Environmental Stress Causes Oxidative Damage to Plant ...

THE JOURNAL OF BIOLOGICAL CHEMISTRY © 2002 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 277, No. 45, Issue of November 8, pp. 42663–42668, 2002 Printed in U.S.A.

Environmental Stress Causes Oxidative Damage to Plant Mitochondria Leading to Inhibition of Glycine Decarboxylase* Received for publication, May 15, 2002, and in revised form, August 29, 2002 Published, JBC Papers in Press, September 3, 2002, DOI 10.1074/jbc.M204761200

Nicolas L. Taylor‡, David A. Day§, and A. Harvey Millar§¶ From the Plant Molecular Biology Group, School of Biomedical and Chemical Sciences, Faculty of Life and Physical Sciences, The University of Western Australia, Crawley, WA 6009, Western Australia, Australia

A cytotoxic product of lipid peroxidation, 4-hydroxy2-nonenal (HNE), rapidly inhibited glycine, malate/ pyruvate, and 2-oxoglutarate-dependent O2 consumption by pea leaf mitochondria. Dose- and timedependence of inhibition showed that glycine oxidation was the most severely affected with a K0.5 of 30 !M. Several mitochondrial proteins containing lipoic acid moieties differentially lost their reactivity to a lipoic acid antibody following HNE treatment. The most dramatic loss of antigenicity was seen with the 17-kDa glycine decarboxylase complex (GDC) H-protein, which was correlated with the loss of glycine-dependent O2 consumption. Paraquat treatment of pea seedlings induced lipid peroxidation, which resulted in the rapid loss of glycine-dependent respiration and loss of Hprotein reactivity with lipoic acid antibodies. Pea plants exposed to chilling and water deficit responded similarly. In contrast, the damage to other lipoic acidcontaining mitochondrial enzymes was minor under these conditions. The implication of the acute sensitivity of glycine decarboxylase complex H-protein to lipid peroxidation products is discussed in the context of photorespiration and potential repair mechanisms in plant mitochondria.

The polyunsaturated fatty acids of membrane lipids are susceptible to reactive oxygen species (ROS)1-induced peroxidation yielding various aldehydes, alkenals, and hydroxyalkenals, including the cytotoxic compounds malonaldehyde (MDA) and 4-hydroxy-2-nonenal (HNE). These two end products are commonly measured in studies of lipid peroxidation by the thiobarbituric acid-reactive substances (TBARS) assay (1). Interest in the biological effect of HNE was stimulated by studies showing cytotoxicity due to rapid reaction with sulfydryl groups via Michael addition (2). A number of stress conditions, including cardiac reperfusion injury, increase ROS production and the rate of membrane peroxidation and ultimately lead to inhibition of the respiratory rate in mammals (3). HNE * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ‡ Recipient of a University of Western Australia postgraduate scholarship. § Supported by the Australian Research Council Discovery Program. ¶ To whom correspondence should be addressed. Tel.: 61-8-93807245; Fax: 61-8-93801148; E-mail: [email protected]. 1 The abbreviations used are: ROS, reactive oxygen species; AOX, alternative oxidase; GDC, glycine decarboxylase complex; HNE, 4-hydroxy-2-nonenal; MDA, malondialdehyde; OGDC, 2-oxoglutarate dehydrogenase complex; PDC, pyruvate dehydrogenase complex; TBARS, thiobarbituric acid-reactive substances; TES, N-Tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid. This paper is available on line at http://www.jbc.org

has been shown to directly inhibit respiration of isolated mammalian mitochondria through modification of the lipoic acid moieties of 2-oxo acid dehydrogenases, forming HNE-Michael adducts (Fig. 1) (4). These targets have also been shown to be damaged in vivo under conditions that induce lipid peroxidation (5). A range of biotic and abiotic stresses also raise ROS levels in plants due to perturbations of chloroplastic and mitochondrial metabolism and the generation of ROS in defense responses (6). Such accumulation of ROS in plants will clearly result in a wide variety of deleterious effects through oxidation reactions involving proteins, lipids, and nucleic acids. Lipid peroxidation in plants has been known for many years, but direct evidence of the pathways involved and the identification of HNE production in plants has only been reported recently (7). However, the exact molecular targets, the relative sensitivity of target enzymes, the mechanisms of repair, and the impact on plant function have not been extensively investigated. Research on the impact of plant lipid peroxidation on metabolism has centered on investigations of sensitive reactions in the chloroplast and the antioxidant systems present in this organelle (8, 9). Our understanding of the impact of lipid peroxidation on plant mitochondrial functions is still very limited. Recently we have shown that HNE specifically inhibits both pyruvate dehydrogenase (PDC) and 2-oxoglutarate dehydrogenase (OGDC) complexes through modification of lipoic acid moieties (Fig. 1) in mitochondria from potato tubers. However, in photosynthetic plant tissues, these tricarboxylic acid cycle enzymes are dwarfed by the presence of the glycine decarboxylase complex (GDC). This enzyme also contains a lipoic acid moiety, can account for up to 50% of matrix protein, and is responsible for the most prominent metabolic activity in the mitochondria of illuminated leaves, photorespiration (10). GDC is a multienzyme complex composed of four component enzymes, the P-, H-, T-, and L-proteins and is responsible for the conversion of glycine produced in the peroxisome to serine in the mitochondria during operation of the photorespiratory cycle (11). The H-protein plays a pivotal role as a mobile substrate that commutes between the other subunits, allowing its lipoic acid “arm” to visit the active sites of the other three components (10). Here we show that HNE rapidly inhibits glycine-dependent respiration of isolated pea leaf mitochondria and that the site of this inhibition is the H-protein of GDC. This site is much more sensitive to oxidative damage than those of related 2-oxo acid dehydrogenases. Further, we show that in vivo lipid peroxidation, induced by the herbicide paraquat and by the environmental stresses of chilling and water deficit, specifically inhibits the ability of mitochondria to oxidize glycine. GDC is, therefore, a major target for oxidative damage in leaf mitochondria.

42663

42664

Environmental Stress Inhibits Glycine Decarboxylase

FIG. 1. Reaction of HNE with lipoic acid, forming lipoic acidHNE Michael adducts. EXPERIMENTAL PROCEDURES

Materials—Tryquat 200 (Crop Care Australasia Pty. Ltd., Pinkenba, Australia) is a commercially available herbicide combination of paraquat (437.5 mg/L) and diquat (225 mg/L). Studies on the mode of action of these two components appear to indicate that both function identically (12), and this combination is referred to here simply as paraquat. HNE was obtained from Calbiochem, and all other chemicals were at least reagent-grade. Pea (Pisum sativum L. cv Green Feast) plants were germinated in vermiculite and grown in controlled environment chambers with a light intensity of 700 !mol/m2/sec at 24 °C and 65% humidity for 10 days on a 16/8-hour day/night cycle. Isolation of Pea Mitochondria—Mitochondria were isolated according to the methods of Day et al. (13) and Zhang and Wiskich (14) from !60 g of leaves. Leaves were disrupted with a Polytron (Kinematica, Kriens, Switzerland) in 250 ml of cold extraction medium (0.3 M sucrose, 25 mM tetra-sodium pyrophosphate, 10 mM KH2PO4, 2 mM EDTA, 1 mM glycine, 1% (w/v) PVP-40, 1% (w/v) bovine serum albumin, 20 mM ascorbate, pH 7.5). The homogenate was filtered through 4 layers of miracloth and centrifuged for 5 min at 1100 " g. The supernatant was centrifuged for 20 min at 18,000 " g and the pellet resuspended in 200 ml of wash medium (0.3 M sucrose, 10 mM TES, 1 mM glycine, 0.1% (w/v) bovine serum albumin, pH 7.5) and centrifuged for 5 min at 1100 " g. The supernatant was centrifuged for 20 min at 18,000 " g and the pellet resuspended in 10 ml of wash medium. 5-ml aliquots were then layered over 27.5 ml of solution (0.3 M sucrose, 10 mM TES, 1 mM glycine, 0.1% (w/v) bovine serum albumin, 28% (v/v) Percoll, and a linear gradient of 0 –10% (w/v) PVP-40, pH 7.5) in a centrifuge tube and centrifuged for 40 min at 40,000 " g. The mitochondria were found as a tight light yellow-brown band near the bottom of the tube. The mitochondrial fraction was removed and diluted in 250 ml of wash medium and centrifuged at 31,000 " g for 15 min. The supernatant was removed, and this wash was repeated. The final mitochondrial pellet was resuspended in !1 ml of wash medium. In Vitro HNE Treatment—Isolated pea mitochondria were treated with HNE and a number of co-factors and substrates before assays were carried out. Mitochondria for glycine oxidation assays were incubated for 60 min on ice with 10 mM glycine and 0.5 mM ATP to maximize GDC activity as described by Zhang and Wiskich (14). These mitochondria were then incubated with 250 !l of assay buffer, glycine (10 mM), NAD (0.5 mM), ATP (0.5 mM), ADP (0.25 mM), and various concentrations of HNE (1.5–250 !M) for 10 min. 750 !l of assay buffer was then added and electrode traces started, with additions of ADP at appropriate times. For pyruvate (# malate) oxidation assays, mitochondria were incubated for 10 min with pyruvate (10 mM), malate (1 mM), NAD (0.5 mM), TPP (0.05 mM), CoA (0.06 mM), ATP (0.5 mM), ADP (0.25 mM), and HNE (1.5–250 !M). For 2-oxoglutarate oxidation assays, mitochondria were incubated with 2-oxoglutarate (10 mM), NAD (0.5 mM), TPP (0.05 mM), CoA (0.06 mM), ATP (0.5 mM), ADP (0.25 mM), and HNE (1.5–250 !M). Again, 750 !l of assay buffer was then added and electrode traces started, with additions of ADP at appropriate times. Abiotic Treatments—Plants were sprayed evenly with paraquat (662.5 mg/L), allowed to dry, and then returned to controlled environment chambers and exposed to light until harvest. For low temperature treatment, plants were placed at 4 °C for 36 h prior to harvest while maintaining a normal day/night light cycle. For drought treatment,

plants were not watered for 7 days prior to harvest while maintaining a normal humidity, temperature, and day/night light cycle. Oxygen Consumption—O2 consumption was measured in an O2 electrode (Rank Bros., Cambridge, UK) in 1 ml of reaction medium containing 0.3 M sucrose, 10 mM TES-KOH, pH 7.5, 5 mM KH2PO4, 10 mM NaCl, 2 mM MgSO4, and 0.1% (w/v) bovine serum albumin. Glycine (10 mM), pyruvate (10 mM), malate (0.5 mM), 2-oxoglutarate (10 mM), succinate (10 mM), NADH (1 mM), NAD (0.5 mM), TPP (0.05 mM), CoA (0.06 mM), ATP (0.5 mM), ADP (0.1–1 mM), and HNE (1.5–250 !M) were added as indicated. Protein concentrations were determined by the method of Peterson (15) using bovine serum albumin as standard. TBARS Assay—We used the method of Hodges et al. (1), which takes into account the presence of anthocyanins and sucrose. Approximately 1 g of tissue was homogenized in 80% (v/v) ethanol and inert sand in a pestle and mortar, followed by centrifugation at 3000 " g for 10 min. Six 1-ml aliquots of the supernatant were placed in 2-ml cryovials and 1 ml of $TBA solution (20% (w/v) trichloroacetic acid, 0.01% (w/v) butylated hydroxytoluene) was added to three tubes and #TBA (0.65% (w/v) thiobarbituric acid, 20% (w/v) trichloroacetic acid, 0.01% (w/v) butylated hydroxytoluene) added to the other three. The samples were vortexed and heated to 95 °C for 25 min, followed by cooling on ice for 5 min. The samples were then spun at 3000 " g for 10 min and absorbances of each sample read at 440, 532, and 600 nm. The amount of MDA equivalents was calculated from Equation 1, MDA equivalents (nmol ml$1) " %A-B/157000)106

(Eq. 1)

where A & [(A532#TBA)-(A600#TBA)-(A532-TBA-A600-TBA)] and B & [(A440#TBA-A600#TBA)0.0571]. Western Blotting—Mitochondria from in vitro assays were treated with cysteine (10 mM) to remove any unreacted HNE and spun at 17,000 " g for 10 min. The supernatant was removed and the pellet resuspended in the original mitochondria volume (20 !l). 5 !l of mitochondrial proteins were separated by electrophoresis under denaturing reducing conditions on 0.1% (w/v) SDS-12% (w/v) polyacrylamide gels according to Laemmli (16). Similarly, mitochondria from abiotic stress treatments were electrophoresed under denaturing, reducing conditions on 0.1% (w/v) SDS-12% (w/v) polyacrylamide gels according to Laemmli (16). For immunoreaction experiments, proteins were electroblotted from SDS-PAGE gels onto nitrocellulose membranes and blocked. 1/10,000 dilution of the anti-lipoic acid serum (17), 1/1000 dilution of the anti-H-protein of GDC serum, and 1/50 dilution of the anti-alternative oxidase (AOX) serum (18) were used as primary antibodies. Chemiluminescence was used for detection of horseradish peroxidase-conjugated secondary antibodies and visualized using a LAS 1000 (Fuji). The blots were quantified using the Image Gauge v3.0 software (Fuji) with the control band denoted as 100%; the other bands were calculated relative to that value. Two-dimensional Gel Protein Separation and Mass Spectrometry— Two-dimensional isoelectric focusing/SDS-PAGE of pea mitochondrial samples was performed according to Millar et al. (19). Western blotting and detection were performed as above. Quadrupole time-of-flight mass spectrometry (Q-TOF MS/MS) was performed on an Applied Biosystems Q-STAR Pulsar using an IonSpray source. Proteins to be analyzed were cut from the two-dimensional PAGE gel, dried at 50 °C in a dry block heater, and stored at $70 °C. For the sequencing analysis, the proteins were digested with trypsin (19) and injected in 50% acetonitrile, and selected doubly charged peptides were fragmented by N2 collision and analyzed by MS/MS. Mass spectra and collision MS/MS data were analyzed with BioAnalyst software (Applied Biosystems, Sydney, Australia). Data is presented as the m/z of the peptide followed by its charge state and the deduced amino acid sequence (X & undetermined; (L/I) & leucine or isoleucine residue indistinguishable on mass). RESULTS

Effect of HNE on Pea Mitochondrial Respiratory Rates with Different Substrates—Recent demonstrations of inhibition of lipoic acid-containing tricarboxylic acid cycle enzymes by HNE, in both mammalian (4, 17) and plant (20) mitochondria, prompted us to investigate the effect of this lipid peroxidation product on the photorespiratory enzyme, GDC, in pea leaves. Isolated pea leaf mitochondria were incubated for 10 min in the presence of ADP, cofactors, various substrates (including glycine), and HNE at concentrations from 0 to 250 !M. Following incubation, O2 consumption was recorded (Fig. 2A). At concentrations up to 50 !M, HNE had no significant effect on pyru-


FIG. 2. Effect of varying HNE concentration on substrate-dependent respiration rate of isolated pea mitochondria. A, rates of mitochondrial oxygen consumption with various concentrations of HNE (0 –250 !M) and different substrates, glycine (‚), pyruvate (# malate) (!), 2-oxoglutarate (E), and all co-factors required for activity. B, rates of mitochondrial oxygen consumption with 85 !M HNE and different substrates, glycine (‚), pyruvate (# malate) (!), 2-oxoglutarate (E), and all co-factors required for activity following incubations of various time periods. The values shown are the state 3 rates in the presence of ADP compared with control plants. Control rates for O2 consumption stimulated by glycine, 163.0 ' 18.0 nmol O2 min$1 mg$1 protein; pyruvate (# malate), 231.7 ' 21.1 nmol O2 min$1 mg$1 protein; and 2-oxoglutarate, 114.3 ' 13.5 nmol O2 min$1 mg$1 protein. These values were obtained from at least three treatments (mean ' S.E.).

vate- or 2-oxoglutarate-dependent respiration but inhibited GDC by 75%. At 250 !M HNE, pyruvate and 2-oxoglutaratedependent respiration decreased by !45%, while glycine-dependent respiration was inhibited by more than 90%. To learn more about the nature of this inhibition, we incubated isolated mitochondria with 85 !M HNE in the presence of cofactors and various substrates for different lengths of time before measuring O2 consumption (Fig. 2B). After 30 min of incubation, 2-oxoglutarate-dependent respiration was decreased by !10%, pyruvate-dependent O2 consumption by !30%, and glycine-dependent respiration by more than 80%. The data presented in Fig. 2B indicate that the inhibition of respiration by HNE increases with time. This means that oxidative damage will be cumulative in vivo. We also showed that free lipoic acid could protect these lipoic acid-containing proteins from inhibition by HNE. Inclusion of 500 !M free lipoic acid during incubation of isolated mitochondria with 85 !M HNE for 10 min decreased inhibition to only 20% (data not shown), compared with 80% inhibition in incubations without free lipoic acid (Fig. 2A). Immunodetection of SDS-PAGE-separated pea mitochondrial proteins, with antibodies raised to lipoic acid, showed four protein bands of apparent molecular masses 78, 54, 50, and 17 kDa (Fig. 3). Previous reports in potato identified these proteins, respectively, as the two acetyltransferase subunits of PDC, the succinyltransferase subunit of OGDC, and the Hprotein of GDC. The GDC H-protein (!17 kDa) can be distinguished from the acyl carrier protein (!14 kDa) by size and antibody reaction and from the ! 50-kDa subunits of PDC and OGDC by comparison with the purified PDC and OGDC from potato (20). Previous work on partially purified pea PDC has shown that this enzyme also contains two acetyltransferase

42665

FIG. 3. Effect of varying HNE concentration on anti-lipoic acid antibody reactivity and anti-H-protein reactivity of pea mitochondrial proteins. Mitochondria from in vitro assays (Fig. 2) were treated to remove unreacted HNE, and proteins were separated by SDS-PAGE. Numbers on the left represent apparent molecular masses in kDa, numbers above the blots represent concentrations of HNE, and those below the blots represent relative band intensity. The 78- and 54-kDa bands are the E2 subunits of PDC, the 50-kDa is the E2 of OGDC, and the 17-kDa (Lip A) is the H-protein of GDC, all using the anti-lipoic acid antibody (17). The H-Pro panel (17 kDa) is the H-protein of GDC detected with an anti-H-protein antibody.

subunits of !80 and 55 kDa (21, 22) and that a contaminating 50-kDa protein was likely to be the succinyltransferase of pea OGDC (21). To confirm that the lower of the two !50-kDa proteins in pea is indeed the succinyltransferase of OGDC and that the 17-kDa protein is the H-protein of GDC, we undertook two-dimensional gel separation and mass spectrometry identification of these lipoic acid-containing proteins. Three peptides derived from the 17-kDa protein (849.48 [M#2H]2#, (L/I)KPTSPDE(L/I)ESS(L/I)(L/I)GAK; 558.77 [M#3H]2#, XAPSHEWVK; 423.27 [M#2H]2#, XXSNV(L/I)DG(L/I)K) were identical to the H-protein sequence of GDC from P. sativum (GenBankTM accession number 1923203A). Three peptides derived from the 50-kDa protein (639.34 [M#3H]2#, PSEPQ(L/I)PPKER; 696.38 [M#2H]2#, DY(L/I)D(L/I)S(L/I)AVGTPK; 525.32 [M#2H]2#, APVVGGVVVPR) matched 73– 85% with a dihydrolipoamide S-succinyltransferase from Arabidopsis thaliana (GenBankTM accession number T04814). Having confirmed these identities, aliquots of mitochondria were sampled during oxidation of different substrates in the presence of added HNE, treated with cysteine (10 mM) to quench any remaining HNE, separated by SDS-PAGE, electroblotted, and probed with antibodies raised against lipoic acid and against the pea H-protein (Fig. 3). Anti-lipoic acid antibodies revealed a decrease in antigenicity of PDC and OGDC E2 subunits upon treatment with HNE, consistent with the declines in respiratory rate noted in Fig. 2A. The decline in lipoic acid antigenicity of the H-protein was much more rapid. A plot of anti-lipoic acid antibody reactivity of H-protein versus glycinedependent respiration rates showed a strong positive correlation with an initial slope of 0.9 (r2 & 0.9352) (analysis not shown). This suggests that the amount of active lipoic acid associated with H-protein has a very high control coefficient over the activity of GDC in intact mitochondria. Immunoblotting with antibodies raised to the pea H-protein itself revealed

42666


FIG. 4. In vivo paraquat-induced lipid peroxidation and the impact on mitochondrial respiration rate. Plants were treated with paraquat and mitochondria isolated at the times shown. Oxygen uptake was measured with the following substrates: glycine (‚), pyruvate (# malate) (!), 2-oxoglutarate (E), and all co-factors required for activity. The values shown are the state 3 rates in the presence of ADP compared with control plants. ", MDA equivalents measured using the TBARS assay in whole leaf extracts at the various time points after herbicide treatment, compared with controls.

that there was no direct loss of H-protein following HNE addition. At very high HNE concentrations, when nearly all lipoic acid had been modified, H-protein began to separate into several bands on SDS-PAGE (Fig. 3), suggesting the formation of further modifications of H-protein polypeptides. These forms, which appear to have a higher molecular mass, may arise from the attachment of HNE to amino acid side chains of H-protein or covalent attachment of an unknown compound produced at high HNE concentrations. The exact nature of these modified forms of the H-protein is yet to be confirmed. In Vivo Lipid Peroxidation Induced in Pea Leaves by Paraquat—The herbicide paraquat acts as an artificial electron acceptor from chloroplast Photosystem I (12) and also inhibits and uncouples the mitochondrial electron transport chain (23), thereby exacerbating the production of superoxide radicals in both locations. Through a cascade of reactions stimulated by transition metals (24) and by reaction with hydrogen peroxide, this produces hydroxyl radicals, which initiate lipid peroxidation (25). We used paraquat to induce oxidative stress in pea leaves. Pea plants sprayed with paraquat (662.5 mg/L) were maintained in the light, and leaves were harvested at 0, 2, 4, 6, 8, 10, or 12 h later. Leaves from all times points were used for TBARS assays to assess the abundance of lipid peroxidation end products (MDA equivalents). Leaves from 0, 2, 6, and 12 h were also used for isolation of mitochondria for assessment of respiratory activities (Fig. 4). MDA equivalents remained relatively constant during the first 6 h before rising 4-fold and peaking at 10 h post-treatment. Pyruvate- and 2-oxoglutaratedependent respiration were unaffected by this treatment of leaves, but glycine-dependent respiration declined sharply to !40% of control after 12 h. Immunoblotting with the same antibodies used in Fig. 3 revealed no change in PDC or OGDC E2 subunit reactivity, whereas H-protein lipoic acid reactivity decreased to 44% of the control (Fig. 5). The amount of Hprotein per se did not change during the first 6 h but did decrease marginally between 6 and 12 h. Concomitant with the appearance of MDA following paraquat treatment, AOX protein, which initially was expressed at very low levels, increased dramatically (Fig. 5). AOX is often proposed as a preoxidant defense in plants, induced to prevent radical formation from the mitochondrial electron transport pathway (26 –28). Its induction in pea leaves by paraquat application agrees with this idea, but apparently it was unable to protect GDC from damage.

FIG. 5. Effect of paraquat on anti-lipoic acid antibody, anti-Hprotein, and anti-AOX antibody reactivity against pea mitochondrial proteins. Mitochondria were isolated from herbicidetreated pea plants at the times shown, and their proteins were separated by SDS-PAGE. Numbers on the left represent apparent molecular masses in kDa, those above the blots represent time of herbicide treatment, and those below the blots represent relative band intensity. The 78- and 54-kDa bands are the E2 subunits of PDC, the 50-kDa is the E2 of OGDC, and the 17-kDa (Lip A) is the H-protein of GDC, all using the anti-lipoic acid antibody (17). The H-Pro panel (17 kDa) is the H-protein of GDC detected with an anti-H-protein antibody. The 31and 32-kDa bands in the AOX panel are AOX proteins detected with the anti-AOX antibody (18).

In Vivo Lipid Peroxidation Induced in Pea Leaves by Environmental Stress—A number of environmental stresses have been shown to induce oxidative stress in plants. These include salinity (29, 30), chilling (31), and water deficit (32). The mechanism of the production of ROS under these conditions remains unclear, but measurements of both ROS and lipid peroxidation products, as well as responses of antioxidant enzymes, have confirmed the oxidative state of the exposed plants. We investigated the effect of two such environmental stresses on pea leaf mitochondria: low temperature and drought. For the low temperature treatment, plants were chilled at 4 °C for 36 h prior to harvest. For drought treatment, plants were waterrestricted for 7 days prior to harvest. Leaves were subsequently harvested from three trays of plants from each treatment, as well as from three trays of 12-h paraquat-treated peas, and used for isolation of mitochondria for assessment of respiratory activities (Fig. 6B). Glycine-dependent respiration was significantly inhibited by all treatments: 58% by low temperature, 63% by water stress, and 72% by the herbicide treatment. Thesesamemitochondriawerealsoassayedforsuccinate/NADHdependent O2 consumption, with the treatment groups showing no significant difference to the control (p ( 0.3, data not


FIG. 6. Effect of environmental stresses on lipoic acid H-protein moieties of GDC and glycine-dependent respiration in isolated mitochondria. A, mitochondria were isolated from stresstreated pea plants and their proteins separated by SDS-PAGE. Numbers on the left represent apparent molecular masses in kDa and those below the blots the percentage intensity of band relative to the control. The Lip A panel (17 kDa) is the H-protein of GDC detected using an anti-lipoic acid antibody (17). The H-Pro panel (17 kDa) is the H-protein of GDC detected with an anti-H-protein antibody. The 31and 32-kDa bands in the AOX panel are AOX proteins detected with the anti-AOX antibody (18). B, rate of oxygen consumption by mitochondria isolated from the leaves of stress-treated pea plants with glycine as substrate. Values shown are the state 3 rates in the presence of ADP and are the average of three treatments (mean ' S.E.). All three treatments are significantly less than the control (p )0.05).

shown). Immunoblotting with the same antibodies used in Fig. 3 revealed only minor changes in PDC or OGDC E2 subunit reactivity, whereas H-protein lipoic acid reactivity decreased to 49% of the control in chilled plants, 17% in water-stressed plants, and 21% in the herbicide-treated plants. (Fig. 6A). The amount of H-protein per se did not change significantly in any of the treatments. The effect of these treatments on AOX protein was also examined. Its induction in pea leaves was associated with the oxidative conditions generated by the environmental stresses but, as in the herbicide-treated plants, its induction did not protect GDC from damage. DISCUSSION

Plant mitochondrial lipid peroxidation has been reported previously (33–35), but the functional implications of this damage for mitochondrial operation was not clear from these studies. It has also been shown previously that HNE has a significant inhibitory effect on tricarboxylic acid cycle function through modification of lipoic acid moieties of key enzymes (4, 17, 20). Here we have shown for the first time that HNE has an even greater inhibitory effect on the lipoic acid-containing en-

42667

zyme, glycine decarboxylase, a key enzyme involved in photorespiration in photosynthetic plant tissues and essential for the regeneration of Calvin cycle substrates in the photorespiratory pathway. The effect of HNE was much more dramatic on glycinedependent respiration than on tricarboxylic acid cycle-dependent respiration, suggesting that GDC is particularly sensitive to HNE modification and may be a key target during oxidative stress in photosynthetic plant tissues. Consistent with this suggestion, we observed a substantial decline in GDC activity following in vivo oxidative stress induced by paraquat treatment, by chilling, and by water deficit. Taken together, the results with isolated mitochondria and intact plants suggest that environmental stress leads directly to lipid peroxidation, the products of which can inhibit mitochondrial function and, in particular, photorespiratory metabolism. In pea plants treated with paraquat, MDA equivalents were detected only after several hours. Lipid peroxidation is expected to commence immediately following paraquat application; the delayed detection of MDA accumulation may indicate that different aldehyde end products take varying times to accumulate to measurable levels because of the presence and function of detoxification pathways in the plant. However, as time passes, these defenses are presumably overwhelmed and lipid peroxidation products rise sharply. The high degree of damage to the H-protein of GDC before MDA equivalents were detected by the methods employed here suggests extreme susceptibility of the H-protein subunit to low concentrations of lipid peroxidation end products, before dilution and/or detoxification takes effect. At this early time point, there were no noticeable effects on plant appearance by the treatments. In this context, GDC inhibition may be an early marker of oxidative damage in photosynthetic plant cells. Glycine metabolism by GDC is an essential step in the photorespiratory cycle, and its inhibition or suppression in leaves is lethal under ambient conditions (36). The extreme sensitivity of GDC to the treatments applied here suggests, therefore, that photorespiration is among the first casualties of oxidative damage in leaves. The mechanisms for detoxifying cytotoxic lipid peroxidation products in plants are only beginning to be understood. Recently, an aldehyde reductase (37) has been identified in both the mitochondria and the cytoplasm that reacts with HNE. A glutathione S-transferase, which uses HNE as a substrate, has been identified in mammalian mitochondria (38, 39) and in the monocot Sorghum bicolour (40). It has also been proposed that a recently identified maize T cytoplasm cytoplasmic male sterility (CMS) restorer gene, rf2, which encodes an aldehyde dehydrogenase, may act in mitochondria to remove the end products of lipid peroxidation such as MDA and HNE (41, 42). In the leaves of the pea plants used in the present study, the putative anti-oxidant protein, AOX, was induced about 6 h after herbicide treatment, coincidentally with the observed rise in MDA (Fig. 5). This suggests three things. First, it shows that nuclear-encoded mitochondrial defense proteins can be readily made and imported during the extreme oxidative stress following paraquat treatment, indicating that this treatment does not simply lead to rapid death and necrosis of the plant tissues studied. Second, this induction of AOX correlated with total cell lipid damage rather than the mitochondrial-specific damage of GDC, raising some questions about the degree to which AOX induction is due solely to radical formation in mitochondria. Third, it shows that such gene induction-based defense of mitochondria, at least as exemplified by AOX, is too slow to save highly susceptible sites such as the H-protein of GDC. However, induction of AOX may have helped to delay damage to the less susceptible tricarboxylic acid enzymes.

42668


We have previously postulated (20) that the well documented inhibition of PDC by phosphorylation in the light (43) might aid in saving the lipoic acid moieties of this enzyme during higher rates of lipid peroxidation likely to be experienced during photosynthesis. There would be no such reprieve for GDC, because glycine is likely to be the dominant respiratory substrate in the light under photorespiratory conditions. Rapid catalytic turnover of GDC will provide ready availability of the reduced form of lipoic acid, which is required for modification by HNE (4). The identification of pathways for lipoic acid attachment to acyltransferase subunits in mitochondria (44) and the discovery of a pathway for de novo lipoic acid synthesis in plant mitochondria from photosynthetic tissues (45) have largely been viewed in terms of the biogenesis of matrix enzymes. However, an enzyme able to remove damaged lipoic acid, regenerating the apoprotein and allowing reattachment of an unmodified lipoate to Lys-63 of P. sativum H-protein, a socalled but as yet undiscovered ‘lipoate lyase,’ would allow this biogenesis pathway to play an equally vital role in damage repair. Such an approach would eliminate the need for complicated removal and reassembly of E2 subunits in large oligometric complexes and greatly reduce the cost of lipoic acid modification. Acknowledgements—We thank Dr. K. M. Humphries and Prof. P. I. Szweda, Department of Physiology and Biophysics, School of Medicine, Case Western Reserve University, Cleveland, OH for the gift of antilipoic acid antibodies. Dr. S. Rawsthorne, Brassica & Oilseeds Research Department, John Innes Centre, Norwich, UK is thanked for the gift of anti-H-protein antibodies, and Dr. T. E. Elthon, Department of Biological Sciences, University of Nebraska, Lincoln, NE is thanked for the gift of anti-AOX antibodies. Dr. Joshua Heazlewood, University of Western Australia, is thanked for assistance with mass spectrometry. Pia Sappl, University of Western Australia, is thanked for assistance during initial investigations. REFERENCES 1. Hodges, D. M., DeLong, J. M., Forney, C. F., and Prange, R. K. (1999) Planta 207, 604 – 611 2. Esterbauer, H., Schaur, R. J., and Zollner, H. (1991) Free Radic. Biol. Med. 11, 81–128 3. Lucas, D. T., and Szweda, L. I. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 510 –514 4. Humphries, K. M., Yoo, Y., and Szweda, L. I. (1998) Biochemistry 37, 552–557 5. Lucas, D. T., and Szweda, L. I. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 6689 – 6693 6. Van Camp, W., Van Montagu, M., and Inze, D. (1998) Trends Plant Sci. 3, 330 –334 7. Takamura, H., and Gardner, H. W. (1996) Biochim. Biophys. Acta 1303, 83–91 8. Mano, J., Ohno, C., Domae, Y., and Asada, K. (2001) Biochim. Biophys. Acta

1504, 275–287 9. Kraus, T. E., McKersie, B. D., and Fletcher, R. A. (1995) J. Plant Physiol. 145, 570 –576 10. Douce, R., Bourguignon, J., Neuburger, M., and Rebeille, F. (2001) Trends Plant Sci. 6, 167–176 11. Vauclare, P., Diallo, N., Bourguignon, J., Macherel, D., and Douce, R. (1996) Plant Physiol. 112, 1523–1530 12. Dodge, A. D. (1971) Endeavour 30, 130 –135 13. Day, D. A., Neuburger, M., and Douce, R. (1985) Aust. J. Plant Physiol. 12, 219 –228 14. Zhang, Q., and Wiskich, J. T. (1995) Arch. Biochem. Biophys. 320, 250 –256 15. Peterson, G. L. (1977) Anal. Biochem. 83, 346 –356 16. Laemmli, U. K. (1970) Nature 227, 680 – 685 17. Humphries, K. M., and Szweda, L. I. (1998) Biochemistry 37, 15835–15841 18. Elthon, T. E., Nickels, R. L., and McIntosh, L. (1989) Plant Physiol. 89, 1311–1317 19. Millar, A. H., Sweetlove, L. J., Giege, P., and Leaver, C. J. (2001) Plant Physiol. 127, 1711–1727 20. Millar, A. H., and Leaver, C. J. (2000) FEBS Lett. 481, 117–121 21. Millar, A. H., Leaver, C. J., and Hill, S. A. (1999) Eur. J. Biochem. 264, 973–981 22. Thelen, J. J., Muszynski, M. G., David, N. R., Luethy, M. H., Elthon, T. E., Miernyk, J. A., and Randall, D. D. (1999) J. Biol. Chem. 274, 21769 –21775 23. Palmeira, C. M., Moreno, A. J., and Madeira, V. M. (1995) Biochim. Biophys. Acta 1229, 187–192 24. Halliwell, B., and Gutteridge, J. M. C. (1999b) Free Radicals in Biology and Medicine, pp. 36 –104, Oxford University Press, Oxford 25. Halliwell, B., and Gutteridge, J. M. C. (1999a) Free Radicals in Biology and Medicine, pp. 246 –350, Oxford University Press, Oxford 26. Maxwell, D. P., Wang, Y., and McIntosh, L. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 8271– 8276 27. Purvis, A. C. (1997) Physiol. Plant. 100, 165–170 28. Wagner, A. M., and Krab, K. (1995) Physiol. Plant. 95, 318 –325 29. Hernandez, J. A., Corpas, F. J., Gomez, M., del Rio, L. A., and Sevilla, F. (1993) Physiol. Plant. 89, 103–110 30. Gomez, J. M., Hernandez, J. A., Jimenez, A., del Rio, L. A., and Sevilla, F. (1999) Free Radical Res. 31, S11-S18 31. Prasad, T. K., Anderson, M. D., Martin, B. A., and Stewart, C. R. (1994) Plant Cell 6, 65–74 32. Iturbe-Ormaetxe, I., Escuredo Pedro, R., Arrese-Igor, C., and Becana, M. (1998) Plant Physiol. 116, 173–181 33. Vianello, A., Macri, F., Cavallini, L., and Bindoli, A. (1986) J. Plant Physiol. 125, 217–224 34. Vianello, A., Macri, F., and Bindoli, A. (1987) Plant Cell Physiol. 28, 1263–1270 35. De Santis, A., Landi, P., and Genchi, G. (1999) Plant Physiol. 119, 743–754 36. Wingler, A., Lea, P. J., Quick, W. P., and Leegood, R. C. (2000) Philos. Trans. R. Soc. Lond. B Biol. Sci. B. 355, 1517–1529 37. Oberschall, A., Deak, M., Torok, K., Sass, L., Vass, I., Kovacs, I., Feher, A., Dudits, D., and Horvath, G. V. (2000) Plant J. 24, 437– 446 38. Ålin, P., Danielson, U. H., and Mannervik, B. (1985) FEBS Lett. 179, 267–270 39. Hubatsch, I., Ridderstrom, M., and Mannervik, B. (1998) Biochem. J. 330, 175–179 40. Gronwald, J. W., and Plaisance, K. L. (1998) Plant Physiol. 117, 877– 892 41. Liu, F., Cui, X. Q., Horner, H. T., Weiner, H., and Schnable, P. S. (2001) Plant Cell 13, 1063–1078 42. Møller, I. M. (2001) Trends Plant Sci. 6, 560 43. Budde, R. J. A., and Randall, D. D. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 673– 676 44. Jordan, S. W., and Cronan, J. E. (1997) J. Biol. Chem. 272, 17903–17906 45. Gueguen, V., Macherel, D., Jaquinod, M., Douce, R., and Bourguignon, J. (2000) J. Biol. Chem. 275, 5016 –5025