Enzyme Distribution in Potato Mitochondria

Journal of Experimental Botany, Vol. 30, No. 116, pp. 539-549, June 1979

Enzyme Distribution in Potato Mitochondria DAVID A. DAY1, GEOFFREY P. ARRON2, AND GEORGE G. LATIES

Received 12 July 1978

ABSTRACT Enzyme distribution in potato mitochondria was investigated by selectively disrupting the outer and inner membranes with digitonin. Antimycin-insensitive NADH-cytochrome c reductase, an outer membrane marker, was released at low digitonin concentrations (0-1 mg mg~' mitochondrial protein). Soluble matrix enzymes, fumarase and malate dehydrogenase were released at 0-3-0-4 mg digitonin mg~' protein, as the inner membrane ruptured. Very little (about 10%) cytochrome oxidase activity was released, even at higher digitonin concentrations, in accord with this enzyme being an integral inner membrane protein. By this criterion adenylate kinase is also firmly bound to the inner membrane. Evidence indicates that it faces the intermembrane space. Malic enzyme activity was released by the same digitonin concentration that released fumarase and malate dehydrogenase, indicating that malic enzyme is a soluble matrix enzyme. No activity was released at low digitonin concentrations which selectively break the outer membrane, showing that malic enzyme is not present in the intermembrane space. Considerable catalase activity (20—40 fimo\ O2 min~' mg"1 protein) was associated with washed mitochondrial preparations, but 95% of this was lost upon purification of mitochondria. The remaining activity was firmly bound to the mitochondrial membranes. INTRODUCTION

The distribution of enzymes in animal mitochondria is well documented (Parsons, Williams, Thompson, Wilson, and Chance, 1967; Sottocasa, Kuylenstierna, Ernster, and Bergstrand, 1967; Schnaitman and Greenawalt, 1968). Although plant mitochondrial membranes have been separated (Moreau and Lance, 1972; Douce, Mannella, and Bonner, 1973; Day and Wiskich, 1975), detailed investigations of enzyme distribution in plant mitochondria have not been made. Thus, while the localization of enzymes of the TCA cycle and the mitochondrial NADH dehydrogenases are well established (Palmer, 1976), controversy surrounds the location of enzymes such as catalase, malic enzyme, and adenylate kinase. Despite much research and discussion (see Palmer, 1976, and Wiskich, 1977, 'Present address: Division of Plant Industry, CSIRO, P.O. Box 1600, Canberra City, A.C.T. 2601, Australia. 2 Present address: Department of Horticulture, University of Wisconsin, Madison, Wisconsin 53706, U.S.A.

Downloaded from http://jxb.oxfordjournals.org/ at Flinders University of South Australia on July 4, 2013

Department of Biology and Molecular Biology Institute, University of California, Los Angeles, CA 90024

MATERIALS AND METHODS Isolation of mitochondria Mitochondria were isolated as previously described (Laties, 1973), and finally suspended in a medium containing 0-4 M mannitol, 25 mM TES buffer (pH 7-4), and 0-1% BSA. Purification was by centrifugation on linear (20-60%, w/v) sucrose gradients, for 3 h at 25 000 rev. min"1 (82 000 g) in a Beckman L5-5O ultracentrifuge (SW 27-1 rotor). The single band of mitochondria (at density 1 • 186) was removed with a large-bore syringe, carefully diluted with 10 mM TES + 0-1% BSA (pH 7-2) over a period of 30 min to give a final sucrose concentration of 0-4 M, and centrifuged for 15 min at 10 000 g in a Sorvall RC-2 centrifuge. Final resuspension was in 0-4 M mannitol + 0-1% BSA. Oxygen consumpton Oxygen uptake was measured in 2 ml standard reaction medium (0-4 M mannitol, 10 mM TES buffer, 5 mM KH 2 PO 4 , 5 mM MgCl2, and 0-1% BSA, pH 7-2) in a Rank oxygen electrode at 25 °C. Approximately 1 mg mitochondrial protein was used per assay. Release of enzymes from mitochondria Mitochondrial membranes were dispersed with digitonin as described by Day and Wiskich (1975). After protein content had been determined, potato mitochondria were divided into aliquots and incubated with appropriate volumes of digitonin solution (prepared using recrystallized digitonin as described by Day and Wiskich, 1974a). Incubation was for 30 min at 0 °C, following which the suspensions were centrifuged for 30 min at 30 000 g in a Beckman L5-5O ultracentrifuge. This speed and time have previously been shown to pellet large membrane entities (i.e. unbroken mitochondria, mitoplasts, and inner membrane fragments) but not outer membrane vesicles (Day and Wiskich, 1975). The 30 000 g supernatant was assayed for the various enzyme activities. It must be stressed that, whereas at low concentrations (below 0-3 mg digitonin g~' mitochondrial protein) digitonin serves selectively to disrupt the outer mitochondrial membrane, disruption need


540 Day, A iron, and Laties—Mitochondrial Enzyme Distribution for reviews), it has yet to be demonstrated conclusively whether malic enzyme exists in the intermembrane space, as well as in the mitochondrial matrix. Knowledge of the locus of malic enzyme is essential to the interpretation of exogenous NAD effects on plant mitochondria (Day and Wiskich, 1974 a, b, 1978; Coleman and Palmer, 1972; Palmer and Arron, 1976), and to the interpretation of transport phenomena (Day and Hanson, 1977). Catalase has been considered a non-mitochondrial enzyme, largely on the basis of studies involving density gradient centrifugation of whole tissue homogenates, and cytochemical microscopy (Tolbert, 1971; Huang and Beevers, 1971). Investigations with isolated mitochondria, however, have yielded conflicting data (Plesnicar, Bonner, and Storey, 1967; Rich, Boveris, Bonner, and Moore, 1976). Clear-cut evidence for the intramitochondrial localization of adenylate kinase in plants has not been provided either, although it is generally assumed to be restricted to the intermembrane space on the basis of animal studies (Ernster and Kuylenstierna, 1969). The present paper reports on attempts to directly determine the intramitochondrial location of the enzymes discussed above, by selectively disrupting the mitochondrial membranes and observing which enzymes are thereby released to the medium. The results indicate that malic enzyme is confined to the matrix, whereas adenylate kinase is tightly bound to the outer face of the inner membrane. Although much of the catalase activity present in washed mitochondrial preparations can be removed by purification, a significant quantity remains bound to the membranes.

Day, Arron, andLaties—Mitochondrial Enzyme Distribution

541

not be complete. Thus, when care is taken to confine disruption to the outer membrane, a considerable fraction of the mitochondrial population may be unscathed or caused to aggregate. Under the circumstances, retrieval in the supernatant solution of an enzyme associated with the outer membrane or situated in the intermembrane space will be incomplete. The same can be said of soluble matrix enzymes at higher digitonin concentrations where inner membranes may not be totally disrupted. The persistance of unbroken mitochondria or mitoplasts will lead to less than full recovery in the supernatant solution. Thus the matter of importance in what follows is not the complete recovery of one or another enzyme following digitonin treatment, but rather the demonstration that the release of certain enzymes is linked to characteristic digitonin concentrations.

Estimation of recovery The activity of each enzyme estimated in the supernatant solution following digitonin treatment and centrifugation was compared with the total mitochondrial activity of that enzyme. To this end, intact mitochondria were disrupted with 0-02% Triton X-100 detergent. The activity of the Triton extract was taken as 100%, and the supernatant activity was expressed as a fraction thereof (see Fig. 1). Protein determination Protein was estimated by the method of Lowry, Rosebrough, Farr, and Randall (1951) using bovine serum albumin (fraction V) as the standard. Materials Untreated potato tubers (Solanum tuberosum Var. Burbank Russet) were kindly supplied by H. Timm, University of California, Davis. Chemicals were obtained from Sigma Chemical Co. or from Calbiochem.

RESULTS AND

DISCUSSION

Release of known marker enzymes Figure 1 shows the release of known marker enzymes by increasing amounts of digitonin. At low concentrations digitonin selectively ruptures and fragments the outer mitochondrial membrane (Schnaitman and Greenawalt, 1968; Day and Wiskich, 1975). Hence antimycin-insensitive NADH-cytochrome c reductase, a


Enzyme activities Malate dehydrogenase was assayed spectrophotometrically by following the oxidation of NADH at 340 nm in the presence of oxaloacetate (Ochoa, 1955). Malic enzyme activity was measured by the method of Hatch and Kagawa (1974) by following the reduction of NAD in a medium containing 10 mM TES buffer, 5 mM malate, 2 mM NAD, 5 mM DTE, 1 ^ig ml"' antimycin A (to prevent NADH oxidation by the respiratory chain), 75 fiM CoA (an activator of malic enzyme), and 0-2 mM EDTA (to chelate endogenous Mg 2+ ). Under these conditions no NAD reduction is observed until 5 mM MgCl2 is added to start the reaction. NAD reduction rates were linear for at least 10 min. Sucrose (0-4 M) was added when necessary to provide isotonicity. Catalase was assayed polarographically with a Rank oxygen electrode, in 2 ml 10 mM TES buffer (pH 7-2) + 0-01 M H 2 O 2 . The reaction was initiated by adding enzyme, and initial rates of oxygen evolution recorded. Sucrose (0-4 M) was added when isotonicity was desired. Cytochrome oxidase was measured polarographically in 2 ml standard reaction medium containing 0-05 mM cytochrome c, 10 mM ascorbate, and 5 mM TMPD. NADH-cytochrome c reductase was measured spectrophotometrically by following the reduction of cytochrome c at 550 nm. The reaction mixture consisted of 0-05 mM cytochrome c and 1 mM KCN in 3 ml standard reaction medium. The reaction was begun by the addition of 1 mM NADH to the cuvette. Fumarase was assayed spectrophotometrically (Huang and Beevers, 1971) by following the formation of fumarate at 240 nm, in 3 ml 10 mM TES buffer (pH 7-0) plus 10 mM malate. Adenylate kinase was measured spectrophotometrically as described by Sottocasa et al. (1967), by linking it to NADPH formation via glucose-6-phosphate dehydrogenase and hexokinase.

542

Day, Arron, andLaties—Mitochondria! Enzyme Distribution

100 -

mitochondnal protein)

Fio. 1. Release of matrix and membrane-bound enzymes of potato mitochondria. Enzymes were solubilized and assayed as described in Materials and Methods. Activities are expressed as percentages of those in solubilized unfractionated mitochondria. Triton X-100 (0-02%) was used to solubilize control mitochondria. A A, antimycin A-insensitive NADH-cytochrome c reductase; O O, fumarase; # # , malate dehydrogenase; D Q, cytochrome oxidase. Total rates of Triton-treated mitochondria: antimycin A-insensitive NADH cyt c reductase, 25 nmol cyt dehydrogenase, 5200 nmol NADH formed min~ mg" 1 protein; cytochrome oxidase, 1-5 iimo\ O2 mur

marker enzyme for the outer membrane (Schnaitman and Greenawalt, 1968; Douce et al., 1973; Day and Wiskich, 1975), is released from the mitochondria by low (0-1-0-2 mg mg" 1 protein) digitonin concentrations. At higher digitonin concentrations (0-3-0-5 mg mg"1 protein) some fraction of the inner membranes break, and the soluble matrix enzymes fumarase and malate dehydrogenase are released. Integral inner membrane proteins, such as cytochrome oxidase, are not released to any great extent, even at higher digitonin concentrations (Fig. 1). The results shown here for potato mitochondria are very similar to those obtained with turnip mitochondria (Day and Wiskich, 1975). It was suggested earlier that incomplete membrane disruption by digitonin may well account for the partial recovery in the supernatant solution of enzymes of the other membrane, as well as soluble matrix enzymes. Unpublished work by Arron and Palmer with Jerusalem artichoke mitochondria in which selective membrane disruption was achieved by graduated osmotic swelling led to the same conclusion. A large fraction (30-50%) of NADH-cytochrome c reductase and malate dehydrogenase, unaccounted for in the supernatant solution following osmotic disruption and centrifugation, was found to be in the 'agglomerated mitochondria,' a fraction sedimented at low centrifugal forces. Malic enzyme Malic enzyme activity appears in the supernatant at the same concentration of digitonin (0-3 mg mg" 1 protein) that released fumarase and malate dehydrogenase


05 Digitonin (mgmg

Day, Arron, and Laties—Mitochondrial Enzyme Distribution 543 (Fig. 2), indicating that malic enzyme is a soluble matrix enzyme. No activity is released at low digitonin concentrations which break the outer membrane, showing that malic enzyme is not present in the intermembrane space. Whereas neither malate dehydrogenase nor malic enzyme is fully released to the supernatant fraction at high digitonin levels, the preciseness of release at a threshold level of digitonin (0-3 mg mg" 1 protein), together with the asymptotic character of release, points to a true matrix enzyme. The difference in percent release to the supernatant solution of each of the enzymes under examination raises the possibility of a variable degree of adherence of soluble enzymes to insoluble mitochondrial components, in situ, or during mitochondrial disruption (Figs 1 and 2). Incomplete

FIG. 2. Solubilization of malic enzyme and adenylate kinase. Experimental details are given in Fig. 1. • 9 . malic enzyme; O Q, adenylate kinase; , malate dehydrogenase; • • • •, NADH-cytochrome c reductase. Total rates of Triton-treated unfractionated mitochondria: malic enzyme, 128 nmol NADH formed min"1 mg" 1 protein; adenylate kinase, 80 nmol NADPH formed min"1 mg" 1 protein. Malate dehydrogenase and antimycin A-insensitive NADH-cyt c reductase: see Fig. 1.

disruption or adherence is indicated rather than destruction, since the sum of the released activity and the activity remaining in the pellet normally accounts for the full mitochondrial complement of the enzyme in question. Adherence does not necessarily bespeak an intrinsic inner membrane protein. Nonetheless, we cannot rule out on the basis of the data presented the possibility that some malate dehydrogenase and malic enzyme is membrane-bound in situ. In any event, whatever malic enzyme is bound must face the matrix space, since very little activity is detected in intact mitochondria (Table 1). Addition of Triton to break the inner membrane (the outer membrane is permeable to added NAD) and allow NAD free access to the enzyme, stimulates the activity approximately 16-fold (Table 1). It seems therefore that all malate oxidation by isolated potato mitochondria occurs in the matrix. Yet intact potato mitochondria respond to NAD and rotenone in much the same manner as cauliflower (Day and Wiskich, 1974a) and artichoke (Coleman and Palmer, 1972) mitochondria. That is, addition of NAD


OS Digitonin (mg mg"' mitochondrial protein)

544

Day, A iron, and Laties—Mitochondrial Enzyme Distribution T A B L E 1. Effect of mitochondrial disruption on enzyme activities Enzymes were assayed as described in Materials and Methods. Sucrose (0-4 M) was present to prevent swelling. 0-02% Triton X-100 detergent was used as indicated. Sonication was for 1 min at 40 W. Adenylate kinase and malic enzyme activities are expressed as nmol NADPH and NADH formed respectively min" 1 mg" 1 protein. Catalase activity is expressed as /rniol O 2 min"1 mg" 1 protein. Treatment

Enzyme activities Adenylate kinase

75 72 85

8-5 127-5

Catalase Washed

Purified

111

0-41 0-67

150

relieves rotenone-inhibited oxygen consumption with malate as substrate (Fig. 3). Two hypotheses have been proposed to explain this phenomenon. Palmer and colleagues (Palmer, 1976) suggest that malic enzyme is localized in the intermembrane space, whereas Day and Wiskich (1974a, b) propose a transmembrane transfer of hydrogen from malate-linked internal NADH to external NAD, probably by means of a transhydrogenase. Such a transhydrogenase may be linked either to malate dehydrogenase or malic enzyme in the matrix, but in either case its presence would result in formation of external NADH and the relief of rotenone inhibition by the oxidation of external NADH via the externally facing dehydrogenase (see Palmer, 1976). Since, in potato mitochondria at least, very little external malic enzyme can be detected, and this is not nearly enough to account for NAD-stimulated increase in malate oxidation, Fig. 3 is interpreted to provide evidence for the operation of a transmembrane transhydrogenase, here probably linked to matrix malate dehydrogenase, since glutamate is present. It is unlikely that NAD itself crosses the inner membrane since little external NAD reduction is observed prior to disruption of the membrane by Triton (Table 1). The measured activity of malate dehydrogenase in the presence of antimycin in intact washed potato mitochondria is virtually nil—thus ruling out contaminating malate dehydrogenase as an explanation of the release of rotenone inhibition by exogenous NAD. Adenylate kinase The release of adenylate kinase shows a similar profile to that of cytochrome oxidase, only 10% of its activity being released by 0-5 mg digitonin mg"1 protein (Fig. 2). To ensure that enzyme activity has not been destroyed by the digitonin treatment, the 30 000 g pellets (see Materials and Methods) were also assayed (Table 2). The results show that only a small amount of the total adenylate kinase activity is released, the remaining activity being bound to the membranes. Soluble fumarase, on the other hand, is completely released by 0-5 mg digitonin mg"1 protein (Table 2). Yet, when adenylate kinase activity is measured in intact mitochondria, its activity is not affected by addition of Triton, and only slightly stimu-


Control Triton X-100 Sonication

Malic enzyme

Day, Arron, and Laties—Mitochondrial Enzyme Distribution

545

T A B L E 2. Enzyme distribution in digitonin fractions ofpotato mitochondria Purified mitochondria were incubated either with standard reaction medium (control), or reaction medium plus 0-5 mg digitonin mg~' protein, for 30 min at 0 °C, and then centrifuged for 30 min at 30 000 g. Ten milligrams mitochondrial protein was used as starting material for the control and digitonin-treated fractionations respectively. Approximately 1-67 mg protein was used in each case in estimates of catalase distribution. The pellets were resuspended in 0-4 M sucrose and solubilized with 0-02% Triton. Enzyme assays were performed as described in Materials and Methods. Adenylate kinase activity is expressed as nmol NADPH formed min"1, fumarase as absorbance units x 10 min"1, and catalase as //mol O 2 min"1. Activities represent the entire fractions. Adenylate kinase Fumarase

Catalase

Control pellet Control supernatant Digitonin pellet Digitonin supernatant

700 25-5 615 134

0-52 0-19 0-28 0-44

10-6 0 0 12-8

lated by sonication (Table 1), showing that there is no permeability barrier to the reagents used in the assay, including NADP. We interpret these results to mean that adenylate kinase is tightly bound to the inner membrane of potato mitochondria, facing the intermembrane space. This is in contrast to the situation in liver (Sottocasa et al., 1967; Schnaitman and Greenawalt, 1967) and yeast (Bandlow, 1972; Velours, Guerin and Duvert, 1977) mitochondria, where adenylate kinase is a soluble intermembrane-space enzyme. Catalase Although only a fraction (12%) of the total catalase activity of potato tubers is found in the washed mitochondrial pellet, washed mitochondrial preparations have a relatively high specific activity (compare, for example, their respiratory activity, Fig. 3, with catalase activity, Tables 1 and 3). More than 95% of the catalase associated with washed mitochondria can be removed by density gradient purification (Table 3 and Fig. 4). Fractionation of the gradient shows that most of the activity in washed preparations is due to peroxisomal contamination (Fig. 4). Although peroxisomes contribute only slightly to the protein content of washed preparations (data not shown), the high turnover rate of catalase results in high specific activities. Rich et al. (1976) reported that the catalase activity associated with mung bean mitochondria is stimulated by a process which disrupts the inner membrane, suggesting that catalase is an intramitochondrial enzyme. We also observe this stimulation, both in washed and purified potato mitochondria (Table 1). However, microbody catalase activity is also stimulated by detergent treatment (Bieglmayer, Nahler and Ruis, 1974) and it is probable that much of the stimulated catalase activity in washed mitochondrial preparations results from peroxisomal disruption (see Fig. 4). Stimulation of catalase activity upon disruption of purified mitochondria is not as readily accounted for. It may signify removal of a permeability barrier to intra-


Fraction

546

Day, A iron, and Lades—Mitochondrial Enzyme Distribution

mitochondria! catalase. Alternatively, detergent may release an activator of catalase from the mitochondrial matrix, as suggested for chloroplasts (Allen, 1978), or simply release the enzyme from the membrane. In any case, stimulation of activity upon disruption of mitochondria cannot alone be considered conclusive evidence that catalase is a bona fide mitochondrial enzyme. TABLE

3. Effect of successive washes on the specific activity of mitochondrial

enzymes Mitochondria were prepared (Mw) and purified (Mp) as described in Materials and Methods. After purification, the mitochondria were successively washed in 25 ml of a medium containing 0-4 M mannitol, 50 mM TES buffer (pH 7-2), and 0-1% BSA. Enzyme activities were measured after each wash. Catalase activity is expressed as //mol O 2 min"1 mg" 1 protein, and NADHcytochrome c reductase as nmol cyt c reduced min~' mg" 1 protein. The mitochondria were disrupted with 0-02% Triton X-100 prior to the catalase measurements. Mitochondrial preparation

Catalase

Antimycin A-insensitive NADHcytochrome c reductase

Mw M n

36-72 108 0-73 0-68 0-64

25-0 27-2 27-6 30-5 33-3

washed lx washed 2 x M P washed 3 x M

P

M

P


FIG. 3. Effect of rotenone and NAD on malate oxidation by potato mitochondria. Oxygen uptake was measured as described in Materials and Methods. Glutamate was included in the reaction medium (final concentration = 10 mM), and 10 mM malate was substrate. ADP, 15 //M rotenone, and 1 mM NAD were added as indicated. Rates are expressed as nmol O 2 min"1 mg" 1 protein.

Day, Arron, andLaties—Mitochondrial Enzyme Distribution

25

30

35

Fraction number FIG. 4. Catalase and cytochiome oxidase profiles from density gradient purification of washed potato mitochondria. Mitochondria were purified, and enzymes assayed, as described in Materials and Methods. Fractionation of the gradients was with a Gilson micro-fractionator (1-25 ml per fraction). • # , cytochrome oxidase; O O, catalase.

About 30% of the catalase activity remaining after purification can be removed by simply washing the mitochondria repeatedly. The easily removed catalase presumably represents contaminating soluble enzyme adhered to the outer membrane, and accounts for a high zero digitonin value in the dispersion profile of catalase (Fig. 5). That is, the incubation of purified mitochondria for 30 min in the

00

0 1

0-2

03

0-4

0-5

Digitonin (mg mg~' mitochondrial protein) Fio. 5. Solubilization of catalase in purified potato mitochondria. Assay conditions are described in Fig. 1, except that purified mitochondria were used. The broken line represents the percent of total activity released, calculated after the zero value has been subtracted from the other values (see text for further explanation).


20

15

547

CONCLUSIONS

We conclude that (1) malic enzyme is localized in the mitochondrial matrix but is absent from the intermembrane space, (2) adenylate kinase is bound to the inner membrane, facing the intermembrane space, and (3) a small but significant quantity of catalase is firmly attached to the mitochondria and may be an intrinsic mitochondrial enzyme. ACKNOWLEDGEMENTS

This work was supported by grants from the U.S. Public Health Service and the U.S. Department of Energy to G. G. Laties.


548 Day, Arron, and Laties—Mitochondrial Enzyme Distribution absence of digitonin (the control for the digitonin series) has the same effect as repeated washing, removing some 30% of the residual catalase activity. When this activity is subtracted from the total activity released by digitonin at each digitonin concentration, the dispersion profile for catalase suggests that a significant fraction thereof is located on the outer membrane (Fig. 5, broken line). Thus the outer membrane of isolated mitochondria seemingly possesses catalase both as a contaminant and as a bona fide constituent. With prolonged washing (Table 3) the removal of the last bit of contaminating catalase from the outer membrane together with some extraneous protein may be the reason that the catalase-specific activity appears to stay steady, while that of another enzyme associated with the external membrane, antimycin A insensitive NADH-cytochrome c, rises slightly. In any event the fact that catalase activity is recovered in the 30 000 g pellet at high (0-5 mg mg~' protein) digitonin concentrations (Table 2) suggests that some catalase is bound to the inner membrane as well. It is not likely that mitochondrial catalase originates in the matrix and subsequently adheres to one or both of the mitochondrial enzymes since release is monotonic with digitonin concentration and there is no threshold at 0-3 mg digitonin mg"1 protein (Fig. 5, cf. Fig. 1). Altogether, mitochondrial catalase comprises about 0-36%, and mitoplast catalase 0-2%, of the total cell catalase. Although the mitochondrial catalase activity is very low compared with that of the peroxisomes (Fig. 4), it is significant when compared with respiratory rates of isolated mitochondria, and might be thought to have important implications for cyanide-insensitive respiration. For example, it has been proposed (Rich et al., 1976) that the alternative oxidase of higher plant mitochondria produces H 2 O 2 . Were that so, in the presence of significant quantities of catalase in plant mitochondria (Rich et al., 1976; present communication) addition of cyanide to antimycin-inhibited mitochondria should stimulate O2 consumption by preventing H 2 O 2 breakdown by catalase. Since such stimulation does not occur, it is unlikely that H2O2 is the end product of the alternative oxidase (see also Huq and Palmer, 1978). Nonetheless, some superoxide anion is formed by side reactions of the respiratory chain (Boveris, Oshino and Chance, 1972), and superoxide dismutase (which catalyses the formation of peroxide from superoxide) is present in plant mitochondria (Arron, Henry, Palmer and Hall, 1976). It is therefore reasonable that some catalase is also present to provide membranes protection from peroxide.

Day, Arron, and Laties—Mitochondrial Enzyme Distribution

549

LITERATURE CITED ALLEN, J.F., 1978. FEBS Lett. 84,221-4. ARRON, G. P., HENRY, L., PALMER, J. M., and HALL, D. O., 1976. Biochem. Soc. Trans. 4, 618—

20. BANDLOW, W., 1972. Biochem. biophys. Acta, 282,105-22. BEEGLMAYER, C , NAHLER, G., AND RUIS, H., 1974. Z. Physiol. Chem. 355,1121-8. BOVERIS, A., OSHINO, N., and CHANCE, B., 1972. Biochem. J. 128,617-30. COLEMAN, J. O. D., and PALMER, J. M., 1972. Eur. J. Biochem. 26,499-509.

DOUCE, R., MANNELLA, C. A., and BONNER, W. D., JR., 1973. Ibid. 292,105-16.

ERNSTER, L., and KUYLENSTIERNA, B., 1969. FEBS Symp. 17,5-31. HATCH, M. D., and KAGAWA, T., 1974. Aust. J. PI. Physiol. 1,357-69. HUANG, A. H. C , and BEEVERS, H., 1971. PL Physiol., Lancaster, 48,637-41. HUQ, S., and PALMER, J. M., 1978. PL Sci. Lett, 11,351-8. LATIES, G. G., 1973. Biochemistry, 12,3350-5. LOWRY, O. H., ROSEBROUGH, N. J., FARR, A. L., and RANDALL, R. J., 1951. J. biol. Chem. 193,

265-75. MOREAU, F., and LANCE, C , 1972. Biochimie, 54,1335-48.

OCHOA, S., 1955. In Methods in enzymology. Ed. S. P. Colowick and N. O. Kaplan. Vol. 1, Pp. 735-9. PALMER, J. M., 1976. A. Rev. PI. Physiol 27,133-57. and ARRON, G. P., 1976. J. exp. Bot. 27,418-30. PARSONS, D. F., WILLIAMS, G. R., THOMPSON, W., WILSON, D., and CHANCE, B., 1967. In Mito-

chondrial structure and compartmentation. Ed. E. Quagliariello, S. Papa, E. C. Slater, and J. M. Tager. Adriatica, Editrice, Ban. Pp. 29-70. PLESNICAR, M., BONNER, W. D., JR., and STOREY, B. T., 1967. PL Physiol., Lancaster, 42,366-70. RICH, P. R., BOVERIS, A., BONNER, W. D. JR., and MOORE, A. L., 1976. Biochem. biophys. Res.

Commun. 71,695-703. SCHNAITMAN, C , and GREENAWALT, J. W., 1968. J. Cell Biol. 38,158-75. SOTTOCASA, G. L., KUYLENSTIERNA, B., ERNSTER, L., and BERGSTRAND, A., 1967. In Methods

in enzymology. Ed. S. P. Colowick and N. O. Kaplan. Vol. 10, Pp. 448-63. TOLBERT, N. E., 1971./1. Rev. PL Physiol. 22,45-74. VELOURS, J., GUERIN, B., and DUVERT, M., 1977. Archs Biochem. Biophys. 182, 295-304.

WISKICH, J. T., 1977. A. Rev. PL Physiol. 28,45-69.


DAY, D. A., and HANSON, J. B., 1977. PL Physiol., Lancaster, 59,630-5. and WISKICH, J. T., 1974a. Ibid. 53,104-9. 1974/). Ibid. 54,360-3. 1975. Archs Biochem. Biophys. 171,117-23. 1978. Biochim. biophys. Acta, 501,396-404.