Planta (1990)180:278-284

P l a n t a 9 Springer-Verlag1990

Subcellular distribution of multiple forms of glutathione reduetase in leaves of pea (Pisum sativum L.) E. Anne Edwards*, Stephen Rawsthorne, and Philip M. Mullineaux John Innes Institute and AFRC Institute of Plant Science Research, Colney Lane, Norwich, NR4 7UH, UK

Abstract. On sodium-dodecyl-sulfate polyacrylamide gels, purified glutathione reductase (GR; EC 1.6.4.2) from the leaves of two- to three-week-old pea (Pisum sativum L. cv. Birte) seedlings was represented by a single band with an apparent molecular weight of 55 kilodaltons. This polypeptide was resolved to multiple isoforms by two-dimensional electrophoresis. Fraetionation of protoplasts and purification of subcellular organelles has shown that enzyme activity is associated with the chloroplasts, mitochondria and cytosol (in this order, approx. 77%, 3%, and 20% of the total activity). Distinct multiple isoforms of the enzyme, which differed in isoelectric point and were compartment-specific, were resolved from purified mitochondria and chloroplasts. The latency of the glutathione reductase activity which co-purified on Percoll gradients with the mitochondrial marker enzyme, cytochrome-c oxidase (EC 1.9.3.1.), indicated that this enzyme was within the mitochondrion. The mitochondrial glutathione reductase activity was strongly dependent on N A D P H and not N A D H .

Key words: Chloroplast

- Glutathione reductase (isoforms, localization) - Mitochondrion - Pisum (glutathione reductase)

The enzyme was first reported in plants almost 40 years ago (Conn and Vennesland 1951; Mapson and Goddard 1951) and has subsequently been noted in numerous plant species and tissues (see Rennenberg 1982 for a review). It is thought to play a key role in protection of plants against oxidative stress induced by, for example, electron leakage during photosynthesis (Foyer and Halliwell 1976), sublethal doses of certain herbicides (Kunert and Dodge 1984) and the presence of gaseous oxidants such as ozone (Tanaka et al. 1988). In pea, subcellular localisation and distribution studies on leaves (Bielawski and Joy 1986a; Gillham and Dodge 1986) have indicated that the majority of cellular G R activity is located in the chloroplast, and purification and kinetic studies have focussed on the enzyme isolated from this compartment. Mitochondrial-associated G R activity was noted as long ago as 1956 (Young and Conn 1956) and more recent work (Bielawski and Joy 1986a) has provided some evidence to indicate the presence of G R activity in the cytosol and possibly in the mitochondria of pea leaves. However, it remains to be conclusively demonstrated that an intramitochondrial G R form exists. We have utilized cell-fractionation and organellepurification methods to investigate the intracellular compartmentation of G R in pea leaf cells and have identified compartment-specific multiple forms of the enzyme which are separable on the basis of charge.

Introduction Glutathione reductase (GR; EC 1.6.4.2.) is a highly active and widely distributed enzyme in both eukaryotes and prokaryotes. It has been implicated in a wide range of metabolic processes of vital importance to the cell (see Meister and Anderson 1983 for a review). * To whom correspondence should be addressed

Material and methods Reagents. ADP-Sepharose, diethylaminoethyl (DEAE)-Sephadex

A50 and CNBr-activated Sepharose 4B were obtained from Pharmacia Fine Chemicals, London, UK. All other reagents were from Sigma Chemical Corp., London, or Bio-Rad Laboratories, Richmond, Cal., USA.

Abbreviations: Da = dalton; GR = glutathione reductase; IEF =

Plant material and growth conditions. Pea (Pisum sativum L. cv.

isoelectric focussing; IgG = immunoglobulin G ; pI =isoelectric point; SDS-PAGE=sodium dodecyl sulfate-polyacrylamide gel electrophoresis

Birte) seeds were obtained from the John Innes Institute pea germplasm collection and were grown in trays of vermiculite in controlled-environment chambers at a temperature of 20~ C/15~

E.A. Edwards et al. : Isoforms of glutathione reductase in pea (light/dark) with a daily light period of 18 h and at a photon fluence rate of 300 lamol.m-2, s-1. Leaves for all experiments were harvested from two- to three-week-old plants.

Enzyme purification. Glutathione reductase was purified from pea leaves according to the method of Connell and Mullet (1986) with two additional steps: (i) Prior to ion-exchange chromatography on DEAE-Sephadex, the cleared homogenate was fractionated by the addition of solid ammonium sulphate. Material precipitating at 50% saturation was discarded. The supernatant was brought up to 80% saturation and the precipitate from this solution was dissolved and dialysed against 50 mM 2-amino-2-(hydroxymethyl)1,3-propanediol (Tris) pH 8.0 prior to running the column. (ii) The eluate from the ADP-Sepharose (affinity) column was further purified by electrophoresis on non-denaturing polyacrylamide gels. Gels were stained for enzyme activity according to the method of Davis et al. (1982). The yellow band was excised and the protein released by electroelution.

Cellfractionation. Protoplasts were isolated from pea leaves using the procedure of Walton and Woolhouse (1983) except that the digestion was done in 20 mM 2-(N-morpholino)ethanesulfonic acid (Mes) pH 5.8 containing 1 mM CaCIz, 0.6 M sorbitol, 3% (w/v) cellulase, 0.5% (w/v) macerozyme and 0.5% (w/v) bovine serum albumin (BSA). Protoplasts were purified on sucrose-sorbitol step gradients according to the method of Edwards et al. (1978) except that the sucrose fraction contained 5% (w/v) dextran. Purification of organelles. Chloroplasts were prepared according to the method of Gillham and Dodge (1986) and further purified on 10% 80% (v/v) linear Percoll gradients (Bartlett et al. 1982). Washed mitochondria and peroxisomes were isolated according to the method of Day et al. (1985), and mitochondria and peroxisomes were subsequently purified on 0-10% (w/v) linear polyvinylpyrrolidone (PVP)-25 gradients in 28% Percoll (Neuberger et al. 1982; Day et al. 1985).

Preparation of the antiserum. Purified pea leaf GR was used to generate antibodies by subcutaneous injection of New Zealand white rabbits. Before beginning the immunisation programme a sample of pre-immune (null) serum was taken. Protein (200 lag) was emulsified with complete adjuvant for the first injection. A second injection of 100 gg in incomplete adjuvant was given two weeks later. Subsequent injections (100 txg each) were given at four, six and eight weeks. Serum samples were taken at weekly intervals after the last injection. Antiserum was purified by passage down a column of CNBr-activated Sepharose 4B to which a crude extract of pea leaf proteins had been coupled. This removed immunoglobulin G (IgG) molecules which recognised cellular proteins other than glutathione reductase which was present as only a small proportion of the crude extract. Anti-GR IgG was concentrated by ammonium-sulphate precipitation.

Gel electrophoresis and Western blotting. Denaturing sodium-dodecyl-sulphate (SDS) polyacrylamide gels were run using the discontinuous buffer system of Laemmli (1970). Gels consisted of a 7.5% stacking gel and an 11% or ! 5% resolving gel. Protein molecularweight markers (daltons, Da) were: (i) Low range (14000-70000; Sigma): bovine albumin, 66000; egg albumin (ovalbumin) 45000; glyceraldehyde-3-phosphate dehydrogenase (rabbit muscle), 36000 (subunit); bovine erythrocyte carbonic anhydrase, 29000; bovine pancreas trypsinogen (treated with phenylmethylsulfonyl fluoride), 24000; soybean trypsin inhibitor, 20100; bovine milk lactalbumin, 14200. (ii) High range (30000-200 000; Sigma): rabbit muscle myosin, 205000 (subunit) ; Escherichia coli fl-galactosidase, 116000 (subunit); rabbit muscle phosphorylase B, 97400 (subunit); bovine plasma albumin, 66000; egg albumin, 45000; bovine erythrocyte carbonic anhydrase, 29000. (iii) Prestained (Biorad; apparent molecular weights are given

279 since the addition of the dye causes the proteins to migrate differently from their true molecular weight): phosphorylase b, 1 |0000; bovine serum albumin, 75 000; ovalbumin, 50000; carbonic anhydrase, 39 000; soybean trypsin inhibitor, 27 000; lysozyme, 17000. Two-dimensional gels were run according to the method of O'Farrell (1975). Western blotting was performed according to the method of Burnette (1981). Immunodetection was by the alkalinephosphatase-conjugated anti-antibody method of Blake et al. (1984).

Proteolytic digestion. Isoelectric-focussing gels were digested with staphylococcal V8 protease (1 ~tg enzyme: 10 I-tgprotein) for 30 min at room temperature in Laemmli sample buffer. Performic-acid oxidation of glutathione reductase was done by a modification of the method of Hirs (1967). A 1.4-ml aliquot of performic-acid reagent (30% (w/v) hydrogen peroxide: 88% (v/v) formic acid, 1:9), was added to ice-cold protein dissolved in 0.1 ml 88% (v/v) formic acid and incubated for 4 h on ice. The reaction was terminated by dialysis overnight against several changes of distilled water. Carboxymethylation. The alkylation reaction of Crestfield et al. (1963) with iodoacetamide was used to carboxymethylate proteins. Detection of glycosylation. For the detection of glycosylated residues, proteins were transferred to nitrocellulose membranes. Fixing, washing and blocking of membranes was done according to the method of Dietz et al. (1988). Biotin-conjugated conconavalin A was applied at 1 ~tg/ml in TBSCT (15 mM Tris pH 7.5, 150 mM NaCl, 1 mM MnC12, 1 mM CaCI2, 0.05% (v/v) Tween 20) for 1 h at room temperature. After four washes with TBSCT, the biotin conjugate was detected with avidin-alkaline phosphatase conjugate used in accordance with the manufacturers instructions.

Dephosphorylation. Protein was dephosphorylated by incubation with E. coli alkaline phosphatase (Boehringer, Mannheim, FRG) at a concentration of 1 unit/10-15 lag protein, in 50 mM Tris pH 8.0, 0.1 mM ethylenediaminetetraacetic acid (EDTA), at 37 ~ C for 24 h. Enzyme assays. All enzyme assays were carried out spectrophotometrically at 25 ~ C in 1-ml reaction volumes. The component used to start the reaction is given last. Glutathione reductase (EC 1.6.4.2) was assayed in 0.1 M 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (Hepes) pH 7.8 containing 0.2 mM NADPH and 0.5 mM oxidised glutathione (Connell and Mullet 1986). Activity was determined by the fall in absorbance at 340 nm as NADPH was oxidised. NADPH-dependent glyceraldehyde-3-phosphate dehydrogenase (EC 1.2.1.12) was assayed as the conversion of 1,3-diphosphoglycerate to glyceraldehyde-3-phosphate according to Wu and Racker (1959). Reactions were carried out in 0.1 M Hepes pH 7.8 containing 0.25 mM NADPH, 9 mM MgCI2, 3 mM ATP, 25 units phosphoglycerate kinase (EC 2.7.2.3) and 5 mM 3-phosphoglycerate. Cytochrome-c oxidase (EC 1.9.3.1) was assayed according to the method of Tolbert (1974) at 550 nm. Reactions contained 0.08 ~tM reduced cytochrome c, 100 mM phosphate buffer pH 7.0, 0.01% (v/ v) Triton X-100 and extract. Hydroxypyruvate reductase (EC 1.1.1.29) was assayed at 340 nm according to the method of Kohn and Utting (1982). Pyrophosphate-dependent phosphofructokinase (EC 2.7.1.90) was assayed according to the method of Journet and Douce (1985). Reactions were in 100 mM 3-(N-morpholino) propanesulfonic acid (Mops) pH 7.4 containing 0.2 mM NADH, 1.5mM MgCI2, 15 ~tM fructose-2,6-bisphosphate, 10mM fructose-6-phosphate, 0.1 unit aldolase (EC 4.1.2.13), 1 unit glycerol-3phosphate dehydrogenase (EC 1.1.1.8), 10 units triose-phosphate isomerase (EC 5.3.1.1) and 1 mM sodium pyrophosphate. Unless otherwise indicated the activities of all the enzymes were assayed in the absence of osmotic protection.

280

E.A. Edwards et al. : Isoforms of glutathione reductase in pea

Latency of enzyme activity. Enzyme activity associated with the intact organelles was determined by carrying out the assays in the presence of 0.3 M mannitol osmoticum. Total organelle activity was measured after lysis of the organelle by the addition of Triton X-100 to a final concentration of 0.01% (v/v). Percentage latency was defined by the equation: % latency= 1-(activity associated with intact organelle/activity associated with lysed organelle) • 100 Organelle intactness was varied by changing the concentration of the osmoticum prior to assaying the enzyme activity. Enzyme kinetics. The Michaelis constants (Kin) of mitochondrial and chloroplastic GR for oxidised glutathione (GSSG), NADPH and NADH were each determined in the presence of a saturating concentration of the other substrate (0.2 mM NADPH or 0.5 mM GSSG). Limiting substrate concentrations varied between 0.4 laM and 10 laM for NADPH and NADH and between 8 laM and 50 laM for GSSG. For each assay, purified organetle extracts containing 0.05 nkat of GR activity were used. Average Km values were determined using Lineweaver-Burk plots.

Fig. 1. A Analysis by SDS-PAGE of purified pea-leaf GR (2 lag) stained with Coomassie blue. B Analysis by two-dimensional electrophoresis (SDS-PAGE and isoelectric focussing, IEF) of purified GR (5 lag) stained with Coomassie blue

Results The purification scheme employed yielded protein which gave a single band with Coomassie blue upon SDS-polyacrylamide gel electrophoresis (PAGE). F r o m the position of the band, a subunit molecular weight of approx. 55 k D a could be deduced for the enzyme (Fig. IA). When this protein was analysed by two-dimensional electrophoresis, the single band was resolved into eight spots varying in isoelectric point (pI) from 6.5 to 5.2 (Fig. 1 B). These eight polypeptides did not represent different oxidation states since neither performic acid nor carboxymethylation treatment of the unresolved enzyme had any observed effect on the resolved pattern after two-dimensional gel electrophoresis (data not shown). No carbohydrate or phosphate moieties were detected in the unresolved protein, indicating that the multiple forms were not the result of differences in post-translational modification (data not shown). An antiserum was raised against purified, unresolved total pea leaf G R . This recognised a single band with an apparent molecular weight of 55 k D a from a crude extract on Western blots of one-dimensional SDS polyacrylamide gels and all eight forms of the enzyme resolved from a crude extract by two-dimensional electrophoresis (Fig. 2A, B). Furthermore, when purified and resolved G R was digested with protease V8 (from Staphylococcus aureus) the individual peptides, identified by immunoblotting, indicated a close relationship and only a few minor differences between some of the lower-molecular-weight peptides (Fig. 3). The null serum did not give a reaction with either one- or two-dimensional gels (data not shown). In order to determine the subcellular distribution of pea leaf G R , protoplast-fractionation experiments were performed. Gentle rupture of protoplasts produced minimal damage to organelles which were then separated by differential centrifugation. The purity of each fraction was assessed using the enzymes NADP+-glyceralde hyde-3-phosphate dehydrogenase, cytochrome-c oxidase and pyrophosphate-dependent phosphofructokinase as

Fig. 2A, B. Western blot of total pea-leaf proteins after separation by A SDS-PAGE (30 lag) and B two-dimensional electrophoresis (120 lag). Anti-GR IgG was used at a concentration of 80 pg/ml

Fig. 3. Peptides of isoelectrically focussed pea GR generated by limited proteolysis with Staphylococcus aureus V8 protease, separated by SDS-PAGE and detected by Western blotting. Anti-GR IgG was used at a concentration of 80 pg/ml

E.A. Edwards et al. : Isoforms of glutathione reductase in pea

281

1. Subcellular fractionation of pea leaf protoplasts. Activities in the 2500.g (chloroplast) pellet, 12000-g (mitochondrial) pellet and 12000.g (cytosolic) supernatant are presented as percentages of the activity in the unfractionated protoplasts. The figures are the results from two separate protoplast preparations (I and II). The activities (nmol.min-1) in the unfractionated extract were: glutathione reductase (GR) 135 (I), 150 (II); glyceraldehyde-3phosphate dehydrogenase (GAPDH) 214 (I), 245 (II); cytochromec oxidase (CCO) 183 (I), 206 (II); pyrophosphate-dependent phosphofructokinase (PFP) 8 (I), 15 (II). The recovery of enzyme activities is also shown Table

2. Distribution of total enzyme activities fractionated from washed mitochondrial preparations loaded onto Percoll/PVP gradients. Values are the results of two separate preparations (I and II). Total enzyme activities (gmol suhstrate oxidised/min) loaded onto the gradients were: glutathione reductase (GR) 145 (I), 205 (II); glyceraldehyde-3-phosphate dehydrogenase (GAPDH) 60 (I), 75 (II); hydroxypyruvate reductase (HPR) 140000 (I), 175 000 (II); cytochrome-c oxidase (CCO) 200000 (I), 250000 (II). Percentage recovery is also shown Table

Fraction

Percentage distribution of initial enzyme activity

Fraction

GR I

2500.g pellet 12000.g pellet

GR

GAPDH

CCO

I

II

I

II

I

II

15 60 nd

nd nd 89

nd nd 97

Top Thylakoid Intermediate Mitochondria Peroxisomes % recovery

75

89

97

nd = not detected

66 2 32

68 3 30

86 nd

88 nd

13

12

12 75 nd

100

101

99

100

87

12 000 .g

supernatant % recovery

II

Percentage distribution of total enzyme activity

PFP

GAPDH

CCO

HPR

I

II

I

II

I

I1

I

II

3 17 27 35 2 84

4 20 25 38 2 89

12 75 10 nd nd 97

17 67 28 nd nd 112

3 10 32 56 2 103

2 15 29 54 2 100

1 3 4 9 28 45

2 2 5 10 35 54

nd = not detected

"~176176176176176176 " . ~ 1 7 6 ~ ~ 1 7 6 1 ~176 76

::2:':'::::::-:-:. ~176

*

Fig. 4. Purification of pea-leaf mitochondria and peroxisomes from washed mitochondria. Schematic representation of the separation of thylakoids and organdies on a gradient generated by centrifugation of 28% Percoll containing a gradient of 0-10% PVP-25 (Day et al. 1985). Fraction 1, top;fraction 2, thylakoids ;fraction 3, intermediate;fraction 4, mitochondria; fraction 5, peroxisomes

markers for chloroplasts, mitochondria and cytosol, respectively. About 87% of the chloroplasts were recovered in the 2500.g pellet and this contained approx. 67% of the total G R activity. Two to three percent of the G R activity was located in the 12000.g pellet which had 80% of the mitochondrial marker but no detectable amounts of the chloroplast or cytosolic m a r k e r enzymes. The remainder of the G R activity (31%) was in the 12000-g supernatant which contained all the cytosolic marker enzyme (Table 1). Since the cytosolic fraction only contained approx. 10% of the chloroplast marker, chloroplast contamination cannot account for all the G R in this fraction. In order to determine whether the G R activity in the 12000.g pellet was associated with

the mitochondria or with the co-sedimenting peroxisomes, these organelles were prepared on a large scale and purified by isopycnic centrifugation on Percoll/PVP25 gradients. Figure 4 illustrates the separation of mitochondria, thylakoids and peroxisomes into defined fractions after centrifugation of a crude mitochondrial suspension. This has been previously documented by D a y et al. 1985. The mitochondria appeared as a tight band near the b o t t o m of the tube, the peroxisomes were also recovered near the b o t t o m of the centrifuge tube and overlapped slightly with the mitochondria. Thylakoid membranes remained near the top. A clean separation of organelles from thylakoid m e m b r a n e fragments was obtained, as shown by the distribution of the activity o f NADPH-glyceraldehyde-3phosphate dehydrogenase which was closely associated with the distribution of chlorophyll on the gradient (data not shown). Glutathione reductase activity in the fractions was positively correlated to the mitochondrial marker, cytochrome-c oxidase and not to the peroxisomal marker, hydroxypyruvate reductase (Table 2). When purified mitochondria were lysed with Triton X-100, the latency of G R activity was found to be in excess of 90%, whereas the latency of this enzyme in the thylakoid-membrane fraction was always close to zero. This was not simply a detergent-activation effect, since the addition of Triton X-100 to purified G R had no effect on the rate of enzyme reaction and a similar latency was found when the organelles were lysed by osmotic shock (data not shown). Intact chloroplasts were prepared on a large scale from leaf homogenates and purified on Percoll gradients. These were judged to be approx. 80% intact by phase-contrast microscopy. The kinetic properties of the chloroplast and mitochondrial G R activities in total organelle extracts are

282

E.A. Edwards et al. : lsoforms of glutathione reductase in pea

Table 3. Apparent K m values for pea chloroplast and pea mitochondrial G R activities. From three separate determinations using purified organelle extracts, average K m values were calculated using Lineweaver-Burk plots Substrate

five forms with pIs ranging from 5.6 to 6.3 (Fig. 6A). Three mitochondrial forms were resolved and these ranged from pI 6.3 to pI 6.5 (Fig. 6B).

Km (gM)

Discussion

Oxidised glutathione NADPH NADH

Chloroplast enzyme

Mitochondrial enzyme

27 2.8 300

12 2.4 100

Fig. 5. Western blot of chloroplast (lane 1, 30 lag) and mitochondrial (lane 2, 30 lag) proteins and purified pea-leaf GR (lane 3, 5 ng) separated by SDS-PAGE. Anti-GR IgG was used at a concentration of 80 pg/ml

Fig. 6A, B. Western blot of organelle proteins from pea leaves separated by two-dimensional gel electrophoresis. A Chloroplast proteins (90 lag). B Mitochondrial proteins (240 rtg). Anti-GR IgG was used at a concentration of 80 pg/ml

compared in Table 3. Both showed high affinities for oxidised glutathione and for NADPH, but much lower affinities for NADH. The Km values for oxidised glutathione and for N A D P H are similar to those reported by Bielawski and Joy (1986b) for pea chloroplast and pea-root G R activities. The anti-GR antibody recognised a single band of 55 kDa on Western blots of SDS polyacrylamide gels of both mitochondrial and chloroplast proteins (Fig. 5). This band was resolved into multiple proteins on two-dimensional gels. Chloroplast G R proteins were more abundant and were resolved into

We have purified total pea-leaf GR and shown by SDSPAGE, that it corresponds to a single band of molecular weight 55 kDa. On two-dimensional gels, this band was resolved into eight spots varying in pI between 6.5 and 5.2. The occurrence of multiple forms of plant enzymes distinguishable by charge is not unknown; for instance, Ellefson and Krogmann (1979) reported that, on purification, five to six forms of the enzyme ferredoxin-NADP oxidoreductase from spinach, were apparent on isoelectric-focussing gels. They concluded that at least some of the forms arose as a result of oxidation-reduction differences since treatment of the unresolved enzyme with performic acid, which would be expected to oxidise all sulfur-containing amino acids and other readily oxidizable groups to a uniform oxidation state, reduced the gel pattern to two forms. However, this was not the case for glutathione reductase since neither performic-acid treatment nor carboxymethylation of the unresolved enzyme had any effect on the number of resolved forms. Furthermore, an antiserum raised against unresolved GR, recognised all eight forms of the enzyme in a total pea leaf protein extract separated by twodimensional electrophoresis, indicating that the multiple forms did not arise as artefacts of purification. Tests for the most common types of post-translational modification - glycosylation and phosphorylation - were also carried out, but were negative, indicating that these invivo modifications were not responsible for multiple forms. When peptides of each form were prepared, using V8 protease, they were found to be very similar, indicating that the forms are closely related in amino acid sequence and that glutathione reductase in the leaves of two- to three-week-old peas comprises a protein family. This phenomenon has been reported before in higher plants; for instance the enzymes glutamine synthetase, which plays a key role in nitrogen metabolism (McNally et al. 1983), and glutathione S-transferase, involved in detoxification reactions (Edwards and Owen 1986), both have several compartment-specific isoforms. In this paper, subcellular-distribution studies, based on organelle and cytosol marker enzymes, have provided evidence for cytosolic, chloroplastic and mitochondrial forms of GR in pea leaves. After correcting for organelle lysis, approx. 20%, 77%, and 3% of the total cellular activity was found to be located in each fraction, respectively. Previous investigations have already shown that the majority of GR activity in pea leaves is located in the chloroplast (Gillham and Dodge 1986; Bielawski and Joy 1986a) and this is consistent with the enzyme having a major role in preventing oxidative damage during photosynthesis (Foyer and Halliwell 1976). However, its presence elsewhere in the cell has not been thoroughly investigated. We have been able to show that G R activity

E.A. Edwards et al. : Isoforms of glutathione reductase in pea is latent in pea leaf mitochondria and is highly dependent on the cofactor, N A D P H . Why mitochondria should contain a G R preferring N A D P H , rather than N A D H is at present unclear. A matrix transhydrogenase has recently been characterised from potato-tuber mitochondria (Carlenor et al. 1988) and a similar enzyme could provide N A D P H for pea-leaf mitochondrial GR. The role of the enzyme in mitochondria must, at present, remain speculative but some recent work with animal mitochondria has shown that reduced glutathione could play a critical role in the defense against membrane peroxidation and in the regulation of inner-membrane permeability by maintaining intramitochondrial sulfhydryl groups in the reduced state (Olafsdottir et al. 1988; Martensson and Meister 1989). When the G R forms from chloroplasts and mitochondria were compared on two-dimensional gels, they were found to differ in number and in ionic charge. The pI of the five chloroplast GRs varied between 5.6 to 6.3, whereas the three mitochondrial forms ranged from 6.3 to 6.5. The eight isoforms detected in resolved total pea leaf extract spanned a pI range of 6 . 5 - 5 . 6 . The five forms between 5.6 and 6.3, presumably corresponding to the chloroplastic forms, were far more abundant than those between pI 6.3 and 6.5 (presumably the mitochondrial forms). Isoforms in the crude extract detected between pI 5.6 and 5.2 could represent cytoplasmic G R forms, though these could also comigrate with plastidial forms. The differences in charge may be important in protein targetting since it has been suggested that this factor plays a role in the recognition phenomenon for the uptake of a particular isoenzyme into an organelle (Simcox and Dennis 1978). The physiological significance of multiple forms of G R has yet to be determined. Higher-plant metabolism appears to be regulated in part by subcellular compartmentation (see Oaks and Bidwell 1970 for a review) and there is evidence for duplication of metabolic pathways which differ in their subcellular location and expression during various physiological states (Anderson and Advani 1970; Schnarrenberger and Oeser 1974). This may be important in view of several recent reports into the effects of environmental and other stresses on the activity of GR, for example high levels of atmospheric ozone (Tanaka et al. 1988) and air pollutants (Mehlhorn et al. 1987). The biosynthesis of multiple forms of G R might be necessitated by the physiological needs of the plant. Guy and Carter (1984) studied G R activity in nonhardened and cold-hardened spinach leaves. They found that G R activity in crude extracts of cold-hardened tissue increased over that of nonhardened tissue. Furthermore, the partially purified enzyme from the two sources showed different kinetic characteristics, heat inactivation, freezing inactivation and electrophoretic mobilities. Hardened leaves contained different forms of GR. Bielawsky and Joy (1986 b) have shown that G R isolated from pea roots differed in kinetic properties from the enzyme isolated from chloroplasts. Recently, Drumm-Herrel et al. 1989 have shown that a plastidic isoform o f G R from mustard (Sinapis alba L.)

283 cotyledons exhibited sensitivity to photo-oxidative treat= ment whereas the cytoplasmic protein was not directly affected. Future work must establish whether the different forms in pea respond to different stress factors. It also remains to be ascertained whether the various forms are products of different genes or represent post-transcriptional or post-translational modifications of a single gene product, and how the expression of the different forms is controlled. We are grateful to Dr. Alison M. Smith for advice on enzyme assays, Dr. Ian Dry for helpful suggestions during the preparation of the manuscript and Tarn Dalzell for typing the manuscript. E.A.E. acknowledges a grant from the John Innes Institute. S.E. and P.M.M. are supported by the Agricultural and Food Research Council by a grant-in-aid to the John Innes Institute. References Anderson, L.E., Advani, J.R. (1970) Chloroplast and cytoplasmic enzymes: three distinct enzymes associated with the reductive pentose phosphate cylce. Plant Physiol. 45, 583-585 Bartlett, S.G., Grossman, A.R., Chua, N.H. (1982) In vitro synthesis and uptake of cytoplasmically-synthesized chloroplast proteins. In: Methods in chloroplast molecular biology, pp. 10811091, Edelman, M., Hallick, R.B., Chua, N.H., eds. Elsevier Biomedical Press, Amsterdam Bielawski, W., Joy, K.W. (1986a) Reduced and oxidised glutathione and glutathione reductase activity in tissues of Pisum sat# rum. Planta 169, 267-272 Bielawski, W., Joy, K.W. (1986b) Properties of glutathione reductase from chloroplasts and roots of pea. Phytochemistry 25, 2261-2265 Blake, M.S., Johnston, K.H., Russell-Jones, G.J., Gotschlich, E.C. (1984) A rapid, sensitive method for detection of alkaline phosphatase-conjugated anti-antibody on western blots. Anal. Biochem. 136, 175-179 Burnette, W.N. (1981) "Western blotting" : electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal. Biochem. 112, 195-203 Carlenor, E., Persson, B., Glaser, E., Andersson, B., Rydstrom, J. (1988) On the presence of a nicotinamide nucleotide transhydrogenase in mitochondria from potato tuber. Plant Physiol. 88, 303 308 Conn, E.E., Vennesland, B. (1951) Glutathione reductase of wheat germ. J. Biol. Chem. 192, 17-28 Connell, J.P., Mullet, J.E. (1986) Pea chloroplast glutathione reductase: Purification and characterization. Plant Physiol. 82, 351-356 Crestfield, A.A., Moore, S., Stein, W.H. (1963) The preparation and enzymatic hydrolysis of reduced and S-carboxymethylated proteins. J. Biol. Chem. 252, 1102 1106 Davis, N.K., Greer, S., Jones-Mortimer, M.C., Perham, R.N. (1982) Isolation and mapping of glutathione reductase-negative mutants of Escherichia coli KI2. J. Gen. Microbiol. 128, 16311634 Day, D.A., Neuburger, M., Douce, R. (1985) Biochemical characterization of chlorophyll-free mitochondria from pea leaves. Aust. J. Plant Physiol. 12, 219-228 Dietz, K.-J., Kaiser, G., Martinoia, E. (1988) Characterization of vacuolar polypeptides of barley mesophyll cells by two-dimensional gel electrophoresis and by their affinity to lectins. Planta 176, 362-367 Drumm-Herrel, H., Gerhausser, U., Mohr, H. (1989) Differential regulation by phytochrome of the appearance of plastidic and

284

E.A. Edwards et al. : Isoforms of glutathione reductase in pea cytoplasmic isoforms of glutathione reductase in mustard (Sinapis alba L.) cotyledons. Planta 178, 103-109

Edwards, R., Owen, W.J. (1986) Comparison of glutathione Stransferases of Zea mays responsible for herbicide detoxification in plants and suspension-cultured cells. Planta 169, 208215 Edwards, G.E., Robinson, S.P., Tyler, N.J.C., Walker, D.A. (1978) Photosynthesis by isolated protoplasts, protoplast extracts, and chloroplasts of wheat. Influence of orthophosphate, pyrophosphate and adenylates. Plant Physiol. 62, 313-319 Ellefson, W.L., Krogmann, D.W. (1979) Studies of the multiple forms of ferredoxin-NADP oxidoreductase from spinach. Arch. Biochem. Biophys. 194, 593 599 Foyer, C.H., Halliwell, B. (1976) Presence of glutathione and glutathione reductase in chloroplasts: a proposed role in ascorbic acid metabolism. Planta 133, 21-25 Gillham, D.J., Dodge, A.D. (1986) Hydrogen-peroxide-scavenging systems within pea chloroplasts. Planta 167, 246--251 Guy, C.L., Carter, J.V. (1984) Characterization of partially purified glutathione reductase from cold-hardened and nonhardened spinach leaf tissue. Cryobiology 21, 454-464 Hirs, C.H.W. (1967) Determination of cystine as cysteic acid. Methods Enzymol. 11, 59-62 Journet, E., Douce, R. (1985) Enzymic capacities of purified cauliflower bud plastids for lipid synthesis and carbohydrate metabolism. Plant Physiol. 79, 458-467 Kohn, L.D., Utting, J.M. (1982) Hydroxypyruvate reductase (Dglyerate dehydrogenase) from Pseudomonas. Methods Enzymol. 89, 341-345 Kunert, K.J., Dodge, A.D. (1984) Herbicide induced radical damage and antioxidant systems. In: Target sites of herbicide action, B6ger, P.H., Sandmann, H., eds. CRC Press, Boca Raton, Fla., USA Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685 Mapson, L.W., Goddard, D.R. (1951) Reduction of glutathione by coenzyme II. Nature 167, 975-977 Martensson, J., Meister, A. (1989) Mitochondrial damage in muscle occurs after marked depletion of glutathione and is prevented by giving glutathione monoester. Proc. Natl. Acad. Sci. USA 86, 471-475 McNally, S.F., Hirel, B., Gadal, P., Mann, A.F., Stewart, G.R. (1983) Glutamine synthetases of higher plants. Evidence for a specific isoform content related to their possible physiological

role and their compartmentation within the leaf. Plant Physiol. 72, 22-25 Mehlhorn, H., Cottam, D.A., Lucas, P.W., Welburn, A.R. (1987) Induction of ascorbate peroxidase and glutathione reductase activities by interactions of mixtures of air pollutants. Free Radical Res. Commun. 3, 193 197 Meister, A., Anderson, J.W. (1983) Glutathione. Annu. Rev. Biochem. 52, 711 760 Neuberger, M., Journet, E.P., Bligny, R., Carve, J.-P., Douce, R. (1982) Purification of plant mitochondria by isopycnic centrifugation in density gradients of Percoll. Arch. Biochem. Biophys. 217, 312-323 Oaks, A., Bidwell, R.G.S. (1970) Compartmentation of intermediary metabolites. Annu. Rev. Plant Physiol. 21, 43-46 O'Farrell, P.H. (1975) High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 250, 4007-4021 Olafsdottir, K., Pascoe, G.A., Reed, D.J. (1988) Mitochondrial glutathione status during Ca z§ ionophore-induced injury to isolated hepatocytes. Arch. Biochem. Biophys. 263, 226 235 Rennenberg, H. (1982) Glutathione metabolism and possible biological roles in higher plants. Phytochemistry 21, 2771-2781 Schnarrenberger, C., Oeser, A. (1974) Two isoenzymes of glucosephosphate isomerase from spinach leaves and their intracellular compartmentation. Eur. J. Biochem. 45, 77 82 Simcox, P.D., Dennis, D.T. (1978) Isoenzymes of the glycolytic and pentose phosphate pathways in proplastids from the developing endosperm of Ricinus communis L. Plant Physiol. 61, 871 877 Tanaka, K., Saji, H., Kondo, N. (1988) Immunological properties of spinach glutathione reductase and inductive biosynthesis of the enzyme with ozone. Plant Cell Physiol. 29, 637-642 Tolbert, N.E. (1974) Isolation of subcellular organelles of metabolism on isopyenic sucrose gradients. Methods Enzymol. 31A, 734-746 Walton, N.J., Woolhouse, H.W. (1983) Effects of illumination and glycollate oxidation in promoting glyoxylate decarboxylation by pea protoplasts. Planta 158, 469-471 Wu, R., Racker, E. (1959) Regulatory mechanisms in carbohydrate metabolism. III Limiting factors in glycolysis of ascites tumor cells. J. Biol. Chem. 234, 1029-1035 Young, L.C.T., Conn, E.E. (1956) The reduction and oxidation of glutathione by plant mitochondria. Plant Physio131,205-211 Received 17 July; accepted 24 August 1989