Reconstitution of the Deinococcus radiodurans aposuperoxide dismutase

ARCHIVES

OF BIOCHEMISTRY

AND

BIOPHYSICS

Vol. 286, No. 1, April, pp. 257-263, 1991

Reconstitution Aposuperoxide

of the Deinococcus Dismutase

radiodurans

Jia-Ying

Juan, Scott N. Keeney,l and Eugene M. Gregory2 Department of Biochemistry, Virginia Polytechnic Institute and State University,

Blacksburg,

Virginia

24060

Received August 31, 1990, and in revised form October 23, 1990

Deinococcus radioduram, a radiation-resistant aerobe, synthesized a 43,000 M, dimeric superoxide dismutase. The holoenzyme, sp act 3300 U/mg, contained 1.5 g-atoms Mn, 0.6 g-atom Fe, and 0.1 g-atom Zn per mole dimer. Apoprotein, prepared by dialysis of the holoenzyme in denaturant plus chelator and then renatured in chelextreated Tris chloride buffer, rapidly regained superoxide dismuting activity upon incubation in 1 mM MnC12. Reconstitution was dependent on Mn concentration and pH. The Mn-reconstituted protein, sp act 3560 U/mg, contained 1.7 g-atoms Mn per mole dimer. The holoenzyme and Mn-reconstituted apoprotein migrated with the same patterns in 10% acrylamide gels and focused to the same pattern upon isoelectric focusing. Fluorescence emission maxima of the holoenzyme, Mn-reconstituted apoprotein, and the renatured apoprotein were 329 + 1 nm but differed from the denatured apoprotein (352 nm). Apoprotein bound 1.7 g-atoms Zn and from 3-7 g-atoms Fe per mole dimer on incubation with 1 mM ZnSOl and Fe(NH&(S04)2, respectively. Although neither Zn nor Fe restored superoxide dismuting activity, the ferrous and the zinc salt inhibited reconstitution of the apoprotein with manganese. Metal addition to renatured aposuperoxide dismutase offers a novel approach to reconstituo 1991 Academic tion of procaryote superoxide dismutases. Press.

Inc.

Superoxide dismutases are metalloproteins that catalyze the disproportionation of superoxide radicals to hydrogen peroxide and oxygen. These enzymes are part of the antioxidant defense system present in almost all cells (1). Three classes of superoxide dismutase (SOD)3 have 1 Present address: Department of Biochemistry, University of California, Berkeley, CA. ‘To whom correspondence should be addressed at Department of Biochemistry, 101 Engel Hall, Virginia Polytechnic Institute and State University, Blacksburg, VA 24060. FAX (703) 231-9070. a Abbreviation used: SOD, superoxide dismutase. 0003-9861/91$3.00 Copyright 0 1991 by Academic Press, All rights of reproduction in any form

been described based on the transition metals bound at the catalytic site. The SOD found predominantly in the cytosol of eucaryotes contains copper and zinc (Cu/ ZnSOD) (2). SODS from the cytosol of procaryotes contain either iron (FeSOD) or manganese (MnSOD) (3,4). There is a large degree of sequence homology among the Cu/Zn SODS and among Fe- and Mn-containing SODS but little homology between the Cu/Zn and the iron or manganese SODS (5). Moreover, the enzymes usually exhibit strict specificity for the metal bound at the active site (6). Ose and Fridovich (7) were first to demonstrate reversible removal and replacement of Mn from the Escherichia coli MnSOD. Denatured apoprotein was prepared in acidic guanidinium chloride. Catalytic activity and tightly bound Mn were restored upon addition of MnCl, and neutralization of the solution. These observations were extended to other Mn- and Fe-containing superoxide dismutases. In each case, the denatured apoSOD was reconstituted only upon addition of the intrinsic transition metal (5). Similar studies were applied to the superoxide dismutases from Bacteroides (8-10). Anaerobically maintained Bacteroides fragilis and Bacteroides thetaiotaomicron synthesize FeSOD. The iron-containing superoxide dismutases isolated from these organisms are less specific with respect to active site transition metal. The denatured apoproteins were reconstituted by dialysis in buffer containing either Fe or Mn. Exposure of the anaerobes to air induced MnSOD whose amino acid composition was virtually identical to that of the corresponding FeSOD. The MnSOD and FeSOD from each species comigrated in native acrylamide gels and had the same isoelectric point. Denatured apoproteins from the induced MnSODs were reconstituted with Fe. In each case cited above, reconstitution was carried out using the denatured apoprotein. During studies with the superoxide dismutase of Deinococcus radiodurans, a radiation-tolerant aerobe, we observed that aposuperoxide dismutase, freed of denaturant, rapidly regained superoxide dismuting activity on addition of manganese. This report details that observation. 257

Inc. reserved.

258

JUAN,

EXPERIMENTAL

KEENEY,

PROCEDURES

D. FCZ&O&LFU~ZS (ATCC 13939) was grown aerobically at 30°C in medium containing 0.5% peptone, 0.5% tryptone, 1% yeast extract, and 1% glucose supplemented with hemin (11). Cell were grown for 18 h, harvested by centrifugation, washed with 50 mM potassium phosphate, 1 mM EDTA (pH 7.8), and lyophilized. The lyophilized cells were stored at -20°C until used. Protein in crude extracts was estimated by absorbance at 280 nm corrected for interference by nucleic acids (12). Protein content of the purified SOD was measured based on an extinction coefficient of 1.8 ml mg-i cm-i at 280 nm. The value was calculated from the absorbance at 280 nm of a solution whose protein content had been determined by quantitative amino acid analysis. Superoxide dismutase activity was measured and units were defined as described by McCord and Fridovich (2). Proteins were separated in 0.75-mm gels containing 10% acrylamide (13) and were stained with Coomassie blue. Superoxide dismutase activity was visualized in gels by a modification of the Beauchamp and Fridovich method (14). Proteins were focused to their isoelectric point in acrylamide gels containing the appropriate ampholytes. The pH gradient was calibrated with proteins and dyes of known isoelectric points. The purified holoenzyme was brought to sedimentation equilibrium in an analytical ultracentrifuge and the molecular weight determined using a partial specific volume of 0.721 ml g-i, calculated from the amino acid composition. Molecular weight of the renatured apoprotein and the Mnreconstituted SOD was estimated by gel exclusion chromatography on a Bio-Rad Superose 6 FPLC column calibrated with appropriate molecular weight standards. Subunit molecular weight was determined as described by Laemmli (15). The spectra of the proteins in 50 mM potassium phosphate (pH 7.0) were obtained with a Shimadzu UV-265 spectrophotometer. Amino acid compositional analysis of the protein was determined on a Waters PicoTag amino acid analysis system. Tryptophan content was measured on protein hydrolyzed in 4 M methanesulfonic acid (performed at the Department of Biochemistry and Biophysics, Virginia Commonwealth University) and was also verified as described by Edelhoch (16). Apoprotein was prepared by dialysis overnight against 5.0 M guanidinium chloride, 20 mM 8-hydroxyquinoline, pH 3.5 (7). Chelator was removed by dialysis in 5.0 M guanidinium chloride (two changes, 500 ml). The denatured apoprotein was subsequently freed of guanidinium chloride by dialysis in 20 mM Tris chloride, pH 7.5 (two changes, 4 liters). The renatured apoprotein was concentrated to approx 0.7 mg/ml and stored at -20°C in O.l-ml aliquots in acid-washed plastic tubes. Tris chloride buffer stock (1 M) was passed twice through Chelex 100 to diminish trace metal content. Renatured apoprotein was diluted into 20 mM Tris chloride (pH 7.5) and reconstitution initiated by the addition of 10 mM MnCl, in water. Final concentration of protein was typically 75-80 pg per milliliter and MnCl, was 1 mM. In some studies, the Mn concentration was varied by using an appropriately diluted MnCl, solution. Reconstitution was terminated by diluting aliquots of the incubation mixture into 50 mM potassium phosphate, 2 mM EDTA (pH 7.8). Plasticware used in the reconstitution studies was acid-washed and rinsed in distilled deionized water. Metal content of the buffers and protein samples was determined on a PerkinElmer Model 560 atomic absorption spectrophotometer by aspiration into an acetylene flame. Fluorescence emission spectra were obtained with a Perkin-Elmer Model 650-40 fluorescence spectrophotometer.

RESULTS

Isolation

of D. radiodurans

Superoxide Dismutase

Eight grams of lyophilized cells was rehydrated with 200 ml of 50 mM potassium phosphate, 1 mM EDTA (pH 7.8) in a blender and disrupted by 3-s bursts (70 W) of sonic oscillation applied through the microtip. Total sonication time was 30 min per 100 ml of suspension. This and subsequent steps were at 4°C. The suspension was

AND

GREGORY

clarified by centrifugation (23,OOOg,30 min), stirred with 0.2% protamine sulfate for 45 min, and centrifuged at 23,000g for 15 min. The supernatant was brought to 50% of saturation by addition of solid ammonium sulfate (313 g per liter) and stirred for 30 min, and the mixture was clarified by centrifugation. The supernatant was brought to 80% of saturation by addition of solid ammonium sulfate (214 g per liter) and was stirred 30 min. The pellet, obtained by centrifugation, was dissolved in 35-50 ml of 5 mM potassium phosphate, 1 mM EDTA (pH 7.0) and was dialyzed in two 4-liter changes of the same buffer. The dialysate was applied to DE-53 (1.5 X 19 cm) equilibrated in 5 mM potassium phosphate (pH 7.0) and was eluted with a linear KC1 gradient in phosphate buffer (O200 mM KCl). Fractions (4.8 ml) with superoxide dismutase activity greater than 350 U/ml were pooled and dialyzed in the phosphate buffer for rechromatography on DE-53 under conditions identical to those described above. Fractions with superoxide dismutase activity greater than 1200 U/mg were concentrated to approximately 2 ml over a YM-5 ultrafilter (Amicon) under N2 pressure. The protein was applied to a Bio-Gel P-100 column equilibrated in 50 mM phosphate (pH 7.0) and was eluted in this buffer. The fractions (2 ml) containing superoxide dismutase activity were pooled and concentrated over a YM-5 membrane. The superoxide dismutase activity from D. radiodurans was purified 31O-fold to a specific activity of 3300 U/mg with a 46% overall yield (Table I). This purification scheme consistently yielded enzyme with specific activity 3300 f 480 U/mg. Characterization

of the D. radiodurans

SOD

On nondenaturing polyacrylamide gels, the isolated superoxide dismutase migrated as a major band and two faint minor bands of protein (3.2 pg applied) (Fig. lA, Lane 1). A separate gel stained for SOD activity (1 pg applied) revealed that the most intense activity migrated coincidentally with the major protein band (Fig. lB, Lane 1). The minor protein bands each displayed a faint but discernible activity. The denatured protein migrated as a single peptide in acrylamide gels containing sodium dodecyl sulfate, whether or not 2-mercaptoethanol was present, and comigrated with the subunit of the E. coli MnSOD [22,900 Da (17)]. The molecular weight of the native protein was 42,300 determined by sedimentation equilibrium. Molecular weights of the native enzyme, renatured apoprotein, and the Mn-reconstituted SOD were estimated by gel exclusion chromatography to be 43,000. Isoelectricfocusing of the SOD revealed two protein bands, each of which had superoxide dismutase activity. The major band focused to pI 4.0 and the minor band to pI 4.7 (Fig. 2, Lane 2). The spectrum of the enzyme (Fig. 3, solid line) shows a broad band between 400 and 600 nm with a distinct peak at 415 nm and an absorbance at 280 nm with a shoulder at 288 nm. The extinction at 280 nm,

D. radioduram

APOSUPEROXIDE TABLE

Purification

Step Crude extract Protamine sulfate 50% (NH&SO, 80% (NH&SO4 (dialyzed pool) First DE 53, pool Second DE 53, pool P-100

of

I

D. rudiodurans SOD

Volume (ml)

Units

Protein (md

180 200 230

23,800 19,200 22,200

2540 1530 1290

68 55 28 16

18,300 16,600 13,300 10,900

based on quantitative amino acid composition, was 1.8 ml mg-’ cm- ‘. The amino acid composition is shown in Table II. The SOD contained 1.5 g-atoms manganese per mole dimer and 0.5-0.8 g-atom iron per mole dimer but zinc content was less than 0.1 g-atom per mole (Table 3). NaNs (10 mM) in the assay mixture inhibited the superoxide dismutase activity by 50% (data not shown). Characterization

259

DISMUTASE

of the Apoprotein

Renatured apoprotein, freed of denaturant, had no more than 2% of the catalytic activity of the native enzyme but retained 0.3 g-atom Mn, 0.4 g-atom Fe, and 0.05 g-atom Zn per mole dimer (Table III). The apoprotein retained the absorbances at 280 and 288 nm but lost the broad

90.7 10.9 5.1 3.3

Specific activity (U/m) 9.4 12.6 17.3 202 1520 2600 3300

Fold

% Yield 100

1.3 1.8

93.2

21.5 162 278 351

76.9 69.7 55.8 45.5

absorbance feature at 400-600 nm. The peak at 415 nm was markedly diminished (Fig. 3, dotted line). The apoprotein migrated to the same position and gave the same pattern of electromorphs as the holoenzyme (Fig. lA, Lane 2) and yielded a small superoxide dismutase activity band (Fig. lB, Lane 2). The isoelectric pattern of the apoprotein is shown (Fig. 2, Lane 3). Reconstitution Dismutase

of D. radiodurans

Aposuperoxide

The renatured apoprotein was diluted in 20 mM Tris chloride (pH 7.5) and reconstitution was initiated by the addition of 10 mM MnCl, in distilled water (final MnCl, concentration, 1 mM). Samples of the reconstitution mixture were withdrawn periodically, diluted into 50 mM po-

A

FIG. 1. Electropherograms of the D. radiodurans native, apo-, and Mn-reconstituted superoxide dismutase. (A) Protein samples were applied to 0.75.mm gels containing 10% acrylamide, separated electrophoretically, and stained with Coomassie blue G-250. Lane 1, purified MnSOD, 3.2 pg. Lane 2, apoprotein, 4.0 fig applied. Lane 3, Mn-reconstituted apoprotein, 5 fig applied. Panel (B) Protein samples were separated as described above but were stained for superoxide dismutase activity by the method of Beauchamp and Fridovich (14). Lane 1, purified MnSOD, 3.2 U, 1.0 @Lg.Lane 2, apoprotein, 1.1 kg. Lane 3, Mn-reconstituted apoprotein, 3.6 U, 1.1 pg.

260

JUAN,

KEENEY,

AND

GREGORY .5

.OS

.4

.04

s x

.3

2$

.2

.02

.I

.Ol

~

200

.03

300

350

400 450 Wavelength

500 (nd

550

600

s f$ ;: 8

658

FIG. 3. UV-visible spectra of D. radiodurans native and aposuperoxide dismutases. Uv and visible spectra were recorded on a Shimadzu UV265 spectrophotometer. The native MnSOD and apoprotein were prepared as described in the text. Protein concentrations of the native MnSOD were 0.98 and 0.24 mg/ml for the visible and ultraviolet spectrum, respectively. The apoprotein concentrations were 1.1 (visible) and 0.26 mg/ml (uv). Native MnSOD, solid trace; apoprotein, dotted trace.

FIG. 2. Isoelectric focusing of the D. radiodurans superoxide dismutase. Protein samples were diluted 1:l in application solution (4% ampholytes, pH 3-10,60% glycerol, and patent blue V marker dye) and were loaded onto gels containing 5.5% acrylamide and the pH 3-10 ampholytes. The pH gradient was determined with a pZ marker protein kit obtained from Pharmacia. After the methyl red and patent blue V markers had separated at the bottom of the gel, the gel was washed consecutively in 10% trichloroacetic acid and in 25% ethanol, 10% acetic acid through several repetitions and then stained with Coomassie blue G-250. The marker dyes were removed during the staining procedure, but their positions are noted by the arrows. Lane 1, pZ markers. Lane 2, purified MnSOD, 3.2 pg. Lane 3, apoprotein, 4.0 wg. Lane 4, Mnreconstituted apoprotein, 4.0 pg.

At pH 8.0, the apoprotein was reconstituted to 1400 U/ mg with 1 j&M MnClz and to full catalytic activity with 100 I.LM MnC& (Fig. 5). Mn content of the SOD reconstituted at pH 8.0 with 1 mM MnCl, was 1.7 g-atoms per mole dimer. At pH 8.5, catalytic activity was fully restored in 10 min, but a pink precipitate developed during the incubation. Reconstitution at this pH was not further explored. Zinc inhibited the reconstitution of denatured aposuperoxide dismutase prepared from B. fragilis or B. the-

TABLE

tassium phosphate, 2 mM EDTA (pH 7.8), and assayed for superoxide dismutase activity (Fig. 4). The EDTA, present in the buffer, completely suppressed any superoxide scavenging activity caused by free metal. The enzyme was reconstituted with 0.6 g-atom Mn per mole dimer to a specific activity of approximately 2100 U/mg after 1 h incubation. Incubation of the renatured apoprotein for 3 h in 1 mM MnClz at pH 7.5 restored 100% of the initial specific activity (3560 f 100 U/mg, 10 trials) with the binding of 1.7 g-atoms Mn per mole dimer (Table III). Electrophoretic separation of the reconstituted protein revealed a protein and enzymatic activity pattern virtually identical to the holoenzyme (Figs. lA, lB, Lane 3). Two electromorphs were also observed upon isoelectric focusing (Fig. 2, Lane 4). In the reconstituted enzyme, each of the electromorphs was enzymatically active (data not shown). Incubation of apoprotein with 1 PM MnC& restored catalytic activity but not as effectively as 1 mM MnClz. With 1 PM MnC12, the specific activity increased to approximately 500 U/mg after 60 min incubation and remained unchanged after 3 h. The rate of apoprotein reconstitution with MnC12 was pH-dependent. Maximum specific activity was achieved in 3 h at pH 7.5 (Fig. 4), and in 20 min at pH 8.0 (Fig. 5).

Amino Amino acid CM-cys’ As(n) Thr Ser Glfnl Glr Ala Val Met Ile Leu Tyr Phe His LYS

Arg Pro Trp

Acid

Composition

II of D. radiodurans

Amino acid content” 0 48.9 14.7 14.9 36.7 33.7 47.6 22.3 5.2 14.2 32.1 14.2 16.3 14.1 26.7 16.2 16.6 10.6d 10.ge

SOD Residues/m01 b 0 49 15 15 37 34 48 22 5 14 32 14 16 14 27 16 17 11

a Moles per mole enzyme based on 43,000 M, (average of six analyses). b Composition given as nearest integer. ’ Determined on reduced, carboxymethylated protein. ‘Triplicate analysis, protein hydrolyzed in 4 M methanesulfonic acid. e Duplicate analysis, Edelhoch method (16).

D. radiodurans

APOSUPEROXIDE TABLE

261

DISMUTASE

III

Metal Content of D. rudioduruns SOD g-atoms per mole dimer Specific activity (U/w3

Treatment Native Apoprotein Mn-reconst.itution Mn-reconstitution Zn-reconstitution Fe-reconstitution

(1 h) (3 h) (1 h)

3300 50 2060 3560 50 50

f 484” + 42 + 113” 31 106 f 42 f 42

Mn

Fe

Zn

1.50 i 0.03 0.25 IL 0.16 0.60 f 0.1 1.70 f 0.08 n.d. 0.3 + 0.08

0.64 f 0.13 0.41 + 0.12 0.62 rtr 0.14 n.d. n.d. 5.2 k 2’

0.10 +- 0.025 0.05 f 0.02

Note. Data are means f SD. n.d. not determined. ’ Five samples, all others, three samples. These data were measured on protein

taiotaomicron SOD (g-10). We tested the effect of Zn and Fe, alone or in combination with Mn, on reconstitution of the D. radiodurans aposuperoxide dismutase. The aposuperoxide dismutase was incubated at pH 7.5 with 1 mM MnCl* and either 5 or 50 PM ZnSOl (Fig. 4). ZnSOl (50 pM) inhibited reconstitution of enzymatic activity by 91% whereas 5 PM ZnS04 inhibited reconstitution by 56%. Zn (1.7 g-atoms per mole dimer) was tightly bound to the protein (Table III). Iron, added as ferrous ammonium sulfate, did not reconstitute enzymatic activity but inhibited reconstitution with Mn (Fig. 4). Incubation of the apoprotein with 1 mM iron salt plus 1 mM MnClz in 20 mM Tris chloride (pH 7.5) yielded superoxide dismutase whose specific activity varied, in five separate experiments, from 1200 to 2200 U/mg. Aposuperoxide dismutase (0.43 mg/ml), incubated with 1 mM ferrous ammonium sulfate, was devoid of superoxide dismutase activity but contained 5.2 + 2 g-atoms Fe per mole dimer (Table III). The fluorescence emission maximum of the holoenzyme, renatured apoprotein, and the Mn-reconstituted apoprotein was 329 * 1 nm (Fig. 6). The fluorescence emission maximum of apoprotein in 5 M guanidinium chloride was 352 nm. The renatured apoprotein was reconstitutable with Mn to a specific activity of 3500 U/mg immediately following measurement of the fluorescence spectrum. These data are consistent with the absence of metal at the catalytic site of the apoSOD used to measure the fluorescence spectrum. DISCUSSION Reconstitution of apoprotein with metal is one criterion of metal specificity and function in metalloproteins. One approach to reconstitution of procaryotic superoxide dismutases requires denaturation to remove the intrinsic metal and then addition of various metals with simultaneous removal of denaturing conditions. Reconstitution of the denatured apoprotein to 7-90% of initial specific activity has been reported for the following bacterial en-

reconstituted

in 20

mM

n.d. 1.7 + 0.04 0.08 + 0.03

Tris (pH 7.5).

zymes: E. coli MnSOD (6, 7), B. frugilis FeSOD (8), and B. thetuiotaomicron FeSOD (10). In contrast, the apoprotein from D. radiodurans MnSOD, in the absence of denaturant, bound Mn with full restoration of catalytic activity. Addition of 1 mM MnC& to the renatured apoprotein in Tris chloride buffer at pH 7.5 or 8.0 restored catalytic activity (3500 U/mg) and protein-bound Mn (1.7 g-atoms per mole dimer) to approximately the same levels found in the native holoenzyme. The rate of reconstitution was greater at pH 8.0 than at pH 7.5, perhaps due to titration of critical residues at the metal binding sites. Zn and Fe each inhibited the Mn-dependent reconstitution of the apoprotein but were bound with different stoichiometries. Zn was bound to the renatured apoprotein in approximately 2:l mole per mole dimer whereas Fe binding varied from 3 to 7 g-atoms per mole dimer. Neither Fe nor Zn imparted catalytic activity to the apoprotein. The variability of Fe binding to apoprotein was reflected in the competition between Fe and Mn for apoprotein. Specific activity of the SOD incubated with equimolar Fe and Mn varied, in five separate experiments, from 1200 to 2200 U/mg (Fig. 4). The variability may result from the oxidation of iron in the reconstitution mixture. Significantly less variability was noted upon reconstitution with Mn in competition with 5 pM ZnSOl (five experiments) or 50 yM ZnSOl (three experiments) (Fig. 4). The native MnSOD, apoSOD, and Mn-reconstituted SOD migrated identically on acrylamide gels and focused to the same isoelectric pattern. The intrinsic tryptophan fluorescence emission of the holo-, apo-, and Mn-reconstituted SODS occurred at the same wavelength whereas the fluorescence emission of the denatured apoprotein differed markedly. These data are consistent with a model in which the apoprotein, freed of denaturant, assumes a gross conformation which readily accepts Mn with restoration of catalytic activity. The Mn-reconstituted SOD, like the isolated enzyme, is stable to dialysis against EDTA-containing buffers. Binding of Mn at the active

262

JUAN,

0

KEENEY,

AND

GREGORY

100

50

Time

150

200

(min)

FIG. 4. Reconstitution of D. radiodurans aposuperoxide dismutase. Apoprotein, prepared as described in the text, was incubated at 23°C with the metals indicated. Aliquots of the incubation mixture were removed periodically, diluted into an equal volume of 50 mM potassium phosphate, was incubated in 20 mM 2 mM EDTA (pH 7.8), and assayed for superoxide dismutase activity. The apoprotein (0.075 mg/ml, final concentration) Tris chloride with 1 mM MnCl, (0); 1 mM MnC& plus 1 mM Fe(NH&(SO,), (V); 1 mM MnCl, plus pM ZnSOd (#); 1 mM MnClr plus 50 pM ZnSO, (X). The data shown are averages of five experiments except for MnClx plus 50 ~.LM ZnSO, (three experiments). The specific activity after 3 h incubation was (mean plus range) 1 mM MnCl,, 3560 + 106 U/mg; 1 mM MnCls plus 1 mM Fe(NH,),(SO&, 1700 + 520 U/mg; 1 mM MnCl, plus 5 pM ZnSO,, 1560 f 200 U/mg; 1 mM MnCl, plus 50 pM ZnSO,, 315 + 45 U/mg.

site may induce local conformational rearrangements that cause the metal to be more tightly bound. Binding of Fe and Zn, and their competition with Mn, is accommodated by the model as follows. Mn, Zn, and Fe may each compete directly for the active site on the apoprotein. Binding of the catalytically inactive Fe or Zn prevents binding of Mn. Alternatively, each metal may bind to a unique site on the apoprotein and preclude binding of the other metals. The model is similar to that we proposed to explain Fe, Mn, and Zn binding to the denatured B. fragilis superoxide dismutase apoprotein U3,9). Preparation and reconstitution of aposuperoxide dismutases have been studied using other approaches. Yamakura (18) prepared aposuperoxide dismutase from Pseudomonas ovalis FeSOD using 2 mM EDTA and 10 mM dithiothreitol in sodium carbonate (pH 11). Iron, added to the apoenzyme in the alkaline thiol-containing buffer, restored both iron and catalytic activity. In buffer at pH 7.8, the apoenzyme was not catalytically reactivated upon addition of the iron salt. Meier et al. (19) used ascorbate and o-phenanthroline at pH 5.5 to remove metal from Propionibacterium shermanii MnSOD. The apoprotein, which retained 34% of the initial catalytic activity, was further activated upon reconstitution with either Fe or Mn at pH 5.5. Dialysis

of the apoprotein at pH 5.5 for longer than 24 h reportedly caused precipitation of the protein. Martin et al. (20) used that approach to reconstitute Streptococcus mutans MnSOD. Apparently the apoprotein was not reconstituted in either case under conditions other than those used to form apoprotein. Procaryotic superoxide dismutases may now be divided into two classes: enzymes having an absolute metal ion

0

10

20

30

40

50

60

Time (min)

FIG. 5. Reconstitution of aposuperoxide dismutase at pH 8. Renatured aposuperoxide dismutase (0.075 mg/ml) was incubated with MnClx at the indicated concentration. Aliquots were assayed as described in the text. MnClx concentrations were 1 pM (0); 10 PM (0); 0.1 mM (A); 1 mM (A).

D. rudioduruns

APOSUPEROXIDE

263

DISMUTASE

ACKNOWLEDGMENTS This research was supported in part by Grants AI 15250 from the National Institutes of Health and J-104 from the Jeffress Trust. We thank R. E. Ebel and W. G. Niehaus for critical review and W. F. Beyer, Jr., for sedimentation equilibrium analysis of the protein.

REFERENCES Wavelength

(nm)

1. Fridovich, 2. McCord,

FIG. 6. Fluorescence emission spectra of D. radioduruns SOD. Native MnSOD (0.21 mg/ml) and Mn-reconstituted apoSOD (0.31 mg/ml), each in 50 mM potassium phosphate, 1 mM EDTA (pH 7.8) renatured apoprotein (0.2 mg/ml in 10 mM Tris chloride, pH 7.5) were excited at 292 nm. The emission spectra for each protein varied 1 nm from the maximum at 329 nm (dashed line). ApoSOD (1.5 Azso absorbance units) in 5 M guanidinium chloride (Sequanal grade, Pierce Chemical Co.) was measured using the guanidinium chloride to determine background fluorescence (dotted line). The recorder was adjusted to 70% of full scale at the emission maxima for each sample. Excitation spectrum, solid line.

specificity and those active with either Fe or Mn. Among the latter are the FeSODs from the anaerobes B. fragilis, B. thetaiotaomicron, and P. shermanii (8-10, 19) and the MnSOD from S. mutans (20). Apoprotein from each of these enzymes was reactivated upon the addition of either iron or manganese, a property termed “cambialistic” by Martin et al. (20). On the other hand, the renatured apoprotein from D. radiodurans was not active upon binding Fe. To determine if renaturing the apoprotein artifactually precluded Fe-dependent activity, we reconstituted the denatured D. radiodurans apoprotein with manganous chloride or ferrous ammonium sulfate. Although we recovered at least 75% of the original enzymatic activity upon reconstitution with Mn, less than 0.3% of the original activity was recovered upon reconstitution with 1 mM ferrous ammonium sulfate (data not shown). Experiments underway with organisms closely related to D. radiodurans may reveal other native aposuperoxide dismutases which will accept metal with equal facility.

I. (1989) J. Biol. Chm.

264, 7761-7764.

J. M., and Fridovich,

I. (1969) J. Biol. Chem. 244, 6049-

3. Yost, F. J., Jr., and Fridovich,

I. (1973) J. Biol. Chem. 248, 4905-

6055. 4908. 4. Keele, B. B., Jr., McCord, Chem. 245,6176-6181. 5. Fridovich,

J. M., and Fridovich,

I. (1970) J. Biol.

I. (1986) A&. Enzymol. 58, 61-97.

6. Kirby, T., Blum, J., Kahane, I., and Fridovich, Biochem. Biophys. 201,551-555.

I. (1980) Arch.

7. Ose, D. E., and Fridovich, I. (1979) Arch. Biochem. Biophys. 194, 360-364. 8. Gregory, E. M., and Dapper, C. H. (1980) Arch. Biochem. Biophys. 220,293-300. 9. Gregory, E. M. (1985) Arch. Biochem. Biophys. 238,83-89. 10. Pennington, C. D., and Gregory, E. M. (1986) J. Bucteriol. 166, 528-532. 11. Holdeman, L. V., Cato, E. P., and Moore, W. E. C. (1977) Anaerobe Laboratory Manual, 4th ed., Anaerobe Laboratory, Virginia Polytechnic Institute and State University, Blacksburg, VA. 12. Warburg,

O., and Christian,

W. (1941). Biochem. Z. 310, 384-421.

13. Davis, B. J. (1964) Ann. N.Y. Acud. Sci. 121, 404-427. 14. Beauchamp, C. O., and Fridovich,

I. (1971) Anal. Biochem. 44,276-

287. 15. Laemmli,

U. K. (1970) Nature (London)

16. Edelhoch, H. (1967) Biochemistry 17. Steinman, 18. Yamakura,

227,680-685.

6,1948-1954.

H. M. (1978) J. Biol. Chem. 253,8708-8720. F. (1978) J. Biochem. 83,849-857.

19. Meier, B., Barra, D., Bossa, F., Calabrese, L., and Rotilio, J. Biol. Chem. 257, 13,977-13,980.

G. (1982)

20. Martin, M. E., Byers, R. B., Olson, M. 0. J., Salin, M. L., Arceneaux, J. E. L., and Tolbert, C. (1986) J. Biol. Chem. 261,9361-9367.