G. Moro, Productive parameters in Chardonnay andVermentino grapevines infected with ‘‘bios noir’’ and recovered in Sardinia, Bull.Insectol. 60 (2007) 233–234.

[9] R. Osler, N. Loi, L. Carraro, P. Ermacora, E. Refatti, Recovery in plants affected byphytoplasmas, in: Proceedings of the 5th Congress of the European Foundation forPlant Pathology, 1999, pp. 589–592.

[10] R. Musetti, R. Marabottini, M. Badiani, M. Martini, L. Sanita di Toppi, S. Borselli, M.Borgo, R. Osler, On the role of H2O2 in the recovery of grapevine (Vitis vinifera cv.Prosecco) from Flavescence doree disease, Funct. Plant Biol. 34 (2007) 750–758.

[11] R. Garau, G. Tolu, V.A. Prota, A. Sechi, Differential reactivity of grapevine cultivarsto ‘‘Bois Noir’’ infections in Sardinia, J. Plant Pathol. 86 (2004) 320.

[12] A. Bressan, S. Spiazzi, V. Girolami, E. Boudon-Padieu, Acquisition efficiency ofFlavescence doree phytoplasma by Scaphoideus titanus Ball from infected tolerantor susceptible grapevine cultivars or experimental host plants, Vitis 44 (2005)143–146.

[13] P.J. O’Donnell, E.A. Schmelz, P. Moussatche, S.T. Lund, J.B. Jonnes, H.J. Klee,Susceptible to intolerance—a range of hormonal actions in a susceptible Arabi-dopsis pathogen response, Plant J. 33 (2003) 245–257.

[14] N. Ehrenfeld, P. Canon, C. Stange, C. Medina, P. Arce-Johnson, Tobamovirus coatprotein CPCg induces an HR-like response in sensitive tobacco plants, Mol. Cells19 (2005) 1–9.

[15] K. Oshima, S. Kakizawa, H. Nishigawa, H.Y. Jung, W. Wei, S. Suzuki, R. Arashida, D.Nakata, S. Miyata, M. Ugaki, S. Namba, Reductive evolution suggested from thecomplete genome sequence of a plant–pathogenic phytoplasma, Nat. Genet. 36(2004) 27–29.

[16] X. Bai, J. Zhang, A.M. Ewing, A. Sally, R.A. Jancso, D.V. Shevchenko, K. Tsukerman, T.Walunas, A. Lapidus, J.W. Campbell, S.A. Hogenhout, Living with genome instabil-ity: the adaptation of phytoplasmas to diverse environments of their insect andplant hosts, J. Bacteriol. 188 (2006) 3682–3696.

[17] S. Jagoueix-Eveillard, F. Tarendau, K. Guolter, J.L. Danet, J.M. Bove, M. Garnier,Catharanthus roseus genes regulated differentially by mollicute infections, Mol.Plant Microbe Interact. 14 (2001) 225–233.

[18] V. Carginale, G. Maria, C. Capasso, E. Ionata, F. La Cara, M. Pastore, A. Bertaccini, A.Capasso, Identification of genes expressed in response to phytoplasma infectionin leaves of Prunus armeniaca by messenger RNA differential display, Gene 332(2004) 29–34.

[19] P. Pracros, J. Renaudin, S. Eveillard, A. Mouras, M. Hernould, Tomato flowerabnormalities induced by stolbur phytoplasma infection are associated withchanges of expression of floral development genes, Mol. Plant Microbe Interact.19 (2006) 62–68.

[20] M. Nicolaisen, D.P. Horvath, A branch-inducing phytoplasma in Euphorbia pul-cherrima is associated with changes in expression of host genes, J. Phytopathol.156 (2008) 403–407.

[21] M. Hren, M. Ravnikar, J. Brzin, P. Ermacora, L. Carraro, P.A. Bianco, P. Casati, M.Borgo, E. Angelini, A. Rotter, K. Gruden, Induced expression of sucrose synthaseand alcohol dehydrogenase I genes in phytoplasma-infected grapevine plantsgrown in the field, Plant Pathol. 58 (2008) 170–180.

[22] E.A.R. Tattersall, A. Ergul, F. Al Kayal, L. DeLuc, J.C. Cushman, G.R. Cramer,Comparison of methods for isolating high-quality RNA from leaves of grapevine,Am. J. Enol. Vitic. 56 (2005) 400–406.

[23] G.R. Cramer, A. Ergul, J. Grimplet, R.L. Tillett, A.R.E. Tattersall, M.C. Bohlman, D.Vincent, J. Sonderegger, J. Evans, C. Osborne, D. Quilici, K.A. Schlauch, D.A. Schooley,J.C. Cushman, Water and salinity stress in grapevines: early and late changes intranscript and metabolite profiles, Funct. Integr. Genomics 7 (2007) 111–134.

[24] L.G. Deluc, J. Grimplet, M.D. Wheatley, R.L. Tillett, D.R. Quilici, C. Osborne, D.A.Schooley, K.A. Schlauch, J.C. Cushman, G.R. Cramer, Transcriptomic and metabo-lite analyses of Cabernet Sauvignon grape berry development, BMC Genomics 8(2007) 429.

[25] J. Grimplet, L.G. Deluc, R.L. Tillett, M.D. Wheatley, K.A. Schlauch, G.R. Cramer, J.C.Cushman, Tissue-specific mRNA expression profiling in grape berry tissues, BMCGenomics 8 (2007) 187.

[26] S. Pilati, M. Perazzolli, A. Malossini, A. Cestaro, L. Dematte, P. Fontana, A. Dal Ri, R.Viola, R. Velasco, C. Moser, Genome-wide transcriptional analysis of grapevineberry ripening reveals a set of genes similarly modulated during three seasonsand the occurrence of an oxidative burst at veraison, BMC Genomics 8 (2007) 428.

[27] K. Mori, N. Goto-Yamamoto, M. Kitayama, K. Hashizume, Loss of anthocyanins inred-wine grape under high temperature, J. Exp. Bot. 58 (2007) 1935–1945.

[28] R.W.M. Fung, M. Gonzalo, C. Fekete, L.G. Kovacs, Y. He, E. Marsh, L.M. McIntyre,D.P. Schachtman, W. Qiu, Powdery mildew induces defense-oriented reprogram-ming of the transcriptome in a susceptible but not in a resistant grapevine, PlantPhysiol. 146 (2007) 236–249.

[29] X. Daire, D. Clair, W. Reinert, E. Boudon-Padieu, Detection and differentiation ofgrapevine yellows phytoplasmas belonging to the elm yellows group and to thestolbur subgroup by PCR amplification of non-ribosomal DNA, Eur. J. Plant Pathol.103 (1997) 507–514.

[30] E. Angelini, D. Clair, M. Borgo, A. Bertaccini, E. Boudon-Padieu, Flavescence doreein France and Italy—occurrence of closely related phytoplasma isolates and theirnear relationships to palatinate grapevine yellows and an alder yellows phyto-plasma, Vitis 40 (2001) 79–86.

[31] S. Deng, C. Hiruki, Amplification of 16S rRNA genes from culturable and non-culturable Mollicutes, J. Microbiol. Methods 14 (1991) 53–61.

[32] C.D. Smart, B. Schneider, C.L. Blomquist, L.J. Guerra, N.A. Harrison, U. Ahrens, K.H.Lorenz, E. Seemuller, B.C. Kirkpatrick, Phytoplasma-specific PCR primers based onsequences of 16S-23S rRNA spacer region, Appl. Environ. Microbiol. 62 (1996)2988–2993.

[33] K.S. Gibb, A.C. Padovan, B.D. Mogen, Studies on sweat potato little-leaf Phyto-plasma detected in sweet potato and other plant species growing in northernAustralia, Phytopathology 85 (1995) 169–174.

[34] A.C. Padovan, K.S. Gibb, A. Bertaccini, R.E. Bonfiglioli, P.A. Magnet, B.B. Sears,Molecular detection of the Australian grapevine yellows phytoplasma and com-parison with grapevine yellows phytoplasmas from Italy, Aust. J. Grape Wine Res.1 (1995) 25–31.

[35] A. Bertaccini, L. Carraro, D. Davies, M. Laimer Da Camara Machado, M. Martini, S.Paltrinieri, E. Seemuller, Micropropagation of a collection of phytoplasma strainsin periwinkle and other host plants, in: Proc. XIII Congress of IOM, ACROSFukuoka, 14–19 July, (2000), p. 101.

[36] F. Goes da Silva, A. Iandolino, F. Al-Kayal, M.C. Bohlmann, M.A. Cushman, H. Lim, A.Ergul, R. Figueroa, E.K. Kabuloglu, C. Osborne, J. Rowe, E. Tattersall, A. Leslie, J. Xu,J.M. Baek, G.R. Cramer, J.C. Cushman, D.R. Cook, Characterizing the grape tran-scriptome. Analysis of expressed sequence tags from multiple Vitis species anddevelopment of a compendium of gene expression during berry development,Plant Physiol. 139 (2005) 574–597.

[37] R.A. Irizarry, B. Hobbs, F. Collin, Y.D. Beazer-Barclay, K.J. Antonellis, U. Scherf, T.P.Speed, Exploration, normalization, and summaries of high density oligonucleo-tide array probe level data, Biostatistics 4 (2003) 249–264.

[38] C. Espinoza, C. Medina, S. Somerville, P. Arce-Johnson, Senescence-associatedgenes induced during compatible viral interactions with grapevine and Arabi-dopsis, J. Exp. Bot. 58 (2007) 3197–3212.

[39] Y. Benjamini, Y. Hochberg, Controlling the false discovery rate: a practical andpowerful approach to multiple testing, J. R. Stat. Soc. Ser. B 57 (1995) 289–300.

[40] S. Rosen, H. Skaletsky, Primer 3 on the WWW for general users and for biologistprogrammers, in: S. Krawtez, S. Misener (Eds.), Bioinformatics Methods and Pro-tocols: Methods in Molecular Biology, Humana, Totowa, NJ, 2000, pp. 365–386.

[41] O. Jaillon, J.M. Aury, B. Noel, et al., The grapevine genome sequence suggestsancestral hexaploidization in major angiosperm phyla, Nature 449 (2007) 463–468.

[42] F. Vignols, M. Jose-Estanyol, D. Caparros-Ruiz, J. Rigau, P. Puigdomenech, Involve-ment of a maize proline-rich protein in secondary cell wall formation as deducedfrom its specific mRNA localization, Plant Mol. Biol. 39 (1999) 945–952.

[43] Z.H. He, D. He, B.D. Kohorn, Requirement for the induced expression of a cell wallassociated receptor kinase for survival during the pathogen response, Plant J. 14(1998) 55–63.

[44] S. Razin, Mycoplasma taxonomy and ecology, in: J. Maniloff, R.N. McElhaney, L.R.Finch, J.B. Baseman (Eds.), Mycoplasma: Molecular Biology and Pathogenesis,American Society for Microbiology, Washington, DC, 1992, pp. 3–22.

[45] J.R. Alfano, A. Collmer, The type III (Hrp) secretion pathway of plant pathogenicbacteria: trafficking harpins, Avr proteins, and death, J. Bacteriol. 179 (1997)5655–5662.

[46] A.J. Simpson, F.C. Reinach, P. Arruda, F.A. Abreu, M. Acencio, R. Alvarenga, L.M.Alves, J.E. Araya, G.S. Baia, C.S. Baptista, The genome sequence of the plantpathogen Xylella fastidiosa, Nature 406 (2000) 151–157.

[47] A. Andre, M. Maucourt, A. Moing, D. Rolin, J. Renaudin, Sugar import andphytopathogenicity of Spiroplasma citri: glucose and fructose play distinct roles,Mol. Plant Microbe Interact. 18 (2005) 33–42.

[48] P. Gaurivaud, F. Laigret, E. Verdin, M. Garnier, J.M. Bove, Fructose operon mutantsof Spiroplasma citri, Microbiology 146 (2000) 2229–2236.

[49] P. Gaurivaud, J.L. Danet, F. Laigret, M. Garnier, J.M. Bove, Fructose utilization andphytopathogenicity of Spiroplasma citri, Mol. Plant Microbe Interact. 13 (2000)1145–1155.

[50] K. Oshima, S. Kakizawa, R. Arashida, Y. Ishii, A. Hoshi, Y. Hayashi, S. Kagiwada, S.Namba, Presence of two glycolytic gene clusters in a severe pathogenic line ofCandidatus Phytoplasma asteris, Mol. Plant Pathol. 8 (2007) 481–489.

[51] M.T. Cousin, E. Boudon-Padieu, Phytoplasma and phytoplasma diseases: vectors,control, and research topics, Cahiers d’etudes et de recherches francophones/Agricultures 11 (2002) 115–126.

[52] S. Sabanadzovic, N. Abou-Ghanem, M.A. Castellano, M. Digiaro, G.P. Martelli,Grapevine fleck virus-like viruses in Vitis, Arch. Virol. 145 (2000) 553–565.

[53] A. Gibly, A. Bonshtien, V. Balaji, P. Debbie, G.B. Martin, G. Sessa, Identification andexpression profiling of tomato genes differentially regulated during a resistanceresponse to Xanthomonas campestris pv. Vesicatoria, Mol. Plant Microbe Interact.17 (2004) 1212–1222.


[54] S. Restrepo, K.L. Myers, O. del Pozo, G.B. Martin, A.L. Hart, C.R. Buell, W.E. Fry, C.D.Smart, Gene profiling of a compatible intercation between Phytohtora ifestans andSolanum tuberosum suggest a role for a Carbonic anhydrase, Mol. Plant MicrobeInteract. 18 (2005) 913–922.

[55] A. Sturm, G.Q. Tang, The sucrose-cleaving enzymes of plants are crucial for devel-opment, growth and carbon partitioning, Trends Plant Sci. 4 (1999) 401–407.

[56] V. Fotopoulos, M.J. Gilbert, J.K. Pittman, A.C. Marvier, A.J. Buchanan, N. Sauer, J.L.Hall, L.E. Williams, The monosaccharide transporter gene, AtSTP4, and the cell-wall invertase, Atbetafruct1, are induced in Arabidopsis during infection with thefungal biotroph Erysiphe cichoracearum, Plant Physiol. 132 (2003) 821–829.

[57] S. Garcia-Rodriguez, M.J. Pozo, C. Azcon-Aguilar, N. Ferrol, Expression of a tomatosugar transporter is increased in leaves of mycorrhizal or Phytophthora parasitica-infected plants, Mycorrhiza 15 (2005) 489–496.

[58] C. Espinoza, A. Vega, C. Medina, K. Schlauch, G. Cramer, P. Arce-Johnson, Geneexpression associated with compatible viral diseases in grapevine cultivars,Funct. Integr. Genomics 7 (2006) 95–110.

[59] B.X. Zhong, Y.W. Shen, Accumulation of pathogenesis-related type-5 like proteinsin phytoplasma-infected garland chrysanthemum Chrysanthemum coronarium,Acta Biochim. Biophys. Sin. 36 (2004) 773–779.

[60] F. Zhou, Z. Zhang, P.L. Gregersen, J.D. Mikkelsen, E. de Neergaard, D.B. Collinge, H.Thordal-Christensen, Molecular characterization of the oxalate oxidase involvedin the response of barley to the Powdery Mildew fungus, Plant Physiol. 117 (1998)33–41.

[61] A.B. Christensen, H. Thordal-Christensen, G. Zimmermann, T. Gjetting, M.F.Lyngkjær, R. Dudler, P. Schweizer, The Germinlike Protein GLP4 exhibits super-oxide dismutase activity and is an important component of quantitative resis-tance in wheat and barley, Mol. Plant Microbe Interact. 17 (2004) 109–117.

[62] C.M.J. Pieterse, L.C. van Loon, Salicylic acid-independent plant defence pathways,Trends Plant Sci. 4 (1999) 52–58.

[63] A.K. Jacobs, I.B. Dry, S.P. Robinson, Induction of different pathogenesis relatedcDNAs in grapevine infected with Powdery mildew and treated with ethephon,Plant Pathol. 48 (1999) 325–336.

[64] A.S. Renault, A. Deloire, I. Letinois, E. Kraeva, C. Tesniere, A. Ageorges, C. Redon, J.Brierne, B-1,3-glucanse gene expression in grapevine leaves as a response toinfection with Botrytis cinerea, Am. J. Enol. Vitic. 51 (2000) 81–87.

[65] T. Eulgem, Regulation of the Arabidopsis defense transcriptome, Trends Plant Sci.10 (2005) 71–78.

[66] Y. Guo, Z. Cai, S. Gan, Transcriptome of Arabidopsis leaf senescence, Plant CellEnviron. 27 (2004) 521–549.

[67] A. Bonshtien, A. Lev, A. Gibly, P. Debbie, A. Avni, G. Sessa, Molecular properties ofthe Xanthomonas AvrRxv effector and global transcriptional changes determinedby its expression in resistant tomato plants, Mol. Plant Microbe Interact. 8 (2005)300–310.

[68] P.J. Rushton, I.E. Somssich, Transcriptional control of plant genes responsive topathogens, Curr. Opin. Plant Biol. 1 (1998) 311–315.

[69] R. Stracke, M. Werber, B. Weisshaar, The R2R3-MYB gene family in Arabidopsisthaliana, Curr. Opin. Plant Biol. 4 (2001) 447–456.

[70] R.A. Dixon, L. Achnine, P. Kota, C.J. Liu, M.S.S. Reddy, L. Wang, The phenylpropa-noid pathway and plant defence—a genomics perspective, Mol. Plant Pathol. 3(2002) 371–390.

[71] C.S. Bestwick, I.R. Brown, J.W. Mansfield, Localized changes in peroxidase activityaccompany hydrogen peroxide generation during the development of a non-hosthypersensitive reaction in lettuce, Plant Physiol. 118 (1998) 1067–1078.

[72] M.C. Heath, Nonhost resistance and nonspecific plant defenses, Curr. Opin. PlantBiol. 3 (2000) 315–319.

[73] S. DeBolt, D.R. Cook, C.M. Ford, l-Tartaric acid synthesis from vitamin C in higherplants, Proc. Natl. Acad. Sci. U.S.A. 103 (2006) 5608–5613.

Gene expression in grapevine cultivars in response to Bois Noir phytoplasma infection

Documents