Arch. Microbiol. 105, 249-256 (1975) - 9 by Springer-Verlag 1975
Purification and Properties of Thiosulfate Reductase from Desulfovibrio gigas E. C. HATCHIKIAN Laboratoire de Chimie Bact6rienne, C.N.R.S., Marseille Received June 2, 1975
Abstract. Thiosulfate reductase of the dissimilatory sulfatereducing bacterium Desulfovibrio gigas has been purified 415-fold and its properties investigated. The enzyme was unstable during the different steps of purification as well as during storage at -15~ The molecular weight of thiosulfate reductase estimated from the chromatographic behaviour of the enzyme on Sephadex G-200 was close to 220000. The absorption spectrum of the purified enzyme exhibited a protein peak at 278 nm without characteristic features in the visible region. Thiosulfate reductase catalyzed the stoichiometric production of hydrogen
sulfide and sulfite from thiosulfate, and exhibited tetrathionate reductase activity. It did not show sulfite reductase activity. The optimum pH of thiosulfate reduction occurred between pH 7.4 and 8.0 and its K,, value for thiosulfate was calculated to be 5 9 10.4 M. The sensitivity of thiosulfate reductase to sulfhydryl reagent and the reversal of the inhibition by cysteine indicated that one or more sulfhydryl groups were involved in the catalytic activity. The study of electron transport between hydrogenase and thiosulfate reductase showed that the most efficient coupling was obtained with a system containing cytochromes c3 (Mr = 13000) and c3 (Mr = 26000).
Key words." Desulfovibrio gigas - Purification, properties of - Thiosulfate reductase - Sulfate-reducing bacteria.
According to the most recent data the dissimilatory reduction of sulfite catalyzed by sulfate-reducing bacteria, proceeds primarily through a pathway containing trithionate and thiosulfate as intermediate compounds (Kobayashi etal., 1969; Suh and Akagi, 1969; Findley and Akagi, 1970). This pathway involves at least three enzymes (Lee and Peck, 1971; Lee etal., 1973). Lee and Peck (1971) reported the purification of the dissimilatory sulfite reductase (desulfoviridin) from Desulfovibrio gigas and the identification of its product as trithionate. However, Kobayashi etal. (1972) and Jones and Skyring (1975) reported that desulfoviridin formed trithionate from sulfite and, furthermore, that thiosulfate and sulfide were also products of this reaction. Recently, a confirmation of the trithionate pathway was obtained by Akagi et al. (1974) with the bisulfite reductase (P582) isolated from
Desulfotomaculum nigr~'cans. Kobayashi etal. (1969) demonstrated that trithionate and thiosulfate were sequentially accumulated during the reduction of sulfite by extracts of D. vulgaris and Lee (1972) observed the reduction of trithionate by a partially fractionated extract of D. gigas. These results established that trithionate reductase which catalyzes the second step of sulfite reduction is present in extracts of these bacteria; however, this enzyme was not further studied.
The ability of the sulfate reducing bacteria to reduce thiosulfate has been previously reported (Ishimoto etal., 1955) and thiosulfate reductase which catalyzes the terminal step of dissimilatory sulfite reduction, i.e. the reductive cleavage of thiosulfate to sulfide and sulfite, has been studied in detail in D. vulgaris (Ishimoto and Koyama, 1957; Haschke and Campbell, 1971) and D.nigr~'cans (Nakatsukasa and Akagi, 1969). It is to be noted that rhodanese activity (thiosulfate :cyanide sulfur transferase; EC 2.8.11) has been found in D.nigrificans by Burton and Akagi (1971). The authors postulated that such an activity might be necessary for the cleavage of the thiosulfate molecule during the reductive process. The electron tra.nsport between hydrogenase and thiosulfate reductase has been the subject of a number of studies. Cytochrome c3 (Mr = 13000) has been reported to act as an electron carrier in the reduction of thiosulfate and sulfite in D. vulgaris (Ishimoto and Koyama, 1955; Postgate, 1956) and more recently, ferredoxin and flavodoxin have been implicated in the same reactions in D.gigas (Le Gall and Dragoni, 1966; Le Gall and Hatchikian, 1967; Guarraia etal., 1968; Irie etal., 1973). A comparative study demonstrated that the reduction of thiosulfate by D.gigas extracts devoid of electron carriers can be stimulated
250
Arch. Microbiol., Vol. 105, No. 3 (1975)
by at least three proteins, a non heme-iron protein ferredoxin, a flavoprotein, flavodoxin and a heme-iron protein, cytochrome c3 (Mr = 26000)1; both ferredoxin and flavodoxin are also carriers for the reduction of sulfite whereas cytochrome c3 (Mr = 26000) is specific for the reduction of thiosulfate (Bruschi et al., 1969; Hatchikian etal., 1972). This paper reports the purification and some properties of the thiosulfate reductase from D.gigas and also its physiological relationship with several electron carriers.
Materials and Methods Cultivation of Bacteria and Preparation of Extracts. Desulfovibrio gigas was cultivated and harvested according to Le Gall etal. (1965). To extract cytochrome c3 (Mr = 13000) and hydrogenase the bacteria were washed as previously described (Le Gall etal., 1965; Hatchikian and Le Gall, 1972). The washed cells were suspended in 0.02 M potassium phosphate buffer, pH 7.6, and the extract as well as soluble and particulate fractions were prepared as already reported (Hatchiklan and Le Gall, 1970).
Enzyme Assays. Thiosulfate reductase activity was determined using the Warburg manometric method under hydrogen atmosphere in a system which contained methyl viologen, excess hydrogenase and thiosulfate at 37~ by following the consumption of hydrogen. The main compartment of each flask contained potassium phosphate (150 ~tmoles), methyl viologen (10 lamoles), hydrogenase preparation (500 lag protein), enzyme solution (0.02-0.15 ml) and the center wel! contained NaOH l0 N (0.05 ml). Na2S2Q (20 lamoles) was introduced from the side arm to give a total volume of 3 ml. The hydrogenase preparation utilized in our experiments was devoid of thiosulfate and sulfite reductase activities. It was prepared by the following procedure: the active fi'action not adsorbed on silica gel (Hatchikian and Le Gall, 1972) was passed through a Sephadex G-100 column; this treatment was sufficient to eliminate the reductases which are high molecular weight proteins. In some cases a purified hydrogenase was used, prepared from D.gigas as already reported (Bell, 1973). The electron carriers, cytochrome c3 (Mr = 13000), flavodoxin and cytochrome c3 (M~ = 26000) used in the coupling assays between hydrogenase and thiosulfate reductase were purified as previously described (Le Gall etal., ~1965; Le Gall and Hatchikian, 1967; Bruschi etal., 1969). Assays of sulfite reductase and tetrathionate reductase activities were carried out manometricalty as described for the thiosulfate reductase activity except that 20 lamoles of sulfite were used in the case of sulfite reductase and 5 lamoles of tetrathionate and 2 lamoles of benzyl viologen instead of methyl viologen in the case of the tetrathionate reductase. Rhodanese activity was measured according to the method of S6rbo (1957) modified by Wang and Volini (1968). Analytical Methods. The stoichiometry of the enzymatic formation of hydrogen sulfide and sulfite from thiosulfate in the manometric system was established as already reported (Hatchikian et al., 1972). The content of H2S in the center well was. measured by the colorimetric method of Fogo and Popowsky (1949) ; the content of sulfite in the reaction mixture of the main compartment of the Warburg vessel was deter1The name c% that was first given to cytochrome c3 (M. W. = 26000) cannot be accepted in the nomenclature of cytochromes.
mined according to the method of Grant (1947). The enzymatic formation of thiosulfate from tetrathionate in the manometric system was measured by iodometric titration using starch as indicator.
Protein Determidation. Protein was estimated by the biuret method (Gornall etal., 1949) and the method of Lowry etal. (1951).
Electrophoresis. Analytical gel electrophoresis was performed in 7 ~ polyacrylamide gel and the Tris-HCl glycine buffer was at pH 8.8. The current used was 3 mA per tube. Each gel was loaded with 30-50 lag protein, and protein bands were revealed with CoomasSie blue. For the determination of thiosulfate reductase directly in the gel after electrophoresis, the gels were first stained with reduced methyl viologen and then dipped into a solution of 40 mM potassium phosphate buffer (pH 7.6) containing 8 mM sodium thiosulfate. A clear band in the gel revealed enzyme activity. Absorption Spectra. Absorption spectra were measured in cells of 1 cm path length at room temperature with a Cary model 14 recording spectrophotometer.
Results Purification Scheme. The washed bacteria (600 g) were suspended in 350 ml of 20 m M potassium phosphate buffer (all buffers used in the purification were kept at p H 7.6) and the soluble protein extract was prepared as already described. The soluble protein fraction contained 6 5 - 7 0 ~ o of the thiosulfate reductase activity present in the crude extract. The first purification step, the chromatography of the suspension on DEAE-cellulose column (4 c m x 3 cm) essentially removed the acidic electron carriers of Desulfovibrio gigas [ferredoxin, flavodoxin, rubredoxin and cytochrome c3 (Mr = 26000)] which were eluted with 1 M Tris-HC1 buffer and stored for later purification. Most of the thiosulfate reductase activity was not retained on the DEAE-cellulose column. To this not-retained fraction solid a m m o n i u m sulfate was added to 20 ~o saturation. After centrifugation at 30 000 x g for 20 min the pellet was discarded and a m m o n i u m sulfate added to the supernatant to 4 0 ~ saturation. The precipitate obtained by centrifugation was dissolved in 150 ml of 50 m M phosphate buffer and dialyzed overnight against 10 1 10 m M of the same buffer. 75 ml of wet DEAE-cellulose equilibrated with 30 m M phosphate buffer were poured into the dialyzed solution (234 ml) and the mixture was stirred overnight at 4 ~ C. The DEAE-cellulose was separated by centrifugation, packed into a column and washed with 30 m M phosphate buffer. The enzyme activity was not retained on the DEAE-cellulose while most of the green pigment, desulfoviridin, previously present in the extract was adsorbed. The active fraction (292 ml) was then concentrated to 180 ml in a 350 ml ultrafiltration cell with a PM-10 filter (Amicon), centrifuged to remove a slight precipitate, a n d 30 ml fractions
E. C. Hatchikian: Thiosulfate Reductase from D.gigas applied to a Sephadex G-200 column (5 x 100 cm) equilibrated with 20 mM Tris-HC1 buffer. The column was eluted with the same buffer at a flow rate of 0.6 ml per min. The yellow coloured active fractions eluted from Sephadex G-200 columns were pooled and adsorbed on a DEAE-cellulose column (4 x 22 cm) equilibrated with 10 mM phosphate buffer. The enzyme was eluted by means of a non-linear phosphate gradient from 1 0 - 2 5 0 raM, concentrated to 72 ml as previously described, then applied to a Sephadex G-200 column. The active fraction, still containing flavoproteins was adsorbed on a calcium phosphate gel column (4 x 12 cm) equilibrated with 5 mM phosphate buffer and eluted by means of a phosphate gradient from 1 0 - 1 5 0 mM. Most of the flavoproteins, mainly adenosine 5'-phosphosulfate reductase (APSreductase), were retained on the column. The fraction containing thiosulfate reductase activity was adsorbed on a DEAE-cellulose column (3 x 7 cm) equilibrated with 10 m M phosphate buffer and the proteins were eluted with a phosphate gradient from 10-150 mM. The enzyme was eluted by 8 0 - 1 0 0 mM phosphate buffer. At this point, the enzyme had been purified 415-fold. The purification scheme is summarized in Table 1. The purified thiosulfate reductase was colourless. It exhibited a great instability and lost activity when frozen and thawed. The reductive reagents (/Lmercaptoethanol or dithiothreitol) had no effect on the instability.
Test of Homogeneity. As shown in Fig. 1 electrophoresis of the partially purified enzyme revealed the presence of four bands. The activity test in the gel followed by staining with Coomassie blue indicated that thiosulfate reductase activity was located between the two protein bands nearest the origin. It is noteworthy that after such an electrophoresis performed at pH 8.8, the enzymatic activity was very weak and was not located on an apparent band of protein. As reported later, the pH optimum of thiosulfate reductase occurs between 7.5 and 8 and at a pH value of 8.8 the activity is only 8 ~o of that obtained at pH 7.6. When the electrophoresis was run for 3 hrs with phosphate buffer at pH 7.6 instead of pH 8.8, the proteins just penetrated into the gel and showed one main band (Fig. 1). In this latter case, where the pH used is near the isoelectric point of thiosulfate reductase, the activity test in the gel with reduced methyl-viologen indicated a very strong enzymatic activity. Molecular Weight. The molecular weight of thiosulfate reductase was not accurately determined; however, the chromatographic behaviour of the enzyme on Sephadex G-200 during the purification compared with that of known D. gigas proteins allowed an
251 Table 1. Purification scheme of thiosulfate reductase Step
Volume Protein Total (ml) (mg/ml) units
Soluble proteins
Specific activity
520
70
7 200
0.20
After DEAE-cellulose column 534
48
5 874
0.23
After ammonium sulfate precipitation (20- 40 ~)
234
60
2 600
0.184
DEAE-eluant
292
34
2450
0.25
1280
0.75
1.5
800
t.4
After Sephade• G-200 (2nd) 298
0.75
500
2.3
After calcium phosphate gel column After DEAE column (3rd)
0.115 0.017 0.059
358 16.3 71.7 83 96.7 41
After SephadexG-200 446 After DEAE column (2nd) 388
151 [51 [ 40
3.85
Enzyme activity was determined as reported in "Materials and Methods". The thiosulfate reductase unit of activity was defined as the amount of enzyme which catalyzed the consumption of 1 ~tmoleof hydrogen per min.
estimation of its value. Thiosulfate reductase was eluted from the Sephadex G-200 column just before desulfoviridin which has a molecular weight of 200 000 (Lee and Peck, 1971) and at the same time as APSreductase which has a molecular weight of 220 000 in D. vulgaris (Michaels et at., 1970). These observations suggest that the value for the molecular weight of D.gigas thiosulfate reductase is about 220000.
Spectral Properties. Fig.2 gives the spectrum of a 280-fold purified thiosulfate reductase preparation. This enzyme fraction has an absorption spectrum without any special features; it shows a protein peak at 278 nm and a weak absorption at 380-400 nm. The weak absorption in the visible region may possibly be attributed to the presence of traces of APS-reductase (Peck etal., 1965) which has a chromatographic behaviour very similar to that of thiosulfate reductase. Substrates, Products and Stoichiometry. With thiosulfate as substrate and reduced methyl-viologen as electron donor, thiosulfate reductase from D.gigas catalyzed the stoichiometric production of hydrogen sulfide and sulfite. The enzyme also exhibited a slight catalytic activity with dithionite. This could be related to the formation of a small amount of thiosulfate detected after solution of the dithionite by the method of S6rbo (1957).
252
Arch. Microbiol., Vol. 105, No. 3 (1975)
Fig. 1. Disc acrylamide-gel electrophoresis of purified preparation of thiosulfate reductase from Desulfovibriogigas at several pH values. 35 lag protein of purified thiosulfate reductase (specific activity 83) were applied to each gel for electrophoresis: gel 1 was run at pH 8.8 in Tris-glycine buffer and gel 3 at pH 7.6 in phosphate buffer and stained for protein with Coomassie blue. Gels 2 and 4 were run respectively at pH 8.8 and 7.6 and thiosulfate reductase visualized after 7 min for gel 2 and 3 min for gel 4 by the thiosulfate dependant oxidation of reduced methyl viologen as described in "Materials and Methods"
Effect ofpH and Buffers. The pH optimum of reduced methyl viologen linked thiosulfate reductase activity was investigated with phosphate buffer (pH 6 . 6 - 8 ) , Tris-HC1 buffer (pH 7 . 2 - 8 . 7 ) and barbital-HC1 buffer (pH 7.0-8.7). The maximum activity of thiosulfate reductase from D.gigas was in the pH range 7 . 4 - 8 . 0 as shown in Fig. 3 and activity decreased rapidly above and below these values. The buffer in which thiosulfate reduction functioned most effectively was phosphate buffer. In this system, the enzyme activity was more than two times that in Tris-HC1 buffer. Assays with potassium phosphate, sodium phosphate, sodium chloride and potassium chloride indicated that phosphate anion was the effective ion.
/
0.7!
)-
0.5
Z [II nO
m 0.3
250
350
/-,50
550
650
WAVELENGTH (nm)
Fig. 2. The absorption spectrum of partially purified thiosulfate reductase from D.gigas. The protein concentration was 0.9 mg/ml and the specific activity was 56
Michaelis Constant. F r o m the data analyzed by a Lineweaver-Burk plot, the Km of the purified enzyme for thiosulfate was calculated to be 5. J0 -4 M at pH 7.6. With high concentrations of thiosulfate (about 10-fold the Km value) there was inhibition by excess substrate. Inhibition by Sulfhydryl Reagents. The data concerning
The partially purified enzyme exhibited tetrathionate reductase activity with reduced benzyl viologen as electron donor; tetrathionate is reduced to thiosulfate. However, the thiosulfate reductase was devoid of rhodanese activity and had no activity With sulfite and sulfate.
the inhibition of thiosulfate reductase are presented in Table 2. The results indicate that all the tested sulfhydry! reagents had an inhibitory effect. These compounds had no effect on hydrogenase activity at the concentrations used. The heavy metals, p-chloromercuribenzoate (pCMB), HgC12, AgNO3 and mer-
E. C. Hatchikian: Thiosulfate Reductase from D.gigas
253 Table 3: Thiosulfate reductase activity with the natural electron carriers
e-
g
m 40
System
lamoles H 2 consumed in 20 rain
Complete minus MV minus S202minus MV + I minus MV + II minus MV + III minus MV + I + II minus MV + I + I I I minus MV + II + I I I minus MV + I + II + I I I minus MV and $20~- + I + II + I I I
7.12 0 0 0.43 0.35 0.04 1.16 0.55 0.62 1.13 0
:zk
~30
c~20 o
-r-10
61 6
'
7.0 '
'
7./, '
'
7.'8
'
8 i2
'
8.6 ' pH
Fig. 3. The effect of pH on the activity of thiosulfate reductase. Assay mixtures were as described in "Materials and Methods" with potassium phosphate buffer, 150 Ixmoles (A -A) or Tfis-HC1 buffer, 150 lamoles (O O) or barbital buffer, 150 lamoles ( O - - - - Q ) and 35 gg of thiosulfate reductase (specific activity, 41)
Table 2. The effect of sulfhydryl inhibitors Inhibitor
Concentration ~ Inhibition
(M) p-chloromercuribenzoate
HgC12
2.10 s
100
8 - 10 . 6
55
5 - 10 .6
36
10 .5 5" 10 . 6
AgNO 3
10 5
100 66 90
Mersalyl
10 .5
35
Iodacetate
5.10 -4 10-4
70 33
N-ethylmaleimide
5 - 10-* 10 4
55 36
5 - 10 .5 10 -s 3.10 -3
52 22 45
o-iodosobenzoate Sulfite
Reaction mixtures were as described in "Materials and Methods" except that inhibitors at the indicated concentrations were incubated for 30 min with buffer and enzyme before starting the reaction. Purified thiosulfate reductase (30 lag protein) of specific activity 83 was utilized in all determinations.
salyl inhibited activity. In addition, alkylating reagents, iodacetate and N-ethylmaleimide as well as the oxidizing agent, o-iodosobenzoate, inhibited the activity of thiosulfate reductase. The enzyme activity was a!so inhibited by sulfite. As indicated in Table 2, 2 . 1 0 -5 M p C M B completely inhibited thiosulfate reductase activity, and the inhibition could be partially or completely reversed by cysteine (10 -4 M) according
The complete system contained the reagents as indicated in "Materials and Methods" except for MV (methyl viologen), 100 nmoles, pure hydrogenase, 400 lag protein and thiosulfate reductase (specific activity, 83) 30 lag protein. Electron carriers from D.gigas were added in the main compartment of the systems instead of MV as indicated. These systems contained (in nmoles): cytochrome c3 (M, = 13000) [I], 10, cytochrome c3 (Mr = 26000) [II], 10 and flavodoxin [III], 100.
to the incubation time of the enzyme with the inhibitor. Cysteine itself had no stimulatory effect on thiosulfate reductase. These results further substantiate that one or more sulfhydryl groups are involved in the catalytic activity of this enzyme. It is noteworthy that important variations in the inhibitory effect of reagents were observed with the experimental conditions (incubation time, p H and substrate concentration) more especially in the case of N-ethylmaleimide, o-iodosobenzoate and pCMB. These observations can be interpreted to indicate that more than one sulfhydryl group is involved in the activity of thiosulfate reductase and that the accessibility of these SH-groups are different.
Electron Donors. Reduced methyl viologen was the most efficient electron donor used by the thiosulfate reductase. With benzyl viologen the enzyme activity was only 8 ~o of that obtained with methyl viologen. No H2S was produced when the electron donors were cysteine, reduced glutathione, dihydrolipoate, reduced nicotinarnide adenine dinucleotide or reduced nicotinamide adenine dinucleotide phosphate.
Electron Carriers. In order to confirm the previous results obtained with a crude soluble protein fraction containing both sulfite and thiosulfate reductase activities separated from the electron carriers by a m m o n i u m sulfate precipitation (Hatchikian etal., 1972), assays
254
Arch. MicrobioL, Vol. i05, No. 3 (1975)
were performed with the partially purified thiosulfate red uctase, pure hydrogenase and the pure e!ectron carriers, cytochrome c3 (M~ = 26000), cytochrome c3 (M~ = 13 000) and flavodoxim The results are presented in 'Fable 3~ The best activity was observed with the systems containing both cytochrome ca (M~ = 13000) and cytochrome c3 (Mr = 26000) whereas the other systems had either no activity or only a slight activity. The activity exhibited by the most efficient system which contained cytochromes c3 (M~ = 13000) and c3 (M~ = 26000) reached about 16~o of the activity of the complete system with methyl viologen (100 nmoles).
Discussion The results of several investigators indicate that trithionate and thiosulfate are intermediates in the reduction of sulfite by the sulfate reducing bacteria (Kobayashi etal., !969; Findley and Akagi, 1970; Lee and Peck, 1971; Le Gall and Postgate, 1973; Akagi etaL, 1974). Thiosulfate reductase which catalyzes the reduction of thiosulfate into sulfide and sulfite as eariier reported by Ishimoto and Koyama (1957) is the terminal enzyme of the cyclic process of sulfite reduction postulated by Kobayashi etaL (1969). The thiosulfate reductase from Desulfovibrio gigas which had been purified 415-fold possessed a high catalytic activity in contrast to bisulfite reductase (Lee and Peck, 1971). The results obtained by analytical electrophoresis at pH 8.8 and 7.6 and by the activity test in the gels suggested that in our conditions of electrophoresis at pH 8.8 there was a dissociation of the thiosulfate reductase molecule and only a slight activity appeared where traces of the different subunits of the enzyme were present. In contrast the greater activity at pH 7.6 might be due to an unchanged enzyme structure. It is to be noted that Oltmann et aL (1974) have reported that in Proteus mirabilis, the thiosulfate reductase consists of two subunits. In addition to thiosnlfate reductase activity, the purified enzyme exhibits tetrathionate reductase activity as described recently for the thiosu~fate reduc~.ase from P.mirabilis (Oltmann etaL, 1974) whereas such an activity was not detected with the enzyme of DesuL Jbtomaculum ,~ligr(ficans (Nakatsukasa and Akagi, !969) and not investigated with the thiosulfate reductase of D. vulgaris (Haschke and Campbell, 197i). Whether tetrathionate reductase activity present in D.gigas is due to the tetrathionate reductase activity of thiosuifate reductase alone cannot be determined from our experiments. D. gigas thiosulfate reductase is devoid of rhodanese activity. This result shows that the reductive cleavage of thiosulfate is catalyzed by a
single protein and suggests that rhodanese is not implied in the thiosulfate reduction process of D.nigrificans as postulated by Burton and Akagi (1971). D.gigas thiosulfate reductase differs also from the other dissimilatory thiosulfate reductases (Nakatsukasa and Akagi, 1969; Haschke, 1969) by its narrow optimum pH range and its K~ value for thiosulfate. This latter value (5 9 10 .4 M) shows that the affinity of D.gigas thiosulfate reductase for thiosulfate is 200-fold greater than that of the similar enzyme of D. nigrificans (Nakatsukasa and Akagi, 1969). As reported for other thiosulfate reductases (Ishimoto and Koyama, 1961; Nakatsukasa and Akagi, 1969) the sensitivity of the enzyme from D.gigas to sulfhydryl reagents and the reversal of the inhibition by cysteine indicate that one or more sutfhydryl groups are involved in the cata!ytic activity. In contrast to assimilatory type thiosulfate reductases (Kaji and McElroy, 1959; Leinweber and Monty, 1963), D.gigas thiosulfate reductase as well as the enzyme from D. nigrificans (Nakatsukasa and Akagi, !969) were inactive with reduced glutathione and cysteine as electron donors. Perhaps the failure of the reductases of sulfate-reducing bacteria to function with these electron donors is a characteristic feature of the dissimilatory type thiosulfate reductases. The role of cytochrome c3 (Mr = 13000) in the electron transport between hydrogenase and thiosulfate reduc~:ase in extracts of D. vulgaris was earlier described (Ishimoto and Koyama, 1955; Postgate, 1956; Ishimoto etaL, 1958). Our previous results (Hatchikian etaL, 1972) indicated that cytochrome c3 (Mr = 26000), flavodoxin or ferredoxin, but not cytochrome c3 (Mr = 13000), stimulated thiosulfate reductase activity of an extract of D~gigas which contained reductase and hydrogenase activities separated from electron carriers by ammonium sulfate precipitation. With the purified thiosu!fate reductase and pure hydrogenase, the most efficient restoration of activity was obtained with the presence of both cytochrome c3 (Mr = 13000) and cytochrome c3 (Mr = 26 000) and the observations suggested that flavodoxin was inactive with this purified system. This discrepancy in our results concerning the implication of cytochrome c3 (M~ = 13 000) and flavodoxin in the coupling between hydrogenase and thiosutfate reductase deserves further comment. Firstly, the preparations obtained by ammonium sulfate precipitation of soluble protein extract were not completely devoid of cytochrome c3 and traces of this carrier could be sufficient to saturate the enzymatic extract. Secondly, the stimuiatory effect of flavodoxin could be explained by the presence of both thiosulfate and sulfite reductase activities in our previous enzymatic preparation. Thus the differences between our earlier observations and the
E. C. Hatchikian: Thiosulfate Reductase from D.gigas
results reported here could be due to the impure enzymatic preparations previously utilized. Another point to notice is the weak activity obtained with the natural electron carriers, a b o u t 16 ~ in the best case, as c o m p a r e d with the system containing methyl viologen. These observations suggest that another electron carrier m a y be implicated in the coupling between hydrogenase and thiosulfate reductase. H o w ever, it must be emphasized that our system is p r o b a b l y vastly different f r o m that occurring intra-cellularly.
Acknowledgements. The author expresses histhanks to Dr. J. Le Gall for his valuable advice and discussions throughout this work. I gratefully acknowledge the skillful technical assistance of Mrs. N. Forget. Thanks are due to Dr. M. Scandellari and Mr. R. Bourrelli for growing the sulfate reducing bacteria used in this study.
References Akagi, J. M., Chan, M., Adams, V.: Observations on the bisulfite reductase (P 582) isolated from Desulfotomaculum nigrificans. J. Bact., 120, 240-244 (1974) Bell, G. R. : Studies on electron transfer systems in Desulfovibrio: Purification and characterization of hydrogenase, catalase and superoxide dismutase. Ph.D. thesis, University of Georgia, Athens (1973) Bruschi, M., Le Gall, J., Hatchikian, E. C., Dubourdieu, M. : Cristallisation et propri6t~s d'un cytochrome intervenant dans la raduction du thiosulfate. Bull. Soc. Fr. Physiol. V6g. 15, 381 - 390 (1969) Burton, C. P., Akagi, J. M. : Observations on the rhodanese activity of Desulfotomaculum nigrificans. J. Bact. 107, 375- 376 (1971) Findley, J. E., Akagi, J. M. : Role of thiosulfate in bisulfite reduction as catalyzed by Desulfovibrio vulgaris. J. Bact. 103, 741 - 744 (1970) Fogo, J. K., Popowsky, M. : Spectrophotometric determination of hydrogen sulfide. Analyt. Chem. 21, 732-734 (1949) Gornall, A. G., Bardawill, G. J., David, M. M. : Determination of serum proteins by means of the biuret reaction. J. biol. Chem. 177, 751-766 (1949) Grant, W. M. : Colorometric determination of sulfur dioxide. Analyt. Chem. 19, 345- 346 (1947) Guarraia, L. J., Laishley, E. J., Forget, N., Peck, H. D., Jr. : The role of ferredoxin and flavodoxin in the sulfur metabolism of Desulfovibrio gigas. Bact. Proc. 1968, 129 Haschke, R. H. : Hydrogenase and thiosulfate reductase from Desulfovibrio vulgaris. Ph.D. thesis, University of Illinois, Urbana (1969) Haschke, R. H., Campbell, L. L.: Thiosulfate reductase of Desulfovibrio vulgaris. J. Bact. 106, 603- 607 (1971) Hatchikian, E. C., Le Gall, J.: Etude du m~tabolisme des acides dicarboxyliques et du pyruvate chez les bactOries sulfato-r6ductriees. I. Etude de l'oxydation enzymatique du fumarate en ac6tate. Ann. Inst. Pasteur 118, 125-142 (1970) Hatchikian, E. C., Le Gall, J. : Evidence for the presence of a b-type cytochrome in the sulfate-reducing bacterium Desulfovibrio gigas, and its role in the reduction of fumarate by molecular hydrogen. Biochim. biophys. Acta (Amst.) 267, 479-484 (1972) Hatchikian, E. C., Le Gall, J., Bruschi, M., Dubourdieu, M, : Regulation of the reduction of sulfite and thiosulfate by
255 ferredoxin, flavodoxin and cytochrome ec; in extracts of the sulfate reducer Desulfovibrio gigas. Biochim. biophys. Acta (Amst.) 258, 701-708 (1972) Irie, K., Kobayashi, K., Kobayashi, M., Ishimoto, M. : Biochemical studies on sulfate-reducing bacteria. XII. Some properties of flavodoxin from Desulfovibrio vulgaris. J. Biochem. (Tokyo)73, 353-366 (197.3) Ishimoto; M., Kondo, Y., Kameyama, T., Yagi, T., Shiraki, M. : The role of cytochrome in the enzyme system of sulfate-reducing bacteria. In: "Proceedings of the International Symposium on Enzyme Chemistry", Tokyo and Kyoto, p. 229-234, Maruzen, Tokyo (1958) Ishimoto, M., Koyama, T. : On the role of acytochrome in the thiosulfate reduction by a sulfate-reducing bacterium. Bull. chem. Soc. Jap. 28, 231-232 (1955) Ishimoto, M., Koyama, J.: Biochemical studies on sulfate reducing bacteria. VI. Separation of hydrogenase and thiosulfate reductase and partial purification of cytochrome and green pigment. J. Biochem. (Tokyo) 44, 233- 242 (1957) Ishimoto, M., Koyama, J.: Biochemical studies on sulfatereducing bacteria. IX. Sulfite reductase. J. Biochem. 49, 103-109 (1961) Ishimoto, M., Koyama, J., Nagai, Y.: Biochemical studies on sulfate reducing bacteria. IV. Reduction of thiosulfate by cell free extracts. J. Biochem. (Tokyo) 42, 4 1 - 5 3
(1955)
Jones, H. E., Skyring, G. W. : Effect of enzymic assay conditions on sulfite reduction catalyzed by desulfoviridin from Desulfovibrio gigas. Biochim. biophys. Acta (Amst.) 377, 5 2 - 60 (1975) Kaji, A., McElroy, W. D. : Mechanism of hydrogen sulfide formation from thiosulfate. J. Bact. 77, 630-637 (1959) Kobayashi, K., Tachibana, S., Ishimoto, M.: Intermediary formation of trithionate in sulfite reduction by a sulfatereducing bacteria. J. Biochem. (Tokyo)65, 155-157 (1969) Kobayashi, K.E., Takahashi, Ishimoto, M. : Biochemical studies on sulfate-reducing bacteria. XI. Purification and some properties of sulfite reductase, desulfoviridin. J. Biochem. (Tokyo) 72, 879-887 (1972) Lee, J. P.: Sulfite reduction in sulfate reducing bacteria. Ph.D. thesis, p. 39, University of Georgia, Athens (1972) Lee, J. P., Le Gall, J., Peck, H. D., Jr.: Isolation of assimilatory and dissimilatory-type sulfite reductases from Desulfovibrio vulgaris. J. Bact. 115, 529-542 (1973) Lee, J. P., Peck, H. D., Jr. : Purification of the enzyme redt~cing bisulfite to trithionate from Desulfovibrio gigas and its identification as desulfoviridin. Biochem. biophys. Res. Commun. 45, 583-589 (1971) Le Gall, J., Dragoni, N. : Dependance of sulfite reduction on a crystallized ferredoxin from Desutfovibrio gigas. Biochem. biophys. Res. Commun. 23, 145-149 (1966) Le Gall, J., Hatchikian, E. C.: Purification et propri6t6s d'une flavoprot6ine intervenant dans la r6duction du sulfite par Desulfovibrio gigas. C. R. Acad. Sci. 264, 2580- 2583 (1967) Le Gall, J., Mazza, G., Dragoni, N.: Le cytochrome c3 de Desulfovibrio gigas. Biochim. biophys. Acta (Amst.) 56, 385- 387 (1965) Le Gall, J., Postgate, j. R. : The physiology of sulfate-reducing bacteria. In: Advances in microbial physiology, Vol. 10, A. H. Rose, D. W. Tempest, eds., pp. 81 - 133. London: Academic Press 1973 Leinweber, F.J., Monty, K.J.: The metabolism of thiosulfate in Salmonella typhimurium. J. biol. Chem. 238, 3775- 3780 (1963)
256 Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R. J. : Protein measurement with folin phenol reagent. J. biol. Chem. 193, 265-275 (1951) Michaels, G. B., Davidson, J. T., Peck, H. D., Jr. : A flavinsulfite adduct as an intermediate in the reaction catalyzed by adenylyl sulfate reductase from Desulfovibrio vulgaris. Biochem. biophys. Res. Commun. 39, 321-328 (1970) Nakatsukasa, W., Akagi, J. M.: Thiosulfate reductase isolated from Desulfotomaculum nigrifieans. J. Bact. 98, 4 2 9 - 433 (1969) Oltmann, L., Schoenmaker, G. S., Stouthamer, A. H. : Solubilisation and purification of a cytoplasmic membrane bound enzyme catalyzing tetrathionate and thiosulfate reduction in Proteus mirabilis. Arch. Microbiol. 98, 1 9 - 30 (1974) Dr. C. Hatchikian Laboratoire de Chimie Bact6rienne, C.N.R.S. F-13274 Marseille Cedex 2, France
Arch. Microbiol., Vol. 105, No. 3 (1975) Peck, H. D., Jr., Deacon, T. E., Davidson, J. T." Studies on adenosine 5'-phosphosulfate reductase from Desulfovibrio desulfuricans and Thiobacillus thioparus. I. The assay and purification. Biochim. biophys. Acta (Amst.) 96, 429-446 (1965) Postgate, J. R. : Cytochrome Ca and desulfoviridine; pigments of the anaerobe D.desulphuricans. J. gen. Microbiol. 14, 5 4 5 - 572 (1956) S6rbo, B. : A colorimetric method for the determination of thiosulfate. Biochim. biophys. Acta (Amst.) 23, 412-416 (1957) Sub, B., Akagi, J. M. : Formation of thiosulfate from sulfite by Desulfovibrio vulgaris. J. Bact. 99, 210-215 (1969) Wang, S., Volini, M. : Studies on the active site of rhodanese. J. biol. Chem. 243, 5465-5470 (1968)
Purification and Properties of Thiosulfate Reductase from Desulfovibrio gigas E. C. HATCHIKIAN Laboratoire de Chimie Bact6rienne, C.N.R.S., Marseille Received June 2, 1975
Abstract. Thiosulfate reductase of the dissimilatory sulfatereducing bacterium Desulfovibrio gigas has been purified 415-fold and its properties investigated. The enzyme was unstable during the different steps of purification as well as during storage at -15~ The molecular weight of thiosulfate reductase estimated from the chromatographic behaviour of the enzyme on Sephadex G-200 was close to 220000. The absorption spectrum of the purified enzyme exhibited a protein peak at 278 nm without characteristic features in the visible region. Thiosulfate reductase catalyzed the stoichiometric production of hydrogen
sulfide and sulfite from thiosulfate, and exhibited tetrathionate reductase activity. It did not show sulfite reductase activity. The optimum pH of thiosulfate reduction occurred between pH 7.4 and 8.0 and its K,, value for thiosulfate was calculated to be 5 9 10.4 M. The sensitivity of thiosulfate reductase to sulfhydryl reagent and the reversal of the inhibition by cysteine indicated that one or more sulfhydryl groups were involved in the catalytic activity. The study of electron transport between hydrogenase and thiosulfate reductase showed that the most efficient coupling was obtained with a system containing cytochromes c3 (Mr = 13000) and c3 (Mr = 26000).
Key words." Desulfovibrio gigas - Purification, properties of - Thiosulfate reductase - Sulfate-reducing bacteria.
According to the most recent data the dissimilatory reduction of sulfite catalyzed by sulfate-reducing bacteria, proceeds primarily through a pathway containing trithionate and thiosulfate as intermediate compounds (Kobayashi etal., 1969; Suh and Akagi, 1969; Findley and Akagi, 1970). This pathway involves at least three enzymes (Lee and Peck, 1971; Lee etal., 1973). Lee and Peck (1971) reported the purification of the dissimilatory sulfite reductase (desulfoviridin) from Desulfovibrio gigas and the identification of its product as trithionate. However, Kobayashi etal. (1972) and Jones and Skyring (1975) reported that desulfoviridin formed trithionate from sulfite and, furthermore, that thiosulfate and sulfide were also products of this reaction. Recently, a confirmation of the trithionate pathway was obtained by Akagi et al. (1974) with the bisulfite reductase (P582) isolated from
Desulfotomaculum nigr~'cans. Kobayashi etal. (1969) demonstrated that trithionate and thiosulfate were sequentially accumulated during the reduction of sulfite by extracts of D. vulgaris and Lee (1972) observed the reduction of trithionate by a partially fractionated extract of D. gigas. These results established that trithionate reductase which catalyzes the second step of sulfite reduction is present in extracts of these bacteria; however, this enzyme was not further studied.
The ability of the sulfate reducing bacteria to reduce thiosulfate has been previously reported (Ishimoto etal., 1955) and thiosulfate reductase which catalyzes the terminal step of dissimilatory sulfite reduction, i.e. the reductive cleavage of thiosulfate to sulfide and sulfite, has been studied in detail in D. vulgaris (Ishimoto and Koyama, 1957; Haschke and Campbell, 1971) and D.nigr~'cans (Nakatsukasa and Akagi, 1969). It is to be noted that rhodanese activity (thiosulfate :cyanide sulfur transferase; EC 2.8.11) has been found in D.nigrificans by Burton and Akagi (1971). The authors postulated that such an activity might be necessary for the cleavage of the thiosulfate molecule during the reductive process. The electron tra.nsport between hydrogenase and thiosulfate reductase has been the subject of a number of studies. Cytochrome c3 (Mr = 13000) has been reported to act as an electron carrier in the reduction of thiosulfate and sulfite in D. vulgaris (Ishimoto and Koyama, 1955; Postgate, 1956) and more recently, ferredoxin and flavodoxin have been implicated in the same reactions in D.gigas (Le Gall and Dragoni, 1966; Le Gall and Hatchikian, 1967; Guarraia etal., 1968; Irie etal., 1973). A comparative study demonstrated that the reduction of thiosulfate by D.gigas extracts devoid of electron carriers can be stimulated
250
Arch. Microbiol., Vol. 105, No. 3 (1975)
by at least three proteins, a non heme-iron protein ferredoxin, a flavoprotein, flavodoxin and a heme-iron protein, cytochrome c3 (Mr = 26000)1; both ferredoxin and flavodoxin are also carriers for the reduction of sulfite whereas cytochrome c3 (Mr = 26000) is specific for the reduction of thiosulfate (Bruschi et al., 1969; Hatchikian etal., 1972). This paper reports the purification and some properties of the thiosulfate reductase from D.gigas and also its physiological relationship with several electron carriers.
Materials and Methods Cultivation of Bacteria and Preparation of Extracts. Desulfovibrio gigas was cultivated and harvested according to Le Gall etal. (1965). To extract cytochrome c3 (Mr = 13000) and hydrogenase the bacteria were washed as previously described (Le Gall etal., 1965; Hatchikian and Le Gall, 1972). The washed cells were suspended in 0.02 M potassium phosphate buffer, pH 7.6, and the extract as well as soluble and particulate fractions were prepared as already reported (Hatchiklan and Le Gall, 1970).
Enzyme Assays. Thiosulfate reductase activity was determined using the Warburg manometric method under hydrogen atmosphere in a system which contained methyl viologen, excess hydrogenase and thiosulfate at 37~ by following the consumption of hydrogen. The main compartment of each flask contained potassium phosphate (150 ~tmoles), methyl viologen (10 lamoles), hydrogenase preparation (500 lag protein), enzyme solution (0.02-0.15 ml) and the center wel! contained NaOH l0 N (0.05 ml). Na2S2Q (20 lamoles) was introduced from the side arm to give a total volume of 3 ml. The hydrogenase preparation utilized in our experiments was devoid of thiosulfate and sulfite reductase activities. It was prepared by the following procedure: the active fi'action not adsorbed on silica gel (Hatchikian and Le Gall, 1972) was passed through a Sephadex G-100 column; this treatment was sufficient to eliminate the reductases which are high molecular weight proteins. In some cases a purified hydrogenase was used, prepared from D.gigas as already reported (Bell, 1973). The electron carriers, cytochrome c3 (Mr = 13000), flavodoxin and cytochrome c3 (M~ = 26000) used in the coupling assays between hydrogenase and thiosulfate reductase were purified as previously described (Le Gall etal., ~1965; Le Gall and Hatchikian, 1967; Bruschi etal., 1969). Assays of sulfite reductase and tetrathionate reductase activities were carried out manometricalty as described for the thiosulfate reductase activity except that 20 lamoles of sulfite were used in the case of sulfite reductase and 5 lamoles of tetrathionate and 2 lamoles of benzyl viologen instead of methyl viologen in the case of the tetrathionate reductase. Rhodanese activity was measured according to the method of S6rbo (1957) modified by Wang and Volini (1968). Analytical Methods. The stoichiometry of the enzymatic formation of hydrogen sulfide and sulfite from thiosulfate in the manometric system was established as already reported (Hatchikian et al., 1972). The content of H2S in the center well was. measured by the colorimetric method of Fogo and Popowsky (1949) ; the content of sulfite in the reaction mixture of the main compartment of the Warburg vessel was deter1The name c% that was first given to cytochrome c3 (M. W. = 26000) cannot be accepted in the nomenclature of cytochromes.
mined according to the method of Grant (1947). The enzymatic formation of thiosulfate from tetrathionate in the manometric system was measured by iodometric titration using starch as indicator.
Protein Determidation. Protein was estimated by the biuret method (Gornall etal., 1949) and the method of Lowry etal. (1951).
Electrophoresis. Analytical gel electrophoresis was performed in 7 ~ polyacrylamide gel and the Tris-HCl glycine buffer was at pH 8.8. The current used was 3 mA per tube. Each gel was loaded with 30-50 lag protein, and protein bands were revealed with CoomasSie blue. For the determination of thiosulfate reductase directly in the gel after electrophoresis, the gels were first stained with reduced methyl viologen and then dipped into a solution of 40 mM potassium phosphate buffer (pH 7.6) containing 8 mM sodium thiosulfate. A clear band in the gel revealed enzyme activity. Absorption Spectra. Absorption spectra were measured in cells of 1 cm path length at room temperature with a Cary model 14 recording spectrophotometer.
Results Purification Scheme. The washed bacteria (600 g) were suspended in 350 ml of 20 m M potassium phosphate buffer (all buffers used in the purification were kept at p H 7.6) and the soluble protein extract was prepared as already described. The soluble protein fraction contained 6 5 - 7 0 ~ o of the thiosulfate reductase activity present in the crude extract. The first purification step, the chromatography of the suspension on DEAE-cellulose column (4 c m x 3 cm) essentially removed the acidic electron carriers of Desulfovibrio gigas [ferredoxin, flavodoxin, rubredoxin and cytochrome c3 (Mr = 26000)] which were eluted with 1 M Tris-HC1 buffer and stored for later purification. Most of the thiosulfate reductase activity was not retained on the DEAE-cellulose column. To this not-retained fraction solid a m m o n i u m sulfate was added to 20 ~o saturation. After centrifugation at 30 000 x g for 20 min the pellet was discarded and a m m o n i u m sulfate added to the supernatant to 4 0 ~ saturation. The precipitate obtained by centrifugation was dissolved in 150 ml of 50 m M phosphate buffer and dialyzed overnight against 10 1 10 m M of the same buffer. 75 ml of wet DEAE-cellulose equilibrated with 30 m M phosphate buffer were poured into the dialyzed solution (234 ml) and the mixture was stirred overnight at 4 ~ C. The DEAE-cellulose was separated by centrifugation, packed into a column and washed with 30 m M phosphate buffer. The enzyme activity was not retained on the DEAE-cellulose while most of the green pigment, desulfoviridin, previously present in the extract was adsorbed. The active fraction (292 ml) was then concentrated to 180 ml in a 350 ml ultrafiltration cell with a PM-10 filter (Amicon), centrifuged to remove a slight precipitate, a n d 30 ml fractions
E. C. Hatchikian: Thiosulfate Reductase from D.gigas applied to a Sephadex G-200 column (5 x 100 cm) equilibrated with 20 mM Tris-HC1 buffer. The column was eluted with the same buffer at a flow rate of 0.6 ml per min. The yellow coloured active fractions eluted from Sephadex G-200 columns were pooled and adsorbed on a DEAE-cellulose column (4 x 22 cm) equilibrated with 10 mM phosphate buffer. The enzyme was eluted by means of a non-linear phosphate gradient from 1 0 - 2 5 0 raM, concentrated to 72 ml as previously described, then applied to a Sephadex G-200 column. The active fraction, still containing flavoproteins was adsorbed on a calcium phosphate gel column (4 x 12 cm) equilibrated with 5 mM phosphate buffer and eluted by means of a phosphate gradient from 1 0 - 1 5 0 mM. Most of the flavoproteins, mainly adenosine 5'-phosphosulfate reductase (APSreductase), were retained on the column. The fraction containing thiosulfate reductase activity was adsorbed on a DEAE-cellulose column (3 x 7 cm) equilibrated with 10 m M phosphate buffer and the proteins were eluted with a phosphate gradient from 10-150 mM. The enzyme was eluted by 8 0 - 1 0 0 mM phosphate buffer. At this point, the enzyme had been purified 415-fold. The purification scheme is summarized in Table 1. The purified thiosulfate reductase was colourless. It exhibited a great instability and lost activity when frozen and thawed. The reductive reagents (/Lmercaptoethanol or dithiothreitol) had no effect on the instability.
Test of Homogeneity. As shown in Fig. 1 electrophoresis of the partially purified enzyme revealed the presence of four bands. The activity test in the gel followed by staining with Coomassie blue indicated that thiosulfate reductase activity was located between the two protein bands nearest the origin. It is noteworthy that after such an electrophoresis performed at pH 8.8, the enzymatic activity was very weak and was not located on an apparent band of protein. As reported later, the pH optimum of thiosulfate reductase occurs between 7.5 and 8 and at a pH value of 8.8 the activity is only 8 ~o of that obtained at pH 7.6. When the electrophoresis was run for 3 hrs with phosphate buffer at pH 7.6 instead of pH 8.8, the proteins just penetrated into the gel and showed one main band (Fig. 1). In this latter case, where the pH used is near the isoelectric point of thiosulfate reductase, the activity test in the gel with reduced methyl-viologen indicated a very strong enzymatic activity. Molecular Weight. The molecular weight of thiosulfate reductase was not accurately determined; however, the chromatographic behaviour of the enzyme on Sephadex G-200 during the purification compared with that of known D. gigas proteins allowed an
251 Table 1. Purification scheme of thiosulfate reductase Step
Volume Protein Total (ml) (mg/ml) units
Soluble proteins
Specific activity
520
70
7 200
0.20
After DEAE-cellulose column 534
48
5 874
0.23
After ammonium sulfate precipitation (20- 40 ~)
234
60
2 600
0.184
DEAE-eluant
292
34
2450
0.25
1280
0.75
1.5
800
t.4
After Sephade• G-200 (2nd) 298
0.75
500
2.3
After calcium phosphate gel column After DEAE column (3rd)
0.115 0.017 0.059
358 16.3 71.7 83 96.7 41
After SephadexG-200 446 After DEAE column (2nd) 388
151 [51 [ 40
3.85
Enzyme activity was determined as reported in "Materials and Methods". The thiosulfate reductase unit of activity was defined as the amount of enzyme which catalyzed the consumption of 1 ~tmoleof hydrogen per min.
estimation of its value. Thiosulfate reductase was eluted from the Sephadex G-200 column just before desulfoviridin which has a molecular weight of 200 000 (Lee and Peck, 1971) and at the same time as APSreductase which has a molecular weight of 220 000 in D. vulgaris (Michaels et at., 1970). These observations suggest that the value for the molecular weight of D.gigas thiosulfate reductase is about 220000.
Spectral Properties. Fig.2 gives the spectrum of a 280-fold purified thiosulfate reductase preparation. This enzyme fraction has an absorption spectrum without any special features; it shows a protein peak at 278 nm and a weak absorption at 380-400 nm. The weak absorption in the visible region may possibly be attributed to the presence of traces of APS-reductase (Peck etal., 1965) which has a chromatographic behaviour very similar to that of thiosulfate reductase. Substrates, Products and Stoichiometry. With thiosulfate as substrate and reduced methyl-viologen as electron donor, thiosulfate reductase from D.gigas catalyzed the stoichiometric production of hydrogen sulfide and sulfite. The enzyme also exhibited a slight catalytic activity with dithionite. This could be related to the formation of a small amount of thiosulfate detected after solution of the dithionite by the method of S6rbo (1957).
252
Arch. Microbiol., Vol. 105, No. 3 (1975)
Fig. 1. Disc acrylamide-gel electrophoresis of purified preparation of thiosulfate reductase from Desulfovibriogigas at several pH values. 35 lag protein of purified thiosulfate reductase (specific activity 83) were applied to each gel for electrophoresis: gel 1 was run at pH 8.8 in Tris-glycine buffer and gel 3 at pH 7.6 in phosphate buffer and stained for protein with Coomassie blue. Gels 2 and 4 were run respectively at pH 8.8 and 7.6 and thiosulfate reductase visualized after 7 min for gel 2 and 3 min for gel 4 by the thiosulfate dependant oxidation of reduced methyl viologen as described in "Materials and Methods"
Effect ofpH and Buffers. The pH optimum of reduced methyl viologen linked thiosulfate reductase activity was investigated with phosphate buffer (pH 6 . 6 - 8 ) , Tris-HC1 buffer (pH 7 . 2 - 8 . 7 ) and barbital-HC1 buffer (pH 7.0-8.7). The maximum activity of thiosulfate reductase from D.gigas was in the pH range 7 . 4 - 8 . 0 as shown in Fig. 3 and activity decreased rapidly above and below these values. The buffer in which thiosulfate reduction functioned most effectively was phosphate buffer. In this system, the enzyme activity was more than two times that in Tris-HC1 buffer. Assays with potassium phosphate, sodium phosphate, sodium chloride and potassium chloride indicated that phosphate anion was the effective ion.
/
0.7!
)-
0.5
Z [II nO
m 0.3
250
350
/-,50
550
650
WAVELENGTH (nm)
Fig. 2. The absorption spectrum of partially purified thiosulfate reductase from D.gigas. The protein concentration was 0.9 mg/ml and the specific activity was 56
Michaelis Constant. F r o m the data analyzed by a Lineweaver-Burk plot, the Km of the purified enzyme for thiosulfate was calculated to be 5. J0 -4 M at pH 7.6. With high concentrations of thiosulfate (about 10-fold the Km value) there was inhibition by excess substrate. Inhibition by Sulfhydryl Reagents. The data concerning
The partially purified enzyme exhibited tetrathionate reductase activity with reduced benzyl viologen as electron donor; tetrathionate is reduced to thiosulfate. However, the thiosulfate reductase was devoid of rhodanese activity and had no activity With sulfite and sulfate.
the inhibition of thiosulfate reductase are presented in Table 2. The results indicate that all the tested sulfhydry! reagents had an inhibitory effect. These compounds had no effect on hydrogenase activity at the concentrations used. The heavy metals, p-chloromercuribenzoate (pCMB), HgC12, AgNO3 and mer-
E. C. Hatchikian: Thiosulfate Reductase from D.gigas
253 Table 3: Thiosulfate reductase activity with the natural electron carriers
e-
g
m 40
System
lamoles H 2 consumed in 20 rain
Complete minus MV minus S202minus MV + I minus MV + II minus MV + III minus MV + I + II minus MV + I + I I I minus MV + II + I I I minus MV + I + II + I I I minus MV and $20~- + I + II + I I I
7.12 0 0 0.43 0.35 0.04 1.16 0.55 0.62 1.13 0
:zk
~30
c~20 o
-r-10
61 6
'
7.0 '
'
7./, '
'
7.'8
'
8 i2
'
8.6 ' pH
Fig. 3. The effect of pH on the activity of thiosulfate reductase. Assay mixtures were as described in "Materials and Methods" with potassium phosphate buffer, 150 Ixmoles (A -A) or Tfis-HC1 buffer, 150 lamoles (O O) or barbital buffer, 150 lamoles ( O - - - - Q ) and 35 gg of thiosulfate reductase (specific activity, 41)
Table 2. The effect of sulfhydryl inhibitors Inhibitor
Concentration ~ Inhibition
(M) p-chloromercuribenzoate
HgC12
2.10 s
100
8 - 10 . 6
55
5 - 10 .6
36
10 .5 5" 10 . 6
AgNO 3
10 5
100 66 90
Mersalyl
10 .5
35
Iodacetate
5.10 -4 10-4
70 33
N-ethylmaleimide
5 - 10-* 10 4
55 36
5 - 10 .5 10 -s 3.10 -3
52 22 45
o-iodosobenzoate Sulfite
Reaction mixtures were as described in "Materials and Methods" except that inhibitors at the indicated concentrations were incubated for 30 min with buffer and enzyme before starting the reaction. Purified thiosulfate reductase (30 lag protein) of specific activity 83 was utilized in all determinations.
salyl inhibited activity. In addition, alkylating reagents, iodacetate and N-ethylmaleimide as well as the oxidizing agent, o-iodosobenzoate, inhibited the activity of thiosulfate reductase. The enzyme activity was a!so inhibited by sulfite. As indicated in Table 2, 2 . 1 0 -5 M p C M B completely inhibited thiosulfate reductase activity, and the inhibition could be partially or completely reversed by cysteine (10 -4 M) according
The complete system contained the reagents as indicated in "Materials and Methods" except for MV (methyl viologen), 100 nmoles, pure hydrogenase, 400 lag protein and thiosulfate reductase (specific activity, 83) 30 lag protein. Electron carriers from D.gigas were added in the main compartment of the systems instead of MV as indicated. These systems contained (in nmoles): cytochrome c3 (M, = 13000) [I], 10, cytochrome c3 (Mr = 26000) [II], 10 and flavodoxin [III], 100.
to the incubation time of the enzyme with the inhibitor. Cysteine itself had no stimulatory effect on thiosulfate reductase. These results further substantiate that one or more sulfhydryl groups are involved in the catalytic activity of this enzyme. It is noteworthy that important variations in the inhibitory effect of reagents were observed with the experimental conditions (incubation time, p H and substrate concentration) more especially in the case of N-ethylmaleimide, o-iodosobenzoate and pCMB. These observations can be interpreted to indicate that more than one sulfhydryl group is involved in the activity of thiosulfate reductase and that the accessibility of these SH-groups are different.
Electron Donors. Reduced methyl viologen was the most efficient electron donor used by the thiosulfate reductase. With benzyl viologen the enzyme activity was only 8 ~o of that obtained with methyl viologen. No H2S was produced when the electron donors were cysteine, reduced glutathione, dihydrolipoate, reduced nicotinarnide adenine dinucleotide or reduced nicotinamide adenine dinucleotide phosphate.
Electron Carriers. In order to confirm the previous results obtained with a crude soluble protein fraction containing both sulfite and thiosulfate reductase activities separated from the electron carriers by a m m o n i u m sulfate precipitation (Hatchikian etal., 1972), assays
254
Arch. MicrobioL, Vol. i05, No. 3 (1975)
were performed with the partially purified thiosulfate red uctase, pure hydrogenase and the pure e!ectron carriers, cytochrome c3 (M~ = 26000), cytochrome c3 (M~ = 13 000) and flavodoxim The results are presented in 'Fable 3~ The best activity was observed with the systems containing both cytochrome ca (M~ = 13000) and cytochrome c3 (Mr = 26000) whereas the other systems had either no activity or only a slight activity. The activity exhibited by the most efficient system which contained cytochromes c3 (M~ = 13000) and c3 (M~ = 26000) reached about 16~o of the activity of the complete system with methyl viologen (100 nmoles).
Discussion The results of several investigators indicate that trithionate and thiosulfate are intermediates in the reduction of sulfite by the sulfate reducing bacteria (Kobayashi etal., !969; Findley and Akagi, 1970; Lee and Peck, 1971; Le Gall and Postgate, 1973; Akagi etaL, 1974). Thiosulfate reductase which catalyzes the reduction of thiosulfate into sulfide and sulfite as eariier reported by Ishimoto and Koyama (1957) is the terminal enzyme of the cyclic process of sulfite reduction postulated by Kobayashi etaL (1969). The thiosulfate reductase from Desulfovibrio gigas which had been purified 415-fold possessed a high catalytic activity in contrast to bisulfite reductase (Lee and Peck, 1971). The results obtained by analytical electrophoresis at pH 8.8 and 7.6 and by the activity test in the gels suggested that in our conditions of electrophoresis at pH 8.8 there was a dissociation of the thiosulfate reductase molecule and only a slight activity appeared where traces of the different subunits of the enzyme were present. In contrast the greater activity at pH 7.6 might be due to an unchanged enzyme structure. It is to be noted that Oltmann et aL (1974) have reported that in Proteus mirabilis, the thiosulfate reductase consists of two subunits. In addition to thiosnlfate reductase activity, the purified enzyme exhibits tetrathionate reductase activity as described recently for the thiosu~fate reduc~.ase from P.mirabilis (Oltmann etaL, 1974) whereas such an activity was not detected with the enzyme of DesuL Jbtomaculum ,~ligr(ficans (Nakatsukasa and Akagi, !969) and not investigated with the thiosulfate reductase of D. vulgaris (Haschke and Campbell, 197i). Whether tetrathionate reductase activity present in D.gigas is due to the tetrathionate reductase activity of thiosuifate reductase alone cannot be determined from our experiments. D. gigas thiosulfate reductase is devoid of rhodanese activity. This result shows that the reductive cleavage of thiosulfate is catalyzed by a
single protein and suggests that rhodanese is not implied in the thiosulfate reduction process of D.nigrificans as postulated by Burton and Akagi (1971). D.gigas thiosulfate reductase differs also from the other dissimilatory thiosulfate reductases (Nakatsukasa and Akagi, 1969; Haschke, 1969) by its narrow optimum pH range and its K~ value for thiosulfate. This latter value (5 9 10 .4 M) shows that the affinity of D.gigas thiosulfate reductase for thiosulfate is 200-fold greater than that of the similar enzyme of D. nigrificans (Nakatsukasa and Akagi, 1969). As reported for other thiosulfate reductases (Ishimoto and Koyama, 1961; Nakatsukasa and Akagi, 1969) the sensitivity of the enzyme from D.gigas to sulfhydryl reagents and the reversal of the inhibition by cysteine indicate that one or more sutfhydryl groups are involved in the cata!ytic activity. In contrast to assimilatory type thiosulfate reductases (Kaji and McElroy, 1959; Leinweber and Monty, 1963), D.gigas thiosulfate reductase as well as the enzyme from D. nigrificans (Nakatsukasa and Akagi, !969) were inactive with reduced glutathione and cysteine as electron donors. Perhaps the failure of the reductases of sulfate-reducing bacteria to function with these electron donors is a characteristic feature of the dissimilatory type thiosulfate reductases. The role of cytochrome c3 (Mr = 13000) in the electron transport between hydrogenase and thiosulfate reduc~:ase in extracts of D. vulgaris was earlier described (Ishimoto and Koyama, 1955; Postgate, 1956; Ishimoto etaL, 1958). Our previous results (Hatchikian etaL, 1972) indicated that cytochrome c3 (Mr = 26000), flavodoxin or ferredoxin, but not cytochrome c3 (Mr = 13000), stimulated thiosulfate reductase activity of an extract of D~gigas which contained reductase and hydrogenase activities separated from electron carriers by ammonium sulfate precipitation. With the purified thiosu!fate reductase and pure hydrogenase, the most efficient restoration of activity was obtained with the presence of both cytochrome c3 (Mr = 13000) and cytochrome c3 (Mr = 26 000) and the observations suggested that flavodoxin was inactive with this purified system. This discrepancy in our results concerning the implication of cytochrome c3 (M~ = 13 000) and flavodoxin in the coupling between hydrogenase and thiosutfate reductase deserves further comment. Firstly, the preparations obtained by ammonium sulfate precipitation of soluble protein extract were not completely devoid of cytochrome c3 and traces of this carrier could be sufficient to saturate the enzymatic extract. Secondly, the stimuiatory effect of flavodoxin could be explained by the presence of both thiosulfate and sulfite reductase activities in our previous enzymatic preparation. Thus the differences between our earlier observations and the
E. C. Hatchikian: Thiosulfate Reductase from D.gigas
results reported here could be due to the impure enzymatic preparations previously utilized. Another point to notice is the weak activity obtained with the natural electron carriers, a b o u t 16 ~ in the best case, as c o m p a r e d with the system containing methyl viologen. These observations suggest that another electron carrier m a y be implicated in the coupling between hydrogenase and thiosulfate reductase. H o w ever, it must be emphasized that our system is p r o b a b l y vastly different f r o m that occurring intra-cellularly.
Acknowledgements. The author expresses histhanks to Dr. J. Le Gall for his valuable advice and discussions throughout this work. I gratefully acknowledge the skillful technical assistance of Mrs. N. Forget. Thanks are due to Dr. M. Scandellari and Mr. R. Bourrelli for growing the sulfate reducing bacteria used in this study.
References Akagi, J. M., Chan, M., Adams, V.: Observations on the bisulfite reductase (P 582) isolated from Desulfotomaculum nigrificans. J. Bact., 120, 240-244 (1974) Bell, G. R. : Studies on electron transfer systems in Desulfovibrio: Purification and characterization of hydrogenase, catalase and superoxide dismutase. Ph.D. thesis, University of Georgia, Athens (1973) Bruschi, M., Le Gall, J., Hatchikian, E. C., Dubourdieu, M. : Cristallisation et propri6t~s d'un cytochrome intervenant dans la raduction du thiosulfate. Bull. Soc. Fr. Physiol. V6g. 15, 381 - 390 (1969) Burton, C. P., Akagi, J. M. : Observations on the rhodanese activity of Desulfotomaculum nigrificans. J. Bact. 107, 375- 376 (1971) Findley, J. E., Akagi, J. M. : Role of thiosulfate in bisulfite reduction as catalyzed by Desulfovibrio vulgaris. J. Bact. 103, 741 - 744 (1970) Fogo, J. K., Popowsky, M. : Spectrophotometric determination of hydrogen sulfide. Analyt. Chem. 21, 732-734 (1949) Gornall, A. G., Bardawill, G. J., David, M. M. : Determination of serum proteins by means of the biuret reaction. J. biol. Chem. 177, 751-766 (1949) Grant, W. M. : Colorometric determination of sulfur dioxide. Analyt. Chem. 19, 345- 346 (1947) Guarraia, L. J., Laishley, E. J., Forget, N., Peck, H. D., Jr. : The role of ferredoxin and flavodoxin in the sulfur metabolism of Desulfovibrio gigas. Bact. Proc. 1968, 129 Haschke, R. H. : Hydrogenase and thiosulfate reductase from Desulfovibrio vulgaris. Ph.D. thesis, University of Illinois, Urbana (1969) Haschke, R. H., Campbell, L. L.: Thiosulfate reductase of Desulfovibrio vulgaris. J. Bact. 106, 603- 607 (1971) Hatchikian, E. C., Le Gall, J.: Etude du m~tabolisme des acides dicarboxyliques et du pyruvate chez les bactOries sulfato-r6ductriees. I. Etude de l'oxydation enzymatique du fumarate en ac6tate. Ann. Inst. Pasteur 118, 125-142 (1970) Hatchikian, E. C., Le Gall, J. : Evidence for the presence of a b-type cytochrome in the sulfate-reducing bacterium Desulfovibrio gigas, and its role in the reduction of fumarate by molecular hydrogen. Biochim. biophys. Acta (Amst.) 267, 479-484 (1972) Hatchikian, E. C., Le Gall, J., Bruschi, M., Dubourdieu, M, : Regulation of the reduction of sulfite and thiosulfate by
255 ferredoxin, flavodoxin and cytochrome ec; in extracts of the sulfate reducer Desulfovibrio gigas. Biochim. biophys. Acta (Amst.) 258, 701-708 (1972) Irie, K., Kobayashi, K., Kobayashi, M., Ishimoto, M. : Biochemical studies on sulfate-reducing bacteria. XII. Some properties of flavodoxin from Desulfovibrio vulgaris. J. Biochem. (Tokyo)73, 353-366 (197.3) Ishimoto; M., Kondo, Y., Kameyama, T., Yagi, T., Shiraki, M. : The role of cytochrome in the enzyme system of sulfate-reducing bacteria. In: "Proceedings of the International Symposium on Enzyme Chemistry", Tokyo and Kyoto, p. 229-234, Maruzen, Tokyo (1958) Ishimoto, M., Koyama, T. : On the role of acytochrome in the thiosulfate reduction by a sulfate-reducing bacterium. Bull. chem. Soc. Jap. 28, 231-232 (1955) Ishimoto, M., Koyama, J.: Biochemical studies on sulfate reducing bacteria. VI. Separation of hydrogenase and thiosulfate reductase and partial purification of cytochrome and green pigment. J. Biochem. (Tokyo) 44, 233- 242 (1957) Ishimoto, M., Koyama, J.: Biochemical studies on sulfatereducing bacteria. IX. Sulfite reductase. J. Biochem. 49, 103-109 (1961) Ishimoto, M., Koyama, J., Nagai, Y.: Biochemical studies on sulfate reducing bacteria. IV. Reduction of thiosulfate by cell free extracts. J. Biochem. (Tokyo) 42, 4 1 - 5 3
(1955)
Jones, H. E., Skyring, G. W. : Effect of enzymic assay conditions on sulfite reduction catalyzed by desulfoviridin from Desulfovibrio gigas. Biochim. biophys. Acta (Amst.) 377, 5 2 - 60 (1975) Kaji, A., McElroy, W. D. : Mechanism of hydrogen sulfide formation from thiosulfate. J. Bact. 77, 630-637 (1959) Kobayashi, K., Tachibana, S., Ishimoto, M.: Intermediary formation of trithionate in sulfite reduction by a sulfatereducing bacteria. J. Biochem. (Tokyo)65, 155-157 (1969) Kobayashi, K.E., Takahashi, Ishimoto, M. : Biochemical studies on sulfate-reducing bacteria. XI. Purification and some properties of sulfite reductase, desulfoviridin. J. Biochem. (Tokyo) 72, 879-887 (1972) Lee, J. P.: Sulfite reduction in sulfate reducing bacteria. Ph.D. thesis, p. 39, University of Georgia, Athens (1972) Lee, J. P., Le Gall, J., Peck, H. D., Jr.: Isolation of assimilatory and dissimilatory-type sulfite reductases from Desulfovibrio vulgaris. J. Bact. 115, 529-542 (1973) Lee, J. P., Peck, H. D., Jr. : Purification of the enzyme redt~cing bisulfite to trithionate from Desulfovibrio gigas and its identification as desulfoviridin. Biochem. biophys. Res. Commun. 45, 583-589 (1971) Le Gall, J., Dragoni, N. : Dependance of sulfite reduction on a crystallized ferredoxin from Desutfovibrio gigas. Biochem. biophys. Res. Commun. 23, 145-149 (1966) Le Gall, J., Hatchikian, E. C.: Purification et propri6t6s d'une flavoprot6ine intervenant dans la r6duction du sulfite par Desulfovibrio gigas. C. R. Acad. Sci. 264, 2580- 2583 (1967) Le Gall, J., Mazza, G., Dragoni, N.: Le cytochrome c3 de Desulfovibrio gigas. Biochim. biophys. Acta (Amst.) 56, 385- 387 (1965) Le Gall, J., Postgate, j. R. : The physiology of sulfate-reducing bacteria. In: Advances in microbial physiology, Vol. 10, A. H. Rose, D. W. Tempest, eds., pp. 81 - 133. London: Academic Press 1973 Leinweber, F.J., Monty, K.J.: The metabolism of thiosulfate in Salmonella typhimurium. J. biol. Chem. 238, 3775- 3780 (1963)
256 Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R. J. : Protein measurement with folin phenol reagent. J. biol. Chem. 193, 265-275 (1951) Michaels, G. B., Davidson, J. T., Peck, H. D., Jr. : A flavinsulfite adduct as an intermediate in the reaction catalyzed by adenylyl sulfate reductase from Desulfovibrio vulgaris. Biochem. biophys. Res. Commun. 39, 321-328 (1970) Nakatsukasa, W., Akagi, J. M.: Thiosulfate reductase isolated from Desulfotomaculum nigrifieans. J. Bact. 98, 4 2 9 - 433 (1969) Oltmann, L., Schoenmaker, G. S., Stouthamer, A. H. : Solubilisation and purification of a cytoplasmic membrane bound enzyme catalyzing tetrathionate and thiosulfate reduction in Proteus mirabilis. Arch. Microbiol. 98, 1 9 - 30 (1974) Dr. C. Hatchikian Laboratoire de Chimie Bact6rienne, C.N.R.S. F-13274 Marseille Cedex 2, France
Arch. Microbiol., Vol. 105, No. 3 (1975) Peck, H. D., Jr., Deacon, T. E., Davidson, J. T." Studies on adenosine 5'-phosphosulfate reductase from Desulfovibrio desulfuricans and Thiobacillus thioparus. I. The assay and purification. Biochim. biophys. Acta (Amst.) 96, 429-446 (1965) Postgate, J. R. : Cytochrome Ca and desulfoviridine; pigments of the anaerobe D.desulphuricans. J. gen. Microbiol. 14, 5 4 5 - 572 (1956) S6rbo, B. : A colorimetric method for the determination of thiosulfate. Biochim. biophys. Acta (Amst.) 23, 412-416 (1957) Sub, B., Akagi, J. M. : Formation of thiosulfate from sulfite by Desulfovibrio vulgaris. J. Bact. 99, 210-215 (1969) Wang, S., Volini, M. : Studies on the active site of rhodanese. J. biol. Chem. 243, 5465-5470 (1968)
