Purification and characterization of monodehydroascorbate reductase from soybean root nodules

ARCHIVES

OF BIOCHEMISTRY

Vol. 292, No.

1,

AND

BIOPHYSICS

January, pp. 28-286,

1992

Purification and Characterization of Monodehydroascorbate Reductase from Soybean Root Nodules’ David A. Dalton,2 Lorene Langeberg, and Michael Department

of Biology, Reed College, Portland,

Robbins

Oregon 97202

Received July 24, 1991, and in revised form September 24, 1991

Soybean (GZycine max (L.) Merr.) root nodules contain the enzymes of the ascorbate-glutathione cycle as an important defense against activated forms of oxygen. A key enzyme in this cyclemonodehydroascorbate reductase (MR)-was purified 646-fold and appeared as a single band on SDS-PAGE with silver or Coomassie blue staining. Purified MR contained 0.7 mol FAD/mol enzyme and had a specific activity of 288 pmol NADH oxidized * min-’ - mg protein’. The enzyme was a single subunit occurring as two isozymes (MR I and MR II) with M, values of 39,000 and 40,000. Isoelectric focusing revealed that each isozyme consisted of two forms with pl values of 4.6 to 4.7. Ferricyanide and 2,6-dichlorophenol-indophenol were effective as electron acceptors. The purified enzyme did not possess leghemoglobin reductase activity. Inhibition by p-chloromercuribenzoate indicated the involvement of a thiol group in MR activity. The K,,, values were 5.6, 150, and 7 pM for NADH, NADPH, and monodehydroascorbate, respectively. The pH optimum was 8 to 9. The N-terminal sequence of 10 amino acids of MR II had little homology to known protein sequences. 0 1992 Acad~emic Press, Inc.

concentrations in legume nodules and can generate superoxide and hydroxyl radicals (2). Oxygen toxicity is minimized by defense enzymes such as superoxide dismutase and the enzymes of the ascorbate-glutathione cycle (1) which are abundant in nodules. In the initial reaction of this cycle, HzOz is scavenged using the reducing power of ascorbate (ASC)3 in a reaction catalyzed by ascorbate peroxidase (EC 1.11.1.11). This enzyme is widespread in various plant materials including chloroplasts (3) and root nodules (4). The product of this reaction is monodehydroascorbate (MDHA), a free radical that has two possible fates. MDHA may be converted directly to ASC by monodehydroascorbate reductase (MR, EC 1.6.5.4, also called ascorbate free radical reductase) in the following reaction: MDHA

+ NADH

+ H+ + ASC + NAD+.

Alternatively, MDHA may spontaneously disproportionate to dehydroascorbate (DHA) and ASC. The DHA is then converted to ASC in a reaction catalyzed by DHA reductase (EC 1.8.5.1): DHA + 2GSH + ASC + GSSG.

Nitrogen-fixing organisms are especially vulnerable to oxygen damage due to O2 inactivation of nitrogenase and to the strong reducing conditions that can lead to the production of hydrogen peroxide and superoxide and hydroxyl free radicals. In legume root nodules, activated forms of oxygen may be produced by several proteins, including ferredoxin, leghLemoglobin, hydrogenase, uricase, and others (1). Of these, leghemoglobin is probably the most important since leghemoglobin is present in high i Supported by NSF Grant DCB-8903254 and a PPG Industries Foundation Grant of Research Corporation. ’ To whom correspondence should be addressed. 0003-9861/92 $3.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form

ill

PI

This reaction converts reduced glutathione (GSH) to the oxidized form (GSSG). GSH is regenerated by glutathione reductase (EC 1.6.4.2): GSSG + BNADPH

+ 2H+ --f 2GSH + 2NADP+.

[31

3 Abbreviations used: ASC, ascorbate; DCPIP, 2,6dichlorophenoLindophenol; DHA, dehydroascorbate; GSH, reduced glutathione; GSSG, oxidized glutathione; MDHA, monodehydroascorbate; MR, monodehydroascorbate reductase; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; BSA, bovine serum albumin; DTT, dithiothreitol; PMSF, phenylmethylsulfonyl fluoride; FAD, flavin adenine dinucleotide; FMN, flavin mononucleotide. 281

Inc. reserved.

282

DALTON.

LANGEBERG,

MDHA may be produced in plant cells by at least four processes other than the ASC peroxidase reaction: (a) oxidation of ASC by superoxide and hydroxyl radicals (5); (b) reaction of ASC with products formed by the reaction of lipid peroxyl radicals and a-tocopherol (6); (c) the metal-catalyzed autoxidation of ASC (7); and (d) oxidation of ASC by ASC oxidase. It is noteworthy that leghemoglobin is a multiple threat in regard to ASC oxidation and hence MDHA formation because oxyleghemoglobin can slowly degrade to produce superoxide which disproportionates to H,Oz (8). Through Fenton chemistry, the Hz02 can react with free Fe to produce hydroxyl radicals. This situation is aggravated by the ability of Hz02 to cause the breakdown of heme, thus releasing Fe ions that further promote the formation of hydroxyl radicals (2). MR is widespread in plant tissues, where it is present in roots, chloroplasts, cytosol, and mitochondria (9). Although DHA reductase is common in plants, it is not as widespread as MR (9). MR has been purified from spinach chloroplasts (lo), cucumber fruit (ll), potato tubers (12), Euglena cytosol (13), and Neurosporu (14); however, the properties and importance of MR in root nodules have not been investigated. The lack of attention to the enzymes of the ascorbate-glutathione cycle in root nodules represents a substantial gap in the understanding of nodule physiology, especially in light of the increasing awareness of the importance of oxygen relations of nodules and the critical role of oxygen regulation through leghemoglobin and the variable oxygen diffusion barrier. MATERIALS

AND

METHODS

Growth of plants. Soybean (Glycine max (L.) Merr. cv. Williams) plants were inoculated with Brudyrhizobium juponicum 122DES and grown as described previously (1). Nodules were harvested between 25 to 40 days after planting and frozen at -7O’C for later use. Enzyme assay. MR activity was assayed by following the decrease in Aad due to the oxidation of NADH (10). Monodehydroascorbate was generated by ascorbate oxidase. The reaction mixture contained 0.6 nmol NADH, 0.625 pmol ascorbic acid, 0.5 units ascorbate oxidase (Sigma), and sufficient enzyme preparation to result in a AAaJmin between 0.02 and 0.180. The reaction mixture final volume was 1 ml. All reaction components were prepared fresh daily in 50 mM Tris-HCl, pH 7.8, except ascorbate oxidase which was frozen in l-ml aliquots (10 units/ml) and thawed as needed. Activity was expressed in units of pmol NADH oxidized * mini’. Leghemoglobin reductase activity was examined using the procedure described by Becana and Klucas (15). Protein concentrations were determined by Coomassie blue binding (16), with BSA as the standard. Purification of MR. All work was performed at 4°C. All buffers and reagents were prepared using distilled, deionized water. Unless specifically stated otherwise, the buffer used throughout was 50 mM TrisHCl, pH 7.8, containing 0.2 mM EDTA and 10 mM @-mercaptoethanol. Frozen nodules (25 g) were ground in liquid nitrogen with a mortar and pestle. Immediately after the nitrogen had boiled away, the dry powder was added to 100 ml of buffer containing 6.25 g of insoluble polyvinylpolypyrrolidone. This slurry was stirred gently until all ice crystals had melted and the plant material was thoroughly suspended. The macerate was then squeezed through four layers of cheesecloth. Usually 100 to 150 g of nodules was processed in 25-g batches and then combined into

AND

ROBBINS

one solution after passage through cheesecloth. This preparation was then centrifuged at 10,OOOgfor 15 min. The resultant supernatant is referred to as the crude extract in this report. Ammonium sulfate precipitation was used to obtain a fraction between 25 and 80% saturation which retained most of the MR activity. The final pellet was dissolved in 75 ml of buffer and dialyzed extensively for 2 to 3 days with four to six changes of buffer. The dialyzed sample was then added to a 2.5- X 35-cm column of DEAE-Sephacel (Pharmacia) equilibrated with the same buffer. The column was washed with 100 ml of buffer followed by a linear KC1 gradient constructed by mixing 200 ml buffer and 200 ml of 250 mM KC1 in buffer. Fractions with MR activity (fraction Nos. 16-34, Fig. 1) were pooled and then concentrated to 5.0 ml using a stirred ultrafiltration cell (Amicon Model 8050) with a 20-kD molecular weight cutoff cellulose filter (Spectrum Spectra/par type C). The sample was loaded onto a 2.5- X 92-cm column of Sephacryl S-200 (Pharmacia) equilibrated with standard buffer plus 100 mM KCl. For molecular weight determinations, the column was standardized with sodium azide, ribonuclease (M, 13,700), chymotrypsinogen (25,000), ovalbumin (43,000), BSA (67,000), and Blue Dextran. Fractions with MR activity were combined, dialyzed extensively to remove KCl, and concentrated to 1.0 ml. The sample was then added to a 1.5- X 13.5.cm column of reactive blue 2-Sepharose CL-6B (Sigma). The column was preconditioned by passage of 10 ml of a solution of BSA (2 mg/ml) followed by washing with 100 ml of buffer. MR bound only weakly and high recovery was achieved by washing with 20 to 40 ml of buffer alone. Fractions with high MR activity were pooled and concentrated to about 2 ml prior to final purification by nondenaturing polyacrylamide gel electrophoresis (17) in a 1.5. X 140. X 170-mm slab gel. The sample was dialyzed overnight against 15 mM Tris-H3P0,, pH 6.9, containing 1 mM DTT, 0.2 mM EDTA, and 10 mM /3-mercaptoethanol. Both 10% (w/v) sucrose and 0.01% (w/v) bromphenol blue were added prior to loading in a single well across the full width of the gel; 1 mM DTT was included in both the upper and lower gels and 2 mM thioglycolate was included in the running buffer to ensure reducing conditions in the gel. The gel was run at constant voltage at 150 V for 1.5 h followed by 180 V for 15 h. Two 4-mm-wide vertical slices were removed and stained by Coomassie blue to identify major protein bands. Horizontal slices were removed from the corresponding regions of the unstained gel and MR was recovered by diffusion and electroelution into 25 mM Tris and 192 mM glycine, pH 8.5, containing 0.2 mM EDTA and 10 mM /l-mercaptoethanol in a Bio-Rad Model 422 electroeluter. Other procedures. Values for Km were determined by using a nonlinear regression analysis with a computer program called Enzyme (Zeba Kimmel and Ronald W. McClard, Reed College). In determination of the K,,, for MDHA, the [MDHA] was adjusted by altering the concentration of ASC between 0 and 0.05 mM in the reaction mixture. The steady-state [MDHA] was determined by monitoring the absorbance at 360 (11, 20) in a separate mixture lacking NADH. Concentrations of

0

10

20

fraction

30

no.

FIG. 1. Elution profile of monodehydroascorbate reductase on DEAESephacel. The initial preparation consisted of 165 ml with a protein concentration of 3.30 mg/ml and a specific activity of 0.205 amol NADH oxidized. min-’ . mg protein-‘.

SOYBEAN

ROOT

NODULE

MONODEHYDROASCORBATE

MDHA thus generated were found to be in the range of 0 to 3 pM and were constant over the course of each assay period (up to 3 min) as judged by the stability of the AsGo.The [MDHA] could not be increased above 3 pM despite additional manipulations of the concentration of ASC and of ASC oxidase. K,,, values for NADH and NADPH were determined at a constant [MDHA] of 2 @M and over a range of substrate concentrations of NADH or N.4DPH from 0 to 62 or 0 to 200 pM, respectively. Flavin content was determined by comparison of relative fluorescence at pH values of 7.7 and 2.6 using a Perkin-Elmer Model 203 fluorescence spectrophotometer (18). SDS-PAGE (12% gel) was performed using a Bio-Rad Mini-PROTEAN II cell following the manufacturer’s operating procedures, including staining with Coomassie blue or silver reagent. Isoelectric focusing was performled with a Bio-Rad Model 111 Mini-IEF apparatus over a pH range of 3 to 7 following the manufacturer’s suggested procedures. For detection of enzyme activity directly in gels, a minigel (1 X 60 X 100 mm) was electrophoresed as described above for nondenaturing gels, except the power was maintained at 40 V for 21 h. MR was visualized with a staining procedure based on diaphorase activity (19). N-terminal amino acid sequencing was performed using an Applied Biosystems 477 pulsed liquid phase protein sequencer with on-line PTH analyzer. Amino acid composition was done usin;: a PICO-TAG system (Waters) with a Perkin-Elmer Nelson 1020 for data acquisition and analysis. The homology search was done usin,g the computer program PROSCAN (DNASTAR, Madison, WI).

RESULTS

Purification Specific activity of MR in crude extracts typically ranged from 0.15 to 0.55 pmol NADH oxidized. min-l . mg protein-l. Fresh and frozen nodules (stored at -70°C for up to 12 years) yielded crude extracts with comparable activity. MR was only pa.rtially retained on DEAE-cellulose (Fig. 1). Substantial. amounts of MR eluted trailing the initial flow-through of protein and additional MR eluted at a KC1 concentration of 40-60 mM. No attempt was made to distinguish lbetween the properties of MR from separate DEAE fractions. Further purification was performed on pooled fractions from the trailing edge of the first activity peak and fractions eluted by KCl. After passage through Sephacryl S-200, MR eluted as a single peak with a M, of 43,000.. Dye-affinity chromatography with reactive blue Sepharose was effective because MR was sufficiently retained by the column to allow the bulk of contaminating proteins to pass through first. MR then eluted in a single peak. Initial attempts at purification with reactive blue Sepharose were unsuccessful due to extremely low recovery of MR. This problem was overcome by pretreatment with BSA. The addition of KCl, NADH, or NAD+ to the elution buffer did not result in recovery of additional MR in either the absence or the presence of BSA pretreatment. Following chromatography on reactive blue Sepharose, the preparation appeared to consist of about 80% MR with minor contaminants
REDUCTASE

283

97.4 66.2 42.7

FIG. 2. SDS-PAGE of monodehydroascorbate reductase (MR). Lane 1, molecular weight standards: rabbit muscle phosphorylase b (97,400), BSA (66,200), hen egg white ovalbumin (42,699), bovine carbonic anhydrase (31,000), soybean trypsin inhibitor (21,500), and hen egg white lysozyme (14,400). Lane 2, 5.5 /Lg of partially purified preparation following chromatography with reactive blue 2-Sepharose CL-GB, showing broad MR band at Mr values of 39,000 to 40,000. Lane 3, 2.5 pg of MR I (M, 39,000) recovered from nondenaturing gel. Lane 4, 2.5 pg of MR II (M, 40,000) recovered from nondenaturing gel. All lanes stained with Coomassie blue.

electrophoresis yielded two distinct major bands, designated MR I and MR II, each of which displayed diaphorase activity (Fig. 3). The proteins were recovered from the nondenaturing gel by diffusion and electroelution and found to have MR activity of 214 and 288 units * mg protein-‘, respectively (Table I). The recovered MR I and MR II preparations each showed a single band after SDSPAGE with Mr values of 39,000 for MR I and 40,000 for MR II (Fig. 2). Staining with Coomassie blue or silver reagent gave identical results. Since M, values approximate that determined by gel filtration, it is evident that MR I and MR II are composed of a single subunit. Characterization Purified preparations stored at 4°C in the standard buffer maintained undiminished activity for up to 3 months; however, activity was lost within a few hours in the absence of /$mercaptoethanol. Freezing resulted in complete loss of activity, but extracts containing 50% glycerol (v/v) could be stored at -20°C for at least 3 months without loss of activity. The relative abundance of MR I and MR II was not affected by inclusion of protease inhibitors (10 PM antipain and 1 mM PMSF) in all buffers used for extraction and purification. Isoelectric focusing of MR I and MR II revealed two separate bands for each form of MR (Fig. 4). On IEF gels,

284

DALTON,

LANGEBERG,

AND

ROBBINS

6.6 5.9 5.4 5.1 MR I MR II 4.6 4.2

+ FIG. 3. Nondenaturing PAGE of monodehydroascorbate reductase stained for diaphorase activity. Lane 1, 3 pegof partially purified preparation following chromatography with reactive blue 2-Sepharose CL6B (identical sample to Fig. 2, Lane 2). Lane 2,2 bg of MR II recovered from earlier nondenaturing gel and reelectrophoresed (identical sample to Fig. 2, Lane 4). Lane 3, 2 Kg of MR I recovered from earlier nondenaturing gel and reelectrophoresed (identical sample to Fig. 2, Lane 3). Gels were stained by immersing in 11 ml of 1.3 mM NADH, 1.2 mM 3[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide, and 0.06 mM DCPIP in 0.25 M Tris-HCl, pH 8.4, for 10 min.

FIG. 4. Isoelectric focusing gel of monodehydroascorbate reductase. Lane 1, pl standards: carbonic anhydrase (6.6, 5.9, and 5.41, fl-lactoglobulin A (5.11, trypsin inhibitor (4.6), glucose oxidase (4.21, and amyloglucosidase (3.6). Lane 2, 0.3 pg of purified MR I. Lane 3, 0.3 pg of purified MR II. Stained by silver reagent.

ration by chromatofocusing yielded two peaks of MR activity in fractions with pH values of 4.2-4.7, but recovery was low and this procedure was not pursued further. The K,,, values for preparations of highly purified MR I and MR II were 5.6,150, and 7 PM for NADH, NADPH, and MDHA, respectively. Rates with NADH were much higher than those with NADPH, with V,,, being 6.7-fold higher with NADH. Purified MR was tested with a variety of possible substrates as alternative electron acceptors to MDHA. MR did not catalyze the reduction of leghemoglobin (Lb3’ to Lb’+. CO). This reaction is catalyzed by free flavins and some flavoproteins (15) so it was hypothesized that MR

MR I consisted of bands with pl values of 4.8 and 4.7. The two bands for MR II were closely spaced near a pl value of 4.7. The ~14.7 band of MR I was slightly higher than either of the bands of MR II, thus indicating the uniqueness of each of the four bands from MR I and MR II. The same pattern was detected in three repetitions and with either silver or Coomassie blue G-250 staining. The low pl values suggested that chromatofocusing might be a convenient procedure for purification. Sepa-

TABLE Purification

Purification

Total protein (md

step

Crude extract (NHMO4 ppt DEAE-Sephacel Sephacryl S-200 Reactive blue 2-Sepharose Nondenaturing PAGE MRI MR II ’ Unit = gmol NADH

of Monodehydroascorbate

oxidized * min-‘.

3.6

I

Reductase

Specific activity (units” * mg protein-‘)

from

Soybean Root Nodules

Total activity (units)

Recovery (%)

Purification (-fold)

2790 2180 737 49.8 2.52

0.446 0.525 1.37 12.9 130

1240 1140 900 642 328

91.9 72.6 51.8 26.4

1 1.2 3.1 28.9 291

0.128 0.170

214 288

21 47

2.2 3.8

480 646

SOYBEAN

ROOT

NODULE

MONODEHYDROASCORBATE

285

REDUCTASE

‘TABLE II

Catalytic Activity of Purified Monodehydroascorbate Reductase from Soybean Root Nodules with Various Electron Acceptors Rate (pm01 - min-’ mg proteinm’)o acceptor

Concentration

Monodehydroascorbate KFe(CN)G DCPIP*

2 100 100

Electron

(PM)

MRI

MR II

90.4 203 30.5

111 234 27.2

lo!

. 55

Note. The assay mixture contained the electron acceptor, 60 nmol NADH, and 125 ng of purified MR in a total volume of 1.0 ml. Values were corrected for nonenzymatic rates. ’ Activity units are pmol NADH oxidized- min-’ * mg protein-‘, except when DCPIP was used, in which case the units are pmol DCPIP reduced. min-’ - mg protein-‘. * 2,6-Dichlorophenol-indophenol (sodium salt).

might function in this reaction and that this could be a secondary physiological role for this enzyme. The assay required removal of P-mercaptoethanol from the MR preparation by repeated washing in a centrifuge microconcentrator. Although MR is not stable in the absence of reducing agents, activity (as judged by the previously described assay for MR activity) was not lost during the short time (60-90 min) required for the leghemoglobin reduction assay. Furthermore, CO did not inhibit activity in the standard assay based on NADH oxidation in the presence of MDHA. The inclusion of up to 25 pg of purified MR I and II did not increase the rate of leghemoglobin reduction above th.e control (nonenzymatic) rate. MR I and MR II are capable of catalyzing the reduction of DCPIP and ferricyanide (Table II). Cytochrome c, dehydroascorbate, menadione, methylene blue, and benzoand naphthoquinone were ineffective as electron acceptors. DCPIP reduction provided a useful principle for activity staining in nondenaturing gels (Fig. 3). The effect of pH on MR activity was examined in a pH range from 5.6 to 10.1. Activity was highest in a pH range from 8.0 to 9.0 (Fig. 5). Use of different buffers with overlapping pH ranges indicated no buffer effect. Purified MR I and II were found to contain FAD (0.71 mol FAD * mol enzyme-l). No FMN was detected. Free flavins (FAD, FMN, or riboflavin) in concentrations from 1.0 to 100 ELM did not catalyze the oxidation of NADH in the presence of MDHA. Incubation of purified MR in buffer containing FAD (100 PM) did not affect MR activity. Inhibition by thiol reagents indicated the essentiality of thiol groups for MR activity. Ninety percent inhibition was observed in the presence of 0.025 mM p-chloromercuribenzoate and complete inhibition resulted at a concentration of 2.5 mM. Thiol-related inhibition was also

, 65

.

I 7.5

, 85

, 95

.',

I 10 5

PH FIG. 5. Effects of pH on monodehydroascorbate reductase activity. Assay was as described under Materials and Methods, except for changes in buffer composition as indicated. All buffer concentrations were 50 mM. MR was added as 10 ~1 of a partially purified preparation in 25 mM Tris, pH 7.8. Each value is the mean of three replicates.

observed in the presence of 5 mM N-ethylmaleimide (36% inhibition) and 5 mM iodoacetate (15% inhibition). The amino acid composition of MR II is presented in Table III. A comparison of MR from nodules and cucumber fruit reveals a similar amino acid composition, except MR from cucumber has substantially more serine, glycine, and glutamate (Table III). The N-terminal sequence of nodule MR II was AKTFKYIILG. Comparison of this sequence to known protein sequences revealed no close homology. No sequence information on MR from other sources is available.

TABLE

III

Amino Acid Composition

of Monodehydroascorbate Reductase from Soybean Root Nodules (MR II) and from Cucumber Fruit Residues/m01 Amino acid

Soybean nodules

Cucumber fruit (Ref. (11))

Ser Gb His Ax Thr Ala Pro Tyr Val Met Half-cystine Ile Leu Phe LYS AsGl-

14.1 33.0 11.4 16.1 14.4 36.4 13.7 12.4 36.0 0 n.d. 19.7 28.6 18.4 20.5 56.4 (Asp + Asn) 35.5 (Gln + Glu)

42.8 60.0 7.3 18.5 19.6 41.6 19.5 9.7 32.1 2.0 2.8 19.9 31.5 21.6 28.8 35.6 (Asp only) 47.0 (Glu only)

Note. n.d., not determined.

286

DALTON,

LANGEBERG.

DISCUSSION

The ascorbate-glutathione cycle is an important mechanism in soybean root nodules for minimizing damage from activated forms of oxygen. The importance of this cycle is indicated by the high concentrations of ASC (typically 600 to 800 nmol . g fresh weight-l) and total glutathione (2.0 to 3.0 ymol . g fresh weight-‘, unpublished observations of our laboratory) which are present in nodules. The high levels of enzymes such as ASC peroxidase and MR in nodules are further indications of the importance of this cycle. The presence of MR in nodules conveys considerable physiological flexibility since it provides for a separate pathway (vis-a-vis dehydroascorbate reductase and glutathione reductase) for ASC regeneration that does not involve glutathione or NADPH. The calculation for FAD content was based on the Coomassie blue dye binding assay for protein content with BSA as the standard. As with most procedures for protein determination, there is a substantial degree of variation depending on the protein standard used. This variation could explain why the FAD content was found to be less than 1.0. It is also likely that purified preparations contain unknown amounts of apoprotein. MR is so widespread in plants, algae, and animals that Arrigoni et al. (9) described it as a “ubiquitous” enzyme. With such a wide range of occurrence, it is not surprising to find considerable differences in the properties of this enzyme depending on the source. Among the various forms from plant sources, MR has been shown to be a single subunit with a range of molecular weights from 52,000 in Euglena (13) to 42,000 in Solanum tubers (12). MR from cucumber fruit has a M, of 47,000 (11). Other MR’s have not been studied for pl values or isozyme forms. The FAD content of nodule MR is shared by MR from cucumber fruit, but not by MR from Neurospora, which contains neither flavin nor heme (14). The pattern of effectiveness of various electron acceptors for nodule MR is similar to that of MR from cucumber fruit (11). The involvement of thiol groups in nodule MR activity has also been reported for MR from spinach chloroplasts (10) and from potato tubers (21). The strong inhibition by p-chloromercuribenzoate and the lack of activity with electron acceptors such as quinones and methylene blue are both characteristics which distinguish nodule MR from the FAD enzyme menadione reductase (DT diaphorase, EC 1.6.99.2, Ref. (22)). The ubiquity of MR applies not only to its phylogenetic distribution, but also to its subcellular location within a given species. For example, in cucumber fruit, MR is present in chloroplasts, mitochondria, microsomes, and cytosol(23). Such a broad distribution further underscores the essential role of MR. In Nz-fixing nodules, this role is especially critical due to the large capacity for the production of activated forms of oxygen. The present study provides background for continued investigation of the

AND

ROBBINS

ascorbate-glutathione cycle in nodules, especially concerning the important issues of subcellular location, immunological comparison to MRs from other sources, and regulation of expression of defense genes. ACKNOWLEDGMENTS Dr. Harold Evans (Department of Botany and Plant Pathology, Oregon State University) generously provided most of the nodules used in this study. Protein sequencing and amino acid analysis were performed by Dr. Charles Mitchell (Genetic Engineering Facility, University of Illinois). Dr. Manuel Becana (Department of Biochemistry, University of Nebraska) provided purified leghemoglobin. The authors are indebted to Dr. Bob Doss (USDA-AR& Corvallis, OR) for use of the fluorescence spectrophotometer.

REFERENCES 1. Dalton, D., Russell, S. A., Hanus, F. J., Pascoe, G. A., and Evans, H. J. (1986) Proc. N&l. Acad. Sci. USA 83, 3811-3815. 2. Puppo, A., and Halliwell, B. (1988) Planto 173,405-410. 3. Nakano, Y., and Asada, K. (1987) Plant Cell Physiol. 28,131-140. 4. Dalton, D. A., Hanus, F. J., Russell, S. A., and Evans, H. J. (1987) Plant Physiol. 83, 789-794. 5. Cabelh, D. E., and Bielski, B. H. J. (1983) J. Phys. Chem. 87,18091812. 6. Packer, J. E., Slater, T. F., and Willson, 737-738. 7. Scarpa, M., Stranato, R., Viglino, Chem. 258, 6695-6697.

R. L. (1979) Nature 278,

P., and Rigo, A. (1983) J. Biol.

8. Puppo, A., Rigaud, J., and Job, D. (1981) Plant Ski. Lett. 22, 353360. 9. Arrigoni, O., Dipierro, 125, 242-244.

S., and Borraccino,

G. (1981) FEBS L&t.

10. Hossain, M. A., Nakano, Y., and Asada, K. (1984) Plant Cell Phystil. 25,385-395. 11. Hossain, M. A., and Asada, K. (1985) J. Biol. Chem. 260, 12,92012,926. 12. Borraccino, 521-526.

G., Dipierro,

S., and Arrigoni,

0. (1986) Planta

13. Shigeoka, S., Yasumoto, R., Onishi, T., Nakano, S. (1987) J. Gen. Microbial. 133, 227-232. 14. Schulze, A. U., Schott, H. M., and Staudinger, Seyler’s 2. Physiol. Chen. 353, 1931-1942.

167,

Y., and Kitaoka, H. (1972) Hoppe-

15. Becana, M., and Klucas, R. V. (1990) Proc. N&l. Acad. Sci. USA 87,7295-7299. 16. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254. 17. Brewer, J. M., and Ashworth, R. B. (1969) J. C/rem. Educ. 46,4145. 18. Faeder, E. J., and Siegel, L. M. (1973) Anal. Biochem. 53, 332-336. 19. Kaplan, J.-C., and Beuter, E. (1967) Btichem. Biophys. Res. Commun. 29,605-610. 20. Bielski, B. H. J., Cornstock, D. A., and Bowen, R. A. (1971) J. Am. Chem. Sot. 93, 5624-5629. 21. Borraccino, G., Dipierro, 28,715-717.

S., and Arrigoni,

22. Ernster, L., Danielson, L., and Ljunggren, phys. Acta 58, 171-188.

0. (1989) Phytochemisty M. (1962) Biochim. Bio-

23. Yamauchi, N., Yamawaki, K., Ueda, Y., and Chachin, K. (1984) J. Jpn. Sot. Hortic. Sci. 53, 347-353.