EPR Properties of the Ni-Fe-C Center in an Enzyme Complex with

THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1987 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 262, No. 32, Issue of November 15, pp. 15392-15395,1987 Printed in U.S.A.

EPR Properties of the Ni-Fe-C Center in an Enzyme Complexwith Carbon Monoxide DehydrogenaseActivity from Acetate-grown Methanosarcina thermophila EVIDENCE THAT ACETYL-COA IS A PHYSIOLOGICAL SUBSTRATE* (Received for publication, June 15, 1987)

Katherine C. TerleskyS, MichaelJ. Barbed,David J. Aceti$, and JamesG . Ferry$# From the $Department of Anaerobic Microbwlogy, VirginiaPolytechnic Institute and State University, Blacksburg, Virginia24061 and the Wepartment of Biochemistry, University of South Florida, College of Medicine, Tampa, Florida 33612

The carbon monoxide dehydrogenase complex from enzyme forms a unique paramagnetic Ni-Fe-C species when acetate-grown Methanosarcina thermophila was fur- reduced with CO (6). The CO dehydrogenase from M. thertherstudied byEPR spectroscopy.The as purified mophila is a complex of five subunits and containsNi, Fe, Co, enzyme exhibited no paramagnetic species at 113 K; Zn, and a corrinoid compound (3). The CO-reduced enzyme however, enzyme reduced with CO exhibited a complexdisplays a nickel EPR signal nearly identical to that of the EPR spectrum comprised oftwo paramagnetic species clostridial CO dehydrogenase which suggests a similar nickel with g values of gl= 2.089, g2= 2.078, andg,= 2.030 environment for both enzymes. (signal I) and gl= 2.057, gz = 2.049, and gs = 2.027 Based on thermodynamic considerations, it is predicted (signal 11). Isotopic substitution with 'j1Ni,"Fe, or 13C0 that acetate requires activation before cleavageof the carbonresulted in broadening of EPR the spectra indicating a carbon bond (7). Acetyl-coA is a likely substrate for the CO Ni-Fe-C spin-coupled complex.Pure signal I1 was obdehydrogenase complex in M. thermophila since it is synthetained following treatment of the CO-reduced enzyme sized by an enzyme with CO dehydrogenase activity in C, with acetyl-coA but not by addition of acetyl phosphate or CoASH. Acetate-grown cells were highly enriched thermoaceticum (5). However, methanogens also contain in acetate kinase (EC 2.7.2.1) and CoASH-dependent Component B which is postulated to be a functional analog phosphotransacetylase (EC 2.3.1.8) activities. These of CoASH based on partial structural similarity (8). results suggest acetyl-coA is a physiological substrate Inthis study, we demonstrate a paramagnetic Ni-Fe-C for the carbon monoxide dehydrogenase complexsyn- center in the CO-reduced M . thermophil5 CO dehydrogenase complex and show that theEPR signal is perturbed by acetylthesized in acetate-growncells of M, thermophila. CoA. This result combined with the presence of high levelsof -~ acetate kinase and CoASH-dependent phosphotransacetylase in cells suggest acetyl-coA is a physiological substrate for the CO dehydrogenase complex. The pathway of methanogenesis from acetate in Methnnosarcina thermophil5 involves transfer of the methyl group to EXPERIMENTALPROCEDURES 2-mercaptoethanesulfonicacid followed by reductive demethOrganism and Culture Conditwns-M. thermophila (9) was culylation of the CH3-S-CoM' to methane and 2-mercaptoethanesulfonic acid (1, 2). It is hypothesized that carbon-carbon turedon acetate (10) or methanol (11) as described. Growth on acetate in the presence of 61Ni wasas described (3). For growth on acetate in bond breakage may be catalyzed by an enzyme complex with the presence of "Fe, trace elements were prepared without FeSO,. CO dehydrogenase activity (3). It is further postulated that 7Hz0 and "Fe was dissolved in concentrated HCI and added to a the carbonyl group of acetate may bind to nickel in the final concentration of 27 p M Fe3+. complex followed by oxidation to CO, to supply electrons for Enzyme Assays-Cell extracts were prepared anaerobically as dethe demethylation of CH3-S-CoM to methane (4). The pro- scribed (2) except in a N2 atmosphere. Acetate kinase (EC 2.7.2.1) posed mechanism is similar to a reversal of acetyl-coA syn- and phosphotransacetylase (EC 2.3.1.8) were assayed aerobically at 'C unless otherwise noted. Units were micromoles of product/min. thesis from CoASH, a methylated corrinoid compound, and 37 Acetate kinase was assayed in the forward direction by coupling ADP CO which is catalyzed by an enzyme with CO dehydrogenase formation to the oxidation of NADH (cue = 6.22 mM"cm") with activity from Clostridium thermoaceticum(5). The clostridial pyruvate kinase and lactate dehydrogenase. The assay mixture (0.5 ~

~~

~

~

~~~~~

* This work was supported by Grant 5086-260-1255from the Gas Research Institute. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solelyto indicate this fact. 4 To whom all inquires and reprint requests should be addressed Department of Anaerobic Microbiology, Virginia Polytechnic Institute and StateUniversity, Blacksburg, VA 24061. The abbreviations used are: CH3-S-CoM,methyl-coenzyme M, 2(methy1thio)ethanesulfonicacid Component B, 7-mercaptoheptanoylthreonine phosphate; DTNB, 5,5'-dithjobis-(2-nitrobenzoicacid); EPR, electron paramagnetic resonance; TES, N-tris[hydroxymethyl] methyl-2-aminoethanesulfonic acid Tricine,N-[2-hydroxy-l,l-bis(hydroxymethyl)ethyl]glycine.

ml) contained (in final concentrations): 100 mM Tricine, pH 8.2, 60 mM potassium acetate, 1.5 mM ATP, 2 mM glutathione, 2 mM MgClp, 2 mM phosphoenolpyruvate, 0.4 mM NADH, 1unit ofpyruvate kinase, 5 units of lactic dehydrogenase. Assays were initiated with 2.5 pl of enzyme solution or cell extract. Phosphotransacetylase was routinely assayed in the forward direction by following formation of the thiol ester of acetyl-coA = 5.55 mM" cm", experimentally determined) as described (12). A second method, used where noted, employed DTNB which reacts with the sulfhydryl form (non-acetylated) of CoASH or Component B releasing the nitrobenzenethiol anion (tllZ = 13.6 mM" cm" (13)). Anaerobic assay mixtures contained (in final concentrations): 2 mM acetyl phosphate, 20 mM potassium phosphate buffer, pH 8,0.4 mM CoASH, or 0.4 mM reduced Component B. Component B, supplied as thedisulfide, was reducedwith 10fold excess sodium borohydride to the sulfhydryl form. The enzyme

15392

15393

Ni-Fe-C Center in CO Dehydrogenase from Methunosarcina reaction was stopped by boiling 100pl for 5 min followingacidification with 5 pl of 1 N acetic acid. The sample was brought to pH 8 anaerobically with 8 pl of 1 N sodium hydroxide followed byaddition of 290 pl of 20 mM potassium phosphate buffer, pH 8,and 100 p1of DTNB/EDTA solution (2 mM DTNB and 6 mM EDTA in 20 mM potassium phosphate buffer, pH 7). Enzyme Purifications-Partial purifications of acetate kinase and phosphotransacetylase were accomplished by fast proteinliquid chromatography using a model GP-250 gradient programmer (Pharmacia Biotechnology Inc). Extract (113mg of protein) from acetate-grown cells was chromatographed using a Mono Q HR 10/10 anion exchange column (Pharmacia) previously equilibrated with buffer A (50 mM TES buffer, pH 6.8,with 10% (v/v) ethylene glycol and 10 mM MgCl,). Ethylene glycol and MgCIZ wereincluded to stabilize activity during purification and storage. Protein was eluted with a linear gradient (110ml) from 0.0 to 0.4 KCl. The preparations were stored in liquid Nz. The CO dehydrogenase complex from acetate-grown cells was purified to electrophoretic homogeneity as described (3).The enzyme was further concentrated by adsorption and elution from a Mono Q column as described (3),except ammonium chloride replaced potassium chloride in the elution buffer. The concentrated enzyme solution eluted in 0.33 M ammonium cbloride. Enzyme manipulations were performed under anaerobic conditions in a Coy Anaerobic Chamber (Coy Laboratory Products, Ann Arbor, MI) or as described (3). Exchange of CO for a Nz atmosphere above the enzyme solutions was done by flushing the head space with Nt at a flow rate of 50 ml/min for 5 min. EPR Spectroscopy-EPR spectra were recorded using a Varian E109 Century Series spectrometer (Varian Associates, Palo Alto, CA) operating at 9 GHz and 100 kHz modulation and equipped with a variable temperature accessory. Spectra were routinely obtained within the temperature range 77-113 K using an incident microwave power of 20 mW and a modulation amplitude of 0.4 millitesla. Spectra were calibrated using a,a-diphenyl-j3-picryylhydrazyl-Mnz* as a g value standard (14). Double integrations of experimental spectra were carried out as described by Wyard (15)using CuEDTA as standard. Computer simulations of experimental spectra were calculated using the program described byLowe (16).Microwave power saturation measurements of EPR spectra were recorded over a 1-200 mW range of microwave power.Data of EPR signal amplitude (S) as a function of incident microwavepower (P) were analyzed as described by Barber et al. (17)using the semiempirical equation:

299

315

33 1

MAGNETIC FIELD C m T ) FIG. 1. EPR spectra of the CO dehydrogenasecomplex from M. thennophila. Enzyme preparations in 50 mM TES, pH 7.0, 10 mM MgCl2, 0.33 M NH&l, and 10% (v/v) ethylene glycol. Spectra were recorded at 113 K and 20 mW power (except where indicated). The gain was 20,000.A, enzyme at 5.2 mg/ml preincubated with W O . B , 67Fe-enrichedenzyme at 4.2 mg/ml preincubated with "CO. C, "Ni-enriched enzyme at 12.2 mg/ml preincubated with '*CO. The power was 10 mW. D,enzyme at 5.2 mg/ml preincubated with %O. The field scale corresponds to a microwave frequency of 8.988 GHz. Arrows indicate g values. mT,millitesla.

Examination of the EPR spectrum of the CO-reduced CO dehydrogenase complex from cells cultured in a medium enriched in 57Fe( I = %) (Fig. 1B) showed significant broadening S = k &'/[I + P/P++]"."* (1) of the overall spectrum, although this was most pronounced where b, the "inhomogeneity parameter," varies from 1.0 for inhom- for the g = 2.028 feature. Spectral broadening was also obogeneously broadened lines to 4.0 for the homogenous case. P,hin served for the CO-reduced enzyme isolated from cells cultured Equation 1 is the power at half-saturation and k is a constant. Saturation curves for species containing more than one relaxing in a medium enriched in 61Ni (I= 3/2) (Fig. IC). However, in component were constructed by adding together individual saturation this case, broadening was most pronounced in the g = 2.074 curves obtained for each of the components in various proportions and 2.049 features. Incubations with ' T O ( T , I = Vi) as until thebest fit was obtained. reductant (Fig. 1D) also resulted in changes in the line shape Chemicals-Acetyl-coA, CoASH, the lithium-potassium salt of of the EPRsignal although the degree of broadening was not acetyl phosphate, sodium phosphoenolpyruvate, ATP, NADH, DTNB, sodium borohydride, pyruvate kinase (rabbit muscle), and as substantial as for 67Feand 61Nisubstitution. While individlactic dehydrogenase (rabbit muscle) were obtained from Sigma. ual hyperfine interactions were not resolved under these conChemically synthesized Component B (the disulfide form) (18)was ditions, the results indicated Fe, Ni, and C were present in kindly provided by K. Noll, University of Illinois, Urbana, Illinois. these paramagnetic species, the carbon was derived from CO, All other chemicals were of reagent grade. W O (99% isotopic purity) and thesignals arose following reduction of a Ni-Fe-C center, was obtained from Merck, Sharp andDohme isotopes and "Fe (93% isotopic purity) was obtained from Oak Ridge National Laboratory, similar to thatdescribed for the clostridial enzyme catalyzing acetyl-coA synthesis. Oak Ridge, TN. The EPRsignal of the aspurified, CO-reduced, enzyme was unchanged following replacement of the CO atmosphere with RESULTS The EPRspectrum observed at 113 K of CO dehydrogenase N2. However, changes in the EPR signal were observed folcomplex, as purified under anaerobic conditions and reduced lowing addition of acetyl-coA to the CO-reduced enzyme with CO, is shown in Fig. lA. The spectrum exhibited three contained under Nz (Fig. 2 0 ) . Treatment with acetyl-coA features with g values of 2.074,2.049, and 2.028, with no other resulted in disappearance of the feature observed at g = 2.074 resonances detected within the g values range of 1.85-4.0. In (Fig. 2D), yielding a paramagnetic species of near axial symthe absence ofCO, no paramagnetic species were detected metry which was simulated (Fig. 2E) using g, = 2.057, g, = within this range. All CO-reduced samples of the CO dehy- 2.049, and g3 = 2.027.We have designated the EPR signal drogenase complex exhibited this overall line shape although observed in the presence of acetyl-coA as signal 11. Replacethe signal amplitudes of the individual features varied among ment of acetyl-coA with equimolar concentrations of acetyl enzyme preparations suggesting this spectrum was a compos- phosphate, CoASH, or dithiothreitol failed to generate signal ite of multiple species. 11. Addition of acetyl-coA to the aspurified CO dehydrogen-

Ni-Fe-C Center in CO Dehydrogenase from Methanosarcina

15394

1

2

3 4 LOG P

5

6

LW)

FIG. 3. Microwave power saturation behavior of the Ni-FeC center in the CO-reduced CO dehydrogenase from M. thermophifa.Untreated CO-reduced enzyme, 5.2 mg/ml, in 50 rnM TES buffercontaining 10 mM MgC12, 0.33 M NHF1, and 10% (v/v) ethylene glycol, pH 6.8. EPR spectra were recorded at 113 K using

299

311

323

MAGNETIC FIELD ( m T ) FIG. 2. Experimental and simulated EPR spectra of the CO dehydrogenase complex from M. thermophila. Enzyme preparations in 50 r n TES, ~ pH 6.8, 10 mM MgClz, 0.33 M NH,Cl, and 10% (v/v) ethylene glycol. Experimentallyobtainedspectra were measured at 113 K and 20 mW power. The gain was 20,000 except where indicated. A, enzyme at 5.2 mg/mlincubatedwith"CO. B, simulationof signal I using the following EPR parameters: g, = 2.089, gz = 2.078, g3 = 2.03, W I = 1.5, wz = 1.5, w3 = 0.45. C, simulation of composite signal with EPR parameters obtained from E and B. D , "CO-reduced enzyme at 2 mg/ml preincubated with 20 mM acetylCoA. The gain was 50,000. E, simulation of spectrum in I) using the following EPR parameters: g, = 2.057, g2 = 2.049, g3 = 2.027, w1 = 1.18, wz 0.73, w3 = 0.45. Referred to as signal 11. The fieldscale corresponds to a microwave frequency of8.988 GHz. Arrowsindicate g values. mT, millitesla.

ase complex under 1 atm of CO had no effect on the EPR signal. Computer subtraction of the EPR signal observed in the presence of acetyl-coA (signal 11) from that of the untreated CO-reduced enzyme yielded a second species of nearly axial symmetry that was simulated using gl = 2.089, g, = 2.078, and g, = 2.030 (Fig. 2B) and is referred to as signal I. Good agreement was obtained between the experimental spectrum for the CO-reduced untreated enzyme andthesimulated spectrum obtainedby adding 23% of signal I to signal I1 using the computer (Fig. 2 C ) . These results indicated the signal from untreated enzyme was a composite of two overlapping species. Examination of the microwave power saturation characteristics of signals I and I1 suggested different saturation properties for the two species. The resultsof the saturation studies performed a t 113 K are shown in Fig. 3.As the microwave power was increased, signal I1 saturated more readily with a P,,%of90 mW while signal I was more resistant to saturation with a P14of 150 mW. The saturation behavior of the g3 feature of the experimental spectrum, which is a composite of signals I and 11, was fit well by a composite curve consisting of the two independently saturating components: 70% of the signal exhibiting a P14of 150 mW and30% exhibiting a PIA of 90 mW. Cell extracts of acetate-grown M. thermophil5 contained

0.4 milliteslamodulationamplitude.Experimentalpoints,corre. sponding to signal I (A),and signalI1 (0)were fit by Equation 1. For the spectral feature comprising signals1 and I1 (m), Equation 1 was

modified t.0 allowfortworelaxingcomponentswith corresponding to 70% signal I and 30% signal 11.

parameters

TABLE I Acetate kinase and phosphotransacetylase activitiesin acetate-grown or methanol-grown cells of M.thermophilu Acetate Phosphotransacetylase kinase

K,

Growth substrate specific activityb

Acetate

Specific activityb

K,

bM)'

Acetyl ohosohate ~

~~

~~

~

Acetate 2.42 k 0.23 24 0.04 53.ck 14.2 0.65 0.18 Methanol 0.12 k 0.02 ND' ND 3.0 k 0.5 ND ND "Determined from double-reciprocal plotsof initial velocity versus substrate concentrations using partially purified preparations of acetate kinase (50.2 units/mg) and phosphotransacetylase (451 units/ mg). The range of substrate concentrations used were: acetate, 1-400 mM; ATP, 0.001-0.8 mM; acetyl phosphate,0.05-3.5 mM; and CoASH, 0.02-0.3 mM.

'

Expressed as micromoles of product/min/mg cell extract protein. Values are the mean of three determinations, with standard error, Assays were initated with 2.5 pl of cell extract (10.9-13.8 mg of protein/ml). ND, not determined.

high specific activities of acetate kinase and phosphotransacetylase suggesting involvement of these enzymes in a major catabolic pathway (Table I). Furthermore, acetate kinase and phosphotransacetylase activities were 20- and 18-fold higher in acetate-grown cells when compared with methanol-grown cells (Table I). When component B was substituted forcoenzyme A as a substrate for thephosphotransacetylase, no activity was detected in extracts orwith the partiallypurified enzyme using either assay method (data not shown). These results suggest that acetyl-coA is the activated form of acetate involved in the pathway of acetate conversion to methane in M.thermophila. DISCUSSION

The CO-reduced CO dehydrogenase complex, as purified from M.thermophib, showed paramagnetic properties which suggest a Ni-Fe-C spin-coupled complex. The EPRproperties were similar to the as purified CO dehydrogenase from C.

Ni-Fe-C Center in

CO Dehydrogenase from Methanosarcina

15395

thermoaceticum (6)suggesting similar reaction centers in the trophic and N2-fixing organisms have been described. The two enzymes. The results also showed that, like the Clostridial methyl-coenzyme M methylreductase catalyzing the final reenzyme, the EPRspectrum of the CO dehydrogenase complex ductive step in methanogenesis, contains the nickel-tetrapyrwhich contains aparamagnetic Ni center from M. thermophila is a composite of two signals that were role, coenzymeF430r resolved by interaction of the enzyme with acetyl-coA. How- withg values greater than 2.1 (26). Extended x-ray absorption ever, differences between the Clostridial and methanogen en- fine structure reveals a Ni(I1) center liganded to four pyrrole zymes areapparent.(i) In the presence of acetyl-coA or N atoms and one or two axial 0 or N atoms, but the actual CoASH the Clostridial enzyme shows only one signal (signal oxidation state of nickel in the protein is believed to be Ni(1) I) with g values of 2.074, 2.074, and 2.028 (6). The second or Ni(1II) (27, 28). The configuration of the Ni-Fe-C center of CO dehydrogensignal (signal II), with g values of 2.062, 2.047, and 2.028, is not experimentally observed separately from signal I and is ase in acetogenic clostridia and acetotrophk methanogens is only resolved by simulation (6). In contrast, treatment of the unknown. It is unlikely that nickel in the M. thermophilu CO methanogen enzyme with acetyl-coA resulted in one EPR dehydrogenase complex is present in coenzyme FdS0since the signal (signal 11) but with g values similar to simulated signal enzymedoes not exhibit strong absorption in the 430-nm 11of the clostridial CO dehydrogenase; the other methanogen region and g values were different than previously observed enzyme signal (signal I) was only obtained separately from for coenzyme F430 or coenzyme Flao-containing methyl-coensignal I1 by simulation but was similar to theexperimentally zyme M methylreductase (26). Extended x-ray absorption fine obtained pure signal I of the acetyl-CoA-treated Clostridial structure studiesindicate that nickel in theCO dehydrogenase CO dehydrogenase. (ii) CoASH or acetyl-coA influences the from C. thermoaceticum is liganded to N or 0 and S atoms EPR signal of the Clostridial enzyme; however, CoASH had (29). Similar studies are necessary with the methanogen enno effect on the signal of the methanogen enzyme. (iii) Ex- zyme to elucidate the nickel ligands. change ofCO for a N2 atmosphere above the CO-reduced Acknowledgment-We wish to thank Dr. Steve Ragsdale for helpful methanogen enzyme solution was necessary for acetyl-coA perturbation of the EPR signal; in contrast, acetyl-coA influ- comments in preparation of the manuscript. ences the EPR signal of the Clostridial enzyme under CO. REFERENCES These differences are unexplained but may reflect the ability 1. Lovley, D. R., White, R. H.,and Ferry, J. G. (1984) J. Bacteriol. 160,521525 of the enzymes to catalyze similar reactions albeit in opposite 2. Nelson, M. J. K., and Ferry, J. G. (1984) J. Bacteriol. 160,526-532 directions. 3. Terlesky, K. C., Nelson, M. J. K., and Ferry, J. G. (1986)J. Bncteriol. 1 6 8 , 1053-1058 The influence of acetyl-coA on the EPR signal of the M. 4. Nelson, M . J. K. Terlesky, K. C., and Ferry, J. G. (1987) Microbiul Growth thermophilo CO dehydrogenase may result from direct interon C-1 c o m p ~ n d spp. , 70-76, Martinus Mijhoff, Dortdrecht 5. Ragsdale, S. W., and Wood, H. G . (1985) J. Biol. Chem. 260,3970-3977 action with the Ni-Fe-C center or by allosteric changes in6. Ragsdale, S. W. Wood, H. G., and Antholine, W. E. (1985) Proc. Nntl. duced by acetyl-coA binding at an alternate site. Although Acnd. Sei. U. A. A. 82,6811-6814 7. Eikmanns, B., and Thauer, R. K. (1984) Arch. Microbiol. 138, 365-370 the mechanism is unresolved, the results clearly show that 8. Noll, K. M., Rinehart, K. L., Jr., Tanner, P. S., and Wolfe, R. S. (1986) acetyl-coA binds to the methanogen enzyme and induces Proc. Natl. Acad. Sei. U. S. A. 83, 4238-4242 S. H., Sowers, K. R.. and Ferry, J. G . (1985) Int. J. Syst. Bacteriol. 9. changes in the environment of the Ni-Fe-C center. This result Zinder, 35.622423 ","_ and the presence of an acetate-inducible acetate kinase and 10. Sowers, K. R., Nelson, M. J., and Ferry, J. G. (1984) Curr. Microbwl. 11, 227-230 CoASH-dependent phosphotransacetylase in cells suggest 11. Murray, P. A,, and Zinder, S. H. (1985) Appl. Enuiron. Microbiol. 60,49that acetyl-coA is likely to be the activated form of acetate 55 cleaved by the CO dehydrogenase complex from this organism. 12. Oberlies, G., Fuchs, G., and Thauer, R. K. (1980) Arch. Microbiof. 1 2 8 , 248-252 Acetate kinase and phosphotransacetylase are also induced in 13. Ellman, G. L. (1958) Arch. Biochem. Biophys. 74,443-450 14. Bolton, J. R., Borg, D. C., and Swartz, H. M. (1972) in Bwlogical Applicnacetate-grown cells of Methanosarcina barkeri (19,20). tions of Electron Spin Resonance (Swartz, H. M.,Bolton, J. R., and Borg, The required absence of a CO atmosphere for binding of D. C.,eds) John Wiley & Sons, New York 15. Wyard, S. J. (1965) J. Sci. Instrum. 42,768-769 acetyl-coA to the methanogen enzyme is unexplained. The 16. Lowe, D. J. (1978) Biochern. J . 171,649-651 extent to which CO was removed from the enzyme or enzyme 17. Barber,M. J., Salerno, J. C., and Siegel, L. M. (1982) Biochemistry 2 1 , 1648-1656 solution was not determined; thus, following addition of ace18. Noll, K. M.,Donnelly, M. I., and Wolfe, R. S. (1987) J. Ed. Chem. 262, tyl-coA, itwas unknown if the origin of carbon in the Ni-Fe513-515 C center was other than from CO. Nonetheless, the results 19. Kenealy, W. R., and Zeikus, J. G. (1982) J. Bacteriol. 151,932-941 20. Laufer, K., Eikmanns, B., Frimmer, U., and Thauer, R. K. (1987) 2. could explain why we were unsuccessful in attempts to demNaturforsch. 42, 360-372 onstrate cleavage of the acetyl group of acetyl-coA by ex- 21. Graf, E.:G., and Thauer, R. K. (1981) FEBS Lett. 1 3 6 , 165-169 w. L., Whitman, W. B., and Wolfe, R.S. (1982) Pmc. Nntl. Acad. change of the carbonyl carbon with atmospheric C02 as re- 22. Ellefson, S e i 11 ~7n7-371n ~ -. -.S A 7",Q ".". -.~" 23. Hausinger, R. P. (1987) Microbiol. Reu. 5 1 , 22-42 ported for the Clostridial enzyme (5). 24. Albracht, S. P. J., Graf, E.-G., and Thauer, R.K. (1982) FEBS Lett. 1 4 0 , CO dehydrogenase is the third nickel-containing enzyme 311-313 described among methanogenic bacteria following hydrogen- 25. Lindahl, P. A,, Kojima, N., Hausinger, R. P., Fox, J. A., Teo, B. K., Walsh, C. T., and Orme-Johnson, W. H. (1984) J. Am. Chem. SOC.1 0 6 , 3062ase (21) and methyl-coenzyme M methylreductase (22, 23). 3064 26. Albracht, S. P. J., Ankel-Fuchs, D., van der Zwann, J. W., and Fontijn, R. Methanogen hydrogenases contain nickel paramagnetic resD., and Thauer, R. K. (1986) Bwchem. Bwphys. Acta 870,50-57 onance in certain oxidation states with characteristic g values 27. Diakun, G. P., Plggott, B., Tinton M. J., Ankel-Fuchs, D., and Thauer, R. K. (1985) Biochem. J. 232,281y284 of 2.3, 2.23, and 2.02 (24). Extended x-ray absorption fine M. K., Sullivan, R. J., Schwartz, J. R. Hartzell P. L., Wolfe R. structure studies indicate the nickel is liganded to 3 or 4 S 28. Eidsness, S., Flank, A,"., Cramer, S. P.. and Scott. R. A. (19&36) J. Am. &m. Soc. 108.. 3120-3121 atoms (25). Similar hydrogenases from other hydrogeno~"

~~~~

* K. C. Terlesky and J. G. Ferry; unpublished results.

"~

29. Cramel, S. P., Eidsness, M. K., Pan, W.-H., Morton, T. A. Ragsdale S. W., Der Vartanian, D. V., Ljungdahl, L. G., and Scott, R. A: (1987) I&g. Chem. 26,2477-2479