The Activity of the Enzyme Was Hardly Affected by Metal-Complexing

J. Biochem., 79, 661-671 (1976)

Properties of Purified Hydrogenase from the Particulate Fraction

of Desulfovibrio ƒÊulgaris, Miyazaki1

Tatsuhiko YAGI,* Keisaku KIMURA ,** Hidehiro DAIDOJI,*** Fumiko SAKAI,*** Shohei TAMURA ,*** and Hiroo INOKUCHI** *Department of Chemistry , Shizuoka University, Oya-836, Shizuoka 422, **Institute for Molecular Science , Okazaki 444, and ***The Institute for Solid State Physics, the University of Tokyo , Roppongi, Minato-ku, Tokyo 106

Received for publication, September 30 , 1975

The properties of purified hydrogenase [EC 1.12.2.1] solubilized from particulate

fraction of sonicated Desulfovibrio ƒÒulgaris cells are described . The enzyme was a brownish iron-sulfur protein of molecular weight 89,000, composed of two different subunits (mol. wt.: 28,000 and 59,000), and it contained 7-9 iron atoms and 7-8 labile sulfide ions. Molybdenum was not detected in the preparation. The absorp tion spectrum of the enzyme was characteristic of iron-sulfur proteins. The milli molar absorbance coefficients of the enzyme were about 164 at 280 nm, and 47 at 400 nm. The absorption spectrum of the enzyme in the visible region changed upon incubating the enzyme under H2 in the presence of cytochrome c3 , but not in its absence. This spectral change was due to the reduction of the enzyme. The absorbance ratio at 400 nm of the reduced and the oxidized forms of the enzyme was 0.66. The activity of the enzyme was hardly affected by metal-complexing agents such as cyanide, azide, 1, 10-phenanthroline, etc., except for CO, which was a strong inhibitor of the enzyme. The activity was inhibited by SH-reagents such as p- chloromercuribenzenesulfonate. The enzyme was significantly resistant to urea, but susceptible to sodium dodecyl sulfate. These properties were very similar to those of clostridial hydrogenase [EC 1.12.7.1], in spite of differences in the acceptor specificity and subunit structure.

Hydrogenases are bacterial enzymes which fering specificities for the electron acceptors, participate in the production or consumption NAD+ (1, 2), cytochrome c3 (3, 4), and fer of H2 in bacterial metabolism. Three kinds redoxin (5, 6), though still another hydro of hydrogenases have been reported, with dif- genase of unknown specificity (7) is known. We have reported solubilization and puri 1 This study was supported in part by a grant (No. fication procedures for the particulate hydro 88024, 1971) from the Ministry of Education, Science genase of Desulfovibrio vulgaris, Miyazaki [EC and Culture, of Japan. 1.12.2.1], which is specific to cytochrome c3 Abbreviation : SDS, sodium dodecyl sulfate. (4), and derived some kinetic constants for

Vol. 79, No. 3, 1976 661 662 T. YAGI, K. KIMURA, H. DAIDOJI, F. SAKAI, S. TAMURA, and H. INOKUCHI the paraH2-orthoH2 conversion and isotope ex- tained was nearly homogeneous as determined change reactions catalyzed by the purified hy by disc electrophoresis (10) at pH 7.3 (see drogenase (8). Fig. 2). The specific activity of this prepa In the present communication, some prop ration was about 330 units/A28o nm, or 610 erties of this enzyme and the effects of some units/mg. metabolic inhibitors on the enzyme are re- Assay of Protein and Intensity of the ported. Brown Color-The concentration of protein was expressed in terms of the absorbance at

MATERIALS AND METHODS 280 nm (A280 nm). The intensity of the brown color was expressed in terms of the absorb Chemicals and Reagents - Trypsin [EC ance at 400 nm (A4oo nm). Desulfoviridin and 3.4.21.41, twice recrystallized, was a product cytochromes which strongly absorb light at of Worthington. Cytochrome c3 was purified 400 nm were eliminated in the early stages as described by Yagi and Maruyama (9 ). Its of the purification and did not interfere with molar concentration was expressed on a pro the assay of the intensity of the brown color. tein basis instead of a heme basis. Methyl Assay of Hydrogenase-Activity of hydro

viologen was a product of BDH. genase was assayed either by the H2-evolution Hydrogenase [H2 : ferricytochrome c3 oxi technique (4) or by the enzymic electric cell doreductase, EC 1.12.2.11 was purified from method (11, 12) recently developed in our the cells of Desulfovibrio vulgaris, Miyazaki, laboratories. by the following procedure, which is a modi Molecular Weight-The molecular weight fication of that reported in our previous paper and partial specific volume of the enzyme pro

(4 ). The particulate fraction of the bacterial tein were kindly determined by Prof. N. Ui sonicate was digested with trypsin (1 g for of Gunma University by the low-speed sedi 400 ml of suspension containing 100 g wet mentation equilibrium method (13) with a

precipitate) at 4? for 20 hr with stirring under photoelectric scanner (14 ). The molecular N2. The solubilized hydrogenase was then weight of the enzyme subunit was determined

precipitated with ammonium sulfate at 70% by electrophoresis on polyacrylamide gel in saturation, and the precipitate was dissolved the presence of sodium dodecyl sulfate (SDS) in H2O, concentrated by ultrafiltration using and 2-mercaptoethanol as described by Weber a Diaflo cell (Aminco Corporation) with a PM- and Osborn (15). 30 membrane, and chromatographed on a Se Amino Acid Analysis-The protein sam

phadex G-150 column (2.7•~72 cm) with 0.02 ple was dried in a stream of N2 and hydro M Tris-HC1 (pH 7.3) containing 0.08 M NaCl lyzed with 6 M HC1 in evacuated glass tubes as an eluting buffer. The effluent fractions at 110? for 24 hr or 48 hr. The hydrolysates containing hydrogenase activity were collected, were analyzed for amino acid contents with a concentrated by means of a Diaflo cell with a Hitachi KLA-3B, or JEOL automatic amino PM-30 membrane, and charged onto a column acid analyzer by the procedure described by

(1.3•~12 cm) of DEAE-Sephadex A-50 previous Spackman et al. (16). Tryptophan content ly equilibrated with the above buffer. Then was determined spectrophotometrically (17 ). the enzyme was eluted from the column by Metal Analysis-Metal contents were de the concentration gradient technique starting termined by atomic absorption spectrometry. with the above buffer and ending with 0.05 M Sulfur Analysis-The sulfur atoms in the Tris-HCl (pH 7.3) containing 0.2 M NaCl. The enzyme were converted into H2S by the tin(II)- active fractions of the effluent were concen strong phosphoric acid reduction method de trated by partial lyophilization, chromato scribed in this paper, and the H2S thus ob

graphed again on a column of Sephadex G-150, tained was led into an absorbent solution con and the resulting active brownish fractions taining cadmium acetate by passing N2. It were collected (see Fig. 1), dialyzed thorough- was determined by the colorimetric method ly, and lyophilized. The preparation thus ob- using N, N•Œ-dimethyl-p-phenylenediamine (18).

J. Biochem. PROPERTIES OF DESULFOVIBRIO HYDROGENASE 663

The assay method for labile sulfide is de enzyme preparation (Fig. 3). The molecular scribed later in the text. weight of the larger component was estimated Electrofocusing-This was carried out as to be 59,000, and that of the smaller compo described in the instruction manual for LKB nent, 28,000 (Fig. 4). These two components 8100 Ampholine, with 0.8% carrier ampholite were always detected upon electrophoresis (pH 5-7) on an LKB 8101 electrofocusing col in the presence of SDS, even in the absence umn. of 2-mercaptoethanol. Spectral Properties-The absorption spec

RESULTS trum of the purified hydrogenase was shown in Fig. 5, Curve 1. The specific absorbance Molecular Weight of the Hydrogenase- coefficients were 1.84 at 280 nm, and 0.53 at Low-speed sedimentation equilibrium analysis 400 nm. This means that the millimolar ab

of the hydrogenase preparation revealed linear sorbance coefficients of the enzyme were about In c : ƒÁ2 relations (13) in both H20 and 90% D20 media containing phosphate-NaC1 buffer (pH 6.5, p=0.2). This indicates that the en zyme is homogeneous, and from the slopes of the plots (i.e., d In c/d ƒÁ2), the partial specific volume and the molecular weight of the en zyme protein were calculated to be 0.751, and 89,000, respectively. Electrophoresis on polyacrylamide gel in the presence of SDS and 2-mercaptoethanol revealed the presence of two subunits in the

Fig. 2. Disc electrophoretic patterns of hydrogenase.

Three tubes, A, B, and C (5 x 50 mM), of polyacryl amide gel were prepared. In tube A, hydrogenase

(52 ƒÊg) was electrophoresed for 50 min at 130 volts, then the gel was stained with Amido Black solution in 7% acetic acid for 20 min and destained by

washing with 7% acetic acid. In tubes B and C, hydrogenase (100 ƒÊg per tube) was electrophoresed under the same conditions. The gel was removed

from tube C, immersed in 0.6 mM methylviologen in Fig. 1. Elution profile of the hydrogenase from a 0.02 M phosphate buffer (pH 6.9) which had been Sephadex G-150 column. The hydrogenase prepara saturated with Ha• When the blue color developed tion (5 ml) was chromatographed on a column (2.7 x as a result of enzymatic reduction of methylviologen, 72 cm) of Sephadex G-150. The elution buffer was the gel was quickly removed from the solution and 0.02 M Tris-HC1 (pH 7.3) containing 0.08 M NaC1. •œ a photograph was taken immediately. The gel in , concentration of brown pigment expressed in tube B was removed from the tube and a photo terms of absorbance at 400 nm ; O , protein concen graph was taken immediately. The mobility of the tration expressed in terms of absorbance at 280 nm ; brownish pigment in tube B was identical to that and 4, the activity of hydrogenase assayed by the of hydrogenase activity in tube C, and this was the enzymic electric cell method expressed as unit•m1-1 only area stained by Amido Black in tube A. of the effluent.

Vol. 79, No. 3, 1976 664 T. YAGI, K. KIMURA, H. DAIDOJI, F. SAKAI, S. TAMURA, and H. INOKUCHI

164 at 280 nm, and 47 at 400 nm. The ab The brownish proteinaceous component sorbance ratio, A400nm/A280 nm, of the most was not separable from the hydrogenase ac highly purified hydrogenase preparation was tivity either by Sephadex G-150 column chro 0.29. matography (Fig. 1), or by disc electrophoresis

Fig. 4. Relative mobility of hydrogenase compared with reference proteins upon SDS-gel electrophoresis

in the presence of 2-mercaptoethanol. The concen

tration of the crosslinker was half of that used in Fig. 3. SDS-gel electrophoretic pattern of hydro the standard conditions described by Weber and genase. The SDS-gel electrophoretic pattern of Osborn (15). •ü: marker proteins, i.e., a, alcohol hydrogenase (8 mg) was obtained by the procedure dehydrogenase [EC 1. 1. 1. 1] (subunit weight=37,000) ; of Weber and Osborn (15) in the presence of 2- b, bovine serum albumin (mol. wt. =68,000) ; g, glu- mercaptoethanol. Omission of 2-mercaptoethanol tamate dehydrogenase [EC 1.4.1.3] (mol. wt.= from the procedure did not affect the results. 53,000), and •œ : hydrogenase subunits.

Fig. 5. Absorption spectra of hydrogenase. Curve 1: Hydrogenase (0.59 mg.ml-1 in 20 mM phosphate buffer, pH 7.0). No spectral change was observed upon incubating the solution under an atmosphere of H2 (36 Torr) at room temperature for 36 hr. Curve 2: Hydrogenase (0.59 mg.ml-1 in 20 mM phosphate buffer, pH 7.0) containing 6 nM cytochrome c2 , incubated for 12 hr under an atmosphere of H2 (36 Torr) at room temperature. The spectrum of the Na2S2O4-reduced hydrogenase was identical with Curve 2 in the visible region, and that of the reoxidized enzyme was similar to Curve 1.

. Biochem.

J PROPERTIES OF DESULFOVIBRIO HYDROGENASE 665 ,

on polyacrylamide gel at pH 8.3, as, shown in 0.58% (9.2 moles Fe/89 kg protein) , while that Fig. 2. Upon electrofocusing, both the activ of enzyme which had been dialyzed against ity and the brownish protein were concen 1 mM tiron (sodium catechol-3 , 5-disulfonate) trated at pI 6.2, as shown in Fig. 6. The ab was 0.45% (7.2 moles Fe/89 kg protein). There sorption spectrum of the enzyme was hardly was a tendency for the enzyme to absorb iron affected by dialyzing the enzyme against 0.1 from the environment. When the enzyme was mM 1, 10-phenanthroline for 24 hr. concentrated by means of a Diaflo cell with a When the hydrogenase preparation was PM-30 membrane, its iron content rose to incubated under H2 in the presence of cyto 3.0%, though when this preparation was chrome c9, both the enzyme and the cyto thoroughly dialyzed against glass - distilled chrome were reduced, as determined from the water, the iron content fell to 1.3%. When spectral change. Figure 5 shows the absorp the same preparation was dialyzed against 1 tion spectrum of the enzyme (Curve 1) and mm 1, 10-phenanthroline, the iron content fell that of the reduced form of the enzyme pre to the original level, i.e., 0.60% . Molybdenum pared by reduction with H2 in the presence of was not detected in the preparation . a trace amount of cytochrome ca (Curve 2). Amino Acid Composition-The amino acid No spectral change was observed even after composition of the purified hydrogenase prepa incubation of the enzyme for 24 hr under H2 ration is shown in Table I. in the absence of cytochrome c2. Metal Contents-The iron content of puri fied hydrogenase preparation which had been TABLE I. Amino acid composition of the hydro thoroughly dialyzed against glass - distilled genase. water was 0.62% (9.9 moles Fe/89 kg protein). The iron content of enzyme which had been dialyzed against 1 mM 1, 10-phenanthroline was

Fig. 6. Elution pattern of the purified hydrogenase from the electrofocusing column (LKB 8101) after isoelectric focusing with 0.8% carrier ampholite (pH 5-8) for 5 days at 600 volts. The absorbance of each fraction was corrected by subtracting that of the corresponding fraction obtained when the electrofocusing was run without the enzyme. •œ, (a) Percentage of amino acid residues in the protein. absorbance at 400 nm ; •ü , absorbance at 280 nm ; (b) Moles per mole of protein to the nearest integer. a Determined spectrophotometrically. and L. activity of hydrogenase in units.ml-1.

Vol. 79, No. 3, 1976 666 T. YAGI, K. KIMURA, H. DAIDOJI, F. SAKAI, S. TAMURA, and H. INOKUCHI

Sulfur Contents-(ƒ¿) Total sulfur: The reduced directly with tin(‡U)-strong phosphoric enzyme was first decomposed with chromium- acid, only 1/8 of the total sulfur atoms could

(‡W)-strong phosphoric acid at 150? to oxidize be recovered as H2S. all the sulfur atoms to sulfate ions, which (b) Labile sulfide: Different assay meth were then reduced to H2S by adding tin(‡U)- ods gave different results. (i) Assay of H2S strong phosphoric acid at 300? after decom derived from the acidified enzyme. The en

posing the chromium(‡Y)-strong phosphoric zyme was first incubated in 0.2 M NaOH at 50? acid to the chromium(‡V) compound ( 19). The for 60 min, then acidified by the addition of sulfur content thus determined was 1.6%d (44 0.2 M HC1, and the H2S produced was deter- moles S/89 kg protein). If the enzyme was mined. The content of labile sulfide ions thus determined was 0.14% (3.9 moles S2-/89 kg

TABLE II. Effect of metabolic inhibitors on the protein). If the enzyme was directly acidified hydrogenase activity. The enzyme and cytochrome c3 with 2 M HC1, only a half of the labile sulfide were preincubated with the inhibitor in 3.0 ml of ions could be liberated as HS.

20 mM phosphate buffer, pH 7.0, for 30 min, and (úA) Direct assay. The enzyme was in the activity was measured by the H2-evolution tech cubated with zinc acetate-NaOH for 2 hr, and nique. then dimethyl-p-phenylenediamine and FeCl2 reagents were added as described by Suhara et al. (20). Any turbidity was removed by centrifugation, and the blue color developed due to the labile sulfide of protein was deter= mined colorimetrically. The content of labile sulfide ions thus determined was 0.25-0.28%

(6.9-7.9 moles S2-/89 kg protein). (c) Sulfur in the apoprotein: The sulfur content calculated from the number of cysteine and methionine residues in the hydrogenase molecule was 1.08% (30 moles S/89 kg protein). Effects of Metabolic Inhibitors and Chemi cal Modifications-The enzyme is not inhibited

a The activity was measured by the H 2-evolution technique and the enzymic electric cell method, both of which gave similar results. The figures are cited from Ref. 12. b The blue color of the reduced methylviologen disappeared when these compounds were added to the reaction mixture in the H2-evolution technique. Probably, these inhibitors oxidized the reduced methylviologen, and thus it is not clear Fig. 7. Inhibition of hydrogenase by CO. Hydro whether these reagents prevented the evolution of genase activities were assayed by the H2-evolution H2 by blocking the SH-groups of the enzyme or by technique under the standard assay conditions eliminating the electron donor of the reaction system . except that N2 in the gas phase was replaced by P The H2-evolution technique and the enzymic electric CO-N2 mixtures. pco is the partial pressure of CO cell method gave inconsistent results. We chose the in the gas phase ; ƒÒo, the activity when Pco is 0 ; latter for the reasons discussed in Ref . 12. and vi, the activity when CO is present.

J. Biochem PROPERTIES OF DESULFOVIBRIO HYDROGENASE 667

Fig. 9. Inhibition of hydrogenase by urea. Hy

drogenase activity was measured by H2-evolution

Fig. 8. Inactivation of hydrogenase in H202 solution. assay under the standard assay conditions except Hydrogenase (0.30 mg) was incubated in 8.3 mM that the reaction mixture contained urea. The

phosphate buffer, pH 7.0, containing 16.7 mM H2O2 enzyme was preincubated in the urea solution for at 25?, 0.1-m1 aliquots were withdrawn at intervals, at least 60 min before adding Na2S204. ----: water dilt ted, and the activity •assayed . by the enzymic concentration (volume %) in the urea solution, •ü :

electric cell method. The concentration of H202 in the observed residual activity in the urea solu the assay mixture was only 8.3 ƒÊM. At the same tion relative to that in the standard assay solution,

time, the intensity of the brown color of the incuba and •œ : the corrected residual activity, i.e., the

tion mixture was measured. ƒ¢ : the residual activity observed activity divided by the water concentration. relative to the original one, and •œ : the intensity The reason for the correction is discussed in the text

of the brown color expressed in A400nm-A6oanm (see also Fig. 14 in Ref. 12). relative to the original intensity.

by metal-complexing agents such as cyanide, 25?, and the concomitant spectral change is

azide, 1, 10-phenanthroline, tiron, and EDTA, shown in Fig. 8.

as shown in Table ‡U. The absorption spectra The activity of hydrogenase, like those of

and the specific activities of enzyme which many other enzymes, decreased when meas

had been treated with 1, 10-phenanthroline or ured in the presence of urea (Fig. 9). How

tiron were not very different from those of ever, when the enzyme was incubated in 9 M

the untreated enzyme. The enzyme was in urea for 24 hr at room temperature and the

hibited to some extent by SH-blocking reagents activity was measured immediately after 15-

such as iodoacetamide, iodoacetate, N-methyl- fold dilution, no inactivation at all was ob-

maleimide, p - chloromercuribenzenesulfonate, served.

or p-chloromercuribenzoate (Table II). The enzyme was also inactivated by 86-

CO strongly inhibited the H2-evolution re- 99% in 0.1% SDS (12). When the enzyme

action. The reciprocal of the residual activity was incubated in 0.1% SDS for 120 min at 30?

was plotted against the partial pressure of CO, either in the presence or absence of cyto

as shown in Fig. 7. From this figure, the in chrome c3, and the activity was measured im

hibitor constant (Ki) for CO was calculated to mediately after 300-fold dilution, 60% of the

be 8.9 Torr. Upon flushing out CO with a original activity was recovered.

stream of N2, the reaction mixture recovered

the original H2 evolution activity, as in the DISCUSSION

case of the soluble hydrogenase of the same

organism. (3). Hydrogenase catalyzes the transfer of electrons reversibly between H2 and an electron acceptor Hydrogenase was gradually inactivated

upon prolonged incubation with dilute H202 at as well as the conversion between paraH2 and

Vol. 79, No. 3, 1976 TABLE III. Comparison of the properties of hydrogenase preparations of various origins.

a Soluble enzyme. b Solubilized enzyme from particulate fraction. 668J. Bioche m. PROPERTIES OF DESULFOVIBRIO HYDROGENASE 669 orthoH2, and isotope exchange between H2 and of 7-8, but the measurement of H2S derived D20 or D2 and H2O. Three kinds of hydro from the acidified enzyme solution gave a con genases have been reported, differing in their tent of only 2-4. Presumably the enzyme specificity for electron acceptors. The electron contains 7-8 iron atoms and the same number acceptor for hydrogen dehydrogenase [EC of labile sulfide ions in the molecule. Mor 1.12.1.2] is NAD+ (1, 2 ), that for cytochrome tenson. and his collaborators also encountered c3 hydrogenase [EC 1.12.2.1] is cytochrome c3 difficulty in determining the iron and sulfide (3, 4), and that for ferredoxin hydrogenase contents. They first reported that the clos [EC 1.12.7.1] is ferredoxin (5, 6) ; still another tridial hydrogenase contained only 4 iron and hydrogenase of unknown specificity was puri sulfide (21), but revised the values to 12 later fied from Chromatium (7). These hydro (6). The sum of the cysteine, methionine, genases were shown to be ironsulfur proteins and labile sulfide contents (19+11+8) is less (2, 6, 7, 21-24). Chen and Mortenson (6) than the total sulfur content (44) meas reported that the clostridial hydrogenase [EC ured by direct analysis of the enzyme. This 1.12.7.1] contained 12 iron atoms and 12 acid- discrepancy could be caused by possible con labile sulfide ions per molecule (mol. wt.= tamination by sulfate ions used in the early 60,000) and did not dissociate into subunits. stage of the purification process. Haschke and Campbell (22) reported that the The absorption spectra of the hydrogenase hydrogenase of Desulfovibrio vulgaris, NCIB preparations so far reported (2, 6, 7, 22, 23) 8303 [EC 1.12.2.1] contained 0.72 iron atom and that reported in this paper are essential and 0.35 acid-labile sulfide ion per molecule ly similar, in spite of the specificity differences for electron acceptors. All of these spectra (mol. wt.=45,000). LeGall et al. (23) reported have two peaks, one near 280 nm, and the that a hydrogenase from Desulfovibrio vulgaris, Hildenborough [probably EC 1.12.2.1] con other near 400 nm. The latter is a broad tained 3.5 iron atoms and 3.2 acid-labile sul peak, and is observed as a mere shoulder in fide ions per molecule (mol. wt. =60,000), and most cases. These spectral features are char was a dimer of two equivalent subunits. Hy acteristic of iron-sulfur proteins such as fer redoxins. The specific absorbance coefficients drogenase solubilized and purified from partic and absorbance ratios (A400nm/A280 nm), as well ulate fraction of Chromatium was also re as the molecular weights and the iron and ported to contain 4 iron atoms and 4 acid labile sulfide ions per molecule (mol. wt. = sulfide contents, are different, however (see Table III for comparison). Since the common 98,000), which was a dimer of two equivalent ability of hydrogenases is to activate H2, iron subunits (7). The molecular weight of the hydrogenase sulfur structure is probably involved in this described in this paper is 89,000. This en activation process. Gradual inactivation of zyme is composed of two different subunits hydrogenase by H202 treatment, accompanied with molecular weights of 59,000 and 28,000. by decolorization, also supports this view. The iron atoms must alternately be re Since the enzyme was purified from tryptic duced and oxidized, as suggested by the spec digest of the bacterial particulate fraction, it tral changes of the enzyme upon reduction and is not certain whether the subunit structure .reoxidation. It is noteworthy that the enzyme is native or an artifact induced by tryptic alone was not reduced under an atmosphere digestion. of H2 unless the electron acceptor (cytochrome The iron content of hydrogenase which c3) was present. This observation is in con had been thoroughly dialyzed against a solu trast to that of LeGall et al. (23), who ob tion containing chelating agent was 7-9. How served a spectral change of the enzyme itself ever, the labile sulfide content of the hydro in less than 1 hr on incubation under an at genase depended on the assay technique. Di mosphere of H2. Probably, flavoprotein(s) pres rect measurement of labile sulfide by incubat ent as contaminants in their preparation be ing the enzyme in H2S-absorbing solution con haved as acceptor(s) to reduce the enzyme in taining Zn2+ and alkali, gave a sulfide content

Vol. 79, No. 3, 1976 670 T. YAGI, K. KIMURA, H. DAIDOJI, F. SAKAI, S. TAMURA, and H. INOKUCHI the absence of exogenously added acceptor. tive as the native enzyme. Under these con It was not stated whether a recently purified ditions, the molar concentration of H2O is only hydrogenase preparation from D. gigas (25), 59% of that under the standard assay condi which was free from flavoproteins and was tions. Assuming that the rate of hydrogenase specific to cytochrome c3, was reducible by it- catalyzed H2 production is proportional to the self or not under an atmosphere of H2. Clos water concentration in the reaction medium, tridial hydrogenase was apparently not reduced as suggested in our earlier observations (4), by H2 in the absence of added electron ac the inhibition was 67% (Fig. 9). Full recovery ceptor (6). of the activity immediately after dilution of It has been repeatedly observed that hy the enzyme from 9 M urea solution indicates drogenase by itself, either in aqueous solution that the enzyme is very resistant to the in

(1, 7, 8, 26, 27) or in the dry state (28), fluence of urea, a hydrogen bond-breaking re catalyzes the conversion of paraH2 to normal agent. Clostridial hydrogenase was also very H2 and the exchange between H2 and D20 and resistant to urea treatment (24). Although D2 and H2O. These together with the above the enzyme was almost inactive in 0.1% SDS, observations mean that the hydrogenase has as much as 60% of the original activity was the ability to combine with H2 but lacks the recovered upon diluting the preparation 300- ability to reduce itself using the bound H fold. Failure to produce H2S by mere reduc atoms in the absence of exogenously added tion of the enzyme with H8P04-Sn(‡U) solution electron acceptors. could also be due to the rigidity of the en Although the amino acid compositions of zyme. the hydrogenases so far reported are not very We are indebted to Prof. N. Ui of Gunma University, similar, generally speaking, they are rich in who kindly determined the molecular weight of the acidic (including acidoamidic) and hydrophobic hydrogenase and advised us on the homogeneity of amino acids and poor in cyst(e)ine and methi the protein from the viewpoint of centrifugal analy onine. The relatively higher molar absorb sis. We are indebted to Prof. N. Tamiya of Tohoku ance coefficient at 280 nm (ƒÃ280nm) of our University and Prof. K. Takahashi of Kyoto Univer preparation is accounted for by the high trypto sity for amino acid analyses. Thanks are also due phan and tyrosine contents. to Miss N. Masuda, Miss H. Tsujimura, and Miss Observations as to the effects of metabolic Y. Ide for conducting some of the inhibition experi inhibitors on hydrogenase are inconsistent, ments. varying with different preparations and ob- servers, except in the case of CO, which is a REFERENCES strong inhibitor for any hydrogenase prepa 1. Bone, D.H. (1963) Biochim. Biophys. Acta 67, ration. Even if references are restricted to 589-598 those published in the last decade, metal-com- 2. Pfitzner, J., Linke, H.A.B., & Schlegel, H.G. plexing agents such as cyanide and 1, 10- (1970) Arch. Mikrobiol. 71, 67-78 phenanthroline have been reported to be in 3. Yagi, T., Honya, M., & Tamiya, N. (1968) hibitory (22, 29, 30) and not inhibitory (6, Biochim. Biophys. Acta 153, 699-705 12, 24, 31, 32, and this paper). Probably, the 4. Yagi, T. (1970) J. Biochem. 68, 649-657 results have been dependent on the assay con 5. Mortenson, L.E., Valentine, R.C., & Carnahan, ditions, as suggested by earlier workers (1, J.E. (1962) Biochem. Biophys. Res. Commun. 7, 33, 34). Chen and Mortenson (6) explained 448-452 6. Chen, J.-S. & Mortenson, L.E. (1974) Biochim. the conflicting observations in terms of slow Biophys. Acta 371, 283-298 reactivity of essential iron in the enzyme with 7. Gitlitz, P.H. & Krasna, A.I. (1975) Biochemistry 1, 10-phenanthroline. 14, 2561-2568 It seems that the particulate hydrogenase 8. Yagi, T., Tsuda, M., & Inokuchi, H. (1973) J. of D. ƒÒulgaris has a very rigid structure. It Biochem. 73, 1069-1081 is very resistant to urea treatment. Even in 9. Yagi, T. & Maruyama, K. (1971) Biochim. 9 M urea solution, the enzyme was 20% as ac- Biophys. Acta 243, 214-224

J. Biochem. PROPERTIES OF DESULFOVIBRIO HYDROGENASE 671

10. Reisfeld, R.A., Lewis, U.J., & Williams, D.E. 22. Haschke, R.H. & Campbell, L.L. (1971) J.

(1962) Nature 195, 281-283, cited in Y. Nagai, Bacteriol. 105, 249-258 in the Methods in Biochemical Experiments, Pt. ‡\ 23. LeGall, J., DerVartanian, D.V., Spilker, E.,

, Tampakushitsu-Kakusan Koso (1967) 3-14 (in Lee, J.-P., & Peck, H.D., Jr. (1971) Biochim. Bio Japanese) phys. Acta 234, 525-530 11. Kimura, K., Inokuchi, H., & Yagi, T. (1972) 24. Nakos, G. & Mortenson, L.E. (1971) Biochemistry Chem. Lett. 693-697 10, 2442-2449

12. Yagi, T., Goto, M., Nakano, K., Kimura, K., 25. DerVartanian, D.V. & LeGall, J. (1974) Biochim. & Inokuchi, H. (1975) J. Biochem. 78, 443-454 Biophys, Acta 346, 79-99

13. Edelstein, S.J. & Schachman, H.K. (1967) J. 26. Krasna, A.I. & Rittenberg, D. (1954) J. Am. Biol. Chem. 242, 306-311 Chem. Soc. 76, 3015-3020

14. Schachman, H.K. & Edelstein, S.J. (1973) Meth. 27. Tamiya, N. & Miller, S.L. (1963) J. Biol. Chem. Enzymol. 27, 3-59 238, 2194-2198

15. Weber, K. & Osborn, M. (1969) J. Biol. Chem. 28. Yagi, T., Tsuda, M., Mori, Y., & Inokuchi, H. 244, 4406-4412 (1969) J. Am. Chem. Soc. 91, 2801

16. Spackman, D.H., Stein, W.H., & Moore, S. 29. Eberhardt, U. (1966) Arch. Mikrobiol. 53, 288-

(1958) Anal. Chem. 30 1190-1206 302 17. Bencze, W.L. & Schmid, K. (1957) Anal. Chem. 30. Gogotov, I.N., Zorin, N.A., & Ushakov, V.M. 29, 1193-1196 (1.973) Mikrobiologiya 42, 21-25, cited in Chem.

18. Boltz, D.F. (1958) Colorimetric Determination of Abst. 79, 682e Nonmetals pp. 273 ff. Interscience Publ. Inc., 31. Yakovlev, V.A. & Mitsova, I.Z. (1970) Biokhimiya

New York 35, 675-678, cited in Chem. Abst. 74, 19697 g

19. Kiba, T., Akaza, I., & Taki, S. (1957) Bull. 32. Thauer, R.K., Kaufer, B., Zahringer, M., & Chem. Soc. Japan 30, 482 Jungermann, K. (1974) Eur. J. Biochem. 42,

20. Suhara, K., Kanayama, K., Takemori, S., & 447-452 Katagiri, M. (1974) Biochim. Biophys. Acta 336, 33. Farkas, L. & Fischer, E. (1947) J. Biol. Chem.

309-317 167, 787-805

21. Nakos, G. & Mortenson, L.E. (1971) Biochim, 34. Atkinson, D.E. & McFadden, B.A. (1954) J. Biol. Biophys. Acta 227, 576-583 Chem. 210, 885-893

Vol. 79, No. 3, 1976