Minireview

Stress response, cell death and signalling: the many faces of reactiveoxygen species

Ramamurthy Mahalingam and Nina Fedoroff*

Biology Department and Life Sciences Consortium, Pennsylvania State University, University Park, PA 16802, USA*Corresponding author, e-mail:nvf1@psu.edu

Received 2 January 2003; revised 25 February 2003

Plants respond to pathogens and abiotic stresses by transientincreases in the production of reactive oxygen species (ROS)and ion fluxes, which activate both local programmed celldeath and systemic increases in stress- and pathogen-

resistance. The present essay explores the emerging comp-lexity of the multiple roles that ROS play in intra- andintercellular communication in both stressed and unstressedorganisms.

Introduction

Contemporary organisms live in an oxidizing environ-ment and generate highly reactive partially reduced oxy-gen species while converting light energy into chemicalenergy and chemical energy into biomass and motion.Common oxidants produced in organisms include reac-tive oxygen species (ROS), such as hydrogen peroxide(H2O2), superoxide (

*O2–), hydroxyl radical (OH–), and

singlet oxygen, as well as reactive sulphur and nitrogenspecies, particularly nitric oxide (NO), and lipid hydro-peroxides. The hazards of reactive oxygen species (ROS)have long been recognized, as has their role in diseaseand ageing in animals (Beal 2002, Melov 2002, Perryet al. 2002). Comprehensive overviews of the extensiveplant stress response literature are provided by a recentvolume on oxidative stress in plants (Inze and VanMontagu 2002), as well as a special issue of the Journalof Experimental Botany devoted to ROS and antioxidants(Smirnoff 2002). But although the concept of redox regu-lation is an old one in enzymology, there is a new andgrowing awareness that ROS and redox regulation arecentral to cell signalling and to both transcriptional andpost-transcriptional regulation (Sauer et al. 2001, Droge2002, Ermak and Davies 2002). The focus of the presentreview is on ROS in plants, but we will also refer torelevant studies in animals, because plants and animals

use analogous systems to monitor and react to oxidants(Inze and Van Montagu 2002, Martindale and Holbrook2002).

Oxidative stress and the oxidative burst

It is conventional wisdom that the oxidative stressresponse is triggered by an imbalance in the productionand metabolism of ROS. Almost paradoxically, how-ever, a variety of stress conditions, both biotic and abi-otic, trigger a transient, enzymatically mediated increasein ROS, predominantly *O2

– and H2O2, in plant andanimal cells. Initially termed the ‘oxidative burst’ or‘respiratory burst’, the transient output of ROS waslong believed to be responsible for the ability of phago-cytic cells to kill engulfed micro-organisms throughdirect ROS toxicity (Babior et al. 1973). This provideda widely accepted paradigm that was adopted by plantbiologists when a similar oxidative burst was discoveredin plant cells exposed to pathogens (Doke 1983, Mehdy1994, Doke et al. 1996).The oxidative burst in plant cells is a biphasic response

that extends over many hours after exposure to a patho-gen (Lamb and Dixon 1997). It comprises a primarypeak 1–2 h after infection, followed by a secondarypeak of greater magnitude 3–6 h after infection. The

PHYSIOLOGIA PLANTARUM119: 56–68. 2003 Copyright# Physiologia Plantarum 2003Printed in Denmark – all rights reserved ISSN 1399-3054

Abbreviations – ABA, abscisic acid; DAG, diacylglycerol; DPI, diphenylene iodonium; NO, nitric oxide; HR, hypersensitive response; IP3,inositol 1,4,5,-trisphosphate; JA, jasmonic acid; LOX, lipoxygenase; H2O2, hydrogen peroxide; NAADP, nicotinic acid adenine dinucleotide;PA, phosphatidic acid; PLC, phospholipase C; PLD, phospholipase D; RNS, reactive nitrogen species; ROS, reactive oxygen species; SA,salicylic acid; SAR, systemic acquired resistance; XOD, xanthine oxidase.

56 Physiol. Plant. 119, 2003

second phase is triggered by avirulent pathogens, whichdo not infect the plant, but instead evoke a localized celldeath program, termed the hypersensitive response(HR). The second phase is not observed with virulentpathogens, which are able to infect the plant (Lamb andDixon 1997, Scheel 2002). A biphasic oxidative burst isalso induced by the air pollutant ozone, considered to bean abiotic trigger of the pathogen defence response(Sandermann et al. 1998, Schraudner et al. 1998).Increased ROS production is also observed in responseto a variety of other abiotic stressors, including extremesof temperature, high light levels, water deficit, herbicides,cycloheximide, amines and air pollutants, such as SO2and NO2 (Allan and Fluhr 1997, Tenhaken and Rubel1998, Scheel 2002). Thus the endogenous production ofROS is a central feature of a generalized plant stressresponse.The emerging picture of the plant oxidative burst is

complex and its physiological significance is by no meansclear. Although there is substantial overlap between theresponses observed with pathogens and with abiotic elicit-ors, such as amines and the ROS derived from ozonegas, increasingly detailed studies show that they can bedistinguished enzymatically, as well as temporally andspatially (Overmyer et al. 2000, Scheel 2002). The enzym-atic sources of ROS are both extracellular and intra-cellular (Bolwell et al. 1995, Allan and Fluhr 1997,Pellinen et al. 1999, Bolwell et al. 2002). Major ROSsources are peroxidases and amine oxidases located incell walls, NADPH oxidase located in the plasma mem-brane, and intracellular oxidases and peroxidases inmitochondria, chloroplasts, peroxisomes and nuclei(Allan and Fluhr 1997, Bolwell and Wojtaszek 1997,Bowler and Fluhr 2000, Laurenzi et al. 2001, Bolwellet al. 2002, del Rio et al. 2002, Scheel 2002). Plantsproduce NO, a reactive nitrogen species (RNS), whichcan interact with ROS to form the highly reactive peroxy-nitrite anion, ONOO– (Neill et al. 2002). In addition,highly reactive lipid hydroperoxides are generated bydirect ROS lipid oxidation and by lipoxygenase activity(Jalloul et al. 2002, Porta and Rocha-Sosa 2002). BothNO and lipid hydroperoxides are important in elabora-tion of the defence response (Delledonne et al. 1998,Rusterucci et al. 1999, Delledonne et al. 2001).The magnitude and rapidity of the increase in ROS, as

well as whether its origin is intra- or extracellular,depends on the eliciting agent, the tissue and the organ-ism (Allan and Fluhr 1997, Bolwell et al. 2002). Further-more, the intracellular localization of the initial ROSburst can differ. For example, it was reported thatL-arginine-evoked ROS are concentrated at the peripheryof tobacco epidermal cells, while cryptogein, a smallfungal protein produced by Phytophthora cryptogeaelicitedROSpredominantly in chloroplasts and the nucleus(Allan and Fluhr 1997). The reasons for these differencesare not yet well-understood. However, it is already clearthat different ROS-producing enzymes are activated bydifferent input signals and that input signals are commu-nicated by different routes. Thus, the L-arginine-evoked

ROS burst in tobacco epidermal cells was reported to beabrogated by catalase and superoxide dismutase imply-ing that the primary response is due to activation ofextracellular enzymes, such as cell wall peroxidases(Allan and Fluhr 1997). By contrast, the ROS burstelicited by cryptogein in the same tissue was sensitive toinhibitors of signal transduction, including okadaic acid(a phosphatase 2 A inhibitor) and staurosporine (a proteinkinase C inhibitor). It was also sensitive to diphenyleneiodonium (DPI), an inhibitor of flavin-containing oxi-dases, among them NADPH oxidase and xanthineoxidase (Allan and Fluhr 1997). The foregoing observa-tions imply that there are at least two qualitatively dif-ferent mechanisms that trigger the oxidative burst: directenzyme activation and indirect activation triggered byreceptor-mediated signal transduction.

Receptor-mediated signalling

A plant’s response to a pathogen is based on its ability torecognize signature molecules produced by the pathogen.These molecules are called elicitors and are primarilyproteins and cell wall-derived oligosaccharides. In atype of host–pathogen interaction termed ‘gene-for-gene’, each race-specific elicitor encoded by an avirulence(Avr) gene in the pathogen is specifically recognized by aprotein encoded by a resistance (R) gene in the plant(Bonas and Lahaye 2002). Most, but not all, proteinsencoded by R genes recognize Avr proteins througheither extra- or intracellular leucine-rich repeat (LRR)domains and initiate signalling through kinase domainsor by interacting with kinases (Blumwald et al. 1998).Recent studies on the signal transduction pathway trig-gered by flagellin, a bacterial elicitor of a non-race spe-cific or innate defence response, have identified an LRRreceptor kinase, as well as members of the MAP kinasecascade that convey the signal to the transcriptionalapparatus (Asai et al. 2002).Although it has been known for some time that patho-

gen recognition by the interaction between an elicitor anda receptor results in the rapid activation of ion fluxes andan oxidative burst, the proteins and small molecules thattransduce these signals are not fully characterized. Earlyevidence for the involvement of a heterotrimeric G proteinwas quite indirect (Legendre et al. 1992, Vera-Estrella et al.1994, Beffa et al. 1995), but has received support frommore recent studies that link G protein signalling withactivation of both Ca12 channels and the membrane-bound NADPH oxidase in the plant defence response(Xing et al. 1997a, Aharon et al. 1998). The heterotrimericG-protein is located in the plasma membrane, as well asthe nucleus and may be in structures analogous to mam-malian lipid rafts (Peskan andOelmuller 2000, Peskan et al.2000). Although detailed information about how G pro-teins exert their regulatory effects on Ca12 channels is notyet available in plants, Ca12 channels in animals areknown to be regulated both by direct interaction with theGbg heterodimer and indirectly through secondmessengers(Dascal 2001).

Physiol. Plant. 119, 2003 57

The membrane-bound NADPH oxidase of mamma-lian neutrophils comprises two membrane-bound flavo-cytochrome subunits, gp91phox and p22phox, as well as acytosolic complex comprising three proteins designatedp47phox, p67phox, and p40phox, and the small GTPaseRac1/Rac2 (Forman and Torres 2001). Upon stimula-tion of neutrophils, the cytosolic components assembleon the membrane to form the active enzyme that usesNADPH to reduce O2 to

*O2–. Initial studies using

antibodies to mammalian NADPH oxidase proteins sug-gested that similar proteins were assembled on theplasma membranes of elicited tomato cells (Xing et al.1997b). But although two of the six recently identifiedArabidopsis homologs of the gp91phox catalytic compon-ent are required for production of ROS, there appear tobe no homologs of the p47phox or p67phox regulators inthe Arabidopsis genome (Torres et al. 1998; Torres et al.2002). As well, the plant catalytic subunit has anextended N-terminal region containing an EF-handcalcium-binding motif consistent with its observedregulation by calcium (Sagi & Fluhr 2001; Torres et al.1998). However, there is evidence that the regulatory linkbetween the heterotrimeric G protein and the NADPHoxidase is mediated by a Rac GTPase in plants, as it is inanimals (Suharsono et al. 2002). Rice mutants in a geneencoding a Ga subunit neither produce an oxidative burstin response to elicitation, nor do they exhibit an HR inresponse to pathogen infection. Expression of a constitu-tively active rice Rac1 gene restores elicitor-mediateddefence signalling, ROS production, and pathogenresistance (Suharsono et al. 2002).Receptor and G protein activation also stimulate the

oxidative burst indirectly through phospholipid secondmessengers. Phospholipase C (PLC) hydrolyses phosphati-dylinositol 4,5-bisphosphate to inositol 1,4,5,-trispho-sphate (IP3) and diacylglycerol (DAG) (Munnik et al.1998, Laxalt and Munnik 2002). IP3 stimulates releaseof calcium from intracellular stores in guard cells(MacRobbie 1998) and there is evidence that pathogensignalling stimulates IP3 production and intracellularcalcium release (Ortega and Perez 2001). Pathogen sig-nalling has also been reported to stimulate the phospho-lipase C-dependent production of phosphatidic acid (PA)via phosphorylation of DAG (Laxalt and Munnik 2002).PLC inhibitors neomycin and U73122 suppress the oxi-dative burst, whereas PA induces it (Sang et al. 2001,Laxalt and Munnik 2002).PA is also generated by phospholipase D (PLD),

which hydrolyses phophatidyl-choline to choline andPA. PLD is activated by the plant stress hormone abscisicacid (ABA), pathogens, osmotic stress and wounding(Laxalt and Munnik 2002). PA activates a kinase thatphosphorylates two cytoplasmic subunits of the neutro-phil NADPH oxidase (McPhail et al. 1999). Althoughsuch activity has not been identified in plants, MAPKcascade activation by PA has been reported (Lee et al.2001). There is also some evidence that phospholipasesA1 and A2, which hydrolyse phospholipids to produce alyso-phospholipid and a free fatty acid, are activated in

the pathogen defence response (Dhondt et al. 2000, Laxaltand Munnik 2002). Fatty acid oxidation by a lipoxy-genase (LOX) is the first step in the synthesis ofoxylipin second messengers, such as jasmonic acid (JA)(Farmer et al. 1998, Schaller 2001).

The ‘ROS signature’

The foregoing discussion shows that the oxidative burstis not a simple unitary response, but rather reflects theactivation of a spatially localized subset of the cellularenzymes that produce ROS in response to a specificstimulus. Because the ROS-producing enzymes arelocated in different cell compartments, their activationdepends on the nature of the triggering stimulus and theroute by which it is perceived. We suggest the uniquespatial and temporal pattern of ROS production bedesignated the ‘ROS signature.’ Fig. 1 shows a diagram-matic representation of the various intra- and extracel-lular sources of ROS that can contribute to such an ROSsignature. The potential complexity of an ROS signatureshould not be underestimated, because differences in thelocation, duration and intensity of the initial ROS burstcan, in turn, generate secondary signals that differ intheir chemical identity, subcellular localization, timing,and intensity, as well as their potential for intercellularpropagation.The challenge is to determine the subcellular location

and molecular identity of the ROS-producing enzymesactivated by a particular stimulus. This is complicatedboth by the multiplicity of enzymes that produce ROSand the paucity of means to discriminate among them, aswell as their ability to carry out multiple reactions.Although it is often assumed that inhibition by DPIuniquely identifies the membrane-bound NADPH oxi-dase, this is not the case (Baker et al. 1998, Li and Trush1998). Plant peroxidases, for example, oxidize NADHand NADPH to produce ROS and it was recentlyreported that this reaction is sensitive to DPI (Bakeret al. 1998). It should also be borne in mind that peroxi-dases are enzymes that are capable of multiple reactions(Chen et al. 2000). A soybean plasma membrane-boundNADH oxidase has also been described whose activitychanges to that of a protein disulphide isomerase inresponse to auxin (Chueh et al. 1997).

The ‘calcium signature’

Rapid influx of Ca21 is correlated with elicitor-mediatedinduction of the pathogen defence response and withactivation of the stress response by ROS and other abi-otic stressors (Price et al. 1994, Levine et al. 1996, Cha-ndra et al. 1997, Jabs et al. 1997, Clayton et al. 1999,Davies et al. 1999). The lag time, magnitude, peak time,intensity, and duration of the increase in [Ca21]cytdepend on the eliciting agent and can differ even betweentwo different pathogen-derived proteinaceous elicitorsthat bind to the same cellular protein (Bourque et al.1998, Lebrun-Garcia et al. 1999, Grant et al. 2000,

58 Physiol. Plant. 119, 2003

Lecourieux et al. 2002). The initial increase in Ca21 isfrom the extracellular medium and both ROS productionand other components of the defence response can beinhibited by calcium chelators in the extracellular medium(Lecourieux et al. 2002). Neomycin, an inhibitor ofphospholipase C and IP3-mediated release of Ca

21 frominternal stores, also changes the calcium signature(Lecourieux et al. 2002). This implies that Ca21 releasefrom internal stores and Ca21 influx from the extracel-lular medium contribute to the calcium signature.Because the H2O2-induced Ca

21 increase can be inhibitedby La31, a Ca21 channel inhibitor, the seeminglyparadoxical observation that exogenous ROS can inducea typical biphasic increase in [Ca21]cyt is likely to be dueto a redox feedback loop that modulates channel activity(Lecourieux et al. 2002). In animal cells, H2O2 has beenshown to activate the ryanodine-sensitive Ca21 releasechannel of the sarcoplasmic reticulum (Hamilton andReid 2000, Hara et al. 2002) and a recent report describesa Ca21-permeable cation channel that is activated by lowconcentrations of H2O2 and agents that produce ROSand RNS, leading to cell death (Hara et al. 2002). More-over, activation of NADPH oxidase has been reported toincrease the sensitivity of intracellular Ca21 stores to thesecond messenger IP3 (Hu et al. 2000).

Compartmentalization: what is a ‘signature’?

The results of a recent study on the effect of hyperosmo-tic stress on Fucus rhizoid cells illustrate the spatial andtemporal dimensions of the ROS and Ca21 signatures.Hyperosmotic stress elicits increases in both [Ca21]cytand ROS production that initiate at the growing cellapex and travel as waves through the cell. Based oninhibitor studies, the first ROS component arises by

activation of the NADPH oxidase at the rhizoid apexat sites of adhesion between the plasma membrane andcell wall (Coelho et al. 2002). The NADPH oxidaseproduces extracellular *O2

–, which is then rapidly dis-mutated to H2O2. The H2O2 activates Ca

21 channels atthe rhizoid apex, initiating a Ca21 wave that travelsthrough the cell, activating ROS production by mito-chondria. Curiously, only the peripheral ROS produc-tion is required for osmotic adaptation (Coelho et al.2002).The foregoing examples begin to reveal the rich com-

plexity of the ROS and calcium ‘signatures.’ The natureof the initiating signal and how and where it is receiveddetermine the subsequent cellular responses both quali-tatively and quantitatively. Because the plant NADPHoxidase is calcium-regulated, either directly or indirectlythrough a calmodulin-regulated NAD kinase (Pou DeCrescenzo et al. 2001, Sagi and Fluhr 2001) and becausecalcium channels are redox regulated (Hamilton andReid 2000, Hara et al. 2002, Lecourieux et al. 2002), aninitial perturbation of either signature can alter both.The various enzymes that generate ROS are spatiallylocalized, as are calcium stores with different properties(Bowler and Fluhr 2000, Lee 2000, Pani et al. 2001).Moreover, secondary signalling molecules, includingIP3, cyclic ADP ribose, nicotinic acid adenine dinucleo-tide phosphate (NAADP), and calcium itself can act onspecific calcium stores, amplifying calcium release(Galione and Churchill 2000, Lee 2000, Navazio et al.2000, Patel et al. 2001).Thus Ca21 and ROS might be viewed as the cell’s

‘common currency’, connecting specific cues in the exter-nal environment with the cell’s genetic and biochemicalnetworks to implement precise and appropriate physio-logical responses. As calcium-binding proteins are very

Fig. 1. Extra- and intracellular sources of ROS in plants. XOD, xanthine oxidase.

Physiol. Plant. 119, 2003 59

abundant and because ROS are both inherently short-lived and subject to rapid enzymatic conversion to lessreactive molecules, both signals operate over very shorttimes and distances required for communication withinindividual cells and between adjacent cells. But here isthe riddle of common currency: how does the buyer(signal) find the appropriate seller (response)? That is,what is the underlying molecular apparatus and how is itorganized in time and space?The emerging view of the cell as a highly structured

assembly (or perhaps network) of gel-like compartmentscomprising one to many proteins and associated mem-branous structures, each with specific functions, providesthe basis for a physical interpretation of ROS and Ca21

signatures (Pollack and Reitz 2001, Pollack 2001). Wesuggest that a ‘signature’ is the manifestation of thecombined outputs – in time and space – of the particularsubset of compartments activated in response to eachcue. A cue can be a chemical that directly activates anenzyme (or ion channel) or it can be one whose effect ismediated by a cell-surface receptor. Signatures are oftencharacterized by oscillatory changes in concentration ofCa21, ROS, or other metabolites, a subject consideredbelow. A common feature of ROS and Ca21 is that theycan modify protein structure directly to activate andinactivate proteins and channels. Changes in both ROSand [Ca21] can also be transmitted through other pro-teins, such as thioredoxin and calmodulin. In turn, thesesignals trigger a variety of more specialized responsesmediated by enzymes that either synthesize additionalmessengers or modify specific target proteins.

The oxidative burst: is it really a killer?

The localized programmed cell death of the HR is closelycorrelated with the oxidative burst and it was initiallyattributed to the toxicity of ROS (Levine et al. 1994,Wojtaszek 1997, Desikan et al. 1998). However, evidenceis accumulating that the connection between ROS andcell death is less direct and more complex than initiallyconceived (Hoeberichts and Woltering 2003). Cell deathis not always associated with a strong oxidative burstand there are elicitors that cause cell death, but do notevoke a strong oxidative burst (Yu et al 1998, Dorey et al.1999, Jabs 1999). The Arabidopsis dnd1 mutation in agene that encodes a cyclic nucleotide-gated ion channelin some measure uncouples cell death from the defenceresponse (Yu et al. 1998; Clough et al. 2000). Moreover,double mutants in both AtrbohD and AtrbohF genesencoding the Arabidopsis homologs of gp91phox subunitof the NADPH oxidase do not produce ROS, but exhibita residual HR (Torres et al. 2002).Several observations suggest that one function of ROS

in the plant stress and defence responses is that of anintercellular messenger. Diffusion of H2O2 from culturedtobacco cells treated with an elicitor into unelicited cellsactivates gene expression (Levine et al. 1994). Intracellu-lar ROS generated by exposing tobacco epidermal cellsloaded with rose bengal (4,5,6,7-tetrachloro-20,40,50,70-

tetraiodofluoroscein) to a focused light beam quicklytravel to neighbouring cells (Allan and Fluhr 1997).The ROS signal is sensitive to catalase, implying that itis H2O2 and can be detected in cells with no directsymplastic connection to the source cell over a distanceof several cell diameters (Allan and Fluhr 1997). Studieson how the local introduction of a pathogen leads to arapid systemic increase in pathogen resistance haveimplicated ROS as an essential component of the distrib-uted response (Alvarez et al. 1998). It was reported that alocal source of enzymatically produced H2O2 induced theaccumulation of H2O2 in small groups of cells at distantsites, often near veins, and sufficed for the developmentof systemic resistance. However, H2O2 is a relativelyshort-lived molecule, hence it appears likely that alonger-lived secondary signal mediates the systemicresponse. Results of recent experiments to distinguishdirect from indirect effects suggest that ROS induceproduction of the intercellular messenger, but are notthemselves the second messengers (Costet et al. 2002).

Second messengers

Although both ROS and Ca21 have been called ‘secondmessengers’, the foregoing observations suggest thatROS and Ca21 have rather different functions thanother molecules given the same designation (Dong1998, McAinsh and Hetherington 1998, Orozco-Cardenaset al. 2001). Among the latter are salicylic acid (SA),jasmonic acid (JA), and systemin, as well the classicalplant hormones ABA and ethylene (Dong 1998, Ryanand Pearce 2001, Kunkel and Brooks 2002). There isabundant evidence that SA accumulation is associatedwith the development of pathogen resistance and it haslong been postulated to be the signal that initiates sys-temic acquired resistance (SAR) (Gaffney et al. 1993,Chen et al. 1995, Lawton et al. 1995). However, severalobservations cast doubt on the systemic signalling func-tion of SA. Using grafts between wild-type tobacco plantsand plants expressing a bacterial salicylate hydroxylasegene (NahG) that encodes an enzyme that degrades SA,it was demonstrated that a signal for SAR could be com-municated from inoculated NahG-expressing leaves tografted wild-type leaves, but not from wild-type leaves tografted NahG-expressing leaves (Ryals et al. 1995). Thisimplies that SA could be a local cell-to-cell signal, but isnot likely to function as the systemic signal initiating thedevelopment of SAR.The observation that SA binds to and inhibits catalase

gave rise to the hypothesis that H2O2 is the direct inducerof defence gene expression (Conrath et al. 1995). How-ever, SA binds to a variety of iron-containing enzymesother than catalase, including lipoxidase and peroxidase(Ruffer et al. 1995). There is evidence that SA induces*O2

– and H2O2 through peroxidase-catalysed oxidationof SA to the SA*, which in turn generates *O2

–. by thedirect reduction of O2 (Kawano and Muto 2000). Loss ofthe SA-potentiated oxidative burst decreases resistanceto an avirulent pathogen (Mur et al. 2000). At the

60 Physiol. Plant. 119, 2003

cellular level, SA production appears to be necessary fordefence gene activation and for limiting the extent of celldeath in response to both avirulent pathogens and ozone,an abiotic stressor (Ryals et al. 1995, Dong 1998, 2001,Rao et al. 2002). Genetic resistance to pathogens is sup-pressed in plants that are SA-deficient either by virtue ofmutations or expression of a salicylate hydroxylasetransgene and such plants also exhibit enhanced tissuedamage in response to ozone (Ryals et al. 1995, Rao andDavis 1999).JA and ethylene are also produced in response to both

pathogens and ozone and are important determinants ofthe biological response (Sandermann et al. 1998, Kunkeland Brooks 2002). As noted earlier, JA is a fatty acidderivative and has been implicated in defence geneexpression in response to the peptide hormone systemin,induced by wounding, and to pathogen-derived oligosac-charide elicitors (Doares et al. 1995). Mutations thatinterfere with JA biosynthesis or perception increasepathogen susceptibility and increase ozone sensitivity(Overmyer et al. 2000, Kunkel and Brooks 2002).Ozone and pathogens induce ethylene biosynthesis byactivating expression of genes encoding ACC synthase,the rate-limiting enzyme of ethylene biosynthesis (Knoesteret al. 1995, Moeder et al. 2002, Rao et al. 2002). Inturn, ethylene production influences the plant defenceresponse, as judged by the altered disease susceptibilityand ozone-sensitivity of plants with mutations in ethyl-ene perception (Overmyer et al. 2000, Kunkel andBrooks 2002, Rao et al. 2002). Mutants that overproduceethylene show enhanced sensitivity to ozone, whereasmutants that cannot perceive ethylene are less sensitive(Overmyer et al. 2000). Ozone-induced ethylene bio-synthesis depends on SA and is not observed in NahGplants (Rao et al. 2002).The foregoing molecular signals are involved in the

spatial evolution of the initial small lesions that ariseeither on elicitation, pathogen infection or ozone treat-ment (Greenberg et al. 2000). The relationships amongthese signals are far from clear. However, the currentlevel of understanding suggests that SA and ROS aresignals that initiate lesions, and both SA and ethyleneare necessary for their propagation, whereas JA attenu-ates lesion propagation (Overmyer et al. 2000; Rao et al.2000). The observation that extracellular *O2–

is produced at the outward edges of lesions, suggests thata positive feedback loop involving ROS, SA and ethylenepropagates the lesion (Jabs et al. 1996, Overmyer et al.2000, Aviv et al. 2002).Insight into the relationship between lesion formation

and the biphasic oxidative burst is beginning to emergefrom studies on lesion development after ozone treat-ment (Lamb and Dixon 1997, Overmyer et al. 2000,Moeder et al. 2002, Scheel 2002). By contrast with locallesion induction by pathogens or elicitors, ozone treat-ment results in the uniform exposure of the entire plantto the initiating oxidative stress. Nonetheless, ozoneinduces ROS production in clusters of cells near thevasculature, just as is observed with the microbursts

reported after pathogen elicitation. Transcription of anumber of genes, including the ozone-inducible ACCsynthase gene, is bi-phasic, with an early peak atapproximately 1 h after ozone treatment and a secondlater peak which depends on ethylene evolution (Moederet al. 2002). Thus it appears likely that the second phaseof the oxidative burst, which is ethylene-induced, medi-ates cell death.Details of the temporal and spatial relationships

among the cells, signals and responses remain to beelucidated. In view of the known links between ethyleneand ABA signalling and synthesis (Fedoroff 2002a), it issurprising that few studies have addressed the role ofABA in the stress and defence responses (Audenaertet al. 2002). Moreover, the role of lipid signallingremains poorly investigated. Massive production of freepolyunsaturated fatty acid hydroperoxides has beenobserved during HR in response to cryptogein(Rusterucci et al. 1999). Expression of lipoxygenaseantisense constructs in transgenic tobacco plantssuppresses the HR to an avirulent pathogen (Rance et al.1998). The Arabidopsis EDS4 and PAD4 genes, whichencode lipase-like proteins, are also required for patho-gen resistance and affect the HR (Feys et al. 2001,Rusterucci et al. 2001). Thus it appears likely that there arelipid signalling molecules other than JA and they maytrigger cell death.

Phase transitions

The foregoing discussion reveals the complexity of thesignalling systems underlying plant stress and defenceresponses. However, it also raises questions about theprecise role of ROS in the progression of biochemicalevents that allows plants to resist pathogen attack andrespond to changes in environmental conditions. Beforeaddressing potential primary regulatory targets of ROS,we briefly discuss recent experiments that reveal somenew aspects of how ROS are produced and used. Thelong-standing concept that mammalian neutrophils pro-duce ROS to kill bacteria is being challenged. Neutro-phils deficient in their ability to produce proteases, butcapable of a normal oxidative burst, are unable todestroy bacteria (Reeves et al. 2002). The results ofthese studies suggest that the important consequence ofactivating the vacuolar membrane NADPH oxidase,which moves electrons across the membrane to reduceoxygen, is to raise the pH and ionic strength of thevacuole. This releases granule-bound proteases fromtheir anionic proteoglycan matrix, thereby activatingthem to kill and digest bacteria (Reeves et al. 2002).These striking observations open a new aspect of ROS

production: the potential to trigger transitions – or phasechanges – that can restructure cell compartments andwhole cells. Cryptogein, a 10-kDa peptide elicitorsecreted by the oomycete Phytophthora cryptogea,induces a rapid and extensive disruption of the micro-tubular network in tobacco cells (Binet et al. 2001).Moreover, ABA induction of stomatal cell closure

Physiol. Plant. 119, 2003 61

similarly requires ROS production, inactivation of Rac1GTPase, and involves massive cytoskeletal restructuring(Pei et al. 2000, Lemichez et al. 2001, Murata et al. 2001).Cytoskeletal changes in response to oxidants are wellknown and may be mediated by changes in the redoxstate of actin itself or they may be catalysed indirectly byredox-sensitive enzymes (Dalle-Donne et al. 2001).

Oscillations

Yet another dimension of mammalian neutrophils is theoscillatory release of ROS during cell migration.Although it was reported some time ago that extracellu-lar ROS production by migrating neutrophils is oscilla-tory (Kindzelskii et al. 1998), the recent application ofhigh-speed microscopy revealed that the neutrophil ispolarized and exhibits a travelling wave of NADPHthat moves parallel to the cell’s long axis (Kindzelskiiand Petty 2002). An ROS plume is released when thewave arrives at the lamellipodium, presumably because itdelivers substrate to the membrane-bound NADPH oxi-dase. The authors infer that the spatial and temporalproperties of the NADPH waves are transformed intothe pulsatile release of superoxide into the extracellularenvironment. The pulsatile release of ROS may alsoaccount for the previous observation of periodic mem-brane disruption in target tumour cells undergoing cyto-toxic destruction by neutrophils (Kindzelskii and Petty1999). However, the pulsatile release of ROS may also bean intercellular signal to activate other cells.Regular oscillations have been reported in the intra-

cellular concentrations of a variety of small molecules,including Ca21 ions, NADPH, ATP, ROS and glucose.A well-studied example is provided by the correlatedoscillations among intracellular oxygen consumption,intracellular free calcium and glucose consumption inpancreatic b-cells secreting insulin (Jung et al. 2000,Kennedy et al. 2002). The oscillations are set up by thepositive and negative feedback loops between Ca21, gly-colytic and respiratory enzymes and result in the pulsa-tile release of insulin. Thus a change in the externalglucose concentration stimulates increased respiration,increasing the ATP/ADP ratio. This increase, in turn,suppresses glycolysis, decreasing glucose consumption,blocking the KATP channel and depolarizing the cell. Asa consequence, Ca21 entry is stimulated, which enhancesrespiration and increases oxygen consumption by activ-ating mitochondrial dehydrogenases. The elevatedintracellular Ca21 concentration leads to a decrease inthe ATP/ADP ratio and cell repolarization, which inturn decreases oxygen consumption and inhibition ofglycolysis, initiating a new cycle (Jung et al. 2000,Kennedy et al. 2002).The information content of short-period oscillatory

processes in plants is just beginning to be explored,perhaps because techniques that permit the monitoringof oscillatory processes in single plant cells are not yetwell-developed (Allen and Schroeder 2001). The most

extensively characterized rapid oscillations in plant cellsare those involving intracellular [Ca21]. Using bothcalcium-sensitive dyes and a green fluorescent protein-basedcalcium sensor, stomatal guard cells have been shown toexhibit characteristic [Ca21]cyt oscillations in response tochanges in external calcium concentration ABA and H2O2(Allen et al. 2000).A number of molecular mechanisms have been postu-

lated to account for calcium oscillations in diverse typesof animal cells, and it appears likely that multiple inputscontribute to and modify the oscillations. The postulatedcontributors include agonist-stimulated dissociation ofthe Ga subunit from the Gbg complex, activation ofPLC and the consequent increase in IP3, which, in turn,stimulates calcium-induced Ca21 release (Kummer et al.2000). In some cases the instability is attributed to theproperties of the Ca21 channel itself and the autocataly-tic process of Ca21-induced Ca21 release, whereas inother cases it is attributed to the oscillatory productionof IP3 by a feedback loop involving desensitization of thereceptor by protein kinase C, or the self-enhanced acti-vation of the Ga subunit of the heterotrimeric G protein(Berridge et al. 2000, Kummer et al. 2000, Haberichteret al. 2001, Nash et al. 2001, 2002).The observation that the Ga subunit is redox regulated

suggests a possible feedback loop connecting theNADPH oxidase to the signalling complex (Nishidaet al. 2002). Another potential oscillatory link betweenthe signal receptor and ROS production is provided byobservations on Rac1, the small GTPase whose acti-vation by Gbg through the Rac exchange factor P-Rex1triggers the assembly and activation of the NADPHoxidase (Weiner 2002). The ROS produced by the acti-vated NADPH oxidase, in turn, stimulate the proteasome-mediated degradation of Rac1 (Kovacic et al. 2001).Finally, it is perhaps worth pointing out that some of

the classical single enzyme oscillators are plant oxidasesand peroxidases (Kummer et al. 1996, Morre and Morre1998, Morre et al. 1999). This includes the intensivelystudied oscillatory horseradish peroxidase–oxidase reac-tion, which uses molecular oxygen to oxidize NADH,dihydroxyfumaric acid and indole acetic acid and formsROS as intermediates (Kirkor and Scheeline 2000, Hauseret al. 2001, Olsen et al. 2001). Sustained oscillationsare observed in the presence of methylene blue and cer-tain phenolic or aromatic compounds (Hauser and Olsen1998, Olsen et al. 2001). Cell wall peroxidases promoteH2O2-dependent polymerization of hydroxycinnamylalcohols in lignin biosynthesis and rapid cross-linkingof cell wall proteins, in addition to the foregoingNADH, NADPH, and IAA oxidase reaction (Bestwicket al. 1998). It has been reported that a plant membrane-bound NADH oxidase that also functions as a proteindisulphide isomerase exhibits regular oscillatory behav-iour that can be entrained by light (Morre et al. 1999).Interestingly, a similar mammalian enzyme has beenshown to exhibit inverse oscillations of its oxidase anddisulphide isomerase activities (Chueh et al. 2002).It seems quite likely, although not yet experimentally

62 Physiol. Plant. 119, 2003

established, that such fundamental enzymatic oscillatorscan also drive calcium oscillations directly by controllingthe supply of NADPH and ROS, and indirectly by alter-ing the activity of disulphide oxidoreductases.

ROS: mediators and targets

While major categories of enzymes known to be redoxregulated are phosphatases and cysteine proteases, evi-dence is rapidly accumulating that structural changesdriven by the reduction and oxidation of disulphidebridges within and between proteins is as fundamentala regulatory principle as protein phosphorylation (Fig. 2).In some cases, ROS oxidize target proteins directly, par-ticularly peroxiredoxins and other peroxidases, but alsoincluding transcription factors (Aslund et al. 1999, Parket al. 2000, Delaunay et al. 2002, Goyer et al. 2002).However, most redox regulation is mediated by a familyof protein disulphide oxidoreductases, including thio-redoxins, peroxiredoxins, glutaredoxins, and protein disul-phide isomerases (Nishiyama et al. 2001, Jacquot et al.2002a, b, Rouhier et al. 2002). These are enzymes thatcommonly, although not always, contain pairs of reac-tive cysteines that function to create and reduce disul-phide bonds in other proteins. Thioredoxins are small(approximately 12kDa), ubiquitous proteins with disul-phide-reducing activity. They are oxidized either directlyby ROS or as a consequence of their ability to reducedisulphides on target proteins such as peroxiredoxins

(thioredoxin peroxidases). Thioredoxins are themselvesreduced by thioredoxin reductase, an NADPH-dependentenzyme.There is growing evidence that thioredoxins and simi-

lar proteins are enzymatic mediators of the regulatoryeffects of ROS at the biochemical, transcriptional andcellular levels (Nishiyama et al. 2001). Thioredoxin bindsto the p40phox subunit of the mammalian NADPH oxi-dase, suggesting direct redox regulation of the latter(Nishiyama et al. 1999). Stimuli such as UV irradiationpromote translocation of thioredoxin to the mammaliannucleus, where it activates stress transcription factorsNF-kB and AP-1 by enhancing DNA binding (Hirotaet al. 1999, Karimpour et al. 2002). Thioredoxin bindsdirectly to NF-kB p50, but interacts indirectly with AP-1through Redox factor-1 (Ref-1) (Karimpour et al. 2002).By contrast, thioredoxin suppresses degradation of theNF-kB inhibitor IkB in the cytoplasm (Hirota et al.1999). In mammalian cells, a variety of cytotoxic stresses,including H2O2 and tumour necrosis factor-a (TNF-a),activate the c-Jun N-terminal protein kinase family (JNKMAPK) (Kamata and Hirata 1999, Barr and Bogoyevitch2001). Apoptosis signal-regulating kinase (ASK) 1is a MAP kinase kinase kinase (MAPKKK) that acti-vates the JNK MAPK cascades (Ichijo et al. 1997).Reduced, but not oxidized, thioredoxin binds directlyto the N-terminus of ASK1, inhibiting its kinase activityas well as ASK1-dependent apoptosis (Saitoh et al.1998).

Fig. 2. Potential routes by which ROS can influence gene expression. (ROS, reactive oxygen species; MAPK, mitogen-activated protein kinase;MAPKK, MAPKkinase; MAPKKK, MAPKK kinase; TF, transcription factor, coTF, transcriptional cofactor).

Physiol. Plant. 119, 2003 63

Plants have distinct chloroplastic, mitochondrial andcytoplasmic redox regulatory systems (Jacquot et al.2002b, Rouhier et al. 2002). In chloroplasts, the iron-sulphur protein ferredoxin is photoreduced by photosys-tem I. The reducing power is transferred by ferredoxinthioredoxin reductase to thioredoxin, which then inter-acts with target enzymes (Jacquot et al. 2002a). There aredistinct cytosolic and mitochondrial thioredoxins inplants, and these are reduced by NADPH-dependentthioredoxin reductases (Rouhier et al. 2002). The targetsof plant thioredoxins are just beginning to be identifiedand are likely to include the same categories of proteins,including enzymes, signalling proteins, and transcriptionfactors, that have been identified in mammalian cells.Among the best-studied redox regulatory plant proteinsare several RNA-binding proteins that control the trans-lation or stability of specific chloroplastic mRNAs inresponse to light by way of the ferredoxin-thioredoxinsystem (Danon 2002, Fedoroff 2002b).

Summary

This brief essay brings us to the conclusion that ROS andredox regulation are far more integral to both intra- andintercellular communication than was apparent even adecade ago. Plant studies on ROS began with the detec-tion of a bi-phasic oxidative burst in plants exposed topathogens (Doke 1983, Lamb and Dixon 1997). Todaywe know that ROS production is a central aspect of howplants defend themselves against pathogens and abioticstress, but we do not yet know in molecular detail how itseffects are translated into the biochemical and transcrip-tional changes that kill some cells, while reinforcing theplant’s ability to withstand further stresses. We know, aswell, that the production of ROS is central to both ABA-and auxin-mediated physiological responses (Pei et al.2000; Joo et al. 2001; Murata et al. 2001; Rodriguezet al. 2002). Many important and exciting questionsremain for future research. For example, what are thedirect protein targets oxidized by ROS and the indirecttargets regulated by association with the redox-regulatedprimary targets? What transcription factors are redox-regulated and what are their target genes? What is thespatial and temporal organization of the ROS and Ca12

signatures? That is, how are the redox- and Ca12-regu-lated proteins compartmentalized in cells and how do thecompartments communicate chemically to translateexternal signals into changes in cell structure, viability,and function?

References

Aharon GS, Gelli A, Snedden WA, Blumwald E (1998) Activationof a plant plasma membrane Ca21 channel by TGa1, aheterotrimeric G protein a-subunit homologue. FEBS Lett424: 17–21

Allan AC, Fluhr R (1997) Two distinct sources of elicited reactiveoxygen species in tobacco epidermal cells. Plant Cell 9: 1559–1572

Allen GJ, Schroeder JI (2001) Combining genetics and cell biologyto crack the code of plant cell calcium signaling. Sci STKE 2001:RE13

Allen GJ, Chu SP, Schumacher K, Shimazaki CT, Vafeados D,Kemper A, Hawke SD, Tallman G, Tsien RY, Harper JF,Chory J, Schroder JI (2000) Alteration of stimulus-specificguard cell calcium oscillations and stomatal closing in Arabidop-sis det3 mutant. Science 289: 2338–2342

Alvarez ME, Pennell RI, Meijer PJ, Ishikawa A, Dixon RA, Lamb C(1998) Reactive oxygen intermediates mediate a systemicsignal network in the establishment of plant immunity. Cell 92:773–784

Asai T, Tena G, Plotnikova J, Willmann MR, Chiu WL, Gomez-Gomez L, Boller T, Ausubel FM, Sheen J (2002) MAP kinasesignalling cascade in Arabidopsis innate immunity. Nature 415:977–983

Aslund F, Zheng M, Beckwith J, Storz G (1999) Regulation of theOxyR transcription factor by hydrogen peroxide and the cel-lular thiol-disulfide status. Proc Natl Acad Sci USA 96:6161–6165

Audenaert K, De Meyer GB, Hofte MM (2002) Abscisic aciddetermines basal susceptibility of tomato to Botrytis cinereaand suppresses salicylic acid-dependent signaling mechanisms.Plant Physiol 128: 491–501

Aviv DH, Rusterucci C, Holt BF, Dietrich RA, Parker JE, DanglJL (2002) Runaway cell death, but not basal disease resistance.lsd1 is SA- and NIM1/NPR1-dependent. Plant J 29: 381–391

Babior BM, Kipnis RS, Curnutte JT (1973) Biological defencemechanisms: the production by leukocytes of superoxide, apotential bactericidal agent. J Clin Invest 52: 741–744

Baker CJ, Deahl K, Domek J, Orlandi EW (1998) Oxygen metab-olism in plant/bacteria interactions: effect of DPI on the pseudo-NAD(P)H oxidase activity of peroxidase. Biochem Biophys ResCommun 252: 461–464

Barr RK, Bogoyevitch MA (2001) The c-Jun N-terminal proteinkinase family of mitogen-activated protein kinases (JNKMAPKs). Int J Biochem Cell Biol 33: 1047–1063

Beal MF (2002) Oxidatively modified proteins in aging and disease.Free Radic Biol Med 32: 797–803

Beffa R, Szell M, Meuwly P, Pay A, Vogeli-Lange R, Metraux JP,Neuhaus G, Meins F Jr, Nagy F (1995) Cholera toxin elevatespathogen resistance and induces pathogenesis-related geneexpression in tobacco. EMBO J 14: 5753–5761

Berridge MJ, Lipp P, Bootman MD (2000) The versatility anduniversality of calcium signalling. Nat Rev Mol Cell Biol 1:11–21

Bestwick CS, Brown IR, Mansfield JW (1998) Localized changes inperoxidase activity accompany hydrogen peroxide generationduring the development of a nonhost hypersensitive reaction inlettuce. Plant Physiol 118: 1067–1078

Binet M, Humbert C, Lecourieux D, Vantard M, Pugin A (2001)Disruption of microtubular cytoskeleton induced by cryptogein,an elicitor of hypersensitive response in tobacco cells. PlantPhysiol 125: 564–572

Blumwald E, Aharon GS, Lam BC-H (1998) Early signal transduc-tion pathways in plant –pathogen interactions. Reviews 3:342–346

Bolwell GP, Wojtaszek P (1997) Mechanisms for the generation ofreactive oxygen species in plant defence – a broad perspective.Physiol Mol Plant Path 51: 347–366

Bolwell GP, Bindschedler LV, Blee KA, Butt VS, Davies DR,Gardner SL, Gerrish C, Minibayeva F (2002) The apoplasticoxidative burst in response to biotic stress in plants: a three-component system. J Exp Bot 53: 1367–1376

Bolwell GP, Butt VS, Davies DR, Zimmerlin A (1995) The origin ofthe oxidative burst in plants. Free Radic Res 23: 517–532

Bonas U, Lahaye T (2002) Plant disease resistance triggered bypathogen-derived molecules: refined models of specific recogni-tion. Curr Opin Microbiol 5: 44–50

Bourque S, Ponchet M, Binet MN, Ricci P, Pugin A, Lebrun-Garcia A (1998) Comparison of binding properties and earlybiological effects of elicitins in tobacco cells. Plant Physiol 118:1317–1326

Bowler C, Fluhr R (2000) The role of calcium and activated oxy-gens as signals for controlling cross-tolerance. Trends Plant Sci5: 241–246

Chandra S, Stennis M, Low PS (1997) Measurement of Ca21 fluxesduring elicitation of the oxidative burst in aequorin-transformedtobacco cells. J Biol Chem 272: 28274–28280

64 Physiol. Plant. 119, 2003

Chen JW, Dodia C, Feinstein SI, Jain MK, Fisher AB (2000) 1-Cysperoxiredoxin, a bifunctional enzyme with glutathione peroxi-dase and phospholipase A2 activities. J Biol Chem 275:28421–28427

Chen Z,Malamy J, Henning J, Conrath U, Sanchez-Casas P, Silva H,Ricigliano J, Klessig DK (1995) Induction, modification, andtransduction of the salicylic acid signal in plant defenseresponses. Proc Natl Acad Sci USA 92: 4134–4137

Chueh PJ, Morre DM, Morre DJ (2002) A site-directed mutagen-esis analysis of tNOX functional domains. Biochim BiophysActa 1594: 74–83

Chueh PJ, Morre DM, Penel C, DeHahn T, Morre DJ (1997) Thehormone-responsive NADH oxidase of the plant plasma mem-brane has properties of a NADH: protein disulfide reductase. JBiol Chem 272: 11221–11227

Clayton H, Knight MR, Knight H, McAinsh MR, Hetherington AM(1999) Dissection of the ozone-induced calcium signature. Plant J17: 575–579

Clough SJ, Fengler KA, Yu IC, Lippok B, Smith RK, Jr., Bent AF(2000) The Arabidopsis dnd1 ‘‘defense, no death’’ gene encodes amutated cyclic nucleotidegated ion channel. Proc Natl Acad SciUSA 97: 9323–9328

Coelho SM, Taylor AR, Ryan KP, Sousa-Pinto I, Brown MT,Brownlee C (2002) Spatiotemporal patterning of reactive oxygenproduction and Ca (21) wave propagation in Fucus rhizoidcells. Plant Cell 14: 2369–2381

Conrath U, Chen Z, Ricigliano JR, Klessig DF (1995) Two inducersof plant defense responses, 2,6-dichloroisonicotinec acid andsalicylic acid, inhibit catalase activity in tobacco. Proc NatlAcad Sci USA 92: 7143–7147

Costet L, Dorey S, Fritig B, Kauffmann S (2002) A pharmaco-logical approach to test the diffusible signal activity of reactiveoxygen intermediates in elicitor-treated tobacco leaves. PlantCell Physiol 43: 91–98

Dalle-Donne I, Rossi R, Milzani A, Di Simplicio P, Colombo R(2001) The actin cytoskeleton response to oxidants: from smallheat shock protein phosphorylation to changes in the redoxstate of actin itself. Free Radic Biol Med 31: 1624–1632

Danon A (2002) Redox reactions of regulatory proteins: do kineticspromote specificity? Trends Biochem Sci 27: 197–203

Dascal N (2001) Ion-channel regulation by G proteins. TrendsEndocrinol Metab 12: 391–398

Davies E, Shimps B, Brown K, Stankovic B (1999) Gravity, stress,calcium and gene expression. J Gravit Physiol 6: P21–P22

Delaunay A, Pflieger D, Barrault MB, Vinh J, Toledano MB (2002)A thiol peroxidase is an H2O2 receptor and redox-transducer ingene activation. Cell 111: 471–481

Delledonne M, Xia Y, Dixon RA, Lamb C (1998) Nitric oxidefunctions as a signal in plant disease resistance. Nature 394:585–588

Delledonne M, Zeier J, Marocco A, Lamb C (2001) Signal interac-tions between nitric oxide and reactive oxygen intermediates inthe plant hypersensitive disease resistance response. Proc NatlAcad Sci USA 98: 13454–13459

Desikan R, Reynolds A, Hancock JT, Neill SJ (1998) Harpin andhydrogen peroxide both initiate programmed cell death but havedifferential effects on defence gene expression in Arabidopsissuspension cultures. Biochem J 330: 115–120

Dhondt S, Geoffroy P, Stelmach BA, Legrand M, Heitz T (2000)Soluble phospholipase A2 activity is induced before oxylipinaccumulation in tobacco mosaic virus-infected tobacco leavesand is contributed by patatin-like enzymes. Plant J 23: 431–440

Doares SH, Syrovets T, Weiler EW, Ryan CA (1995) Oligogalac-turonides and chitosan activate plant defensive genes throughthe octadecanoid pathway. Proc Natl Acad Sci USA 92:4095–4098

Doke N (1983) Involvement of superoxide anion generation in thehypersensitive response of potato-tuber tissues to infection withan incompatible race of Phytophthora infestans and to thehyphal wall components. Physiol Plant Pathol 23: 345–357

Doke N, Miura Y, Sanchez LM, Park HJ, Noritake T, Yoshioka H,Kawakita K (1996) The oxidative burst protects plants againstpathogen attack: mechanism and role as an emergency signal forplant bio-defence – a review. Gene 179: 45–51

Dong X (1998) SA, JA, ethylene, and disease resistance in plants.Curr Opin Plant Biol 1: 316–323

Dong X (2001) Genetic dissection of systematic acquired resistance.Curr Opin Plant Biol 4: 309–314

Dorey S, Kopp M, Geoffroy P, Fritig B, Kauffmann S (1999)Hydrogen peroxide from the oxidative burst is neither necessarynor sufficient for hypersensitive cell death induction, phenylala-nine ammonia lyase stimulation, salicylic acid accumulation, orscopoletin consumption in cultured tobacco cells treated withelicitin. Plant Physiol 121: 163–172

Droge W (2002) Free radicals in the physiological control of cellfunction. Physiol Rev 82: 47–95

Ermak G, Davies KJ (2002) Calcium and oxidative stress: from cellsignaling to cell death. Mol Immunol 38: 713–721

Farmer EE, Weber H, Vollenweider S (1998) Fatty acid signaling inArabidopsis. Planta 206: 167–174

Fedoroff NV (2002a) ‘Cross-talk’ in abscisic acid signaling. SciSTKE 2002: RE10

Fedoroff NV (2002b) RNA-binding proteins in plants: the tip of aniceberg? Curr Opin Plant Biol 5: 452–459

Feys BJ, Moisan LJ, Newman MA, Parker JE (2001) Direct inter-action between the Arabidopsis disease resistance signaling pro-teins, EDS1 and PAD4. EMBO J 20: 5400–5411

Forman HJ, Torres M (2001) Signaling by the respiratory burst inmacrophages. IUBMB Life 51: 365–371

Gaffney T, Friedrich L, Vernooij B, Negrotto D, Nye G, Uknes S,Ward E, Kessmann H, Ryals J (1993) Requirement of salicyclicacid for the induction of systemic acquired resistance. Science261: 754–756

Galione A, Churchill GC (2000) Cyclic ADP ribose as a calcium-mobilizing messenger. Sci STKE 2000: PE1

Goyer A, Haslekas C, Miginiac-MaslowM, Klein U, Le Marechal P,Jacquot JP, Decottignies P (2002) Isolation and characteriza-tion of a thioredoxin-dependent peroxidase from Chlamydomonasreinhardtii. Eur J Biochem 269: 272–282

Grant M, Brown I, Adams S, Knight M, Ainslie A, Mansfield J(2000) The RPM1 plant disease resistance gene facilitates arapid and sustained increase in cytosolic calcium that is neces-sary for the oxidative burst and hypersensitive cell death. Plant J23: 441–450

Greenberg JT, Silverman FP, Liang H (2000) Uncoupling salicylicacid-dependent cell death and defense-related responses fromdisease resistance in the Arabidopsis mutant acd5. Genetics156: 341–350

Haberichter T, Marhl M, Heinrich R (2001) Birhythmicity, tri-rhythmicity and chaos in bursting calcium oscillations. BiophysChem 90: 17–30

Hamilton SL, Reid MB (2000) RyR1 modulation by oxidation andcalmodulin. Antioxid Redox Signal 2: 41–45

Hara Y, Wakamori M, Ishii M, Maeno E, Nishida M, Yoshida T,Yamada H, Shimizu S, Mori E, Kudoh J, Shimizu N, Kurose H,Okada Y, Imoto K, Mori Y (2002) LTRPC2 Ca21-permeablechannel activated by changes in redox status confers suscept-ibility to cell death. Mol Cell 9: 163–173

Hauser MJ, Kummer U, Larsen AZ, Olsen LF (2001) Oscillatorydynamics protect enzymes and possibly cells against toxic sub-stances Faraday Discuss 120: 215–227

Hauser MJ, Olsen LF (1998) The role of naturally occurring phe-nols in inducing oscillations in the peroxidase-oxidase reaction.Biochemistry 37: 2458–2469

Hirota K, Murata M, Sachi Y, Nakamura H, Takeuchi J, Mori K,Yodoi J (1999) Distinct roles of thioredoxin in the cyto-plasm and in the nucleus. A two-step mechanism of redoxregulation of transcription factor NF-kappaB. J Biol Chem274: 27891–27897

Hoeberichts FA, Woltering EJ (2003) Multiple mediators of plantprogrammed cell death: Interplay of conserved cell deathmechanisms and plant-specific regulators.. Bioessays 25: 47–57

Hu Q, Zheng G, Zweier JL, Deshpande S, Irani K, Ziegelstein RC(2000) NADPH oxidase activation increases the sensitivity ofintracellular Ca21 stores to inositol 1,4,5-trisphosphate inhuman endothelial cells. J Biol Chem 275: 15749–15757

Ichijo H, Nishida E, Irie K, ten Dijke P, Saitoh M, Moriguchi T,Takagi M, Matsumoto K, Miyazono K, Gotoh Y (1997) Induc-tion of apoptosis by ASK1, a mammalian MAPKKK that acti-vates SAPK/JNK and p38 signaling pathways. Science 275: 90–94

Inze D, Van Montagu M (eds) (2002) Oxidative Stress in Plants.Taylor & Francis, London.

Physiol. Plant. 119, 2003 65

Jabs T (1999) Reactive oxygen intermediates as mediators of pro-grammed cell death in plants and animals. Biochem Pharmacol57: 231–245

Jabs T, Dietrich RA, Dangl JL (1996) Initiation of runaway celldeath in an Arabidopsis mutant by extracellular superoxide.Science 273: 1853–1856

Jabs T, Tschope M, Colling C, Hahlbrock K, Scheel D (1997)Elicitor-stimulated ion fluxes and O2- from the oxidative burstare essential components in triggering defense gene activationand phytoalexin synthesis in parsley. Proc Natl Acad Sci USA94: 4800–4805

Jacquot JP, Gelhaye E, Rouhier N, Corbier C, Didierjean C, Aubry A(2002a) Thioredoxins and related proteins in photosyntheticorganisms: molecular basis for thiol dependent regulation. Bio-chem Pharmacol 64: 1065–1069

Jacquot JP, Rouhier N, Gelhaye E (2002b) Redox control bydithiol-disulfide exchange in plants. I. The chloroplastic systems.Ann NY Acad Sci 973: 508–519

Jalloul A, Montillet JL, Assigbetse K, Agnel JP, Delannoy E,Triantaphylides C, Daniel JF, Marmey P, Geiger JP, Nicole M(2002) Lipid peroxidation in cotton: Xanthomonas interactionsand the role of lipoxygenases during the hypersensitive reaction.Plant J 32: 1–12

Joo JH, Bae YS, Lee JS (2001) Role of auxin-induced reactiveoxygen species in root gravitropism. Plant Physiol 126:1055–1060

Jung SK, Kauri LM, Qian WJ, Kennedy RT (2000) Correlatedoscillations in glucose consumption, oxygen consumption, andintracellular free Ca (21) in single islets of Langerhans. J BiolChem 275: 6642–6650

Kamata H, Hirata H (1999) Redox regulation of cellular signalling.Cell Signal 11: 1–14

Karimpour S, Lou J, Lin LL, Rene LM, Lagunas L, Ma X, Karra S,Bradbury CM, Markovina S, Goswami PC, et al. (2002)Thioredoxin reductase regulates AP-1 activity as well as thio-redoxin nuclear localization via active cysteines in response toionizing radiation. Oncogene 21: 6317–6327

Kawano T, Muto S (2000) Mechanism of peroxidase actions forsalicylic acid-induced generation of active oxygen species and anincrease in cytosolic calcium in tobacco cell suspension culture. JExp Bot 51: 685–693

Kennedy RT, Kauri LM, Dahlgren GM, Jung SK (2002) Metabolicoscillations in beta-cells. Diabetes 51: S152–S161

Kindzelskii AL, Petty HR (1999) Early membrane rupture eventsduring neutrophil-mediated antibody-dependent tumor cellcytolysis. J Immunol 162: 3188–3192

Kindzelskii AL, Petty HR (2002) Apparent role of traveling metab-olic waves in oxidant release by living neutrophils. Proc NatlAcad Sci USA 99: 9207–9212

Kindzelskii AL, Zhou MJ, Haugland RP, Boxer LA, Petty HR(1998) Oscillatory pericellular proteolysis and oxidant deposi-tion during neutrophil locomotion. Biophys J 74: 90–97

Kirkor ES, Scheeline A (2000) Nicotinamide adenine dinucleotidespecies in the horseradish peroxidase-oxidase oscillator. Eur JBiochem 267: 5014–5022

Knoester M, Bol JF, Vanloon LC, Linthorst H (1995) Virus-induced gene expression for enzymes of ethylene biosynthesisin hypersensitively reacting tobacco. Mol Plant Microbe Inter-act 8: 177–180

Kovacic HN, Irani K, Goldschmidt-Clermont PJ (2001) Redoxregulation of human Rac1 stability by the proteasome inhuman aortic endothelial cells. J Biol Chem 276: 45856–45861

Kummer U, Olsen LF, Dixon CJ, Green AK, Bornberg-Bauer E,Baier G (2000) Switching from simple to complex oscillations incalcium signaling. Biophys J 79: 1188–1195

Kummer U, Valeur KR, Baier G, Wegmann K, Olsen LF (1996)Oscillations in the peroxidase-oxidase reaction: a comparison ofdifferent peroxidases. Biochim Biophys Acta 1289: 397–403

Kunkel BN, Brooks DM (2002) Cross talk between signaling path-ways in pathogen defense. Curr Opin Plant Biol 5: 325–331

Lamb D, Dixon RA (1997) The oxidative burst in plant diseaseresistance. Annu Rev Plant Physiol Plant Mol Biol 48: 251–275

Laurenzi M, Tipping AJ, Marcus SE, Knox JP, Federico R,Angelini R, McPherson MJ (2001) Analysis of the distributionof copper amine oxidase in cell walls of legume seedlings. Planta214: 37–45

Lawton K, Weymann K, Friedrich L, Vernooij B, Uknes S, Ryals J(1995) Systemic acquired resistance in Arabidopsis requires sali-cylic acid but not ethylene. Mol Plant Microbe Interact 8:863–870

Laxalt AM, Munnik T (2002) Phospholipid signalling in plantdefence. Curr Opin Plant Biol 5: 332–338

Lebrun-Garcia A, Bourque S, Binet MN, Ouaked F, Wendehenne D,Chiltz A, Schaffner A, Pugin A (1999) Involvement ofplasma membrane proteins in plant defense responses. Analysisof the cryptogein signal transduction in tobacco. Biochimie 81:663–668

Lecourieux D, Mazars C, Pauly N, Ranjeva R, Pugin A (2002)Analysis and effects of cytosolic free calcium increases inresponse to elicitors in Nicotiana plumbaginifolia cells. PlantCell 14: 2627–2641

Lee HC (2000) Multiple calcium stores: separate but interacting. SciSTKE 2000: PE1

Lee S, Hirt H, Lee Y (2001) Phosphatidic acid activates a wound-activated MAPK in Glycine max. Plant J 26: 479–486

Legendre L, Heinstein PF, Low PS (1992) Evidence for parti-cipation of GTP-binding proteins in elicitation of the rapidoxidative burst in cultured soybean cells. J Biol Chem 267:20140–20147

Lemichez E, Wu Y, Sanchez JP, Mettouchi A, Mathur J, Chua NH(2001) Inactivation of AtRac1 by abscisic acid is essential forstomatal closure. Genes Dev 15: 1808–1816

Levine A, Pennell RI, Alvarez ME, Palmer R, Lamb C (1996)Calcium-mediated apoptosis in a plant hypersensitive diseaseresistance response. Curr Biol 6: 427–437

Levine A, Tenhaken R, Dixon R, Lamb C (1994) H2O2 from theoxidative burst orchestrates the plant hypersensitive diseaseresistance response. Cell 79: 583–593

Li Y, Trush MA (1998) Diphenyleneiodonium, an NAD (P) Hoxidase inhibitor, also potently inhibits mitochondrial reactiveoxygen species production. Biochem Biophys Res Commun 253:295–299

MacRobbie EA (1998) Signal transduction and ion channels inguard cells. Philos Trans R Soc Lond B Biol Sci 353: 1475–1488

Martindale JL, Holbrook NJ (2002) Cellular response to oxidativestress: signaling for suicide and survival. J Cell Physiol 192: 1–15

McAinsh MR, Hetherington AM (1998) Encoding specificity inCa21 signalling systems. Trends Plant Sci 3: 32–36

McPhail LC, Waite KA, Regier DS, Nixon JB, Qualliotine-Mann D,Zhang WX, Wallin R, Sergeant S (1999) A novel protein kinasetarget for the lipid second messenger phosphatidic acid. BiochimBiophys Acta 1439: 277–290

Mehdy MC (1994) Active oxygen species in plant defense againstpathogens. Plant Physiol 105: 467–472

Melov S (2002) Animal models of oxidative stress, aging, andtherapeutic antioxidant interventions. Int J Biochem Cell Biol34: 1395–1400

Moeder W, Barry CS, Tauriainen AA, Betz C, Tuomainen J,Utriainen M, Grierson D, Sandermann H, Langebartels C,Kangasjarvi J (2002) Ethylene synthesis regulated by biphasicinduction of 1-aminocyclopropane-1-carboxylic acid synthaseand 1-aminocyclopropane-1-carboxylic acid oxidase genes isrequired for hydrogen peroxide accumulation and cell death inozone-exposed tomato. Plant Physiol 130: 1918–1926

Morre DJ, Morre DM (1998) NADH oxidase activity of soybeanplasma membranes oscillates with a temperature compensatedperiod of 24 min. Plant J 16: 277–284

Morre DJ, Morre DM, Penel C, Greppin H (1999) NADH oxidaseperiodicity of spinach leaves synchronized by light. Int J PlantSci 160: 855–860

Munnik T, Irvine RF, Musgrave A (1998) Phospholipid signallingin plants. Biochim Biophys Acta 1389: 222–272

Mur LA, Brown IR, Darby RM, Bestwick CS, Bi YM,Mansfield JW,Draper J (2000) A loss of resistance to avirulent bacterialpathogens in tobacco is associated with the attenuation of asalicylic acid-potentiated oxidative burst. Plant J 23: 609–621

Murata Y, Pei ZM, Mori IC, Schroeder J (2001) Abscisic acidactivation of plasma membrane Ca (21) channels in guardcells requires cytosolic NAD (P) H and is differentially disruptedupstream and downstream of reactive oxygen species productionin abi1–1 and abi2–1 protein phosphatase 2C mutants. PlantCell 13: 2513–2523

66 Physiol. Plant. 119, 2003

Nash MS, Schell MJ, Atkinson PJ, Johnston NR, Nahorski SR,Challiss RA (2002) Determinants of metabotropic glutamatereceptor-5-mediated Ca21 and inositol 1,4,5-trisphosphateoscillation frequency. Receptor density versus agonist concen-tration. J Biol Chem 277: 35947–35960

Nash MS, Young KW, Challiss RA, Nahorski SR (2001) Receptor-specific messenger oscillation. Nature 413: 381–382

Navazio L, Bewell MA, Siddiqua A, Dickinson GD, Galione A,Sanders D (2000) Calcium release from the endoplasmic reticulumof higher plants elicited by the NADP metabolite nicotinic acidadenine dinucleotide phosphate. Proc Natl Acad Sci USA 97:8693–8698

Neill SJ, Desikan R, Clarke A, Hurst RD, Hancock JT (2002)Hydrogen peroxide and nitric oxide as signalling molecules inplants. J Exp Bot 53: 1237–1247

NishidaM, ScheyKL,TakagaharaS,KontaniK,KatadaT,UranoY,Nagano T, Nagao T, Kurose H (2002) Activation mechanismof Gi and Go by reactive oxygen species. J Biol Chem 277:9036–9042

Nishiyama A, Masutani H, Nakamura H, Nishinaka Y, Yodoi J(2001) Redox regulation by thioredoxin and thioredoxin-bindingproteins. IUBMB Life 52: 29–33

Nishiyama A, Ohno T, Iwata S, Matsui M, Hirota K, Masutani H,Nakamura H, Yodoi J (1999) Demonstration of the interaction ofthioredoxin with p40phox, a phagocyte oxidase component, usinga yeast two-hybrid system. Immunol Lett 68: 155–159

Olsen LF, Lunding A, Lauritsen FR, Allegra M (2001) Melatoninactivates the peroxidase-oxidase reaction and promotes oscilla-tions. Biochem Biophys Res Commun 284: 1071–1076

Orozco-Cardenas M, Narvaez-Vasquez J, Ryan C (2001) Hydrogenperoxide acts as a second messenger for the induction of defensegenes in tomato plants in response to wounding, systemin, andmethyl jasmonate. Plant Cell 13: 179–191

Ortega X, Perez LM (2001) Participation of the phosphoinositidemetabolism in the hypersensitive response of Citrus limonagainst Alternaria alternata. Biol Res 34: 43–50

Overmyer K, Tuominen H, Kettunen R, Betz C, Langebartels C,Sandermann H Jr, Kangasjarvi J (2000) Ozone-sensitive Arabi-dopsis rcd1 mutant reveals opposite roles for ethylene and jas-monate signaling pathways in regulating superoxide-dependentcell death. Plant Cell 12: 1849–1862

Pani G, Bedogni B, Colavitti R, Anzevino R, Borrello S, Galeotti T(2001) Cell compartmentalization in redox signaling. IUBMBLife 52: 7–16

Park SG, Cha MK, Jeong W, Kim IH (2000) Distinct physiologicalfunctions of thiol peroxidase isoenzymes in Saccharomyces cer-evisiae. J Biol Chem 275: 5723–5732

Patel S, Churchill GC, Galione A (2001) Coordination of Ca21signalling by NAADP. Trends Biochem Sci 26: 482–489

Pei ZM, Murata Y, Benning G, Thomine S, Klusener B, Allen GJ,Grill E, Schroeder JI (2000) Calcium channels activated byhydrogen peroxide mediate abscisic acid signalling in guardcells. Nature 406: 731–734

Pellinen R, Palva T, Kangasjarvi J (1999) Subcellular localization ofozone-induced hydrogen peroxide production in birch (Betulapendula) leaf cells. Plant J 20: 349–356

PerryG,NunomuraA,HiraiK,ZhuX,PrezM,Avila J,CastellaniRJ,Atwood CS, Aliev G, Sayre LMet al. (2002) Is oxidativedamage the fundamental pathogenic mechanism of Alzheimer’sand other neurodegenerative diseases? Free Radic Biol Med 33:1475–1479

Peskan T, Oelmuller R (2000) Heterotrimeric G-protein beta-subunitis localized in the plasma membrane and nuclei of tobacco leaves.Plant Mol Biol 42: 915–922

Peskan T, Westermann M, Oelmuller R (2000) Identification oflow-density Triton X-100-insoluble plasma membrane micro-domains in higher plants. Eur J Biochem 267: 6989–6995

Pollack GH (2001) Is the cell a gel – and why does it matter? Jpn JPhysiol 51: 649–660

Pollack GH, Reitz FB (2001) Phase transitions and molecularmotion in the cell. Cell Mol Biol 47: 885–900

Porta H, Rocha-Sosa M (2002) Plant lipoxygenases. Physiologicaland molecular features. Plant Physiol 130: 15–21

Pou De Crescenzo MA, Gallais S, Leon A, Laval-Martin DL (2001)Tween-20 activates and solubilizes the mitochondrial

membrane-bound, calmodulin dependent NAD1 kinase ofAvena sativa L. J Membr Biol 182: 135–146

Price AH, Taylor A, Ripley SJ, Griffiths A, Trewavas AJ, KnightMR(1994) Oxidative signals in tobacco increase cytosolic calcium.Plant Cell 6: 1301–1310

Rance I, Fournier J, Esquerre-Tugaye MT (1998) The incompatibleinteraction between Phytophthora parsitica var. nicotianae race 0and tobacco is suppressed in transgenic plants expressing anti-sense lipoxygenase sequences. Proc Natl Acad Sci USA 95:6554–6559

Rao MV, Davis KR (1999) Ozone-induced cell death occurs via twodistinct mechanisms in Arabidopsis: the role of salicylic acid.Plant J 17: 603–614

Rao MV, Lee H, Creelman RA, Mullet JE, Davis KR (2000)Jasmonic acid signaling modulates ozone-induced hypersensitivecell death. Plant Cell 12: 1633–1646

Rao MV, Lee HI, Davis KR (2002) Ozone-induced ethylene pro-duction is dependent on salicylic acid, and both salicylic acidand ethylene act in concert to regulate ozone-induced cell death.Plant J 32: 447–456

Reeves EP, Lu H, Jacobs HL, Messina CG, Bolsover S, Gabella G,Potma EO, Warley A, Roes J, Segal AW (2002) Killing activityof neutrophils is mediated through activation of proteases byK1 flux. Nature 416: 291–297

del Rio LA, Corpas FJ, Sandalio LM, Palma JM, Gomez M,Barroso JB (2002) Reactive oxygen species, antioxidant systemsand nitric oxide in peroxisomes. J Exp Bot 53: 1255–1272

Rodriguez AA, Grunberg KA, Taleisnik EL (2002) Reactive oxygenspecies in the elongation zone of maize leaves are necessary forleaf extension. Plant Physiol 129: 1627–1632

Rouhier N, Gelhaye E, Jacquot JP (2002) Redox control by dithiol-disulfide exchange in plants. II. The cytosolic and mitochondrialsystems. Ann NY Acad Sci 973: 520–528

Ruffer M, Steipe B, Zenk MH (1995) Evidence against specificbinding of salicylic acid to plant catalase. FEBS Lett 377:175–180

Rusterucci C, Aviv DH, Holt BF 3rd, Dangl JL, Parker JE (2001)The disease resistance signaling components EDS1 and PAD4are essential regulators of the cell death pathway controlled byLSD1 in Arabidopsis. Plant Cell 13: 2211–2224

Rusterucci C, Montillet JL, Agnel JP, Battesti C, Alonso B, Knoll A,Bessoule JJ, Etienne P, Suty L, Blein JP, Triantaphylides C(1999) Involvement of lipoxygenase-dependent production offatty acid hydroperoxides in the development of the hypersensi-tive cell death induced by cryptogein on tobacco leaves. J BiolChem 274: 36446–36455

Ryals J, Lawton KA, Delaney TP, Friedrich L, Kessmann H,Neuenschwander U, Uknes S, Vernooij B, Weymann K (1995)Signal transduction in systemic acquired resistance. Proc NatlAcad Sci USA 92: 4202–4205

Ryan CA, Pearce G (2001) Polypeptide hormones. Plant Physiol125: 65–68

Sagi M, Fluhr R (2001) Superoxide production by plant homolo-gues of the gp91 (phox) NADPH oxidase. Modulation of acti-vity by calcium and by tobacco mosaic virus infection. PlantPhysiol 126: 1281–1290

Saitoh M, Nishitoh H, Fujii M, Takeda K, Tobiume K, Sawada Y,Kawabata M, Miyazono K, Ichijo H (1998) Mammalian thio-redoxin is a direct inhibitor of apoptosis signal-regulating kinase(ASK) 1. EMBO J 17: 2596–2606

Sandermann HJ, Ernst E, Heller W, Langebartels C (1998) Ozone:an abiotic elicitor of plant defence reactions. Trends Plant Sci 3:47–49

Sang Y, Cui D, Wang X (2001) Phospholipase D and phosphatidicacid-mediated generation of superoxide in Arabidopsis. PlantPhysiol 126: 1449–1458

Sauer H, Wartenberg M, Hescheler J (2001) Reactive oxygen spe-cies as intracellular messengers during cell growth and differen-tiation. Cell Physiol Biochem 11: 173–186

Schaller F (2001) Enzymes of the biosynthesis of octadecanoid-derived signalling molecules. J Exp Bot 52: 11–23

Scheel D (2002) Oxidative burst and the role of reactive oxygenspecies in plant–pathogen interactions. In: Inze D, VanMontagu M (eds) Oxidative Stress in Plants. Taylor & Francis,New York, pp. 137–153

Physiol. Plant. 119, 2003 67

Schraudner M, Moeder W, Wiese C, Van Camp W, Inze D,Langebartels C, Sandermann H Jr (1998) Ozone-induced oxidativeburst in the ozone biomonitor plant, tobacco Bel W3. Plant J 16:235–245

Smirnoff N (2002) Antioxidants and reactive oxygen species inplants. J Exp Bot 53: Special issue. no. 372

Suharsono U, Fujisawa Y, Kawasaki T, Iwasaki Y, Satoh H,Shimamoto K (2002) The heterotrimeric G protein alpha sub-unit acts upstream of the small GTPase Rac in disease resistanceof rice. Proc Natl Acad Sci USA 99: 13307–13312

Tenhaken R, Rubel C (1998) Induction of alkalinization and anoxidative burst by low doses of cycloheximide in soybean cells.Planta 206: 666–672

Torres MA, Dangl JL, Jones JD (2002) Arabidopsis gp91phoxhomologues AtrbohDandAtrbohF are required for accumulationof reactive oxygen intermediates in the plant defense response.Proc Natl Acad Sci USA 99: 517–522

Vera-EstrellaR,BarklaBJ,HigginsVJ,BlumwaldE (1994)Plantdefenseresponse to fungal pathogens. II.G-protein-mediated changes in hostplasma membrane redox reactions. Plant Physiol 106: 97–102

Weiner OD (2002) Rac activation: P-Rex1 – a convergence point forPIP (3) and Gbetagamma? Curr Biol 12: R429–R431

Wojtaszek P (1997) Oxidative burst: an early plant response topathogen infection. Biochem J 322: 681–692

Xing T, Higgins VJ, Blumwald E (1997a) Identification of G pro-teins mediating fungal elicitor-induced dephosphorylation ofhost plasma membrane H1-ATPase. J Exp Bot 48: 229–237

Xing T, Higgins VJ, Blumwald E (1997b) Race-specific elicitors ofCladosporium fulvum promote translocation of cytosolic compon-ents of NADPH oxidase to the plasma membrane of tomato cells.Plant Cell 9: 249–259

Yu IC, Parker J, Bent AF (1998) Gene-for-gene disease resistancewithout the hypersensitive response in Arabidopsis dnd1 mutant.Proc Natl Acad Sci USA 95: 7819–7824

Edited by E.-M. Aro

68 Physiol. Plant. 119, 2003