lable at ScienceDirect

Plant Physiology and Biochemistry 49 (2011) 917e922

Contents lists avai

Plant Physiology and Biochemistry

journal homepage: www.elsevier .com/locate/plaphy

Research article

Hydrogen peroxide formation in cacao tissues infected by the hemibiotrophicfungus Moniliophthora perniciosa

Cristiano Villela Dias a,b, Juliano Sales Mendes a, Anderson Carvalho dos Santos a,Carlos Priminho Pirovani a, Abelmon da Silva Gesteira a,c, Fabienne Micheli a,d,*,Karina Peres Gramacho e, John Hammerstone b, Paulo Mazzafera f,Júlio Cézar de Mattos Cascardo a,1

aUESC, DCB, Centro de Biotecnologia e Genética, Rodovia Ilhéus-Itabuna Km 16, 45662-900 Ilhéus-BA, BrazilbMars Center for Cocoa Science, Barro Preto, Bahia, Brazilc EMBRAPA Mandioca e Fruticultura Tropical, Rua Embrapa, s/n� , 44380-000 Cruz das Almas-BA, BrazildCIRAD, UMR DAP, Avenue Agropolis TA96/03, 34398 Montpellier Cedex 5, FranceeCocoa Research Center, CEPLAC/CEPEC, 45600-970 Itabuna-BA, Brazilf IB/UNICAMP, Campinas, São Paulo, Brazil

a r t i c l e i n f o

Article history:Received 28 January 2011Accepted 10 May 2011Available online 18 May 2011

Keywords:Ascorbic acidHydrogen peroxideOxalic acidResistanceTheobroma cacaoWitches’ broom disease

Abbreviations: ABA, abcisic acid; Apx, ascorbate pCaOx, calcium oxalate; DAB, 3-30diaminobenzideneDhar, dehydroascorbate reductase; Glp, germin-likecacao; hai, hour after inoculation; JA, jasmonic acid; Ooxidase from wheat; ROS, reactive oxygen species; SA* Corresponding author. UESC, DCB, Centro de Biote

Ilhéus-Itabuna Km 16, 45662-900 Ilhéus-BA, Brazil. Te73 36805226.

E-mail address: fabienne.micheli@cirad.fr (F. Mich1 In memoriam.

0981-9428/$ e see front matter � 2011 Elsevier Masdoi:10.1016/j.plaphy.2011.05.004

a b s t r a c t

In plantepathogen interaction, the hydrogen peroxide (H2O2) may play a dual role: its accumulationinhibits the growth of biotrophic pathogens, while it could help the infection/colonization process ofplant by necrotrophic pathogens. One of the possible pathways of H2O2 production involves oxalic acid(Oxa) degradation by apoplastic oxalate oxidase. Here, we analyzed the production of H2O2, the presenceof calcium oxalate (CaOx) crystals and the content of Oxa and ascorbic acid (Asa) e the main precursor ofOxa in plants e in susceptible and resistant cacao (Theobroma cacao L.) infected by the hemibiotrophicfungus Moniliophthora perniciosa. We also quantified the transcript level of ascorbate peroxidase (Apx),germin-like oxalate oxidase (Glp) and dehydroascorbate reductase (Dhar) by RT-qPCR. We report that theCaOx crystal amount and the H2O2 levels in the two varieties present distinct temporal and genotype-dependent patterns. Susceptible variety accumulated more CaOx crystals than the resistant one, andthe dissolution of these crystals occurred in the early infection steps and in the final stage of the diseasein the resistant and the susceptible variety, respectively. High expression of the Glp and accumulation ofOxa were observed in the resistant variety. The content of Asa increased in the inoculated susceptiblevariety, but remained constant in the resistant one. The susceptible variety presented reduced Dharexpression. The role of H2O2 and its formation from Oxa via Apx and Glp in resistant and susceptiblevariety infected by M. perniciosa were discussed.

� 2011 Elsevier Masson SAS. All rights reserved.

1. Introduction

Reactive oxygen species (ROS) are continuously produced inplants as a result of the aerobic metabolism. ROS are originated

eroxidase; Asa, ascorbic acid;; dai, day after inoculation;oxalate oxidase protein fromxa, oxalic acid; OXO, oxalate, salicylic acid; Tub, tubuline.cnologia e Genética, Rodovial.: þ55 7336805196; fax: þ55

eli).

son SAS. All rights reserved.

from the excitement of the molecular O2, forming simple molecularoxygen (1O2), or from the transfer of one, two or three electrons tothe O2, leading to the formation of superoxide radical (O2

�),hydrogen peroxide (H2O2) or hydroxyl radical (OH�), respectively.The fast increase of ROS levels in cells is called oxidative burst andconstitutes one of the fastest responses to pathogen attack. Thesemolecules can be harmful to the cell, but on the other hand they canactivate defense pathways or metabolic responses to biotic andabiotic stresses [1]. Hydrogen peroxide, the chemically more stableROS, is involved in a series of cell processes related to plant defenseto pathogens via the biosynthesis of other important signalingmolecules such as salicylic acid (SA), abcisic acid (ABA), jasmonicacid (JA), ethylene and Ca2þ [1]. However, depending on theplantepathogen interaction type, the H2O2 may play a dual role; itsaccumulation inhibited the growth of biotrophic pathogens, while

Fig. 1. DAB staining for H2O2 detection in susceptible and resistant cacao varieties.Apical meristems of infected (24 hai) susceptible (A) and resistant (B) varieties infil-trated with distilled water. DAB staining of susceptible (C) and resistant (D) cacaovarieties at 24 hai (100�). Dark field micrography of susceptible (E) and resistant (G)varieties at 72 hai (20�). Dark field micrography of resistant variety at 0 hai (F; 100�).DAB staining of the resistant variety at 48 hai (H; 400�). White arrows show CaOxcrystal surrounded by H2O2.

C.V. Dias et al. / Plant Physiology and Biochemistry 49 (2011) 917e922918

it has been ignored by necrotrophic pathogens or reported asa helper during the infection/colonization process of plant by theseones [2e5]. In the case of the hemibiotrophic fungus Septoria tritici,the accumulation of H2O2 inhibited the infection during the initialstage (biotrophic phase) of the disease in wheat [6]. Hydrogenperoxide is produced in plants via two possible pathways: i)cell wall oxidases catalyze the oxidation of the NADH to NADþ,which in turn enables reduction of O2 to O2

�. Then, the dismutationof O2

� produces O2 and H2O2 [7]; ii) apoplastic oxalate oxidase andamine oxidase have been proposed to generate H2O2 [8,9].

In 2007, Ceita et al. [10] analyzed the interaction betweena susceptible variety of cacao (Theobroma cacao L.; Catongo variety)and the hemibiotrophic fungus Moniliophthora perniciosa. Duringthe time course disease, two distinct phases were observed, i)a biotrophic phase, in which the mycelia grows only in the apoplastand presents monokaryotic hyphae; and ii) a saprotrophic phase, inwhich the mycelia invades the cell, presenting dikaryotic hyphaeand formation of clamp connections. At the end of the saprotrophicphase, plant cell death occurs, followed by basidiocarp productionand releasing of spores. The authors also reported an increase ofcalcium oxalate (CaOx) crystals concomitant with the funguscolonization (biotrophic phase). During the necrotrophic phase, anincrease in H2O2 level was observed, associated with the dissolu-tion of the CaOx crystals and the degradation of the generated freeoxalic acid (Oxa) by the germin-like oxalate oxidase (Glp). Althougha considerable knowledge on the whole fungus cycle and in -omicsof T. cacao has been accumulated in the last 10 years [11], little isknown about the cellular mechanisms of the resistantcacaoeM. perniciosa interaction.

In this study, we analyzed the production of H2O2 and thecontent of free Oxa and ascorbic acid (Asa) emain precursor of Oxain plants [12,13]. We also quantified the expression of ascorbateperoxidase (Apx; which converts Asa into dehydroascorbate),dehydroascorbate reductase (Dhar; which uses dehydroascorbateand glutathione to form Asa and glutathione disulfide) and Glpgenes in two cacao varieties (susceptible vs resistant) infected andnon infected with M. perniciosa. We showed that the CaOx crystalamount and the H2O2 levels in the two varieties present distincttemporal and genotype-dependent patterns. High expression of theGlp and accumulation of Oxawere observed in the resistant variety.The content of Asa increased in the inoculated susceptiblevariety, but remained constant in the resistant one. The susceptiblevariety presented reduced Dhar expression. The role of H2O2 and itsformation from Oxa via Apx and Glp in resistant and susceptiblevariety infected by M. perniciosa were discussed.

2. Results

2.1. Detection of H2O2 and CaOx crystals in resistantand susceptible cacao varieties

An increase of H2O2 production was observed at early stages(24 hai, Fig. 1D; 48 and 72 hai, data not shown) in the resistantcacao variety inoculated with M. perniciosa, particularly in thetissues adjacent to the vascular system (Fig. 1D and H). No presenceof H2O2 in the susceptible variety was observed (Fig. 1C); at 24 haithe susceptible variety did not differ from the control inoculated byM. perniciosa and infiltrated with water (Fig. 1A and B).

CaOx crystal were visible on sections of both varieties as brightwhite spots (arrows; Fig. 1EeG). However, at 72 hai, much moreCaOx crystals were observed in the susceptible variety (Fig. 1E) incomparison to the resistant one (Fig. 1G). The resistant varietypresented less CaOx crystals at 72 hai (Fig. 1G) than before theinoculation (0 hai, Fig. 1F), suggesting a dissolution of the CaOxcrystals after the plant infection. This CaOx crystal dissolution was

associated with the production of H2O2 in the resistant plants(Fig. 1H). At 24 hai, some remaining CaOx crystals were stillobserved, surrounded by adjacent cells containing H2O2 (DABstaining; Fig. 1G). In the susceptible variety, no reduction of CaOxcrystal number was observed.

2.2. Quantification of Oxa and Asa in resistant and susceptiblecacao varieties inoculated by M. perniciosa

The basal level of Oxawas about 2.5 times higher in the resistantvariety than in the susceptible one (Fig. 2A and B, 0 hai). Bothinoculated varieties showed an increase of Oxa at 3 dai. However, inthe inoculated susceptible variety, the Oxa level decreased at 15 dai(returned to basal level), then increased at 30 dai and finallydecreased at 45 and 60 dai (with no significant difference with theplant control; Fig. 2A). In the inoculated resistant variety, the Oxa

Fig. 2. Quantification of free Asa and Oxa in resistant and susceptible cacao varieties. Oxa level on susceptible (A) and resistant (B) varieties. Asa level on susceptible (C) andresistant (D) varieties. Results are the average of three biological replicates � standard error. Different letters indicate significant statistical difference between samples by theDuncan test. Dark gray bars: plant control; light gray bars: plant inoculated com M. perniciosa.

C.V. Dias et al. / Plant Physiology and Biochemistry 49 (2011) 917e922 919

level remainedhigh at 15 dai, then decreased at 30 dai and increasedat 45 and 60 dai (with no significant difference with the plantcontrol at 45 dai; Fig. 2B).

The basal level of Asa was about 1.5 times higher in the resistantvariety than in the susceptible one (Fig. 2C and D, 0 hai). In theinoculated susceptible variety, the Asa level increased from 3 to 30dai, decreased at 45 dai (without significant difference with theplant control) and then increased at 60 dai (same level than at 30dai; Fig. 2C). In the inoculated resistant plant, the Asa level at 3 and15 dai remained constant and equal to the basal level (withoutsignificant difference with the plant control), then slightlydecreased at 30 dai and finally slightly increased at 45 and 60 dai(without significant difference with the plant control for 45 and 60dai; Fig. 2D).

2.3. Transcript level of Apx, Glp and Dhar in susceptibleand resistant cacao varieties inoculated with M. perniciosa

The expression patterns of Apx, Glp and Dhar were obtained inthe early stages of the infection (up to 72 hai). No significant vari-ation of the expression of the three genes studied was observed inthe inoculated susceptible variety (Fig. 3B). In the inoculatedresistant variety, a significant increase of Apx, Glp and Dhar tran-script levels was observed at 48 hai. At 72 hai, Apx transcripts levelwas still high (and significantly different from the one observedbefore inoculation; Fig. 3A, 0 hai) while Glp and Dhar transcriptlevels were as low as the ones observed at 0 hai (Fig. 3A). Theexpression of the Dhar and Glp in the resistant variety followed thesame patterns as the Apx throughout the time course, witha correlation coefficient of 0.96 and 0.95 for Dhar/Apx and Glp/Apx,respectively (data not shown).

3. Discussion

Hydrogen peroxide plays an important role in plants understress conditions as a signaling molecule which intermediates

a series of important cellular responses [1,14]. In the resistant cacaovariety infected by M. perniciosa, we observed a significant accu-mulation of H2O2 throughout the stem apex vascular system in thefirst 72 hai (Fig. 1D) while no H2O2 accumulation was observed atthe early infection stages in the susceptible variety (Fig. 1C). Ina previous work, we showed a significant accumulation of H2O2 inthe advanced stages of infection (45 dai) in the susceptible cacaoinfected by M. perniciosa facilitating the transition phase of thefungus from biotrophic to necrotrophic [10] and corroboratingother works involving necrotrophic or hemibiotrophic pathogens[2,5,6]. For example, the infection of Arabidopsis thaliana by Botrytiscinerea led to the accumulation of H2O2, which killed the host cellsfacilitating the pathogen invasion [2].

The successive sequence of production of Oxa (higher after 72hai), transcript level of Glp (higher at 48 hai), dissolution of CaOxcrystals andH2O2 accumulation (visible up to 72 hai) in the resistantinfected variety suggested that the Glp participates in Oxa degra-dation leading to the formation of CO2 andH2O2. These observationscorroborate with previous results obtained for infected susceptiblecacao plants in which the dissolution of CaOx crystals was followedby a massive accumulation of H2O2 in infected tissues at the lateinfection times [10]. Togetherwith the increase of the Glp transcriptlevel in the resistant inoculated cacao variety, we observed theincrease of the Apx and Dhar transcript levels (Fig. 3A). Apx usesH2O2 to convert Asa into dehydroascorbate, and Dhar uses dehy-droascorbate to react with glutathione and form Asa and gluta-thione disulfide. Therefore, at the same time, Asa was used assubstrate of Apx, and was regenerated by Dhar reaction, whichmayexplain the similar Asa contents found in the inoculated and non-inoculated resistant plantlets (Fig. 2D). Despite the existence ofseveral reports describing the function of the CaOx in the freecalcium amount regulation process in plants [15,16], a novel andmore dynamic role for this molecule was raised by the discovery ofthewheat germins [17], opening to other possibilities regarding thefunctionality on the formation and/or dissolution of CaOx in plants.The over-expression of an oxalate oxidase (OXO) in transgenic

Fig. 3. Expression pattern of Apx, Glp and Dhar in cacao varieties inoculated withM. perniciosa. A. Resistant inoculated variety. B. Susceptible inoculated variety. Resultsare the average of three biological replicates � standard deviation. t-test was made bycomparison to 0 hai for each gene. ns: non significant (p > 0.05); **: significant atp < 0.001; *: significant at p < 0.05. Dark gray bars: Apx; light gray bars: Glp; whitebars: Dhar.

C.V. Dias et al. / Plant Physiology and Biochemistry 49 (2011) 917e922920

plants triggered the production of H2O2 and defense responses bythe plant to the fungal infection [18e20]. It was also demonstratedthe involvement of OXO in the reduction of the progression of somefungal diseases, conferring, in some cases, partial resistance [21].It was also reported the existence of a crystal matrix protein asso-ciated with CaOx precipitation [22]; our group recently identifiedcacao proteins associatedwith crystal formationwhich are involvedin the formation/dissolution of CaOx crystals (data not shown). Theparticipation of Asa in the generation of Oxa has been previouslyproposed [23], and since then, has stimulated the interest of theresearchers [13]. By using Asa, Oxa, eritroascorbic acid, galactoseand glycolate labeled with 14C, it has been demonstrated that Asaacts as precursor of Oxa and CaOx in idioblasts of Pistia stratiotes,and that glycolate is a very poor substrate for Oxa synthesis andCaOx generation. This work confirmed the results of Wheeler et al.[24], which proposed that the L-galactosemay be a key intermediatein the conversion of the D-glucose to Asa in plants.

In the susceptible cacao variety, we observed that the CaOxcrystals were not dissolved after the infection by M. perniciosa(Fig. 1E). In addition an increase of Asa in the susceptible plants

(Fig. 2C) was observed but without variation of Glp, Apx and Dhartranscripts levels (Fig. 3B) suggesting a relation between Asa andCaOxcontents. Ceita et al. [10] already suggested the participation ofAsa in the interaction of cacaoeM. perniciosa as a pro-oxidantmolecule [25]. In Medicago truncatula mutants, a relationshipbetween Asa levels and the presence of CaOx crystals was demon-strated, stating that, in this plant model, Asa is the main source ofOxa [26]. Other studies showed that Oxa itself is involved in pro-grammed cell death (PCD) and is responsible for the increase of ROSin plant. However, when ROS production was inhibited, theapoptotic-like-cell death induced by Oxa does not occur [27].According to these findings Glp, CaOx and Oxa are involved in CaOxcrystal dissolution and subsequent H2O2 formation, and implicated,in one hand, in plant resistance process (acting at the early stages ofthe interaction) or, on the other hand, contribute to pathogen lifecycle in the susceptible variety (at later stages of the disease). It isinteresting to note that even M. perniciosa also produces CaOxcrystals [28], these probably are not involved in the earliest stages ofdisease, but contribute, in the susceptible variety, to the develop-ment of the pathogen. The lowexpression of Dhar in the susceptiblevariety explains the differences found in the behavior of ascorbateand oxalate between the contrasting varieties. In conclusion, theincreased H2O2 levels in the two cacao varieties present distincttemporal and functional patterns. Once produced at the beginningof the infection by the resistant variety, it contributes to the infec-tion control and to plant resistance. In advanced stages of thedisease in the susceptible variety, it promotes the pathogen devel-opment and the finalization of its life cycle.

4. Material and methods

4.1. Plant material

Plantlets of T. cacao L. susceptible (Catongo variety) and resistant(TSH 1188 variety) toM. perniciosawere cultivated in greenhouse atCEPEC/CEPLAC (Centro de Pesquisas do Cacau da Comissão Execu-tiva do Plano da Lavoura Cacaueira, Ilhéus, Bahia State, Brasil) undernatural light and 90% of relative humidity as described by [29].Apical meristems of 6-week old-plantlets were inoculated by thespraying method [30] using a 106.ml�1 basidiospore suspensionfrom the M. perniciosa Cp1441 CEPEC/CEPLAC strain. After inocu-lation, plantlets were kept during 24 h at 25 �C � 2 �C in a water-saturated atmosphere to allow M. perniciosa spore germination,penetration and infection [30]. A spore viability test was conductedin the humid chamber (25 �C) 24 h after inoculation and comparedto spore viability obtained before inoculation. Apical meristems ofthree individual inoculated plantlets of each variety (biologicalreplicates) were collected before inoculation (0), at 24 h, 48 h, 72 hafter inoculation (hai) and at 15, 30, 45 and 60 days after inocula-tion (dai). In the time course disease, 24e72 hai corresponded tothe early stages of the infection, 30 dai to wilt and necrosis of theyoung leaves, and 45 dai to hypertrophy of the stem (“greenbroom”) and first microscopic necrosis symptoms [10]. At 60 dai,the infected plant presented macroscopic symptoms called “drybroom” and at 90 dai the plant was considered as completelynecrotic. Plantlets inoculated with water, and maintained andharvested in the same condition as the inoculated ones, were usedas control. Inoculated and control samples were frozen in liquidnitrogen, ground to a fine powder and stored at �80 �C.

4.2. Detection of H2O2 by DAB staining methodand visualization of CaOx crystals

Apical meristems of TSH 1188 and Catongo varieties harvestedat 24, 48 and 72 hai were immersed in 1 mg/ml of 3e30

C.V. Dias et al. / Plant Physiology and Biochemistry 49 (2011) 917e922 921

diaminobenzidene (DAB) e HCl, pH 3.8 (Sigma) and infiltratedunder vacuum for 4 h, as described by [31]. Samples were cleared inboiling ethanol (96%) for 20 min, then hand-sectioned with a razorblade, mounted in 50% glycerol and examined using an opticalmicroscope (Olympus CX41). Images were obtained using a digitalcamera Olympus C-7070. H2O2 was visualized as a reddish-browncoloration. Control samples were immersed and infiltrated withdistilled water.

For CaOx crystal visualization, apical meristems were cleared inboiling ethanol (96%) for 20 min, then hand-sectioned with a razorblade, mounted in 50% glycerol and examined using opticalmicroscope with dark field (Olympus CX41). Crystals were visual-ized as bright spots at lowmagnification and perfect faceted crystalat high magnification.

4.3. Quantification of ascorbic and oxalic acid by HPLC

Organic acid extraction and quantification was carried out asdescribed by [13] with some modifications. In 2 ml tubes, 6 mg offrozen material were mixed with 1 ml of extraction buffer (4 mMH2SO4, 5 mMDTT, PVPP added to a concentration of 10 mg/ml). Thesample was vortexed for 15 min and centrifuged at 26,200g for20 min at room temperature. One hundred microliters of filteredsupernatant (0.45 mm) was injected into the HPLC (ÄKTAbasic�,GE-Health Care). Asa and Oxa were separated in a Bio-Rad AminexHPX-87H column (300 � 7.8 mm) at a flow rate of 0.7 ml/min, withdetection of Oxa at 210 nm (Rt ¼ 8 min) and Asa at 243 nm(Rt ¼ 10 min). The chromatogram peaks were integrated andanalyzed using the UNICORN� version 5.0 software. Oxa and Asaconcentrations in cacao samples were obtained by comparisonwitha calibration curve obtained by separation in the HPLC of knownamounts of pure standards (SigmaeAldrich). Three biologicalreplicates (3 meristems for each cacao variety at a given point of thetime course) and three experimental replicates (3 quantifications ofAsa and Oxa for each meristem) were performed. The quantifica-tion of Oxa e Asa were made concomitantly, in the same run andfrom the same sample to minimize bias in the analysis. Statisticalanalysis was performed with the software SASM e Agri whichtested the experiments as a completely randomised design. Anal-ysis of variance (ANOVA) was applied and for means comparisonDuncan’s test was employed, with a critical value of P ¼ 0.05. CaOxquantification was not performed since its formation/dissolutionpattern was well established by [10] in both varieties.

4.4. RNA extraction and RT-qPCR analyses

Total RNA from frozen inoculated and control tissues was iso-lated as described by [32], and cleaned using the Rneasy Plant Minikit as described by themanufacturer (Qiagen). The RNAwas treatedwith DNase RNase-free according to the manufacturer (Fermentas),and then separated on 1% DEPC-treated agarose gel and stainedwith ethidium bromide to confirm RNA integrity and the estimatedRNA concentration. The first strand cDNA was synthesized in 10 mlfrom 1 mg of total RNA using the SuperScript II kit (Invitrogen)according to the manufacturer’s instructions. RT-qPCR analysis wascarried out for the germin-like oxalate oxidase (Glp), ascorbateperoxidase (Apx) and dehydroascorbate reductase (Dhar) cacaogenes [29]. The tubulin gene (Tub) from cacao was used as refer-ence (endogenous control) in the qPCR experiments (Supplemen-tary material 1). The qPCR reaction containing 150 ng of cDNA,0.2 mM of each primer and 10 ml of Applied Mix was performedusing the Syber Green Kit PCRmaster mix on the Applied 7500 RealTime PCR (Applied Biosystems). The expression of each gene wasobtained in comparison to the expression level of the Tub gene(relative quantification/2ΔΔCT). Cycling parameters were as follows:

95 �C for 10 min and 45 cycles of 95 �C for 15 s and 60 �C for 45 s.Automated gene expression analysis was performed using theequipment’s software (Applied Biosystems 7500 System SDS Soft-ware v1.3.1.21).

Acknowledgements

This research was supported by the Moniliophthora perniciosaProteomic Network granted by the Financiadora de Estudos eProjetos (FINEP; project n� 01.07.0074-00) and the Fundação deAmparo à Pesquisa da Bahia (FAPESB; project n� 1431080017116).Thework of C. Villela Dias and J. SalesMendeswas supported by theFAPESB and the Conselho Nacional de Desenvolvimento Científico eTecnológico do Brasil (CNPq). We thank Dr. Marcio G.C. Costa(UESC) for advises about statistical analysis and Dr. Claudia FortesFerreira (Embrapa) for critical reading of the manuscript. Wededicate this work to Julio Cezar de Mattos Cascardo: the seedsplanted in our minds will make his presence will always remainalive in our hearts.

Appendix. Supplementary data

Supplementary data associated with this article can be found inthe online version, at doi:10.1016/j.plaphy.2011.05.004.

References

[1] L.-J. Quan, B. Zhang, W.-W. Shi, H.-Y. Li, Hydrogen peroxide in plants:a versatile molecule of the reactive oxygen species network, J. Integr. PlantBiol. 50 (2008) 2e18.

[2] E.M. Govrin, A. Levine, The hypersensitive response facilitates plant infectionby the necrotrophic pathogen Botrytis cinerea, Curr. Biol. 10 (2000) 751e757.

[3] J.A.L. van Kan, Licensed to kill: the lifestyle of a necrotrophic plant pathogen,Trends Plant Sci. 11 (2006) 247e253.

[4] N. Temme, P. Tudzynski, Does Botrytis cinerea ignore H2O2-induced oxidativestress during infection? Characterization of Botrytis activator protein 1, MPMI22 (2009) 987e998.

[5] S. Deller, K.E. Hammond-Kosack, J.J. Rudd, The complex interactions betweenhost immunity and non-biotrophic fungal pathogens of wheat leaves, J. Plant.Physiol. 168 (2011) 63e71.

[6] N.P. Shetty, R. Mehrabi, H. Lütken, A. Haldrup, G.H.J. Kema, B. David,D.B. Collinge, H.J.L. Jorgensen, Role of hydrogen peroxide during the interac-tion between the hemibiotrophic fungal pathogen Septoria tritici and wheat,New Phytol. 174 (2007) 637e664.

[7] S. Bhattacharjee, Reactive oxygen species and oxidative burst: roles in stress,senescence and signal transduction in plant, Curr. Sci. 89 (2005) 1113e1121.

[8] G.P. Bolwell, P. Wojtaszek, Mechanisms for the generation of reactive oxygenspecies in plant defense broad perspective, Physiol. Mol. Plant. Pathol. 51(2005) 347e366.

[9] X. Hu, D.L. Bidney, N. Yalpani, J.P. Duvick, O. Crasta, O. Folkerts, G. Lu, Over-expression of a gene encoding hydrogen peroxide generating oxalate oxidaseevokes defense responses in sunflower, Plant Physiol. 133 (2003) 170e181.

[10] G.O. Ceita, J.N.A. Macêdo, T.B. Santos, L. Alemanno, A.S. Gesteira, F. Micheli,A.C. Mariano, K.P. Gramacho, D.C. Silva, L. Meinhardt, P. Mazzafera,G.A.G. Pereira, J.C.M. Cascardo, Involvement of calcium oxalate degradationduring programmed cell death in Theobroma cacao tissues triggered by thehemibiotrophic fungus Crinipellis perniciosa, Plant Sci. 173 (2007) 106e117.

[11] F. Micheli, M. Guiltinan, K.P. Gramacho, A.V. Figueira, J.C.M. Cascardo,S. Maximova, C. Lanaud, Functional genomics of cacao, Adv. Bot. Res. 155(2010) 119e177.

[12] H.T. Horner, A.P. Kausch, B.L. Wagner, Ascorbic acid: a precursor of oxalate incrystal idioblasts of Yucca torreyi in liquid root culture, Int. J. Plant Sci. 161(2007) 861e868.

[13] S.E. Keates, N.M. Tarlyn, F.A. Loewus, V.R. Franceschi, L-ascorbic acid andL-galactose are sources for oxalic acid and calcium oxalate in Pistia stratiotes,Phytochemistry 53 (2000) 433e440.

[14] Y. Cheng, C. Song, Hydrogen peroxide homeostasis and signaling in plant cells,Sci. China Ser. C 49 (2006) 1e11.

[15] P.A. Nakata, Advances in our understanding of calcium oxalate crystalformation and function in plants, Plant Sci. 164 (2003) 901e909.

[16] V.R. Franceschi, P.A. Nakata, Calcium oxalate in plants: formation and func-tion, Ann. Rev. Plant. Biol. 56 (2005) 41e71.

[17] J.M. Dunwell, J.G. Gibbings, T. Mahmood, S.M. Saqlan Naqvi, Germin andgermin-like proteins: evolution, structure, and function, Crit. Rev. Plant Sci. 27(2008) 342e375.

C.V. Dias et al. / Plant Physiology and Biochemistry 49 (2011) 917e922922

[18] X. Dong, R. Ji, X. Guo, S.J. Foster, H. Chen, C. Dong, Y. Liu, Q. Hu, S. Liu,Expressing a gene encoding wheat oxalate oxidase enhances resistance toSclerotinia sclerotiorum in oilseed rape (Brassica napus), Planta 228 (2008)331e340.

[19] J. Banerjee, N. Das, P. Dey, M.K. Maiti, Transgenically expressed rice germin-like protein1 in tobacco causes hyper-accumulation of H2O2 and reinforce-ment of the cell wall components, Biochem. Biophys. Res. Commun. 402(2010) 637e643.

[20] K. Knecht, M. Seyffarth, C. Desel, T. Thurau, I. Sherameti, B. Lou, R. Oelmüller,D. Cai, Expression of BvGLP-1 encoding a germin-like protein from sugar beetin Arabidopsis thaliana leads to resistance against phytopathogenic fungi,MPMI 23 (2010) 446e457.

[21] E.R. Cober, S. Rioux, I. Rajcan, P.A. Donaldson, D.H. Simmonds, Partial resis-tance to white mold in a transgenic soybean line, Crop Sci. 43 (2003) 92e95.

[22] X. Li, D. Zhang, V.J. Lynch-Holm, T.W. Okita, V.R. Franceschi, Isolation ofa crystal matrix protein associated with calcium oxalate precipitation invacuoles of specialized cells, Plant Physiol. 133 (2003) 549e559.

[23] G. Wagner, F. Loewus, The biosynthesis of and (þ)-tartaric acid in Pelargoniumcrispum, Plant Physiol. 52 (1973) 651e654.

[24] G.L. Wheeler, M.A. Jones, N. Smirnoff, The biosynthetic pathway of vitamin Cin higher plants, Nature 393 (1998) 365e369.

[25] M.A. Green, S.C. Fry, Vitamin degradation in plant cells via enzymatichydrolysis of 4-O-oxalyl-L-threonate, Nature 433 (2005) 83e87.

[26] P.A. Nakata, M. McConn, Isolated Medicago truncatula mutants with increasedcalcium oxalate crystal accumulation have decreased ascorbic acid levels,Plant. Physiol. Biochem. 45 (2007) 216e220.

[27] K.S. Kim, M. Ji-Young, B.D. Martin, Oxalic acid is an elicitor of plant pro-grammed cell death during Sclerotinia sclerotiorum disease development, Mol.PlanteMicrobe. Interact. 21 (2008) 605e612.

[28] M.C.S. Rio, B.V. Oliveira, D.P.T. Tomazella, J.A.F. Silva, G.A.G. Pereira, Productionof calcium oxalate crystals by the basidiomycete Moniliophthora perniciosa,the causal agent of witches’ broom disease of cacao, Curr. Microbiol. 56 (2008)363e370.

[29] A.S. Gesteira, F. Micheli, N. Carels, A.C. Silva, K.P. Gramacho, I. Schuster,J.N.A. Macêdo, G.A.G. Pereira, J.C.M. Cascardo, Comparative analysis ofexpressed genes from cacao meristems infected by Moniliophthora perniciosa,Ann. Bot. 100 (2007) 129e140.

[30] G.A. Frias, L.H. Purdy, R.A. Schmidt, An inoculation method for evaluatingresistance of cacao to Crinipellis perniciosa, Plant Dis. 79 (1995) 787e791.

[31] H. Thordal-Christensen, Z.G. Zhang, Y.D. Wei, D.B. Collinge, Sub-cellularlocalization of H2O2 in plants. H2O2 accumulation in papillae and hypersen-sitive response during the barley-powdery mildew interaction, Plant J. 11(1997) 1187e1194.

[32] A. Gesteira, F. Micheli, C.F. Ferreira, J.C.M. Cascardo, Isolation and purification offunctional total RNA from different organs of cocoa tree during its interactionwith the pathogen Crinipellis perniciosa, Biotechniques 35 (1995) 494e500.