Planta
Planta (Berl.) 128, 225-231 (1976)
9 by Springer-Verlag 1976
Hydrogen Peroxide Production and the Release of Carbon Dioxide during Glycollate Oxidation in Leaf Peroxisomes* B. Grodzinski and V.S. Butt Botany School, Oxford University, South Parks Road, Oxford OX1 3RA, U.K.
Summary. The rate at which H202 becomes available during glycollate oxidation for further oxidation reactions, especially that of glyoxylate to formate and CO2, in peroxisomes from spinach-beet (Beta vulgaris L., vat. vulgaris) leaves has been determined by measuring Oz uptake in the presence and absence of added catalase. The rates observed under air and pure 02 were sufficient to account for the 14COz released from [1-~4C]glycollate under these conditions; the two reactions showed similar characteristics. In the course of the reaction, a fall in catalase activity was observed concomitant with an increase in 14C02release. There is no evidence that catalase was disproportionately lost f r o m the peroxisomes during isolation, and it is argued that the COz release observed contributes to the photorespiratory CO2 loss in intact leaves.
destroy all of the H202 generated during glycollate oxidation (Tolbert, 1971), this CO2 release is a consequence of the non-enzymic oxidation of glyoxylate by HzO2 which has escaped catalase action (Halliwell and Butt, 1974). Further evidence for the action of H202 during the oxidative decarboxylation of glycollate in peroxisomes is presented here. A method has been devised to estimate the rate o f H 2 0 2 generation in excess of that destroyed by catalase and this rate has been related to that of CO2 release during the course of the peroxisomal oxidation of glycollate. Its significance in relation to CO2 release in photorcspiration is discussed.
Materials and Methods
Introduction The CO2 released in the photorespiration of leaves, especially at higher temperatures, has been shown to arise predominantly from the carboxyl group of glycollate (Zelitch, 1966). Two processes, which could proceed at rates adequate to account for this release, have been demonstrated. The immediate substrate could be glycine, decarboxylated in leaf mitochondria (Yolbert, 1971; Kisaki etal., 1971), or glyoxylate, oxidised in illuminated chloroplasts (Zelitch, 1972) under the action of H202 and the superoxide free radial anion, - O z - (Elstner and Heupel, 1973). It has however been shown that 14C02 is released from [1-14C]glycollate in the peroxisomes from spinach-beet leaves (Halliwell et al., 1972) and further that, although there is generally assumed to be sufficient catalase present in the peroxisomes to
* Abbreviations: DCPIP = 2,6-dichlorophenolindophenol ; FMN = flavin mononucleotide.
Chemicals. Reagents used were of the highest purity available from BDH ChemicalsLtd., Poole, Dorset, U.K. or from Sigma Chemical Co., Kingston-upon-Thames, Surrey. U.K. Pyrid-2-yl-~-hydroxymethanesulphonate (PHMS) was purchased from Aldrich Chemical Co. Ltd., Wembley, London, U.K. and bovine liver catalase from Boehringer Corporation, London W5, U.K. 5-(4-Biphenylyl)-2-(4t-butylphenyl)-l-oxa-3,4-diazole (butyl-PBD) for liquid scintillation counting was obtained from CIBA-GeigyLtd., Duxford, Cambridge, U.K. Sodium [1-1~C] glycollate and sodium [1-14C]glyoxylatewere obtained from Radiochemical Centre, Amersham, Bucks, U.K. and stored for 1-2 weeks in the deep-freeze at -20~ until required. Plants. Leaves of spinach beet (Beta vulgaris vat. vulgaris) were harvested from plants grown from seed outdoors in the summer, and in the greenhouse, receiving supplementary mercury lighting for 4 h each day, during the winter. Glycollate oxidase (Glycollate:oxygen oxidoreductase, EC 1.1.3. I) was extracted and purified by a procedure, modified from that of Zelitch and Ochoa (1953): A. Extraction. About 400 g of spinach-beet leaves, from which the midribs and coarser veins had been removed, were homogenized in 1,200ml of chilled 0.1 M Tris-HC1 buffer, pH 8.0, containing 2% (w/v) polyvinylpyrollidone (Polyclar AT) for two successive 20 s periods at top speed in an MSE Ato-Mix homogenizer. All subsequent operations were carried out at 2 4 ~C.
226
B. Grodzinski and V.S. Butt: H202 Production and CO2 Release
The homogenate was strained through four layers of muslin and centrifuged at 500 x g for 10rain. The pH of the supernatant liquid was carefully adjusted to pH 5.3 by the slow addition of prechilled 10% (v/v) acetic acid with constant stirring. After standing for 30 min, the precipitate was removed by centrifugation at 20,000 x g for 20 min. B. Ammonium sulphate precipitation. 140 g of (NH4)2SO~ was added to each litre of supernatant slowly with continuous stirring. The precipitate was removed by centrifugation and a further 80 g of (NH~)zSO4 added to each litre of supernatant. The resulting precipitate was collected by centrifugation and resuspended in 0.1 M Tris-HC1 buffer, pH 8.3. These two (NH4)2SO4 precipitations were repeated after raising the pH to 8.3 with KOH, and the second precipitate suspended in 10mM Tris-HC1 buffer, pH 7.5, containing 2 r a m flavin mononucleotide (FMN). C. Ion-Exchange Chromatography. The solution was applied to a column (15 cm long x2.0 cm diam.) of Sephadex G-25 (fine), which had been equilibrated with 10 mM Tris-HC1 buffer, pH 7.5. The first protein band collected contained most of the glycollate oxidase and catalase activity. FMN was added to this fraction, which was then applied to a column (20 cm long x 2.5 cm diam.) of carboxymethyl (CM)-cellulose, which had also been equilibrated with 10 mM Tris-HC1 buffer, pH 7.5. The first protein band containing glycollate oxidase activity was collected and then layered on a similar column of DEAE-cellulose (DE-52), equilibriated with the same buffer. The first protein band to emerge from this column again contained most of the glycollate oxidase activity, The enzyme was precipitated from the eluate by the slow addition of 0.6 vol. of saturated (NH4)2SO 4 solution and stored overnight in suspension. D. Sephadex G-200 Chromatography. The enzyme was sedimented by centrifugation, resuspended in 0.1 M Tris-HCI buffer, pH 8.0, containing 5 mM FMN, and layered on a column (36 cm long x 3.0 cm diam.) of Sephadex G-200. The eluate was collected and two protein peaks, accounting for 80% of the glycollate oxidase applied to the column, recovered. The first, smaller peak contained almost half of the activity but also some catalase activity; the second, major peak contained no measurable cata~ase activity. Fractions containing glycollate oxidase free of contaminant catalase were pooled and used in subsequent assays. Preparation of Peroxisomes. Homogenates of spinach-beet leaves were prepared and fractionated essentially by the method of Leek et al. (1972). The homogenate was centrifuged at 500 x g for 25 rain, and the supernatant then centrifuged at 6,000 x g for 20 rain in a Servall Superspeed RC2 centrifuge. The 6,000 x g precipitate was resuspended in 0.5 M sucrose containing 20 mM glycylglycine-KOH buffer, pH 7.5, and layered on a step gradient of 10 ml of 1.2 M sucrose over 20 ml of 1.8 M sucrose, each contain-
ing the same buffer. The gradients were centrifuged at 65,000 x g for 4 h in the swingout head of a MSE Superspeed 65 centrifuge. The peroxisome pellet at the bottom of the tube was resuspended in 0.5 M sucrose containing 20 mM glycylglycine-KOH buffer, pH 7.5. Enzyme Assays. Glycollate oxidase was assayed at 25~ by the method of Zelitch and Ochoa (1953) in which the reduction of 2,6-dichlorophenolindophenoI (DCPIP) is measured spectrophotometrically by the fall in absorbance at 600 nm. Glycollate oxidase activity was also assayed using a Rank oxygen electrode, in which the 02 uptake following the addition of 30 ~tmol of sodium glycollate to a reaction mixture containing 200 gmol of Tris-HC1 buffer, pH 8.3, 3 lamol of FMN and the enzyme in a final volume of 3.0 ml was measured. Where indicated, 3 gmol of sodium azide was included in the reaction mixture to inhibit catalase activity. Catalase (I-I20 2 : H2 Oz oxidoreductase, EC 1.11.1.6) activity was assayed at 25~ either spectrophotometrically by measuring the decrease in absorbance at 240 nm (Liick, 1963) or by measuring oxygen release with the oxygen electrode from a reaction mixture containing the enzyme, 40 gmol of H20 2 and 60 ~tmol of KHzPO4KzHPO 4 buffer, pH 7.0, in a total volume of 3.0 ml. Decarboxylation. The release of 1~CO2 from 14C-labelled substrates was measured as described by Halliwell and Butt (1974), but using smaller volumes of reaction mixture contained in small tubes, closed with self-seal rubber stoppers, through which solutions could be injected or an aeration tube introduced. For glycollate, 6gmol of sodium glycollate, containing 0.4~tCi of [1-14C] glycollate with 30 nmol of FMN, 30 gmol of sodium barbitone-HC1 buffer, pH 8.3, and 0.05 ml of extract in a total volume of 0.6 ml were incubated at 25~ with shaking for 2 h (unless otherwise stated). For glyoxylate, 0.3 gmol of sodium glyoxylate, containing 0.25 gCi of [1-14C]glyoxylate, were substituted for glycollate. 14CO2 was collected in 0.2 ml of 20% (w]v) KOH contained in a small tube within the reaction tube, and 50 ~tl samples were taken and counted in the scintillation counter. Protein was measured by the method of Lowry, as modified by Leggett Bailey (1962), using desiccated bovine serum albumin as standard. Chlorophyll was determined spectrophotometrically, using the data of Arnon (t949).
Results
Oxygen Uptake and H202 Production during Gtycollate Oxidation with Purified Enzyme During the purification of glycollate oxidase by 700 times on a protein basis, catalase activity was removed at successive stages and completely in the final stage
Table 1. Purification of glycoltate oxidase from spinach-beet leaves. Spinach-beet leaves (400 g) were homogenised and the extracted enzyme purified as described in the Methods section. Glycollate oxidase and catalase activities were measured spectrophotometrically Purification stage
Glycollate oxidase
Catalase
Specific activity Recovery (units/rng protein) (%)
Specific activity Recovery (units/rag protein) (%)
A. Crude extract After pH 5.3 precipitation
0.062 0.088
100 98
38.2 31.1
100 96
B. (NF[g)2SO 4 precipitate (pH 5.3) (NH4)2SO4 precipitate (pH 8.3)
0.364 0.746
67 53
23.1 15.1
47 29
C. Eluate from Sephadex G-25 and CM-cellulose Eluate from DEAE-cellulose
4.01 23.2
D. Eluate from Sephadex G-200
43.5
35 28 9.3
6.72 0.91 ~
'0
8.3 2.1 0
B. Grodzinski and V.S. Butt: H202 Production and CO2 Release
was reduced by half, due to the destruction o f the H 2 0 2 p r o d u c e d in glycollate oxidation to p r o d u c e 0 2 (Fig. 1). W h e n 1 m M azide or 20 m M 3-amino1,2,4-triazole were included in the reaction mixture together with catalase, the 0 2 uptake was completely restored. These concentrations o f catalase inhibitor had no effect on glycollate oxidase activity in the absence o f catalase. W h e n catalase was absent, a b o u t 85% o f the glycollate oxidised was decarboxylated (Table 2). Catalase almost totally inhibited the CO2 release, whereas azide and aminotriazole had little effect. W h e n a m o u n t s o f [1-1*C]glyoxylate equivalent to less than 5% o f the glycollate supplied were included in the reaction mixture, similar quantities o f 14CO2 were released, showing that sufficient H 2 0 2 was generated during glycollate oxidation to react with glyoxylate and release CO2.
R
_~ ~oo
e,
5(
I
1
2
227
3
Time (rain.)
Fig, 1. Effect of catalase and catalase inhibitors on oxygen uptake during glycollate oxidation with purified glycollate oxidase. Oxygen uptake was monitored using an O2-electrode after addition of sodium glycollate (10raM) to purified enzyme incubated with FMN (6mM) in 50mM sodium barbitone-HCt buffer, pH8.3, under air at 25~ (A), in the presence of added (B) liver catalase (18,000 units), (C) 20 mM 3-am~no-l,2,4-triazole, (D) 1 mM azide. (E) was similar to (A) but with twice the quantity of enzyme
14C02release from [1-~4C]glycollate by purified glycollate oxidase. Purified glycollate oxidase (0.163 units) was incubated with glycollate (10 raM) in the presence of bovine serum albumin (1.5 mg/ml) under pure oxygen and the conditions described in the Methods section; 02 uptake was measured with the O2-electrode and available H202 production from the rate with added catalase (1,800 units). ~4CO2 release from gIycollate (30 ~tmol) containing [l-~aC]glycollate (0.56 laCi) was measured after 15 min incubation. Results are expressed as ~mol rain- 1 mg protein i Table 2. Rate of production of available peroxide and
Reagents added
02 uptake
Available H202
COz released
None 1 mM azide 20 mM aminotriazole Catalase (480 units) Catalase (1,800 units) 5 mM Pyrid-2-yl-~-hydroxymethane sulphonate
43 42 41 23 22 2.4
42 40 38 2 0 -
36.1 34.3 32.8 0.6 0.1 1.3
Oxygen Uptake and C O 2 Release during Glycollate Oxidation with Peroxisomes Glycottate-dependent 0 2 uptake was demonstrated with peroxisome preparations f r o m spinach-beet leaves (Fig. 2). 0.05 m M F M N was included in the assay mixture; 02 uptake fell by 20% in its absence. 0 . 5 m M P H M S completely, inhibited glycollate-dependent 02 uptake. W h e n 1 m M azide or 20 m M aminotriazole were included in the reaction mixture, the rate o f 02 uptake
m
by passage t h r o u g h Sephadex G-200 (Table 1). The purified enzyme required the addition o f 0.05 m M F M N for m a x i m u m 02 uptake; no significant 02 u p t a k e was observed with F M N alone or when added to the boiled enzyme. 5 m M P H M S almost completely inhibited 0 2 uptake. By following the change in O2 u p t a k e during glycollate oxidation in a sealed system, the K m for 02 was f o u n d to be 0.17 m M . W h e n a large excess o f bovine liver catalase was added to the reaction mixture, the rate o f 02 uptake
0.1C
B
I 0,5
I 1.0
i'ime (rain.) Fig. 2. E f f e c t o f c a t a l a s e a n d a m i n o t r i a z o l e o n o x y g e n u p t a k e d u r i n g glycollate oxidation with spinach-beet leaf peroxisomes.
Oxygen uptake was measured under the conditions in Fig. l with the substitution of spinach-beet peroxisomes (containing 0.38 mg protein) for enzyme (A), in the presence of added (B) liver catalase (4,600 units) or (C) 20 mM 3-amino-l,2,4-triazole
228
B. Grodzinski and V.S. Butt: H202 Production and CO2 Release Table 3. Rate of production of available H20 / and 14CO2 release
100
"ac
=l
o
f
50
= . _*
from [1-14C]glycollate by spinach-beet peroxisomes. Spinach-beet peroxisomal preparation was incubated with glycollate (10 mM) as described in the Methods section. The rates of production of available H2Oz were calculated from 02 uptake with added catalase (4,600 units) under pure 02; the figures in parentheses give the rates under air. ~CO2 production was measured under air. Results are expressed as gmol h- 1 mg protein- 1. Reagents added
Expt.
None
A B
1 mM azide
A B A B
~.
G,
.m
I ?
I 8
i 9
pH
Fig. 3. Effect ofpH on the ~ates of production of available peroxide and 14CO2 release from ['l-a4C]glycollate by peroxisomes. The rate of production of available :peroxide . ( e - - e ) was calculated from oxygen uptake during glycollate oxidation in the presence and absence of added catalase (4,600 units) under the ,conditions in Fig. 2. 14CO2 release ( o - - o ) w a s measured over a ~period of 2 h after addition of [1-1~C]gly~ollate to the mixture. 50 m M sodium barbitone-HCl buffers over the range pH 7.0 to 9:0 were used
was considerably increased but was less than ,doubled (Fig. 2), as would be expected if peroxisomat catalase activity were sufficient to destroy completely the peroxide produced by glycollate oxidation. Furthermore, the addition of bovine liver catalase in quantities equivalent to 50 to 200 times the catalase activity of the peroxisomes reduced the rate of O2 uptake to half of that observed when either of the catalase inhibitors were present. The almost exact stoichiometry suggests that this reduction in the rate of O2 uptake was due, not to inhibition by the added catalase, but to the destruction of H 2 0 2 which had escaped attack by peroxisomal catalase. The rate at which this H2O 2 was made available for other reactions was given by the difference between the control rate and the rate of 02 uptake with added catalase. When the rate of production of "available" H2Oz was measured by this method over a range of pH, the maximum rate was observed at pH 8.3 (Fig. 3) identical with that recorded for glycollate oxidase (Zelitch and Ochoa, 1953) and for the peroxisomal enzyme. This maximum was markedly different from that of peroxisomal catalase at about pH 7.0. It did however correspond with the pH optimum for the release of 14CO2 from [1-14C]glycollate (Fig. 3). The relation between the production of"available" H202 and CO2 release from glycollate was further evident from a comparison of their rates (Table 3). Although the rate of production o f " available" H 202 was determined over a period of only a few minutes with
Catalase (4,600 .units)
Available HzO2 1.36 (0.76) 2.18 14.2 i9.8) 17.8 -
CO2 released 0.52 0.92 9.5 11,6 0.08 0.12
the Oz electrode, this is shown to be rather more than sufficient over 2 h to :account for the 14CO2 released from [1-~4C]glycollate over that period both with different peroxisomal preparations, at 21% O2 and 100% Oz, and when catalase activity was inhibited by azide. The ,addition of a large excess of bovine liver catalase reduced the available peroxide to very low levels, but even with a 200-fold excess, some 14COz release was detectable. Under pure oxygen, the rate of HzOz production was increased along with an increase in the rate of CO2 release.
Time Course of COz Release from Glycollate by Peroxisomes The rates of C O 2 release recorded in Table 3 and those given by Halliwell and Butt (1974) were calculated from the release measured over the first 2 h of incubation. With peroxisomes, however, the reaction course was not linear with time but attained a constant rate after a lag period of 60 to 90 rain. In order to assess the rates more exactly, the reaction course was examined in detail with special reference to any changes in enzyme activity which may occur. Fig. 4a shows the typical time course for 14CO2 release from [1-t4C]glycollate with spinach-beet peroxisomes. The usual increase in the rate of CO/ release to a maximum after 90 min incubation was observed. However, glycollate appeared to be oxidised at close to its maximum rate from zero time. The activity of glycollate oxidase fell slowly with time to about 85% of its initial activity after 3 h. By comparison, catalase activity fell rapidly to about 80% of its initial activity after 1 h, but thereafter more slowly. The rapid fall in catalase activity coincided with the increase in 14COz release from [1-14C]glycol-
B. Grodzinskiand V.S. Butt: H202 Productionand CO2 Release
Distribution of Glycollate Oxidase and Catalase in Subcellular Fractions
A 6
4
-~ 2
B v
~8 4 n
I
I
I
~ 100~~~~~%~~ ;~-,~
80
g 2 0
229
1
2 Time(h)
5
4
Fig. 4. T i m e course of l~COz release f r o m [1-14C]gtycollate and
[1-14C]glyoxylate, and of production available peroxide_ Spinachbeet peroxisomes(A, 91 lag.; B, 76 lag. protein) were incubated with (A) [1-14C]glycollate (5raM), (B) [1-14C]glyoxylate (5 mM; o--o), and sodium glycollate(5 raM) with [1-~4C]glyoxylate (0.5 raM; o--o), under the conditionsin Fig. 2. Samples (200 to 410 gg protein) were extracted at intervals from incubationvessels containingglycollate(5 mM) for assay of production of available peroxide (A--A glycollateoxidase activity(a--zx), and cata lase activity in the presence of 5 mM PHMS (n--D), using the 02 electrode
late during the lag period and was also reflected in an increased rate of production of "available" H202, which reached a maximum after 2 h (Fig. 4c). Over the period of maximum H202 production, 1~CO2 release was most rapid and declined as the H202 production declined after 3 h. When trace quantities of [1- l 4C]glyoxylate equivalent to 5% of unlabelled glycollate were included in the reaction mixture, the time course of 14CO2 release was similar to that observed when [1-14C]glycollate was supplied. It is unlikely that the lag response observed could be due to the accumulation of glyoxylate to some necessary level before decarboxylation became maximal, nor that the catalase was undergoing a simple thermal destruction, since there was a linear release of 14CO2 when the same quantity of [1-~4C]glyoxylate was supplied in the absence of any glycollate (Fig. 4b).
The data presented substantiate the view that the 14CO2 released from [1-14C]glycollate by leaf peroxisomes is due to the attack on glyoxylate by H202 escaped from catalase action. That this was not because catalase was lost preferentially during the preparation of peroxisomes is shown in Table 4, in which the proportions of catalase and glycollate oxidase were very similar in each fraction. The recovery of both enzymes in the peroxisomes was low, but in agreement with the data of Lips (1975), the proportion of catalase recovered in this pellet was rather higher than that of glycollate oxidase. The data presented in Table 4 allow the calculation of the rates of CO2 release from glycollate by the homogenate. From the initial rate of production of "available" H202, about 10% of the H202 produced in glycollate oxidation was not destroyed (Fig. 2), giving a maximum rate of 1 gmol of CO2 released/h per mg of protein, or, with a 4% recovery of glycollate oxidase, a total rate of 10 ~tmol of CO2 released/h per mg of chlorophyll in the homogenate. The maximum rate, observed over the 90 rain to 180 min period of incubation, corresponded to 18 gmol of CO2 released/h per mg of chlorophyll in the homogenate. Discussion
Glycollate oxidase catalyses the oxidation of glycollate to glyoxylate by molecular oxygen with the formation of H202 (Reaction 1), which non-enzymically oxidises the glyoxylate further to formate and CO2 (Reaction 2), the COz being derived uniquely from the carboxyl group (Zelitch and Ochoa, 1953): C H 2 O H "C O O H + 02 - * C H O . C O O H + H 2 0 2
(1)
C H O - C O O H + H202--* H- C O O H + CO2.
(2)
One mole of oxygen is consumed for each mole of glycollate oxidised. The enzyme is located mainly, and perhaps entirely, in the leaf peroxisomes (Tolbert et al., 1969), in which catalase is also concentrated. Catalase competes with Reaction 2 for H202, and both inhibits CO2 release and reduces the net 02 uptake: H202 - + H 2 0 -/- 1 0 2 .
(.3)
The balance between Reactions 2 and 3 has been measured here by determining the changes in 02 uptake when catalase activity was completely inhibited by 1 mM azide or 20 m M aminotriazole and when excess bovine liver catalase was added. The reduction in the rate of 02 uptake when excess cata-
230
B. Grodzinski and V.S. Butt: H202 Production and C02 Release
Table 4. Distribution of glycollate oxidase and catalase in subcellular fractions of spinach-beet leaves. Spinach-beet leaves (about (150 g) were homogenised and the extract (containing 195 mg chlorophyll) fractionated as described in the Methods section. Glycollate oxidase and catalase activities were determined using the O2 electrode Fraction
(a) Differential centrifugation Crude homogenate 500 x g precipitate 6,00 x g supernatant 6,000 x g precipitate (b) Centrifugation of 6,000 • precipitate through 1.8 M sucrose Peroxisomal pellet Supernatant
Glycollate oxidase
Catalase
Total activity (units)
Recovery (%)
Total activity (units x 10- 3)
Recovery (%)
234 9 194 37
100 4 82 16
128 5 92 25
100 4 72 20
10 23
4 10
11 13
9 10
lase was added is a measure of the H202 available for Reaction 2 and not broken down in Reaction 3. With the purified enzyme, the addition of catalase exactly halved the rate of 02 uptake; this was restored by adding azide or aminotriazole. On the other hand, the rate of 02 uptake with leaf peroxisomes was decreased by little more than 10% when catalase was added, but raised to twice this value when azide or aminotriazole were added. It is clear therefore that as much as 90% of the H202 produced in Reaction 1 by the action of peroxisomal glycollate oxidase is destroyed by peroxisomal catalase or other reactions. It has been claimed that catalase activity in the peroxisomes is sufficient to destroy completely the H202 produced in Reaction 1 (Kisaki and Tolbert, 1969), but the concentrations of glycollate and glyoxylate used may have been too low to detect the release of 14CO 2 from Reaction 2. In rat liver preparations it has been estimated that 40% to 80% of the H202 produced by peroxisomes is destroyed by their catalase (Boveris et al., 1972). Further, the leaf catalase activity may be overestimated since at pH 8.3, it is about one-third of its maximum at pH 7.0, while the enzyme may be relatively ineffective with low concentrations of H202, if its Km is above 20 mM as reported for bovine catalase (von Euler and Josephson, 1927). Peroxisomal catalase activity at pH 8.3 may therefore be sufficient to hold H202 at non-toxic levels but not to destroy it completely. Reaction 2 proceeds with this residual H202, which amounts to at least 10% of that generated in Reaction 1, and it should be noted that the addition of a very large excess of bovine liver catalase failed to suppress Reaction 2 completely. This reaction is therefore likely to proceed during glycollate oxidation in leaves and hence to contribute to the CO2 released in photorespiration.
Any relation between the rates of CO2 release in photorespiration and from glycollate oxidation in peroxisomes is difficult to establish because the conditions under which the peroxisomal oxidations take place in vivo are not known. Thus, the rates of glycollate oxidation and CO2 production from peroxisomes have usually been measured in air but, when 02 is rapidly evolved during photosynthesis in leaves, a more exact estimate of these reactions might be obtained under pure 02. These rates were found to be appreciably higher, as would be expected since the Km for 02 observed with purified glycollate oxidase was 0.17 mM (cf. 0.13 mM for pea-leaf glycollate oxidase, Kerr and Groves, 1975) and the enzyme has been reported to be saturated in atmospheres containing 60% 02 and above (see Andrews et al., 1973). Again, the time course of CO2 release from peroxisomes is not linear apparently because of the inactivation of peroxisomal catalase during the reaction, perhaps due to the high steady-rate of H202 production (Chance, 1950). Estimates of catalase activity in leaf extracts may therefore be high and of CO2 release from glycollate in peroxisomes correspondingly low; some measurement of catalase activity in vivo is necessary to obtain relevant estimates for CO2 release in illuminated leaves from the peroxisomal oxidation of glycollate. With these reservations, some comparison can be made between the rates of CO2 release during the peroxisomal oxidation of glycollate and in the photorespiration of leaves. From his survey of the rates of photorespiration in the leaves of a number of species, Zelitch (1975) concluded that any biochemical mechanism must account for rates of at least 76 gmol of COg released/h per mg of chlorophyll. However, the highest rates observed in the experiments reported here,' carried out at 25 ~ C, are barely one-quarter of
B. Grodzinski and V.S. Butt: H2Oa Production and CO 2 Release
this requirement. The extension of these observations to species with high photorespiratory rates and the determination of the rate of photorespiration of spinach-beet leaves at high light intensities may indicate the extent of the discrepancy between these observations. As well as CO2, both formate and glyoxylate are formed in the reaction mixture. Formate can be oxidised to CO2 in peroxisomes (Leek et al., t972) and other organelles (Halliwell, 1974) but the CO2 here was derived from C-2 of glyoxylate. Halliwell (1973) also showed spinach-beet leaf homogenates to possess sufficient formyl-tetrahydrofolatesynthetase (formate: tetrahydrofolate ligase (ADP), EC 6.3.4.3) activity to convert any formate produced in a first step leading to the hydroxymethylation of glycine to serine. Up to 80% of the glycollate oxidised in peroxisomes can be isolated as glyoxylate, which can be further oxidised to oxalate (Halliwell and Butt, 1974), aminated by serine or glutamate to form glycine (Tolbert, 1971) or oxidised in illuminated chloroplasts to formate and CO2 (Zelitch, 1972). While these further reactions also give rise to CO2, the CO2 released in the peroxisomal reactions reported here must provide some part of the total CO2 release in photorespiration.
References Andrews, T.J., Lorimer, G.H., Toibert, N.E.: Ribulose diphosphate oxygenase. I. Synthesis of phosphoglycolate by fraction-1 protein of leaves. Biochemistry 12, 11-18 (1973) Arnon, D.I. : Copper enzymes in isolated chloroplasts. PoIyphenol oxidase in Beta vulgaris. Plant Physiol. 24, 1-15 (1949) Boveris, A., Oshino, N., Chance, B.: The cellular production of hydrogen peroxide. Biochem. J. 138, 77-85 (1974) Chance, B. : The reactions of catalase in the presence of the notatin system. Biochem. J. 46, 387402 (1950) Elstner, E.F., Heupel, A. : On the decarboxylation of ~-keto acids by isolated chloroplasts. Biochim. biophys. Acta (Amst.) 325, 182-188 (1973) Euler, H. von, Josephson, K.: Uber Katalase. II. Ann. Chem. 455, 1 16 (1927)
231 Halliwell, B. : The role of formate in photorespiration. Trans. Biochem. Soc. 1, 1147-1150 (1973) Halliwell, B. : Oxidation of formate by peroxisomes and mitochondria from spinach leaves. Biochem. J. 138, 77-85 (1974) Halliwell, B., Butt, V.S.: Oxidative decarboxylation of glycollate and glyoxylate by leaf peroxisomes. Biochem. J. 138, 217~24 (1974) Halliwell, B., Leek, A.E., Butt, V.S.: Oxidative decarboxylation of glycollate and glyoxylate by leaf peroxisomes. Biochem. J. 128, 87p-88p (1972) Kerr, M.W., Groves, D. : Purification and properties of glycollate oxidase from Pisum sativum leaves. Phytochemistry 14, 359-362 (1975) Kisaki, T., Tolbert, N.E.: Glycolate and glyoxylate metabolism by isolated peroxisomes or chloroplasts. PIant Physiol. 44, 242250 (1969) Kisaki, T., Yoshida, N., Imai, A. : Glycine decarboxylase and serine formation in spinach leaf mitochondrial preparation with reference to photorespiration. Plant Cell Physiol. 12, 275-288 (1971) Leek, A.E., Halliwell, B., Butt, V.S.: Oxidation of formate and oxalate in peroxisomal prearations from leaves of spinach beet (Beta vulgaris L,). Biochim. biophys. Acta (Amst.) 286, 299-311 (1972) Leggett Bailey, J. : Techniques in protein chemistry, p. 293. Amsterdam: Elsevier 1962 Lips, S.H. : Enzyme content of plant microbodies as affected by experimental procedures. Plant Physiol. 55, 5984501 (1975) Ltick, H. : Catalase. In : Methods of Enzymatic Analysis. pp. 886888, Ed.: Bergmeyer, H.U., New York: Academic Press 1963 Tolbert, N.E. : Microbodies-peroxisomes and glyoxysomes. Ann. Rev. Plant Physiol. 22, 45-74 (1971) Tolbert, N.E., Oeser, A., Yamazaki, R.K., Hageman, R.H., Kisaki, T. : A survey of plants for leaf peroxisomes. Plant Physiol. 44, 135-147 (1969) Zelitch, I.: Increase rate of net photosynthetic carbon dioxide uptake caused by the inhibition of glycolate oxidase. Plant Physiol. 41, 1623-1631 (1966) Zelitch, I.: Photosynthesis, photorespiration and plant productivity. New York-London: Academic Press 1971 Zelitch, I.: The photooxidation of glyoxylate by envelope-free spinach chloroplasts and its relation to photorespiration. Arch. Biochem. 150, 698-707 (1972) Zelitch, I.: Improving the efficiency of photosynthesis. Science 188, 6264533 (1975) Zelitch, I., Ochoa, S.: Oxidation and reduction of glycolic and glyoxylic acids in plants. I. Glycolic acid oxidase. J. biol. Chem. 201, 707 718 (1953)
Received 22 September; accepted 14 October 1975
Planta (Berl.) 128, 225-231 (1976)
9 by Springer-Verlag 1976
Hydrogen Peroxide Production and the Release of Carbon Dioxide during Glycollate Oxidation in Leaf Peroxisomes* B. Grodzinski and V.S. Butt Botany School, Oxford University, South Parks Road, Oxford OX1 3RA, U.K.
Summary. The rate at which H202 becomes available during glycollate oxidation for further oxidation reactions, especially that of glyoxylate to formate and CO2, in peroxisomes from spinach-beet (Beta vulgaris L., vat. vulgaris) leaves has been determined by measuring Oz uptake in the presence and absence of added catalase. The rates observed under air and pure 02 were sufficient to account for the 14COz released from [1-~4C]glycollate under these conditions; the two reactions showed similar characteristics. In the course of the reaction, a fall in catalase activity was observed concomitant with an increase in 14C02release. There is no evidence that catalase was disproportionately lost f r o m the peroxisomes during isolation, and it is argued that the COz release observed contributes to the photorespiratory CO2 loss in intact leaves.
destroy all of the H202 generated during glycollate oxidation (Tolbert, 1971), this CO2 release is a consequence of the non-enzymic oxidation of glyoxylate by HzO2 which has escaped catalase action (Halliwell and Butt, 1974). Further evidence for the action of H202 during the oxidative decarboxylation of glycollate in peroxisomes is presented here. A method has been devised to estimate the rate o f H 2 0 2 generation in excess of that destroyed by catalase and this rate has been related to that of CO2 release during the course of the peroxisomal oxidation of glycollate. Its significance in relation to CO2 release in photorcspiration is discussed.
Materials and Methods
Introduction The CO2 released in the photorespiration of leaves, especially at higher temperatures, has been shown to arise predominantly from the carboxyl group of glycollate (Zelitch, 1966). Two processes, which could proceed at rates adequate to account for this release, have been demonstrated. The immediate substrate could be glycine, decarboxylated in leaf mitochondria (Yolbert, 1971; Kisaki etal., 1971), or glyoxylate, oxidised in illuminated chloroplasts (Zelitch, 1972) under the action of H202 and the superoxide free radial anion, - O z - (Elstner and Heupel, 1973). It has however been shown that 14C02 is released from [1-14C]glycollate in the peroxisomes from spinach-beet leaves (Halliwell et al., 1972) and further that, although there is generally assumed to be sufficient catalase present in the peroxisomes to
* Abbreviations: DCPIP = 2,6-dichlorophenolindophenol ; FMN = flavin mononucleotide.
Chemicals. Reagents used were of the highest purity available from BDH ChemicalsLtd., Poole, Dorset, U.K. or from Sigma Chemical Co., Kingston-upon-Thames, Surrey. U.K. Pyrid-2-yl-~-hydroxymethanesulphonate (PHMS) was purchased from Aldrich Chemical Co. Ltd., Wembley, London, U.K. and bovine liver catalase from Boehringer Corporation, London W5, U.K. 5-(4-Biphenylyl)-2-(4t-butylphenyl)-l-oxa-3,4-diazole (butyl-PBD) for liquid scintillation counting was obtained from CIBA-GeigyLtd., Duxford, Cambridge, U.K. Sodium [1-1~C] glycollate and sodium [1-14C]glyoxylatewere obtained from Radiochemical Centre, Amersham, Bucks, U.K. and stored for 1-2 weeks in the deep-freeze at -20~ until required. Plants. Leaves of spinach beet (Beta vulgaris vat. vulgaris) were harvested from plants grown from seed outdoors in the summer, and in the greenhouse, receiving supplementary mercury lighting for 4 h each day, during the winter. Glycollate oxidase (Glycollate:oxygen oxidoreductase, EC 1.1.3. I) was extracted and purified by a procedure, modified from that of Zelitch and Ochoa (1953): A. Extraction. About 400 g of spinach-beet leaves, from which the midribs and coarser veins had been removed, were homogenized in 1,200ml of chilled 0.1 M Tris-HC1 buffer, pH 8.0, containing 2% (w/v) polyvinylpyrollidone (Polyclar AT) for two successive 20 s periods at top speed in an MSE Ato-Mix homogenizer. All subsequent operations were carried out at 2 4 ~C.
226
B. Grodzinski and V.S. Butt: H202 Production and CO2 Release
The homogenate was strained through four layers of muslin and centrifuged at 500 x g for 10rain. The pH of the supernatant liquid was carefully adjusted to pH 5.3 by the slow addition of prechilled 10% (v/v) acetic acid with constant stirring. After standing for 30 min, the precipitate was removed by centrifugation at 20,000 x g for 20 min. B. Ammonium sulphate precipitation. 140 g of (NH4)2SO~ was added to each litre of supernatant slowly with continuous stirring. The precipitate was removed by centrifugation and a further 80 g of (NH~)zSO4 added to each litre of supernatant. The resulting precipitate was collected by centrifugation and resuspended in 0.1 M Tris-HC1 buffer, pH 8.3. These two (NH4)2SO4 precipitations were repeated after raising the pH to 8.3 with KOH, and the second precipitate suspended in 10mM Tris-HC1 buffer, pH 7.5, containing 2 r a m flavin mononucleotide (FMN). C. Ion-Exchange Chromatography. The solution was applied to a column (15 cm long x2.0 cm diam.) of Sephadex G-25 (fine), which had been equilibrated with 10 mM Tris-HC1 buffer, pH 7.5. The first protein band collected contained most of the glycollate oxidase and catalase activity. FMN was added to this fraction, which was then applied to a column (20 cm long x 2.5 cm diam.) of carboxymethyl (CM)-cellulose, which had also been equilibrated with 10 mM Tris-HC1 buffer, pH 7.5. The first protein band containing glycollate oxidase activity was collected and then layered on a similar column of DEAE-cellulose (DE-52), equilibriated with the same buffer. The first protein band to emerge from this column again contained most of the glycollate oxidase activity, The enzyme was precipitated from the eluate by the slow addition of 0.6 vol. of saturated (NH4)2SO 4 solution and stored overnight in suspension. D. Sephadex G-200 Chromatography. The enzyme was sedimented by centrifugation, resuspended in 0.1 M Tris-HCI buffer, pH 8.0, containing 5 mM FMN, and layered on a column (36 cm long x 3.0 cm diam.) of Sephadex G-200. The eluate was collected and two protein peaks, accounting for 80% of the glycollate oxidase applied to the column, recovered. The first, smaller peak contained almost half of the activity but also some catalase activity; the second, major peak contained no measurable cata~ase activity. Fractions containing glycollate oxidase free of contaminant catalase were pooled and used in subsequent assays. Preparation of Peroxisomes. Homogenates of spinach-beet leaves were prepared and fractionated essentially by the method of Leek et al. (1972). The homogenate was centrifuged at 500 x g for 25 rain, and the supernatant then centrifuged at 6,000 x g for 20 rain in a Servall Superspeed RC2 centrifuge. The 6,000 x g precipitate was resuspended in 0.5 M sucrose containing 20 mM glycylglycine-KOH buffer, pH 7.5, and layered on a step gradient of 10 ml of 1.2 M sucrose over 20 ml of 1.8 M sucrose, each contain-
ing the same buffer. The gradients were centrifuged at 65,000 x g for 4 h in the swingout head of a MSE Superspeed 65 centrifuge. The peroxisome pellet at the bottom of the tube was resuspended in 0.5 M sucrose containing 20 mM glycylglycine-KOH buffer, pH 7.5. Enzyme Assays. Glycollate oxidase was assayed at 25~ by the method of Zelitch and Ochoa (1953) in which the reduction of 2,6-dichlorophenolindophenoI (DCPIP) is measured spectrophotometrically by the fall in absorbance at 600 nm. Glycollate oxidase activity was also assayed using a Rank oxygen electrode, in which the 02 uptake following the addition of 30 ~tmol of sodium glycollate to a reaction mixture containing 200 gmol of Tris-HC1 buffer, pH 8.3, 3 lamol of FMN and the enzyme in a final volume of 3.0 ml was measured. Where indicated, 3 gmol of sodium azide was included in the reaction mixture to inhibit catalase activity. Catalase (I-I20 2 : H2 Oz oxidoreductase, EC 1.11.1.6) activity was assayed at 25~ either spectrophotometrically by measuring the decrease in absorbance at 240 nm (Liick, 1963) or by measuring oxygen release with the oxygen electrode from a reaction mixture containing the enzyme, 40 gmol of H20 2 and 60 ~tmol of KHzPO4KzHPO 4 buffer, pH 7.0, in a total volume of 3.0 ml. Decarboxylation. The release of 1~CO2 from 14C-labelled substrates was measured as described by Halliwell and Butt (1974), but using smaller volumes of reaction mixture contained in small tubes, closed with self-seal rubber stoppers, through which solutions could be injected or an aeration tube introduced. For glycollate, 6gmol of sodium glycollate, containing 0.4~tCi of [1-14C] glycollate with 30 nmol of FMN, 30 gmol of sodium barbitone-HC1 buffer, pH 8.3, and 0.05 ml of extract in a total volume of 0.6 ml were incubated at 25~ with shaking for 2 h (unless otherwise stated). For glyoxylate, 0.3 gmol of sodium glyoxylate, containing 0.25 gCi of [1-14C]glyoxylate, were substituted for glycollate. 14CO2 was collected in 0.2 ml of 20% (w]v) KOH contained in a small tube within the reaction tube, and 50 ~tl samples were taken and counted in the scintillation counter. Protein was measured by the method of Lowry, as modified by Leggett Bailey (1962), using desiccated bovine serum albumin as standard. Chlorophyll was determined spectrophotometrically, using the data of Arnon (t949).
Results
Oxygen Uptake and H202 Production during Gtycollate Oxidation with Purified Enzyme During the purification of glycollate oxidase by 700 times on a protein basis, catalase activity was removed at successive stages and completely in the final stage
Table 1. Purification of glycoltate oxidase from spinach-beet leaves. Spinach-beet leaves (400 g) were homogenised and the extracted enzyme purified as described in the Methods section. Glycollate oxidase and catalase activities were measured spectrophotometrically Purification stage
Glycollate oxidase
Catalase
Specific activity Recovery (units/rng protein) (%)
Specific activity Recovery (units/rag protein) (%)
A. Crude extract After pH 5.3 precipitation
0.062 0.088
100 98
38.2 31.1
100 96
B. (NF[g)2SO 4 precipitate (pH 5.3) (NH4)2SO4 precipitate (pH 8.3)
0.364 0.746
67 53
23.1 15.1
47 29
C. Eluate from Sephadex G-25 and CM-cellulose Eluate from DEAE-cellulose
4.01 23.2
D. Eluate from Sephadex G-200
43.5
35 28 9.3
6.72 0.91 ~
'0
8.3 2.1 0
B. Grodzinski and V.S. Butt: H202 Production and CO2 Release
was reduced by half, due to the destruction o f the H 2 0 2 p r o d u c e d in glycollate oxidation to p r o d u c e 0 2 (Fig. 1). W h e n 1 m M azide or 20 m M 3-amino1,2,4-triazole were included in the reaction mixture together with catalase, the 0 2 uptake was completely restored. These concentrations o f catalase inhibitor had no effect on glycollate oxidase activity in the absence o f catalase. W h e n catalase was absent, a b o u t 85% o f the glycollate oxidised was decarboxylated (Table 2). Catalase almost totally inhibited the CO2 release, whereas azide and aminotriazole had little effect. W h e n a m o u n t s o f [1-1*C]glyoxylate equivalent to less than 5% o f the glycollate supplied were included in the reaction mixture, similar quantities o f 14CO2 were released, showing that sufficient H 2 0 2 was generated during glycollate oxidation to react with glyoxylate and release CO2.
R
_~ ~oo
e,
5(
I
1
2
227
3
Time (rain.)
Fig, 1. Effect of catalase and catalase inhibitors on oxygen uptake during glycollate oxidation with purified glycollate oxidase. Oxygen uptake was monitored using an O2-electrode after addition of sodium glycollate (10raM) to purified enzyme incubated with FMN (6mM) in 50mM sodium barbitone-HCt buffer, pH8.3, under air at 25~ (A), in the presence of added (B) liver catalase (18,000 units), (C) 20 mM 3-am~no-l,2,4-triazole, (D) 1 mM azide. (E) was similar to (A) but with twice the quantity of enzyme
14C02release from [1-~4C]glycollate by purified glycollate oxidase. Purified glycollate oxidase (0.163 units) was incubated with glycollate (10 raM) in the presence of bovine serum albumin (1.5 mg/ml) under pure oxygen and the conditions described in the Methods section; 02 uptake was measured with the O2-electrode and available H202 production from the rate with added catalase (1,800 units). ~4CO2 release from gIycollate (30 ~tmol) containing [l-~aC]glycollate (0.56 laCi) was measured after 15 min incubation. Results are expressed as ~mol rain- 1 mg protein i Table 2. Rate of production of available peroxide and
Reagents added
02 uptake
Available H202
COz released
None 1 mM azide 20 mM aminotriazole Catalase (480 units) Catalase (1,800 units) 5 mM Pyrid-2-yl-~-hydroxymethane sulphonate
43 42 41 23 22 2.4
42 40 38 2 0 -
36.1 34.3 32.8 0.6 0.1 1.3
Oxygen Uptake and C O 2 Release during Glycollate Oxidation with Peroxisomes Glycottate-dependent 0 2 uptake was demonstrated with peroxisome preparations f r o m spinach-beet leaves (Fig. 2). 0.05 m M F M N was included in the assay mixture; 02 uptake fell by 20% in its absence. 0 . 5 m M P H M S completely, inhibited glycollate-dependent 02 uptake. W h e n 1 m M azide or 20 m M aminotriazole were included in the reaction mixture, the rate o f 02 uptake
m
by passage t h r o u g h Sephadex G-200 (Table 1). The purified enzyme required the addition o f 0.05 m M F M N for m a x i m u m 02 uptake; no significant 02 u p t a k e was observed with F M N alone or when added to the boiled enzyme. 5 m M P H M S almost completely inhibited 0 2 uptake. By following the change in O2 u p t a k e during glycollate oxidation in a sealed system, the K m for 02 was f o u n d to be 0.17 m M . W h e n a large excess o f bovine liver catalase was added to the reaction mixture, the rate o f 02 uptake
0.1C
B
I 0,5
I 1.0
i'ime (rain.) Fig. 2. E f f e c t o f c a t a l a s e a n d a m i n o t r i a z o l e o n o x y g e n u p t a k e d u r i n g glycollate oxidation with spinach-beet leaf peroxisomes.
Oxygen uptake was measured under the conditions in Fig. l with the substitution of spinach-beet peroxisomes (containing 0.38 mg protein) for enzyme (A), in the presence of added (B) liver catalase (4,600 units) or (C) 20 mM 3-amino-l,2,4-triazole
228
B. Grodzinski and V.S. Butt: H202 Production and CO2 Release Table 3. Rate of production of available H20 / and 14CO2 release
100
"ac
=l
o
f
50
= . _*
from [1-14C]glycollate by spinach-beet peroxisomes. Spinach-beet peroxisomal preparation was incubated with glycollate (10 mM) as described in the Methods section. The rates of production of available H2Oz were calculated from 02 uptake with added catalase (4,600 units) under pure 02; the figures in parentheses give the rates under air. ~CO2 production was measured under air. Results are expressed as gmol h- 1 mg protein- 1. Reagents added
Expt.
None
A B
1 mM azide
A B A B
~.
G,
.m
I ?
I 8
i 9
pH
Fig. 3. Effect ofpH on the ~ates of production of available peroxide and 14CO2 release from ['l-a4C]glycollate by peroxisomes. The rate of production of available :peroxide . ( e - - e ) was calculated from oxygen uptake during glycollate oxidation in the presence and absence of added catalase (4,600 units) under the ,conditions in Fig. 2. 14CO2 release ( o - - o ) w a s measured over a ~period of 2 h after addition of [1-1~C]gly~ollate to the mixture. 50 m M sodium barbitone-HCl buffers over the range pH 7.0 to 9:0 were used
was considerably increased but was less than ,doubled (Fig. 2), as would be expected if peroxisomat catalase activity were sufficient to destroy completely the peroxide produced by glycollate oxidation. Furthermore, the addition of bovine liver catalase in quantities equivalent to 50 to 200 times the catalase activity of the peroxisomes reduced the rate of O2 uptake to half of that observed when either of the catalase inhibitors were present. The almost exact stoichiometry suggests that this reduction in the rate of O2 uptake was due, not to inhibition by the added catalase, but to the destruction of H 2 0 2 which had escaped attack by peroxisomal catalase. The rate at which this H2O 2 was made available for other reactions was given by the difference between the control rate and the rate of 02 uptake with added catalase. When the rate of production of "available" H2Oz was measured by this method over a range of pH, the maximum rate was observed at pH 8.3 (Fig. 3) identical with that recorded for glycollate oxidase (Zelitch and Ochoa, 1953) and for the peroxisomal enzyme. This maximum was markedly different from that of peroxisomal catalase at about pH 7.0. It did however correspond with the pH optimum for the release of 14CO2 from [1-14C]glycollate (Fig. 3). The relation between the production of"available" H202 and CO2 release from glycollate was further evident from a comparison of their rates (Table 3). Although the rate of production o f " available" H 202 was determined over a period of only a few minutes with
Catalase (4,600 .units)
Available HzO2 1.36 (0.76) 2.18 14.2 i9.8) 17.8 -
CO2 released 0.52 0.92 9.5 11,6 0.08 0.12
the Oz electrode, this is shown to be rather more than sufficient over 2 h to :account for the 14CO2 released from [1-~4C]glycollate over that period both with different peroxisomal preparations, at 21% O2 and 100% Oz, and when catalase activity was inhibited by azide. The ,addition of a large excess of bovine liver catalase reduced the available peroxide to very low levels, but even with a 200-fold excess, some 14COz release was detectable. Under pure oxygen, the rate of HzOz production was increased along with an increase in the rate of CO2 release.
Time Course of COz Release from Glycollate by Peroxisomes The rates of C O 2 release recorded in Table 3 and those given by Halliwell and Butt (1974) were calculated from the release measured over the first 2 h of incubation. With peroxisomes, however, the reaction course was not linear with time but attained a constant rate after a lag period of 60 to 90 rain. In order to assess the rates more exactly, the reaction course was examined in detail with special reference to any changes in enzyme activity which may occur. Fig. 4a shows the typical time course for 14CO2 release from [1-t4C]glycollate with spinach-beet peroxisomes. The usual increase in the rate of CO/ release to a maximum after 90 min incubation was observed. However, glycollate appeared to be oxidised at close to its maximum rate from zero time. The activity of glycollate oxidase fell slowly with time to about 85% of its initial activity after 3 h. By comparison, catalase activity fell rapidly to about 80% of its initial activity after 1 h, but thereafter more slowly. The rapid fall in catalase activity coincided with the increase in 14COz release from [1-14C]glycol-
B. Grodzinskiand V.S. Butt: H202 Productionand CO2 Release
Distribution of Glycollate Oxidase and Catalase in Subcellular Fractions
A 6
4
-~ 2
B v
~8 4 n
I
I
I
~ 100~~~~~%~~ ;~-,~
80
g 2 0
229
1
2 Time(h)
5
4
Fig. 4. T i m e course of l~COz release f r o m [1-14C]gtycollate and
[1-14C]glyoxylate, and of production available peroxide_ Spinachbeet peroxisomes(A, 91 lag.; B, 76 lag. protein) were incubated with (A) [1-14C]glycollate (5raM), (B) [1-14C]glyoxylate (5 mM; o--o), and sodium glycollate(5 raM) with [1-~4C]glyoxylate (0.5 raM; o--o), under the conditionsin Fig. 2. Samples (200 to 410 gg protein) were extracted at intervals from incubationvessels containingglycollate(5 mM) for assay of production of available peroxide (A--A glycollateoxidase activity(a--zx), and cata lase activity in the presence of 5 mM PHMS (n--D), using the 02 electrode
late during the lag period and was also reflected in an increased rate of production of "available" H202, which reached a maximum after 2 h (Fig. 4c). Over the period of maximum H202 production, 1~CO2 release was most rapid and declined as the H202 production declined after 3 h. When trace quantities of [1- l 4C]glyoxylate equivalent to 5% of unlabelled glycollate were included in the reaction mixture, the time course of 14CO2 release was similar to that observed when [1-14C]glycollate was supplied. It is unlikely that the lag response observed could be due to the accumulation of glyoxylate to some necessary level before decarboxylation became maximal, nor that the catalase was undergoing a simple thermal destruction, since there was a linear release of 14CO2 when the same quantity of [1-~4C]glyoxylate was supplied in the absence of any glycollate (Fig. 4b).
The data presented substantiate the view that the 14CO2 released from [1-14C]glycollate by leaf peroxisomes is due to the attack on glyoxylate by H202 escaped from catalase action. That this was not because catalase was lost preferentially during the preparation of peroxisomes is shown in Table 4, in which the proportions of catalase and glycollate oxidase were very similar in each fraction. The recovery of both enzymes in the peroxisomes was low, but in agreement with the data of Lips (1975), the proportion of catalase recovered in this pellet was rather higher than that of glycollate oxidase. The data presented in Table 4 allow the calculation of the rates of CO2 release from glycollate by the homogenate. From the initial rate of production of "available" H202, about 10% of the H202 produced in glycollate oxidation was not destroyed (Fig. 2), giving a maximum rate of 1 gmol of CO2 released/h per mg of protein, or, with a 4% recovery of glycollate oxidase, a total rate of 10 ~tmol of CO2 released/h per mg of chlorophyll in the homogenate. The maximum rate, observed over the 90 rain to 180 min period of incubation, corresponded to 18 gmol of CO2 released/h per mg of chlorophyll in the homogenate. Discussion
Glycollate oxidase catalyses the oxidation of glycollate to glyoxylate by molecular oxygen with the formation of H202 (Reaction 1), which non-enzymically oxidises the glyoxylate further to formate and CO2 (Reaction 2), the COz being derived uniquely from the carboxyl group (Zelitch and Ochoa, 1953): C H 2 O H "C O O H + 02 - * C H O . C O O H + H 2 0 2
(1)
C H O - C O O H + H202--* H- C O O H + CO2.
(2)
One mole of oxygen is consumed for each mole of glycollate oxidised. The enzyme is located mainly, and perhaps entirely, in the leaf peroxisomes (Tolbert et al., 1969), in which catalase is also concentrated. Catalase competes with Reaction 2 for H202, and both inhibits CO2 release and reduces the net 02 uptake: H202 - + H 2 0 -/- 1 0 2 .
(.3)
The balance between Reactions 2 and 3 has been measured here by determining the changes in 02 uptake when catalase activity was completely inhibited by 1 mM azide or 20 m M aminotriazole and when excess bovine liver catalase was added. The reduction in the rate of 02 uptake when excess cata-
230
B. Grodzinski and V.S. Butt: H202 Production and C02 Release
Table 4. Distribution of glycollate oxidase and catalase in subcellular fractions of spinach-beet leaves. Spinach-beet leaves (about (150 g) were homogenised and the extract (containing 195 mg chlorophyll) fractionated as described in the Methods section. Glycollate oxidase and catalase activities were determined using the O2 electrode Fraction
(a) Differential centrifugation Crude homogenate 500 x g precipitate 6,00 x g supernatant 6,000 x g precipitate (b) Centrifugation of 6,000 • precipitate through 1.8 M sucrose Peroxisomal pellet Supernatant
Glycollate oxidase
Catalase
Total activity (units)
Recovery (%)
Total activity (units x 10- 3)
Recovery (%)
234 9 194 37
100 4 82 16
128 5 92 25
100 4 72 20
10 23
4 10
11 13
9 10
lase was added is a measure of the H202 available for Reaction 2 and not broken down in Reaction 3. With the purified enzyme, the addition of catalase exactly halved the rate of 02 uptake; this was restored by adding azide or aminotriazole. On the other hand, the rate of 02 uptake with leaf peroxisomes was decreased by little more than 10% when catalase was added, but raised to twice this value when azide or aminotriazole were added. It is clear therefore that as much as 90% of the H202 produced in Reaction 1 by the action of peroxisomal glycollate oxidase is destroyed by peroxisomal catalase or other reactions. It has been claimed that catalase activity in the peroxisomes is sufficient to destroy completely the H202 produced in Reaction 1 (Kisaki and Tolbert, 1969), but the concentrations of glycollate and glyoxylate used may have been too low to detect the release of 14CO 2 from Reaction 2. In rat liver preparations it has been estimated that 40% to 80% of the H202 produced by peroxisomes is destroyed by their catalase (Boveris et al., 1972). Further, the leaf catalase activity may be overestimated since at pH 8.3, it is about one-third of its maximum at pH 7.0, while the enzyme may be relatively ineffective with low concentrations of H202, if its Km is above 20 mM as reported for bovine catalase (von Euler and Josephson, 1927). Peroxisomal catalase activity at pH 8.3 may therefore be sufficient to hold H202 at non-toxic levels but not to destroy it completely. Reaction 2 proceeds with this residual H202, which amounts to at least 10% of that generated in Reaction 1, and it should be noted that the addition of a very large excess of bovine liver catalase failed to suppress Reaction 2 completely. This reaction is therefore likely to proceed during glycollate oxidation in leaves and hence to contribute to the CO2 released in photorespiration.
Any relation between the rates of CO2 release in photorespiration and from glycollate oxidation in peroxisomes is difficult to establish because the conditions under which the peroxisomal oxidations take place in vivo are not known. Thus, the rates of glycollate oxidation and CO2 production from peroxisomes have usually been measured in air but, when 02 is rapidly evolved during photosynthesis in leaves, a more exact estimate of these reactions might be obtained under pure 02. These rates were found to be appreciably higher, as would be expected since the Km for 02 observed with purified glycollate oxidase was 0.17 mM (cf. 0.13 mM for pea-leaf glycollate oxidase, Kerr and Groves, 1975) and the enzyme has been reported to be saturated in atmospheres containing 60% 02 and above (see Andrews et al., 1973). Again, the time course of CO2 release from peroxisomes is not linear apparently because of the inactivation of peroxisomal catalase during the reaction, perhaps due to the high steady-rate of H202 production (Chance, 1950). Estimates of catalase activity in leaf extracts may therefore be high and of CO2 release from glycollate in peroxisomes correspondingly low; some measurement of catalase activity in vivo is necessary to obtain relevant estimates for CO2 release in illuminated leaves from the peroxisomal oxidation of glycollate. With these reservations, some comparison can be made between the rates of CO2 release during the peroxisomal oxidation of glycollate and in the photorespiration of leaves. From his survey of the rates of photorespiration in the leaves of a number of species, Zelitch (1975) concluded that any biochemical mechanism must account for rates of at least 76 gmol of COg released/h per mg of chlorophyll. However, the highest rates observed in the experiments reported here,' carried out at 25 ~ C, are barely one-quarter of
B. Grodzinski and V.S. Butt: H2Oa Production and CO 2 Release
this requirement. The extension of these observations to species with high photorespiratory rates and the determination of the rate of photorespiration of spinach-beet leaves at high light intensities may indicate the extent of the discrepancy between these observations. As well as CO2, both formate and glyoxylate are formed in the reaction mixture. Formate can be oxidised to CO2 in peroxisomes (Leek et al., t972) and other organelles (Halliwell, 1974) but the CO2 here was derived from C-2 of glyoxylate. Halliwell (1973) also showed spinach-beet leaf homogenates to possess sufficient formyl-tetrahydrofolatesynthetase (formate: tetrahydrofolate ligase (ADP), EC 6.3.4.3) activity to convert any formate produced in a first step leading to the hydroxymethylation of glycine to serine. Up to 80% of the glycollate oxidised in peroxisomes can be isolated as glyoxylate, which can be further oxidised to oxalate (Halliwell and Butt, 1974), aminated by serine or glutamate to form glycine (Tolbert, 1971) or oxidised in illuminated chloroplasts to formate and CO2 (Zelitch, 1972). While these further reactions also give rise to CO2, the CO2 released in the peroxisomal reactions reported here must provide some part of the total CO2 release in photorespiration.
References Andrews, T.J., Lorimer, G.H., Toibert, N.E.: Ribulose diphosphate oxygenase. I. Synthesis of phosphoglycolate by fraction-1 protein of leaves. Biochemistry 12, 11-18 (1973) Arnon, D.I. : Copper enzymes in isolated chloroplasts. PoIyphenol oxidase in Beta vulgaris. Plant Physiol. 24, 1-15 (1949) Boveris, A., Oshino, N., Chance, B.: The cellular production of hydrogen peroxide. Biochem. J. 138, 77-85 (1974) Chance, B. : The reactions of catalase in the presence of the notatin system. Biochem. J. 46, 387402 (1950) Elstner, E.F., Heupel, A. : On the decarboxylation of ~-keto acids by isolated chloroplasts. Biochim. biophys. Acta (Amst.) 325, 182-188 (1973) Euler, H. von, Josephson, K.: Uber Katalase. II. Ann. Chem. 455, 1 16 (1927)
231 Halliwell, B. : The role of formate in photorespiration. Trans. Biochem. Soc. 1, 1147-1150 (1973) Halliwell, B. : Oxidation of formate by peroxisomes and mitochondria from spinach leaves. Biochem. J. 138, 77-85 (1974) Halliwell, B., Butt, V.S.: Oxidative decarboxylation of glycollate and glyoxylate by leaf peroxisomes. Biochem. J. 138, 217~24 (1974) Halliwell, B., Leek, A.E., Butt, V.S.: Oxidative decarboxylation of glycollate and glyoxylate by leaf peroxisomes. Biochem. J. 128, 87p-88p (1972) Kerr, M.W., Groves, D. : Purification and properties of glycollate oxidase from Pisum sativum leaves. Phytochemistry 14, 359-362 (1975) Kisaki, T., Tolbert, N.E.: Glycolate and glyoxylate metabolism by isolated peroxisomes or chloroplasts. PIant Physiol. 44, 242250 (1969) Kisaki, T., Yoshida, N., Imai, A. : Glycine decarboxylase and serine formation in spinach leaf mitochondrial preparation with reference to photorespiration. Plant Cell Physiol. 12, 275-288 (1971) Leek, A.E., Halliwell, B., Butt, V.S.: Oxidation of formate and oxalate in peroxisomal prearations from leaves of spinach beet (Beta vulgaris L,). Biochim. biophys. Acta (Amst.) 286, 299-311 (1972) Leggett Bailey, J. : Techniques in protein chemistry, p. 293. Amsterdam: Elsevier 1962 Lips, S.H. : Enzyme content of plant microbodies as affected by experimental procedures. Plant Physiol. 55, 5984501 (1975) Ltick, H. : Catalase. In : Methods of Enzymatic Analysis. pp. 886888, Ed.: Bergmeyer, H.U., New York: Academic Press 1963 Tolbert, N.E. : Microbodies-peroxisomes and glyoxysomes. Ann. Rev. Plant Physiol. 22, 45-74 (1971) Tolbert, N.E., Oeser, A., Yamazaki, R.K., Hageman, R.H., Kisaki, T. : A survey of plants for leaf peroxisomes. Plant Physiol. 44, 135-147 (1969) Zelitch, I.: Increase rate of net photosynthetic carbon dioxide uptake caused by the inhibition of glycolate oxidase. Plant Physiol. 41, 1623-1631 (1966) Zelitch, I.: Photosynthesis, photorespiration and plant productivity. New York-London: Academic Press 1971 Zelitch, I.: The photooxidation of glyoxylate by envelope-free spinach chloroplasts and its relation to photorespiration. Arch. Biochem. 150, 698-707 (1972) Zelitch, I.: Improving the efficiency of photosynthesis. Science 188, 6264533 (1975) Zelitch, I., Ochoa, S.: Oxidation and reduction of glycolic and glyoxylic acids in plants. I. Glycolic acid oxidase. J. biol. Chem. 201, 707 718 (1953)
Received 22 September; accepted 14 October 1975
