Electron Paramagnetic Resonance Spectroscopic

0 downloads 0 Views 256KB Size Report
The N5-methyltetrahydromethanopterin:coenzyme M methyltransferase is a membrane-bound ... nine synthase, the methyl group is transferred to homocysteine.
JOURNAL OF BACTERIOLOGY, May 1995, p. 2245–2250 0021-9193/95/$04.0010 Copyright q 1995, American Society for Microbiology

Vol. 177, No. 9

Electron Paramagnetic Resonance Spectroscopic and Electrochemical Characterization of the Partially Purified N5-Methyltetrahydromethanopterin:Coenzyme M Methyltransferase from Methanosarcina mazei Go ¨1† WEI-PING LU,1‡ BURKHARD BECHER,2 GERHARD GOTTSCHALK,2*

AND

STEPHEN W. RAGSDALE1*

¨r Mikrobiologie Department of Biochemistry, University of Nebraska, Lincoln, Nebraska 68583-0718,1 and Institut fu der Georg-August Universita ¨t, Go ¨ttingen, Germany2 Received 14 October 1994/Accepted 16 February 1995

synthesis. Other methyltransferases that cycle between the Co11 and methyl-Co31 states include a methanol:vitamin B12 methyltransferase involved in conversion of methanol to methane (29, 30) and a C/Fe-SP from Methanosarcina thermophila involved in methanogenesis from acetic acid (15). The results reported here demonstrate that the effect of ATP is to lower the redox potential required for reductive activation from approximately 2450 mV in the absence of ATP to ;2235 mV in its presence. We also have characterized the cobamide center and a [4Fe-4S] cluster which are required for activity.

Methane is a major terminal product of the anaerobic degradation of biomass. It is produced by methanogenic archaea from substrates such as acetate, H2 1 CO2, methanol, methylamines, and formate. Two reactions involved in methanogenesis from H2 1 CO2 are coupled to energy-conserving ion translocations: the H2-heterodisulfide reductase functions as a primary proton pump (6) and the corrinoid enzyme N5methyltetrahydromethanopterin (CH3-H4MPT): coenzyme M (CoM-SH) methyltransferase (methyltransferase) as a primary sodium ion pump (4). Methyltransferase is a membrane-associated protein (4, 11) that has been characterized and partially purified from Methanobacterium thermoautotrophicum Marburg (12), Methanosarcina mazei Go ¨1 (4), and M. thermoautotrophicum DH (17). For full activity, the methyltransferase requires ATP and a reductant, such as Ti(III) citrate, for activity (4, 12, 17, 18). Corrinoid-containing methyltransferases that are well characterized include methionine synthase (2, 22) and a corrinoid/ iron-sulfur protein (C/Fe-SP) from Clostridium thermoaceticum (14, 24). These proteins utilize Co11 as a supernucleophile to displace the methyl group of methyltetrahydrofolate/(CH3H4folate) and form H4folate and methyl-Co31. For methionine synthase, the methyl group is transferred to homocysteine to form methionine. In the case of the C/Fe-SP, the methyl group is transferred to another protein, carbon monoxide dehydrogenase, to initiate the final steps in acetyl coenzyme A

MATERIALS AND METHODS Growth of the organism. M. mazei Go ¨1 (DSM 3647) was cultivated in 1 or 60 liters and harvested anaerobically, essentially as described before (4). Solubilization of cell membrane. Cell-free protoplasts and the membrane fraction were prepared in an anaerobic chamber (Coy Laboratory, Ann Arbor, Mich.) as described previously (4). The membrane sample used for electron paramagnetic resonance (EPR) and redox titration of the cobamide center was prepared as follows. The membrane in 40 mM potassium phosphate (KPi) buffer, pH 7.0, was treated with 1% (wt/vol) of CHAPS [3-(3-cholamidopropyl dimethylammonio)-1-propanesulfonate] for 10 min and then centrifuged at 200,000 3 g and 48C for 1 h. The supernatant contained most (90%) of the membrane hydrogenase activity and little cobamide and was discarded. This treatment greatly reduced the viscosity of the membrane fraction, allowing a higher concentration (;100 mg/ml) of membrane sample to be prepared. Octylglucoside was later found to be a more effective solubilizer and less destructive to the enzyme activity than CHAPS. The membrane fraction was incubated with 1% n-octylglucoside for 30 min with gentle stirring at 168C. Vigorous stirring was avoided, since it caused dissociation of cobamide from the methyltransferase. After centrifuging the membrane solution at 200,000 3 g for 2 h, a supernatant (S1), a pellet, and a fraction (S2) between them were decanted separately. The S1 fraction contained about 70% and the S2 fraction contained the rest of the recovered enzyme activity (Table 1). Partial purification of the enzyme. The solubilized enzyme was partially purified by hydroxyapatite (HPT) chromatography. The S1 fraction (40 mg in 2 mM KPi, pH 7.0) was loaded on an HPT column (1.5 by 4 cm) equilibrated with 1 mM KPi, pH 7.0, containing 1 mM CaCl2 and 1 mM dithiothreitol (DTT) and eluted with increasing concentrations of KPi in steps of 1 mM (12 ml), 10 mM (15 ml), 20 mM (15 ml), 40 mM (15 ml), and finally 400 mM KPi containing 1 mM DTT. Although the enzyme usually eluted in the 10 mM fraction, sometimes substan-

* Corresponding author. Mailing address for Stephen W. Ragsdale: Department of Biochemistry, East Campus, University of Nebraska, Lincoln, Nebraska 68583-0718. Phone: (402) 472-2943. Fax: (402) 4727842. Electronic mail address: (Internet) [email protected]. † This paper has been assigned Journal Series number 11036 of the Agricultural Research Division, University of Nebraska. ‡ Present address: Anti-Infective Research, Procter & Gamble Pharmaceuticals, Sharon Woods Technical Center, Cincinnati, OH 45241. 2245

Downloaded from http://jb.asm.org/ on June 21, 2017 by guest

The N5-methyltetrahydromethanopterin:coenzyme M methyltransferase is a membrane-bound cobalamincontaining protein of Methanosarcina mazei Go ¨1 that couples the methylation of coenzyme M by methyltetrahydrosarcinopterin to the translocation of Na1 across the cell membrane (B. Becher, V. Mu ¨ller, and G. Gottschalk, J. Bacteriol. 174:7656–7660, 1992). We have partially purified this enzyme and shown that, in addition to the cobamide, at least one iron-sulfur cluster is essential for the transmethylation reaction. The membrane fraction or the partly purified protein contains a ‘‘base-on’’ cobamide with a standard reduction potential (E*0) for the Co21/11 couple of 2426 mV. The iron-sulfur cluster appears to be a [4Fe-4S]21/11 type with an E*0 value of 2215 mV. We have determined the methyltransferase activity at various controlled redox potentials and demonstrated that the enzyme activity is activated by a one-electron reduction with halfmaximum activity occurring at 2235 mV in the presence of ATP and 2450 mV in its absence. No activation was observed when ATP was replaced by other nucleoside triphosphates or nonhydrolyzable ATP analogs.

2246

LU ET AL.

J. BACTERIOL. TABLE 1. Purification of the methyltransferase

Fraction

Total protein (mg)

Crude vesicles Cytoplasm Membrane S1 (membrane solubilization) 10 mM Pi (HPT chromatography)

10,334 9,588 688 140 25

a b

Sp acta (U/mg)

0.06 (0.010) ND (,0.010) 0.5 (0.084) ND (0.199) 4.0 (0.422)

% Recovery of activity

Cobamide content (nmol/mg)

% Cobamide recovery

100 48 18

NDb ND 0.7 2.6 5.5

100 75 28

The values in parentheses were obtained with CH3-H4folate as the methyl donor in the enzyme assay. ND, not done.

Eh 5 E90 1 59/n 3 log

S D [ox] [red]

(1)

Enzyme assays at controlled redox potentials. The general methodology and the electrochemical cell used for performing enzyme assays at controlled redox potentials have been reported (20). Reaction mixtures contained 0.34 mg of protein from the HPT purification step, 10 mM CH3-H4folate, 5 mM CoM-SH, 1 mM DTT, 0.1% octylglucoside, 1 (or 0) mM ATP, 1 mM MgSO4, and 40 mM KPi buffer, pH 7, in a final volume of 100 ml. For reactions controlled at 2150 to 2350 mV, two redox mediators (50 mg each), benzyl viologen (E90 5 2350 mV) and anthraquinone-1,5-disulfonate (E90 5 2174 mV), were included. For reactions controlled at the range of 2350 mV to 2550 mV, methyl viologen (E90 5 2450 mV) and N,N9-trimethylene-2,29-dipyridinium bromide (TRIQUAT; E90 5 2540 mV) were used. The reaction mixture was poised and equilibrated at the desired redox potential for about 5 to 10 min at 378C before CH3-H4folate was added to start the reaction. The enzyme activity was measured as described above. The same protein sample gave a specific activity of 350 mU/mg when assayed under normal conditions with Ti(III) citrate (0.5 mM) as a reductant. The redox dyes used were found to have no effect on the enzyme activity in

separate experiments. We were unable to establish redox potentials below 2420 mV, because a trace amount of hydrogenase was present in the sample. This was problematic because at these low potentials, hydrogenase uses the reducing equivalents to reduce protons to hydrogen gas. To partially compensate for this, we used a chemical titrant. About 2 ml of 80 mM Ti(III) citrate was added to the mixture to establish a potential of ;2480 mV. The potential could then be maintained at various values between 2430 and 2480 mV for ;5 min with the potentiostat. When potentials lower than 2500 mV were required, more Ti(III) citrate (1 to 2 ml of an 80 mM concentration) had to be added to reach and maintain the acquired potentials. Thus, when the recorded potentials were below 2420 mV, the solutions were not fully at redox equilibrium. Analytical techniques. Protein concentrations were determined by the Bradford assay (8). The amount of protein-bound cobamide was determined spectrophotometrically after conversion of the cobamide to the dicyano form (19) as previously described (24). The metal content was analyzed by plasma emission spectroscopy (16) at the Chemical Analysis Laboratory, University of Georgia. Chemicals. CH3-H4folate, CoM-SH, ATP, CHAPS, and octylglucoside were purchased from Sigma. TRIQUAT was synthesized as previously reported (26). Methyl-H4MPT was a gift from D. A. Grahame (University of Uniformed Services, Bethesda, Md.).

RESULTS AND DISCUSSION Partial purification of the methyltransferase. Crude vesicles of M. mazei Go ¨1 catalyzed the transfer of a methyl group from CH3-H4MPT or from CH3-H4folate to CoM-SH at a specific rate of 0.05 to 0.08 and 0.01 to 0.017 U mg21, respectively. After centrifugation and three wash steps, the specific activity increased to 0.4 to 0.9 U mg21 with CH3-H4MPT or 0.07 to 0.12 U mg21 with CH3-H4folate as the methyl group donor. Solubilization with 1% octyl-b-D-glucopyranoside yielded preparations with specific activities of 0.7 to 1.3 and 0.16 to 0.2 U mg21, respectively. Chromatography of this solution on HPT increased the specific activity 2.3- to 3-fold (Table 1). Metal analysis revealed that the enzyme sample after HPT chromatography contained 7 to 10 g-atom of Fe and ;0.5 g-atom of Co per mol, on the basis of 80% purity and a molecular mass of 100 kDa. These values are similar to those reported for the methyltransferase of M. thermoautotrophicum (12). After SDS-polyacrylamide gel electrophoresis (PAGE) (27) and silver staining (7), two major protein bands corresponding to molecular masses of 55 and 25 kDa and three weakly stained bands at 30, 23, and 18 kDa were observed (Fig. 1). A mild method of denaturating proteins in which the protein is directly applied (without boiling or treatment with SDS) to a standard SDS-polyacrylamide gel and run at low voltage at 168C has been developed (21). In order to observe the pink and brown color of the unstained protein, 12-fold-higher amounts were applied to the gel. The 55-kDa band was brown and the 25-kDa band was pink (data not shown), indicating that the iron-sulfur cluster is a component of the 55-kDa subunit and the cobamide is associated with the 25-kDa subunit. After SDS-PAGE was performed as described by Scha¨gger and von Jagow (27) and the gel was stained with silver nitrate (7), two additional small protein bands, corresponding to molecular masses of 13 and 9 kDa, were observed (18a).

Downloaded from http://jb.asm.org/ on June 21, 2017 by guest

tial activity was found in the 20 and 40 mM fractions. Several other chromatographic methods, including ion exchange on DEAE-Sephacel or Q Sepharose and hydrophobic interaction on phenyl Sepharose, were attempted. These procedures resulted in a more than 70% loss of activity with a minimal increase in specific activity. Assay of the methyltransferase. The reaction mixture contained 10 to 100 mg of protein, 6 mM CoM-SH, 10 mM CH3-H4folate or 3 mM CH3-H4MPT, 0.5 mM ATP, 1 mM MgSO4, 1 mM DTT, 0.7 mM Ti(III) citrate, and 40 mM KPi, pH 7.0, in a final volume of 100 ml. The reaction was run at 378C and was started by adding the pterin. Aliquots (5 ml) were withdrawn at different intervals and analyzed for release of thiol with dithiobisnitrobenzene (DTNB) (10). Ti(III) citrate was prepared from TiCl3 (32). The enzyme unit is defined as the micromoles of CoM-SH methylated by CH3-H4folate or CH3-H4MPT per minute. Reactions performed with nucleoside triphosphates were performed as just described except that they contained 1.6 mM Ti(III) citrate and 2 mM nucleoside phosphates. To the assays containing ITP, GTP, and UTP, 10 mM anoxic solutions of hexokinase (2.8 U) and glucose (10 mM) were added. Inhibition of the enzyme with propyl iodide was reversed by illuminating the sample for 15 to 20 s with a Mineralight (San Gabriel, Calif.) UVGL-25 lamp at a distance of 30 cm. EPR spectroscopy and redox titrations. EPR spectra were recorded on a Bruker ESP 300E spectrometer equipped with an Oxford ITC4 temperature controller, an automatic frequency counter (model 5340A; Hewlett-Packard Co.), and a gaussmeter. Spectroscopic parameters are given in the figure legends. The spin concentration was calculated by comparing the double integral (with the supplied Bruker software) of a spectrum with that of a 1 mM copper perchlorate standard. The saturation behavior of the iron-sulfur cluster was evaluated by measuring the EPR signal intensity at various microwave powers. The value of P1/2, the power for half-saturation with microwave power, was obtained from simulation of experimental datum points with an empirical formula (25). Redox titrations were performed in an electrochemical cell with a fused quartz EPR tube as reported before (13). In a typical experiment, 50 mg of each redox mediator was added to the cell and made anaerobic by several cycles of vacuum and nitrogen. Approximately 400 ml of an anaerobic solution of the methyltransferase (10 to 20 mg) in 50 mM Tris-HCl, pH 7.6, containing 40 mM KCl was then added, and the redox potential was poised with a CV-27 potentiostat (Bioanalytical System Inc., West Lafayette, Ind.). The ambient potential of the solution, Eh, was measured after the solution had reached equilibrium, i.e., the drift in measured potential was less than ;5 mV/min. The sample was then shaken from the cell into the EPR tube and frozen in liquid nitrogen for spectroscopic analysis. Redox titrations were performed in both oxidative and reductive directions. All redox potentials are in reference to the standard hydrogen electrode. Experimental data were fitted to a linearized form of the Nernst equation (equation 1) with the least-squares method to obtain the values of the standard reduction potential, E90, measured in millivolts, and the number (n) of electrons involved in the redox reaction.

VOL. 177, 1995

CH3-H4MPT:METHYLTRANSFERASE FROM M. MAZEI Go ¨1

2247

TABLE 2. Effects of nucleoside triphosphates on the activity of partially purified methyltransferasea Addition

ATP ........................................................................................... None .......................................................................................... ADP........................................................................................... ITP............................................................................................. GTP ........................................................................................... UTP ........................................................................................... AMP-PNPd ............................................................................... ATP-29,39-dialdehyde ..............................................................

Activityb (%)

100c 17 13 18 16 23 12 9

a Partially purified methyltransferase (21.5 mg of protein per 100-ml assay) was preincubated for 5 min with Ti(III) citrate and nucleoside triphosphate before the reaction was started with CH3-H4MPT. b Initial velocity was determined from the linear portion of the reaction curve. c One hundred percent activity corresponds to 3.1 U/mg. d AMP-PNP, 59-adenylylimidodiphosphate.

Downloaded from http://jb.asm.org/ on June 21, 2017 by guest

FIG. 1. SDS-PAGE of the partially purified methyltransferase. Lane 1, 8 mg of protein from HTP chromatography; lane 2, prestained protein markers (molecular masses, 105, 74, 42, 27, 18, and 15 kDa). The gel was stained with Coomassie blue.

Dependence of the enzyme activity on redox potential and the effect of ATP. The methyltransferases from M. mazei Go ¨1, Methanosarcina barkeri, and M. thermoautotrophicum have been shown to require ATP and reductive activation by titanium(III) or reduced ferredoxin (3, 4, 12). In order to better understand these activation processes, we performed the methyltransferase reaction at various redox potentials in the presence and absence of ATP. As shown in Fig. 2, under normal assay conditions in the presence of ATP, the activity increased as the potentials were lowered from 2150 to 2300 mV following the Nernst equation for a one-electron process with half-maximum activity occurring at 2235 mV. However, when ATP was omitted from the reaction mixture, a much lower potential was required for activity. In this case, half-maximum activity was observed at 2450 mV. The same maximum enzyme activity was obtained in the presence or absence of ATP, indicating that the requirement for ATP could be fully supplanted by increasing the strength of the electron donor. Thus, ATP made the enzyme over 4,300-fold easier to reduce, corresponding to a DDG of 20.7 kJ mol21, which compares well with the free energy of hydrolysis of ATP. This is reminiscent

FIG. 2. Effects of redox potential on the activity of partially purified methyltransferase in the presence (F) or absence (E) of ATP. The reaction was performed in an electrochemical cell as detailed in Materials and Methods. The lines were generated with values of n 5 1 and E90 5 2235 mV (F) or 2450 mV (E) versus the standard hydrogen electrode (SHE). These values were obtained from analysis by the Nernst equation.

FIG. 3. EPR spectra of cob(II)amide in the membrane (A) and the partially purified methyltransferase (B). (A) The CHAPS-treated (100 mg ml21) membrane was prepared as described in Materials and Methods. The membrane was in 0.1 M KPi buffer, pH 7, containing 1% CHAPS. The intense triplet centered at g 5 2.002 arises from a species of unknown origin. The spectrometer conditions were as follows: microwave power, 20 mW; modulation amplitude, 10 G; modulation frequency, 100 kHz; microwave frequency, 9.455 GHz; sample temperature, 70 K; and receiver gain, 2 3 104. (B) The enzyme purified from the HPT column (7 mg/ml) was in 20 mM KPi buffer, pH 7, containing 0.1% octylglucoside. EPR conditions were the same as for panel A, except that microwave power was 40 mW. (Inset) Redox titration of the cobamide in the membrane fraction. The titration was performed in the electrochemical and EPR cell as described in Materials and Methods. The titration mixture contained membrane protein (50 mg ml21) and redox mediators (benzyl viologen, methyl viologen, and TRIQUAT) in 0.1 M KPi buffer, pH 7. The peak height at g 5 2.3 was measured at each potential (Eh). The line was generated from the Nernst equation, with E90 5 2426 mV and n 5 1. In a separate experiment, 1 mM ATP was added to the titration mixture, and the datum points are shown as open circles.

2248

LU ET AL.

J. BACTERIOL.

of ATP’s role in the nitrogenase, in which a conformational change of the Fe protein is effected by ATP hydrolysis, facilitating reduction of the Mo-Fe protein by the Fe protein (31). In order to evaluate whether ATP hydrolysis is involved in stimulation of the methyltransferase activity, we performed the

reaction in the presence of ADP, other nucleoside triphosphates, and two nonhydrolyzable analogs of ATP (Table 2). Since only ATP stimulated the methyltransferase activity, it appears that activation requires ATP hydrolysis. We do not yet know how ATP hydrolysis is coupled to decreasing the barrier for reductive activation of methyltransferase. ATP could alter the E90 for the Co21/11 couple of the cobamide, making the Co11 state more easily available for reaction. Although a complete titration was not performed in the presence of ATP, our results (Fig. 2) indicate that ATP did not markedly affect the E90 of the Co21/11 couple. This was unexpected, since when a chemical reaction occurs selectively with the reduced state of a redox couple (Co11 in this case), one expects the apparent E90 to increase. For example, when the midpoint potential for the cobalamin-dependent methionine synthase was determined in the presence of substrate, CH3-H4folate, the redox potential shifted 80 mV in the positive direction (1). In the presence of the activator, S-adenosylmethionine, the reduction of Co21 could be observed at potentials as high as 2200 mV. Possibly the effect of ATP on the Co21/11 couple would not be observed unless the methyl donor is present. Such a requirement would indicate that a ternary complex with ATP, CH3-H4MPT, and enzyme is required for the reaction.

TABLE 3. Inhibition of methyltransferasea

FIG. 5. Redox titration of the iron-sulfur cluster in partially purified methyltransferase. The samples were prepared at various redox potentials. EPR spectra were recorded at the conditions described in the legend to Fig. 4A. (Inset) The amplitude of the feature at g 5 1.92 was plotted versus redox potential. The solid line is a theoretical curve defined by the Nernst equation with n 5 1 and E90 5 2215 mV.

Inhibitor (mM)

Time of incubation (min)

Final concn in reaction mixture (mM)

% Inhibition

Iodoacetamide (200) TPCK (200) Propyl iodideb (400) Ferricyanidec (400)

5 5 15 120

20 20 40 400

100 100 100# 80

a The S1 fraction (specific activity 5 0.12 U mg21) was preincubated with inhibitor before transfer into an assay mixture. b Inhibition was reversed by 65% after illuminating the reaction mixture for 15 to 20 s. c The concentration of Ti(III) citrate was 1.4 mM in this reaction mixture.

Downloaded from http://jb.asm.org/ on June 21, 2017 by guest

FIG. 4. EPR spectrum of the iron-sulfur cluster of partially purified methyltransferase. (A) The partially purified enzyme (5 mg ml21) in KPi buffer, pH 7, was reduced by dithionite (0.5 mM). The spectrum was recorded at the following conditions: temperature, 10 K; microwave power, 40 mW; microwave frequency, 9.4468 GHz; modulation amplitude, 10 G; and receiver gain, 2 3 104. (B and C) The temperature and microwave power, respectively, were varied.

VOL. 177, 1995

CH3-H4MPT:METHYLTRANSFERASE FROM M. MAZEI Go ¨1

(Fig. 5, inset). Resolution of the discrepancy between the iron quantitation and the EPR results will require further studies of highly purified enzyme. The partially purified enzyme displayed a nearly isotropic EPR signal with a g value of ;2.02 and 0.2 spins per mol (data not shown). The signal resembles the spectrum of [3Fe-4S] clusters (5) and disappeared upon reduction. Given the lowspin concentration, it is likely that it represents a small amount of degraded cluster. Further studies of this EPR signal were not performed. Inhibition of the enzyme activity. Treatment with TPCK (N-tosyl-L-phenylalanine chloromethyl ketone) or iodoacetamide (final concentration, 200 mM) for 5 min resulted in complete inactivation of the enzyme (Table 3). The enzyme is also sensitive to ferricyanide; it is likely that ferricyanide inhibits the methyltransferase by disrupting the iron-sulfur cluster, since it is known to convert [4Fe-4S] to [3Fe-4S] clusters (5). The enzyme is extremely sensitive to oxygen, which is characteristic of many proteins that contain [4Fe-4S] clusters. Inhibition by oxygen and ferricyanide indicates that the Fe-S cluster is essential for activity. The activity of the partly purified enzyme was 95% inhibited when the enzyme was incubated with 0.4 mM propyl iodide for 15 min prior to the start of the reaction. About 65% of the total activity was restored upon illumination of the reaction mixture. This provides evidence for an essential role of cobamide in the reaction. Inhibition by propyl iodide and reversal by light was observed earlier in studies with membrane vesicles (4). ACKNOWLEDGMENTS This work was partially supported by Department of Energy grant ER20053 (S.W.R.), NSF EPSCoR cooperative agreement OSR9255225 (S.W.R.), and a Deutsche Forschungsgemeinschaft grant (G.G.). REFERENCES 1. Banerjee, R. V., S. R. Harder, S. W. Ragsdale, and R. G. Matthews. 1990. Mechanism of reductive activation of cobalamin-dependent methionine synthase: an electron paramagnetic resonance spectroelectrochemical study. Biochemistry 29:1129–1135. 2. Banerjee, R. V., and R. G. Matthews. 1990. Cobalamin-dependent methionine synthase. FASEB J. 4:1450–1459. 3. Becher, B., V. Mu ¨ller, and G. Gottschalk. 1992. The methyltetrahydromethanopterin:coenzyme M methyltransferase of Methanosarcina strain Go ¨1 is a primary sodium pump. FEMS Microbiol. Lett. 91:239–244. 4. Becher, B., V. Mu ¨ller, and G. Gottschalk. 1992. N5-methyltetrahydromethanopterin:coenzyme M methyltransferase of Methanosarcina strain Go ¨1 is a Na1-translocating membrane protein. J. Bacteriol. 174:7656–7660. 5. Beinert, H., and A. J. Thompson. 1983. Three-iron clusters in iron-sulfur proteins. Arch. Biochem. Biophys. 222:333–361. 6. Blaut, M., V. Mu ¨ller, and G. Gottschalk. 1992. Energetics of methanogenesis studied in vesicular systems. J. Bioenerg. Biomembr. 24:529–546. 7. Blum, H., H. Meier, and H. J. Gross. 1987. Improved silver staining method of plant proteins, RNA and DNA in polyacrylamide gels. Electrophoresis 8:93–98. 8. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248–254. 9. Drennan, C. L., S. Huang, J. T. Drummond, R. G. Matthews, and M. L. Ludwig. 1994. How a protein binds B12; a 3.0 Å X-ray structure of B12binding domains of methionine synthase. Science 266:1669–1674. 10. Ellman, G. L. 1958. A colorimetric method for determining low concentrations of mercaptans. Arch. Biochem. Biophys. 74:443–450. 11. Fischer, R., P. Ga ¨rtner, A. Yeliseev, and R. K. Thauer. 1992. N5-methyltetrahydromethanopterin:coenzyme M methyltransferase in methanogenic archaebacteria is a membrane protein. Arch. Microbiol. 158:208–217. 12. Ga ¨rtner, P., A. Ecker, R. Fischer, D. Linder, G. Fuchs, and R. K. Thauer. 1993. Purification and properties of N5-methyltetrahydromethanopterin:coenzyme M methyltransferase from Methanobacterium thermoautotrophicum. Eur. J. Biochem. 213:537–545. 13. Harder, S. A., B. F. Feinberg, and S. W. Ragsdale. 1989. A spectroelectrochemical cell designed for low temperature electron paramagnetic resonance titration of oxygen-sensitive proteins. Anal. Biochem. 181:283–287. 14. Harder, S. A., W.-P. Lu, B. F. Feinberg, and S. W. Ragsdale. 1989. Spectro-

Downloaded from http://jb.asm.org/ on June 21, 2017 by guest

EPR spectroscopic and electrochemical properties of the cobamide in the membrane and the partially purified methyltransferase. The membrane of the organism contains from 0.4 to 0.7 nmol of cobamide per mg of protein (3) (Table 1). The well-washed membrane (Fig. 3A) and the partially purified enzyme (Fig. 3B) exhibited similar EPR signals at 100 K characteristic of low-spin (S 5 1/2) cob(II)amide with g values of ;2.0 and ;2.3 (23, 24). Low-spin Co21 has eight absorptionlike features centered at a g value of ;2.0 due to interaction between the unpaired electron and the cobalt nucleus (I 5 7/2). Each of these eight resonances is further split into three components, characteristic of a ‘‘base-on’’ cobamide. The triplet results from superhyperfine interaction between the unpaired electron and a nitrogen (I 5 1) donor ligand that is axially coordinated to cobalt. The value of the hyperfine splitting constant (AII), which represents the strength of the interaction between the unpaired electron and cobalt nucleus, is ;100 G, consistent with coordination of an axial electrondonating group. Base-on cobamides have AII values of between 80 and 100 G, whereas ‘‘base-off’’ cobamides have a significantly larger value of ;150 G (23, 24). Whether the lower ligand is the benzimidazole base or a protein-derived ligand such as histidine has not been determined. Recently, it has been shown that some cobamide-containing proteins, such as methionine synthase (9) and a corrinoid protein from Sporomusa ovata (28), have EPR spectra characteristic of base-on conformation but are actually base-off, with a histidine nitrogen in the lower axial position. Since cobamide-containing transmethylases are active in the Co11 state, we used EPR spectroelectrochemistry to establish the reduction potential of the Co21/11 couple of the membrane-bound cobamide. We performed the redox titration with the membrane fraction that had been washed with CHAPS to reduce the viscosity and remove most of the membrane-associated hydrogenase. The experimental data fit the Nernst equation for a one-electron process with a standard equilibrium reduction potential (E90) of 2426 mV (Fig. 3, inset). Characterization of an iron-sulfur cluster in the methyltransferase preparation. Although methyltransferase was only partially purified, results from inhibition experiments and EPR spectroscopy indicate that the protein contains at least one Fe-S cluster (below). There is a possibility that it contains more than a single cluster, since iron quantitation indicated 7 to 10 Fe per mol of partially purified protein. The fully reduced enzyme exhibited an EPR spectrum at 10 K with g values at 2.1, 1.92, and 1.87 (Fig. 4A), characteristic of an iron-sulfur cluster. Double integration of the spectrum yielded 1.1 spins per mol, assuming 80% homogeneity of the enzyme sample and a molecular mass of 100 kDa. This could indicate the presence of a single Fe-S cluster. The intensity of the spectrum is dependent on temperature (Fig. 4B) and shows strong resistance to microwave power saturation with a P1/2 (the power for halfsaturation with microwave energy) of 415 mW (Fig. 4C). These properties, indicative of a very fast relaxing species, suggest that this is a [4Fe-4S]21/11 cluster rather than a [2Fe-2S] cluster (25). The overall shape of the spectrum in Fig. 4 is broad and poorly resolved. This is reminiscent of that of the [4Fe-4S]21/11 in a C/Fe-SP from C. thermoaceticum (24). The spectrum of the latter protein was transformed into a simpler spectrum with increased axial character by treatment of the protein with 2.5 M urea. However, the same treatment did not affect the spectrum of the cluster of this methyltransferase. The EPR signal of the iron-sulfur cluster exhibited Nernstian redox behavior according to a one-electron transfer and an E90 of 2215 mV

2249

2250

LU ET AL.

23. 24. 25. 26. 27. 28.

29. 30. 31. 32.

ferase, methylenetetrahydrofolate reductase, and methyltetrahydrofolatehomocysteine methyltransferase. Chem. Rev. 90:1275. Pilbrow, J. R. 1982. EPR of B12-dependent enzyme reactions and related systems, p. 431–462. In D. Dolphin (ed.), Vitamin B12, vol. 1. John Wiley & Sons, Inc., New York. Ragsdale, S. W., P. A. Lindahl, and E. Mu ¨nck. 1987. Mo ¨ssbauer, EPR, and optical studies of the corrinoid/Fe-S protein involved in the synthesis of acetyl-CoA by Clostridium thermoaceticum. J. Biol. Chem. 262:14289–14297. Rupp, H., K. K. Rao, D. O. Hall, and R. Cammack. 1978. Electron spin relaxation of iron-sulfur proteins studied by microwave power saturation. Biochim. Biophys. Acta 537:255–269. Salmon, R. T., and F. M. Hawkridge. 1980. The electrochemical properties of three dipyridinium salts as mediators. J. Electroanal. Chem. 112:253–264. Scha ¨gger, H., and G. von Jagow. 1987. Tricine sodium dodecyl sulfate polyacrylamide gel electrophoresis for separation of protein in the range from 1 to 100 kDa. Anal. Biochem. 166:368–379. Stupperich, E., J. J. Eisinger, and S. P. J. Albracht. 1990. Evidence for a super-reduced cobamide as the major corrinoid fraction in vivo and a histidine residue as a cobalt ligand of the p-cresolyl cobamide in the acetogenic bacterium Sporomusa ovata. Eur. J. Biochem. 193:105–109. van der Meijden, P., H. J. Heythuysen, A. Pouwels, F. Houwen, C. van der Drift, and G. Vogels. 1983. Methyltransferases involved in methanol conversion by Methanosarcina barkeri. Arch. Microbiol. 134:238–242. van de Wijngaard, W. M. H., C. van der Drift, and G. D. Vogels. 1988. Involvement of a corrinoid enzyme in methanogenesis. FEMS Microbiol. Lett. 52:165. Watt, G. D., Z.-C. Wang, and R. R. Knotts. 1986. Redox reactions of and nucleotide binding to the iron protein of Azotobacter vinelandii. Biochemistry 25:8156. Zehnder, A. J. B., and K. Wuhrmann. 1976. Titanium(III) citrate as a nontoxic oxidation-reduction buffering system for the culture of obligate anaerobes. Science 194:1165–1166.

Downloaded from http://jb.asm.org/ on June 21, 2017 by guest

electrochemical studies of the corrinoid/iron-sulfur protein from Clostridium thermoaceticum. Biochemistry 28:9080–9087. 15. Jablonski, P. E., W.-P. Lu, S. W. Ragsdale, and J. G. Ferry. 1993. Characterization of the metal centers of the corrinoid/iron-sulfur component of the CO dehydrogenase enzyme complex from Methanosarcina thermophila by EPR spectroscopy and spectroelectrochemistry. J. Biol. Chem. 268:325– 329. 16. Jones, J. B., Jr. 1977. Elemental analysis of soil extracts and plant tissue by plasma emission spectroscopy. Commun. Soil Sci. Plant Anal. 8:349–365. 17. Kengen, S. W. M., P. J. H. Daas, E. F. G. Duits, J. T. Keltjens, C. van der Drift, and G. D. Vogels. 1992. Isolation of a 5-hydroxybenzimidazolyl cobamide-containing enzyme involved in the methyltetrahydromethanopterin: coenzyme M methyltransferase reaction in Methanobacterium thermoautotrophicum. Biochim. Biophys. Acta 1118:249–260. 18. Kengen, S. W. M., J. J. Mosterd, R. L. H. Nelissen, J. T. Keltjens, C. van der Drift, and G. D. Vogels. 1988. Reductive activation of the methyltetrahydromethanopterin:coenzyme M methyltransferase from Methanobacterium thermoautotrophicum strain DH. Arch. Microbiol. 150:405–412. 18a.Lienard, T., B. Becher, and G. Gottschalk. Unpublished data. 19. Ljungdahl, L. G., J. LeGall, and J.-P. Lee. 1973. Isolation of a protein containing tightly bound 5-methoxybenzimidazolylcobamide (factor IIIm) from Clostridium thermoaceticum. Biochemistry 12:1802–1808. 20. Lu, W. P., and S. W. Ragsdale. 1991. Reductive activation of the coenzyme A/acetyl-CoA isotopic exchange reaction catalyzed by carbon monoxide dehydrogenase from Clostridium thermoaceticum and its inhibition by nitrous oxide and carbon monoxide. J. Biol. Chem. 266:3554–3564. 21. Lu, W.-P., I. Schiau, J. R. Cunningham, and S. W. Ragsdale. 1993. Sequence and expression of the gene encoding the corrinoid/iron-sulfur protein from Clostridium thermoaceticum and reconstitution of the recombinant protein to full activity. J. Biol. Chem. 268:5605–5614. 22. Matthews, R. G., and J. T. Drummond. 1990. Providing one-carbon units for biological methylations: mechanistic studies on serine hydroxymethyltrans-

J. BACTERIOL.