Genome-wide transcriptional analysis of grapevine berry ripening reveals a set of genes similarly...

BioMed CentralBMC Genomics

ss

Open AcceResearch articleGenome-wide transcriptional analysis of grapevine berry ripening reveals a set of genes similarly modulated during three seasons and the occurrence of an oxidative burst at vegraveraisonStefania Pilati1 Michele Perazzolli14 Andrea Malossini2 Alessandro Cestaro1 Lorenzo Demattegrave13 Paolo Fontana1 Antonio Dal Ri1 Roberto Viola1 Riccardo Velasco1 and Claudio Moser1

Address 1Department of Genetics and Molecular Biology IASMA Research Center Via E Mach 1 38010 S Michele aAdige (TN) Italy 2Department of Information and Communication Technology University of Trento Via Sommarive 14 38050 Povo (TN) Italy 3Lorenzo Demattegrave Microsoft Research-University of Trento Centre Piazza Manci 17 38050 Povo (TN) Italy and 4Michele Perazzolli SafeCrop Centre Istituto Agrario San Michele aAdige Via E Mach 1 38010 S Michele aAdige (TN) Italy

Email Stefania Pilati - stefaniapilatiiasmait Michele Perazzolli - micheleperazzolliiasmait Andrea Malossini - malossinditunitnit Alessandro Cestaro - alessandrocestaroiasmait Lorenzo Demattegrave - demattecosbieu Paolo Fontana - paolofontanaiasmait Antonio Dal Ri - antoniodalriiasmait Roberto Viola - robertoviolaiasmait Riccardo Velasco - riccardovelascoiasmait Claudio Moser - claudiomoseriasmait

Corresponding author

AbstractBackground Grapevine (Vitis species) is among the most important fruit crops in terms ofcultivated area and economic impact Despite this relevance little is known about thetranscriptional changes and the regulatory circuits underlying the biochemical and physical changesoccurring during berry development

Results Fruit ripening in the non-climacteric crop species Vitis vinifera L has been investigated atthe transcriptional level by the use of the Affymetrix Vitis GeneChipreg which contains approximately14500 unigenes Gene expression data obtained from berries sampled before and after veacuteraison inthree growing years were analyzed to identify genes specifically involved in fruit ripening and toinvestigate seasonal influences on the process From these analyses a core set of 1477 genes wasfound which was similarly modulated in all seasons We were able to separate ripening specificisoforms within gene families and to identify ripening related genes which appeared stronglyregulated also by the seasonal weather conditions Transcripts annotation by Gene Ontologyvocabulary revealed five overrepresented functional categories of which cell wall organization andbiogenesis carbohydrate and secondary metabolisms and stress response were specifically inducedduring the ripening phase while photosynthesis was strongly repressed About 19 of the coregene set was characterized by genes involved in regulatory processes such as transcription factorsand transcripts related to hormonal metabolism and signal transduction Auxin ethylene and lightemerged as the main stimuli influencing berry development In addition an oxidative burstpreviously not detected in grapevine characterized by rapid accumulation of H2O2 starting fromveacuteraison and by the modulation of many ROS scavenging enzymes was observed

Published 22 November 2007

BMC Genomics 2007 8428 doi1011861471-2164-8-428

Received 10 August 2007Accepted 22 November 2007

This article is available from httpwwwbiomedcentralcom1471-21648428

copy 2007 Pilati et al licensee BioMed Central Ltd This is an Open Access article distributed under the terms of the Creative Commons Attribution License (httpcreativecommonsorglicensesby20) which permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

of 22(page number not for citation purposes)

BMC Genomics 2007 8428 httpwwwbiomedcentralcom1471-21648428

Conclusion The time-course gene expression analysis of grapevine berry development hasidentified the occurrence of two well distinct phases along the process The pre-veacuteraison phaserepresents a reprogramming stage of the cellular metabolism characterized by the expression ofnumerous genes involved in hormonal signalling and transcriptional regulation The post-veacuteraisonphase is characterized by the onset of a ripening-specialized metabolism responsible for thephenotypic traits of the ripe berry Between the two phases at veacuteraison an oxidative burst and theconcurrent modulation of the anti-oxidative enzymatic network was observed The large numberof regulatory genes we have identified represents a powerful new resource for dissecting themechanisms of fruit ripening control in non-climacteric plants

BackgroundGrape is among the most ancient widely cultivated fruitcrops The keen interest in the understanding of grapeberry ripening is justified by the economic relevance of thequality of grapes and their processed products such aswine juice and dried fruit The onset of the genomic erahas brought unprecedented progress into our knowledgeof Vitis molecular biology and powerful genetic andgenomic tools are now available or are being developedsuch as mapping populations genetic and physical maps[12] an extensive ESTs collection (342576 sequencesdeposited at the NCBI EST database-March 19 2007 [3])and several tools for gene expression analysis Theseresources enable us to identify grape as a model plant forstudies on non-climacteric anthocyanin accumulatingfruits beside tomato which is currently the model plantfor climacteric carotenoid accumulating fruits Grapeberry development from anthesis to ripening is classicallydivided into three phases on the basis of chemical andmorphological traits [4] berry formation characterizedby exponential growth of the berry and accumulation oforganic acids mainly malate in the vacuole veacuteraison atransition phase during which growth declines and berriesstart to change colour and soften ripening characterizedby an increase in pH marked berry growth and accumula-tion of sugars anthocyanins and flavour-enhancing com-pounds Published studies based on EST analysis [5-7]differential display [8] cDNA-arrays [9] and oligo arrays[10] have shown distinct and extensive changes in thetranscriptome occurring during ripening These studiesreport up to 740 developmentally regulated genes mainlyinvolved in defence stress response primary and second-ary metabolism berry growth and hormonal metabolismThe very recent study of Grimplet et al [11] provides anextensive description of these functional classes investi-gating their tissue-specific expression in the ripe berry bymeans of the Affymetrix Vitis GeneChipreg

With the aim of understanding the transcriptionalchanges underlying grapevine berry ripening and the reg-ulatory mechanisms acting at this level we were able toidentify 1477 genes modulated during berry develop-ment comprising 282 regulatory factors in addition to the

structural genes by the use of the Affymetrix Vitis Gene-Chipreg The results obtained from replicated experimentscarried out in three growing seasons provide also insightsinto the relationship between genetic and environmentaleffects on berry development control

Results and DiscussionGrape berries phenotypic evaluationsFigure 1 shows the changes in average berry weightorganic acids total soluble solids and anthocyanin con-tent of grape berries sampled weekly during the 20032005 and 2006 growing seasons The date of veacuteraison wasset at approximately 8 weeks post-flowering (wpf) in allyears In order to characterize berry ripening we chosethree distinct moments according to [12] the last stage ofberry formation time-point A (TP A stage E-L 33) char-acterized by small hard green berries still accumulatingorganic acids and two stages during berry ripening TP B(stage E-L 34) just before veacuteraison characterized by ber-ries in the green state with maximum acidic content start-ing to soften and TP C (stage E-L 36) when the ripeningprocess is well established and berries are growing fastcolouring and accumulating sugars The identification ofthe corresponding developmental stages in the three yearswas based on the biochemical profiles As average temper-atures during 2003 remained constantly unusually higherfrom April to October [see Additional file 1] floweringand harvest occurred approximately one week earlier in2003

Statistical analysisRNA extracted in triplicate from berries harvested at threetime points was used to hybridize 27 Affymetrix VitisGeneChipsreg Datasets from each season were analyzedseparately The quality of the biological replicates wasevaluated by the R-squared coefficient [see Additional file2] which ranged from 089 to 099 Minimum R-squaredvalues were found for TP B samples suggesting a highersample heterogeneity approaching veacuteraison when dra-matic transcriptional changes occur in less than 48 hours[10]

Page 2 of 22(page number not for citation purposes)


Principal Component Analysis (PCA) confirmed the uni-formity of the biological replicates as the nine groups ofreplicates clusterised tightly (Fig 2) The greatest variancein gene expression was found between samples from thethree growing seasons as they were separated along thefirst component (51 of the total variance) In particularthe largest difference was evident between the 2003expression data compared to the 2005 and 2006 suggest-ing that the environmental conditions greatly affect theberry transcriptome [see Additional file 1] On the otherhand the second principal component (24 of the totalvariance) separated neatly the pre-veacuteraison stages (TP Aand TP B) from the post-veacuteraison one (TP C) confirmingprevious observations of extensive transcriptional changesoccurring from berry formation to berry ripening [10]

Class comparison analysis by means of Significance Anal-ysis of Microarray [13] imposing a minimum fold changeof 2 in at least one comparison (TP A vs TP B and TP B vsTP C) produced three sets of 2632 4238 and 3856 differ-entially expressed probesets for the 2003 2005 and 2006samples respectively The size difference between thethree sets likely reflects inter-seasonal biological differ-ences The comparison of the three datasets identified acommon set of 1700 modulated transcripts and amongthese 1477 transcripts with a conserved profile in the

Principal component analysis (PCA) of the 27 expression datasetsFigure 2Principal component analysis (PCA) of the 27 expres-sion datasets Scaled expression data relative to 2003 (red) 2005 (blue) and 2006 (green) samples are sharply separated according to the sampling year (first principal component) and to the developmental stage (second principal compo-nent) Biological replicates relative to time point A B and C are represented as circles triangles and downwards triangles respectively

Biochemical changes observed during Pinot Noir berry devel-opment in 2003 (red) 2005 (blue) and 2006 (green)Figure 1Biochemical changes observed during Pinot Noir berry development in 2003 (red) 2005 (blue) and 2006 (green) Total acids (expressed as grams of tartaric acid per liter of must) total soluble solids and anthocyanins content and berry weight averaged on a pool of 50 berries are reported The dashed area indicates veacuteraison The weeks post flowering (wpf) corresponding to the samples used for the hybridization experiments are enclosed in boxes



three years (Pearsons correlation coefficient gt05 in anypairwise comparison) This 1477 core dataset which cor-responds to 9 of the chip probesets is listed in Addi-tional file 3 It represents a highly conserved expressionnetwork reasonably involved in an internal basic func-tional program such as berry ripening

Annotation of the differentially expressed genesAutomatic annotation of all the Vitis GeneChipreg

sequences was performed using the Gene Ontology (GO)classification [14] The core set of 1477 differentiallyexpressed genes was subsequently manually checked andintegrated with additional GO biological process termsTranscripts were then grouped into 17 GO functional cat-egories [see Additional file 3 and 4] whose distributionamong the entire chip and the core set was statisticallycompared (Fig 3) Nine categories were differentially rep-resented at significant levels during berry developmentand five of these were overrepresented in the regulatedgene core set These were carbohydrate metabolismcell wall organization and biogenesis response tostimulus secondary metabolism and photosynthe-

sis all processes that ought to be specifically regulatedduring grape berry development The four underrepre-sented categories were nucleobase nucleoside nucle-otide and nucleic acid metabolism proteinbiosynthesis protein metabolism and transportThese classes may either play a minor role in the develop-mental program or their involvement could be restrictedto a small subset of genes

The functional class distribution of the Vitis GeneChipreg

was also compared to that of the Arabidopsis genome avail-able at TAIR ([15] [Additional file 5]) The overall agree-ment in the class distributions confirms the VitisGeneChipreg as a reliable toolkit for the investigation of thegrape expressed genome

Cluster analysis and functional categories modulation during berry ripeningCluster analysis of the gene core set was based on the k-means method using Pearsons correlation distance calcu-lated on the gene expression profiles Transcripts weredivided into eight groups representing the minimum

Functional categories distribution in the core set of the modulated genes (red) and in the entire Vitis GeneChipreg (grey)Figure 3Functional categories distribution in the core set of the modulated genes (red) and in the entire Vitis Gene-Chipreg (grey) Frequencies are calculated as percentage of the whole number of GO biological process terms (1825 and 21890 in the modulated and chip sets respectively) Nine classes marked by an asterisk and written in bold resulted differ-ently represented in the modulated set compared to the entire chip after statistical analysis (p-value lt 0001) In italics are the five categories overrepresented in the modulated set



number of profiles that could be obtained with threetime-points We observed good agreement between clus-tering in the three gene sets (80 of the transcripts fell inthe same cluster in all seasons) We decided to use the2005 expression data for the cluster representation (Fig4) due to the smaller variance among replicates [see Addi-tional file 2] As expected from PCA the two most popu-lated clusters were number 4 and 8 composed by genesmodulated positively and negatively from TP B to TP Crespectively Conversely clusters 2 and 6 consisting ofgenes modulated only during the pre-veacuteraison intervalwere the less populated ones Approximately the samenumber of genes was positively or negatively modulatedalong the whole study interval (cluster 3 and 7) Genesspecifically induced around veacuteraison should fall incluster1 while genes with roles in early and late stages ofberry development should fall in cluster 5

Functional class distribution frequency was then calcu-lated for each cluster and represented as histogram [Addi-tional file 6] To better understand the modulation of thefunctional classes during berry ripening their frequencieswere determined after splitting the experimental intervalinto two phases one from TP A to TP B and one from TPB to TP C The genes modulated during the first phasederived from clusters 1 2 3 (induced) and 5 6 7(repressed) while the genes modulated during the secondphase derived from clusters 3 4 5 (induced) and 1 7 8(repressed) During the pre-veacuteraison phase the number ofthe induced genes (408) exceeded that of the repressedones (270) while the opposite occurred during the fol-lowing phase (651 induced vs 1118 repressed) Asdepicted in Figure 5 all but two categories showed a gen-eral tendency towards an opposite modulation betweenthe first and the second phase Furthermore in the firstphase a general bias towards induction for the GO catego-ries involved in regulatory mechanisms namely nucleo-base nucleoside nucleotide and nucleic acidmetabolism (mainly transcription factors) response toendogenous stimulus (mainly hormone metabolism)signal transduction and protein metabolism wasobserved This suggests a strong cell re-programming tak-ing place in berry cells up to veacuteraison During the secondphase a marked negative regulation is evident for catego-ries involved in cell division such as organelle organiza-tion and biogenesis and protein biosynthesis andphotosynthesis This behaviour is coherent with theslowing down of cell replication and the loss of photosyn-thetic capacity On the other hand the classes amino acidand derivative metabolism carbohydrate metabolismcell wall organization and biogenesis developmentlipid metabolism primary and secondary metabo-lism display a positive regulation This indicates the prev-alence after veacuteraison of metabolic processes involved incell wall loosening and synthesis sugar accumulation and

synthesis and transport of metabolites responsible ofgrape colour and flavour Finally the class response tostimulus appeared induced in both phases suggestingthat throughout the entire berry development process thecell devotes considerable effort to face different kinds ofstimuli presumably including osmotic oxidative andbiotic stress

Gene regulation of grape berry ripeningOur study of the berry ripening process identified 125transcription factors 65 genes involved in hormonemetabolism and response and 92 involved in signal trans-duction corresponding to 85 44 and 62 of theregulated core set respectively These figures and similardata recently obtained for tomato and Citrus fruit [1617]suggest that a consistent fraction of the modulated fruittranscriptome is devoted to the control of the develop-mental program In this section we present our main con-tribution to the knowledge of gene regulation duringberry ripening in grapevine A restricted list of selectedgenes involved in the process and discussed in the text isreported in Table 1 with the specific AffyID whereas thecomplete list is provided in part A of Additional file 7

Hormones metabolism and signallingAmongst the genes related to hormone metabolism in thecore set those related to auxin and ethylene were the mostrepresented followed by those related to abscisic acid andbrassinosteroids Few genes related to cytokinines gib-berellins or jasmonic acid were found These observationsare consistent with our knowledge of the relative impor-tance of these signalling pathways during grape berrydevelopment [1819]

In fruits from both climacteric and non-climacteric spe-cies auxins are known to be more abundant immediatelyafter ovary fertilization and to mediate important signalsfor the onset of fruit development [2021] In grape it hasbeen generally accepted that indole-3-acetic acid (IAA)content reaches its maximal level just after anthesis andthen declines to very low levels in the ripe fruit [22] evenif a recent study on Vitis vinifera cv Cabernet Sauvignondid not confirm it [23] The expression of genes involvedin cell wall metabolism (GRIP4) anthocyanin synthesis(CHS and UFGT) and sugar storage (GIN1) and the accu-mulation of the correlated metabolites were all shown tobe retarded by auxin treatments at veacuteraison [19]

In the present study we have identified a gene homolo-gous to the Arabidopsis amidase AtAMI1 which in vitrosynthesizes IAA from indole-3-acetamide [24] and itsdescending profile during ripening agrees well with thedecline in IAA levels observed in Vitis labrusca berries [22]We have also found a gene homologous to an ArabidopsisIAA-amino acid synthetase which may contribute to IAA




Cluster analysis of the expression profiles of the modulated core setFigure 4Cluster analysis of the expression profiles of the modulated core set The expression profiles of the 1477 modulated genes during P Noir berry ripening were clusterised in eight clusters which represent the minimum number of profiles consid-ering three time points Clusters were obtained by the k-means method using Pearsons correlation distance The representa-tive profile and the number of genes in each cluster are indicated


intracellular homeostasis via amino acid conjugation ofexcess IAA [25] Two auxin carriers (an AUX1-like and aPIN1-like) putatively mediating auxin efflux were alsoexpressed before veacuteraison

We detected numerous auxin-regulated genes belongingto the AuxIAA Small Auxin-Up RNA and GH3 gene fam-ilies These families are characterized by the presence of anAuxin Responsive Element in the promoter region of theirmembers recognized by specific Auxin Response Factors(ARFs) Two ARFs ARF5 and ARF18 and an auxin receptorof the ABP family were found expressed during the pre-veacuteraison stage and then steeply repressed during ripeningWith the exception of two AuxIAA proteins which wereinduced during ripening all the other transcripts pre-sented profiles with either a peak of expression aroundveacuteraison or a decreasing profile after it The auxin respon-sive genes seem therefore timely co-ordinated with thecellular auxin level reflecting its prevalent role in the firststages of berry development We have also detected a gene(1612001_s_at) homologous to a Capsicum chinense GH3gene which has been proposed as point of convergence

between auxin and ethylene signals in non-climactericfruit development [26] While in pepper this GH3 tran-script is up-regulated during ripening due to its ethylene-inducibility our analyses showed a steep repression of thegrape homologue after TP B

The role of ethylene during development and ripening ofsome non-climacteric fruits was suggested since the 1970sand has recently been well documented [27-29] In grapea small and transient increase of endogenous ethyleneproduction occurs just before veacuteraison in concomitancewith an increase in 1-aminocyclopropane-1-carboxylicacid (ACC) oxidase activity [29] Our results well suitthese reports as we observed a peak of expression of ACCoxidase around veacuteraison while ACC synthase and a puta-tive ethylene-forming-enzyme dioxygenase detected withsimilar profile also by others [10] were expressed untilveacuteraison and then repressed

We were not able to identify any receptor or kinaseinvolved in ethylene signal transduction in grape withinthe modulated gene set Instead we detected four tran-

Functional categories distribution before and after-veacuteraisonFigure 5Functional categories distribution before and after-veacuteraison Induced and repressed functional categories during the first stage (from TP A to TP B green histograms) derive from clusters 1 2 3 and 5 6 7 respectively while categories induced and repressed during the second stage (from TP B to TP C violet histograms) derive from clusters 3 4 5 and 1 7 8 respec-tively The relative number of induced (Ni) or repressed (Nr) genes for a specific functional category was calculated with respect to the total number of induced or repressed genes in each stage The estimate of induction or repression within each functional category was then calculated as follows estimate of inductionrepression = (Ni - Nr)(Ni + Nr) Positive values rep-resent an overall induction and negative values an overall repression



Table 1 Selection of genes putatively involved in the regulation of berry ripening

AffyID TIGR TC Clustera UniprotIDb Descriptionb E-value(b)

HORMONE METABOLISM AND SIGNALING

AUXINAuxin synthesis1610647_at TC40710 8 Q9FR37 Amidase (At1g08980 AtAMI1) 389E-152Auxin transport1612060_at TC48033 6 Q76DT1 AUX1-like auxin influx carrier protein 297E-1431621946_at TC50194 7 Q8H0E0 PIN1-like auxin transport protein 105E-51AuxIAA1618875_s_at TC46557 1 Q8LAL2 IAA26 (Phytochrome-associated protein 1) 295E-631617513_at 8 Q8LAL2 IAA26 (Phytochrome-associated protein 1) 604E-621615985_at TC42737 1 O24542 AUX22D 240E-691614660_at TC45938 4 P13088 AUX22 635E-601621754_at TC41728 8 Q84V38 AuxIAA protein 919E-891615728_at TC45614 8 Q84V38 AuxIAA protein 000E+001611390_a_at TC39148 8 Q8RW16 AuxIAA protein 292E-981620512_at TC46080 8 Q9ZSY8 IAA27 (Phytochrome-associated protein 2) 942E-861613468_at TC38800 8 Q8LSK7 Auxin-regulated protein (IAA1) 338E-99SAUR1621201_at TC41864 1 Q6ZKQ7 Auxin-induced protein (SAUR) 599E-15GH31610880_s_at CF371851 8 O22190 Indole-3-acetic acid-amido synthetase GH33 178E-331612001_s_at TC43261 8 Q6QUQ3 Auxin and ethylene responsive GH3-like protein 667E-1111607503_s_at TC38370 8 Q52QX4 Auxin-repressed protein-like protein ARP1 250E-48Other auxin responsive proteins1619856_at TC45354 1 O48629 Putative auxin-repressed protein (dormancy) 641E-441621185_s_at TC45354 1 O48629 Putative auxin-repressed protein (dormancy) 641E-441613204_at TC38341 2 O49235 24-D inducible glutathione S-transferase 174E-981619913_at TC46727 4 Q9SV71 auxin-responsive protein 597E-116ARF (transcription factor)1612180_at TC51304 8 Q6L8T9 Auxin response factor 5 124E-641616225_at TC44210 8 Q9C5W9 Auxin response factor 18 369E-88Auxin receptor1612090_s_at TC45186 8 Q9ZRA4 Auxin-binding protein ABP19a precursor 175E-88ETHYLENEEthylene synthesis1615952_s_at TC38453 1 Q84X67 1-aminocyclopropane-1-carboxylic acid oxidase 1 ACO 120E-1671622308_at TC40451 8 Q8S933 1-aminocyclopropane-1-carboxylate synthase ACC synthase 392E-681612699_at BQ798614 8 Q9XIA5 Similar to ethylene-forming-enzyme-like dioxygenase 503E-29Ethylene responsive transcription factors1617012_at TC45046 3 Q8S9H4 Tomato Ethylene response factor 1 510E-611611910_s_at - 8 Q6TKQ3 Putative ethylene response factor ERF3b 726E-1241609990_at TC39828 8 Q6TKQ3 Putative ethylene response factor ERF3b 145E-1231620278_at TC46710 3 Q6J9Q4 Putative AP2EREBP transcription factor 274E-641621270_at TC47575 8 Q9LYD3 Putative AP2EREBP transcription factor (DREB3) 334E-63Ethylene induced genes1618213_at - 1 Q9SWV2 ER6 protein (Fragment) 389E-181619086_at TC39832 8 Q93W91 Putative ER6 protein 255E-511609780_at TC47621 8 Q94E74 Putative ER6 protein Ethylene-responsive protein 574E-331613123_at CF215236 1 Q8S3D1 bHLH transcription factor (tomato ER33 homolog) 104E-611613177_at TC47351 8 Q9LSP8 Ethylene-induced calmodulin-binding transcription activator 376E-941622850_at TC49495 3 O48631 Ethylene-forming-enzyme-like dioxygenase (tomato E8 homolog) 727E-78Pathogenesis related genes putatively induced by ethylene1611058_at TC47119 4 Q7XAJ6 Putative pathogenesis related protein 1 precursor 155E-731618835_s_at TC38693 4 O81228 PR-4 type protein 162E-801622360_at TC40938 4 Q42966 NtNitrilase 4B (EC 3551) 105E-135ABSCISIC ACIDAbscisic acid synthesis1607029_at TC42536 4 Q8LP14 Nine-cis-epoxycarotenoid dioxygenase4 132E-51Abscisic acid responsive proteins



1618211_at TC47430 4 P93615 ABA induced plasma membrane protein PM 19 137E-601614372_at TC47836 4 Q6H5X2 ABA-responsive protein-like 303E-611614788_at TC45479 4 Q4VT48 Dehydrin 292E-691621592_s_at TC45479 4 Q4VT48 Dehydrin 292E-691606669_s_at - 4 Q5PXH0 Aquaporin (PIP2_1) 987E-741615808_s_at TC38121 4 Q5PXH0 Aquaporin (PIP2_1) 263E-1621610982_at TC38138 8 O24049 Major intrinsic protein C aquaporin 631E-1411614916_at TC38138 8 O24049 Major intrinsic protein C aquaporin 631E-1411612244_s_at TC38281 8 Q9M7B0 Putative aquaporin PIP2-2 240E-158BRASSINOSTEROIDSBrassinosteroid synthesis1608099_at TC40368 8 Q43147 Cyt P450 85A1 C6-oxidase Dwarf protein 289E-74Brassinosteroid receptor1612516_at TC47080 1 Q76FZ8 Brassinosteroid receptor 000E+00GIBBERELLINS1610607_at TC40495 5 Q6NMQ7 Gibberellin-responsive protein (At1g74670) 104E-281609893_at TC38517 8 O24040 LtCOR11 (Gibberellin regulated protein) 162E-361618503_at TC47765 8 Q75V70 Gibberellin 2-oxidase 1 150E-69CYTOKININS1612955_at TC47637 4 Q71BZ3 Type-A response regulator 373E-111619945_at CB345883 8 Q39636 CR9 protein cytokinin-repressed gene 312E-301620306_at TC47826 8 Q67YU0 Cytokinin oxidase 5 (cytokinin degradation) 629E-113LIGHT STIMULUSCircadian rhythm1608397_at TC43962 8 Q6LA43 Two-component response regulator-like APRR2 (TOC2 protein) 277E-341617059_at TC42407 4 Q9CAV6 WNK1 kinase 807E-261609874_at TC43754 8 Q6R3R2 CONSTANS-like protein CO1 180E-761614417_at TC47511 3 O81834 Transcription factor Constans-like family (At4g27310) 132E-321613978_at TC49120 3 Q5GA60 Putative EARLY flowering 4 protein 122E-30Flowering control1614189_at TC47473 8 Q76CC3 Flowering locus T 168E-66Transcription factors1620877_at TC40824 8 Q948G4 Putative GATA-1 zinc finger protein 114E-31Hormone and light signals1612955_at TC47637 4 Q71BZ3 Type-A response regulator 373E-111616694_at TC46350 7 Q94KS0 Histidine-containing phosphotransfer protein 196E-571614802_at TC41815 7 Q94KS0 Histidine-containing phosphotransfer protein 627E-591617513_at 8 Q8LAL2 IAA26 (Phytochrome-associated protein 1) 604E-621618875_s_at TC46557 1 Q8LAL2 IAA26 (Phytochrome-associated protein 1) 295E-631620512_at TC46080 8 Q9ZSY8 IAA27 (Phytochrome-associated protein 2) 942E-86Other light regulated genes1614237_a_at TC38178 1 Q7XAB8 Light-regulated chloroplast-localized protein 516E-231618107_at TC46244 1 Q93WJ7 Leaf Lethal Spot 1-like protein 000E+001614720_at TC46244 1 Q93WJ7 Leaf Lethal Spot 1-like protein 000E+001616605_at TC45785 1 Q8W5A3 Lethal leaf spot 1-like protein 000E+001617173_s_at 4 Q6T6I9 ELIP Early light-induced protein 388E-541610360_at TC38699 8 Q6Q9W1 Ultraviolet-B-repressible protein 247E-361620292_at TC38866 8 Q6Q9W1 Ultraviolet-B-repressible protein 200E-33TRANSCRIPTION FACTORSMADS-BOX TF1621827_at TC39767 1 Q6UGQ6 MADS-box protein 15 195E-471621836_at TC45777 7 Q8LLQ9 MADS-box protein 5 625E-1191607973_at TC38620 8 Q9ZTV9 MADS1-like 487E-1071613748_at TC40277 8 Q8LLR2 Vitis MADS-box protein 2 Ap1-like MADS-RIN-like 173E-77MYB TF1611920_at TC40303 1 Q9M9A3 F27J1520 (Hypothetical protein) (MYB transcription factor) 383E-841618260_s_at TC42111 3 Q8L5N7 Myb-related transcription factor VlMYBB1-2 similar to 447E-431620959_s_at TC48485 4 Q6L973 Myb-related transcription factor VvMYBA1 204E-1191609612_at TC47332 7 Q688D6 Putative myb transcription factor (MYB16 protein) 317E-211612686_at TC48806 8 Q5NDD2 Putative MYB transcription factor 603E-841613486_at TC45686 8 O04544 F20P526 protein (At1g70000) (MYB transcription factor) 160E-741614932_at TC48806 8 Q5NDD2 Putative MYB transcription factor 603E-841618514_at TC51437 8 O22059 CPC (Putative MYB family transcription factor) 470E-11

Table 1 Selection of genes putatively involved in the regulation of berry ripening (Continued)



scription factors of the Ethylene Responsive Factors(ERFs) and AP2EREBP families The transcript homolo-gous to tomato LeERF1 and a putative AP2EREBP tran-scription factor were induced during berry ripening whilethe homologous of ERF3b and another putative AP2EREBP transcription factor were repressed Consideringthat in tomato tobacco and Arabidopsis ERF1 genes areinduced by ethylene and wounding and positively regu-late downstream genes involved in biotic and abioticstresses [30] the induced transcription factors that wehave identified could be involved in ethylene mediatedresponses to plant stresses in grape as well [3132] TheVitis homologues to osmotin PR-1 and PR-4 down-stream genes known to be regulated by ERFs in the speciesabove mentioned were indeed found to be induced inour study suggesting that they could be under the sameregulatory mechanism The homologous of tobacco nitri-lase NtNIT4B [33] another gene related to the transcrip-tional control of ethylene responsive defence genes wasfound to be induced during ripening in our analysis aswell We could not thus exclude that an additional role ofethylene during grape berry development is related tobiotic stress response

Sequence annotation revealed the modulation of sometranscripts homologous to the tomato genes ER6 andER33 whose functions are still unknown but are reportedto be induced by ethylene during tomato ripening [34] Inour experiment however these genes did not show strongmodulation around veacuteraison and they were all repressedduring ripening On the other hand a transcript homolo-gous to tomato E8 was found induced in grape in agree-ment with tomato and strawberry gene expression data[3536] Although E8 resembles ACC synthase structure itapparently exerts a negative control on ethylene biosyn-thesis [35] and its expression seems affected by the home-otic transcription factor LeMADS-RIN [37] E8 similarmodulation in both climacteric and non-climacteric fruitsraises the possibility of a common regulatory mechanismfor ethylene homeostasis control in fleshy fruit ripening

The phytohormone abscisic acid (ABA) regulates proc-esses such as floral transition embryo maturation seeddevelopment and tolerance to abiotic and biotic stress[38] In grape berries ABA begins to accumulate at veacuterais-on and is thus considered a good candidate for triggeringberry ripening [18] The present analysis found three ABA-

1619201_at TC49224 8 Q8GV05 TRIPTYCHON (MYB transcription factor) 293E-29NAC TF1621448_at TC47141 2 Q52QR1 NAC domain protein NAC5 139E-271609172_at TC39120 3 Q52QR2 NAC domain protein NAC4 110E-1261606678_at TC42489 4 Q8LKN7 Nam-like protein 17 (Fragment) 653E-191607120_at TC46243 4 Q84K00 NAC-domain containing protein 78 (ANAC078) 143E-341621255_at TC41700 4 Q8LRL4 Nam-like protein 11 185E-46WRKY TF1609130_at TC46341 3 Q9FGZ4 WRKY DNA-binding protein 48 162E-561622333_at TC40428 3 O22900 WRKY DNA-binding protein 23 568E-461611285_s_at CA809190 7 Q6IEL3 WRKY transcription factor 68 184E-39Homeotic TF1616863_at TC41982 1 Q8LLD9 BEL1-related homeotic protein 29 (Fragment) 120E-781618774_at TC40997 1 Q4VPE9 Lateral organ boundaries-like 1 (Fragment) 651E-521611583_at TC40246 3 Q9FJ90 Similarity to AP2 domain transcription factor 285E-551609295_at TC47034 3 P46897 Homeobox-leucine zipper protein ATHB-7 241E-531620170_at TC42225 7 Q9XHC9 APETALA2 protein homolog HAP2 404E-461607122_at TC51295 8 Q546G6 Homeodomain-leucine zipper protein HAT22 482E-151607284_at TC40589 8 Q7Y0Z9 Bell-like homeodomain protein 3 (Fragment) 312E-1141617931_at TC40185 8 Q9M276 Homeobox-leucine zipper protein ATHB-12 441E-341615625_at TC39581 8 O04136 Homeobox protein knotted-1 like 3 (KNAP3) 659E-861613036_at TC39310 4 Q56R05 Putative pollen specific LIM domain-containing protein 268E-901606561_at TC41613 7 O81384 Putative basic helix-loop-helix DNA binding protein TCP2 125E-201611070_at TC49539 7 Q8L8A5 GRF1-interacting factor 1 (Hypothetical protein At5g28640) 797E-431606591_at TC41789 8 Q8RU28 Putative SHORT-ROOT (SHR) protein (Fragment) 128E-381610187_a_at TC47632 8 Q9FL03 SCARECROW gene regulator 111E-1101619334_at TC47632 8 Q9FL03 SCARECROW gene regulator 111E-1101612362_at TC41970 8 Q6SS00 YABBY-like transcription factor GRAMINIFOLIA 306E-691613445_at TC40338 8 Q6XX20 Mutant cincinnata 364E-161614851_s_at TC40029 8 Q9SIV3 Expressed protein (GPRI1) (Golden2-like protein 1) 210E-391617877_at TC41303 8 Q8SBC9 Transcription factor LIM 171E-92

(a) cluster number as reported in Figure 4(b) Uniprot ID description and E-value of the annotation results obtained by blast and GORetriever analysis




related transcripts induced after veacuteraison one is homolo-gous to the pea nine-cis-epoxycarotenoid dioxygenase 4and thus putatively involved in ABA synthesis the othertwo are homologous to two ABA responsive proteinsRecently a Vitis dehydrin DHN1 [39] and some Arabi-dopsis plasma membrane aquaporins (PIPs) [40] havebeen reported to be regulated by ABA and involved instress and senescence respectively Interestingly in ouranalysis we found VvDHN1 and numerous VvPIP iso-forms modulated in a ripening specific way (clusters 4 and8)

Brassinosteroids accumulate during fruit developmentand seem to play a key role in determining the onset ofripening in fleshy fruits [2341] Three Vitis genes involvedin brassinosteroid biosynthesis and sensing have beenrecently cloned and transcriptionally characterized duringberry development [23] Our analysis identifiedVvBR6OX1 which converts 6-deoxocastasterone tocastasterone the only bioactive brassinosteroid detectedin grape [23] and VvBRI1 a brassinosteroid receptorTheir expression profiles were characterized by a peak ofinduction at TP B in agreement with previous data [23]

Light stimulusWe identified numerous transcripts putatively involved inthe circadian rhythm oscillatory system In particular onemember of the His-to-Asp two-component signal trans-duction family homologous to Arabidopsis APRR2 andone transcription factor of the Constans-like familyshowed a decreasing profile after veacuteraison while a kinasehomologous to WNK1 [42] and two transcription factorsbelonging to the Constans-like and Early flowering familyrespectively [43] displayed a positive regulation moremarked after veacuteraison A transcript homologous to Flow-ering locus T putatively coordinating the circadianrhythm with the flowering switch [44] appearedexpressed until TP B thus suggesting its involvement insignalling circuits during early berry development

Some transcripts putatively involved in light and hormo-nal signalling cross-talk were also identified IAA26phy-tochrome associated protein 1 and IAA27phytochromeassociated protein 2 involved in light and auxin signaltransduction were down regulated after veacuteraison Mem-bers of the His-to-Asp family such as Type A response reg-ulators pseudo response regulators and histidine-containing phosphotransfer proteins are known to medi-ate the responses to light osmotic cytokinin and ethylenestimuli (reviewed in [45]) and were found modulatedduring the process of berry development

Transcription factorsIn our study we identified 125 transcription factorsamong which members of the myb MADS-box NAC and

WRKY families and some homeotic and development-specific genes

Within the myb family we found two previously isolatedVitis genes VvmybA1 which showed a ripening-specificprofile highly correlated to those of UDP-glucoseflavo-noid 3-O-glucosyltransferase glutathione-S-transferase 4and the caffeoyl-CoA O-methyltransferase as described in[46] and a VlmybB-like gene which was induced duringberry development as reported for Vitis labruscana [47]The recently characterized grape Vvmyb5A proposed tobe involved in the regulation of the phenylpropanoidpathway [48] was found to be expressed invariantly atlow levels in all years during this study interval(1621471_s_at) thus not supporting a ripening-specificfunction for this gene

Concerning the MADS-box family of transcription factorsonly one previously characterized grape gene was presentin the regulated set namely VvMADS5 which was highlyinduced during early berry development in agreementwith [49] Both VvMADS1 (1619742_at) and VvMADS4(1614965_at) which were previously reported as abun-dant transcription factors involved in berry development[4950] were highly expressed at all time-points in ourexperiment and thus not included in the regulated set

1613748_at could represent the homologue of tomatoLeMADS-rin (E-value = 82e-50 68 identity) which ispart of the developmental signalling system that initiatesripening in this fruit [51] However while LeMADS-rin isinduced from breaker stage to ripeness its putative grapehomologue was slightly induced until veacuteraison and thenshowed a decreasing profile This behaviour could eithersuggest that the grape transcript is not the homologue ofLeMADS-rin or that its role is different in non-climactericfruits

Five transcripts with homology to NAC transcription fac-tors appeared modulated all in a positive way in the studyinterval This is of interest as recent transcriptome studiesreport the regulation of NAC transcription factors familynot only during biotic and abiotic stress responses butalso in many other processes such as fruit developmentand ABA signalling (reviewed in [52])

Three transcripts homologous to the Arabidopsis WRKYfamily were also modulated In plants the WRKY familyis quite large (74 members in Arabidopsis and 90 in rice)and its members participate in numerous cellular proc-esses such as defence and hormonal signalling develop-mental programs and fruit maturation (reviewed in [53])

Nine putative homeotic genes previously identified in flo-ral development studies and 11 transcripts involved in



other morphological andor developmental processeswere modulated in our study although their connection tofruit development has not yet been reported and will needfurther investigation 1606591_at and 1619334_at corre-spond to two transcriptional regulators SHORT ROOTand SCARECROW which in Arabidopsis are both impli-cated in developmental root layers specification [54] Asthey were co-expressed in our analysis they might interactin berry as well

Though not present in the regulated gene set1616455_s_at corresponding to the grape VvMSA tran-scription factor ought to be mentioned VvMSA is inducedby sugar and ABA and positively regulates VvHT1 expres-sion [5556] In our study 1616455_s_at was very highlyexpressed at all three time-points in agreement with pre-vious proteomic and gene expression studies [75557]

Genes responsible for berry phenotypic traits and stress responseBerry development involves major transcriptionalchanges of genes directly responsible for the biochemicaland phenotypical characteristics of each growth stage Sev-eral protein families linked to these phenotypic traits havebeen expansively described in past studies focussed onsingle or small gene sets [58] and recently in a genome-wide study [11] The present work provides an extensivedescription of gene expression during ripening reportingfor the first time the occurrence of an oxidative burstimmediately after veacuteraison The complete list of the genesdiscussed in the text is in part B of Additional file 7ordered by GO functional category for clarity they arefurther divided according to more specific GO metabo-lisms when possible

Cell wall metabolismThe previously described involvement of specific genefamilies in berry growth and softening [59] was confirmedin the present study Among these there were pectin mod-ifying enzymes (methylesterase invertasepectin methyl-esterase inhibitor pectate lyase pectinesterase) extensinsand expansins grape ripening-induced proteins (GRIP 313 15 22 and 28) and xyloglucan endotransglycosylases(XETs) [859] In particular 10 members of the XET familywere modulated and four of these very strongly during rip-ening (1613415_at 1615809_at 1619355_at1621251_s_at cluster 4) These induced isoforms arehomologous to the tomato SIXTH5 6 and 8 XET iso-forms recently defined as Group 3 XETs [60] Althoughfrom the analysis of the tomato fruit cDNA libraries onlySIXTH5 was found expressed [60] our array-based geneexpression analysis suggests that all the three isoforms ofgroup 3 could be involved in fruit development sharing acommon function Vitis XET isoforms have been proposed

to be implied in brassinosteroids mediated berry cell wallsoftening [23]

Stress responseThe present analysis identified 270 transcripts related toplant stress perception and response which fall into theresponse to stimulus GO category

Sixty-nine transcripts were involved in plant biotic stressresponse and they were mostly accumulating during rip-ening The induction of defence genes including GRIPs(GRIP 22 28 31 32 51 61) thaumatins lipid transferproteins chitinases γ-thionins and pathogenesis relatedproteins during berry ripening confirmed previous results[8]

A particular kind of stress response which has beenreported during fruit development in several species suchas tomato [61] strawberry [20] avocado [62] pineapple[63] and pear [64] is the response to oxidative stress Inrecent years a dual role for reactive oxygen species (ROS)in plants has been identified toxic by-products of the aer-obic metabolism and important regulators of growthdevelopment and defence reactions A complex networkof genes finely tuning ROS concentration as the result oftheir production and scavenging has been documented[65]

The occurrence of oxidative stress during grape berrydevelopment has been rather controversial catalase wasidentified among the abundant proteins in the ripe berry[57] but at the transcriptional level the typical oxidativestress markers seemed absent or negatively regulated [10]The present study clearly indicated that oxidative proc-esses take place also during berry ripening and that anenzyme-mediated scavenging system is activated BerryH2O2 content increased significantly in correspondence ofthe veacuteraison stage reaching its maximum one-two weeksafter and then decreasing at a slower pace towards harvesttime (Fig 6) At least 32 transcripts of the modulated coreset appeared involved in the enzymatic detoxificationfrom H2O2 and the other ROS species which accumulatein the cell starting from veacuteraison (Fig 7) Some of themsuch as ascorbate peroxidase glutathione peroxidase andperoxiredoxins catalyze the direct reduction of hydrogenperoxide to water the others (thioredoxins glutaredoxinsglutathione-S-transferases and metallothioneins) regulatethe balance of the oxidized and reduced forms of the anti-oxidants ascorbic acid and glutathione Although the reg-ulation of the detoxifying enzymes must be very complexisozyme specific and occurring at different levels (tran-scriptional- and post-transcriptional regulation subcellu-lar compartimentalization etc) as previously shown forsome of them [61] most of the enzymes shown in Figure7 appeared to be unchanged from TP A to TP B and be



strongly modulated from TP B to TP C Further studies areneeded to address how ROS molecules are taking part tothese regulatory circuits and what is the main reason ofROS accumulation at veacuteraison The superoxide dismutase(SOD) and catalase (CAT) transcripts which also codifyfor two important detoxifying enzymes were not signifi-cantly modulated in our study The CAT gene(1618705_s_at) was highly expressed in all the three TPsthe SOD encoding transcripts (5 probes) showed very dif-ferent expression levels but no one fell into the modulatedcore set

The Nudix gene family may also be involved in the oxida-tive stress response as three transcripts were found mod-ulated (one in cluster 3 and two in cluster 8) InArabidopsis Nudix hydrolases appear to be involved in thesanitization of the oxidized nucleotide pool accumulatingduring oxidative stress [66]

Ten genes involved in programmed cell death (PCD)appeared to be modulated In particular three transcriptswith putative function of PCD suppression are grouped incluster 1 while a putative death associated protein isinduced after veacuteraison (cluster 4) In addition listed in theprotein metabolism category there are two cysteine pro-teases (cluster 4) and a cystatin-like gene (cluster 5) Ourdata are insufficient to draw any conclusion however theymay suggest the occurrence of a PCD process in grapeberry during ripening as highlighted for strawberry [20]

Sugaracid ratio and water balanceOur results support the model of sugar phloem unloadingvia an apoplastic pathway by means of plasma membranesucrose and hexoses transporters as proposed by [67] andrecently confirmed by [6869] Four sugar transporterswere found modulated the hexose transporter VvHT6(1615257_at and 1619691_at) was positively modulatedduring ripening (cluster 3) as previously described [10]interestingly a transcript homologous to a sugar trans-porter isolated in Citrus fruit was also induced while theforth transcript which is homologous to a sugar trans-porter from tomato root is repressed The sucrose trans-porter VvSUC27 appeared dramatically down-regulatedafter veacuteraison as already reported in the cv Syrah [70]while VvSUC11 and VvSUC12 transporters with a ripen-ing specific expression profile [70] were highly expressedat all time points in our experiment

There is evidence that during berry ripening sucrose ishydrolysed in the cell wall and then re-synthesised in thecytoplasm following hexose uptake [69] Our data corrob-orate this mechanism as three sucrose-phosphate syn-thases (cluster 3 and 4) a sucrose synthase (cluster 4) asucrose-6-phosphate phosphatase (cluster 4) a UDP-sugar pyrophosphorylase (cluster 2) and a beta-phos-phoglucomutase (cluster 4) were all up-regulated duringripening Vacuolar invertases VvGIN1 and VvGIN2 bothresulted highly expressed until veacuteraison (cluster 8) as pre-viously described [5771] The existence of futile cycles ofsucrose synthesis and breakdown has been demonstratedin other fruits such as tomato where the relative rate ofsynthesis and breakdown cycles controls the rate of sugarand starch accumulation [72]

The early expression of a phosphoenolpyruvate carboxy-lase isoforms (cluster 8) well correlated with the rate ofmalate accumulation during the first phase of berry devel-opment Furthermore a vacuolar pyrophosphatase(1614834_at) putatively involved in malate vacuolaruptake was expressed early and then induced during rip-ening (cluster 5) as previously observed [73] At veacuteraisonwhen malate decompartimentalization and breakdowntake place [67] induction of malate degrading enzymessuch as malic enzyme and phosphoenolpyruvate carbox-ykinase (cluster 1) and mitochondrial malate dehydroge-nase (cluster 8) was observed The profiles observed forgrape alcohol dehydrogenase 2 and a putative short chainalcohol dehydrogenase (cluster 4 and 5 respectively) andthree isoforms of aldehyde dehydrogenase (one in cluster4 and two in cluster 8) may be indicative of a shift to anaerobic fermentative metabolism during ripening [74]The alpha subunit of the pyruvate dehydrogenase com-plex was induced after veacuteraison (cluster 5) however thisenzyme is known to be highly regulated post-translation-ally allosterically by NADH and covalently by reversible

H2O2 content during berry ripening in 2005 (blue) and 2006 (green) samplesFigure 6H2O2 content during berry ripening in 2005 (blue) and 2006 (green) samples Data are means plusmn SE of three replicates The weeks post flowering (wpf) corresponding to the samples used for the hybridization experiments are enclosed in boxes Veacuteraison time is delimited within dotted lines



phosphorylation Our analysis identified two mitochon-drial pyruvate dehydrogenase kinase isoforms one ofwhich induced in a ripening-specific way Two enzymes ofthe glycolysisgluconeogenesis pathways such as enolaseand glyceraldehyde-3-phosphate dehydrogenase (cluster

5 and 4 respectively) were induced during ripening sug-gesting their involvement into the acidsugar interconver-sion andor the fermentative metabolism

Genes involved in the response to oxidative stress modulated during ripeningFigure 7Genes involved in the response to oxidative stress modulated during ripening List of the ROS-scavenging genes found in the modulated core set and comparative analysis of their expression in the three sampling TPs Log2 ratios of the expression values observed in TPs A B and C (BA and CB) are visualized by a colour scale where red indicates an increase and green a decrease For each functional group the corresponding enzymatic reaction is reported refer to probesets prob-ably corresponding to the same transcript



About 90 of ripe berries fresh weight is comprised ofwater whose transport into the berry vacuole is mediatedby aquaporins In our study we detected 13 transcriptscoding for aquaporin isoforms eight belong to theplasma membrane intrinsic protein family four to thetonoplast intrinsic protein family and one to the NOD26-like intrinsic protein family All isoforms were highlyexpressed in the pre-veacuteraison phase and three of themcoding for the putative aquaporin PIP2-1 were alsoinduced during ripening A similar modulation has beenrecently observed in Citrus although the induced iso-forms during fruit development belong to the TIP family[16]

Secondary metabolismThe secondary metabolism GO functional categoryincluded genes involved in the synthesis of aromatic andvolatile compounds such as terpenes and benzenoidcompounds antioxidant compounds such as polyphe-nols and vitamin E and genes of the phenylpropanoidpathway Due to its importance for berry quality traits werestricted our analysis to the latter pathway Most struc-tural genes of this branched pathway which starts withphenylalanine ammonia lyase were present among theregulated genes of the current study During ripening thegenes coding for the enzymes acting in flavonols synthesis(flavonol synthase cluster 3) stilbenes synthesis (threeisoforms of stilbene synthase cluster 4) and anthocyaninssynthesis (UDP-glucoseflavonoid 3-O-glucosyltrans-ferase cluster 4) were induced while those involved intannins synthesis (anthocyanidin reductase and leucoan-thocyanidin reductase) were repressed The profiles ofgenes coding for phenylalanine ammonia lyase chalconeisomerase 1 and flavonoid 35-hydroxylase showed adecrease until veacuteraison followed by an induction duringripening confirming previous data [75] Two isoforms of4-coumarate-CoA ligase with opposite profiles weredetected (clusters 3 and 8) and their pattern of expressionis consistent with an involvement into the anthocyaninand lignin synthesis pathways respectively

Eleven members of the ATP-binding cassette (ABC) andmultidrug and toxic compound extrusion (MATE) trans-porters families which are involved in several cellulartransport processes (reviewed in [76]) were modulatedAmong them the three transporters positively modulatedduring berry development (1610275_at 1607763_at1616929_at) represent possible candidates for secondarymetabolites transport

Gene expression profiles highly influenced by seasonal variationThe influence on gene expression by environmental con-ditions is well reported in literature [77] and was evidentin our PCA analysis (Fig 2) The comparison of the three

sampling years highlighted a large number of genes mod-ulated only during one or two seasons and thus excludedby the core set of ripening-specific genes 938 genes in2003 2530 in 2005 and 2143 in 2006 Two examples arethe isoform 2 of the phenylpropanoid biosynthetic geneschalcone synthase and chalcone isomerase whose expres-sion profiles are depicted in Fig 8E and 8F such discrep-ancy in their modulation is indeed expected for non-ripening specific isoforms Although these genes are prob-ably highly affected by changes in the environment theirGO functional categories distribution was not signifi-cantly different to that of the core set genes (data notshown) This result suggests that on a long time framethe plant reacts to the seasonal variations by adjusting thewhole metabolism and not just a part of it to maintainhomeostasis

The group of genes modulated in all three seasons withdifferent transcriptional profiles deserves in our opiniona special discussion as they are likely involved into the rip-ening process but in the meantime highly influenced bythe climatic conditions These 223 genes represent about13 of the modulated gene set (1700 transcripts) Amongthem at least four transcripts are putatively involved inlight perception and signalling cryptochrome 1(1615199_at) and pseudo-response regulator 9 (APRR91616872_at) showed opposite behaviours in 2003 vs2005 and 2006 (Fig 8A and 8B) pseudo-response-regula-tor 7 (APRR7 1608006_at) displayed a V profile in2003 and 2005 and the opposite pattern during 2006finally LONG HYPOCOTYL 5 (HY5 1609930_at) showeda different profile from TP A to TP B in 2003 and 2005 vs2006 while it was repressed in all three seasons from TP Bto TP C Interestingly APRR7 and APRR9 seem involvedin the temperature responsiveness of the Arabidopsis circa-dian clock synchronizing photo- and thermo-cycles [78]and HY5 is a positive regulator of light signalling whichacts downstream of several photoreceptors and mediatesthe response to different types of light (redfar red blueand UV-B [7980])

Out of the three previously characterized grape hexosetransporters present on the chip two (Vvht1 and Vvht2)seemed influenced by the season Vvht1 (1616083_at)showed a V profile in all the three years but with differ-ent slopes suggesting a strong environmental condition-ing along with the reported cultivar specificity [108182]Unlike Vvht1 Vvht2 is considered a ripening specific iso-form [108283] however the present work highlighted acertain degree of variability in its induction time in thethree years (1614764_at Fig 8C) A similar behaviourwas observed for another ripening specific gene namelythe isoform 3 of the chalcone synthase family [7584](Fig 8D)




Real time RT-PCR validation of the expression profiles of six genes highly influenced by the seasonFigure 8Real time RT-PCR validation of the expression profiles of six genes highly influenced by the season A 1615199_at Cryptochrome 1 B 1616872_at Pseudo response regulator 9 C 1614764_at VvHT2 hexose transporter 2 D 1606663_at VvChs3 chalcone synthase isoform 3 E 1607732_at VvChs2 chalcone synthase isoform 2 F 1620424_at VvChi2 chalcone isomerase isofom 2 2003 profiles are represented in red 2005 in blue and 2006 in green Relative expres-sion data of the array hybridization experiments were centered on the average of the log2 values of each season Expression profiles measured by RT-PCR experiments (insets) were first centered on the mean Ct value calculated on the three seasons for each gene and then log2 transformed RT-PCR data are reported as means plusmn SE of three technical replicates


To ensure that the observed differences among seasonsand among gene isoforms were not technical artefacts theexpression profile of six genes was validated by real timeRT-PCR (Fig 8 insets) In all but one case (Vvht2 year2003) there was complete agreement with the array data

ConclusionThe time-course gene expression analysis of Pinot Noirberry development based on a systematic and genome-wide approach has identified the occurrence of two welldistinct phases along the process The pre-veacuteraison phaserepresents a reprogramming stage of the cellular metabo-lism characterized by the expression of numerous genesinvolved in hormonal signalling and transcriptional regu-lation The post-veacuteraison phase is characterized by theonset of a ripening-specialized metabolism responsiblefor the phenotypic traits of the ripe berry A similar switchin fruit metabolism has been recently reported also in Cit-rus [16] Five functional categories are overrepresented inthis ripening-specialized metabolism cell wall carbohy-drate and secondary metabolisms stress response andphotosynthesis the latter being obviously switched offAn oxidative burst previously not detected in grapevineand characterized by a rapid accumulation of H2O2 start-ing from veacuteraison was demonstrated A concurrent mod-ulation of the enzymatic ROS scavenging network wasalso highlighted suggesting a possible role of ROS speciesin the regulation of berry ripening We have found a verylarge number of transcription factors and transcriptsrelated to hormonal metabolism and signal transductionthat appear to be regulated during berry developmentThese represent 19 of the modulated gene set corrobo-rating the widely-accepted hypothesis that fruit develop-ment is under tight transcriptional control These dataprovide the basis for subsequent correlation analysisaimed at highlighting gene co-regulation and metabolicnetworks [85] Our strategy to perform the experimentduring three seasons allowed us to restrict the dataset ofmodulated genes to those more likely to be ripening-spe-cific Nonetheless we were able to separate ripening spe-cific isoforms within gene families and to identifyripening related genes which appeared strongly regulatedalso by the seasonal weather conditions

In perspective the wealth of information we present herewill provide a new extended platform to study the com-plex process of grapevine berry development and more ingeneral of non-climacteric fruits

MethodsPlant Material and Experimental DesignTen grape clusters of Vitis vinifera cv Pinot Noir were col-lected weekly between 8 am and 9 am from flowering toover-ripening during 2003 2005 and 2006 seasons at theIASMA study site ai Molini (San Michele allAdige-TN

Italy) To be more representative of the canopy five clus-ters were collected from the north side (shady) and theother five from the south side (sunny) Berries were rap-idly detached with their petioles and divided randomlyinto groups of 50 that were either immediately frozen forsubsequent transcriptional analysis or subjected to bio-chemical analyses including must acid and sugar contentaverage berry weight determination and anthocyanin con-centration [68] Based on these parameters the peak ofacidity just before veacuteraison was estimated and two addi-tional time-points (two weeks before and three weeksafter veacuteraison) were chosen for gene expression analysisThe three TPs correspond to the 33 34 and 36 of the mod-ified E-L system [12] For each time-point three biologicalreplicates were obtained by performing an independentRNA extraction from three sub-pools of 8 berries derivedfrom the group of 50 berries Consequently nine arrayswere utilized for each season

RNA Preparation Array Hybridization Data AnalysisTotal RNA was extracted from frozen deseeded berriesaccording to [86] and further purified using RNeasy spincolumns (Qiagen) RNA quality and quantity wereassessed by gel-electrophoresis and by absorbance meas-urements Ten μg of each sample were sent to the IFOM(Milan-I) Affymetrix platform facility for probe synthesisGeneChipreg Vitis arrays hybridization washing stainingand scanning with GeneChip Scanner 3000 according tothe Affymetrix GeneChipreg Expression Analysis TechnicalManual Data referred to 2003 2005 and 2006 were ana-lyzed independently in order to obtain one set of differen-tially expressed genes for each season Bioconductorpackages [87] in R v210 [88] were used for the statisticalanalyses Data were pre-processed using the GCRMAmethod [89] which performs background correctionquantile normalization and summarization and returnslog2 converted expression values Biological replicatesquality was assessed by means of R-squared coefficientwhich represents the fraction of variance explained by alinear model For variation assessment of the gene expres-sion values between all 27 data sets data from the threeseasons were scaled and Principal Components Analysis(PCA) was performed In order to identify genes modu-lated during berry development a filtering step based onthe inter-quantile range method (IQR = 025) was usedThis restricted dataset enriched in differentially expressedgenes was suited to run a multiclass comparison methodof Significance Analysis of Microarrays (SAM) with falsediscovery rate =015 [13] SAM output was furtherrestricted to genes with fold-change greater than 2 in atleast one of the two comparisons (TP A vs TP B and TP Bvs TP C) The final datasets relative to the three seasonswere intersected obtaining a dataset of 1700 commonmodulated genes Pearsons correlation coefficientsbetween genes profiles relative each pair of seasons were



then calculated and a threshold of 05 was fixed to obtaina berry ripening-modulated core dataset of 1477 genesExpression profiles clustering was carried out by the k-means method (k = 12) with Pearsons correlation dis-tance using the T-MeV software [90] The number of clus-ters was manually reduced to eight which represents theminimum number of expression profiles consideringthree time-points

Affymetrix Vitis ArrayThe GeneChipreg Vitis vinifera genome array (Affymetrix)consists of 16436 probesets 14496 derived from V vin-ifera transcripts and 1940 derived from other Vitis speciesor hybrids transcripts It can interrogate 12908 GenBankaccessions of V vinifera and 1547 of other Vitis species orhybrids Sequences used in the design of the Vitis Gene-Chipreg were selected from GenBank dbEST and RefSeqThe sequence clusters were created from the UniGenedatabase (Build 7 Oct 2003) V vinifera sequences repre-sented on the chip correspond to 10042 TIGR TentativeConsensus and 1940 Singletons (Release 4 Sept 2004)while 102 GenBank accessions are not present in the TIGRdatabase Overall chip redundancy is estimated to be166

AnnotationThe consensus sequence of each Vitis GeneChipreg probeset(available at [91] was analyzed to identify the codingsequence (cds) by means of an in-house developedmethod which combines the results of two cds predictorsEstscan [92] and FrameFinder (at ESTate [93]) 13923cds were obtainded and used to interrogate the Uniprotdatabase (Sept 05) by means of blastp while theremaining 2513 sequences were used to interrogate thesame database by means of blastx [94] Blast results (E-value lte-10) with GO associated terms ([14]) were ana-lyzed by the in-house developed program GORetrieverfor annotation and 12242 sequences were annotatedwith high confidence (gt 80) The automatic annotationresults for the 1700 modulated genes were manuallyinspected and integrated with GO biological processterms supported by literature evidences

Genes were grouped into 17 functional categories basedon GO biological process terms by means of the in-housedeveloped program GOSlimer [see Additional file 4 forthe list of the GO categories and Additional file 3 for genesannotation] Functional categories distribution in themodulated and chip sequences were compared by meansof Chi square and Fisher statistical tests (p-value lt 0001)

Determination of the H2O2 contentH2O2 content was measured by a fluorimetric assay basedon the substrate 10-acetyl-37-dihydroxyphenoxazinewhich is oxidized in the presence of H2O2 and peroxidase

using the Amplex Red Hydrogen PeroxidePeroxidaseAssay (Molecular Probes Eugene USA) following themanufacturers instructions Ten frozen berries for eachtime point considered were ground in liquid nitrogen and02 g of fine powder were then solubilized for each repli-cate with 05 ml of 50 mM phosphate buffer (pH 74) andkept 5 min on ice After centrifugation at 20000 g for 15min the cleared supernatant was extracted with an equalvolume of 21 (vv) chloroformmethanol mixture andcentrifuged at 12000 g for 5 min To perform the H2O2assay 50 μl of the aqueous phase were added to 50 μl ofthe reagent working solution After 30 min of incubationat 25degC relative fluorescence (excitation at 544 nm andemission at 590 nm) was measured in a black 96 wellsplate Absolute quantification was determined by the useof a H2O2 standard curve

Real time RT-PCRFirst strand cDNA synthesis was performed on one mRNAsample of the three biological replicates used for themicroarray experiments using the SuperScripttrade III ReverseTranscriptase kit (Invitrogen) according to the manufac-turer instructions Primers [see Additional file 8 forsequences] and cDNA were mixed with the Platinumreg

SYBRreg Green qPCR SuperMix-UDG (Invitrogen) and thereaction was carried out on an ABI PRISM 7000 SequenceDetection System (Applied Biosystems) Cycling condi-tions were 50degC for 2 minutes 95degC for 2 minutes then40 cycles of 95degC for 15 seconds and 60degC for 1 minuteRaw data were analyzed with ABI PRISM 7000 SDS soft-ware to extract Ct values and with the LinReg software tocalculate the reaction efficiency [95] Relative expressionof each gene (target) was then calculated according to theequation by [96] using actin for normalization (reference)and centered on the mean Ct calculated on the three sea-sons and on the three time points (control) RelExp = E(tar-

get)exp[ΔCt(target)(sample-control)]E(reference)exp[ΔCt(reference)(sample-control)]

Data AvailabilityAll microarray expression data produced by this work areavailable at EBI ArrayExpress [97] under the series entry E-MEXP-1282

AbbreviationscDNA Complementary DNA

EST Expressed Sequence Tag

GO Gene Ontology

NCBI National Center for Biotechnology Information

SAM Significance Analysis of Microarrays



TIGR The Institute for Genomic Research

TAIR The Arabidopsis Information Resource

Authors contributionsSP has made substantial contribution to conception dataanalysis and manuscript drafting MP participated in theGO annotation of the sequences performed profiles clus-tering and revised critically the manuscript AM has con-tributed to data analysis and revised critically themanuscript AC LD and PF contributed to sequence anno-tation ADR carried out samples collection RNA isolationthe biochemical assays and the real time PCR experi-ments RV contributed to critically revise the manuscriptRV participated to projects design and coordination CMhas made substantial contribution to conception projectcoordination and helped in drafting the manuscript Allauthors read and approved the final manuscript

Additional material

AcknowledgementsWe wish to thank Dr Davide Marchignoli for informatic support and stim-ulating discussion Dr Massimo Bertamini Dr Roberto Zorer for physio-logical data on berry samples and seasonal climate data (IASMA Research Center) Dr Mauro Dallaserra (FBK amp CNR-Istituto di Biofisica Trento-Italy) for H2O2 measurements Dr Etti Or (Volcani Center Israel) and Dr David Neale (University of California Davis ndash USA) for critical reading of the manuscript and Dr Enrico Blanzieri for help with data analysis

This work was supported in part by grants from the Fondo Unico of the Provincia Autonoma di Trento post doc project Profiles (SP) and from the Advanced Biology Project funded by the Fondazione delle Casse di Ris-parmio di Trento e Rovereto

References1 Doligez A Adam-Blondon AF Cipriani G Di Gaspero G Laucou V

Merdinoglu D Meredith CP Riaz S Roux C This P An integratedSSR map of grapevine based on five mapping populationsTheor Appl Genet 2006 113369-382

2 IASMA Genomics [httpgenomicsresearchiasmait]3 NCBI dbest [httpwwwncbinlmnihgovdbEST]4 Coombe BG Mccarthy MG Dynamics of Grape Berry Growth

and Physiology of Ripening Aust J Grape Wine Res 20006131-135

Additional file 1Winkler index Winkler index of the study site calculated for the seasons 2002ndash2006Click here for file[httpwwwbiomedcentralcomcontentsupplementary1471-2164-8-428-S1tiff]

Additional file 2Data quality assessment R-squared coefficient analysis on the 27 hybrid-ization data setsClick here for file[httpwwwbiomedcentralcomcontentsupplementary1471-2164-8-428-S2xls]

Additional file 3Core set of modulated genes during berry ripening Log2 average expres-sion data for 2003 2005 and 2006 of the 1477 modulated genes during berry ripening Some more information such as the corresponding TC or singleton of the TIGR gene index v 40 the expression profile cluster the sequence annotation and the corresponding GO functional category are also includedClick here for file[httpwwwbiomedcentralcomcontentsupplementary1471-2164-8-428-S3xls]

Additional file 4List of the GO functional categories Gene Ontology codes with obsolete and updated descriptions of the functional categories used in this workClick here for file[httpwwwbiomedcentralcomcontentsupplementary1471-2164-8-428-S4xls]

Additional file 5Functional categories distribution in the Vitis GeneChipreg Comparison of the functional categories distribution in the Affymetrix Vitis GeneChip and in the Arabidopsis genomeClick here for file[httpwwwbiomedcentralcomcontentsupplementary1471-2164-8-428-S5tiff]

Additional file 6Functional categories distribution in the expression clusters Functional categories distribution in the eight clusters obtained by the k-means method on the gene expression profiles of the 1477 modulated genesClick here for file[httpwwwbiomedcentralcomcontentsupplementary1471-2164-8-428-S6tiff]

Additional file 7Complete list of genes belonging to the functional categories discussed in the article Genes of the core set involved in the regulation of berry devel-opment (part A) and responsible for berry phenotypic traits (part B) ordered by major GO functional categories (in red) For clarity tran-scripts discussed in the manuscript were further classified according to GO more specific termsClick here for file[httpwwwbiomedcentralcomcontentsupplementary1471-2164-8-428-S7xls]

Additional file 8List of the primers used for the RT-PCR validation experiment Sequence of the primers used in real time reverse transcription-polymerase chain reactionClick here for file[httpwwwbiomedcentralcomcontentsupplementary1471-2164-8-428-S8doc]



5 Ablett E Seaton G Scott K Shelton D Graham MW Baverstock PLee LS Henry R Analysis of grape ESTs global gene expres-sion patterns in leaf and berry Plant Sci 2000 15987-95

6 Moser C Segala C Fontana P Salakhudtinov I Gatto P Pindo MZyprian E Toepfer R Grando MS Velasco R Comparative analy-sis of expressed sequence tags from different organs of Vitisvinifera L Funct Integr Genomics 2005 5208-217

7 da Silva FG Iandolino A Al-Kayal F Bohlmann MC Cushman MA LimH Ergul A Figueroa R Kabuloglu EK Osborne C Rowe J TattersallE Leslie A Xu J Baek J Cramer GR Cushman JC Cook DR Char-acterizing the grape transcriptome Analysis of expressedsequence tags from multiple Vitis species and developmentof a compendium of gene expression during berry develop-ment Plant Physiol 2005 139574-597

8 Davies C Robinson SP Differential screening indicates a dra-matic change in mRNA profiles during grape berry ripeningCloning and characterization of cDNAs encoding putativecell wall and stress response proteins Plant Physiol 2000122803-812

9 Waters DL Holton TA Ablett EM Lee LS Henry RJ cDNA micro-array analysis of developing grape (Vitis vinifera cv Shiraz)berry skin Funct Integr Genomics 2005 540-58

10 Terrier N Glissant D Grimplet J Barrieu F Abbal P Couture CAgeorges A Atanassova R Leon C Renaudin JP Dedaldechamp FRomieu C Delrot S Hamdi S Isogene specific oligo arraysreveal multifaceted changes in gene expression during grapeberry (Vitis vinifera L) development Planta 2005 222832-847

11 Grimplet J Deluc LG Tillett RL Wheatley MD Schlauch KA CramerGR Cushman JC Tissue-specific mRNA expression profiling ingrape berry tissues BMC Genomics 2007 8187

12 Coombe BG Adoption of a system for identifying grapevinegrowth stages Aust J Grape Wine Res 1995 1104-110

13 Tusher VG Tibshirani R Chu G Significance analysis of micro-arrays applied to the ionizing radiation response Proc NatlAcad Sci USA 2001 985116-5121

14 Gene Ontology home [httpwwwgeneontologyorg]15 TAIRThe Arabidopsis Information Resource [httpwwwara

bidopsisorg]16 Cercos M Soler G Iglesias DJ Gadea J Forment J Talon M Global

analysis of gene expression during development and ripeningof citrus fruit flesh A proposed mechanism for citric acid uti-lization Plant Mol Biol 2006 62513-527

17 Alba R Payton P Fei Z McQuinn R Debbie P Martin GB TanksleySD Giovannoni JJ Transcriptome and selected metaboliteanalyses reveal multiple points of ethylene control duringtomato fruit development Plant Cell 2005 172954-2965

18 Coombe BG Hale CR The hormone content of ripening grapeberries and the effects of growth substance treatments PlantPhysiol 1973 51629-634

19 Davies C Boss PK Robinson SP Treatment of grape berries anonclimacteric fruit with a synthetic auxin retards ripeningand alters the expression of developmentally regulatedgenes Plant Physiol 1997 1151155-1161

20 Aharoni A Keizer LC Van Den Broeck HC Blanco-Portales RMunoz-Blanco J Bois G Smit P De Vos RC OConnell AP Novelinsight into vascular stress and auxin-dependent and -inde-pendent gene expression programs in strawberry a non-cli-macteric fruit Plant Physiol 2002 1291019-1031

21 Gillaspy G Ben-David H Gruissem W Fruits a developmentalperspective Plant Cell 1993 51439-1451

22 Cawthon DL Morris JR Relationship of seed number and matu-rity to berry development fruit maturation hormonalchanges and uneven berry ripening in Concord (Vitis labr-usca) grapes J Am Soc Hortic Sci 19821097-1104

23 Symons GM Davies C Shavrukov Y Dry IB Reid JB Thomas MRGrapes on steroids Brassinosteroids are involved in grapeberry ripening Plant Physiol 2006 140150-158

24 Pollmann S Neu D Weiler EW Molecular cloning and charac-terization of an amidase from Arabidopsis thaliana capable ofconverting indole-3-acetamide into the plant growth hor-mone indole-3-acetic acid Phytochemistry 2003 62293-300

25 Staswick PE Serban B Rowe M Tiryaki I Maldonado MT MaldonadoMC Suza W Characterization of an Arabidopsis enzyme fam-ily that conjugates amino acids to indole-3-acetic acid PlantCell 2005 17616-627

26 Liu K Kang BC Jiang H Moore SL Li H Watkins CB Setter TL JahnMM A GH3-like gene CcGH3 isolated from Capsicum chin-ense L fruit is regulated by auxin and ethylene Plant Mol Biol2005 58447-464

27 Hale CR Coombe BG Hawker JS Effects of ethylene and 2-chlo-roethylphosphonic acid on the ripening of grapes Plant Physiol1970 45620-623

28 Trainotti L Pavanello A Casadoro G Different ethylene recep-tors show an increased expression during the ripening ofstrawberries does such an increment imply a role for ethyl-ene in the ripening of these non-climacteric fruits J Exp Bot2005 562037-2046

29 Chervin C El-Kereamy A Roustan JP Latche A Lamon J BouzayenM Ethylene seems required for the berry development andripening in grape a non-climacteric fruit Plant Sci 20041671301-1305

30 Tournier B Sanchez-Ballesta MT Jones B Pesquet E Regad F LatcheA Pech JC Bouzayen M New members of the tomato ERF fam-ily show specific expression pattern and diverse DNA-bind-ing capacity to the GCC box element FEBS Lett 2003550149-154

31 Lorenzo O Piqueras R Sanchez-Serrano JJ Solano R ETHYLENERESPONSE FACTOR1 integrates signals from ethylene andjasmonate pathways in plant defense Plant Cell 200315165-178

32 Ohme-Takagi M Shinshi H Ethylene-inducible DNA bindingproteins that interact with an ethylene-responsive elementPlant Cell 1995 7173-182

33 Xu P Narasimhan ML Samson T Coca MA Huh GH Zhou J MartinGB Hasegawa PM Bressan RA A nitrilase-like protein interactswith GCC box DNA-binding proteins involved in ethyleneand defense responses Plant Physiol 1998 118867-874

34 Zegzouti H Jones B Frasse P Marty C Maitre B Latch A Pech JCBouzayen M Ethylene-regulated gene expression in tomatofruit characterization of novel ethylene-responsive and rip-ening-related genes isolated by differential display Plant J1999 18589-600

35 Kneissl ML Deikman J The tomato E8 gene influences ethylenebiosynthesis in fruit but not in flowers Plant Physiol 1996112537-547

36 Aharoni A OConnell AP Gene expression analysis of straw-berry achene and receptacle maturation using DNA micro-arrays J Exp Bot 2002 532073-2087

37 Giovannoni JJ Genetic regulation of fruit development and rip-ening Plant Cell 2004 16(Suppl)S170-S180

38 Zeevaart JAD Creelman RA Metabolism and physiology ofabscisic-acid Annu Rev Plant Physiol Plant Mol Biol 1988 39439-473

39 Xiao H Nassuth A Stress- and development-induced expres-sion of spliced and unspliced transcripts from two highly sim-ilar dehydrin 1 genes in V riparia and V vinifera Plant Cell Rep2006 25968-977

40 Jang JY Kim DG Kim YO Kim JS Kang H An expression analysisof a gene family encoding plasma membrane aquaporins inresponse to abiotic stresses in Arabidopsis thaliana Plant MolBiol 2004 54713-725

41 Montoya T Nomura T Yokota T Farrar K Harrison K Jones JDKaneta T Kamiya Y Szekeres M Bishop GJ Patterns of Dwarfexpression and brassinosteroid accumulation in tomatoreveal the importance of brassinosteroid synthesis duringfruit development Plant J 2005 42262-269

42 Nakamichi N Murakami-Kojima M Sato E Kishi Y Yamashino TMizuno T Compilation and characterization of a novel WNKfamily of protein kinases in Arabiodpsis thaliana with refer-ence to circadian rhythms Biosci Biotechnol Biochem 2002662429-36

43 Kikis EA Khanna R Quail PH ELF4 is a phytochrome-regulatedcomponent of a negative-feedback loop involving the centraloscillator components CCA1 and LHY Plant J 200544300-313

44 Kardailsky I Shukla VK Ahn JH Dagenais N Christensen SK NguyenJT Chory J Harrison MJ Weigel D Activation tagging of the flo-ral inducer FT Science 1999 2861962-1965

45 Hwang I Chen HC Sheen J Two-component signal transduc-tion pathways in Arabidopsis Plant Physiol 2002 129500-515

46 Ageorges A Fernandez L Vialet S Merdinoglu D Terrier N RomieuC Four specific isogenes of the anthocyanin metabolic path-



way are systematically co-expressed with the red colour ofgrape berries Plant Sci 2006 170372-383

47 Kobayashi S Ishimaru M Hiraoka K Honda C Myb-related genesof the Kyoho grape (Vitis labruscana) regulate anthocyaninbiosynthesis Planta 2002 215924-933

48 Deluc L Barrieu F Marchive C Lauvergeat V Decendit A Richard TCarde JP Merillon JM Hamdi S Characterization of a grapevineR2R3-MYB transcription factor that regulates the phenyl-propanoid pathway Plant Physiol 2006 140499-511

49 Boss PK Buckeridge EJ Poole A Thomas MR New insights intograpevine flowering Funct Plant Biol 2003 30593-606

50 Boss PK Vivier M Matsumoto S Dry IB Thomas MR A cDNAfrom grapevine (Vitis vinifera L) which shows homology toAGAMOUS and SHATTERPROOF is not only expressed inflowers but also throughout berry development Plant Mol Biol2001 45541-553

51 Vrebalov J Ruezinsky D Padmanabhan V White R Medrano DDrake R Schuch W Giovannoni J A MADS-box gene necessaryfor fruit ripening at the tomato ripening-inhibitor (rin) locusScience 2002 296343-346

52 Olsen AN Ernst HA Leggio LL Skriver K NAC transcription fac-tors structurally distinct functionally diverse Trends Plant Sci2005 1079-87

53 Ulker B Somssich IE WRKY transcription factors from DNAbinding towards biological function Curr Opin Plant Biol 20047491-498

54 Paquette AJ Benfey PN Maturation of the ground tissue of theroot is regulated by gibberellin and SCARECROW andrequires SHORT-ROOT Plant Physiol 2005 138636-640

55 Cakir B Agasse A Gaillard C Saumonneau A Delrot S AtanassovaR A grape ASR protein involved in sugar and abscisic acidsignaling Plant Cell 2003 152165-2180

56 Atanassova R Leterrier M Gaillard C Agasse A Sagot E Coutos-Thevenot P Delrot S Sugar-regulated expression of a putativehexose transport gene in grape Plant Physiol 2003 131326-334

57 Sarry JE Sommerer N Sauvage FX Bergoin A Rossignol M AlbagnacG Romieu C Grape berry biochemistry revisited upon pro-teomic analysis of the mesocarp Proteomics 2004 4201-215

58 Robinson SP Davies C Molecular biology of grape berry ripen-ing Aust J Grape Wine Res 2000 6175-188

59 Nunan KJ Davies C Robinson SP Fincher GB Expression patternsof cell wall-modifying enzymes during grape berry develop-ment Planta 2001 214257-264

60 Saladie M Rose JK Cosgrove DJ Catala C Characterization of anew xyloglucan endotransglucosylasehydrolase (XTH) fromripening tomato fruit and implications for the diverse modesof enzymic action Plant J 2006 47282-295

61 Jimenez A Creissen G Kular B Firmin J Robinson S Verhoeyen MMullineaux P Changes in oxidative processes and componentsof the antioxidant system during tomato fruit ripening Planta2002 214751-758

62 Meir S Philosoph-Hadas S Zauberman G Fuchs Y Akerman M Aha-roni N Increased formation of fluorescent lipid-peroxidationproducts in avocado peels precedes other signs of ripening JAmer Soc Hort Sci 1991 116823-826

63 Moyle R Fairbairn DJ Ripi J Crowe M Botella JR Developing pine-apple fruit has a small transcriptome dominated by metal-lothionein J Exp Bot 2005 56101-112

64 Brennan T Frenkel C Involvement of hydrogen peroxide in theregulation of senescence in pear Plant Physiol 1977 59411-416

65 Mittler R Vanderauwera S Gollery M Van Breusegem F Reactiveoxygen gene network of plants Trends Plant Sci 2004 9490-498

66 Ogawa T Ueda Y Yoshimura K Shigeoka S Comprehensive anal-ysis of cytosolic Nudix hydrolases in Arabidopsis thaliana J BiolChem 2005 28025277-25283

67 Terrier N Romieu C Grape berry acidity Kluwer Academic Pub-lisher 200135-57

68 Cheng GW Breen PJ Activity of phenylalanine ammonia-lyase(PAL) and concentrations of anthocyanins and phenolics indeveloping strawberry fruit J Am Soc Hortic Sci 1991116865-869

69 Zhang X-Y Wang X-L Wang X-F Xia G-H Pan Q-H Fan R-C WuF-Q Yu X-C Zhang D-P A shift of phloem unloading from sym-plasmic to apoplasmic pathway is involved in developmentalonset of ripening in grape berry Plant Physiol 2006 142220-232

70 Davies C Wolf T Robinson SP Three putative sucrose trans-porters are differentially expressed in grapvine tissues PlantSci 1999 14793-100

71 Davies C Robinson SP Sugar accumulation in grape berriesCloning of two putative vacuolar invertase cDNAs and theirexpression in grapevine tissues Plant Physiol 1996 111275-283

72 Nguyen-Quoc B Foyer CH A role for futile cycles involvinginvertase and sucrose synthase in sucrose metabolism oftomato fruit J Exp Bot 2001 52881-889

73 Terrier N Sauvage FX Ageorges A Romieu C Changes in acidityand in proton transport at the tonoplast of grape berriesduring development Planta 2001 21320-8

74 Mellema S Eichenberger W Rawyler A Suter M Tadege M Kuhle-meier C The ethanolic fermentation pathway supports respi-ration and lipid biosynthesis in tobacco pollen Plant J 200230329-336

75 Boss PK Davies C Robinson SP Analysis of the expression ofanthocyanin pathway genes in developing Vitis vinifera L cvShiraz grape berries and the implications for pathway regu-lation Plant Physiol 1996 1111059-1066

76 Martinoia E Klein M Geisler M Bovet L Forestier C Kolukisaoglu UMuller-Rober B Schulz B Multifunctionality of plant ABC trans-porters ndash more than just detoxifiers Planta 2002 214345-355

77 Wissel K Pettersson F Berglund A Jansson S What affects mRNAlevels in leaves of field-grown aspen A study of developmen-tal and environmental influences Plant Physiol 20031331190-1197

78 Salome PA McClung CR PSEUDO-RESPONSE REGULATOR7 and 9 are partially redundant genes essential for the tem-perature responsiveness of the Arabidopsis circadian clockPlant Cell 2005 17791-803

79 Ulm R Baumann A Oravecz A Mate Z Adam E Oakeley EJ SchaferE Nagy F Genome-wide analysis of gene expression revealsfunction of the bZIP transcription factor HY5 in the UV-Bresponse of Arabidopsis Proc Natl Acad Sci USA 20041011397-1402

80 Tepperman JM Zhu T Chang HS Wang X Quail PH Multiple tran-scription-factor genes are early targets of phytochrome Asignaling Proc Natl Acad Sci USA 2001 989437-9442

81 Conde C Agasse A Glissant D Tavares R Geros H Delrot S Path-ways of glucose regulation of monosaccharide transport ingrape cells Plant Physiol 2006 1411563-1577

82 Fillion L Ageorges A Picaud S Coutos-Thevenot P Lemoine RRomieu C Delrot S Cloning and expression of a hexose trans-porter gene expressed during the ripening of grape berryPlant Physiol 1999 1201083-1094

83 Vignault C Vachaud M Cakir B Glissant D Dedaldechamp F ButtnerM Atanassova R Fleurat-Lessard P Lemoine R Delrot S VvHT1encodes a monosaccharide transporter expressed in theconducting complex of the grape berry phloem J Exp Bot2005 561409-1418

84 Goto-Yamamoto N Wan GH Masaki K Kobayashi S Structureand transcription of three chalcone synthase genes of grape-vine (Vitis vinifera) Plant Sci 2002 162867-872

85 Harmer SL Hogenesch JB Straume M Chang HS Han B Zhu TWang X Kreps JA Kay SA Orchestrated transcription of keypathways in Arabidopsis by the circadian clock Science 20002902110-2113

86 Moser C Gatto P Moser M Pindo M Velasco R Isolation of func-tional RNA from small amounts of different grape and appletissues Mol Biotechnol 2004 2695-100

87 Gentleman RC Carey VJ Bates DM Bolstad B Dettling M Dudoit SEllis B Gautier L Ge Y Gentry J Hornik K Hothorn T Huber WIacus S Irizarry R Leisch F Li C Maechler M Rossini AJ Sawitzki GSmith C Smyth G Tierney L Yang JY Zhang J Bioconductor opensoftware development for computational biology and bioin-formatics Genome Biol 2004 5R80

88 Ihaka R Gentleman R R a language for data analysis and graph-ics J Comput Graph Stat 1996 5299-314

89 Wu Z Irizarry RA Preprocessing of oligonucleotide arraydata Nat Biotechno 2004 22656-658 author reply 658

90 TM4 Microarray software suite [httpwwwtm4org]91 Affymetrix home site [httpwwwaffymetrixcomindexaffx]92 Nadershahi A Fahrenkrug SC Ellis LB Comparison of computa-

tional methods for identifying translation initiation sites inEST data BMC Bioinformatics 2004 514-



Publish with BioMed Central and every scientist can read your work free of charge

BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime

Sir Paul Nurse Cancer Research UK

Your research papers will be

available free of charge to the entire biomedical community

peer reviewed and published immediately upon acceptance

cited in PubMed and archived on PubMed Central

yours mdash you keep the copyright

Submit your manuscript herehttpwwwbiomedcentralcominfopublishing_advasp

BioMedcentral

93 ESTate [httpwwwbionetbionetmmbio-soft1999-December021776html]

94 Altschul SF Madden TL Schaffer AA Zhang J Zhang Z Miller W Lip-man DJ Gapped BLAST and PSI-BLAST a new generation ofprotein database search programs Nucleic Acids Res 1997253389-3402

95 Ramakers C Ruijter JM Deprez RH Moorman AF Assumption-free analysis of quantitative real-time polymerase chainreaction (PCR) data Neurosci Lett 2003 33962-66

96 Pfaffl MW A new mathematical model for relative quantifica-tion in real-time RT-PCR Nucleic Acids Res 2001 29e45

97 ArrayExpress home [httpwwwebiacukmicroarray-asaerae-main[0]]



Conclusion The time-course gene expression analysis of grapevine berry development hasidentified the occurrence of two well distinct phases along the process The pre-veacuteraison phaserepresents a reprogramming stage of the cellular metabolism characterized by the expression ofnumerous genes involved in hormonal signalling and transcriptional regulation The post-veacuteraisonphase is characterized by the onset of a ripening-specialized metabolism responsible for thephenotypic traits of the ripe berry Between the two phases at veacuteraison an oxidative burst and theconcurrent modulation of the anti-oxidative enzymatic network was observed The large numberof regulatory genes we have identified represents a powerful new resource for dissecting themechanisms of fruit ripening control in non-climacteric plants

BackgroundGrape is among the most ancient widely cultivated fruitcrops The keen interest in the understanding of grapeberry ripening is justified by the economic relevance of thequality of grapes and their processed products such aswine juice and dried fruit The onset of the genomic erahas brought unprecedented progress into our knowledgeof Vitis molecular biology and powerful genetic andgenomic tools are now available or are being developedsuch as mapping populations genetic and physical maps[12] an extensive ESTs collection (342576 sequencesdeposited at the NCBI EST database-March 19 2007 [3])and several tools for gene expression analysis Theseresources enable us to identify grape as a model plant forstudies on non-climacteric anthocyanin accumulatingfruits beside tomato which is currently the model plantfor climacteric carotenoid accumulating fruits Grapeberry development from anthesis to ripening is classicallydivided into three phases on the basis of chemical andmorphological traits [4] berry formation characterizedby exponential growth of the berry and accumulation oforganic acids mainly malate in the vacuole veacuteraison atransition phase during which growth declines and berriesstart to change colour and soften ripening characterizedby an increase in pH marked berry growth and accumula-tion of sugars anthocyanins and flavour-enhancing com-pounds Published studies based on EST analysis [5-7]differential display [8] cDNA-arrays [9] and oligo arrays[10] have shown distinct and extensive changes in thetranscriptome occurring during ripening These studiesreport up to 740 developmentally regulated genes mainlyinvolved in defence stress response primary and second-ary metabolism berry growth and hormonal metabolismThe very recent study of Grimplet et al [11] provides anextensive description of these functional classes investi-gating their tissue-specific expression in the ripe berry bymeans of the Affymetrix Vitis GeneChipreg

With the aim of understanding the transcriptionalchanges underlying grapevine berry ripening and the reg-ulatory mechanisms acting at this level we were able toidentify 1477 genes modulated during berry develop-ment comprising 282 regulatory factors in addition to the

structural genes by the use of the Affymetrix Vitis Gene-Chipreg The results obtained from replicated experimentscarried out in three growing seasons provide also insightsinto the relationship between genetic and environmentaleffects on berry development control

Results and DiscussionGrape berries phenotypic evaluationsFigure 1 shows the changes in average berry weightorganic acids total soluble solids and anthocyanin con-tent of grape berries sampled weekly during the 20032005 and 2006 growing seasons The date of veacuteraison wasset at approximately 8 weeks post-flowering (wpf) in allyears In order to characterize berry ripening we chosethree distinct moments according to [12] the last stage ofberry formation time-point A (TP A stage E-L 33) char-acterized by small hard green berries still accumulatingorganic acids and two stages during berry ripening TP B(stage E-L 34) just before veacuteraison characterized by ber-ries in the green state with maximum acidic content start-ing to soften and TP C (stage E-L 36) when the ripeningprocess is well established and berries are growing fastcolouring and accumulating sugars The identification ofthe corresponding developmental stages in the three yearswas based on the biochemical profiles As average temper-atures during 2003 remained constantly unusually higherfrom April to October [see Additional file 1] floweringand harvest occurred approximately one week earlier in2003

Statistical analysisRNA extracted in triplicate from berries harvested at threetime points was used to hybridize 27 Affymetrix VitisGeneChipsreg Datasets from each season were analyzedseparately The quality of the biological replicates wasevaluated by the R-squared coefficient [see Additional file2] which ranged from 089 to 099 Minimum R-squaredvalues were found for TP B samples suggesting a highersample heterogeneity approaching veacuteraison when dra-matic transcriptional changes occur in less than 48 hours[10]
































































































































GO Gene Ontology








Additional material






















































































































BioMedcentral





































































































































GO Gene Ontology








Additional material






















































































































BioMedcentral











Additional material






















































































































BioMedcentral













































































































BioMedcentral







Genome-wide transcriptional analysis of grapevine berry ripening reveals a set of genes similarly...

Documents

Transcript of Genome-wide transcriptional analysis of grapevine berry ripening reveals a set of genes similarly...