Developmental Expression and Intracellular Localization of Superoxide Dismutases in Maize

Differentiation

Differentiation 13, 133-140 (1979)

0 Springer-Vrrlq 1979

Developmental Expression and Intracellular Localization of Superoxide Dismutases in Maize* JAMES A. BAUM and JOHN G. SCANDALIOS' Department of Genetics, North Carolina State University, Raleigh, North Carolina 27650, USA

Maize superoxide dismutase (superoxide : superoxide oxidoreductase, EC 1.15.1.1 ; abbreviated as

SOD)has been resolved into five major electrophoretic forms by starch gel electrophoresis. All five isozymes are present in the tissues examined. Superoxide dismutase, catalase, and peroxidase activity was measured in the scutellum during kernel development and germination with no obvious correlations being observed. Superoxide dismutase activity is associated with the cytosolic, mitochondrial, and chloroplastic fractions of maize seedlings. Cyanide-sensitive SOD- 1 is associated with chloroplasts while cyanide-resistant SOD-3 is associated with mitochondria.

Introduction

The superoxide dismutases (SOD)comprise a class of metalloproteins which catalyze an unusual oxidation-reduction reaction in which two superoxide radicals are disproportionated to hydrogen peroxide and oxygen:

07 + 0,+ 2 H+

-+

H,O,

+ 0,.

Since the discovery of their function nearly a decade ago [l], these enzymes have been intensively studied with respect to their physical properties and mechanism of action. Thus, the cytosolic copper and zinc-containing enzyme of eukaryotic cells has been purified from a wide variety of sources including chicken [21, Neurospora crussu [31, wheat g e m [41, and green pea (51. The manganese-containing superoxide dismutases have been isolated from such sources as E. coli [61, Streptococcus

Research supported by National Institutes of Health Grant No. GM 2273362 (to J.G.S.) and by N.I.H. Training Grant No. STOlGM296-18 This is Paper No. 5721 of the Journal Series of the North Carolina Agricultural Experiment Station, Raleigh, North Carolina 1 To whom reprint requests should be addressed

murans [7], yeast 181, and rat [ 9 ] . In addition, an ironcontaining superoxide dismutase has been isolated from several microorganisms including E. coli, Photobucterium [lo], and Chromatiurn vinosum i l l ] . It has been proposed that superoxide dismutase serves a protective role in respiring cells through its elimination of the reactive superoxide radical [12]. In recent years, much evidence has been presented in support of this hypothesis 112-151. Presumably, the hydrogen peroxide generated by the dismutation reaction is subsequently decomposed by catalase or by peroxidases within the cell. Relatively little research has been devoted to studying the genetics of superoxide dismutase in eukaryotic organisms, although such work has been done with man [161. We have initiated a study of superoxide dismutase isozymes in Zeu mays L. for the purpose of characterizing them with respect to their genetic basis, their physical properties, and their physiologic role within the cell. Our work with superoxide dismutase was prompted in part by our interest in its physiologic relationship to catalase, and enzyme which we have been studying intensively in maize. The present investigation deals with the intracellular localization of superoxide dismutases in maize and their expression in various tissues during development. 030 1-468 1/79/OO13/0133/101.60

134

J. A. Baum and J. G. Scandalios: Superoxide Dismutases in Maize Development

Methods Growth of Seedlings

Three inbred lines, W64A. 78, and 229 were used. Sceds were surface sterilized for 10 min in 1% hypochlorite and soaked in distilled deionized water for 24 h. The seeds were then placed in plastic containers on several layers of moistened germination paper. Gennination was canied out at 23O C in an incubator set for a 16 h day. Age of the seedlings was counted from day of imbibition. Preparation of C

d Extracts

For studying the time course of enzyme activity during development, tissues were ground with sand in a pre-chilled mortar with 0.025 M glycylglycine buffer pH 7.4 (10 : I volumdg fresh weight) and the

extracts centrifuged at 14,000 rpm for 20 min at 4' C. The supernaused in in vitro assays.

tant was

Starch Gel Electrophoresis Maize superoxide dismutases were separated by starch gel electrophoresis using a lithium hydroxide-boric acid :tris citrate buffer system 1171. Horizontal gel slices wcre incubated at 37OC in the dark for 20 min in a 0.05 M Tris HCI (pH 8.2) solution containing 0.003% riboflavin, 2 x lo-' M EDTA. and 0.010% nitro-blue-tetrazolium (NBT). The gels were then placed on a light box and illuminated for 20-30 min. The electrophoretic banding is stable and the gds can be stored in 50% glycerol at 4" C. For studying the cxpression of SOD isozymes during devdopment, tissues were homogenized in a minimal amount of 0.025 M glycyiglycine buffer pH 7.4. The homogenate was absorbed onto a 10 x 6 mm picce of Whatman 3M filter paper or Beckman electrophoresis paper and the paper inserted into a starch gel for electrophoresis.

layers of Miracloth and the filtrate (12- 14 ml) layered onto a semil i n w sucrose gradient consisting of 2ml 60%. 6ml 43-60%. 8-10 ml43%. and 8 ml25-43% sucrose. The sucrose gradient was prepared in 0.01 M HEPES buffer pH 7.5 containing 0.1 mM EDTA. The gradient was centrifuged at 4,000 rpm (2.900g maximum)for 5 min followed by l0,oOO rpm (18,000 g maximum) for 10 min in a SW-27 rotor at 4OC using a Beckman LS-65 ultracentrifuge. The brake was applied at 5,oOO rpm during deceleration. Fractions (0.5 ml) were collected by puncturing the tube with a dissecting

needk. In an alternate procedure, 20 g of 10 day old maize leaves were homogenized with 80 ml of odd grinding medium in a small Waring blender (three I-s bursts). The homogenate was passed through three layers of Miacloth and the filtrate centrifuged at 1,500g (maximum) for 5 min. The chloroplastcnriched pelkt was resuspended in 10 ml of 0.01 M HEPES, pH 7.5, containing 25% sucrose and l ( r M EDTA. The resuspension was then layered onto a semi-linear sucrose gradient (same as above) and centrifuged as described above. For routine preparations of crude chloroplasts, chloroplastznriched pellets obtained by differential centrifugation were washed once or twice in 25 ml of cold grinding medium and repelleted by centrifugation at 1,500g (maximum)for 5 min. For isolation of mitochondria and glyoxysomes, 9-log of maize scutella from 4-day dark-grown seedlings were minced in 7 ml of cold grinding medium (same as above) in a petri dish. The homogmate was passed through eight layers of cheesecloth and centrifuged at 3,000 rpm (1,100g maximum) for 15 min. The supernatant was layered onto a continuous 5 6 0 % sucrose gradient and the gradient centrifuged at 25.000 rpm (1 13,000g maximum) in a SW-27 rotor for 4 h at 4O C. After centrifugation, the bottom of the tube was punctured with a dissecting needle and 10 drop fractions were collected. Mitochondria were identified by the presence of cytochrome oxidase 1241. The glyoxysomal peak was identified by using isocitrate lyase as a marker enzyme 1251. Intact chloroplasts were identified by the presence of triose phosphate isomerase activity 1261. Sucrox concentrations were measured by refractometry at 20°C.

Enzyme Assays Superoxide dismutasc was assayed photochemically according to the method of Ravindranath and Fridovich 181. Reduction of NBT was linear up to 7 min and the absorbance at 570 nm in the absence of enzyme increased at a rate of 0.035-0.045 absorbance units per minute. The assay was linear up to 50% inhibition of the NBT photorcduction. One unit of superoxide dismutase activity was ddined as that activity causing 50% inhibition. Catalasc was assayed either polarographicdy 1181 or by the mahod of Beers and Sizer 1191. Paoxidam was assayed using dianisidine as the hydrogen donor 1201. Protein was determined by the method of Lowry et al. 1211. Organelle Isolations and Enzyme Assays

For the isolation of plastids. a procedure similar to that described by M i and k e r s [22] WM used. 6 g of 8-10 day old maize leaves were minced with a razor Made in a grinding medium (3 : 1 volumdwaght) modified from the mcdium uscd by Briedenbach and Bccvers [23]. Sucrose (25%) in HEPES (N-2-hydroxyethylpiperazineN'-2&anesulfonic acid) buffer pH 7.5 was substituted for 0.4 M sucrose in Tris buffer. The homogmate was passed through three

Results

Isozyme Patterns Maize superoxide dismutase has been resolved into five major electrophoretic forms, one of which is resistant to millimolar concentrations of cyanide (Fig. 1). These forms have been labeled SOD-1,SOD-2,SOD-3, SOD4, and SOD-5,in order of their migration toward the anode. Screening of a wide variety of inbred lines has revealed two common electrophoretic patterns; one having four zones of enzymatic activity and one having five zones of enzymatic activity. The four-banded phenotype is by far the most common phenotype among inbred lines of maize. In those lines having SOD-5, the activity of SOD-4is significantly reduced. Our preliminary data suggest that this variation in the expression of SOD4 and SOD-5is under nuclear gene control. Hybrids between AA individuals and BB individuals (Fig. 1) exhibit

135

J. A. Baum and J. G. Scandalios: Supcroxide Dismutases in Maize Development

-k 1 m M K C N

CONTROI.

AA

AB

BA

BB

AA

AB

BA

BB

A five zones of enzymatic activity; approximately 50% of the progeny from backcrosses to the AA parent exhibit the AA (4-banded) phenotype. Selfed hybrids give progeny in which 25% of the individuals have the AA phenotype (data not shown). Thus, the appearance of SOD-5 activity and the concomitant decrease in SOD-4activity follow a pattern of single gene inheritance in these lines. A thorough genetic analysis of this phenotypic variation will be reported in a future paper.

Tissue Distribution of SOD Isozymes

Fig. 2. Superoxide diamutase isozyme patterns from tissues of AA (A) and BB (B) individuals. AA individuals have four zones of enzymatic activity while BB individuals have five zones of enzymatic activity. Sc, = xutellum from Day I seedlings; GLf = green leaf, CSh = cdeoptile sheath; Mes = mesocotyl; Rt = root (all from 8day seedlings); Per = pencarp; LE = liquid endospmn; Sc, = scutellum (aU threc from kernds 20 days post-pollination). All live isozymes are prescnt in the tissues that have been examined. Note that S O D 4 of AA individuals is more active than SOD-4 of BB individu-

als

The expression of superoxide dismutase isozymes in various tissues of maize are shown in Fig. 2A and B. The isozyme patterns observed have been found to be very reproducible. AU four or five isozymes (depending on the inbred line used) are present in the liquid endosperm, scutellum, pericarp, root, primary leaf, coleoptile sheath, and mesocotyl. In addition, all five isozymes are present in the silk, husk, ovule, cob, and pollen of adult plants (gel not shown). The activity of SOD-1appears relatively high in leaf tissue while the activity of SOD-3 appears very high in the scutellum. SOD-3is very active in the scutellum 15-25 days after pollination, while the expression of the other isozymes is highly variable during this stage of development.

Intracellular Localization Results of the cell fractionation studies show that superoxide dismutase activity is associated with the cytosolic, mitochondrial, and chloroplastic fractions of maize seedlings. Sucrose gradient centrifugation has revealed the

136

J. A. Baum and J. G. Scanddios: Superoxide Dismutases in Maize Development

I-

I

1-

50 w

1

0

I

i15

0

2

w

20

P

FRACIION

rRACTION

WHBEII

Ftg. 3. Organelles from scutella of D a y 4 etiolated d i g s separated by sucrose gradient centrifugation. Isocitrate lyase and cytochrome oxidase an used as marker enzymes for glyoxysomes and mitochondria, mpatively. Superoxide dismutase is assoCiated with mitochondria whik catalasc is associated with glyoxysomes. Percent sucrosc is represented by the dashed line

LE

Sf

M

Chl

Et

(31

bL

Fig. 4. Subcellular distribution of superoxide dismutase isozyrnes from W64A seedlines. LE = liquid endoqmm crude extract, ='S d u b k fraction; M = mitochondrial fraction; Chl = chloroplastic fraction; EI = etiochloroplastic fraction; GI = glyoxysomal fraction; GL = green leaf crude extract Mitochondria and glyoxysoma were isdated by sucrose gradient centrifugation 01 described under Mcthodk Plastid fractions w m prepared by differential centrifugation: cnrde p l d o b t a i d by cmtrifu@~n at 1,500g (maximum) for 5 min mre resuspended in 25 ml d cold grinding medium and repelleted

NUMBER

Fig. 5. Isolation of intact chloroplasts from 10 day okl maize leaves by sucrose gradient centrifugation. Crude filtrate was laymd directly onto a semi-linear sucrose gradient and centrifuged as d a c n i under Methods.The pcak of triose phosphate isomcnse and superoxide dismutase activity at fraction 14 has an isopycnic density of approximatdy 1.204g/cm'. Two bands of chlorophyll were ob served following centrifugation. The denser chlorophyll band coincided with triose phosphate isomerase activity and thus represents intact chloroplasts. The less dense band of chlorophyll coincided with the peak of supcroxide dismutasc at fraction 29 and represents the broken chloroplast fraction

presence of the cyanide-resistant isozyme (SOD-3) in the mitochondria of maize scutella (Figs. 3 and 4). A faint cyanide-resistant isozyme migrates just behind SOD-3on starch gels and is also assoCiated with mite chondria. Cyanidesensitive SOD-1 is associated with chloroplasts and etiochloroplasts prepared by difTerential centrifugation (Fig. 4). Sucrose gradient centrifugation confinned the presmce of superoxide dismutase in the intact chloroplast fraction of maize leaves (Figs. 5 and 6). The peak of superoxide dismutase activity coinciding with that of triose phosphate isomerase ( 1.204 g/cm3) is due to the presence of SOD-1 (Fig. 7). The second peak of superoxide dismutase activity (fraction 29) has a density of approximately 1.17 g/cm3 and represents the broken chloroplast fraction as judged by the presence of chlorophyll (not quantitated) and the

137

J. A. Baum and J. G. Scandalios: Superoxide Dismutases in Maize Development c

PEPOXIDASE

2

,

w

w

u

0 20

10

t

30

40

50

60

F R A C T I O N NUMBER

Fig. 6. Sucrose gradient centrifugation of a chloroplast-enriched fraction obtained by differential centrifugation. The intact chloroplast fraction represented by the presmn of triose phosphate isomerase activity has an isopycnic density of 1.204 g/cm'. Enzyme units are presented in the same scale as in Figure 5 to emphasize the enhanced recovery of both superoxide dismutase and triose phosphate isomerase activity in the intact chloroplast fraction. Catalase and cytwhrome oxidase w m barely assayable on the gradient due to the initial differential centrifugation step. The small peak of catalase activity in the intact chloroplast fraction probably represents an artifactual association with this organelle

y'4zv-

LE

IC

BC SN

Fig. 7. Zymogram showins the distribution of SOD-1 on the sucrose gradient presented in Figure 6. LE = liquid endosperm marker; IC = intact chloroplasts; BC = broken chloroplasts; SN = supernatant (top of the gradient). Whik SOD- 1 is associated with intact chloroplasts, only traces of this isozyme can be detected in the broken chloroplast and supernatant fractions

DAY s POS 1 - POLLI N A T I O N

DAIS AFTER

lMBl0lTlON

Fig. 8. Time course of superoxide dismutase, catalasc, and peroxidase specific activities in the scutellum during kernel development and e d y germination

absence of a peak of triose phosphate isomerase activity. The yield of intact chloroplasts by sucrose gradient centrifugation techniques is, at best, less than 20% [22, 271. Consequently, the bulk of the chloroplasts (at least 80%) are found in the broken chloroplast fraction. If SOD-I were largely membrane-bound, then the broken chloroplast fraction would be expected to contain considerably more SOD-1 than the intact chloroplast fraction. In fact, just the opposite is true: SOD-I is present in the intact fraction but is virtually absent from the broken chloroplast fraction. Thus, it appears that SOD1 is released from broken chloroplasts and its 8ssociation with intact chloroplasts is not due to artifactual binding to membranes. SOD-2, SOD-4, and SOD-5 a p pear to be localized only in the cytosol. Catalase is associated with glyoxysomes while superoxide dismutase is not (Fig. 3). Furthermore, catalase has not been found to be associated with either chloroplasts or scutellar mitochondria.

138

J. A. Baum and J. G. Scandalios: Superoxide Dismutases in Maize Development

a

22

25

30

35

b

43

1

DS

2

4

6

8

1

0

DAYS

Developmental Expression of SOD in the Scutellum The time course of superoxide dismutase specific activity in the scutellum during kernel and seedling development was determined (Fig. 8). Superoxide dismutase activity is low in the scutellum 15-20 days aRer pollination and gradually increases to a near steady-state level at the dry seed stage. There are no dramatic changes in the expression of the five SOD isozymes in the scutellum during the course of devilopment as judged by starch gel electrophoresis (Figs. 9A and B). Scutellar extracts from germinating seedlings were diluted somewhat to avoid problems with amylase digesting the gel. The activity of SOD-4 appears relatively low in the scutellum during kernel development and is not clearly resolved by starch gel electrophoresis. A sharp peak of catalase activity was observed during germination, in agreement with previous findings (281. The peroxidax time c o r n is peculiar in that soluble peroxidase activity is barely detectable immediately following seed imbibition. This may be a reflection of its association with cell walls. Discussion Our laboratory has devoted considerable research to

characterizing the genetics and biochemistry of catalase isozymes in maize. We have initiated a study of superoxide dismutase to complement our earlier work with catalase and to expand our knowledge of peroxide metabolism within this organism. Maize superoxide dismutases have been resolved into five major electrophoretic forms by starch gel elec-

trophoresis. These have beem designated SOD- 1, SOD2, SOD-3,SOD-4, and SOD-5 in order of their migration to the anode, in accordance with the recommendations of the Subcommitteeon Multiple Molecular Forms of Enzymes of the IUPAC-IUB (197 1). In the lines and tissues that were examined, there was no detectable tissue specificity with respect to these five electrophoretic forms. Quantitative differences in the expression of SOD-1 and SOD-3 seem apparent: SOD-1 and SOD-3 appear to be preferentially expressed in the leaf and in the scutellum, respectively. We are attempting to quantitate this variation in isozyme expression. The time courses of superoxide dismutase, catalase, and peroxidase activities were studied in the scutellum during kernel development and germination. Although no obvious correlations were observed, it should be noted that all three enzymes exhibit highest total specific activity during germination. This may be related to the high rate of metabolism in germinating seeds. The peak of catalase activity observed during germination may be a reflection of the Merentid expression of the Cut1 and Cat2 gene products [28, 291. Under the conditions of our gel assay, there are no serious interferences due to peroxidases. NO significant peroxidase activity was detected in mitochondrial and chloroplast fractions using the spectrophotometric assay. There is no correspondence between the benzidine-specificp e r o x i d e isozymes and the superoxide dismutase isozymes of maize (zymogram not shown). All of the major peroxidases in maize will use benzidine or o-dianisidme as a hydrogen donor. Furthermore, none of the superoxide dismutase isozymes exhibit the tissue specificity characteristic of the maize peroxidases 130; unpublished data, this laboratoryl.

J. A. Baum and J. G.Scandalios: Superoxide Disrnutascs in Maize Development

The cell fractionation studies undertaken demonstrate the presence of superoxide dismutase activity in three subcellular fractions: the mitochondrial fraction, the chloroplastic fraction; and the cytosolic fraction. Furthermore, the association of superoxide dismutase activity with the two organelles can be attributed to the differential localization of distinct isozymes. In agreement with what has been observed in other eukaryotes, maize has a cyanide-resistant superoxide dismutase associated with mitochondria. Cyanide-resistant manganese-containing superoxide dismutases have been reported in the mitochondria of yeast, chicken, rat, and man [2, 8, 3 1, 321. A copper and zinc-containing superoxide dismutase has been reported in spinach chloroplasts (331. We have used a variety of techniques for isolating intact chloroplasts. Differential centrifugation experiments suggested that SOD-1 is associated with chloroplasts and etiochloroplasts; sucrose gradient centrifugation was used to confirm the isozyme’s association with intact chloroplasts. The procedure described by Miflin and Beevers [22] relies on the fact that chloroplasts have a much higher sedimentation coefficient than either mitochondria or microbodies. Thus, after 15 min of centrifugation, only the intact and broken chloroplasts have reached their isopycnic density of 1.204 d c m 3 and 1.17 g/cm3, respectively. This combination of differential and isopycnic density centrifugation allows for clean separation of chloroplasts from other organelles within the cell. The chloroplasts at 1.204 g/cm3 are intact in the sense that they retain enzymes commonly believed to be present in the stroma and which are lost upon breakage of the organelle (e.g., triose phosphate isomerase). Electron micrographs of maize chloroplasts obtained by this procedure have been published [271. The yield of intact chloroplasts obtained by this technique is low but nevertheless provides a means for demonstrating the localization of enzymes within chloroplasts. In a modification of the procedure of Mifin and Becvers (221, a chloroplast-enriched fraction obtained by differential centrifugation was layered onto a semilinear sucrose gradient and subjected to the same 15 min centrifugation period (Fig. 6). The same results were obtained: SOD-1 is present in the intact chloroplast fraction but appears to be absent from the broken chloroplast fraction. We regard this as good evidence that SOD-1 is not artifactually binding to chloroplast membranes and that it is truly localized within plastids. The presumably membrane-bound component exhibiting superoxide dismutase activity in the broken chloroplast fraction has not yet been identified but may in fact be

139

SOD-1. It is worthwhile noting that a significant fraction of the superoxide dismutase activity associated with spinach chloroplasts appears to be membrane bound [331. The superoxide dismutase activity in the intact chloroplast fraction was confirmed by use of the standard assay for superoxide dismutase 111. We are currently investigating the distinction between the chloroplast form of superoxide dismutase and the other cyanide-sensitive isozymes of SOD in maize. No attempt has yet been made to estimate the percentage of superoxide dismutase activity associated with mitochondria and chloroplasts. One purpose of this study was to further evaluate the physiologic relationship between catalase and superoxide dismutase. The cell fractionation studies undertaken so far in maize demonstrate different organelle specificities for these two enzymes. Catalase has been shown to be localized in glyoxysomes I281 but not in scutellar mitochondria or plastids. Conversely, superoxide dismutase in maize is associated with mitochondria and plastids but not with glyoxysomes. This suggests that, in mitochondria and chloroplasts, catalase does not serve a primary role in removing the hydrogen peroxide presumed to be generated by superoxide dismutase. It is conceivable, however, that the level of catalase in mitochondria and plastids is so low that its presence would be difficult to demonstrate 1341. In fact, unpublished data from our laboratory indicate that catalase is associated with mitochondria in maize leaves. Another possibility is that a H,O,-scavenging enzyme such as glutathione peroxidase serves to remove H,O, from these organelles [351. In the future, we plan to further study the relationship between superoxide dismutase and H,O,-scavenging enzymes within the cell. This investigation will involve induction experiments using hyperbaric oxygen and possibly paraquat [IS]. Our preliminary results indicate that variations in the expression of SOD-4 and SOD-5 (Fig. 1) are under nuclear gene control. We have screened a wide variety of exotic maize lines and have recovered apparent electrophoretic variants for both SOD-1 and SOD-3; crosses have been made with these variants to establish their mode of inheritance. Analysis of these crosses is forthcoming. We have begun purification of the superoxide dismutases in maize for the purpose of characterizing them with respect to some of their biochemical properties; this information should aid us in understanding the evolutionary and genetic relationships among these isozymes.

140

J. A. Baum and J. G. Scandalios: Superoxide Dismutases in Maize Development

References 1. McCord. J. M, Fridovich, 1.: Superoxide dismutase. An en-

zymic function for erythrocuprdn (hcmocuprein). J. Biol. Chem 244, 6049 (1969) 2. Weisiga, R. A.. Fridovich, I.: Superoxide dismutase: organelle speciricity. J. Biol. Chcm. 248, 3582 (1973) 3. Misra, H. P., Fridovich, I.: The purifiation and properties of superoxide dismutase from Nrutwpra cmssa. J. Bid. C h m . 247, 3410 (1972) 4. Beauchamp, C. 0..Fridovich, I.: Isozymes of superoxide dism u m e from wheat germ. Biochim. Biophys. Acta 317, 50 (1973) 5. Sawads, Y., Ohyama, T.,Yamazaki. I.: Preparation and physiochemical properties of grcen pea superoxide dismutase. Biochim. Biophys. Acta 268, 305 (1972) 6. Kccle, B. B., Jr., McCord, I. M.,Fridovich. I.: Superoxide dismutase from Escherichio colf B. A new manganese-containing enzyme. J. Biol. Chcm. 245, 6176 (1970) 7. Van- P. G., Kcele, B. B.. Jr., Rajagopalm. K.V.: Superoxide dismutase from Streptocucccw mtans: isolation and characterization of two forms of the enzyme. J. Biol. Chcm. 247, 4782 (1972) 8. Ravindranath, S. D., Fridovich, I.: Isolation and characterization d a manganesecontaining superoxide dismutase from yeast J. Biol. Chem. 250, 6107 (1975) 9. Salin, M. L,Day, E.D, Jr., Crapo, J. D.:Isolation and characterization of a manganutcontaining superoxide dismutase from rat liva. Arch. Biochcm. Biophys. 187, 223 (1978) 10. McCord, J. M.: Iron and manganesccontaining superoxide dismutases: Structure, distribution and evolutionary relationships. In: Advances in experimental. medicine and biology: Iron and Copper proteins. Yasunobu, K. T.. Mower, H.F.. Hayaishi. 0. (cds.),vd. 74, pp. 540-550. New York. London: Plenum P r u s 1976 11. Kanematsu, S.. Asada. K.:Superoxide dismutase from an ans o b i c photosynthetic bacterium, Chmmdum vinosum. Arch. Biochem. Biophys. 185, 473 (1978) 12. Fridovich. I.: Superoxide and evdution. H&z. Biochem. Biophys. I, l(1974) 13. Fridovich. I.: Superoxide dismutases. Annu. Rev. Biochem. 44, 147 (1975) 14. McCord, J. M.: Free radicals and inflammation: protection of synovial nuid by superoxide dismutase. Science 185, 529 (1974) 15. Hasaan. H. M.. Fridovich, I.: Regulation of synthesis of superoxide dismutase in Escherfchb mil. J. Biol. Chem. 252, 7667 (1977) 16. !%met, P.-M.: SOD genes in humans: Chromosome localitation and electrophoretic variants. In: Superoxide and superoxide dismutasts. Michdson, A. M.,McCord J. M..Fridovich. I. (&). pp. 459-465. London, New York, SM Francisco: Academic PM 1977 17. SCMdaliOS, J. G.: Genetic control of multiple molecular forms of enzymes in p l ~ t s :a review. Biochcm. Genet 3, 37 (1969) 18. s c a n d d h , J. G., Liu, E. H.,Campeau, M.A.: The effects of intragenic and intergenic complementation on catdase structure

and function in maize: a molecular approach to hcterosis. Arch. Biochem. Biophys. 153, 695 (1972) 19. Beers. P. F., S i r , 1. W.: A spectrophotometric assay for mcasuring the breakdown of hydrogen peroxide by catalase. J. Biol. Chcm. 195, 133 (1952) 20. Worthington enzyme manual. Freehold, N. J.: Worthington Biochemical Corporation 1972 21. Lowry. 0. H., Rosebrough, N. J., Farr. A. L., Randall, R. J.: Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265 (1951) 22. Mifin, B.. Beevers. H.: Isolation of intact plastids from a range of plant tissues Plant Physiol. 53, 870 (1974) 23. Briedenbach, R. W., Bmcrs, H.:Association of the glyoxylate cycle enzymes in a novel subcellular particle from castor bean endosperm. Biochem. Biophys. Res. Commun. 27, 462 (1967) 24. Smith, L.: Spcctrophotomdric assay of cytochrome C oxidase. In: Methods of biochemical analysis. Glick, D. (4.Vol. ).2, pp. 4 2 7 4 3 4 . New York: Interscience Publishing Inc. 1955 25. Cooper, T. G., Beevers, H.: Mitochondria and glyoxysomes from castor bean endosperm. J. Biol. Chem. 244, 3507 (I 969) 26. Gibbs, M.,Turner. J. F.: Enzymes of glycolysis. In: Modem methods of plant analysis. Linskens, H.F., Sanwal, B. D.. Tracey, M.V.. (eds.). Vol. 7, p. 520. Berlin, Gijttingen, Heidelberg: Springer 1964 27. Bryan, J. K.. Lissik, E. A., Matthews, B. F.: Changes in enzyme regulation during growth of maize: intracdlular localization of homoserine dchydrogenax in chloroplasts. Plant Physiol. 59, 673 (1977) 28. kandalios, J. G.: SubceUular localization of catalasc variants coded by two genetic loci during maize development. J. Hered. 65, 28 (1974) 29. Quail, P. H., Scandalios, J. G.: Turnover of genetically-defined catalase isozymes in maize. Proc. Natl. Acad. Sci. USA 68. 1402 (1971) 30. Brewbaker, J. L.. Hasegawa. Y.: Polymorphisms of the major peroxidases of maize. In: Isozymes. Markert. C. L. (ed.), Vol. 3, p. 659. New York: Academic Press 1975 31. Tyler, D. D.: Pdarographic assay and intracellular distribution of superoxide dismutase in rat liver. Biochcm. J. 147, 493 (1975) 32. Beckman. G., Lundrgrcn. E.. Tarnvik, A.: Superoxide dismutasc isozymcs in different human tissues, their genetic control, and intraccllular localization. Human Hered. 23, 338 (1973) 33. Asada, K., Urano, M., Takehashi, M.:Subcellular localization of superoxide dismutase in spinach leaves and preparation and properties of crystalline spinach superoxide dismutase. Eur. J. Biochem 36, 257 (1973) 34. Rich, P. R., Boveris, A., Bonner. W. D., Ir., Moore. A. L.: Hydrogen peroxide generation by the alternate oxidase of higher plants. Biochem Biophys. Res. Commun. 71, 695 (1976) 35. Wolosiuk, R. A.. Buchanan, B. B.: Thiondoxin and glutathione regulate photosynthesis in chloroplasts. Nature 266, 565 (1977) Received August 1978/Acqted Deccmbcr 1978