The role of the mitochondrion in plant responses to biotic stress

  • Published on

  • View

  • Download


Physiologia Plantarum 129: 253266. 2007 Copyright Physiologia Plantarum 2007, ISSN 0031-9317REVIEWThe role of the mitochondrion in plant responses tobiotic stressSasan Amirsadeghi, Christine A. Robson and Greg C. Vanlerberghe*Department of Life Sciences and Department of Cell and Systems Biology, University of Toronto, Scarborough, 1265 Military Trail, Toronto,ON M1C 1A4, CanadaCorrespondence*Corresponding author,e-mail: gregv@utsc.utoronto.caReceived 24 April 2006; revised 23 May 2006doi: 10.1111/j.1399-3054.2006.00775.xRecent studies suggest that the plant mitochondrion may play a role duringbiotic stress responses, such as those occurring during incompatible plantpathogen interactions. There are indications that signal molecules or pathwaysinitiated by such interactions may directly or indirectly target mitochondrialcomponents and that an important consequence of this targeting is an earlydisruption of mitochondrial homeostasis, resulting in an increased generationof mitochondrial reactive oxygen species (mROS). These mROS may theninitiate further mitochondrial dysfunction and further mROS generation ina self-amplifying manner. The mROS, as well as the graded dysfunction ofthe mitochondrion may act as cellular signals that initiate graded cellularresponses ranging from defense gene induction to initiation of programmedcell death. However, these events may be attenuated by the uniquecomponents of the plant electron transport chain that act to substitute fordysfunctional components, dampen mROS generation or facilitate in definingthe cellular level of ROS and antioxidant defense systems.IntroductionUpon recognition of a pathogen, plants mount a resis-tance response meant to cease pathogen growth anddisease development (Dangl and Jones 2001, Greenbergand Yao 2004, Lam et al. 2001). The resistance responsecan include activation of local and systemic defenses(e.g. expression of pathogenesis-related proteins) andinduction of a localized plant cell death at the site ofinfection called the hypersensitive response (HR). TheHR is a form of programmed cell death (PCD) and sharessome molecular and biochemical similarities withanimal apoptosis.Salicylic acid (SA), nitric oxide (NO) and reactiveoxygen species (ROS) (particularly H2O2) increase inabundance following pathogen recognition and each areimportant signaling molecules that promote and co-ordinate defense and HR responses (Alvarez 2000,Delledonne 2005, Laloi et al. 2004, Neill et al. 2002,Torres and Dangl 2005, Wendehenne et al. 2004). Theincrease in ROS (the so-called oxidative burst) involvesactivation of a plasma membrane-localized nicotinamideadenine dinucleotide phosphate (NADPH) oxidase.During the HR, this is accompanied by an active down-regulation of ROS-scavenging systems to further promoteROS accumulation (Mittler et al. 1998, Vacca et al. 2004).There are also complex synergistic (and possibly antag-onistic) interactions between SA, NO and ROS that definethe responses to biotic stress (Delledonne 2005).Abbreviations AA, antimycin A; ANT, adenine nucleotide translocator; AOX, alternative oxidase; BA, bongkrekic acid; CsA,cyclosporin A; cyt, cytochrome;DCm,mitochondrial transmembrane potential; DPI, diphenylene iodonium; ETC, electron transportchain; GDC, glycine decarboxylase; HR, hypersensitive response; IMM, innermitochondrial membrane; IMS, intermembrane space;mROS, mitochondrial reactive oxygen species; NO, nitric oxide; OMM, outer mitochondrial membrane; PCD, programmed celldeath; PPIX, protoporphyrin IX; PTP, permeability transition pore; ROS, reactive oxygen species; SA, salicylic acid; VDAC, voltage-dependent anion channel.Physiol. Plant. 129, 2007 253It is hypothesized that plant mitochondria act in theperception of biotic stress and take part in initiatingresponses such as the HR (Jones 2000, Lam et al. 2001). Inpart, this hypothesis derives from studies of animalapoptosis, where mitochondria play an active role (seereviews by Bratton and Cohen 2001, Crompton 1999,Kuwana and Newmeyer 2003, Ly et al. 2003, Newmeyerand Ferguson-Miller 2003, van Loo et al. 2002). Animalapoptosis involves activation of an aspartate-specificcysteine protease (caspase) cascade. Activation isachieved by the release of mitochondrial intermembranespace (IMS) proteins, in particular the electron transportchain (ETC) component cytochrome (cyt) c, to the cytosol.Cyt c then combines with other cytosolic components toform a caspase-activating complex. The caspase cascadeacts to amplify the original death-inducing signal andparticipates in the ordered disassembly of the cell. Cyt crelease is tightly regulated: antiapoptotic Bcl-2 familymembers present on the outer mitochondrial membrane(OMM) act to prevent cyt c release, whereas proapoptoticBcl-2 members can translocate from cytosol to the OMMand promote cyt c release.The mechanism by which IMS proteins are released tothe cytosol during animal apoptosis remains a topic ofdebate (Ly et al. 2003). Potential mechanisms arebroadly divided into three types: (1) the inner mitochon-drial membrane (IMM) experiences a large increase inpermeability because of opening of the permeabilitytransition pore (PTP). The PTP resides at contact sitesbetween the inner and outer membranes and itscore components include the IMM-localized adeninenucleotide translocator (ANT), the OMM-localizedvoltage-dependent anion channel (VDAC) and thematrix-localized cyclophilin-D. Pore opening results ina loss of mitochondrial transmembrane potential (DCm),which is followed by an influx of water and solutes to thematrix. This causes matrix swelling and selective ruptureof the OMM (because of its smaller surface area incomparison to the IMM), allowing the release of IMSproteins. Cyclosporin A (CsA) and bongkrekic acid (BA)are pharmacological inhibitors of PTP opening, actingby interaction with cyclophilin-D or ANT, respectively.A key requirement for pore opening is the accumulationof Ca21 in the mitochondrial matrix and susceptibility toCa21-induced opening is influenced by numerous otheraspects of mitochondrial status (Crompton 1999). Also,the pro- and antiapoptotic proteins may act by pro-moting or inhibiting PTPopening; (2) proteins residing inand/or recruited to the OMM can produce a pore thatallows release of IMS proteins to the cytosol. VDAC, aswell as proapoptotic proteins (e.g. Bax) may becomponents of this pore, whereas antiapoptotic proteins(e.g. Bcl-2) may inhibit pore formation; (3) the VDACcloses in response to death stimuli and because VDACand ANT coordinately shuttle adenosine diphosphate(ADP) into the matrix in exchange for adenosinetriphosphate (ATP), this closure depletes matrix ADP.This leads to an initial increase in DCm that promotesenhanced generation of ROS by the ETC (see below).These factors damage the IMM, leading to an influx ofsolutes and water, followed by swelling and rupture ofthe OMM.A distinct feature of plant mitochondria is the presenceof several unique ETC components beside those com-ponents associated with the usual cyt pathway (thatconsists of Complexes IIV and cyt c). Besides Complex I(the rotenone-sensitive NADH dehydrogenase oxidizingmatrix NADH), the IMM contains alternative rotenone-resistant NAD(P)H dehydrogenases (Finnegan et al. 2004,Rasmusson et al. 2004). These include both internalenzymes oxidizing matrix NAD(P)H and externalenzymes that oxidize NAD(P)H on the external side ofthe IMM. The alternative dehydrogenases reduce theenergy yield of respiration because they are non-protonpumping and bypass the proton-pumping Complex I.Several alternative NAD(P)H dehydrogenases possessEF-hand motifs for Ca21 binding, consistent with theobservation that their activity is modulated by Ca21.The IMM also contains an additional terminal oxidase(beside Complex IV or cyt oxidase) called alternativeoxidase (AOX) that catalyzes the oxidation of ubiquinoneand reduction of O2 to H2O (Finnegan et al. 2004). AOXalso reduces the energy yield of respiration because it isnon-proton pumping and bypasses proton-pumpingComplexes III and IV.Mitochondrial electron transport is associated with thegeneration of ROS such as superoxide and H2O2, whichare referred to in this review specifically as mitochondrialROS (mROS). Because ROS can damage macromole-cules, their cellular levels are managed through avoid-ance and scavenging mechanisms (Mittler et al. 2004). Asin animals, Complexes I and III likely represent theprimary sites of mROS generation (Mller 2001). Therelative importance of these two sites of mROS generationand the factors influencing their rates of mROS pro-duction are largely unknown but an important gener-alization is that mROS formation increases as the ETCbecomes more highly reduced. mROS generation byisolated mitochondria is therefore increased under ADP-limiting conditions that increase DCm and decreased byuncouplers that dissipate DCm. mROS formation is alsoincreased by inhibition of specific sites in the ETC suchas inhibition of Complex III by antimycin A (AA) or inhibi-tion of Complex I by rotenone. These inhibitors presum-ably promote mROS formation by promoting overreductionof specific ETC components (Mller 2001).254 Physiol. Plant. 129, 2007The alternative dehydrogenases and AOX may impactthe rate of mROS production. By accepting electronsfrom ubiquinone, AOX may prevent overreduction atComplex I and/or III. This route of electron transport couldbe important in dampening mROS formation underconditions in which cyt pathway components havesuffered stress-induced damage or, because AOX respi-ration is less tightly coupled to ATP production, underconditions in which ADP availability is limiting. Sucha role for AOX is supported by the finding that transgeniccells lacking AOX have more ROS emanating from themitochondrion (Maxwell et al. 1999). How the alterna-tive NAD(P)H dehydrogenases impact ROS generationis unknown. On the one hand, they may themselvesrepresent sites of ROS generation. Alternatively, they mayact to dampen ROS generation because (1) their activitywill bypass Complex I, a known ROS producer and (2)unlike Complex I, their activity will not contributeto DCm.Below, we review recent literature investigating thepotential role of plant mitochondria in biotic stressresponses. Fig. 1 is a summary of the main questionsbeing addressed. We propose some working models toaid further research in this area.Recent studies suggest that plantmitochondriamaybe a target of biotic stressBeside other well-studied signaling roles for SA duringbiotic stress (see Introduction), it has recently beensuggested that SA may directly impact mitochondria. Itwas shown that SA disrupts mitochondrial function ina concentration-dependent manner in tobacco suspen-sion cells (Norman et al. 2004). At low concentrations, itacted as an uncoupler, whereas at higher concentrationsit strongly inhibited electron flow. These effects were seenin both whole cells and isolated mitochondria andprovide a rationale for studies showing that SA coulddramatically inhibit ATP synthesis by tobacco cells (Xieand Chen 1999). It may also provide a rationale for whySA is able to induce AOX because AOX expressionappears to increase in response to disruptions inrespiratory homeostasis induced by diverse means(Finnegan et al. 2004). Norman et al. (2004) found thatSA inhibited electron flow upstream of the ubiquinonepool, perhaps by acting as a quinone analog interactingwith Complex I or II. Significantly, the concentrations ofSA required to induce these dramatic effects are withinthe range often used by investigators when examiningeffects of externally supplied SA. A key unresolvedquestion is whether endogenous localized concentra-tions of SA that accompany pathogen infection aresufficient to impact mitochondrial function. If they are,it opens up the possibility that some signaling functionsof SA act via effects on the mitochondrion.Norman et al. (2004) also found that AOX expressioncorrelated with the ability of SA to disrupt mitochon-drial function. Low concentrations of SA caused onlytransitory increases in cellular SA and this correlatedwell with both transitory mitochondrial dysfunction andtransitory increases in AOX expression. Hence, AOXmay represent an excellent reporter gene to evaluatewhether mitochondrial dysfunction is occurring duringbiotic stress. Several studies suggest that this is the case(see later). For example, AOX was amongst the earlyresponse genes induced in Arabidopsis during bacterialinfection (Lacomme and Roby 1999). AOX inductionwas transient (as expected for the increase in SA) andspecific to an avirulent interaction (as are increasesin SA).Interestingly, recent work with animal mitochondriashows that SA interacts directly with Complex I, causingan increase in Complex Igenerated ROS, which thencontributes to a permeability transition, cyt c release andapoptosis (Battaglia et al. 2005). If SA targets plantmitochondria in a similar fashion, it could play a role inthe early generation of mROS noted in recent studies (seebelow).Biotic stressProgrammedcell deathMitochondrialfunctionDefenseresponses????SignalingpathwaysFig. 1. A framework for investigating the role of plant mitochondria inbiotic stress. The following are the key questions being addressed in thisreview and illustrated in this figure: (1) Do any signaling molecules orpathways initiated by biotic stress impact mitochondrial function? (2)What changes occur in mitochondrial function? (3) Does the mitochon-drion play an active role in programmed cell death events such as thehypersensitive response? (4) Is the induction of any defense responses tobiotic stress dependent upon mitochondrial events?Physiol. Plant. 129, 2007 255Another signal molecule during biotic stress is NO,which along with SA and ROS, has been shown topromote the HR (see Introduction). In animals, NO isa modulator of mitochondrial-mediated apoptosis, in partbecause it causes a strong reversible inhibition of cytoxidase (Vieira and Kroemer 2003). Plant cyt oxidase issimilarly sensitive to NO but whether the physiologicalNO concentrations generated during plantpathogeninteractions are sufficient to inhibit cyt oxidase andwhether such inhibition contributes to defense responsesor the HR remains unknown. An important factor in thisregard may be the cellular source of NO. Animals havea mitochondrial-localized NO synthase. The situation inplants has been less clear but a recent publication hasidentified a NO synthase localizing to mitochondria (Guoand Crawford 2005). Under some conditions, the plantETC may also generate NO from nitrite (Planchet et al.2005). These studies provide potential means by whichNO could be generated in close proximity to cyt oxidase,hence perturbing mitochondrial function.An important set of virulence factors in pathogenicfungi is the so-called host-selective toxins that interactwith host molecules to cause plant cell death andcontribute to disease development. One such toxin,victorin, was shown to bind to and inhibit mitochondrialglycine decarboxylase (GDC), suggesting that GDCinhibition acted to promote cell death (Curtis andWolpert 2002). Victorin treatment of oat leaves resultedin a loss of DCm, followed by an ability of victorin to gainaccess to the mitochondrial matrix. This was interpretedto indicate that a permeability transition had occurredand that victorin used the PTP to gain access to matrixGDC. However, more recent results suggest that celldeath precedes access of victorin to the cell interior andthat victorin likely interacts with a cell surface protein toinitiate defense responses and cell death (Curtis andWolpert 2004, Tada et al. 2005). In this respect, thevirulence of the toxin may reside in its ability to elicita plant PCD pathway. These results shed doubt on theimportance of victorin-induced GDC inhibition in pro-moting cell death, but they do not preclude a role for themitochondrion in this cell death. In particular, Yao et al.(2002) have shown that victorin induces a burst of mROSpreceding death (see below).Ceramides are lipids that act as important secondmessengers in animals, where the balance betweenceramides and their phosphorylated derivatives mayregulate apoptosis. Animal studies indicate that ceramidecan cause a direct inhibition of Complex III, which, bypromoting mROS generation, initiates apoptosis (Gudzet al. 1997, Quillet-Mary et al. 1997). Interestingly, anArabidopsis mutant defective in ceramide kinase (andhence accumulating ceramide) shows excessive PCD inresponse to bacterial infection (Liang et al. 2006). It willbe interesting to examine whether this enhanced celldeath is because of ceramide targeting of the ETC.In summary, a number of molecules commonlyassociated with biotic stress may have a direct impacton ETC components such as Complexes I, III and IV. Asdiscussed next, a common consequence of this targetingmay be an increase in mROS.Recent studies suggest that an increase inmROS formation is an early consequence ofbiotic stressEarlier studies showed that intracellular sources of ROSmight contribute to the pathogen-induced oxidative burst(e.g. Allan and Fluhr 1997, Naton et al. 1996) and a reviewby Bolwell and Wojtaszek (1997) suggested a need toinvestigate whether the mitochondrion represented sucha source. A few recent studies have now directly addressedthis question by using ROS-sensitive fluorescent dyes andother imaging techniques to localize ROS generation invivo and in response to pathogens or their elicitors.Harpins are virulence factors produced by bacterialpathogens such as Pseudomonas syringae. Application ofpurified harpin to plant tissue can elicit a rapid HR-likecell death and some studies have examined the impactof such harpin treatments on mitochondria. By doublestaining Arabidopsis cell cultures with both a mitochon-drial-specific dye and a ROS-indicating dye, it was shownthat a large and early ROS burst associated with harpintreatment emanated specifically from the mitochondrion,suggesting the ETC as the likely source of ROS (Krause andDurner 2004). This burst of mROS was associated witha decline in DCm and cellular ATP levels and theappearance of cytosol-localized cyt c. All these eventspreceded PCD by several hours. The results are consistentwith those of another study in which harpin was shown todramatically inhibit ATP synthesis in tobacco cell cultures(Xie and Chen 2000). That study found that the earlyharpin-induced burst of ROS could be completelyinhibited by diphenylene iodonium (DPI), a findingusually interpreted to indicate that ROS production isoccurring via the DPI-sensitive NADPH oxidase. How-ever, DPI is also a potent inhibitor of Complex I (Mller2001). Hence, another interpretation of the DPI resultcould be that ROS is being generated by the mitochon-drion in response to harpin and that this ROS generationcan be dampened by DPI inhibition of Complex I. Thestudy of Xie and Chen (2000) also found that harpintreatment dramatically reduced the in vivo capacity forcyt pathway electron transport downstream of ubiqui-none. This would be consistent with a loss of cyt c fromthe mitochondrion, although this was not examined.256 Physiol. Plant. 129, 2007The above studies show that harpin has a rapid anddramatic impact on mitochondria, an interesting obser-vation in light of the recent proposal that most P. syringaevirulence factors likely function by targeting the plasmamembrane, chloroplast or mitochondrion of host cells(Greenberg and Vinatzer 2003).Greenberg and colleagues have studied mitochondrialevents associated with HR induction by P. syringae aswell as PCD induced by protoporphyrin IX (PPIX) or bylight treatment of theArabidopsis accelerated cell death 2(ACD2) mutant. ACD2 encodes a protein that attenuatesPCD, probably by sequestering or metabolizing porphy-rin-related molecules (such as PPIX) that can be photo-activated, leading to the production of ROS. Interestingly,the localization of ACD2 shifts from being largelychloroplastic to including the mitochondrion duringPCD-inducing treatments. Yao and Greenberg (2006)reported that a very early event (1.5 h) associated withdeath-inducing treatment of wild-type or ACD2 plantswas a burst of mROS, localized using ROS-sensitivefluorescent dyes. This was followed slightly later by a lossof DCm (quantified using flow cytometry) that, if blockedby CsA or ROS scavengers, was able to attenuate the PCD(Yao and Greenberg 2006, Yao et al. 2004). These elegantstudies provide the most convincing data to date thatmitochondrial events precede and contribute towardplant PCD.In another interesting study,DCm and mROS generationof camptothecin-treated and digitonin-permeabilized pro-toplasts were monitored by flow cytometry (Weir et al.2003). This study also found an early (1.5 h) burst inmROS and this corresponded closely with an increase ofDCm. This was then followed slightly later by a decreasein both these parameters. The initial increase in DCm(similar to that reported in an early study by Naton et al.1996) is of particular interest. It is in keeping with animalmodels in which impaired ATP/ADP exchange betweenthe cytosol and matrix (perhaps because of VDACclosure) promotes an initial increase in DCm that, bypromoting overreduction of the ETC, promotes mROSgeneration and mitochondrial dysfunction. The decreasedexpression of ANT during heat shock or senescenceassociated PCD of Arabidopsis cells provides another hintthat impaired ATP/ADPexchange may be an early event inPCD (Swidzinski et al. 2002). In another study, victorin wasshown to elicit a very rapid (30 min) increase in mROS(Yao et al. 2002). In this case, localization of the ROS wasbased on a cytochemical assay that showed H2O2eruptions at specific sites on the OMM.The above studies indicate that increased mROS is anearly event that clearly precedes PCD and likely alsoprecedes other documented mitochondrial events suchas loss of DCm and cyt c release (see later). As well, theresults suggest that the mROS released is obligatory toPCD in that, in some cases, it was shown that scavengingof the ROS attenuated PCD. We suggest that the earlyburst of mROS being noted in these studies is because ofa disruption of metabolic homeostasis in the mitochon-drion, possibly because of molecules (such as thosedescribed in the previous section) that target the ETC.Also, we suggest that an important consequence of thismROS burst will be a self-amplifying cycle in which theincreased mROS leads to mitochondrial damage, result-ing in further increases in mROS and further damage. Theculmination of these events will be the catastrophicmitochondrial dysfunction associated with changes in thepermeability or integrity of the mitochondrial membranes(see later). This hypothesis is outlined in Fig. 2.Several studies have documented the sensitivity ofmitochondria (particularly components of energy metab-olism) to oxidative stress, suggesting that ROS accumu-lation can promote damage and dysfunction (Bartoli et al.2004, Kristensen et al. 2004, Sweetlove et al. 2002, Tayloret al. 2002). Some of the identified components thatappear particularly susceptible to oxidative stress includeaconitase, GDC, ATP synthase, cyt c and VDAC. Asoutlined more later, the self-amplifying cycle of mROSgeneration and mitochondrial dysfunction may be animportant feature promoting PCD.There is also evidence that the ROS-scavengingcapacity of the mitochondrion is modulated in responseto pathogen infection. In particular, increases in mito-chondrial superoxide-scavenging capacity combinedwith decreases in the H2O2-scavenging components ofthe organelle were seen during Botrytis cinerea infectionof tomato leaves and it was hypothesized that this couldpromote accumulation of mitochondrial H2O2 (Kuzniakand Skodowska 2004). Such results imply an activemechanism to ensure accumulation of specific ROSspecies at the mitochondrion.Recent studies suggest thatmitochondria doplay an active role in plant PCDA possible role of plant mitochondria in PCD wasindicated by studies showing that when pro- or antiapop-totic animal proteins such as Bax or Bcl-2 were expressedin plants, they were able to, respectively, promote orinhibit PCD (Lam et al. 2001). Plants lack clear homologsof these proteins and so the functional relevance of theseobservations remains speculative. However, the studiesdid emphasize that manipulation of components at theOMM impacted PCD, implying that plant mitochondriacould play an active role in the process.Table 1 summarizes some recent literature in whichmitochondrial events were examined during PCD and thePhysiol. Plant. 129, 2007 257reader is referred to this literature for a more in-depthanalysis of this topic. An understanding of how mito-chondria contribute to PCD will depend upon elucidatingthe timing of mitochondrial events, of which we still haveonly a rudimentary knowledge. As summarized inTable 1, numerous studies have documented decreasesinDCm that precede PCD. In some cases (but not all), thisdecrease (and in some cases PCD itself) can be attenuatedby CsA, consistent with the drop in DCm representinga permeability transition. Often closely associated withthe loss of DCm is a loss of cyt c to the cytosol. This mightalso be consistent with a permeability transition becausemany animal models of cyt c release are dependent uponthe permeability transition (see Introduction). However,interpretations of such data remain difficult because themechanism of cyt c release in plants has not beeninvestigated. As noted in the previous section, a break-through in our understanding may reside with studies thathave shown a very early and localized increase in mROS.If, as we suggest, this mROS promotes a self-amplifyingcycle of mitochondrial dysfunction, then this could leadto the often-documented (and often slightly later) eventsof declining DCm and cyt c release. Interestingly, a recentarticle shows that cyt c release can be blocked byantioxidants, perhaps evidence that cyt c release isdependent upon mROS generation (Vacca et al. 2006).A central feature of many models of mitochondrialdysfunction and release of IMS proteins during animalapoptosis is opening of the PTP (see Introduction). Plantmitochondria are known to contain the key components(VDAC, ANT, cyclophilin-D) that constitute the animalPTP. Hence, a key question is whether a similar perme-ability transition occurs in plants. The elegant study ofArpagaus et al. (2002) strongly suggests that sucha permeability transition can indeed occur in plants andthat conditions promoting PTP opening are similar tothose described in animals. Under PTP-inducing con-ditions, swelling of purified potato mitochondria pro-ceeded with kinetics similar to that in animals and thisresulted in selective rupture of the OMM and release ofIMS proteins, including cyt c. Similar to animals, theseevents were absolutely dependent upon the presence ofCa21 (other cations such as Mg21were not effective) andwere potently inhibited by CsA. Similar to animals, theability of Ca21 to induce pore opening was modulated byother key factors. For example, the presence of Pi wasDefense geneexpressionPCDBiotic stressSA, NO,ROS, Ca2+,ceramide,virulence factors,toxinsunknown factorsIncreased ROSfrom ETCSelf-amplifyingloopMembrane disruptionMembrane poresRelease of IMS proteinRespiratory collapseAlternative mitochondrialETC components+CatastrophicmitochondrialdysfunctionEarly disruptionof mitochondrialhomeostasisFig. 2. Aworkingmodel for the role ofmitochondria in biotic stress responses such as the hypersensitive response. Themodel suggests that biotic stress-induced factors such as salicylic acid (SA), nitric oxide (NO), H2O2, Ca21, ceramide, virulence factors or other unknown agents promote defense geneexpression and programmed cell death by inhibiting the function of cytochrome (cyt) pathway electron transport chain (ETC) components such asComplex I, III or IV. This dysfunction promotes increased reactive oxygen species (ROS) generation by the ETC (mitochondrial ROS [mROS]), thus initiatingfurther damage and dysfunction in a self-amplifying manner, and ultimately leading to the catastrophic dysfunction associated with permeabilitytransition, loss of outer membrane integrity and release of intermembrane space proteins (including cyt c) to the cytosol. However, the unique alternativecomponents of the ETC (the rotenone-resistant NAD(P)H dehydrogenases and alternative oxidase) can attenuate these events by functionally replacingcyt pathway components and attenuatingmROS generation. Further, the activity and/or expression of these alternative componentsmay be enhanced bysome of the same factors (e.g. SA, NO, Ca21) responsible for inducing cyt pathway dysfunction. See text for further details.258 Physiol. Plant. 129, 2007Table 1. A summary of some recent studies linking the plant mitochondrion to programmed cell death (PCD). Only a subset of these represents bioticstress-induced PCD, enforcing the idea that the mitochondrion may be a common component amongst diverse PCD pathways. In the majority of thesestudies, PCDwas confirmed bymarkers such as nuclear condensation, cytoplasmic shrinkage or oligonucleosomal cleavage of DNA.DCm,mitochondrialtransmembrane potential; CsA, cyclosporin A; ROS, reactive oxygen species; OMM, outer mitochondrial membrane; AOX, alternative oxidase; ANT,adenine nucleotide translocator; VDAC, voltage-dependent anion channel; IMS, intermembrane space; PPIX, protoporphyrin IX; NO, nitric oxide; cyt,cytochrome.Experimental system Mitochondrial events ReferencePetroselinum crispum; suspension cells infectedwith Phytophthora infestansIncreased DCm and ROS accumulation inindividual fungus-infected cells precedes PCDNaton et al. 1996Helianthus annuus; PCD of tapetal cells incytoplasmic male sterile plantsCyt c release precedes loss of OMM integrity Balk and Leaver 2001Arabidopsis thaliana; PCD of synergid cells Mutant defective in the mitochondrial proteinGFA2 is defective in PCDChristensen et al. 2002Nicotiana tabacum; SA and H2O2-inducedPCD of suspension cellsCells lacking AOX show increased susceptibilityto PCD; cyt c releaseRobson and Vanlerberghe 2002Citrus sinensis; NO-treated suspension cells Decrease in DCm and PCD blocked by CsA Saviani et al. 2002A. thaliana; heat shock or senescence-associatedPCD of suspension cellsDecreased expression of ANT during PCD Swidzinski et al. 2002A. thaliana; oxidative stress-induced PCD ofsuspension cellsMitochondria from cells given oxidative stressgenerate increased ROSTiwari et al. 2002Triticum aestivum; root mitochondriaunder anoxiaAnoxia plus Ca21 induces mitochondrial swellingand cyt c release in CsA-insensitive mannerVirolainen et al. 2002Avena sativa; leaves treated with thehost-selective toxin victorinA burst of mROS clearly precedes a laterdecrease in DCmYao et al. 2002Zinnia elegans; tracheary element differentiation Decrease in DCm and CsA-independentcyt c release prior to PCDYu et al. 2002A. thaliana; isolated nuclear, cytosolic andmitochondrial fractions fromheat-shocked suspension cellsAn IMS-localized nuclease activity promoteshigh molecular weight DNA cleavageand chromatin condensation.Balk et al. 2003Nicotiana benthamiana; leaf PCD activationfollowing virus-induced silencing ofproteasome subunitsHigh ROS production, decreased DCm;cyt c releaseKim et al. 2003N. tabacum; ozone-induced leaf PCD Cyt c release Pasqualini et al. 2003Oryza sativa; lesion mimic mutant Hyperphosphorylation of the mitochondrialprotein prohibitinTakahashi et al. 2003Sugarbeet; camptothecin-induced PCD,digitonin-permeabilized protoplastsEarly increase, followed by later decreasein mROS and DCmWeir et al. 2003A. sativa; leaves treated with thehost-selective toxin victorinA subpopulation of mitochondria loseDCm prior to PCD, whereas othersin the same cell retain DCmCurtis and Wolpert 2004A. thaliana; harpin-treated suspension cells Rapid increase in mROS and decreasein DCmKrause and Durner 2004A. thaliana; heat shock or senescenceassociated PCD of suspension cellsPreferential maintenance of specificmitochondrial proteins(e.g. manganese superoxidedismutase; VDAC) during PCDSwidzinski et al. 2004Papaver rhoeas; pollen PCD duringself-incompatibility responseVery rapid cyt c release Thomas and Franklin-Tong 2004A. thaliana; ceramide, PPIX andelicitor-induced PCD in protoplastsDecrease in DCm is a early marker of PCD;CsA can partially block the decreasein DCm and PCD but not cyt c releaseYao et al. 2004Glycine max; NO and H2O2-inducedPCD of suspension cellsChanges in mitochondrial K1-channelactivityCasolo et al. 2005A. thaliana; dark-induced senescenceof attached leavesIncreased oxidative damage and acceleratedsenescence in mutant lackingmitochondrial NO synthaseGuo and Crawford 2005Physiol. Plant. 129, 2007 259necessary for Ca21-induced opening, the threshold[Ca21] needed for opening was lower at reduced DCmand opening was promoted by compounds capable ofthiol oxidation. In animals, oxidation of critical thiols ofthe ANT promotes pore opening and this may explainwhy ROS are often reported to enhance PTP opening(Kanno et al. 2004). Arpagaus et al. (2002) did notexamine whether ROS could promote pore opening butdid demonstrate that pore opening could occur underanoxia, thus precluding ROS as an absolute requirementfor permeability transition.An important area of future study will be to determinewhether the release of any IMS proteins from mitochon-drion to cytosol plays an active role in plant PCD,analogous to the situation in animals. For example,although the release of cyt c to the cytosol does appear tobe an event often coinciding with plant PCD, there is atpresent little compelling data to indicate that thisrelocalization is an obligatory event for PCD inductionand no indication that cyt c has interacting partners in thecytosol, similar to that seen in animals. One possibility isthat cyt c release in plants simply represents a morepassive (primitive?) mechanism promoting PCD than isdocumented in animals. For example, a progressive lossof cyt c could amplify mROS generation.Although an active role for cytosolic cyt c in PCDremains speculative, one study has provided compellingevidence for another IMS protein promoting PCD. Usinga cell-free system, Balk et al. (2003) have shown that anIMS DNase activity could mediate the generation of30-kb DNA fragments as well as DNA condensation.Such activity is reminiscent of apoptosis-inducing factor,an animal protein that once released from the IMS movesto the nucleus and brings about cleavage of DNA intolarge fragments and chromatin condensation. Severalapoptosis-inducing factor homologs are present in theArabidopsis genome. Finally, activation of caspases isa central feature of the mitochondria-dependent pathwayof animal apoptosis (see Introduction). Accumulatingevidence suggests that caspase-like activities are alsoactivated during plant PCD but this activation has not yetbeen strongly linked to the mitochondrion (Sanmartnet al. 2005).Recent studies suggest that the expressionof some plant defense genes may bemodulated by mitochondrial functionBeside a role in the HR, mitochondria may represent animportant intermediate between the perception of bioticstress and downstream responses such as the induction ofdefense gene expression ( Jones 2000, Lam et al. 2001).Studies that have investigated this hypothesis are outlinedbelow.Polyamines such as spermine are proposed to play a roleduring biotic stress responses. Spermine accumulatesdramatically and in an N-genespecific manner in theapoplast of tobacco mosaic virus (TMV)-infected tobaccobecause of upregulation of genes involved in sperminebiosynthesis (Yamakawa et al. 1998, Yoda et al. 2003).Accumulated polyamines are subsequently degraded in theapoplast by polyamine oxidase, generating H2O2 that maycontribute to plant responses (Yoda et al. 2003). A series ofrecent publications have investigated the series of eventsthat may link this apoplastic degradation of spermine todownstream events that include the mitochondrion. Thefindings of these studies are summarized in Fig. 3.Exogenous application of spermine to tobacco leavescould induce defense responses and cell death, mediatedby a pathway involving activation of mitogen activatedprotein (MAP) kinases and resulting in increased expressionof HR marker genes and transcription factors (Takahashiet al. 2003, 2004, Uehara et al. 2005). Interestingly, acti-vation of the MAP kinase cascade and downstream changesin gene expression could be blocked by BA, the inhibitor ofanimal PTP opening (see Introduction). This suggests thatmitochondrial events (dysfunction leading to PTP?) arerequired for MAP kinase activation by spermine. Sperminealso induced AOX, perhaps indicative of mitochondrialdysfunction. It was also shown that AOX induction andTable 1. ContinuedExperimental system Mitochondrial events ReferenceN. tabacum; protoplasts subjected to salt stress Decrease in DCm and initiation of PCD isdependent upon increases in cytosolicCa21 and is delayed by CsALin et al. 2005A. thaliana; ovule abortion during in responseto salt stressEarly ROS accumulation; early decreasein DCmHauser et al. 2006N. tabacum; heat-shockinduced PCD ofsuspension cellsCyt c release blocked by ROS scavengers;cytosolic cyt c degraded by caspase-like activityVacca et al. 2006A. thaliana; PPIX and P. syringae-inducedPCD in protoplastsEarly (1.5 h) increase in mROS; translocationof PCD modulator ACD2 to mitochondrionYao and Greenberg 2006260 Physiol. Plant. 129, 2007activation of the MAP kinases could be blocked by theantioxidant flavone and by the Ca21 channel blocker La21suggesting that increases in ROS and influx of Ca21 arecellular events upstream of the mitochondrial events.Critically, increased ROS and especially increased mito-chondrial Ca21 are thought to be critical factors promotingPTP opening (Arpagaus et al. 2002, Crompton 1999). Therequirement of spermine and mitochondrial events forchanges in defense gene induction should now beevaluated using recently isolated spermine-deficient mu-tants of Arabidopsis (Imai et al. 2004).As noted above, one source of the ROS generatedduring biotic stress could derive from the metabolism ofamines by cell walllocalized amine oxidases. Interest-ingly, animals have an amine oxidase that localizes to theOMM and which is able to induce the mitochondrialpathway of PCD via H2O2 generation (Maccarrone et al.2001). To our knowledge, there has been no report ofa similar activity in plant mitochondria but this mightrepresent a fruitful area for study.A similar BA-sensitive pathway to that outlined abovewas described by Maxwell et al. (2002) (Fig. 3). Theyfound that treatment of tobacco cells with AA resulted inthe rapid expression of eight different genes, as identifiedusing differential display. Only one of these genesencoded an ETC component (AOX), whereas the othersencoded proteins more generally implicated in eithersenescence or (biotic) stress responses. All eight geneswere also induced by treatment of cells with SA or H2O2.Each of the gene-inducing treatments was associated withincreased cellular levels of ROS and gene inductioncould be partially blocked by antioxidants that loweredROS levels, suggesting ROS as an important intermediaryin gene induction. The authors also showed that pre-treatment of cells with BA blocked induction of all eightgenes regardless of whether AA, SA or H2O2 was used asthe inducing agent (Maxwell et al. 2002).Interpretation of the above studies is still somewhatspeculative because it is not well established that theplant PTP is BA sensitive. Nonetheless, the mostparsimonious explanation of results to date is that a plantBA-sensitive PTPexists, that this pore opens in response tosignaling molecules commonly associated with bioticstress (spermine, SA, H2O2) or mitochondrial dysfunction(AA) and that the state of this pore affects defense geneexpression (Fig. 3). If this is the case, then plantmitochondria may indeed act as a focal point for theperception of and response to biotic stress.Further evidence that mitochondria may act in theperception of and response to stress (albeit abiotic stressin this case) comes from studies of the fro1 mutant ofArabidopsis (Lee et al. 2002). Mutant plants are unableto induce a set of cold-responsive genes, thus reducingtheir capacity to cold acclimate. The defect is specificto cold stress in that expression of the genes in responseto other treatments (abseisic acid, NaCl) is normal.Interestingly, FRO1 encodes a protein similar to the 18-kDa Fe-S subunit of complex I from diverse organismsand was shown to localize to Arabidopsismitochondria.Fro1 plants also displayed constitutively higher levels ofROS, even in the absence of stress. Because Complex I isconsidered a major site of mROS production, onepotential explanation is that mutation of the 18-kDasubunit has increased the ROS-generating activity ofComplex I. The authors hypothesize that the constitutivegeneration of ROS makes mutant plants less responsiveto what would normally be a cold-induced increase inROS generation. Presumably, this cold-induced ROSgeneration would also involve Complex I, thus repre-senting part of a cold-stressactivated signal path fromPTP openingSIPK / WIPKactivationChanges in gene expressionDefense / death responsesBongkrekic acidCa2+ ROSETCdysfunctionFlavoneTMV SpermineAA, SA, ROS, coldCa2+ influxROSApoplastSymplastLa2+Polyamine oxidaseMitochondrionFig. 3. A working model for how mitochondria may act as a focal pointfor the perception of and response to biotic stress. Several studies haveshown that stress-induced changes in gene expression can be blocked bybongkrekic acid, suggesting that opening of amitochondrial permeabilitytransition pore (PTP) is a necessary step for gene induction. Opening of thePTPmay be promoted by stress-induced changes in Ca21, reactive oxygenspecies or mitochondrial function. The results outlined here are primarilybased upon thework of Lee et al. (2002),Maxwell et al. (2002), Takahashiet al. (2003b, 2004) and Uehara et al. (2005). See text for further details.Physiol. Plant. 129, 2007 261mitochondrion to nucleus, with mROS as a key inter-mediate (Fig. 3).Other mutations of Complex I or Complex IV (but inthese case mutations that dramatically compromise theiractivity) have been shown to increase stress geneexpression and/or stress tolerance (Dutilleul et al. 2003,Kuzmin et al. 2004). These mutations illustrate that theconsequence of a major mitochondrial deficiency is notlimited to PCD (perhaps because of compensatoryactivities; see below) but can also be linked to protective(defense) responses.Several studies investigated a possible role for themitochondrion in the SA-mediated development of localand systemic resistance to viruses. Interest in this areastemmed from studies suggesting that AOX played someactive role in the induction of resistance. Results from theuse of transgenic plants with increased and decreasedlevels of AOX have largely negated any direct role forAOX in the development of resistance (Gilliland et al.2003, Ordog et al. 2002).Recent studies suggest that the uniquecomponents of the plant mitochondrial ETCplay a complex role in biotic stress responsesAlthough there is now a large body of circumstantialevidence that plant mitochondria play a regulatory role inPCD (Table 1), analogous to the relatively well-definedactive role of mitochondria in animal apoptosis, it is alsoevident that the regulation in plants must differ from thatin animals. This is best exemplified by the lack of planthomologs of many of the key pro- and antiapoptoticproteins described in animals (Lam et al. 2001). Hence,an important next step will be to define the pro- andantiapoptotic players in a plant mitochondria-dependentPCD pathway.As discussed earlier, the generation of mROS andmitochondrial dysfunction are early events associatedwith biotic stress and preceding PCD. Dysfunction maybe the result of key mitochondrial components (e.g.Complexes I, III, IV) being targeted by biotic stress signals,leading to the self-amplifying cycle of increased mROSand increased dysfunction described earlier (Fig. 2).However, a striking feature of plant mitochondria incomparison to their animal counterparts is the existenceof additional ETC components that increase the points ofentry and exit of electrons in the respiratory chain as wellas providing a high degree of flexibility in terms of thecoupling of electron transport to oxidative phosphoryla-tion (see Introduction). This is significant because thesecomponents (the rotenone-resistant alternative dehydro-genases and AOX) represent a potential means tomodulate a mitochondria-dependent PCD because theycould compensate for dysfunctional ETC components, aswell as providing a means to dampen the mROSgeneration associated with escalating dysfunction(Fig. 2). In other words, they may represent antiapoptoticcomponents of plant mitochondria, the levels of whichcould define cell fate (e.g. defense vs death). It is certainlyclear, for example, that AOX can prevent the PCDinitiated by a loss of cyt pathway function downstreamof ubiquinone (Vanlerberghe et al. 2002).Interestingly, numerous studies have shown increases inAOX expression in response to pathogen infection, sig-naling molecules (SA, NO, H2O2) or elicitors (Bruggmanet al. 2005, Huang et al. 2002, Krause and Durner2004, Maxwell et al. 2002, Vanlerberghe and McIntosh1996, Mizuno et al. 2005, Murphy et al. 2001, Lacommeand Roby 1999, Ordog et al. 2002, Takahashi et al. 2003,Simons et al. 1999, Zottini et al. 2002). As discussedearlier, one interpretation is that increased AOX expressionsimply represents an all-purpose response to disruptions ofrespiratory homeostasis. However, another interpretationis that it represents a defensive response against theinitiation of PCD, much like increases in expression ofROS-scavenging systems. Such a response could beimportant in limiting cell death progression and in thisregard it is interesting that overexpression of AOX has beenshown to reduce the size of HR lesions induced by TMV(Ordog et al. 2002). In the case of ROS-scavengingsystems, however, it has also been shown that an activedecline in their capacity may be an important means tocommit a cell to PCD (see Introduction). It would beinteresting to examine whether capacity of the alternativeETC components might also be declining in suchinstances. It is also possible that once a threshold level ofROS is reached in the mitochondrion, AOX might beinactivated by oxidation of critical sulfhydryl residuesinvolved in a-keto acid activation. This has beendemonstrated to occur when cells are treated exogenouslywith H2O2 (Vanlerberghe et al. 1999). Such inactivationcould again act to amplify mROS levels.Chloroplasts are a large source of ROS and so it is oftenassumed that the steady-state level of cellular ROS as wellas the capacity of cellular ROS-scavenging systems islargely determined by this organelle. However, asreviewed by Foyer and Noctor (2003), recent studiesindicate an unexpectedly influential role for mitochon-dria in determining the cellular level of ROS and capacityof both intra- and extramitochondrial antioxidant de-fenses. Hence, another role for the alternative compo-nents of the mitochondrial ETC during biotic stress(beside a role in compensating for dysfunctional ETCcomponents and/or controlling mROS production afterimposition of the stress) is that they may play a key role inconstitutively defining the cellular level of ROS and262 Physiol. Plant. 129, 2007capacity of antioxidant defenses in the plant. This coulddramatically impact the plant response to biotic stress(e.g. defense vs death) if in fact this response is modulatedby the cellular level of ROS and capacity of antioxidantdefenses (see Introduction). These ideas are summarizedin Fig. 4. Recently, we have investigated these points bymaking use of a collection of transgenic plants andsuspension cells with altered levels of AOX (Amirsadeghi,Robson and Vanlerberghe, unpublished). In both plantsand suspension cells, we found that genetic manipulationof AOX levels altered both the steady-state level of ROSand the capacity of cellular antioxidant defenses. Wefound that susceptibility of these plants or cells to death-inducing stimuli that may act synergistically with ROS(i.e. SA, NO) correlated well with the steady-state level ofROS. These results suggest that susceptibility to cell deathinitiated by these signaling molecules is strongly depen-dent upon the steady-state cellular level of ROS and thatAOX levels clearly contribute to this steady-state byinfluencing the rate of mROS generation and the cellularlevel of antioxidant defenses. It remains to be seen howthese plants will respond to various biotic stress.The potential regulatory role of the rotenone-resistantdehydrogenases in plant PCD has not yet been investi-gated. However, we hypothesize that both the increasesin cytosolic Ca21 associated with biotic stress as well asthe potential targeting of Complex I by SA could act toengage these components of the ETC.Acknowledgements We thank the Natural Sciences andEngineering Research Council of Canada for their continuedsupport of our work (to G. C. V.).ReferencesAllan AC, Fluhr R (1997) Two distinct sources of elicitedreactive oxygen species in tobacco epidermal cells. PlantCell 9: 15591572Alvarez ME (2000) Salicylic acid in the machinery ofhypersensitive cell death and disease resistance. PlantMol Biol 44: 429442Arpagaus S, Rawyler A, Braendle R (2002) Occurrenceand characteristics of the mitochondrial permeabilitytransition in plants. J Biol Chem 277: 17801787Balk J, Leaver CJ (2001) The PET1-CMS mitochondrialmutation in sunflower is associated with prematureprogrammed cell death and cytochrome c release. PlantCell 13: 18031818Balk J, Chew SK, Leaver CJ, McCabe PF (2003) Theintermembrane space of plant mitochondria containsa DNase activity that may be involved in programmedcell death. Plant J 34: 573583Bartoli CG, Gomez F, Martnez DE, Guiamet JJ (2004)Mitochondria are the main target for oxidative damagein leaves of wheat (Triticum aestivum L.) J Exp Bot 55:16631669Battaglia V, Salvi M, Toninello A (2005) Oxidative stress isresponsible for mitochondrial permeability transitioninduction by salicylate in liver mitochondria. J BiolChem 280: 3386433872Bolwell GP, Wojtaszek P (1997) Mechanisms for thegeneration of reactive oxygen species in plant defenseabroad perspective. Physiol Mol Plant Pathol 51: 347366Bratton SB, Cohen GM (2001) Apoptotic death sensor: anorganelles alter ego? Trends Pharmacol Sci 22: 306315ETCcompositionDefinedlevel ofmROSDefinedlevel ofmROSsignalingAntioxidantdefense genesDefined cellularantioxidantdefense levelTargetingto multiplecompartmentsDefined cellularROS levelSA, NOSynergisticinteractionsGraded cellular responseAntioxidant defensesAlternative ETC activityAntioxidant defenses? Alternative ETC activityDefense PCDBioticstresslevelFig. 4. A working model for how mitochondria may impact biotic stressresponses by defining cellular levels of reactive oxygen species (ROS) andantioxidant defense capacity. Recent studies suggest that the rate ofelectron transport chain (ETC)generated ROS (mitochondrial ROS[mROS]) is dependent upon the composition of the ETC (such as thelevel of alternative oxidase [AOX]) and that these mROS influence theantioxidant defense capacity of multiple cellular compartments byinfluencing nuclear gene expression. This antioxidant defense capacity,in turn, defines the steady-state cellular level of ROS. ROS level is criticalduring biotic stress, possibly because of its synergistic (and potentiallyantagonistic) interactions with key biotic stress signaling molecules suchas salicylic acid and nitric oxide. Cellular responses to these signalinteractions can range from defense gene expression to programmed celldeath. Progression toward the later may depend upon active decreases inantioxidant defenses and perhaps antiapoptotic components of the ETC(such as AOX). See text for further details.Physiol. Plant. 129, 2007 263Bruggman R, Abderhalden O, Reymond P, Dudler R (2005)Analysis of epidermis- and mesophyll-specific transcriptaccumulation in powdery mildew-inoculated wheatleaves. Plant Mol Biol 58: 247267Casolo V, Petrussa E, Krajnakova J, Macri F, Vianello A(2005) Involvement of the mitochondrial K1ATP channelin H2O2- or NO-induced programmed cell death ofsoybean suspension cell cultures. J Exp Bot 56: 9971006Christensen CA, Gorsich SW, Brown RH, Jones LG, Brown J,Shaw JM, Drews GN (2002) Mitochondrial GFA2 isrequired for synergid cell death in Arabidopsis. Plant Cell14: 22152232Crompton M (1999) The mitochondrial permeability transitionpore and its role in cell death. Biochem J 341: 233249Curtis MJ, Wolpert TJ (2002) The oat mitochondrialpermeability transition and its implication in victorinbinding and induced cell death. Plant J 29: 295312Curtis MJ, Wolpert TJ (2004) The victorin-inducedmitochondrial permeability transition precedes cellshrinkage and biochemical markers of cell death, andshrinkage occurs without loss of membrane integrity. PlantJ 38: 244259Dangl JL, Jones JDG (2001) Plant pathogens and integrateddefense responses to infection. Nature 411: 826833Delledonne M (2005) NO news is good news for plants.Curr Opin Plant Biol 8: 390396Dutilleul C, Garmier M, Noctor G, Mathieu C, Chetrit P,Foyer CH, de Paepe R (2003) Leaf mitochondria modulatewhole cell redox homeostasis, set antioxidant capacity,and determine stress resistance through altered signalingand diurnal regulation. Plant Cell 15: 12121226Finnegan PM, Soole KL, Umbach AL (2004) Alternativemitochondrial electron transport proteins in higher plants.In: Day DA, Millar AH, Whelan J (eds) Plant Mitochondria:From Genome to Function. Kluwer Academic Publishers,Great Britain, pp 163230Foyer CH, Noctor G (2003) Redox sensing and signalingassociated with reactive oxygen in chloroplasts,peroxisomes and mitochondria. Physiol Plant 119: 355364Gilliland A, Singh DP, Hayward JM, Moore CA, Murphy AM,York CJ, Slator J, Carr JP (2003) Genetic modification ofalternative respiration has differential effects on antimycinAinduced versus salicylic acidinduced resistance toTobacco mosaic virus. Plant Physiol 132: 15181528Greenberg JT, Vinatzer BA (2003) Identifying type III effectorsof plant pathogens and analyzing their interaction withplant cells. Curr Opin Microbiol 6: 2028Greenberg JT, Yao N (2004) The role and regulation ofprogrammed cell death in plantpathogen interactions.Cell Microbiol 6: 201211Gudz TI, Tserng K-Y, Hoppel CL (1997) Direct inhibitionof mitochondrial respiratory chain Complex III bycell-permeable ceramide. J Biol Chem 272: 2415424158Guo F-Q, Crawford NM (2005) Arabidopsis nitric oxidesynthase 1 is targeted to mitochondria and protects againstoxidative damage and dark-induced senescence. PlantCell 17: 34363450Hauser BA, Sun K, Oppenheimer DG, Sage TL (2006)Changes in mitochondrial membrane potential andaccumulation of reactive oxygen species precedeultrastructural changes during ovule abortion. Planta223: 492499Huang X, von Rad U, Durner J (2002) Nitric oxide inducestranscriptional activation of the nitric oxide-tolerantalternative oxidase in Arabidopsis suspension cells. Plant215: 914923Imai A, Akiyama T, Kato T, Sato S, Tabata S, Yamamoto KT,Takahashi T (2004) Spermine is not essential for survivalof Arabidopsis. FEBS Lett 556: 148152Jones A (2000) Does the plant mitochondrion integratecellular stress and regulate programmed cell death? TrendsPlant Sci 5: 225230Kanno T, Sato EF, Muranaka S, Fujita H, Fujiwara T, Utsumi T,Inoue M, Utsumi K (2004) Oxidative stress underlies themechanism for Ca21-induced permeability transition ofmitochondria. Free Radic Res 38: 2735Kim M, Ahn J-W, Jin U-H, Choi D, Paek K-H, Pai H-S (2003)Activation of the programmed cell death pathway byinhibition of proteasome function in plants. J Biol Chem278: 1940619415Krause M, Durner J (2004) Harpin inactivates mitochondriain Arabidopsis suspension cells. Mol Plant MicrobeInteract 17: 131139Kristensen BK, Askerlund P, Bykova NV, Egsgaard H, MllerIM (2004) Identification of oxidized proteins in the matrixof rice leaf mitochondria by immunoprecipitation andtwo-dimensional liquid chromatographytandem massspectrometry. Phytochemistry 65: 18391851Kuwana T and Newmeyer DD (2003) Bcl-2 family proteinsand the role of mitochondria in apoptosis. Curr Opin PlantBiol 15: 691699Kuzmin EV, Karpova OV, Elthon TE, Newton KJ (2004)Mitochondrial respiratory deficiencies signal up-regulationof genes for heat shock proteins. J Biol Chem 279:2067220677Kuzniak E, Skodowska M (2004) The effect of Botrytiscinerea infection on the antioxidant profile ofmitochondria from tomato leaves. J Exp Bot 55:605612Lacomme C, Roby D (1999) Identification of new earlymarkers of the hypersensitive response in Arabidopsisthaliana. FEBS Lett 459: 149153Laloi C, Apel K, Danon A (2004) Reactive oxygen signaling:the latest news. Curr Opin Plant Biol 7: 323328Lam E, Kato N, Lawton M (2001) Programmed cell death,mitochondria and the plant hypersensitive response.Nature 411: 848853Lee B, Lee H, Xiong L, Zhu J-K (2002) A mitochondrialComplex I defect impairs cold-regulated nuclear geneexpression. Plant Cell 14: 12351251264 Physiol. Plant. 129, 2007Liang H, Yao N, Song JT, Luo S, Lu H, Greenberg JT (2006)Ceramides modulate programmed cell death in plants.Genes Dev 17: 26362641Lin J, Wang Y, Wang G (2005) Salt stress-inducedprogrammed cell death via Ca21-mediated mitochondrialpermeability transition in tobacco protoplasts. PlantGrowth Regul 45: 243250Ly JD, Grubb DR, Lawen A (2003) The mitochondrialmembrane potential (DCm) in apoptosis; an update.Apoptosis 8: 115128Maccarrone M, Bari M, Battista N, Di Rienzo M, Falciglia K,Agro` AF (2001) Oxidation products of polyaminesinduce mitochondrial uncoupling and cytochromec release. FEBS Lett 507: 3034Maxwell DP, Wang Y, McIntosh L (1999) The alternativeoxidase lowers mitochondrial reactive oxygen productionin plant cells. Proc Natl Acad Sci USA 96: 82718276Maxwell DP, Nickels R, McIntosh L (2002) Evidence ofmitochondrial involvement in the transduction of signalsrequired for the induction of genes associated withpathogen attack and senescence. Plant J 29: 269279Mittler R, Feng X, Cohen M (1998) Post-transcriptionalsuppression of cytosolic ascorbate peroxidase expressionduring pathogen-induced programmed cell death intobacco. Plant Cell 10: 461473Mittler R, Vanderauwera S, Gollery M, Van Breusegem F(2004) Reactive oxygen gene network of plants. TrendsPlant Sci 9: 490498Mizuno M, Tada Y, Uchii K, Kawakami S, Mayama S (2005)Catalase and alternative oxidase cooperatively regulateprogrammed cell death induced by b-glucan elicitor inpotato suspension cultures. Planta 220: 849853Mller IM (2001) Plant mitochondria and oxidative stress:electron transport, NADPH turnover, and metabolism ofreactive oxygen species. Annu Rev Plant Physiol PlantMol Biol 52: 561591Murphy AM, Gilliland A, Wong CE, West J, Singh DP, Carr JP(2001) Signal transduction in resistance to plant viruses.Eur J Plant Pathol 107: 121128Naton B, Hahlbrock K, Schmelzer E (1996) Correlation ofrapid cell death with metabolic changes in fungus-infected, cultured parsley cells. Plant Physiol 112:433444Neill S, Desikan R, Hancock J (2002) Hydrogen peroxidesignaling. Curr Opin Plant Biol 5: 388395Newmeyer DD, Ferguson-Miller S (2003) Mitochondria:releasing power for life and unleashing the machineriesof death. Cell 112: 481490Norman C, Howell KA, Millar AH, Whelan JM, Day DA(2004) Salicylic acid is an uncoupler and inhibitor ofmitochondrial electron transport. Plant Physiol 134:492501Ordog SH, Higgins VJ, Vanlerberghe GC (2002)Mitochondrial alternative oxidase is not a criticalcomponent of plant viral resistance but may play a role inthe hypersensitive response. Plant Physiol 129: 18581865Pasqualini S, Piccioni C, Reale L, Ederli L, Della Torre G,Ferranti F (2003) Ozone-induced cell death in tobaccocultivar Bel W3 plants. The role of programmed cell deathin lesion formation. Plant Physiol 133: 11221134Planchet E, Gupta KJ, Sonoda M, Kaiser WM (2005) Nitricoxide emission from tobacco leaves and cell suspensions:rate limiting factors and evidence for the involvement ofmitochondrial electron transport. Plant J 41: 732743Quillet-Mary A, Jaffrezou J-P, Mansat V, Bordier C, Naval J,Laurent G (1997) Implication of mitochondrial hydrogenperoxide generation in ceramide-induced apoptosis.J Biol Chem 272: 2138821395Rasmusson AG, Soole KL, Elthon TE (2004) AlternativeNAD(P)H dehydrogenases of plant mitochondria. AnnuRev Plant Biol 55: 2339Robson CA, Vanlerberghe GC (2002) Transgenic plantcells lacking mitochondrial alternative oxidase haveincreased susceptibility to mitochondria-dependentand -independent pathways of programmed celldeath. Plant Physiol 129: 19081920Sanmartn M, Jaroszewski L, Raikhel N, Rojo E (2005)Caspases. Regulating death since the origin of life. PlantPhysiol 137: 841847Saviani EE, Orsi CH, Oliveira JFP, Pinto-Maglio CAF,Salgado I (2002) Participation of the mitochondrialpermeability transition pore in nitric oxide-induced plantcell death. FEBS Lett 510: 136140Simons BH, Millenaar FF, Mulder L, Van Loon LC, Lambers H(1999) Enhanced expression and activation of thealternative oxidase during infection of Arabidopsis withPseudomonas syringae pv tomato. Plant Physiol 120:529538Sweetlove L, Heazlewood J, Herald V, Holtzapffel R, DayDA, Leaver C, Millar AH (2002) The impact of oxidativestress on Arabidopsis mitochondria. Plant J 32: 891904Swidzinski JA, Sweetlove LJ, Leaver CJ (2002) A custommicroarray analysis of gene expression duringprogrammed cell death in Arabidopsis thaliana. Plant J 30:431446Swidzinski JA, Leaver CJ, Sweetlove LJ (2004) A proteomicanalysis of plant programmed cell death. Phytochem 65:18291838Tada Y, Kusaka K, Betsuyaku S, Shinoyi T, Sakamoto M,Ohura Y, Hata S, Mori T, Tosa Y, Mayama S (2005) Victorintriggers programmed cell death and the defense responsevia interaction with a cell surface mediator. Plant CellPhysiol 46: 17871798Takahashi A, Kawasaki T, Wong HL, Suharsono U, Hirano H,Shimamoto K (2003a) Hyperphosphorylation ofa mitochondrial protein, prohibitin, is induced bycalyculin A in a rice lesion-mimic mutant cdr1. PlantPhysiol 132: 18611869Physiol. Plant. 129, 2007 265Takahashi Y, Berberich T, Miyazaki A, Seo S, Ohashi Y,Kusano T (2003b) Spermine signaling in tobacco:activation of mitogen-activated protein kinases byspermine is mediated through mitochondrial dysfunction.Plant J 36: 820829Takahashi Y, Uehara Y, Berberich T, Ito A, Saitoh H, MiyazakiA, Terauchi R, Kusano T (2004) A subset of hypersensitiveresponse marker genes, including HSR203J, is thedownstream target of a spermine signal transductionpathway in tobacco. Plant J 40: 586595Taylor NL, Day DA, Millar AH (2002) Environmental stresscauses oxidative damage to plant mitochondria leading toinhibition of glycine decarboxylase. J Biol Chem 277:4266342668Thomas SG, Franklin-Tong VE (2004) Self-incompatibilitytriggers programmed cell death in Papaver pollen. Nature429: 305309Tiwari BS, Belenghi B, Levine A (2002) Oxidative stressincreased respiration and generation of reactive oxygenspecies, resulting in ATP depletion, opening ofmitochondrial permeability transition, and programmedcell death. Plant Physiol 128: 12711281Torres MA, Dangl JL (2005) Functions of the respiratory burstoxidase in biotic interactions, abiotic stress anddevelopment. Curr Opin Plant Biol 8: 397403Uehara Y, Takahashi Y, Berberich T, Miyazaki A, TakahashiH, Matsui K, Ohme-Takagi M, Saitoh H, Terauchi R,Kusano T (2005) Tobacco ZFT1, a transcriptional repressorwith a Cys2/His2 type zinc finger motif that functionsin spermine-signaling pathway. Plant Mol Biol 59:435448Vacca RA, de Pinto MC, Valenti D, Passarella S, Marra E,De Gara L (2004) Production of reactive oxygen species,alteration of cytosolic ascorbate peroxidase, andimpairment of mitochondrial metabolism are early eventsin heat-shockinduced programmed cell death in tobaccoBright-Yellow 2 cells. Plant Physiol 134: 11001112Vacca RA, Valenti D, Bobba A, Merafina RS, Passarella S,Marra E (2006) Cytochrome c is released in a reactiveoxygen species-dependent manner and is degraded viacaspase-like proteases in tobacco BY-2 cells en route toheat-shock induced cell death. Plant Physiol 141: 208219Vanlerberghe GC, McIntosh L (1996) Signals regulating theexpression of the nuclear gene encoding alternativeoxidase of plant mitochondria. Plant Physiol 111: 589595Vanlerberghe GC, Yip JYH, Parsons HL (1999) In organelloand in vivo evidence of the importance of the regulatorysulfhydryl/disulfide system and pyruvate for alternativeoxidase activity in tobacco. Plant Physiol 121: 793803Vanlerberghe GC, Robson CA, Yip JYH (2002) Induction ofmitochondrial alternative oxidase in response to a cellsignal pathway down-regulating the cytochrome pathwayprevents programmed cell death. Plant Physiol 129:18291842van Loo G, Saelens X, van Gurp M, MacFarlane M, Martin SJ,Vandenabeele P (2002) The role of mitochondrial factorsin apoptosis: a Russian roulette with more than one bullet.Cell Death Differ 9: 10311042Vieira H, Kroemer G (2003) Mitochondria as targets ofapoptosis regulation by nitric oxide. IUBMB Life 55:613616Virolainen E, Blokhina O, Fagerstedt K (2002) Ca21-inducedhigh amplitude swelling and cytochrome c release fromwheat (Triticum aestivum L.) mitochondria under anoxicstress. Ann Bot 90: 509516Weir IE, Pham N-A, Hedley DW (2003) Oxidative stress isgenerated via the mitochondrial respiratory chain duringplant cell apoptosis. Cytometry Part A 54A: 109117Wendehenne D, Durner J, Klessig DF (2004) Nitric oxide:a new player in plant signaling and defense responses.Curr Opin Plant Biol 7: 449455Xie Z, Chen Z (1999) Salicylic acid induces rapid inhibitionof mitochondrial electron transport and oxidativephosphorylation in tobacco cells. Plant Physiol 120:217226Xie Z, Chen Z (2000) Harpin-induced hypersensitive celldeath is associated with altered mitochondrial functions intobacco cells. Mol Plant Microbe Interact 13: 183190Yamakawa H, Kamada H, Satoh M, Ohashi Y (1998)Spermine is a salicylate-independent endogenous inducerfor both tobacco acidic pathogenesis-related proteins andresistance against tobacco mosaic virus infection. PlantPhysiol 118: 12131222Yao N, Greenberg JT (2006) Arabidopsis ACCELERATEDCELL DEATH2 modulates programmed cell death. PlantCell 18: 397411Yao N, Tada Y, Sakamoto M, Nakayashiki H, Park P, Tosa Y,Mayama S (2002) Mitochondrial oxidative burst involvedin apoptotic response in oats. Plant J 30: 567579Yao N, Eisfelder BJ, Marvin J, Greenberg JT (2004) Themitochondrionan organelle commonly involved inprogrammed cell death in Arabidopsis thaliana. Plant J40: 596610Yoda H, Yamaguchi Y, Sano H (2003) Induction ofhypersensitive cell death by hydrogen peroxide producedthrough polyamine degradation in tobacco plants. PlantPhysiol 132: 19731981Yu X-H, Perdue TD, Heimer YM, Jones AM (2002)Mitochondrial involvement in tracheary elementprogrammed cell death. Cell Death Differ 9:189198Zottini M, Formentin E, Scattolin M, Carimi F, Schiavo FL,Terzi M (2002) Nitric oxide affects plant mitochondrialfunctionality in vivo. FEBS Lett 515: 7578Edited by P. Gardestrom266 Physiol. Plant. 129, 2007


View more >