Planta (2002) 214: 537±545 DOI 10.1007/s004250100654
O R I GI N A L A R T IC L E
Anke J. de Jong á Elena T. Yakimova Veneta M. Kapchina á Ernst J. Woltering
A critical role for ethylene in hydrogen peroxide release during programmed cell death in tomato suspension cells Received: 10 February 2001 / Accepted: 13 July 2001 / Published online: 11 October 2001 Ó Springer-Verlag 2001
Abstract Camptothecin, a topo isomerase-I inhibitor used in cancer therapy, induces apoptosis in animal cells. In tomato (Lycopersicon esculentum Mill.) suspension cells, camptothecin induces cell death that is accompanied by the characteristic nuclear morphological changes such as chromatin condensation and nuclear and DNA fragmentation that are commonly associated with apoptosis in animal systems. These eects of camptothecin can eectively be blocked by inhibitors of animal caspases, indicating that, in tomato suspension cells, camptothecin induces a form of programmed cell death (PCD) with similarities to animal apoptosis (A.J. De Jong et al. (2000) Planta 211:656±662). Camptothecininduced cell death was employed to study processes involved in plant PCD. Camptothecin induced a transient increase in H2O2 production starting within 2 h of application. Both camptothecin-induced cell death and the release of H2O2 were eectively blocked by application of the calcium-channel blocker lanthanum chloride, the caspase-speci®c inhibitor Z-Asp-CH2-DCB, or the NADPH oxidase inhibitor diphenyl iodonium, indicating that camptothecin exerts its eect on cell death through a calcium- and caspase-dependent stimulation of NADPH oxidase activity. In addition, we show that ethylene is an essential factor in camptothecin-induced PCD. Inhibition of either ethylene synthesis or ethylene perception by L-a-(2-aminoethoxyvinyl)glycine or silver thiosulphate, respectively, blocked camptothecin-induced
A.J. de Jong á E.T. Yakimova1 á V.M. Kapchina2 E.J. Woltering (&) Agrotechnological Research Institute (ATO), Wageningen University and Research Center, Bornsesteeg 59, 6708 PD Wageningen, The Netherlands E-mail:
[email protected] Fax: +31-317-475347 Present addresses: 1 Institute of Floriculture, Agricultural Academy, 1222 Negovan, So®a, Bulgaria 2 Department of Plant Physiology, Faculty of Biology, University of So®a Blvd. Dr. Tzankov-8, 1421 So®a, Bulgaria
H2O2 production and PCD. Although, in itself, insucient to trigger H2O2 production and cell death, exogenous ethylene greatly stimulated camptothecin-induced H2O2 production and cell death. These results show that ethylene is a potentiator of the camptothecin-induced oxidative burst and subsequent PCD in tomato cells. The possible mechanisms by which ethylene stimulates cell death are discussed. Keywords Apoptosis á Ethylene á Lycopersicon (programmed cell death) á Hydrogen peroxide á Oxidative burst á Programmed cell death Abbreviations AVG: L-a-(2-aminoethoxyvinyl)glycine á DPI: diphenyl iodonium á FDA: ¯uorescein diacetate á MAPK: mitogen-activated protein kinase á PCD: programmed cell death á ROS: reactive oxygen species á STS: silver thiosulphate á TLCK: Na-p-tosylL-lysine chloromethyl ketone á Z-Asp-CH2-DCB: Na-benzyloxycarbonyl-L-aspartic acid a-([2,6-dichlorobenzoyloxy]methyl ketone)
Introduction Programmed cell death (PCD) or apoptosis is a genetically de®ned process associated with characteristic morphological and biochemical changes (Steller 1995). It is well established that PCD is an intrinsic part of the life cycle of all multicellular organisms studied so far, including both animals and plants (Mittler and Lam 1996; Pennell and Lamb 1997). PCD can be induced by a variety of stimuli, including developmental signals and environmental cues (Jones and Dangl 1996; Wertz and Hanley 1996). In recent years, an intensive eort has been made in order to identify the key factors in the regulation of PCD in plants. A key event in animal PCD is the release of cytochrome c from mitochondria into the cytosol, and subsequent complexation with Apaf1, resulting in the activation of caspases, a speci®c family of cysteine
538
proteases (Chu et al. 2001). Mitochondria, in addition, are the major source of reactive oxygen species (ROS) in animal cells and may lead to alteration of the cellular reduction-oxidation (redox) potential, another cellular trigger believed to play an important role in the induction of apoptosis (Grant and Loake 2000). The cytosolic caspase-mediated apoptotic pathway is highly conserved between animal cell types, but such a death cascade has not been found in plant cells thus far. However, recent studies reveal that at least some similarities between plant PCD and animal caspase-mediated apoptotic cell death programs exist. Camptothecin-induced PCD in tomato suspension cells can be inhibited with caspase-speci®c peptide inhibitors, suggesting that caspase-like proteases are involved in the death process (De Jong et al. 2000). Furthermore, caspase-like protease activity was required in N-gene mediated hypersensitive response (HR) in tobacco (Del Pozo and Lam 1998) and in PCD in barley suspension cells (Korthout et al. 2000). Recently, a group of structurally related proteases, designated metacaspases, was identi®ed in animals and homologues were found in plants (Uren et al. 2000). Bcl2, and related death-agonist proteins, promote cell survival by blocking the release of cytochrome c and subsequent activation of caspases (Luo et al. 1998). Homologues of animal Bcl-2 have been detected in plants (Dion et al. 1997) and expression of Bcl-Xl and its homologue from Caenorhabditis elegans, Ced-9, in tobacco inhibited cell death (Mitsuhara et al. 1999). The death-agonist members of the Bcl-2 family, of which Bax is a well-studied example, appear to be able to disrupt the protective eect of Bcl-2. Expression of Bax induced an HR-type cell death in tobacco (Lacomme and Cruz 1999) and a plant homologue of Bax inhibitor-1, AtBI-1, suppressed Bax-induced cell death in yeast and was rapidly upregulated during wounding and pathogen attack (Sanchez et al. 2000). Taken together, these ®ndings suggest that processes involved in plant PCD share similarity to animal PCD. Additional support for this hypothesis is the observation that Xenopus egg extracts were able to induce apoptotic changes in plant nuclei (Fan et al. 1999) and that plant mitochondria and plant cytochrome c were able to activate caspase-3 in Xenopus extracts (Korthout et al. 2000). However, it should be recognised that regulators of PCD in plants may also dier from animal PCD regulators. One of the putative regulators of PCD in plant cells is ethylene (C2H4). PCD, accompanied by TUNEL (terminal deoxynucleotidyl transferase-mediated dUTP nick end labelling)-positive nuclei, during aerenchyma formation in hypoxic maize roots was found to depend on ethylene (He et al. 1996; Gunawardena et al. 2001). Similarly, PCD during maize endosperm development is dependent on ethylene (Young and Gallie 2000). Flower petal and ovary senescence in pea is accompanied by the appearance of TUNEL-positive nuclei and DNA laddering while inhibitors of ethylene action delayed senescence processes and blocked DNA degradation. Furthermore, ethylene treatment induced senescence and DNA breakdown (OrzaÂez and Granell 1997).
A well-studied process in plants that involves PCD is the response of cells to pathogens and the elicitation of cells by abiotic stresses. Plants respond to attack by incompatible pathogens by activating a variety of defence responses, including activation of an ROS-generating enzyme complex responsible for the oxidative burst (Bolwell 1999). The overall response comprises formation of small necrotic lesions, resulting from the hypersensitive cell death. One of the earliest detectable reactions of a cell upon pathogen recognition are the opening of speci®c ion channels, and the formation of superoxide and subsequent accumulation of H2O2, probably through plasma-membrane-associated NADPH oxidases (Kombrink and Somssich 1995). NADPH oxidase generates superoxide anions (O2±), which are readily dismuted into H2O2 either spontaneously or by superoxide dismutase (Sutherland 1991). These initial transient reactions are, at least in part, prerequisites for further signal transduction events, possibly resulting in the build-up of a complex signalling network that triggers the overall defence response (Hammond-Kosack and Jones 1996). The role of ethylene in pathogeninduced cell death has been investigated in ethylene insensitive (NR) tomatoes. Following infection of these mutants with Pseudomonas or Fusarium, largely reduced disease symptoms were observed, indicating that ethylene is involved in the susceptible response to pathogens (Lund et al. 1998). Plant defence reactions can also be elicited by abiotic elicitors, such as ozone (O3). Ozone forms ROS in the apoplast and causes the plant cell itself to produce ROS in an oxidative burst. In sensitive plants this leads to the formation of HR-like lesions, the formation of which has the characteristics of PCD. Ozone exposure upregulates ethylene biosynthesis and, if ethylene biosynthesis or perception is blocked, the incidence of lesions is reduced. It was therefore suggested that ROS and ethylene together are involved in the induction of cell death in O3-exposed plants (Overmyer et al. 2000). Camptothecin is a topo isomerase-I inhibitor and a known inducer of apoptosis in animal cells (Kaufmann 1998; Simizu et al. 1998). In tomato suspension cells, camptothecin induces cell death accompanied by nuclear morphological changes, such as chromatin condensation and nuclear and DNA fragmentation, commonly observed in apoptotic animal cells. Inhibitors of animal caspases were found to inhibit camptothecin-induced PCD and DNA fragmentation in tomato cells (De Jong et al. 2000). The ability to manipulate PCD in tomato suspension cells with speci®c inhibitors makes this system an attractive model with which to identify the processes involved in plant apoptotic cell death. Here we report on the role of ethylene in PCD in tomato suspension cells. We show that ethylene plays a regulatory role in the release of ROS during camptothecin-induced PCD. A model for camptothecin-induced signalling in tomato suspension cells is presented and similarities between HR-type and camptothecin-induced cell death in tomato suspension cells are discussed.
539
Materials and methods
Results
Chemicals
Ethylene is necessary, but not sucient to induce PCD
The caspase-inhibiting peptide Na-benzyloxycarbonyl-L-aspartic acid a-([2,6-dichlorobenzoyloxy]methyl ketone) (Z-Asp-CH2DCB) was obtained from Bachem AG (Bubendorf, Switzerland), and ethylene from Praxair (Oevel, Belgium). All other chemicals were obtained from Sigma.
Cell culture and treatments Tomato (Lycopersicon esculentum Mill.) cell-suspension cultures, line Msk8 (Koornneef et al. 1987; kindly provided by T. Boller, Basel, Switzerland), were grown on a Murashige-Skoog-type liquid medium supplemented with 5 lM a-naphthalene acetic acid, 1 lM 6-benzyladenine and vitamins, as described by Adams and Townsend (1983). Cells were subcultured every 7 days by making a 1:4 dilution in 25 ml of medium in 100-ml ¯asks with aluminium caps. For treatments, cells were used 5 days after subculture. Cell death inducers and inhibitors were added simultaneously to 5 ml of suspension culture in 30-ml ¯asks with a gas-tight screw cap. The viability of the cells was determined by staining with 0.002% ¯uorescein diacetate (FDA). Camptothecin was applied in dimethyl sulfoxide (DMSO; ®nal solvent concentration 0.1% v/v). DMSO had no eect on the viability of the cells. For ethylene treatments, 100 ll/l ethylene was added to the head space of the 30-ml ¯ask with 5 ml of suspension culture. This yielded a ®nal concentration of ethylene in the culture medium of 10±12 ll/l. A 0.2 mM stock solution of silver thiosulphate (STS) was prepared by mixing 100 ml of 2 mM AgNO3 with 100 ml of 16 mM Na2S2O3á5H2O and making up to 1 l with water.
To investigate the role of ethylene in camptothecininduced cell death in tomato suspension cells, ethylene was added together with camptothecin. This resulted in a 127% increase in cell death within 24 h compared with camptothecin alone. Addition of ethylene alone did not induce cell death (Fig. 1a). Application of lower concentrations of ethylene or addition of micromolar concentrations of the ethylene precursor 1-aminocyclopropane-1-carboxylic acid (ACC) gave similar results (Woltering et al. 1999). Hence, ethylene stimulates camptothecin-induced cell death. The involvement of ethylene in camptothecin-induced PCD was con®rmed by the addition of speci®c ethylene inhibitors. Addition of camptothecin together with L-a-(2-aminoethoxyvinyl)glycine (AVG), an ethylene-synthesis inhibitor, resulted in 82% inhibition of cell death. To show that the inhibiting eect of AVG is due to its action on ethylene synthesis, camptothecin was added together with both AVG and ethylene. As shown in Fig. 1a, the addition of
Measurement of H2O2 Hydrogen peroxide release by the tomato suspension cells was measured by chemiluminescence in a ferricyanide-catalysed oxidation of luminol (Warm and Laties 1982). Cells were treated and harvested at dierent time points to measure H2O2 release. The culture medium was removed and cells were washed in washing buer (24.7 mM KNO3, 1.1 mM NaH2PO4, 1 mM MgSO4, 1 mM (NH4)2SO4, 1 mM CaCl2, 20 mM Mes, 2% sucrose, pH 6.6). 0.5 g of cells was resuspended in 5 ml of washing buer. A sample was taken both immediately and after 30 min of incubation on a rotary shaker. Cells were spun down and the H2O2 concentration in the supernatant was measured as described by Schwacke and Hager (1992). Because of interference with the assay we could not measure H2O2 directly in the culture medium. The release of H2O2 measured by us was not an eect of the transfer of the cells to the assay buer since in cells suspended for a prolonged period in the assay buer the addition of camptothecin elicited a similar response (data not shown).
Ethylene measurement To measure the ethylene concentration in the head space of the ¯asks, a 2-ml gas-sample was withdrawn and analysed by gas chromatography. The ethylene concentration in the culture medium was analysed by measuring the ethylene release from a 1-ml sample of the medium incubated in a 100-ml ¯ask. Ethylene production rates of cells were measured following incubation of 5 ml of suspension in 30-ml ¯asks for 1 h. The fresh weight of cells was determined after ®ltration.
Fig. 1a, b Ethylene and camptothecin-induced PCD. a Stimulation of camptothecin-induced PCD by ethylene. Suspension-cultured tomato (Lycopersicon esculentum Mill.) cells were left untreated or were treated with camptothecin (CPT; 5 lM), camptothecin + ethylene (Eth; 100 ll/l in the head space), camptothecin + AVG (10 lM), or camptothecin + STS (20 lM). Cell death was estimated by FDA staining. Data are the means SE of at least three independent experiments using dierent batches of cells. b Production of ethylene in a tomato suspension culture. Cells were left untreated (®lled triangles) or were treated with 5 lM camptothecin (open squares) or10 lg/ml xylanase (®lled diamonds). Ethylene-production measurements were done in triplicate; SE was less than 5% of the mean
540
ethylene abolished the inhibiting eect of AVG on camptothecin-induced cell death. This indicates that the basal ethylene production is a prerequisite for cell death induced by camptothecin. Addition of camptothecin together with STS, an inhibitor of ethylene action, resulted in 66% inhibition of cell death. To show that the inhibiting eect of STS is due to its eect on ethylene action, camptothecin was added together with both STS and ethylene. This did not result in a higher percentage of dead cells, indicating that indeed the ethylene action was blocked by STS. To determine whether camptothecin elicits ethylene production in tomato suspension cells, the ethylene concentration was measured in the head space of the ¯asks. Camptothecin did not stimulate ethylene production in tomato suspension cells. This was not due to a general inability of the cells to produce ethylene, because treatment with, for example, xylanase, a known inducer of ethylene production in tomato cells (Grosskopf et al. 1990), did induce elevated amounts of ethylene (Fig. 1b). Role of calcium, mitogen-activated protein kinases (MAPKs), and (caspase-like) proteases in ethylene-stimulated cell death To investigate the processes involved in the ethylenemediated cell death pathway, several classes of inhibitor were tested for their eect on both cell death induced by camptothecin alone and by simultaneous application of camptothecin and ethylene. Addition of lanthanum chloride (LaCl3), a plasmamembrane calcium-channel blocker, resulted in 75% reduction of cell death induced by camptothecin (Fig. 2). Cell death induced by simultaneous application of camptothecin and ethylene was also inhibited by lanthanum chloride (92%). This indicates that a calcium
Fig. 2 The eect of inhibitors on cell death induced by camptothecin alone or by simultaneous application of camptothecin (CPT) and ethylene. Tomato cells were treated with camptothecin (5 lM) or with camptothecin together with ethylene (100 ll/l in the head space). The inhibiters tested were TLCK (0.2 lM), Z-Asp-CH2DCB (0.1 lM), PD98059 (0.1 lM), catalase (10 U), DPI (10 lM) and LaCl3 (100 lM). Cell death was estimated by FDA staining. Data are the means SE of at least three independent experiments using dierent batches of cells
in¯ux is involved in camptothecin-induced cell death at basal as well as at high ethylene levels. To investigate the role of MAPKs in camptothecininduced cell death, the inhibitor of mammalian MAPKs, PD98059, was added. PD98059 did reduce ethylenestimulated camptothecin-induced PCD (63%), but did not have any eect on camptothecin-induced PCD (Fig. 2). This indicates that MAPKs may play a role in cell death at high ethylene levels but may not be involved at basal ethylene levels. Previous experiments have shown the involvement of proteases and caspases in camptothecin-induced cell death (De Jong et al. 2000). As shown in Fig. 2, the serine protease inhibitor Na-p-tosyl-L-lysine chloromethyl ketone (TLCK) inhibits camptothecin-induced cell death by 57%. By contrast, TLCK does not appreciably inhibit ethylene-stimulated camptothecin-induced cell death. A similar result was obtained with the caspasespeci®c peptide inhibitor Z-Asp-CH2-DCB. This inhibitor reduced camptothecin-induced cell death by 72% but did not appreciably reduce ethylene-stimulated camptothecin-induced cell death. To test the hypothesis that ethylene stimulates camptothecin-induced cell death by increasing the activity of (caspase-like) proteases in the cells, a 10-fold higher concentration of protease or caspase-inhibitors was added. This had no eect on the percentage of cell death induced by simultaneous application of camptothecin and ethylene (data not shown). These results indicate that ethylene does not stimulate camptothecin-induced cell death by regulation of the activity of (caspase-like) proteases. A calcium in¯ux and ethylene are sucient to induce cell death The observation that lanthanum chloride inhibits cell death induced by both application of camptothecin alone and by simultaneous application of camptothecin and ethylene suggests that calcium is an essential component in the cell death pathway. It was hypothesised that one of the main events upon triggering the cells with camptothecin could be a calcium in¯ux. To test this hypothesis, a calcium in¯ux was arti®cially induced by the addition of the antibiotic ionophore A23187, a compound that is widely used to study the regulatory role of calcium ions in biological systems (Rasmussen and Goodman 1977). Application of A23187 together with ethylene induced 47% cell death (Fig. 3). This percentage of cell death was about similar to that induced by simultaneous application of camptothecin and ethylene (Fig. 1). Addition of either ethylene alone or A23187 alone did not induce cell death (Fig. 3). Camptothecin-induced PCD is accompanied by an oxidative burst To investigate the role of H2O2 in PCD, compounds that interfere with H2O2 metabolism were tested for
541
Fig. 3 Induction of cell death by a calcium ionophore and ethylene. Tomato cells were left untreated, or were treated with ethylene (100 ll/l in the head space), A23187 alone (1 lM) or with ethylene + A23187. Cell death was estimated by FDA staining. Data are the means SE of at least three independent experiments using dierent batches of cells
their eect on cell death induced by application of camptothecin alone and by simultaneous application of camptothecin and ethylene. As shown in Fig. 2, catalase reduced camptothecin-induced cell death by 84% and ethylene-stimulated camptothecin-induced cell death by 50%. This indicates that H2O2 is involved in cell death induced by both treatments. Similarly, addition of diphenyl iodonium (DPI), an inhibitor of H2O2-producing NADPH-oxidase, inhibited camptothecin- induced cell death and ethylene-stimulated camptothecin-induced cell death by 59% and 51%, respectively (Fig. 2). These results again emphasise the role of H2O2 in cell death induced by both treatments. H2O2 was measured in the modi®ed culture medium by chemiluminescence from the ferricyanide-catalyzed oxidation of luminol. Treatment of the tomato suspension cells with camptothecin resulted in a transient increase in the production of H2O2. The H2O2 concentration started to increase within 2 h after treatment with camptothecin, reaching a maximum at about 6 h (Fig. 4), and declined within 24 h to the control level
Fig. 4 Camptothecin induces a release of H2O2. Tomato cells were left untreated (®lled diamonds) or treated with camptothecin alone (open squares), or together with either 10 lM DPI (®lled triangles) or 100 lM LaCl3 (crosses). H2O2 was estimated using luminoldependent chemiluminescence. Data are the means SE of at least three independent experiments using dierent batches of cells
(not shown). Untreated cells exhibited a baseline production of active oxygen equivalent to less than 0.6 lM H2O2 in the medium. When the cells were simultaneously incubated with camptothecin and DPI, chemiluminescence was reduced to the baseline level (Fig. 4). Hence, the camptothecin-induced increase in chemiluminescence as measured by the assay system was presumably mainly eected by H2O2 generated by a (plasma membrane) NADPH oxidase. Similarly, when camptothecin was added together with lanthanum chloride, no oxidative burst was measured, indicating that calcium acts upstream of the oxidative burst. In mammalian cells, caspases are involved in H2O2 release during PCD (Simizu et al. 1998). To investigate the hypothesis that caspase-like proteases are involved in the oxidative burst in plants, the caspase-1 inhibitor Z-Asp-CH2-DCB, an ecient inhibitor of camptothecin-induced cell death, was added together with camptothecin, and H2O2 was measured in the culture medium. As shown in Fig. 5, the caspase inhibitor eectively blocked the oxidative burst, suggesting a regulatory role for caspase-like proteases. Ethylene is a potentiator of the oxidative burst The observation that catalase inhibits camptothecin-induced cell death in the presence of additional ethylene suggests a role for H2O2 in ethylene-stimulated camptothecin-induced cell death. H2O2 in the culture medium 6 h after addition of camptothecin and ethylene was 86% higher than levels following treatment with camptothecin alone. Tomato suspension cells treated with ethylene alone exhibited a low, basal production similar to untreated cells (Fig. 6). To investigate if ethylene also plays a role in the oxidative burst in cells treated with camptothecin alone, ethylene synthesis was inhibited by addition of AVG. No oxidative burst occurred in the presence of AVG (Fig. 6). Similarly, in the presence of
Fig. 5 Camptothecin-induced H2O2 release can be inhibited by a caspase-speci®c peptide inhibitor. Tomato cells were left untreated (®lled diamonds), treated with 5 lM camptothecin (open squares) or with camptothecin + 0.1 lM Z-Asp-CH2-DCB (®lled triangles). H2O2 was estimated using luminol-dependent chemiluminescence. Data are the means SE of at least three independent experiments using dierent batches of cells
542
Fig. 6 Ethylene involvement in the release of H2O2 induced by camptothecin. Tomato cells were left untreated (®lled diamonds), treated with ethylene (100 ll/l in the head space) (®lled squares), 5 lM camptothecin (open squares), camptothecin + ethylene (crosses), camptothecin + 10 lM AVG (®lled triangles) or camptothecin + 20 lM STS (®lled circles). H2O2 was estimated using luminol-dependent chemiluminescence. Data are the meansSE of at least three independent experiments using dierent batches of cells
the ethylene-action inhibitor STS, no oxidative burst occurred upon camptothecin treatment either. Taken together, these results indicate that an oxidative burst is an essential part of the camptothecin-induced cell death pathway and that ethylene, although in itself not sucient to induce an oxidative burst, is an important potentiator.
Discussion In tomato suspension cells the anticancer drug camptothecin induces cell death with morphological characteristics of mammalian apoptosis, such as chromatin condensation, and nuclear and DNA fragmentation. This camptothecin-induced PCD can be inhibited by the addition of caspase-speci®c peptide inhibitors (De Jong et al. 2000). In this study we have shown that ethylene is indispensable for PCD in tomato suspension cells. Ethylene is a powerful potentiator of H2O2 accumulation and subsequent PCD in response to camptothecin. Ethylene has been reported previously to play an important role in regulating and modulating plant responses, including cell death, to both abiotic and biotic stresses. Ethylene is involved in the regulation of cell death during pea carpel senescence (Orzaez and Granell 1997), in hypoxia-induced aerenchyma formation in maize roots (He et al. 1996) and during PCD in maize endosperm development (Young and Gallie 2000). Although ethylene is not a prerequisite in cell death during the hypersensitive response (HR), ethylene was shown to promote cell death and to contribute to symptom formation in the susceptible response to virulent pathogens (Bent et al. 1992; Lund et al. 1998). The formation of HR-like lesions during ozone stress has the characteristics of PCD. Ozone forms ROS in the apoplasm and causes the plant cell itself to produce ROS in an
oxidative burst. In sensitive plants, the elevated level of stress-ethylene emission is correlated with the damage level, and if ethylene perception or biosynthesis is prevented, the damage formation is reduced. This suggests that both ROS and ethylene are involved in the induction of cell death in ozone-exposed plants (Overmyer et al. 2000). These observations are in line with our ®ndings that inhibition of ethylene action strongly reduces camptothecin-induced PCD, while exogenous ethylene strongly stimulates PCD. Hence, the regulation of camptothecin-induced PCD exhibits similarity to cell death processes caused by other triggers. Camptothecin-induced PCD in tomato suspension cells is accompanied by a release of ROS into the culture medium. Generation of ROS, such as superoxide (O2±), hydrogen peroxide (H2O2) and hydroxyl radicals (OH±) is a common phenomenon in many stress responses, such as hypo-osmotic stress, drought and cold shock (Foyer et al. 1997). ROS generation leads to cellular damage and ultimately cell death. Pathogen-defence responses are associated with ROS-generating systems based on plasma-membrane-bound NADPH oxidases or peroxidases (Lamb and Dixon 1997). Our observation that the NADPH-oxidase inhibitor DPI strongly inhibits camptothecin-induced cell death suggests that the release of ROS into the medium is mainly due to NADPH activity in the plasma membrane. However, it has been reported that DPI may also be able to inhibit the production of ROS by mitochondria (Li and Trush 1998). Hence, it cannot be excluded that ROS are also produced in mitochondria during camptothecin-induced PCD. Our ®nding that catalase inhibits camptothecininduced cell death suggests that H2O2 is responsible for the induction of PCD. H2O2 has been implicated in hypersensitive cell death. It has been shown that H2O2 drives the cross-linking of cell wall structural proteins and functions as a local trigger of PCD in challenged cells; moreover, H2O2 may act as a diusible signal for the induction of genes encoding cellular protectants in adjacent cells (Levine et al. 1994). H2O2 is a potent activator of certain MAPK cascades, such as those involving wound-induced protein kinase (WIPK), which are components of pathogen-defence signalling (Bolwell 1999). In addition, H2O2 was associated with PCD in barley aleurone cells (Bethke and Jones 2001). In animal cells, the role of H2O2 as an intracellular signal is well known. For example, it activates the NF-jB transcription factor, which mediates in¯ammatory, immune and acute-phase responses in diverse stress stimuli. Several plant disease-resistance genes share some homology with molecules involved in NF-jB-mediated responses (Lamb and Dixon 1997), indicating that similarities may exist between animal and plant stress signalling systems. Additional evidence for similarities between animal and plant stress signalling is the recent identi®cation of a plant homologue of PIRIN, a stabiliser of the NF-jBcomplex. Pirin expression during camptothecin-induced PCD in tomato suspension cells was highly induced,
543
indicating that NF-jB complexes may be activated during PCD (Orzaez et al. 2001). The role of ethylene in camptothecin-induced cell death is complex. Although ethylene in itself is unable to induce cell death, the low levels of (endogenously produced) ethylene are crucial for execution of camptothecin-induced cell death. Fumonisin-B1-induced apoptosis in Arabidopsis protoplasts (Asai et al. 2000) was found to depend on jasmonic acid, salicylic acid (SA) and ethylene signalling pathways. In accordance with our results, blocking ethylene perception using Ag+, or by using protoplasts from the ethylene-insensitive mutant ETR 1-1, impaired fumonisin-B1-induced cell death. Ethylene apparently provides a crucial co-factor for the production or maintenance of increased H2O2 levels. In this respect, the role of ethylene is very similar to the proposed role of SA in H2O2 production following pathogen recognition and cell death (Shirasu et al. 1997). SA is believed to stimulate H2O2 production in dierent ways; it may aect the phosphorylation status of signalling pathway components involved in the activation of NADPH oxidase (Zhang and Klessig 1997) and SA may inhibit directly the enzymes involved in breakdown of H2O2 such as catalase (Durner and Klessig 1995). Ethylene may therefore act through stimulation of NADPH oxidase in a similar manner, i.e. by in¯uencing phosphorylation events or by blocking the activity of ROS-inactivating enzymes such as catalases or ascorbate peroxidases. Similar to our observations in tomato cells, caspase activity is required for H2O2 production in camptothecin-induced apoptosis in human cells (Simizu et al. 1998). In many animal systems, apoptosis is dependent on the balance between pro- and anti-apoptotic factors, the latter aecting, for example, the activity of caspases by prevention of their activation through inhibition of the release of cytochrome c from mitochondria. As ethylene appears to greatly modify the sensitivity of the cells to camptothecin, ethylene may act as an inactivator of existing survival pathways. Simultaneous application of camptothecin and ethylene greatly enhances camptothecin-induced cell death, apparently as a result of the increased production of H2O2. Although inhibitors of H2O2 formation and accumulation are still eective in blocking cell death, protease and caspase inhibitors, even when applied at supra-optimal concentrations, are not. This indicates that at high ethylene concentrations, camptothecin apparently exerts its eect in a caspase-independent manner. Chen and Bleecker (1995) proposed a model for a dose-dependent ethylene response of growing Arabidopsis seedlings and the induction of chitinase expression. The key feature of this model is based on the idea that the various responses to ethylene are mediated by independent downstream pathways that operate over ranges of signal output that are dierent from those of the primary ethylene signal transduction pathway. In line with this assumption is our observation that ethylene-stimulated camptothecin-induced PCD can be reduced by the MAPK inhibitor PD98059, while
camptothecin-induced PCD, which requires low levels of ethylene, cannot be inhibited by this inhibitor. This indicates that other downstream pathways are activated upon addition of high levels of exogenous ethylene to camptothecin-treated cells. In cells treated with a sub-lethal concentration of the calcium ionophore A23187, cell death could be induced by addition of ethylene. This indicates that the caspase-independent cell death pathway depends on calcium and ethylene. Hence, the contribution of camptothecin in camptothecin-induced cell death at high ethylene concentrations may merely re¯ect its eect on ion ¯uxes, as described for various elicitors and signalling molecules (Felix et al. 1994). In conclusion, we show that camptothecin exerts its eect on cell death through a calcium- and caspasedependent stimulation of NADPH oxidase activity. In addition, we show that ethylene is an essential factor in camptothecin-induced PCD. Although in itself insucient to trigger cell death, exogenous ethylene greatly stimulates camptothecin-induced H2O2 production and
Fig. 7 A schematic model for camptothecin-induced cell death signalling in tomato suspension cells. Camptothecin treatment presumably induces ion ¯uxes that result in elevated calcium levels in the cytoplasm. In addition, DNA damage induced by camptothecin may, in combination with increased calcium levels, activate (caspase-like) proteases. Together with low levels of ethylene these factors activate the NADPH oxidase complex resulting in the initial production of ROS. The produced ROS may, e.g. through their eect on mitochondrial membrane potential, stimulate further production of ROS and subsequent downstream processes leading to apoptotic cell death. In the case of high ethylene concentrations, the NADPH oxidase complex is activated in a protease-independent way, presumably through activation of a MAP kinase cascade (as indicated by double arrows)
544
cell death. We propose two partly overlapping cell-death pathways (Fig. 7). These comprise one pathway involving caspases that requires low levels of ethylene and one caspase-independent pathway operative at high ethylene levels. The latter pathway presumably acts through MAPK-like proteins that are not essential in PCD at basal ethylene concentrations. Acknowledgement This work was supported by grants from Wageningen University & Research Center to E.Y. and V.K.
References Adams TL, Townsend JA (1983) A new procedure for increasing eciency of protoplast plating and cloning selection. Plant Cell Rep 2:165±168 Asai T, Stone JM, Heard JE, Kovtun Y, Yorgey P, Sheen J, Ausubel FM (2000) Fumonisin B1-induced cell death in Arabidopsis protoplasts requires jasmonate-, ethylene-, and salicylate-dependent signaling pathways. Plant Cell 12:1823±1835 Bent AF, Innes RW, Ecke JR, Staskawicz BJ (1992) Disease development in ethylene-insensitive Arabidopsis thaliana infected with virulent and avirulent Pseudomonas and Xanthomonas pathogens. Mol Plant-Microbe Interact 5:372±378 Bethke PC, Jones RL (2001) Cell death of barley aleurone protoplasts is mediated by reactive oxygen species. The Plant J 25:19±29 Bolwell GP (1999) Role of active oxygen species and NO in plant defence responses. Curr Opin Plant Biol 2:287±94 Chen QG, Bleecker AB (1995) Analysis of ethylene signal-transduction kinetics associated with seedling-growth response and chitinase induction in wild-type and mutant Arabidopsis. Plant Physiol 108:597±607 Chu ZL, Pio F, Xie Z, Welsh K, Krajewska M, Krajewksi S, Godzik A, Reed JC (2001) A novel enhancer of the Apaf1 apoptosome involved in cytochrome c-dependent caspase activation and apoptosis. J Biol Chem 271:9239±9245 De Jong AJ, Hoeberichts FA, Yakimova ET, Maximova E, Woltering EJ (2000) Chemical-induced apoptotic cell death in tomato cells: involvement of caspase-like proteases. Planta 211:656±662 Del Pozo O, Lam E (1998) Caspases and programmed cell death in the hypersensitive response of plants to pathogens. Curr Biol 8:1129±1132 Dion M, Chamberland H, St-Michel C, Plante M, Darveau A, Lafontaine JG, Brisson LF (1997) Detection of a homologue of bcl-2 in plant cells. Biochem Cell Biol 75:457±461 Durner J, Klessig DF (1995) Inhibition of ascorbate peroxidase by salicylic acid and 2,6-dichloroisonicotinic acid, two inducers of plant defense responses. Proc Natl Acad Sci USA 92:11312± 11316 Fan JZ, Zhu S, Sun YL, Zhai ZH (1999) Induction of apoptosis in puri®ed animal and plant nuclei by Xenopus egg extracts. Cell Res 9:79±90 Felix G, Regenass M, Spanu P, Boller T (1994) The protein phosphatase inhibitor calyculin A mimics elicitor action in plant cells and induces rapid hyperphosphorylation of speci®c proteins as revealed by pulse labeling with [33P]phosphate. Proc Natl Acad Sci USA 91:952±956 Foyer CH, Lopez-Delgado H, Dat JF, Scott IM (1997) Hydrogen peroxide- and glutathione-associated mechanisms of acclimatory stress tolerance and signalling. Physiol Plant 100:241±254 Grant JJ, Loake GJ (2000) Role of reactive oxygen intermediates and cognate redox signaling in disease resistance. Plant Physiol 124:21±29 Grosskopf DG, Felix G, Boller T (1990) K-252a inhibits the response of tomato cells to fungal elicitors and their microsomal protein kinase in vitro. FEBS Lett 275:177±180
Gunawardena AHLAN, Pearce DM, Jackson MB, Hawes CR, Evans DE (2001) Characterisation of programmed cell death during aerenchyma formation induced by ethylene or hypoxia in roots of maize (Zea mays L.). Planta 212:205±214 Hammond-Kosack KE, Jones JDG (1996) Resistance gene-dependent plant defense responses. Plant Cell 8:1773±1791 He C-J, Morgan PW, Drew MC (1996) Transduction of an ethylene signal is required for cell death and lysis in the root cortex of maize during aerenchyma formation induced by hypoxia. Plant Physiol 112:463±472 Jones AM, Dangl JL (1996) Logjam at the styx: programmed cell death in plants. Trends Plant Sci 4:114±119 Kaufmann SH (1998) Cell death induced by topoisomerase-targeted drugs: more questions than answers. Biochim Biophys Acta 1400:195±211 Kombrink E, Somssich IE (1995) Defense responses of plants to pathogens In: Andrews JH, Tommerup IC (eds) Advances in botanical research. Academic Press, New York, pp 1±34 Koornneef M, Hanhart CJ, Martinelli L (1987) A genetic analysis of cell culture traits in tomato. Theor Appl Gen 74:633±641 Korthout HAAJ, Berecki G, Bruin W, Van Duijn B, Wang M (2000) The presence and subcellular localization of caspase 3-like proteinases in plant cells. FEBS Lett 475:139±144 Lacomme C, Cruz SS (1999) Bax-induced cell death in tobacco is similar to the hypersensitive response. Proc Natl Acad Sci USA 96:7956±7961 Lamb C, Dixon RA (1997) The oxidative burst in plant disease resistance. Annu Rev Plant Physiol Plant Mol Biol 48:251±275 Levine A, Tenhaken R, Dixon R, Lamb C (1994) H2O2 from the oxidative burst orchestrates the plant hypersensitive disease resistance response. Cell 79:583±593 Li Y, Trush MA (1998) Diphenyleneiodonium, an NAD(P)H oxidase inhibitor, also potently inhibits mitochondrial reactive oxygen species production. Biochem Biophys Res Commun 253:295±9 Lund TS, Stall RE, Klee HJ (1998) Ethylene regulates the susceptible response to pathogen infection in tomato. Plant Cell 10:371±382 Luo X, Budihardjo I, Zou H, Slaughter C, Wang X (1998) Bid, a Bcl2 interacting protein, mediates cytochrome c release from mitochondria in response to activation of cell surface death receptors. Cell 94:481±490 Mitsuhara, I, Malik KA, Miura M, Ohashi Y (1999) Animal celldeath suppressors Bcl-xl and Ced-9 inhibit cell death in tobacco plants. Curr Biol 9:775±778 Mittler R, Lam E (1996) Sacri®ce in the face of the foes: pathogeninduced programmed cell death in plants. Trends Microbiol 4:10±15 OrzaÂez D, Granell A (1997) DNA fragmentation is regulated by ethylene during carpel senescence in Pisum sativum. Plant J 11:137±144 OrzaÂez D, AJ De Jong and EJ Woltering (2001) The tomato homolog of the human PIRIN is induced during programmed cell death. Plant Mol Biol 46:459±468 Overmyer K, Tuominen H, Kettunen R, Betz C, Langebartels C, Sandermann Jr H, KangasjaÈrvi J (2000) Ozone-sensitive Arabidopsis rcd1 mutant reveals opposite roles for ethylene and jasmonate signaling pathways in regulating superoxide-dependent cell death. Plant Cell 12:1849±1862 Pennell RI, Lamb C (1997) Programmed cell death in plants. Plant Cell 9:1157±1168 Rasmussen H, Goodman DBP (1977) Relationships between calcium and cyclic nucleotides in cell activation. Physiol Rev 57:421±509 Sanchez P, Torres Zabala de M, Grant M (2000) AtBI-1, a plant homologue of Bax inhibitor-1, suppresses Bax-induced cell death in yeast and is rapidly upregulated during wounding and pathogen challenge. Plant J 2:393±399 Schwacke R, Hager A (1992) Fungal elicitors induce a transient release of active oxygen species from cultured spruce cells that is dependent on Ca2+ and protein kinase activity. Planta 187:136± 141
545 Shirasu K, Nakajima H, Krishnamachari Rajasekhar V, Dixon RA, Lamb C (1997) Salicylic acid potentiates an agonist-dependent gain control that ampli®es pathogen signals in the activation of defense mechanisms. Plant Cell:261±270 Simizu S, Takada M, Umezawa K, Imoto M (1998) Requirement of caspase-3-like protease-mediated hydrogen peroxide production for apoptosis induced by various anticancer drugs. J Biol Chem 273:26900±26907 Steller H (1995) Mechanisms and genes of cellular suicide. Science 267:1445±1449 Sutherland MW (1991) The generation of oxygen radicals during host plant responses to infection. Physiol Mol Plant Pathol 39:79±93 Uren AG, O'Rourke K, Aravind L, Pisabarro MT, Seshagiri S, Koonin EV, Dixit VM (2000) Identi®cation of paracaspases and metacaspases. Two ancient families of caspase-like
proteins, one of which plays a key role in MALT lymphoma. Mol Cell 6:961±967 Warm E, Laties GG (1982) Quanti®cation of hydrogen peroxide in plant extracts by the chemiluminescence reaction with luminol. Phytochemistry 21:827±831 Wertz IE, Hanley MR (1996) Diverse molecular provocation of programmed cell death. Trends Biochem Sci 21:359±364 Woltering, EJ, De Jong, AJ, Yakimova, ET (1999) Apoptotic cell death in plants: the role of ethylene. In: A.K. Kanellis, Chang C, Klee H, Bleecker AB, Pech JC, Grierson D (eds), Biology and biotechnology of the plant hormone ethylene, II. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp 209±216 Young TE, Gallie DR (2000) Programmed cell death during endosperm development. Plant Mol Biol 44:283±301 Zhang S, Klessig DF (1997) Salicylic acid activates a 48-kD MAP kinase in tobacco. Plant Cell 9:809±824
